United States
Environmental Protection
Agency
Office of Air Quality
Planning And Standards
Research Triangle Park, NC 27711
EPA-450/4-86-010
Revised
February 1990
AIR
SEPA
COMPILING AIR TOXICS
EMISSION INVENTORIES
SECOND EDITION
-p*
\
\
-------
REPORT USER FEEDBACK AND MAIL KEY REGISTRATION
The U.S. Environmental Protection Agency's (EPA) Office of Air Quality
Planning and Standards (OAQPS) provides technical support to assist State and
local air pollution control agencies in developing and implementing air toxics
programs. One way that OAQPS provides assistance to agencies and other
interested individuals is by compiling and publishing emission data for agencies
and others who are interested in locating potential emitters of toxic air
compounds and in making preliminary estimates of toxic air emissions. These
reports published by EPA are introductory documents only, and they are not
intended to provide exact estimates of air toxics releases from specific
facilities. EPA will update and expand these reports and publish new documents
as toxic air emissions data are obtained. Your comments on the usefulness of
this report and availability of additional data which could be used to extend
and improve it, are important input to this process. Please provide any
information to us that will allow us to improve these reports. The format below
is provided for your convenience.
Please check the appropriate blanks and mail to:
Pollutant Characterization Section
Noncriteria Pollutant Programs Branch (MD-15)
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
I have additional air toxics emission data that would help EPA. Please
contact me.
Other comments on the report or needs for similar reports.
NAME:
POSITION:
COMPANY/AGENCY:
MAILING ADDRESS:
PHONE NUMBER: ( ).
REPORT TITLE:
-------
EPA-450/4-86-O1O
Revised February 1990
COMPILING AIR TOXICS
EMISSION INVENTORIES
SECOND EDITION
Edited and Revised
by
Dallas W. Safriet
Air Quality Management Division
U.S. Environmental Protection Agency
US. ENVIRONMENTAL PROTECTION AGENCY
Office Of Air And Radiation
Air Quality Management Division
Office Of Air Quality Planning And Standards
Research Triangle Park, North Carolina 27711
February 1990
-------
This report has been reviewed by the Office Of Air Quality Planning And
Standards, U.S. Environmental Protection Agency, and approved for publication.
Any mention of trade names or commercial products is not intended to constitute
endorsement or recommendation for use.
EPA-450/4-86-O1O
ii
-------
TABLE OF CONTENTS
Chapter Pane
List of Tables V
List of Figures Vi
1 Introduction 1
2 Anticipated Uses of Air Toxics Emission Inventories . 3
2.1 Uses of Air Toxics Emission Inventories 3
2.2 Effect of Program Use on Inventory Complexity . 5
2.3 Additional Planning Considerations 6
3 Preliminary Screening Studies 8
3.1 Screening Considerations 8
3.2 Overview of Screening Procedures 8
3.3 Screening Study Error 16
3.4 Review of Screening Study Results 17
3.5 Use of Screening Tools Evaluating Permit
Applications 18
4 Pollutant Coverage 19
4.1 “Open-ended” Versus ‘Directed” Approaches 19
4.2 State and Local Agency Lists 21
4.3 Core List of Compounds for Possible
Consideration 21
5 Source Category Coverage 24
6 Point/Area Source Distinctions 32
6.1 Making the Point/Area Source Distinction 32
6.2 Factors Affecting the Point/Area Distinction .. 32
7 Geographic Area of Coverage 34
8 Detailed Data Collection and Emission Estimation
Procedures 35
8.1 Locating Sources 35
8.2 Data Collection 36
8.3 Emission Estimation Procedures 39
i ii
-------
Chaoter page
8.4 Control Device Applicability and Efficiency ... 40
8.5 Problems in Data Collection and Interpretation. 41
8.6 Sources of Inventory Error 42
9 Summary 43
References 44
ADoendix Page
A Example Applications of Screening Tools A-i
B Example Air Toxics Production and Use
Questionnaire B-i
C Glossary of Chemical Name Synonyms C-i
iv
-------
LIST OF TABLES
TABLE PAGE
1 Potential Uses of Air Toxics Emission Inventories ... 4
2 Inventory Complexity as a Function of Application ... 6
3 Features of Screening and Detailed Inventories 9
4 Source of Toxic Emission Factor Information 11
5 Example VOC Profile 14
6 Example Particulate Matter Profile 15
7 Percent of Cancer Risk Associated with Point and
Area Sources 17
8 Substances Considered by Four or More Agencies 22
9 Core List of Compounds for Consideration in an Air
Toxics Emission Inventory 23
10 Nationwide Air Toxics Emissions by Board Source
Category 27
11 Example Cutoff Levels and Exempted Sources 33
V
-------
FIGURE PAGE
1 Sample Material Safety Data Sheet 38
v i
-------
Chapter 1
INTRODUCTION
The purpose of this report is to offer assistance to those agencies who are
getting started in compiling inventories of air toxics emissions. This
compilation process may simply involve the development of a preliminary screening
inventory or may involve the development of a more detailed and comprehensive
inventory, depending on program needs; resources and the nature of the air toxic
probl em.
Emission inventories have generally been compiled for use in criteria
pollutant control programs in order to (1) provide summary information on sources
and emissions, (2) help define source/receptor relationships and (3) develop and
evaluate alternate control strategies. Air toxics emission inventories are now
being developed and used in air toxic control programs for basically the same
reasons.
Recent survey information collected for inclusion in the U.S. Environmental
Protection Agency’s (EPA’s) National Air Toxic Information Clearinghouse (NATICH)
data base indicates that about 36 States and local agencies have compiled air
toxics emission inventories in some form. EPA is strongly emphasizing the
development of State and local inventory and other data collection capabilities
as part of its National Air Toxics Strategy.’
To complement this discussion, several appendices are included which present
examples and other useful data in compil ing air toxics inventories. The emphasis
in these appendices is on presenting various tools and other techniques for
identifying pollutants of potential concern and for estimating emissions.
The contents of these appendices are summarized below:
•Appendix A presents example applications of various screening tools
that are available to develop preliminary emission estimates.
•Appendix B presents an example of a production and use survey,
sometimes used to determine which sources should receive more detailed
follow-up questionnaires or to supply emergency response data.
•Appendix C presents a glossary of chemical names/synonyms For commonly
inventoried substances, useful in interpretating data supplied by
sources.
EPA has publ ished procedures for compil ing emissions inventories of criteria
pollutants. 27 Many of these same procedures will apply to air toxics emission
inventory development and an agency’s existing criteria pollutant inventory may
serve as a good starting point for compiling the air toxics inventory.
-------
The main body provides a general discussion of the issues and concepts
involving air toxic inventories. It advises State and local agencies of various
considerations that should be taken into account when planning and compiling air
toxic inventories. The following topics are discussed that are considered
fundamental to any such effort:
-Anticipated uses of air toxic emission inventories
-What substances to cover
-What source categories to cover
-Point/area source distinctions
-Geographic area of coverage
-Source location, data collection and emission estimation procedures
(screening and detailed)
This report is primarily geared to helping agencies in the planning and
screening portions of their air toxics programs. Such as (1) basic considera-
tions are emphasized that the agency should take into account when planning the
inventory effort and (2) simple tools are presented to help the agency make
prel iminary, screening estimates of source strength without having to make source
contacts.
To some extent, the screening tools presented herein can be extended to
provide more detailed emission estimates and to screen information provided by
sources in permit applications. Also, the references cited in this reDort, and
additionally in the EPA clearinghouse bibliography, 8 will provide some of this
support.
2
-------
CHAPTER 2
ANTICIPATED USES OF AIR TOXICS EMISSION INVENTORIES
The paramount consideration in planning any inventory is how, ultimately,
the inventory will be used in the agency’s air quality control program. Since
air toxics emissions data may also be used in water or solid waste programs or
by other agencies, outside groups, and individuals, the needs of these other
potential users should also be considered.
2.1 Uses of Air Toxics Emission Inventories
Various uses of air toxics inventories by State and local agencies are
summarized in Table 1. Several of the more important of these applications are
discussed below.
A. Defining General Source and Emission Patterns
Air toxics inventories can be used for relatively undemanding purposes
such as developing preliminary screening estimates for various source categories
to help understand broad emission patterns and trends. For example, an agency
may want the capability to determine the largest point source emitters of a
particular substance in a given area, or determine the relative contribution of
certain point and area sources. A screening inventory may help the agency focus
on certain pollutants and sources in its detailed inventory. Or an emergency
response program may desire the ability to identify whether certain cnemicals
are stored or used at various plants. These types of applications would not
necessarily require the collection of extremely detailed source and emissions
data or the development of a sophisticated data handling system to store and
manipulate these data. This abbreviated type of inventory may not support
detailed exposure modeling or risk assessment analyses, but would allow the
agency to start defining its air toxics probi em and prioritizing subsequent data
collection activities. (Screening is discussed in more detail in Chapter 3.)
B. Point Source Emission Model ing
Many agencies use their inventories to provide input to various kinds
of dispersion models for estimating ambient air concentrations and new and/or
existing point sources. Typically, in this approach, these modeled ambient air
levels are then compared with ambient air standards or guidelines (such as
occupational exposure limits divided by a safety factor) to determine what
control may be required. In this type of program, each source is generally
evaluated as if it were an isolated emitter, without consideration of the
aggregate effects of nearby sources.
3
-------
TABLE 1
POTENTIAL USES OF AIR TOXICS EMISSION INVENTORIES
1. To satisfy a legislative or other mandate
2. To identify sources and general emission strengths, patterns, and trends
3. To store data from related programs
e.g. -- permit/registration/compl lance data
- - emergency preparedness data
- - right-to-know data
-- act as “tickler file” for permit review and enforcement
action
4. To site ambient air monitors
5. To provide input to point source dispersion models to predict ambient
air levels
e.g. -- to compare with acceptable ambient air levels (AALs)
-- to determine maximum individual risks and aggregate
incidence
6. To focus subsequent inventory work or other program development efforts
7. To identify multiple source and multiple pollutant problem areas
characterized by high additive risk
8. To develop control strategies and regulations
Note: This is not intended to be all inclusive. It reflects uses indicated by
various State and local agencies and EPA.
In general, the kinds of data needed for modeling air toxics around
point sources are the same as for criteria pollutants since, for most releases,
air toxics are assumed to behave in the same ways as criteria pollutants. (Some
accidental releases may involve such large quantities of heavier-than-air
compounds that conventional Gaussian plume modeling may not be applicable.)
Indeed, the same dispersion models (e.g., SCREEN, CDM, ISC) may be used in
certain air toxics appi ications. To model point sources, detailed stack, exhaust
and sometimes building parameters are needed as well as temporally resolve
emission estimates. Different and more exacting source data may be required for
modeling short-term releases than for projecting annual averages. The input
requirements to each model will vary, of course, and cannot be summarized here.
Various available models are described in Reference 9 and 10.
4
-------
Some models require more detailed source and emissions data than the
agency may feel are reasonable to collect or estimate for all sources and all
pollutants. In this situation, the agency may opt, instead, to prepare a
detailed ad hoc inventory only in those special situations where complex model ing
is necessary.
C. Multiple Source/Pollutant Assessment and Control Strategy Develooment
A few programs are starting to go beyond individual point source
assessment and are beginning to evaluate the impact of all sources of air toxics
in some areas. To evaluate the impact of aggregations of point and area sources
in an area, an agency must consider minor point and area sources as well as major
point sources. Such assessments may entail the use of areawide dispersion and
exposure models which require detailed source and emissions data for both major
and minor point sources as well as spatially resolved estimates for area sources.
The inclusion of minor point and area sources in the air toxics inventory
requires more resource expenditure than just including major point sources. The
use of air toxics emissions data in multiple source/pollutant assessments and
control strategy development represents the most demanding use of these emission
inventories.
2.2 Effect of Program Use on Inventory Complexity
Different levels of detail are needed depending on the anticipated uses of
the inventory. Table 2 shows some of the inventory parameters that increase n
complexity as the inventory applications become more demanding. Because this
is such an important consideration, the agency should determine what uses the
air toxics inventory will be put to at the outset of the inventory planning
process. The important point is that the inventory agency must be aware of all
of the projected uses of the inventory before commencing with data collection.
In general, inventory design and data collection needs will be determined by the
most demanding applications of the inventory.
F-
-------
TABLE 2
INVENTORY COMPLEXITY AS A FUNCTION OF APPLICATION
Inventory Use Level of Comolexity
Less Demanding Applications Simple Data Summaries
Aggregated Source Data
Major Point Source Emphasis
More Source Contacts
Stack and Exhaust Data
Control Equipment Type And Efficiency
Spatial And Temporal Resolution
Inclusion Of Minor And Area Sources
I Control Strategy Simulation Capability
More Demanding Applications Complex Data Handling Capability
Note: Refer to Table 1 for various applications of air toxics emission
inventories. In general, the applications in Table 1 become more
demanding with increasing order of listing.)
2.3 Additional Plannina Considerations
Several other considerations should be made in the early planning stages
that will effect the usefulness of the air toxics inventory. These are discussed
as follows.
A. Actual vs. Allowable Emissions
Agencies typically include estimates of actual emissions in their
inventories. Some agencies also include allowable emissions, perhaps as
reflected in each source’s permit. The inclusion of allowable emissions may be
useful to the agency in certain modeling control or compliance programs.
6
-------
B. Annual vs. Short Term Emissions
Agencies traditionally have compiled estimates of annual emissions. Annual
estimates of air toxic emissions are appropriate when the agency is evaluating
potential cancer risks since carcinogenesis involves long term exposures.
Shorter term emission estimates may also be useful for modeling short term,
maximum concentrations around certain sources, such as when estimating acute
health effects posed by certain pollutants. In addition, allowable emission
limits may be expressed on a short term time frame.
C. Accidental vs. Routine (oredictable) Emissions
Agencies have generally only considered routine, predictable emissions in
their inventories, i.e., those emissions that could be expected to recur in a
certain time frame. Routine emissions would include continuous emissions, both
process and fugitive, and intermittent emissions from scheduled, but discon-
tinuous, operations (e.g., batch processes, startups, shutdown and maintenance
operations). Accidental releases would be short term and unscheduled and may
involve events that are catastrophic in nature. Accidental events could involve
control equipment failure or bypass, pressure relief, equipment openings and
spills. There is a fine line between what some may consider irregularly
occurring, but predictable, events and accidental events. The latter would not
necessarily be expected to happen at every plant within a given reporting year
and may be difficult to handle in a representative manner in the air toxics
inventory.
0. Process vs. Fugitive Emissions
Process, or stack, emissions have traditionally been the focal point of
emission inventories. Fugitive emissions can be a very important component from
some source categories (e.g., organic chemical manufacturing) and should not be
overlooked. Important sources of fugitive emissions include material storage,
loading and unloading operations, process valves, flanges, pump seals, open-
ended lines, sampling ports, etc.
7
-------
CHAPTER 3
PRELIMINARY SCREENING OF SOURCES AND EMISSIONS
It may prove helpful as part of the planning process to conduct some sort
of screening study before commencing with the detailed air toxics inventory.
The general idea of a screening study is to develop preliminary estimates of
emissions and associated risks in order for the agency to focus its program and
resources on the most important sources and pollutants. EPA’s “Six-Month Study”
is an example of a national screening study conducted to evaluate maximum
individual lifetime risks, additive risks and aggregate cancer incidence.” Table
3 contrasts the features of the screening and detailed emission inventories.
3.1 Screening Considerations
As part of the screening study, the agency may consider more than just
emissions - - ambient air concentrations and health data may also be factored into
the screening process. The screening study should afford the agency with enough
perspective to determine the following:
• What pollutants should be covered in more detail?
• What source categories should be emphasized?
• What geographic areas should be covered? Is there is a need to focus
on areas of high additive risk from multiple sources and pollutants?
• What is the relative importance of major and minor sources and area
sources?
To what extent can the existing criteria pollutant inventory and permit
files be used as a foundation for the air toxics inventory?
Ideally, the screening study should be performed quickly and inexpensively
and yet yield results the agency has sufficient confidence in to make decisions
concerning program direction. Screening can be an iterative process, with more
focus placed on certain pollutants and sources as the agency’s perception of its
air toxics problem unfolds and as resources allow.
3.2 Overview of Screening Procedures
To minimize resources and time spent on the screening study, the agency
should try to identify as many sources and to make as many emission estimates
as possible using available data without making expensive and time consuming
source contacts. A number of screening tools are available to help the agency
identify potential emitters of air toxics and develop preliminary emission
estimates. These tools are described as follows, with the applicability of each
tool discussed below. With example applications shown in Appendix A.
8
-------
TABLE 3
FEATURES OF SCREENING AND DETAILED INVENTORIES
Screening Inventory
- Done first
- Preliminary emission estimates Minimal or
No
- Minimal source contacts Source Contacts
- Maximum use of available data
- Can cover a large number of pollutants
- Can cover all potential sources
- Broad geographical coverage
______________________ 1. ________________
Detailed Inventory
- Follows screening inventory
- Detailed emission estimates Extensive
Source Contacts
- Extensive source contacts
- May be selective e.g., many factors
particular areas per screening study
results
Used for .
- e.g., may focus on particular sources
- e.g., may focus on particular pollutants
9
-------
A. Source Category-Pollutant Cross-Indexes
These are tables associating source categories with pollutants, showing
the potential for a given substance to be emitted by a particular facility.
This qualitative identification process can be useful for two purposes:
1. to determine which sources to send questionnaire to if the
agency is focusing on specific pollutants; and
ii. to direct the agency to the appropriate emission factors,
speciation factors, etc., if the agency is going to make
screening estimates itself (as opposed to sending out ques-
tionnaires).
These cross indexes could also be used by the agency to help evaluate
application for new or revised operating permits. Such a cross index is
contained in the report, Toxic Air Pollutant/Source Crosswalk - A Screening
Tool For Locating Possible Sources Emitting Toxic Air Pollutants. Second
Edition (EPA-450/2-89-017). The cross-walk contains pollutant names, CAS
numbers, Standard Industrial Classification (SIC) codes, and emitting source
classification codes (SCC’s). The crosswalk tables are sorted by pollutant,
SIC codes, and SCC’s.
B. Emission Factors
Just as criteria pollutant emission factors can be app.lied to activity
levels or throughput to estimate emissions, so can air toxic emission factors
be applied for this same purpose. The major source of air toxics emission
factors is the EPA series entitled Locating And Estimating Air Emissions From
Sources Of (Substance Or Source Cate orv ) and the document Toxic Air Pollutant
Emission Factors A Compilation For Selected Air Toxic Compounds And Sources
(Table 4 shows a listing). The compilation is a general emission factor
listing for more pollutants than contained in the Locating And Estimating
series. The document list available air toxic emission factors and references
from which the emission factors are obtained.
A database management system for use on IBM compatible personal com-
puters has been developed that contain all crosswalk data and air toxic
emission factors in the report discussed earlier plus recent updates. The
system, XATEF, is now available and was last updated in October, 1989. A
system user’s Manual entitle Crosswalk/Air Toxic Emission Factor Data Base
Management System User’s Manual is also available to explain the use of data
management software.
10
-------
TABLE 4
SOURCE OF TOXIC EMISSION FACTOR INFORMATION
LOCATING AND ESTIMATING SERIES
SUBSTANCE OR
SOURCE CATEGORY
EPA PUBLICATION
NUMBER
NTIS ORDER
NUMBER
DATE
Acrylonitrile
Carbon Tetrachloride
Chloroform
Ethylene Dichioride
EPA-450/4-84-007a
EPA-450/4-84-007b
EPA-450/4-84-007c
EPA-450/4-84-007d
P8-84-200609
P8-84-200625
PB-84-200617
PB-84-239193
1984
1984
1984
1984
Formaldehyde
Nickel
EPA-450/4-84-007e
EPA-450/4-84-007f
PB-84-200633
P8-84-210988
1984
1984
Chromium
Manganese
EPA-450/4-84- 007g
EPA-450/4-84-007h
PB-85-106474
PB-86-117587
1986
Phosgene
Epichlorohydrin
EPA-450/4-84-007i
EPA-450/4-84-0073
PB-86-117595
P8-86-117603
1986
1986
Vinyl idene Chloride
Ethylene Oxide
EPA-450/4-84-007k
EPA-450/4-84-0071
PB-86-117611
P8-87-113973
1986
1987
Chlorobenzenes
EPA-450/4-84-007m
PB-87-189841
1987
ol ychi orinated
Biphenyls (PCBs)
EPA-450/4-84-OO7n
PB-87-209540
1987
Polycyci ic Organic
Matter (POM)
EPA-450/4-84-007p
P8-88-149059
1988
Benzene
EPA-450/4-84-007q
P8-88- 196175
1988
Organic Liquid Storage
Tanks
EPA-450/4-88-0O4
P8-89-129019
Coal and Oil Combustion
Sources
EPA-450/2-89-OO1
PB-89- 194229
1989
Municipal Waste
Combustors
EPA-450/2-89-006
P8-89-195226
1989
Perch 1 oroethyl ene and
Trichi oroethyl ene
EPA-450/2-89-013
PB-89-235501
1989
Chromium (Supplement)
EPA-450/2-89-002
P8-90-103243
1989
1,3 Butadiene
EPA-450/2-89-021
(not assigned)
1989
Sewage Sludge Incineration is to be released in 1990.
OTHER REPORTS
1. Toxic Air Pollutant Emission Factors for Selected Air Toxics Compounds
and Sources, EPA-450/20-88-006a, NTIS, PB 89-135644. October 1988.
ii
-------
2. Compilation And Speciation Of National Emissions Factors For Consumer!
Commercial Solvent Use, EPA-450/2-89-008, NTIS Number: PB-89-207203.
3. Procedures For Estimating And Allocating Area Source Emissions Of Air
Toxics, Working Draft.
Another potential source of emission factor information is EPA’s National
Air Toxic Information Clearinghouse (NATICH) data base. 12 States and local
agencies have submitted permit and source test information to the clearing-
house that may be of assistance in determining emission rates for many kinds
of industrial facilities.
Contact names and phones numbers are also given, allowing the user the
opportunity to get more specific directly from the supplying agency. As more
data are supplied, the Clearinghouse should improve as a source of emission
data.
In general, emission factors are a useful screening technique since they
can be applied to the level of activity specified in the existing criteria
pollutant inventory or permit file to estimate emissions of air toxics.
Emission factors are applicable for estimating emissions of sources for which
factors are available and for which appropriate source information is avail-
able to apply them. An example emission factor application is shown in
Appendix A.
C. Speciation Factors Or Profiles
These are factors that can be applied directly to volatile organic
compounds (VOC) or particulate matter (PM) totals in an existing criteria
pollutant inventory or permit file to yield estimates of air toxics emissions.
They are distinguished from emission factors in that they apply to VOC and PM
emissions rather than activity levels or throughputs.
The VOC and PM speciation factors or profiles can be found in the two
volume Air Emissions Species Manual. 13 Speciation factors are used to estimate
emissions of air toxics from emission factors or estimates of total VOC’s or
PM. Both volumes are similarly organized with speciation data presented by
source category and by SCC. Volumes are updated periodically.
VOC Data Base (Volume I) . The VOC volume is a major revision of an
earlier EPA document entitled VOC Species Data Manual . About 75 percent of
the 250 profiles in’ the’ document are new or- revised. In addition to the
original VOC profiles, this document contains VOC profiles for all SCC’s in
the National Emissions Data System (NEDS) and applicable NEDS area source
codes.
PM Data Base (Volume ii) . The PM volume is a revision of an earlier EPA
document, Receptor Model Source Composition Library . Only minor changes have
been made in data and report format in the new document. There are over 250
12
-------
PM profiles in this document, assigned for all NEDS SCC’s and many area
sources. The format of the PM profiles are similar to the VOC profile tables
except for the presentation of particle size distributions for each species.
A data base management system for use on IBMTh compatible personal
computers is also now available. This data base may contain more recent and
additional information beyond the latest published version. Example of the
species profile are shown in Tables 5 and 6.
D. Conservative Material Balances
For some combustion sources, solvent users and other evaporative loss
sources, conservative assumptions can often be made concerning the fraction of
a material input to or used by a process that is emitted. Commonly, all
makeup solvent used by certain operations is assumed to be evaporated and a
large fraction of metals in fuels is assumed to be exhausted. Hence, for many
operations, screening estimates can be generated by making reasonable assump-
tions about the amount a toxic substance in a solvent, fuel, or process feed
and then applying this to activity level information in the existing criteria
pollutant inventory or permit file. Examples of the application of conserva-
tive material balances are given in Appendix A.
E. Existing Source Data/Emission Estimates
The literature, permit and registration files, and NATICH contain source
data and/or emission estimates (and in some cases, risk assessments) for many
substances that may suffice for screening purposes. This is especially likely
for many large production (e.g., SOCMI) facilities and facilities required to
obtain a permit. For example, references 14 through 19 provide source loca-
tions, operating data and emission estimates for sources of ethylene dichlo-
ride, methylene chloride, carbon tetrachioride, trichioroethylene, chloroform,
and perchiorothylene, respectively. The EPA Air Toxics Clearinghouse stores
source and emissions data on thousands of facilities that have been submitted
by States and local agencies, as well as risk estimates that have been
computed by EPA for certain sources under the NESHAPS process. The Air Toxics
Clearinghouse also publishes a bibliography of references that may be useful
in locating data and emission estimates.
13
-------
VOC Profile Speciation Report
TABLE 5
EXAMPLE VOC PROFILE
43202
43204
43212
43220
43229
43230
43231
43232
43233
43242
43247
43248
43261
43262
43263
43274
43276
43288
43291
43295
43296
43298
43910
43911
43912
45201
45202
45203
45204
90010
TOTAL
74-84-0
74-98-6
106-97-8
109-66-0
107-83-5
96-14-0
110-54-3
142-82-5
111-65-9
287-92-3
108-08-7
110-82-7
108-87-2
96-37-7
591-76-4
565-59-3
79-29-8
1678-91-7
75-83-2
589-34-4
592-27-8
592-27-8
75-28-5
78-78-4
540-84-1
7 1-43-2
108-88-3
100-41-4
95-47-6
Ethane
Propane
N-Butane
N-Pentane
2-Methyl pentane
3-Methyl Pentane
Hexane
Heptane
Octane
Cyci opentane
2, 4-Dimethyl pentane
Cycl ohexane
Methyl cyci ohexane
Methyl cyclopentane
2-Methyl Hexane
2,3 Dimethyl Pentane
2,3 Dimethylbutane
Ethylcyclohexane
2, 2-Dimethyl butane
3-Methyl hexane
2-Methyl heptane
3-Methyl heptane
2-Me-Propane
2-Me-Bwtane
2,2, 4-Trime-Pentane
Benzene
Tol uene
Ethyl benzene
O-Xylene
M-Xylene and P-Xylene
30.07
44.09
58. 12
72.15
86.17
86.17
86. 17
100.20
114.23
70.14
100.20
84.16
98.2
84.16
100.20
100.20
86.1
112.23
86. 17
100.20
114.23
114.23
58.10
72.20
114.20
78.11
92.13
106.16
106.16
106.16
17.52
32.63
21.33
6.46
1 .48
0.33
1.68
0.68
0.68
0.38
0.05
0.68
10.74
1.04
0.24
0.11
70.17
0.09
0.05
0.32
0.23
0.10
6.44
5.56
0.32
0.05
0.09
0.02
0.03
0.12
99.69
Profile Name :
Profile Number:
Data Quality :
Crude Oil Production - Well heads (Gas Drive)
1212
C
Control Device:
Uncontrolled
Reference(s) :
96
Data Source :
Crude comes from the Oklahoma Dutcher formation which is at a
depth of 1,400 feet. Four samples were taken in stainless
steel canisters and analyzed using a GC with cryogenic sample
pre-concentration and flame ionization detection.
SCC Assignments:
31000101
SAROAD CAS
Number Name Spec_MW Percent Weight
14
-------
TABLE 6
EXAMPLE PARTICULATE MATTER
PROFILE
PM Profile Speciation Report
Profile Name Primary Copper Roaster
Profile Number 29209
Data Quality 0
Control Device Uncontrolled(1)
Reference(s) 1.18
Data Source Data taken from Reference 1 For details, consult Reference 18
SCC Assignments 30300502. 30300509
Mass Fraction Data - Size Interval (iun) (0—2 5) (0—6) (0-10)
Mass Fraction 0 970 1 000 1 000
Total Particulate Matter -
0—2.5 an 2 5-10 gm 0-10 uri
Total
Measured (2)
Species
Cas Number Mum Sym % wt unc % t unc % wt unc
% wt unc
429-90-5 13 4L 6300 0 600 2 400 0200 2 621 1 096
2 621 0000
7440-21-3 14 Si 13 500 1 200 4 800 0 500 5 293 2 300
5 293 0 000
7704-34-9 16 S 9 800 0 900 8 100 0 300 8.196 1 830
8 196 0 000
7782-50-5 17 CL 0 000 0 100 0 000 0 100 0.000 0 138
0 000 0 000
7440-09-7 19 K 1 010 0 100 0.420 0 080 0 453 0 201
0 453 0 000
7440-70-2 20 Ca 2 000 0 170 1 520 0 110 1 642 0 393
1 642 0 000
7440-32-6 22 Ii 0 260 0 050 0 160 0.040 0.166 0 083
0 166 0 000
7440-62-2 23 V 0 000 0 050 0 000 0 040 0.000 0 062
0 000 0 000
7440- 7-3 24 Cr 0 066 0 015 0 040 0 020 0 042 0 033
0 042 0 000
7439-96—5 25 Mn 0 137 0 015 0 081 0 018 0 084 0 041
0 084 0 000
7439-89-6 26 Fe 18 600 1 500 13 000 0 500 13.317 3 269
13 317 0 000
7440-02-0 28 Mi 0 110 0 050 0 000 0 080 0 006 0 094
0 006 0 000
7440-50-8 29 Cu 18.600 1 500 18 400 1 900 18 411 3.935
18 411 0 000
7440-66-6 30 Zn 1 090 0 120 1 020 0 080 1 024 0.248
1 024 0 000
7440-38-2 33 As 4 200 0 400 3 700 0 600 3 728 0 998
3 728 0 000
7782-49—2 34 Se 0 000 0 050 0 000 0 030 0 000 0 057
0 000 0 000
7726-95-6 35 Br 0 000 0 300 0 000 0 200 0 000 0 351
0 000 0 000
7440-17-7 37 Rb 0 000 0 040 0 000 0 050 0 000 0 063
0 000 0 000
7440-24—6 38 Sr 0 000 0 050 0 020 0 040 0 003 0 044
0 003 0 000
7440-65-5 39 Y 0 000 0 030 0 000 0 040 0 000 0 049
0 003 0 000
7439-98-7 42 Mo 0 160 0050 0 120 0040 0 122 0075
0 122 0000
440—22-4 47 Ag 0 000 0 200 (1 042 0 010 0 040 0 195
0 040 0 000
7440-43-9 48 Cd 0 000 0 200 0 000 0 020 0 000 0 195
0 000 0 000
7440-75—6 49 [ r I 0 000 0 200 0 000 0 020 0 000 0 195
0 300 0 000
7440-31-5 50 Sn 0 000 0 200 0 000 0 020 0 000 0 195
0 000 0 000
7440-36-0 51 Sb 0340 0 160 0076 0021 0.091 0 166
0091 0000
13494-80-9 52 Te 0 000 0 200 0 000 0 000 0 000 0 218
0 000 0 000
7439-97-5 80 Hg 0 000 0 100 0 000 0 040 0 000 0 105
0 000 0 000
7439-92—1 82 Pb 2 130 0 180 1 750 0 100 1 772 0 413
1 772 0 000
7440-69-9 83 Bi 0.300 0 100 1 060 0 030 0 074 0 112
0 074 0 000
TOTALS 78.653 55.789 57.085
57.085
assumed uncontrolled
(1) No controlled specified thus,
(2) Estimated I extrapolated
15
-------
Another resource is the Superfund Amendments and Reauthorization Act
Section 313 (SARA 313) Inventory. By July 1, 1988 many facilities were
required to complete SARA 313 inventory release forms. Working from a list of
329 compounds and groups of compounds, extensive sources will be covered.
This new information can help States and local agencies to identify affected
sources, pollutants being emitted, and estimates of total plant emissions,
etc. It will not, however, provide individual stack data, the size of
sources, or plant layouts. The industry must provide total plant estimates of
point-sources and nonpoint source (fugitive) air emissions. This information
should be useful as a screening tool to focus on potential air toxics hot
spots, or to help select what air toxics pollutants to include in the inven-
tory permit program or source registration process.
What these screening tools have in common is that they allow the agency to
identify potential emitters and make screening estimates without making time
consuming and expensive source contacts. Once the screening study is com-
plete, the detailed air toxics inventory can commence, if necessary. In some
cases, the screening study may divulge that a detailed inventory may not be
necessary, at least for some pollutants or in some areas. Or it may turn out
that selective coverage, either for specific sources, pollutants or for small
areas, may be sufficient for program planning and control purposes. This is
the point of the screening process -- to help the agency avoid expending
resources on activities that do not yield significant reductions in air toxic
emissions and risks. This allows the bulk of available agency resources to be
devoted to more detailed data collection and analysis and to focus on those
oollutants, sources, and geographic areas that pose most risk to the com-
munity.
Table 7 shows an example of possible results from an inventory screening
study where the emissions data have been modeled to estimate aggregate cancer
incidence. These data show the relative risk associated with various point
and area sources in a given area. This kind of screening exercise would
provide some basis for the agency to focus any subsequent data collection and
control effort. Of course, any screening estimates of emissions and risk
would probably be considered preliminary, to be followed by more detailed
estimates for the more important sources and pollutants.
3.3 Screening Study Error
Because crude techniques may be used in screening studies, errors are
unavoidable. Hence, it is generally advisable to be conservative in such
studies , i.e., to develop estimates that do not underestimate emissions and
associateu risks. Since the point of the screening study is to identify areas
where additional attention is needed, it is better to overestimate emissions
than to underestimate emissions. Presumably, overestimates of emissions will
be put into proper perspective later on in the inventory process, but under-
estimates could conceivably lead to false conclusions that may not be cor-
rected. One way to deal with potential errors is to view the screening
process as an iterative one. That is, instead of assuming that the focus for
16
-------
the entire program can be correctly ascertained from a single screening, the
agency might opt to keep the process open for a second iteration to fine tune
the results.
TABLE 7
PERCENT OF CANCER RISK ASSOCIATED WITH POINT AND AREA SOURCES’
Point Sources % of Total Incidenceb
Chemical production/usage 10
Metal Manufacturing 8
Petroleum Refining 5
Rubber Production 5
Utilities 3
Publicly Owned Treatment Works (POTWs) 2
All Others
Total Point Source 49
Area Sources
Road Vehicles 22
Solvent Usage 10
Gasol me Marketing 9
Waste Oil Burning 8
Woodsmoke (stoves/fireplaces) 0.5
All Other
Total Area Source 51
In 35 counties, from Reference 20.
Because of the uncertainties in the incidence calculations used to derive
these estimates, the numbers should be regarded as rough indicators only.
They will not necessarily apply to any given geographical area.
3.4 Review of Screening Study Results
it may prove helpful at the end of the screening study (or later, after
the detailed inventory) to make the results available for outside review, at
least in summary form. This review could involve having various noninventory
personnel review the results to see if the aata patterns and conclusions
appear reasonable.
Reviewers should include persons that are sufficiently detached from the
inventory process so they will not be reluctant to point out possible errors
and shortcomings. Equally important, these reviewers should include potential
users of the air toxics inventory data such as modelers and planners who will
17
-------
need the data to assist in control strategy development. And, at some point,
comments from outside interests may be desirable, including industry, health
specialists and environmental groups. Often, bringing differing and diverse
viewpoints into a process can be quite helpful in spotting anomalies, mis-
representations, or shortcomings that may not be detected by those directly
involved in inventory effort. This review process will heighten the visi-
bility and credibility of the air toxics inventory, assisting any subsequent
data collection efforts and improving the likelihood of acceptance of any
control program based on these data.
3.5 Use of Screening Tools in Evaluating Permit Applications
The emission inventory screening tools presented in this report may be
useful to agency permitting staff in evaluating permit applications (either
permit renewals or permits for new or modified sources) for air toxics. The
typical permit review period (e.g., 90 days) does not allow for detailed
research; therefore, these tools can be of great importance. Whereas the
screening tools may not always provide the level of detail or accuracy
necessary for complete evaluation of permit applications with respect to air
toxic emissions, they can provide information useful in determining which
applicants need to provide more detailed information. The results of the per-
mit process can feed into the air toxics emission inventory, thus providing an
improving foundation of information useful in future permit reviews and in
developing other aspects of regulatory programs.
The first step in reviewing oermit applications is to determine if all
toxic pollutants of concern potentially emitted by the proposed source are
accounted for and if the stated emission levels are reasonably accurate. Such
information may be needed initially to determine the applicability of the
permitting requirements. Air toxics emissions screening tools may be useful
in determining whether the source potentially emits specific pollutants
subject to regulation under the Act or other toxic compounds of concern to
State or local permitting agencies. For example, the permitting staff could
use the source category/pollutant indices, emission factors or speciation
factors to determine if all pollutants of concern have been included in the
permit app] ication.
Where emission estimates are relatively uncertain (e.g., for new pro-
cesses), agencies with operative programs sometimes require emissions tests or
other means of verification.
18
-------
POLLUTANT COVERAGE
The major difference between air toxics inventories and criteria pollutant
inventories is in substance coverage. Since there are literally thousands of
potentially toxic substances as compared to only a handful of criteria
pollutants, the compilation effort can become formidable since data collection
and data handling resource needs increase quickly as more and more substances
are inventoried. In this regard, limiting the number of substances included
in the detailed inventory to those causing most of the problem may be one of
the most valuable outcomes of the screening study.
4.1 eOpen ended Vs. Oirected Aooroaches
There are two fundamentally different approaches that can be taken when
soliciting air toxics emissions data: The “open-ended” and “directed”
approaches. In the open-ended approach, the agency asks each source to submit
data on whatever air toxics it emits, either without specifying any compounds
of interest or by specifying a very broad list containing perhaps hundreds of
compounds. The open-ended approach is commonly used in permit-based inven-
tories where point source applicants are asked to submit information on what-
ever compounds they expect to emit. By contrast, in the directed approach,
the agency may target a fairly limited number of compounds it wants to focus
on.
A potential advantage to the open-ended approach is that a single ‘one
shot” data collection effort can be employed to get information on most, if
not all , toxics being emitted in an area. In theory, this approach may
provide a better overall perspective of an area’s air toxics problem than will
the direct approach because the agency need not have predetermined what
substances are most important.
The disadvantages to the open-ended approach are several:
-Without some focus on a limited number of compounds, questionnaire
respondents may overlook and not report the use and/or emissions of
many air toxics. This may particularly be the case if such com-
pounds are present at low levels or as contaminants in a facility’s
product, feed stock materials, etc., or if trade name solvents are
used and the respondent does not know what comprises them. tn this
latter regard, many chemical synonyms and trade names may be
reported that the agency cannot readily interpret. For example, a
source may report it uses Dutch liquid in its process, which is
an alternative name for ethylene dichioride (EDC). Or, hypotheti-
cally, a source may report that it uses a material named “Super
Solvent No. 209” which may contain a large number of substances,
including toluene, xylenes, etc. Unless the State or local agency
can associate these synonyms and trade names with specific
compounds of interest in its air toxics inventory, it may find
itself unable to utilize much of the data submitted by various
19
-------
respondents. (Chemical name synonyms for commonly inventoried
substances are presented in Appendix C. The use of Material Safety
Data Sheets and other techniques for determining compositions of
trade name products is discussed in Chapter 8.)
-The inventory agency may receive much more information than it
needs or is prepared to deal with, especially for compounds not
emitted at problem levels.
-Unless sufficient quality checks and followup contacts can be made,
the response rates may be low and the data suspect.
-Industry may resist reporting on hundreds of substances without
regard for the level at which each substance is emitted. One way
to counter this objection is to develop a set of conditions that
allow certain source/emission level/pollutant combinations to be
exempted from detailed reporting. For example, an agency could
exclude from reporting any emissions resulting solely from the
combustion of unadulterated fuels. Or it might exclude from
reporting any emissions of certain compounds discharged below
certain rates. Substances toxicity may also be factored into the
agency’s exemption criteria. (This is discussed more in Chapter
6.)
In the directed approach, an agency may focus on a specific set of
compounds it is interested in based on recommendations from a
health group or possibly as a result of an emissions screening
study. Advantages of the directed approach are:
-It generally allows the agency to focus its data collection effort
on fewer sources, targeting those which it believes are probable
emitters of the particular substances on its list.
-The amount of data obtained may be much more manageable.
Disadvantages are:
-Additional contacts to the same sources may become necessary if the
initial list of pollutants is expanded by the agency. (This may be
acceptable if the additional compounds are phased in over the
course of the normal update cycle.)
-The agency must either have some predetermined idea of what
compounds are important at the outset of the inventory effort or
else it must arbitrarily use a designated list. Unless the agency
has performed a preliminary screening survey of some sort to
determine the major compounds of potential concern, it runs the
risk of wasting resources on relatively unimportant compounds while
possibly neglecting some compounds actually posing considerable
hazard.
2O
-------
4.2 State and Local Agency Lists
Recent surveys of existing air toxics programs indicate that many
agencies use a specific list of substances whereas others accept open-ended
submissions and a few have hybrid programs -- apparently targeting a specific
list of substances but accepting data submitted on any substance.
Several common rationale for selecting a group of compounds are listed
below. All of these rationale are not mutually exclusive and more than one
rationale may be the basis for selection in some agencies.
-Substance toxicity (most common rationale)
-Emissions potential
-Substance toxicity and emissions potential
-Use an EPA list
e.g. -NESHAPS and NESHAPS candidate substances
-Acute Hazard List
-Adopt list used by another State or local agency
-List mandated to agency (e.g., by a legislature)
The actual compounds selected for inclusion in agency programs vary
considerably, and space precludes presenting or summarizing all of the lists
or compounds. However, a number of compounds are commonly included in many
agency lists. Table 8 shows 64 compounds that are common to at least 4
agency’s lists. Reference 21 presents methods for pollutant selection and
prioritization. used by various agencies.
4.3 Core List of Compounds for Possible Consideration
EPA is not recommending a specific list of compounds for inclusion in a
given area’s air toxics inventory. Unlike criteria pollutants, many air
toxics are riot emitted in every locale. Some air toxics, of course, will be
fairly widespread, such as those associated with general solvent use and area
sources (e.g., benzene, toluene, xylenes, formaldehyde, perchioroethylene,
methyl chloroform, inethylene chloride and trichioroethylene). These substan-
ces would be emitted in most urban areas. However, many compounds that are
primarily associated with point sources (e.g., epichiorohydrin, acrylonitrile,
phosgene, etc.), might only be emitted by a few large producers or users in
the entire U.S. and may riot be emitted in significant levels at all in most
areas.
As starting point, in the absence of locale-specific screening results,
an agency may consider inventorying the compounds in Table 9 for which EPA has
set NESHAPS under Section 112 of the Clean Air Act, or has listed or intends
to list in the near future, and/or that were identified as important contri-
butors to aggregate cancer incidence in EPA’s “Six-Month Study.”” Most of
these compounds will be emitted in industrialized urban areas and, because
these compounds are important from a national risk standpoint, a considerable
amount of work has been done or is underway to analyze and characterize them.
21
-------
Hence, these compounds will generally have better data bases associated with
them than are available for many other noncriteria substances. The avail-
ability of these data will greatly assist in inventory preparation.
TABLE 8
SUBSTANCES COt SIDERED BY FOUR OR MORE AGENCIES
Acetaldehyde 3,3-Dichi orobenzidine
Acrolein Dimethyl Sulfate
Acrylonitrile 1,4-Dioxane
Allyl Chloride Dioxins
Arsenic Epichiorohydrin
Asbestos Ethylene Dibromide
Benzene Ethylene Dichioride
Benzidine Ethylene Oxide
Benzo(a)pyrene Formal dehyde
Benzyl Chloride Heptachior
Beryllium Hexachlorocyclopentadiene
Biz(chloromethyl )ether Hydrazine
1,3-Butadiene Hydrogen Sulfide
Cadmium Lead
Carbon Tetrachioride Lindane
Chlordane Maleic Anhydride
Chi orobenzene Manganese
Chloroform Mercury
Chioroprene Methyl Bromide
Chromium Methyl Chloride
Cresol Methyl Chloroform
1 ,4-Oichlorobenzene Methylene Chloride
Beta-Naphthyl amine Polycyci ic Aromatic Hydrocarbons
(PAH)
Nickel Propylene Oxide
Nitrobenzene Radionuci ides
N-nitrosodimethyl amine Styrene
Nitrosomorphol me Tetrahydrofuran
Parathion Toluene
Perchi oroethyl ene Trichi oroethyl ene
Phenol and Chlorinated Phenols Vinyl Chloride
Phosgene Vinyl idene Chloride
Polychlorinated Biphenyis (PCB) Xylene
‘Also defined as “PlC” (or products of incomplete combustion) in EPA’s “Six-
Month Study.” B(a)P is used as a surrogate for calculating risk and
aggregate cancer incidence from PlC exposure in this study.
22
-------
TABLE 9
CORE LIST OF COMPOUNDS FOR CONSIDERATION IN
AN AIR TOXICS EMISSION INVENTORY’
Arsenic Ethylene Oxide
Asbestos Formaldehyde
Benzene Gasol me Vapors
Beryllium Mercury
1,3 Butadine Methylene Chloride
Cadmium Perchioroethylene
Carbon Tetrachioride Polycyclic Organic Matter°
Chloroform Radionuci ides
Chromium Trichl oroethyl ene
Coke Oven Emissions Vinyl Chloride
Ethylene Dibromide Vinyl idene Chloride
Ethylene Dichoride
‘As mentioned in the text, these compounds were selected because EPA has (1)
set NESHAPS for them under Section 112 of the Clean Air Act, or has listed or
intends to list them, or (2) identified them as important contributors to
aggregate cancer incidence in its Six-Month Study.”
‘Also defined as “P lC” or products of incomplete combustion, in EPA’s “Six-
Month Study.” B(a)P is used as a surrogate for calculating risk and aggre-
gate cancer incidence from PlC exposure in this study.
There are compounds which will be locally important in some areas which
will not pose significant national risk. Acrylonitrile, 4,4-methylene diani-
line and propylene oxide are examples of substances which may account for
significant exposures in the proximity of a few larger chemical manufacturing
plants, but which do not account for significant aggregate cancer incidence in
the U.S.” Hence, the agency should not automatically exclude compounds from
consideration because they are not on the above (or any or other) list. Since
EPA continues to evaluate and list compounds for NESHAPS under Section 112 of
the Clean Air Act, the agency may wish to include any newly listed compounds
in its inventory, as well.
23
-------
CHAPTER 5
SOURCE CATEGORY COVERAGE
For the most part, an agency must deal with the same types of sources and
source categories in both criteria and air toxics inventories. The funda-
mental reason for this inherent similarity is that most air toxics are a
subset of either particulate matter (PM) or volatile organic compounds (VOC);
most toxic metals and heavy organics are emitted in the PM fraction while most
toxic light organics are associated with the VOC fraction. For this reason, a
complete criteria pollutant inventory will serve as a good starting point for
the air toxics inventory.
Although most sources of air toxics should already be included in complete
criteria pollutant inventories, every source will not emit toxic compounds.
Table 10 presents a simplified source characterization scheme that may prove
helpful in providing a broad perspective of what kinds of sources emit what
air toxics. Table 10 summarizes available nationwide emission totals for the
substances evaluated in EPA’s “Six-Month Study.” The sources in Table 10 are
lumped by similarity of operations, physical characteristics, mode of release,
etc., into the following generic categories that account for a large fraction
of air toxic emissions.
1. SOCMI (Synthetic organic chemical manufacturing industry)
2.. Solvent evaporation (surface coating, degreasing, printing, etc.)
3. Stationary fuel combustion (boilers, internal combustion engines)
4. Metal processing (mining, refining, production of metals)
5. Mobile sources (automobiles, boats, planes, etc.)
6. Waste oil combustion
7. Other (incineration, in process fuel use, etc.)
Several characterizations can be made from Table 10 that may prove useful
in understanding which compounds are generally associated with what sources:
-Roughly half of the 86 substances are largely emitted from SOCMI
facilities, i.e., producers of synthetic organic chemicals; users of
these chemicals as feed stocks; or facilities that produce these
chemicals indirectly as contaminants or byproducts. Major emitting
points within SOCMI include equipment leaks (from pumps, valves,
compressors, etc.), storage and handling losses (loading, unloading
and breathing losses from railroad cars, storage tanks, etc.)
process vents (distillation towers and reactor processes), and
secondary ad indirect sources (e.g., wastewater treatment).
-Solvent use accounts for the major release of a number of sub
stances including cyclohexane, methyl ethyl ketone, chlorobenzene,
nitrobenzene, perchioroethylene, trichioroethylene, methyl chioro-
form, cresols, acrolein, methylene chloride, and xylenes. Solvent
loss occurs from evaporation during surface coating, degreasing, dry
cleaning, printing, etc. Minor point and area sources are sig-
nificant contributors.
24
-------
-Metal processing includes emissions from mining, refining and the
production of alloy and metal products. Emissions of zinc, man-
ganese, copper, cadmium, nickel and beryllium are all significant.
Although not explicitly identified with this category, chrome
plating is a very important source of chromium emissions.
-Mobile sources are major emitters of benzene, toluene, isomers of
xylene, POM and formaldehyde. Large percentages of all benzene,
toluene, POM, and xylene emanate from mobile sources, primarily from
tailpipe emissions and evaporation from gasoline powered vehicles.
-Stationary fuel combustion is the largest emitter of POM and several
metals (beryllium, chromium and nickel) and an important source of
chlorine, cadmium, copper, formaldehyde and manganese. Generally,
smaller sources such as woodstoves will be the most important
emitters of POM.
-Except for waste oil combustion, so-called “nontraditional sources”
are not covered in Table 10. Nontraditional sources are those that
have not traditionally been included in criteria pollutant inven-
tories, but which may be significant emitters of air toxics.
Potentially important nontraditional sources are listed below:
‘Publicly-owned Treatment Works (POTWs), i.e., municipal sewage
treatment plants.
‘Treatment, Storage and Disposal Facilities (TSDFs) with handle
hazardous waste in the following kinds of operations:
-landfills and land treatment
-surface impoundments and treatment tanks
-waste piles
-drum handi ing
-unloading and storage
-recyci ing
-deep well injection
- incineration
‘Waste oil combustion
Wood combustion (stoves, fireplaces)
Drinking water aeration
Cooling towers
Hospital and medical supply sterilizers
-Methyl ene chloride and methyl chloroform (1,1,1-trichloroethane) are
emitted in important quantities from solvent use, aerosols, etc.,
but may not be accounted for in VOC inventories that have been
adjusted to remove non-photochemically reactive compounds.
It must be emphasized that Table 10 is based on fragmented and sometimes
old, inconsistent data, since reliable air toxics emission estimates are only
25
-------
slowly forthcoming. Hence, the conclusions drawn therefrom should not be
considered absolute. Moreover, since the underlying data in Table 10 are
representative of the entire nation, the proportions will not necessarily
apply to any particular region or urban area -- especially for point sources.
Nevertheless, the broad perspective offered by this summary may offer some
help in the screening process in understanding and characterizing the general
distribution of air toxics emissions.
26
-------
TABLE 10
NAFIONWIDE AIR roxic EMISSIONS IIY BROAD SOURCE CATEGORY 22
SUBS IAI4CE
lIRCENT
OF EMISSIONS
Bi
lAth
SOURCE CATEGORY
TOTAL
(1415510115
a b
CONFIDENCE
SGt’dU iI
STATIONARY
FUEL
1 411* 1.
WASTE
011
50C14 1 USE
COlIOUST ION
PROCESSING
MOBILE
CC USTI(14 OTHER SOURCE OF OTHER (KKG/vR)
Ac t .dd hyde 22.0 36 0 42.0 1k 1 .702 C
Acetic ic id 35.8 64.2 72 O46 £
Acelonitrile 100.0 591 C
Acro lein 41.8 50.2 41.5 C
kryIa Ude 91.4 2.6 3.8 B
kry ion itr$Ie 100.0
Allyl chloride 100.0 408 B
99.9 0.1 TR 110,109 C
8 arIum carbonata 100.0 160 C
Ben eiie 6.0 10.2 1k 23.8 coke ovens, 118,145 A
gas marketing
Benzyl chlorIde 99.0 38 A
Berylitua 83.1 3.8 1.4 0.8 10.4 coke ovens 163 C
1,3-Outadlene 99.8 0.2 fungIcide 4,137 C
Cadelue 10.0 23.1 65.6 0.6 475 B
Carbon disulflde 99.9 55,549 C
Carbunyl sulfide 100.0 23.423 C
Carbon tetrachloride 15.6 4.2 80.2 fu 1gant, lab use, 21,402 C
storage and
marketing
ChIo,ine 98.2 0.2 glass manufacturIng. 615.972 C
wood rocess4ng,
brick kilos
Cliluroacetic acid 100.0 24.9 C
Chlo,obenzene .8 99.2 19,720 C
0 NoLe that emissions do nut represent a particular yea:’. Note also that these data are based on fragmented and
soiii times old, inconsistent data, since reliable air toxics emission estimates are only slowly forthcomin9.
The proportions In this table will not necessarily apply to any particular region or urban area —- especially
for point sources.
bA qualitative ranking where “A” represents best agreement among sources and “C” represents poor agreement or
perhaps only one source for the estimate.
-------
TABLE 10 (continued)
NATIONWIDE AIR TOXIC LMLSSIONS BY BROAD SOURCE CATEGORY 22
SLJBSTANCL
PLKCLNI OF 0415510115 BV LACK
SOURCE CATEGORY
WASTE TOTAL
OIL EMISSIONS
CO 1JSTION OTHER SOURCE OF OTHER (nG/vR)
COIIFWENCE
SCORE
SOLVENT
SOCHI USE
SIAIIOI IAIIY
FULL 11(1*1.
C04 USTI0N PROCESSING
K 8I1L
Cliloroethane 100.0 4 .541 C
Chloroform 11. 7 5 41.1 chlorination, wastewatar 6.9140 C
Chioroprene 100.0 838 C
ChromIum 89.6 0.2 1.4 8.9 cement productIon, 2,819 C
Co’e oven ealss lons 100.0 coke ovens NA C
Copper 24.2 75.8 10,425 C
u-Cresol 20.9 60.8 9.1 9.2 disinfectant, cleaning 2.047 8
cor ound
m-Cresol 11.1 68.3 10.3 dIsinfectant, cleaning 4,980
COOQOUnd
9.9 69.3 10.4 10.4 disinfectant, cleanIng 4,146 8
c xi ourid
Cumene 100.0 879 C
Cyc1otie ane 100.0 428 C
Dibenzofuran 100.0 co ust$on and pyrolysls NA C
of chlorinated
conpounds
I,2-Oibro.xieftane 100.0 TM 885 C
p-D lch lorobenzene 0.8 99.2 space deodorant, 22,682 8
th control
1.2-Dichloroethylene 100.0 46.9 C
Ulmethyl nitrosamlne 93.4 6.6 98 8
Diocty lphthalate 100.0 hA C
Ipichiorohydrin 100.0 217 B
Ethyl benzene 48.1 51.9 1.04 C
Ethylene 98.6 TM 0.3 artifIcial ripening 209,524 C
of fruits and
vegetables
-------
TABLE 10 (continued)
NATIONWIDE AIR TOXIC EMISSIONS BY BROAD SOURCE CATEGORY 22
SUBSTANCE
PERCENt OF ENISSIONS
8!
EACH
SOURCE CATEGORY
SOURCE OF OTHER
TOTAL
EMISSIONS
(KKGJYR)
CONFIDENCE
SCORE
SOLVENT
50CM! 11SF
STA IIOHAhiY
FUEL NITAL
CO UST10N PROCESSING
MOBILE
WASTE
OIL
CU UST1ON OTHER
Ethylene d lcbloride 90.0 2 I) 2.0 fumIgant. pestIcide. 64,800 C
color film
Ethylene glycol 100.0 982 C
Ethylene glycol 100.0 14.0 C
noethyi ether
Ethylene oxide 313 t i ant , eteri)ant 5,580
Ethyl ester 100.0 203 C
acrylic acid
hnaaldebyde 49.9 17.5 32.6 510,000 C
iie . chlorocyclo- 100.0 8
pentadiene
hksahydro-2b- 100.0 4.221 C
azepln-I-one
Isopropyl alcohol 100.0 1,310 C
4 4-IsopropylI- 100.0 HA C
‘Jenediphenol
H..*leIc anhydrlde 100.0 1,504 C
K in janese 1.8 15.8 76.2 6.2 battery productIon, 16,356 C
coke ovens
iielam lne 100.0 NA C
I ethanol 98.6 1.4 23,582 C
Kctliyl chloroform 0.1 78.6 21.3 adhesIves, aerosols, 212.174 A
paint
t4ethyl chlorIde 100.0 23,982 8
1jctiiylene chloride 56.1 43.6 aerosol, foam blowIng 200,409 C
H Liiyi ethyl ketone 100.0 1k 5.317 C
4,4-Nethylene- 100.0 26.1 C
dianlilne
&tiiyl methacrylate 100.0 728 C
-------
TABLE 10 (coi’itinued)
NATIONWIDE MR TOXIC EMISSIONS BY BROAI) SOURCE CATEGORY 22
SU IIS IMCE
PERCENT OF EMISSIONS
bY
EACH
SOuRCE
CMIGORY
TOTAL
ENlSSI0Ii
COME WINCE
SQu WkNI
STATIONARy
Foil.
METAL
WASTE
OIL
50CM! USE
COP180STION PROCESSING
NUbILE
COuIMISJIOII OTHER SOURCE OF OTHER ( G/YR)
SCORE
Itapihalene 100.0 119.4
Nickel 84.1 7.3 8.0 0.) 0.1 coke ovens 10,182 8
N ltrobenzene 4.0 96.0 3,032 B
Nitrosomorphollne 0.2 99.8 corrosion lnhthitor, 4,558 8
polishes, optical
brightener
Peutach lorophenol 11.9 88.1 fungIcide, home and 101 B
garden applications,
preservative
PelLh loroethylene 6.4 93.6 JR 144,083 A
P i nol 100.0 T B 2,406 C
Phosyene 100.0 91 8
I’hthalic anhydride 100.0 1,059 C
Polychlorinated 100.0 transfort rs, waste ( IA C
biphen3rls Incineration
Polycyclic organic 51.8 42.6 5.6 waste Incineration, NA C
matter charcoal produUion .
coke ovens
Propene 100.0 TN 31.540 C
Propylene oxIde 100.0 608 C
Styrene 100.0 5,625 C
ier phtholic acid 100.0 NA C
Tit nlum dioxide 100.0 pt nent use 4.218 C
1,2-loluene 100.0 51.0 C
di Isocyanate
luluene 31.1 0.1 61.6 1.0 coke ovens 1,0(3,379 C
Irichioroethylene 0.1 9 1.4 TB 6.5 adhesives, paInt 85,917 A
Vinyl acetate 100.0 682 C
-------
TABLE 10 (continued)
tIATIONWIDE AIR TOXIC EMISSIONS I Y UROAD SOURCE CATEGORY 22
PIkC(N1 OF EMISSIONS 01 EACH SOIJI1CE CATEG0
STATIOMARI — WASTE
FUEL META l. OiL
C0I USTION PRIXESSING MOdULE C(JIBUSTU$
SUBSIANCE
TOTM.
SOCI4I
SOLVENT
USE
OTHER
SOURCE
OF
OTHER
(MISSIONS
(K G/vR)
CONFIDENCE
SCORE
Vinyl idene
chlorIde
100.0
61.0
C
o-Ky iene
1.8
52.3
46.9
122,044
8
w4y lene
0.8
54.0
45.2
206,152
8
p-Xy$ene
3.8
48.6
47.1
108.759
8
Zinc
3.1
96.3
55.512
C
Zinc oxide
I.?
983
p1 ent use
1.208
C
-------
CHAPTER 6
POINT/AREA SOURCE DISTINCTIONS
Mast emission inventories distinguish between point sources and area
sources. Point sources are those sources large enough to permit and/or to
keep individual records on. Area sources are sources too small and/or
numerous to permit or to warrant individual recordkeeping. (Some inventories
also distinguish between major point and minor point sources.) The rationale
for making this distinction is to minimize the reporting burden on smaller
sources and to reduce the volume of point source inventory records.
6.1 Making the Point/Area Source Distinction
The point/area source distinction can be made in several ways:
A. By defining all sources within particular source categories as either
point or area sources. For example, all gasoline service stations, dry-
cleaners, etc., might be handled as area sources and exempted from reporting,
and, conversely, all hospital sterilizers and chrome platers may be required
to submit emissions data as point sources.
B. By establishing emission cutoff or “de minimis” levels. For example,
sources might only be required to report emissions greater than “x” tons per
year.
Various de minimis levels and exemption criteria are presently being
employed by States and local agencies in their air toxics inventories. Some
of these are listed in Table 11 as examples. At least one agency has defined
cutoff levels for each toxic substance covered in its inventory.
6.2 Factors Affecting the Point/Area Source Distinction
The agency needs to make several decisions regarding the point/area source
distinction. For example, the point/area source distinctions for criteria
pollutants need to be evaluated carefully if the agency plans to base the air
toxics inventory on the criteria pollutant inventory. Small industrial
sources such as dry cleaners, degreasers, chrome platers, paint coaters, waste
oil boilers, etc., are often considered collectively as area sources in
criteria pollutant inventories since their emissions of PM or VOC may not
exceed typical cutoff levels of 5-100 tons per year. In some cases, these
smaller sources may even be excluded altogether from the ex1stinq inventory
and not even covered as area sources. If the agency decides that it needs to
permit and keep individual records on certain of these smaller sources to
afford a proper evaluation of emissions and risk, additional contacts may be
necessary to obtain the necessary source and/or emissions data. Conceivably,
the agency might even go to a “zero ’ emission cutoff level for a few important
sources, thereby designating fl sources within particular categories (e.g.,
all chrome platers or all benzene storage tanks) as point sources regardless
of size, throughput or emissions.
32
-------
The principal advantage of considering sources individually rather than
collectively is that it results in more accurate estimates of emissions,
ambient air concentrations and risks. It also improves the accuracy of any
detailed control strategy projections. A negative aspect of considering many
small sources as point sources is that more resources are required both by the
agency and the industrial community to collect and digest the greater volume
of point source data. In this regard, industry may strongly resist any
requirements to acquire information on very small sources without some
reporting exemptions based on emission rates and/or substance toxicities. Of
course, if an agency is focusing on all sources within relatively small
geographic areas, the added expense of considering small sources individually
may be necessary.
One potential problem should be mentioned with using cutoff or de minimis
levels in air toxics inventories, a problem that exists in criteria pollutant
inventories, as well. If the cutoff level chosen allows a significant number
of sources to escape consideration as point sources, these sources will have
to be considered collectively as area sources if the agency is seeking a
comprehensive emission inventory. For example, if the agency exempts coin-
operated (self-service) dry cleaners and small, cold-cleaner degreasers from
reporting, a significant amount of perchloroethylene, methylene chloride and
methyl chloroform will not be inventoried unless these small emitters are
accounted for in the area source totals. (Procedures for accounting for
sources below cutoff levels as area sources are discussed in Reference 2.)
TABLE 11
EXAMPLE CUTOFF LEVELS AND EXEMPTED SOURCES
Example Cutoff Levels
°None
TPY of any toxic substance
TPY of any regulated pollutant
0.25 lb emission of any toxic substance per 8 hr time period
Cutoff level for each inventoried compound (as function of TLV)
Examoles of Exempted Sources
Boilers <100 MMBtu/hr exempted
•Boilers firing wood or unadulterate fossil fuels
Certain small sources exempted (e.g., dry cleaners, service stations)
33
-------
CHAPTER 7
GEOGRAPHIC AREA OF COVERAGE
An important consideration during the screening process regards how big a
geographic area should be covered. Typically, criteria pollutant inventories
cover fairly board areas such as entire states or nonattainment areas. This
extent of coverage has been necessary since criteria pollutants are ubiquitous
and emitted by a diverse collection of point and area sources.
There are no well defined boundaries for air toxics inventories such as
exist for nonattainment areas. Because inventorying a multitude of toxic
substances can be expensive and time consuming, the relevant question is
whether there are justifiable rationale for excluding certain areas from
coverage in order to save resources. The answer to this question may be based
on the underlying approach the agency has taken in its air toxics control
program.
The most common approach in air toxics programs is to model ambient air
levels around new and/or existing point sources, using source emission
estimates, and then impose controls based on how much the modeled levels
exceed some acceptable ambient standards or guidelines. Since the nature of
this type of program is to focus solely on major point sources wherever they
occur -- often regardless of the existence of other nearby sources - the
concept of excluding some areas from concern is really not germane as the
entire State (or local jurisdiction) becomes the area of coverage. Thus,
extent of coverage may not be an issue in this type of program.
In programs that are (1) defining relative emission strengths of small,
non-permitted point sources and area sources, or (2) starting to consider the
additive risks from these smaller sources along with major, permitted sources,
the concept of limiting geographic coverage has more relevance. Several
studies 1123 have concluded that small sources are major contributors to
aggregate incidence in urban areas; hence, this focus seems appropriate.
However, due to the extra burden of compiling point source data on many
smaller sources and developing area source emission estimates at an appro-
priate level of spatial resolution, it makes sense to try to limit the area of
coverage to the extent justifiable.
This matter is an important one to consider during the planning stages of
inventory development and one that a screening study can address. Those
agencies that plan to compile more than a crude, aggregate inventory of small
point and area sources -- especially those agencies concerned with additive
risk assessments and control strategy evaluations based on risk management
techniques -- should predetermine where to focus their inventory effort.
Presumably, such emphasis would be placed on industrial and/or urban areas, or
high risk areas therein, where the screening study indicated high emission
concentrations and/or high ambient air levels along with elevated risks.
34
-------
CHAPTER 8
DETAILED DATA COLLECTION AND EMISSION ESTIMATION PROCEDURES
Chapter 3 discussed the concept of the screening inventory as a means of
locating potential emitters of air toxics and developing screening estimates
of emissions. The purpose of the screening study is to focus subsequent
agency activities and resources on those sources, pollutants and geographic
areas causing most risk. At some point in the air toxics program, the
screening inventory will probably need to be refined before any control pro-
gram is instituted. During this follow-up effort, more detailed and credible
source data and emissions estimates will be needed, probably requiring some
direct source contacts to provide this information. Various procedures are
set forth in this report to help those agencies going beyond the screening
inventory. In this regard, some of the approaches for compiling the detailed
inventory will be the same as or extensions of those used for screening. For
example, the location of sources will largely be the same, using the basic
tools outlined in Chapter 3. The major differences are in compiling detailed
source data and making emission estimates.
8.1 locating Sources
Since most sources of air toxics substances should already be in the
criteria pollutant inventory, these sources should not have to be located
again as if the agency were starting from scratch. Instead, the agency needs
only to associate emissions of specific air toxic compounds of interest with
existing sources in the existing inventory. As mentioned before, certain
“nontraditional” sources may not be included and certain sources may be
excluded because their VOC were considered photochemically nonreactive. Two
important nonreactive VOC of particular interest are methyl chloroform (1,1,1
trichioromethane) and methylene chloride.
Many of the screening tools outlined in Chapter 3 will be appropriate for
use in locating sources in the detailed inventory and, thus, are not repeated
here. Since none of these screening tools requires source contacts, some
errors will occur in identifying actual emitters of particular toxic sub-
stances. One additional locating tool that some agencies employ that involves
making source contacts is the production/use survey. Some agencies have
utilized production/use questionnaires which basically just ask sources to
identify whether each substance is purchased, used or produced. From this,
the agency may decide which facilities should receive a more detailed,
followup questionnaire.
In general, the emphasis in the detailed inventory will not be on locating
sources, since this has presumably been done in the screening process. This
step will be necessary, however, where the agency has foregone the screening
process.
35
-------
8.2 Data CpUection
Before sending out questionnaires or making plant visits, the agency
should carefully evaluate the need to contact all sources. Based on the
results of the screening study, certain sources may have such low emissions or
pose such little risk that further information would not prove useful. Or, in
some cases, the screening estimate developed by the agency based on material
balances or emission factors may prove as accurate as any estimates likely to
be provided by certain industrial segments. The latter may be the case for
small solvent users and combustion sources. For those sources where the
agency believes it needs more information, data collection may proceed using
questionnaires, plant visits, etc., just as with any criteria pollutant. A
great deal of data can also be obtained through the new source review and
permit renewal process.
A. Use of questionnaires
As mentioned previously, a fundamental decision that has to be made during
planning is whether the questionnaires should be open-ended or directed toward
a limited number of compounds. This decision will affect both the design of
the questionnaires and the nature of data collected. In general, if a
directed approach is taken, and the number of compounds is kept to a minimum,
the resulting number of source contacts that has to be made can be greatly
reduced. Moreover, the quality of the data received may also improve --
especially if the questionnaires can be tailored to the principal source
categories of concern. Data collected from industry-specific questionnaires
will not only be more accurate for the respondent sources themselves, but will
also allow the agency to develop emission factors and emission-to-employee
ratios to apply to other, perhaps nonresponding, sources in the same source
categories. Of course, tailoring questionnaires to specific industries can be
costly and time consuming unless another agency’s questionnaires can be
adopted. This time and effort may be recovered since less followup will often
be required.
Conversely, if one uses the open-ended approach and is not targeting
information on specific sources or a limited group of compounds, a much larger
number of contacts will be- necessary using questionnaires that are more
generic in design. This approach may require less time and effort for
questionnaire design, but the resulting responses may be commensurately less
detailed and possibly not as useful. For example, if an agency is conducting
an open-ended survey of air toxics associated with solvent use, a fairly
general survey form will have to be sent out to a large number of coaters,
degreasers, printers, drycleaners, etc., with a voluminous and varied response
expected. As mentioned previously, a particular danger in this type of survey
is that many (especially smaller) sources may simply report trade names of
solvents, coatings, etc., or may not report anything at all if they don’t know
what compounds are present in the materials they use. Either possibility can
lead to under reporting and an underestimating of substance use and emissions
by the sources and the inventory agency. Various kinds of questionnaires have
been developed to collect air toxics emissions data, with some being quite
detailed and others being very general. EPA has developed a document,
-------
Compilation Of Air Toxics Emission Inventory Questionnaires (EPA-450/4-88 -
QQp).!4 This document provides a resource for agencies interested in develop-
ing air toxics emission inventory mail out questionnaires. The questionnaires
represent various formats dependent on planned use of data and the agency’s
air toxic program.
B. Interpreting and Augmenting Questionnaire Responses
Whatever questionnaire is used, some responses will be incomplete and/or
suspect. To this end, several other avenues may be explored to corroborate,
complement or help interpret the questionnaire responses. For example, if the
questionnaire respondent does not know the makeup of its company’s emissions,
it may pay to contact his purchasing department to see if any composition data
can be inferred from purchasing specifications. Or, if Material Safety Data
Sheets are available, they might be solicited in the data collection effort as
they list specific compounds and weight percentages of hazardous ingredients
present in coatings, paints, preservatives, thinners, resins, etc. Figure 1
shows an example Material Safety Data Sheet. This kind of information should
be more available with time as a result of State and Federal right-to-know
laws which require that information be made available on the composition of
products being used in the work place. Specifically, as of May 25, 1986,
OSHA’s Hazard Communication, or right-to-know, standard requires users,
distributors, repackagers, producers, importers, etc., (within SICs 20-39) to
supply all known hazard data, including species composition data and some
toxicity data, on Material Safety Data Sheets. ’
State OSHAs can also be a valuable source of information even though the
Federal Hazard Communication standard supersedes States’ right-to-know laws.
For example, some State agencies may have collected work place ambient air
samples that could suggest what air toxics are being emitted. Also, these
agencies may have information on compositions of trade name products.
Various published product indexes and formularies contain information
product compositions that could assist in this regard. The Colour Index, 26
Pesticide index, 2 ’ and Formulary of Paints and Other Coati ngs 2 ° are examples of
compilations of the makeup of products in certain industrial segments that may
help the agency corroborate or interpret the questionnaire responses. Other
publications that may be consulted for this kind of information are cited in
References 29 through 32.
Finally, the National Institute for Occupational Safety and Health (NIOSH)
has compiled an extensive data base containing information on the chemical
composition of approximately 75,000 products from all segments of industry and
commerce. This data base, called the Trade Name Ingredient Clarification File
(TNIC) file, was developed by NIOSH under its National Occupational Hazard
Survey from 1972-1974, and contains data on both metals and organics that were
obtained by NIOSH visits to the various product manufacturers. There is no
report, per Se, summarizing the information in this file, but NIOSH personnel
will develop appropriate retrievals upon request. Any prospective user should
contact the Chief of the Hazard Section at NIOSH at (513) 841-4491 or (FTS)
684-4491. It should be noted that about one-third of the product data in the
37
-------
FIGURE 1
SAMPLE MATERIAL SAFETY DATA SHEET
J. 0 e?
U I O* tC 1. ris l 1.t4M 14
I&ucnM PU P v VO ç17TAJ . prv
• — * *1U.
p .rt, ffi. t & . 1o m ld fl
Ag l.OS b ( O pI et 11 P &4’ J
-
IAI ø• 4. UICü W f L p r* tad
ii.. ii ta J’ C4 X U’ i•t&1 iAj ID ’
EPtWI td ø4 t L .a 9I”*’
J t tg t t ..I ‘a ‘ a .rt .d4
I LPQ II __________ — - —. -
NOTE: Companies are allowed to use any reporting format as long as the
necessary data are contained. Hence, differing formats would be
expected from most companies.
1
MATESLAL SAFETY DATA SHEET
pØ a7I *. *$D * A **LAT*D AT* E.$
I —
—ta
— — .•‘
P r 1i 201
sv ’.*n
$sc on *— MAZLROOU T
‘J
N.th71 i lPl$1 tI Dr%t . 1.11 2.
-8 ty3 AIC ’IO) Is 1% ss
10 I X 1. I
M*t t. i’
L
L —
-
ta S9 l i i
E S.c on i v — FU E AUO 4.OSiO 4 $*ZARO O&rA
-
38
-------
TNIC file is considered confidential and therefore unavailable. NIOSH
repeated this survey in 1981-83 on an expanded list of products. These data
base may prove useful in matching possible pollutants with sources if more
direct procedures are not successful.
Synonyms for commonly inventoried substances are presented in Appendix C.
C. Use of Information in Permit Applications
Many agencies have new source review and permit renewal programs that
cover air toxics. These activities should be viewed as a potential source of
information for use in the air toxics emission inventory. Some agencies
directly incorporate certain permit conditions, such as allowable emissions,
in the emission inventory.
The permit process should be viewed as a good way to collect air toxics
information because of the leverage the agency has on the permit applicant.
However, because the duration of many States’ permit renewal cycle is typi-
cally from 3 to 5 years, the agency normally cannot use this process to
develop or update the air toxics inventory quickly.
8.3 Emission Estimation Procedures
Air toxics emissions are estimated in largely the same manner as are
criteria pollutant emission estimates. The two main approaches involve the
use of emission factors and material balances. The use of speciation factors,
as described in Chapter 3, is generally considered more of a screening tool
although it may also find some application in the detailed inventory. Source
testing is another procedure that may be applicable in certain situations.
Currently, the literature contains limited information on air toxic
emission factors for various substances. The main source is the EPA series
entitled Locating And EstimatinQ Air Emissions From Sources Of (Substance ) and
the Toxic Air Pollutant Emission Factors - A Compilation For Selected Air
Toxic Compounds And Sources . (See Table 5 in Chapter 3).
Another potential source of emission factor information is EPA’s National
Air Toxics Clearinghouse data base.’ 2 States and local agencies have submitted
permit and source test information to the Clearinghouse that may be of assist-
ance in determining emission rates for many kinds of industrial facilities.
Contact names and phone numbers are also given, allowing the user the oppor-
tunity to get more information directly from the supplying agency. As more
data are supplied, the Clearinghouse should improve as a source of emissions
data.
The second major technique for estimating air toxics emissions is the
material balance. In a material balance, emissions are determined by dif-
ference by knowing the amount of a certain material that enters a process and
the amount that (1) leaves either in liquid or solid waste, (2) is contained
as part of a product, (3) is sold to a recovery facility or (4) is converted
to another substance. Material balance techniques are particularly useful for
39
-------
operations using solvents, but can also be used for metals, as well. Fre-
quently, when doing a material balance for screening purposes, the conserva-
tive and simplifying assumption is made that all or some fraction of a solvent
or metal used or consumed in a process is emitted to the air. Of course, any
such conservative screening estimates may have to be refined later in the
program if the Inventory is going to be used for modeling or control strategy
development purposes.
Source testing -- either by the agency or by the source itself --
represents a third technique for estimating air toxics emissions from a
source. While source tests are generally considered to yield more accurate
emission estimates than emission factors, they are not as widely employed for
air toxics for several reasons. First, as with criteria pollutants, is the
cost factor; source testing can be expensive. Second, validated sampling and
analytical procedures are not yet readily available for many air toxics.
Third, there are many situations where emission factors and material balance
procedures will yield acceptable results at much less cost. Finally, there
are certain sources which are not readily amenable to routine source testing,
such as fugitive leaks from valves, pumps, etc., and certain intermittent
operations characterized by varying flow characteristics and concentrations.
Available source test methods are summarized in the Locating and Estimating --
documents for each respective substance. Other methods may be cited in
Reference 8.
The use of emission factors and material balances for estimating air toxic
emissions is exemplified in Appendix A.
8.4 Control Device Applicability and Efficiency
An important consideration when calculating current emission levels, as
well as when projecting emission reductions, is control device applicability.
EPA has published a manual which should help agency personnel select and
evaluate costs of control techniques for reducing air toxics from industrial
and commercial sources. 3 A generic treatment is given both to pollutants and
sources. Air toxics are categorized and handled as (1) organic vapors, (2)
organic particulate (3) inorganic vapors (4) inorganic particulate. Emission
sources are classified as (1) process point sources, (2) process fugitive
sources and (3) area fugitive sources. General guidelines are provided that
match specific control devices with specified emission stream properties
(e.g., pollutant content in stream, temperature, moisture, heat content,
particle size and flow rate.) Basic design parameters are then determined to
provide- general design conditions that should be’ met or exceeded for each
selected control technique to achieve some specified level of control effi-
ciency. This manual can also identify which control techniques will not meet
specified reduction requirements.
4 o_
-------
8.5 Problems iii O&tp Collection and Interpretation
The agency may encounter problems in a number of areas when compiling the
air toxics inventory. Some potential problem areas are identified below.
A. Too Many Substances
If the agency utilizes the open ended approach (discussed in Section
4.1) and does not limit the number of substances, it may receive much more
information than it can adequately process and store in its data handling
system. Also, it may encounter source reluctance to report on a large number
of substances.
B. No Cutoff Levels or Exemptions
Many agencies employ cutoff levels to preclude reporting of low
emission rates and/or exemptions to preclude reporting by certain sources.
(See Chapter 6.) If an agency does not provide for any such exclusion, it
runs the same risks as if it is trying to handle too many substances -- it may
obtain more data than it can efficiently handle and it may encounter source
resistance to reporting low emission levels for many substances.
C. Nomencla ture Problems
Sources may report trade names or chemical synonyms that the agency
may have difficulty in interpreting. In some cases, the composition of trade
name products can be ascertained from the source or supplier by soliciting
Material Safety Data Sheets; however, this is not always successful and such
followup requires additional resources. Some of this can be avoided if the
agency asks sources to report in terms of specific substances.
0. Sources Can’t SuDDly Emissions Data
Some sources, for various reasons, may claim not to be able to supply
the agency with adequate emissions data. This may particularly be the case
with smaller operations that may not have the expertise or resources to deter-
mine emissions. Such determinations may even prove difficult for some large,
complicated manufacturing processes. The agency will want to consider what
services, if any, it may provide to such sources, or if it will make emission
estimates itself.
E. Need For Agency Followup
Several agencies have pointed out the inevitable need for foflowup to
clarify the responses from some sources or to obtain additional or missing
data. This might be the case, for example, where the agency’s screening
estimates for a large source differ significantly from those supplied by the
source, or where the agency believes a source is emitting a certain pollutant
it doesn’t report. The agency will probably want to hold some resources in
reserve to make follow-up contacts to a fraction of its sources.
41
-------
8.6 Source of Inventory Error
As in criteria pollutant inventories, the potential exists for many
sources of error in the inventory of air toxics. Some of these errors will
originate with the sources supplying the data and some will be a result of the
data and methods used by the agency to estimate emissions. A summary of some
of these errors is given below:
‘Missing sources.
.ilnaccurate/insufficient/uninterpretable
data supplied by sources.
‘Missing or inaccurate emission factors.
•Missing or inaccurate VOC and/or particulate
matter apportioning factors.
‘Erroneous material balance or source tests.
A number of corrective measures that may be employed to reduce these
errors is summarized below:
•Perform followup visits, calls, and tests at important sources.
•Checks for missing sources.
-Apply pollutant/source category cross indexes to
criteria pollutant inventories to identify potential
sources.
-Check EPA emission factor and source assessment reports.
-Check permit, registration, compliance, right-to-know
and other State or local data bases.
.Check EPA publications and the EPA National Air Toxics Clearinghouse
for better emission factors and apportioning factors, or develop
such based on emissions data submitted for similar sources.
•Rev.iew material balance and source test data, making sure that all
material pathways have been identified, the best test procedures
have been used and the results are reconcilable with estimates based
on emission factors and/or VOC or PM speciation factors.
It is very difficult to develop rigorous, quantitative estimates of the
precision and accuracy of emission inventories because many of the components
are not amenable to mathematical error analysis. The best means to minimize
error is to use the best data and procedures possible, within existing
resources, and to subject the results to review by those that will either use
the data or be affected by subsequent application thereof.
42
-------
CHAPTER 9
SUMMARY
Many State and local agencies have compiled inventories of air toxic
emissions or are starting to develop them. Several different directions are
being taken in the design of these inventories as well as in how they are
being used in the agencies’ source evaluation and control programs. Because
these inventories can be expensive due to the large number of pollutants that
potentially can be included, careful planning is necessary -- before initiat-
ing any data collection -- to optimize the use of available resources. In
this regard, all end uses of the inventories should be identified during the
planning stages, realizing that the most demanding modeling and risk assess-
ment applications will dictate how detailed and extensive the inventories must
be.
If possible, some type of screening study might be performed to focus the
inventory on those pollutants, sources and geographic areas that pose the
greatest risk. Various tools are available to compile preliminary inventories
for screening purposes without having to make extensive source contacts.
These tools include EPA published emission factors, speciation factors and
source assessment documents along with existing criteria pollutant emission
inventories and permit files. Federal and State right-to-know standards may
also result in useful information for this purpose.
The detailed emission inventory of air toxics is compiled using the same
basic techniques employed in criteria pollutant inventories. Many agencies
are, in fact, directly coupl ing their air toxics inventory activities with
their criteria pollutant programs. The data in the criteria pollutant
inventory and permit system can serve as a useful starting point for develop-
ing the detailed air toxics inventory, either to help locate potential sources
of air toxics or to make emission estimates therefrom. For less important
sources, the agency may not need to go beyond the preliminary emission
estimates made in the screening inventory; however, for more important
sources, questionnaires or other direct source contacts may be necessary to
determine emissions. The mechanics of making source contacts and determining
emissions will be largely the same as the criteria pollutant inventory.
Most sources of air toxics will already have been inventoried as part of
particulate or VOC programs. Additional emphasis will be needed to include
non-traditional sources, sources below the criteria pollutant cutoff level(s),
and sources of photochemically nonreactive compounds (particularly methylene
chloride and methyl chloroform.) And further spatial or temporal resolution
may be needed in the air toxics inventory for certain area sources than is
afforded in the criteria pollutant inventory.
43
-------
REF ERENCES
1. Statement of Lee M. Thomas, Administrator, U.S. Environmental Protection
Agency, before the Subcommittee on Health and the Environment Committee on
Energy and Commerce, U.S. House of Representative, June 11, 1985.
2.
3.
Procedures for the Preparation of Emission Inventories for Precursors of
Ozone. Volume 1, EPA-450/4-88-021, Third Edition, U.S. Environmental
Protection Agency, Office of Air Quality Planning and Standards, Research
Triangle Park, NC, December 1988.
Procedures for the Preparation of Emission Inventories for Volatile
Organic Compounds. Volume II: Emissions Inventory Requirements for
Photochemical Air quality Simulation Models. EPA-450/4-79-018. U.S.
Environmental Protection Agency, Research Triangle Park, NC, September
1979.
4. Procedures for Emission Inventory Preparation, U.S. Environmental
Protection Agency, Office of Air Quality Planning and Standards, Research
Triangle Park, NC:
a. Volume I: Emission Inventory Fundamentals , EPA-450/4-81-026a,
September 1981.
b. Volume II: Point Sources , EPA-450/4-81-026b, September 1981.
c. Volume III: Area Sources , EPA-450/4-81-026c, September 1981.
d. Volume IV: Mobile Sources , EPA-450/4-81-026d (Revised), July 198g.
e. Volume V: Bibliography , EPA-450/4-81-026e, September 1981.
5. Example Emission Inventory Documentation for Post-1987 Ozone State
Implementation Plans (SIPSI , EPA-450/4-89-018. U.S. Environmental
Protection Agency, Research Triangle Park, NC, October 1989.
6. Development of Questionnaires for Various Emission Inventory Uses . EPA-
450/3-78-122. U.S. Environmental Protection Agency, Research Triangle
Park, NC, June 1979.
7. Development of an Emission Inventory Quality Assurance Program . EPA-
450/4-79-006. U.S. Environmental Protection Agency, Research Triangle
Park, NC, December 1978.
8. Bibliography of Selected EPA Reports and Federal Register Notices Related
to Air Toxics, U.S. Environmental Protection Agency, Research Triangle
Park, NC:
a. Volume 1: Citations , EPA-450/5-87-005, NTIS PB 88-136601, July 1987.
44
-------
b. Volume 2: Citations , EPA-450/5-88-005, NTIS PB 89-103436/AS, July
1988.
c. Volume 3: Citations , EPA-450/3-89-25, NTIS PB 90-1135/AS, July 1989.
d. Index : EPA-450/3-89-25a, NTIS PB 90-113423/AS, July 1989.
9. A Workbook of Screening Techniques for Assessing Impacts of Toxic Air
Pollutants , EPA-450/4-88-009, NTIS PB 89-134340, U.S. Environmental
Protection Agency, Research Triangle Park, NC, September 1988.
10. Guideline on Air quality Models (Revised) and Suoplement A , EPA-450/2-78-
027R, NTIS PB 86-245248, PB 88-150958, U.S. Environmental Protection
Agency, Research Triangle Park, NC, July 1986 and July 1987.
11. Haemisegger, E., et al., The Air Toxics Problem in the United States: An
Analyses of Cancer Risks from Selected Pollutants , U. S. Environmental
Protection Agency, Research Triangle Park, NC, May 1985.
12. NATICH Data Base Report on State. Local and EPA Air Toxics Activities ,
U.S. Environmental Protection Agency and STAPPA/ALAPCO, EPA-450/3-89-29,
U.S. Environmental Protection Agency, Research Triangle Park, NC, July
1989.
13. Air Emission Species Manual , EPA-450/2-88-003a and b. NTIS PB 88-225792,
PB 88-225,800, U.S. Environmental Protection Agency, Research Triangle
Park, NC, April 1988.
14. Survey of Ethylene Dichioride Emission Sources , EPA-450/3-84-018, U.S.
Environmental Protection Agency, Research Triangle Park, NC, December
1984.
15. Survey of Methvlene Chloride Emission Sources , EPA-450/3-85-015, U.S.
Environmental Protection Agency, Research Triangle Park, NC, June 1985.
16. Survey of Carbon Tetrachloride Emission Sources , EPA-450/3-85-018, U.S.
Environmental Protection Agency, Research Triangle Park, NC, July 1985.
17. Survey of Trichloroethvlene Emission Sources , EPA-450/3-85-021, U.S.
Environmental Protection Agency, Research Triangle Park, NC, July 1985.
18. Survey of Chloroform Emission Sources , EPA-450/3-85-026, U.S Environmen-
tal Protection Agency, Research Triangle Park, NC, October 1985.
19. Survey of Perchloroethvlene Emission Sources , EPA-450/3-85-017, U.S.
Environmental Protection Agency, Research Triangle Park, NC, June 1985.
20. A Preliminary Exposure and Preliminary Risk Appraisal for 35 U. S.
Counties . Prepared for U.S. Environmental Protection Agency under
Contract Number 68-01-6715 by Versar, Inc., and American Management
Systems, September 1984.
45
-------
21. Methods for Pollutant SelectipiLand Prioritiza tion , EPA-450/5-86-O1O,
U.S. Environmental Protection Agency, Research Triangle Park, NC, July
1986.
22. Lahre, Thomas F., “Characterization of Available Nationwide Air Toxics
Emissions Data.” U.S. Environmental Protection Agency, Research Triangle
Park, NC, June 13, 1984. (Unpublished).
23. Air Toxics Controlhbilit Study , Prepared for U.S. Environmental
Protection Agency by E. H. Pechan & Associates, Inc. and Radian Corpora-
tion under Contract Number 68-01-7047, Work Assignment Number 26, October
1985.
24. Compilation of Air Toxics Emission Inventory Ouestionnaires , EPA-450/4-
88-008, U.S. Environmental Protection Agency, Research Triangle Park, NC,
June 1988.
25. “Hazard Communication,” Final Rule. Federal Register, Volume 48, Number
228, Friday, November 25, 1983, pp. 53280-53348.
26. Colour Index , Third Edition, Society of Dryers and Colourists, American
Association of Textile Chemists and Colourists. Bradford, West York-
shire, 1982.
27. Wiswesser, W.J., Pesticrne Index , 5th Edition, Entomological Society of
America, College Park, Maryland, 1976.
28. Ash, M. and I. Ash, Formulary of Paints and Other Coatings , Volume 1, New
York, NY, Chemical Publishing Company, 1978.
29. Merck Index: An Encyclopedia of Chemicals and Drugs , Ninth Edition,
March 1976, Rahway, NJ.
30. Handbook of Chemical Synonyms and Trade Names , Eight Edition, Edited by
Edward I. Cooke, CRC Press, 1978.
31. Condensed Chemical Dictionary , Tenth Edition, Gessner Hawley, Edited by
Van NosReinhold, 1981.
32. Clinical Toxicology of Commercial Products , Fifth Edition, Robert E.
Gesselin. Williams and Wilkins, 1984.
33. Evaluation of Control Technologies for Hazardous Air Pollutants , EPA-
625/6-86-Olla, U.S. Environmental Protection Agency, Research Triangle
Park, NC, February 1986.
46
-------
APPENDIX A
EXAMPLE APPLICATIONS OF SCREENING TOOLS
-------
APPENDIX A
EXAMPLE APPLICATIONS OF SCREENING TOOLS
Examples of each of the screening tools discussed in Chapter 3 of this
report and how they can be applied are given in Appendix A. The screening tools
described are:
-SIC/pollutant, pollutant/SIC crosswalk
-Emission factors
-Speciation/apportioning factors
-Conservative mass balance
The strengths and limitations of each tool as related to air toxics emission
inventory development and permitting/new source review are discussed. For either
of these applications, the objectives of using a screening tool are:
-To identify potential toxic air pollutants associated with a source
category;
-To estimate the quantity of air toxics emissions; and
-To focus or direct follow-up, detailed analyses.
UsThg combinations of each of the four types of screening tools can allow
State/local agencies to reach these three objectives.
A.1 USE OF SIC/POLLUTANT CROSSWALK INDICES
In this section, uses of the SIC/pollutant crosswalk indices are discussed.
While the crosswalk provide an indication of the pollutants reasonably
anticipated to be associated with a given SIC code, they do not provide definite
proof that the pollutants are emitted. One limitation of using the SIC pollutant
crosswalk indices is that any given source may emit one, all, or none of the
pollutants listed in the index. A source may emit other pollutants not covered
in the Toxic Air Pollutant/Source Crosswalk document or not permitted by
reporting agencies. An example is provided.
Example: Identification of a Potential Emitted of Air Toxic
A state air pollution control agency needs to evaluate which toxic
pollutants are emitted in the state.
As a first step, the agency could identify the source categories the
area of concern by SIC code: for example, a gray and ductile iron
foundry - SIC. 3321. The existence of this source may have been
A-i
-------
TABLE A-i
SIC
CODE
SCC
SIC DESCRIPTION
CODE ASSOCIATED POLLUTANTS
3321 Gray and Ductile Iron Acetaldehyde (21, 22, 23, 28, 29)
Iron Foundaries Acetone (21, 22, 27, 28)
Acrolein (21, 22, 23, 24, 28, 29)
Aluminum (21, 22, 27, 28)
Aluminum Oxide (27)
Ammonia (21, 22, 23, 27, 29)
Aniline (21, 22, 28)
Antimony (27)
Arsenic (20, 21, 23, 27, 29)
Asbestos (27)
Barium (27)
Benzene (21, 22, 23, 24, 27, 28, 29)
Benzene, Ethyl- (27)
1,2,4-Benzenetricarboxyl Ic Acid
1,2-Anhydride (28)
Beryllium (21, 23, 29)
Biphenyls Polychlorinated (PCBS)
(27)
1-Butanol (27)
Butyl Acetate (22)
Cadmium (20. 21. 22, 27, 29)
Chlorine (27)
Chloroform (23)
Chiorophenols (21, 24, 27, 28)
Chromic Acid (VI) (22)
Chromium (20, 21, 22, 23, 27, 29)
Cobalt (27)
Copper (21, 22, 23, 26, 27, 29)
Corticosterone (28)
Cumene (27)
Dibutyl Phthalate (27)
1,2-Dichioroethylene (28)
Oiethanaolamine (27)
Oimethyl Phthalate (27)
Dioxins, Polychiorinated Dibenzo-P-,
Total (27)
Dioxins, Polychlorinated, Total (27)
Diphenylmethanedlisocyanate, 4,4’ -
(27)
Ethane, Dichlorotetrafluoro (28)
Ethane, 1,1,1-Trichloro (21, 22, 23,
27, 28, 29)
Ethane, 1,1,2-Trichioro (27)
Ethanol, 2-Ethoxy- (27)
Ethylene Glycol (27)
Formaldehyde (21, 22, 23, 24, 27,
Iron Foundaries 28, 29)
1-2
-------
TABLE A-i (continued)
SIC
CODE
5CC
SIC DESCRIPTION
CODE ASSOCIATED POLLUTANTS
3321 Gray and Ductile Iron Furan, 1,2,3,4,7,8-Hexachioro-
dibenzo- (28)
Furan, 1 ,2,3,6,7,8-Hexachloro-
dibenzo- (27)
Furan, 1,2,3,7,8,9-Hexachioro-
dibenzo- (27)
Furfuryl Alcohol (22)
Hydrogen Cyanide (21, 22, 28)
Hydrogen Sulfide (21, 22, 23, 29)
Iron (22)
Isobutyl Alcohol (22)
Isopropanol (27)
Lead (21, 23, 27, 29)
Manganese (21, 23, 26, 27, 29)
Mercury (3)
Methanol (27)
Methylene Chloride (27)
Methyl Ethyl Ketone (21, 22, 27)
Methyl Isobutyl Ketone (27)
Naphthalene (21, 22, 23, 24, 27, 28,
29)
Nickel (20, 21, 22, 23, 27, 29)
1-Nonene (28)
Phenol (22, 22, 23, 24, 27, 28)
Phenol, 2,4-Dimethyl (28)
Phosphoric Acid (27)
Phthalate, Olisooctyl (29)
Polycyclic Organic Matter (3)
Polyvinyl Chloride Latex (28)
Propane, 1,2,3-Trichioro- (29)
Saccharin (27)
Silica, Crystalline - Quartz (22)
Silicon Carbide (22)
Sodium Hydroxide (27)
Styrene (21, 22, 23, 29)
Sulfuric Acid (27)
Tetrachl oroethyl ene (27)
Toluene (21, 22, 23, 24, 27, 28, 29)
Trichi oroethyl ene (27)
Uranium (22)
Vanadium (27)
Xylene, M- (21, 23, 24, 27)
Xylene, 0- (21, 23, 24)
Xylenes (Mixed Isomers) (27, 29)
Xylidine (29)
Zinc (21, 11, 23, 27, 28, 29)
A- 3
-------
determined from an existing criteria pollutant inventory or permit!
registration file. The SIC/pollutant crosswalk document (Table A-i)
shows 79 pollutants potentially associated with SIC 3321. Based on
this information, the agency could decide to inventory and further
investigate these pollutants.
Conversely, a pollutant/SIC index can be used to identify potential
emitters of particularly toxics. For example, if an agency were
interested in acetaldehyde, there are 50 potential source categories
associated with potentially emit acetaldehyde (see Table A-2).
A-4
-------
TABLE A-2
Acenaphthene 2431, 2491, 2752, 2759, 2812, 2821,
2833, 2842, 2865, 2869, 3089, 3411,
3471, 3663, 3721, 3728, 4952, 8731,
9223, 9711
Acenaphthylene 2491
Acephenanthrylene, Benz(E)- 2491, 25 , 251 , 252 , 253
Acetaldehyde 2013, 2087, 2095, 2099, 2297, 2421,
2429, 2431, 2435, 2436, 2621, 265
2656, 267 , 2673, 2674, 2812, 2819,
2821, 2822, 2823, 2824, 2844, 2851,
2865, 2869, 287 , 2873, 2879, 2891,
2892, 2899, 2911, 2951, 308 , 32
3312, 3321, 3479, 3679, 3826, 3827,
3829, 3861, 4911, 4953, 516 , 6512,
8059, 8062
Acetaldehyde, Chioro- 07 , 2869
Acetaldehyde, Diethyl Acetal 2844
Acetamide 2819, 2821, 2833, 2842, 2865, 2869,
3479, 3674
Acetamide, Methyl -N-
( [ methyl carbamoyl )oxy]thio- 5153
Acetamide, N-9H-Fluoren-2-Yl 289 , 3861
Acetamide, N,N-Dimethyl- 2819, 2821, 308 , 3585, 3861
Acetanilide 2851, 2865, 3411
Acetanil ide, 2-Chloro-2’ ,6’ -
Diethyl-N-(Methoxymethyl)- 2431, 2879, 308 , 3524, 3731
Acetate, 2-Butoxyethyl- 2869, 308 , 3241, 3312, 3484, 3585,
3629, 3641, 3821
Acetate, Ethyl Chioro- 2865, 2869, 2879
Acetate, Ethyl Cyano- 2833, 2865, 2869
A- 5
-------
A.2 USE OF EMISSION FACTOR LISTING
When source - specific data are not available for a facility, emission
factors are useful tools for making emission estimates. Emission factors may
also be used to cross check data submitted by a source during the permitting
process. As pointed out in Chapter 3, the major sources of toxic emission
factors are the “Locating and Estimating.... series” and the Air Toxics Emission
Factor Compilation documents (see Table 4 in Chapter 3).
Example: Preparation of An Air Toxics Emission Estimate
Suppose an agency has decided to prepare a quick, preliminary
estimate of cadmium emissions from municipal waste combustors.
To illustrate the use of emissions factors for this source
category, Table A-3 from the compilation document shows emission
factors for this type of combustors. In order to calculate an
emissions estimate, additional information is needed (e.g., type
of combustor and kilograms of refuse burned per year). Such data
may already exist in a permit file or from inspection reports.
If not, an inquiry made to the facility or facilities or a site
inspection may be needed. Assume there are two combustors in the
area of concern, one a mass burn water wall facility burning
2,100,000 kilograms per year and, a mass burn refractory unit
burning 800,000 kilograms per year. The estimated emissions are:
Cadium Emissions = 686 mg/kg x 2,100,000 kg/yr + 1228 mg/kg x
1,500,00 kg/yr
Cadium Emissions = 1440 x 106 + 1842 x l0
Cadium Emissions = 3282 x 106 mg/yr or 3282 kg/yr
A-6
-------
TABLE A-3
IIOU SI5IAL PROCLSS
Secondary zinc productIon
tiec I rop l at I rig
(lectro slatlr i g
Nickel csthlia battety
manufacture
Mte Icipal waste cattaj5t tori
Mtaitcipat waste cottajstion
Mtsuictpcl waste cmnI .rct ton
IttsitctpoI wasie cuitj.jsiion
Hieitctpot waste cantaist ion
Sewagi. slizlge incineration
Stwage sludge incineration
Hospital waste incineration
AlriioriiLet sulfate prix5 ct toil
caprolactae by pronkict
S,itihettc fibers niscuse
rayon
Molts
Controlled with cyclone • baghouse
stick saepl trig • engineer tng
silt mate
Controlled with three cyclones scsI
bsgttouse
cm-strolled with three cyclones
baghouss stack sasipling
engineering estimate
Uncontrolled
ttncontrol led
tktcont rolled, measured
Survey of carpantes
Controlled by IS P , based on smircH
tests
Controlled by (5?, based on source
iests
Controlled by ISP, based on source
lest
Cont rot led by t5P based on stance
eests
Controlled by spe-ay baffle
1 scrttter, based on material
balsnce for nedel irtctnerator
Average ol three tests, scrt er
Average of 13 tests, scri4 r,
range 0 26 63
tlnconsrol led, test date
Based on test data
uncontrolled, low quality lactor
POCIulsdt
CAS
Sli lSt t
/4404 : 19
sic
C c c l
334 1
I$ISSIIM SUJRcE
7440439 3341
74404)9 334%
Sccorflaiy lead smelting
Secondary lead seal t trig
Sceorniary lead smelt Ins
Reverberelory furnace 30400402
5CC tMlSSttsl fACtOR
CsuS ’tiuo
C anSfi I tar
Cs4ni un
C ado as
Cocsn I tan
Cstsniutt
Caciatta
Cstsitiua
C suits as
Cactittua
C sun tin
(stint tat
C stint tan
Cscanl tat
C ants I tin
C api 01 lam
CarL n nitsut tide
74404 )9
7440439
1 640439
7440439
7440439
7440439
74 40439
7440439
7440439
7440439
744 0439
7440439
105602
75150
3341
347,
341 $
3691
4953
4953
495 ,
4953
4953
4953
4953
806
2869
2824
Reverberatory furnace
Blast lurnace
lotal etalas ions
Plating tank
(nttro process
total etatselona
Mass turn, water salt Iscility
Mass burn, refractory focility
Refuse dectved fuel faislity
itu .alar facility
Incinerator stack
I lusdioed bed furnace
MultipLe hearth furnace
Inc I n c rat or
Rotary di per
General estissions
1.6 s lOt 6 lb/ton Pb
30400402 5 a t0l ? lb/ton Pb
30400403 2 a lOt 6 lb/ton Pb
304005 I a 104 2 lb/ton Zn produced
5 0 a lO t 5 gJhr/anp
0 00005 g/hr/a’tp
304005 2 lb/ton Cd charged
301001 686 sq/kg controlled
particulate eats
501001 1228 eq/kg control led
particulate eats
501001 106 atg/kg controlled
part tculate mais
501001 3510 mjttg cosittotled
pert tculate emis
501001 0 004 k 0 /Mg ma-tic ao(td
waste ldry wt
50100506 0 74 mg/kg dry sIt.aJgt
50100506 9 9 sq/kg dry sludge
315020 35 2 a 10 4 lb/ton Iced
30121009 I I kg/hr
30102501 55 Itt/ton fibers
ttIt5 INCl
34
38
38
36
147
92
38
Its
118
118
lid
Qa
to ’
to’
9
t i c
9 ,
-------
A.3 USE OF SPECIES FACTORS
Another technique that may be used to estimate emission of air toxics is
to apply species factors to existing estimates of particulate matter (PM) or
volatile organic compound (VOC).
Two major speciation manuals are currently available as pointed out in
Chapter 3, one for volatile organic compounds and one for particulate matter.
The VOC species manual shows percent by weight of specific chemicals in emissions
from specific processes. The sources covered in this manual include point and
area sources. SCC’s are presented where applicable. The manual for particulate
matter was designed to provide particle composition data that are useful not only
for air toxics emission inventory, but for receptor modeling applications work
as well. The manual lists composition data as percent by weight for four
particle size ranges, for specific metals, organic carbon and elemental carbon,
and for particulate emissions from point and area sources. SCC’s provided when
applicable. Example of how these tables amy be used in air toxics emission
inventory are described below.
Example: quantifying Species in Particulate Matter Emissions
Suppose that an agency wished to identify particulate elements that
comprise the particulate matter emissions from a primary copper
roaster. The PM speciation manual lists species present, or weight
percent basis for four particle size ranges from a primary copper
roaster as shown in profile 29209 (see Table 6 in Chapter 3 for
profile).
If the total PM emissions of the subject plant were 150 tons of
particulate per year and the agency s only interest in the fine
fraction (i.e., <2.5 jan) for arsenic, then from the profile 29209,
arsenic is shown to be 4.2 percent by weight. The estimated arsenic
emissions would be 150 ton/year x 4.2 percent or 6.3 tons per year.
Similarly, the profile (profile 29209) shows that in the 0-10 jan range,
chromium is .042 percent by weight. Then if total PM emissions of
sources is 150 ton/year, then the estimated chromium emissions would
be 150 ton/year x .042 percent or .063 tons per year.
Before applying the species profile data, the agency should verify that
the source is similar to the typical source in that category. Such information
and data may be obtained from’ permit applications, inspection reports, or
registration files.
Examole: quantifying Species of Volatile 0r anic Compounds
An agency could estimate species emissions from crude oil production -
well heads by using the VOC Species Manual. From the VOC manual
(profile number 1212, Table 5 in Chapter 3) you may locate a list of
A-8
-------
30 compound emitted from this source category. If the agency wants
only benzene and toluene, the profile list these compound as .05 and
.09 percent by weight, respectively. Based on the total VOC emission
from the criteria pollutant inventory, estimates of the two species
can be made. Assume the source emitted 100 ton/year total VOC then
benzene and toluene can be estimated by multiplying 100 tons/year VOC
by the respective percent weight, yielding .05 tons/year for benzene
and .09 tons per/year for toluene.
The primary limitations when using species factors are that (1) materials
used at each facility will vary, and (2) speciation data-are not available for
all source categories. For example, a surface coating operation may be known
to use varnish and lacquer, but the specific type and its composition are not
know. The speciatian factors can be successfully used to get a rough emissions
estimate of specific pollutants. That level of accuracy may be all that is
needed for screening purposes. If a more detailed analysis is needed, the source
could be contacted to obtain national safety data sheets for specific materials
used. The same calculation methods presented here could then be made using the
site-specific species data.
A.4 USE OF MATERIAL SAFETY DATA SHEETS
Suppose an agency needs to speciate VOC emissions from a furniture
manufacturing plant. When reviewing permit application files for the source,
the agency found information indicating that Supplier X’s Enamel 209 and Primer
387 were the raw materials used at the plant and also found that Material Safety
Data Sheets for these two products were included in the files. Material safety
data sheets generally list the composition of the material by percent weight
(Figure A-2).
By obtaining total VOC emissions for the plant from the criteria pollutant
inventory and/or permit applications, the emissions of specific toxic pollutants
can be estimated using the information on the material safety data sheet. If
the data in Figure A-2 represent the enamel and primers used at the plant, and
assuming that the plants total VOC emissions were 50 tons per year, then 7.3 tons
per year toluene and 1.9 tons per year of xylene would be an estimate of the air
toxics emitted.
A- 9
-------
II HAZARDOUS INGREDIENTS
MATERIAL OR
COMPONENT
%
HAZARD DATA
Toluene
14.5
(w)
Xylene
3.8
(w)
Ethyl Acetate
5.9
(w)
Figure A-2. Sample Material Safety Data Sheet: Hazardous Ingredient Section.
A-1O
-------
A.5 USE OF MATERIAL BALANCES
A material balance can be used to estimate emissions from processes where
emission factors, speciation data, or source assessment data are not available.
Even where these other tools can be used, the material balance may be the best
approach in some situations. The simplest material balances (most conservative)
are those that assume most or all of a material used or fed into a process is
emitted to the atmosphere. In many cases, this assumption will overestimate
emissions; however, for screening purposes, such as assumption may be reasonable.
Examples of how a material balance may be used to estimate emissions are provided
in this section.
Example: Estimating Emissions of VOC from Air Stripping of Contaminated
Groundwater
Suppose that an agency needs to estimate emissions of two solvents
(1,1,1-trichloroethane and carbon tetrachioride) from an air stripper
used to decontaminate groundwater. Using simple material balances
based on concentrations of the solvents in the groundwater and the
operating parameters of the air stripper, emissions (lb/hr) of the two
solvents can be estimated.
Assume that the groundwater concentrations of the two solvents were:
1,1,1-Trichioroethane 38 mg/i
Carbon Tetrachioride 12.6 mg/i
Then assume that all of the solvent contained in the groundwater is
emitted to the atmosphere and that the stripper cleaned 50,000 1 of
water per hour. Then emissions could be estimated for 1,1,1-trichioro-
ettiane as follows:
50,000 1 water/hr x 38 mg solvent/i x 1 lb/454 g x 1
g/100 mg 42. lb solvent emitter/hr
Similarly, the est mated emissions of carbon tetrachioride would be
1.4 lb emitted/hr.
Fxample: Estimating Emissions of Metals and Organics from Waste Oil
Combust ion
Suppose that an agency needed to estimate the areawide emissions of
cadmium, chromium, and perchioroethylene from waste oil combustion.
Using simple material balances based on the concentrations of metals
and organics in the waste oil and the amount of waste oil burned
annually, and by estimating the amount of metals and organics contained
in the fly ash, emissions (tons/year) can be estimated.
A-il
-------
Assuming that about 75 percent of the metals in waste oil are emitted
(conservative estimate from the general literature), 739,000 gallons
of waste oil are burned per year and the oil contains 2.7 mg/i cadmium,
the emissions could be estimated as follows for cadmium:
739,000 gal waste oil burned/yr x 2.7 mg Cd/i waste oil x 75%
metal emitted x 2.785 1 waste oil/galloon waste oil x 1 g/i000
mg x lb/454 g x 1 ton/2000 lb 0.006 ton/year Cd.
For chromium, the emissions would be 0.085 tons/year, based on 37 mg
Cr/i in waste oil.
For perchloroethylene, assume that 1 percent of the organics in waste
oil are emitted (99 percent destruction), then:
739,000 gal waste oil burned/year x 448 mg perc/1 waste oil x 1%
organic emitted x 3.785 1 waste oil/gal waste oil x 1 g/1000 mg
x 1 lb/454 g x 1 ton/2000 lb = 0.014 ton/year.
For the processes where reasonable assumptions can be made about the fate
of a pollutant during the process, the conservative material balance provides
a reasonable first estimate of emissions for screening. For processes where a
material is reacted to form a product or otherwise significantly chemically
changed, the conservative material balance technique is not appropriate.
Consulting the literature for more detailed data about the process may provide
enough information to allow assumptions to be made about the emission of a
specific pollutant.
A-12
-------
APPENDIX B
EXAMPLE AIR TOXICS PRODUCTION AND USE QUESTIONNAIRE
-------
APPENDIX B
This appendix is an example of a general production and use survey form
designed to acquire information on chemical production, storage, and use at
facilities. This type of form may be used as a screening survey to determine
who should receive more detailed followup questionnaires. Or, it may also be
utilized to gather emergency preparedness data.
B-i
-------
To the President or Chief Environmental Officer:
As part of our continuing efforts to improve air quality in __________
the Bureau of Air Pollution Control (Bureau) is now engaged in evaluating the
effect of potentially toxic chemical pollutants. The enclosed Chemical Substance
Survey is a vital part of that evaluation. We request your full cooperation in
this effort.
Your are requested to report materials which are stored or produced on site
in sufficient quantities that could potentially pose an immediate or chronic
threat to off-site life and health, including emissions from accidental releases.
In reporting this chemical substance data, do not report chemicals such as those
used in laboratory work, or for weed and pest control, or virgin fuels such as
coal or oil.
The questionnaire was designed for a wide variety of operations.
Accordingly, some sections may not apply to your facil ity; simply mark those “not
applicable.” Remember, it is just as important for this inventory to alleviate
concerns about the unknowns by documenting that potentially serious hazardous
air emissions are taking place.
If there is insufficient room on Form 2 to record all of the chemical
substances to be reported, please make sufficient additional copies of Form 2
before filling it out.
The Bureau further requests that .these forms be filled out separately for
each plant or facility operated by the company. Please return the completed
forms to the Bureau, at the above address, as soon as possible, but no later than
Questions should be referred to:
Your cooperation in filling out the questionnaire is greatly appreciated.
Sincerely,
B -2.
-------
COUNTY HEALTh DEPARTMENT
BUREAU OF AIR POLLUTION CONTROL
Chemical Substance Survey
Attached are two survey forms. These forms request two kinds of data:
1. General information such as company name and address, plant
Standard Industrial Classification (SIC) code, name of the person
filling out the form, and general plant description. These data
are requested on Form 1. Use one set of forms for each plant.
Please fill out and return Form 1 even if you have nothing to
report on Form 2.
2. Chemical substance information: We are interested in determining
what chemicals or substances are used, made or stored at this
site.
For the purposes of this study, the scope of the survey is being
defined as pertaining to chemica’ substances which have been found
to pose a threat to life and health. The enclosed list of
chemical substances are of concern to the Health Department. Use
Form 2 to report any of these chemicals which are stored or
produced on site in sufficient quantities that could potentially
pose an immediate or chronic threat to off-site life or health.
The chemicals have been grouped into categories, similar to the
ones below to assist you in locating the chemicals that you use:
Acids Metals and Compounds
Catalysts and Reagents Monomers or used for Organic
Synthesis
Chemical Intermediates Pesticide Compounds
Chemical Warfare Agents Plasticizing Compounds
Cyan ides Pharmaceuticals
Flavors and Fragrances Refrigerants
Fumigants Solvents
General Use Chemicals Miscellaneous Chemicals
Industrial Gases
The basic chemical data are to be recorded on Form 2.
Please fill in all data to the maximum extend possible. We will be glad
to answer any questions about this form or the information we are requesting.
Just call ( ) ________ - _______ . Return completed forms to:
8-3
-------
COUNTY HEALTh DEPARTMENT
BUREAU OF AIR POLLUTION CONTROL
CHEMICAL SUBSTANCE SURVEY
FORM 1 - COMPANY INFORMATION
Plant/Facility Location
1. Company Name: _________________________________
2. Company Address: ______________________________
3. City/Zip Code: _________________________________
4. Municipality (or Township): ___________________
5. Telephone Number: ______________________________
6. Contact Person (Name/Title): ___________________
Mailing Information, if Different from Above:
7. Company Name: __________________________________
8. Company Address: _______________________________
9. City/Zip Code: _______________________________________________________
10. Plant SIC Code: ______________________________________________________
11. Please give a brief, general description of operations at this plant.
B-4
-------
12. Is there an emergency response program to deal with releases of chemical
substance in use or storage on your premises? _________________________
13. If the answer to No. 12 is yes, is the program coordinated with local
emergency management agencies (fire and police departments, or other) and
i so, which ones? ____________________________________________________
14. Give the plant UTM coordinates, if available: ________________________
15. Federal Employer I.D. Number for plant/facility name in Item 1.
I certify that to the best of my knowledge, the above information is
correct.
Signature (or person filling out these
forms)
Name (type or print)
Title
Telephone Number where above person
can be reached.
-------
COUNTY HEALTh DEPARTMENT
BUREAU OF AIR POLLUTION CONTROL
CHEMICAL SUBSTANCE SURVEY
FORM 2 - INSTRUCTIONS
If there is insufficient room on Form 2 to record all of the chemical
substances to be reported, please make sufficient additional copies of Form 2
before filling it out. Use this form to report any chemical substances from the
enclosed list, that are stored or produced on-site in sufficient quantities that
could pose an immediate or chronic threat to off-site life or health, including
emissions from accidental releases.
Item 20 - Comercial Name, Comon Name, or CAS Name
Chemical substances should be listed by their
For instance, methyl ethyl ketone (common name)
(another common name), 2-Butanone (CAS name) or
example, Jones super solvent). For substances
chemicals, list all know ingredients separately
ation available on the individual ingredients.
derived from the nomenclature used by the American
Chemical Abstract Service in the 8th Collection
Abstracts.
The Chemical Abstract Service (CAS) Registry Number is a numeric
designation assigned by the American Chemical Society’s Chemical Abstracts
Service and uniquely identifies a specific chemical compound. This entry
allows one to conclusively identify a substance regardless of the name or
naming system used. CAS numbers for most of the substances of interest
here are on the attached list.
Item 22 - Quantity Purchased by the Facility Annually in Pounds
Give the quantity annually purchased by the facility in pounds. If the
chemical substance listed in Item 20 is part of a mixture of other chemicals
(such as a solution, suspension, emulsion, etc.), list information only for
the designated chemical substance in Columns 21 through 26. In Column 20,
along with the chemical substance name, give pertinent information on the
mixture (such as: “stored as a 1/2% solution in mineral oil”, or “chemical
is 1000 ppm in an inert compressed gas product”). For this Item, and for
Item 23, the amount reported should be the higher of the following two
figures:
1. Average of last three (3) years
2. Amount used in 1984.
Chemicals/Substances In Use.
most commonly used name.
may be referred to as MEK
some commercial name (for
consisting of two or more
unless there is no inform-
The CAS name is that name
Item 21 - CAS Number
Chemical Soclety’s
Index of Chemical
B-6
-------
Item 23 - Quantity Produced or Manufactured Annual, in Pounds
Give the quantity produced or manufactured by the facility annually, in
pounds.
Item 24 - Maximum Quantity Stored in the Plant at any one time, in Pounds
Item 25 - Indicate by using one or more of the following codes the type of
airborne emissions of this substance during the last three (3) years.
A - Accidental (only) I - Intermittent
B - During Breakdown S - During Start-up/Shutdown
only
C - Continuous During Operations N - There were no air/
emissions
Item 25 - Indicate with “Yes” if there is air pollution control equipment on the
source of emissions and with “No” if there is no air pollution control
equipment on the source of emissions. (Note: Emissions control for
the substance in question.)
If it is more convenient to report your data in units other than pounds,
you may do so. However, we request that you also provide adequate information
to convert those units to pounds.
B-7
-------
COUNTY HEALTH DEPARTMENT
BUREAU OF AIR POLLUTION CONTROL
Cheuzicai Subatance Survey — Form 2
Chemical Subatances in U8e
Commercial
Uain or CAS
Name,
Name
Common
CAS
Number
Quantity Quantity Produced
Purchased or Manufactured
Haximum Quanti—
ty Stored in
Chemical Substance Emissions
Type of
Air Pollution
Annually Annually
Plant at
any
Emissions
Control Equip.
(Pounds) (Pounds)
one Time
(lbs.)
Code
Ye8 No
20
21
22 23
24
25
26
I certify that, to the best o my knowledge, the above Information Ia correct.
Date Signature Title
-------
APPENDIX C
GLOSSARY OF CHEMICAL NAME SYNONYMS
-------
APPENDIX C
GLOSSARY OF CHEMICAL NAME SYNONYMS
A brief glossary of common synonyms for the substances addressed in this
report is listed in Appendix C. The more frequently used names are listed in
the left hand column and synonyms in the right hand column. Other synonyms may
exist, but more common ones are listed here. Additional synonyms can be found
in the EPA CASR or Chemical Activities Status Report (EPA-560/130-80-040a).
C - i
-------
GLOSSARY OF CHEMICAL NAMES SYNONYMS’
Substance
yn 0 nym
Oxychioride
Acetaldehyde
Acrolein
Acrylonitrile
Allyl Chloride
Amonia
Benzene
Benzo(a)pyrene
Benzyl Chloride
1 ,3-Butadiene
Carbon Tetrachioride
Chioroethane
Chi oro form
Chi oroprene
C re sal
p-Dichl orobenzene
Dioxane
Ep ichi orohydrin
Ethylene Dibromide
Ethylene Dichioride
Ethylene Oxide
Formaldehyde
Hexachl orobenzene
Maleic Anhydride
Methyl Bromide
Methyl Chloride
Methyl Chloroform
Methyl Methacryl ate
Methylene Chloride
Naphthal ene
Ni trobenzene
Perchl oroethyl ene
Phenol
Phosgene
Phthal Ic Anhydricle
Propylene Oxide
Styrene
Tetrahydrofuran
Toluene
Toluene Diisocyanate
Freon 113
Trichioroethylene
Xylene
Vinylidene Chloride
Vinyl Chloride
Ethanal
Acrylic Aldehyde
Propenenitrile
3 -Chi oropropene
Anhydrous Ammonia
Benzol, Phenyl Hydride
3, 4-Benzpyrine
Aipha-Chiorotol uene
Div inyl, Biethylene
Tetrachioromethane
Ethyl Chloride, Hydrochloric Ether
Tr ichioromethane
2-Chloro-1 ,3-Butadiene
Cresylic Acid
1 ,4-Dichlorobenzene
Diethylene Dioxide, Dioxan, 1,4-Dioxane
Chioropropylene Oxide
I ,2-Dibromoethane
1 ,2-Dichloroethane
1 ,2-Epoxyethane
Methylene Oxide, Formalin
Perch 1 orobenzene
2, 5-Furaned ione
Bromome thane
Chi oromethane
1,1, 1-Trichloroethane
Methyl Ester, Pegalan
Dichioromethane
Naphthal in
Nitrobenzol
Tetrachioroethylene
Carbolic Acid
Carbonyl Chloride, Carbon
PAN
1,2-Epoxy Propane, Propene Oxide
Vinyl Benzene, Phenyl Ethylene
Diethylene Oxide
Toluol, Methyl Benzene
TOt, 2,4-Tolylene Oiisocyanate
1,1 ,2-Trichloro-1 , 2,2-Trifluoroethane
Ethylene Trichioride
Dimethyl Benzene
1, 1-Dichi oroethyl ene
Chioroethylene, Vinyl C. Monomer
based on NIUSH/USHA Pocket liulde to Lhemical Hazards, 19I , UHLW (NIUSH) No.
78-210 and EPA Chemical Activities Status Report, 1980 EPA-560/130-80-040(a).
C-2
-------
I
1 REPORT NO.
fl1(
: • TITLE AND SUBTITLE
This report contains technical materials that will assist state and local control
agencies in compiling air toxics emission inventories. It contains a aiscussion of
(1) various considerations that should be nade in planning and beginning an inventory
and (2) various tools an agency can use to locate potential sources and to estinate
emissions theref’rom. The appendices contain the data needed to use these inventory
tools and some example applications of them.
17. EY WORDS ANO OOCUMENT ANALYSIS
DESCRIPTORS
b. I0ENTIFIEMS,OPEN ENDED TERMS
C COSATI Fie ld/Group
Air Toxics
Emission Inventories
Toxic Emissions
Screening Inventory
Inventory Procedures
Hazardous Air Pollutants
18. DISTRIBUTION STATEMENT
19 sEcURITY CLASS fli.sReporr
21 NO OF PAGES
78
20 SECURITY CLASS (Th.s po e
22 PRICE
12.
TECHNICAL REPORT DkTA
(F? r, d In:uucuonz on the revene befo,e completing)
3 REC!PI€NTS ACCESSION NO
Compiling Air Toxics Emission Inventories
8 REPORT DATE
February 1990 (Revised)
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Dallas W. Safriet
U.S. Environmental Protection Agency
I. PERFORMING ORGANIZATION REPORT NO
I. PERFORMING ORGANIZATION NAME ANO ADDRESS
IC. PROGRAM ELEMENT NO
11 CONTRACT/GRANT NO.
12. SPONSORING AGENCY NAME AND ADORESS
Pollutant Characterization Section (MD—l5)
Noncriteria Pollutant Programs Branch
Air Quality Management Division
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOO COVERED
14. SPONSORING AGENCY COOE
15. SUPPLEMENTARY NOTES
16. M5 I IIMC 1
EPA Form 2220 — I (R.v. 4...77)
n viou O,rION S oe,oLEr
------- |