United States
Environmental Protection
Agency
Research and Development
c/EPA
Atmospheric Research and Exposure
Assessment Laboratory
Research Triangle Park, NC 27711
EPA/600/R-95/027
^-
March 1995
ERA'S Urban Area Source
Research Program
- A Status Report on Preliminary Research -
Principal Authors
Larry T. Cupitt
Atmospheric Research and Exposure Assessment Laboratory
Office of Research and Development
X
1la,L Cote
Health Effects Research Laboratory
Office of Research and Development
Joellen Lewtas
Health Effects Research Laboratory
Office of Research and Development
Thomas F. Lahre
Air Quality strategies and Standards Division
Office of Air Quality Planning and standards
Julian W. Jones
Air and Energy Engineering Research Laboratory
Office of Research and Development
Office of Research and Development
U.S. Environmental Protection Agency
401 Ml Street, S.W.
Washington, D.c. 20460
-------
Disclaimer
The information in this document has been funded wholly by the United States Environmental
Protection Agency. It has been subjected to the Agency's peer and administrative review, and it has
been approved for publication as an EPA document. Mention of trade names or commercial products
does not constitute endorsement or recommendation for use.
11
-------
Contents
Tables , iv
Figures v
Acronyms and Abbreviations vi
Executive Summary vii
Acknowledgements xi
Section 1. Introduction 1
Section 2. Hazardous Air Pollutant Assessment 3
2.1 Overview 3
2.2 Environmental Health Paradigm 3
2.2.1 Exposure Assessment 6
2.2.1.1 Emission Sources 6
2.2.1.2 Environmental Concentrations 9
2.2.1.3 Human Exposures 10
2.2.1.4 Complicating Factors 13
2.2.2 Effects Assessment 16
2.2.2.1 Internal Dose 16
2.2.2.2 Health Effects 17
2.2.2.3 Complicating Factors 19
Section 3. Previous Assessments 23
Section 4. Research Needs 29
4.1 Research on Exposure Assessment 29
4.2 Research on Effects Assessment 31
Section 5. Summary of Preliminary Findings 33
References 37
Appendix 41
Glossary 55
111
-------
Tables
Table E-l. The HAPs with the most extensive available data needed for evaluation
of the Environmental Health Paradigm ix
Table 2-1. Typical median ambient outdoor concentrations of some of the 189
listed HAPs. Concentrations are in micrograms per cubic meter (^ig/m3) of air. 12
Table 2-2. Occurrence and biological test results indicating carcinogenicity of air-
borne chemicals, for the 2,827 chemicals that have been reported to exist in
the air 22
Table 5-1. The HAPs with the most extensive available data needed for a risk
assessment 36
Table A-l. Availability of data on the 189 listed HAPs 45
Table A-2. Number of hazardous air pollutants that have been reported to produce
health effects in humans or animals by inhalation exposure 55
IV
-------
Figures
Figure E-l. The components of the Environmental Health Paradigm viii
Figure 2-1. The components of the Environmental Health Paradigm and their
relationship to Exposure Assessment and Effects Assessment 4
Figure 2-2. Summary of the available data on emissions of HAPs from all source
types. (Table A-l, Appendix A, categorizes the data for each of the 189
HAPs.) 8
Figure 2-3. Summary of available data on ambient outdoor concentrations of
HAPs. (Table A-l, Appendix A, categorizes the available data for each of the
189 HAPs.) 10
Figure 2-4. Summary of the number of HAPs that have been measured in a
variety of U.S. cities or towns 11
Figure 2-5. Effects of photochemical reactions on the mutagenicity of wood smoke
and auto exhaust, two common pollutant sources in populated areas. (Mutage-
nicity was measured using two different bacterial reversion assays.) 15
Figure 2-6. Evidence of carcinogenicity of the HAPs. (Table A-l, Appendix A,
categorizes the data for each of the 189 HAPs.) 18
Figure 2-7. Availability of validated Reference Concentrations (RfCs) for the 189
listed HAPs. RfCs are available for more chemicals, but several are grouped
under a single listed HAP. (See Table A-l for data on each chemical.) .... 19
Figure 3-1. Relative contribution of various hazardous air pollutants to the
estimate of nationwide cancer cases (from Cancer Risk From Outdoor Expo-
sure to Air Toxics) 24
Figure 3-2. Relative contribution by source to the estimate of nationwide cancer
cases per year caused by all sources, as reported in Cancer Risk From Outdoor
Exposure to Air Toxics 25
Figure 3-3. Results of a screening study to identify air pollutants with potential
noncancer health effects 26
Figure 5-1. Summary of the available data on the 189 listed HAPs. (Table A-l,
Appendix A, categorizes the data for each of the 189 HAPs.) 35
-------
Acronyms and Abbreviations
BTX Benzene, toluene, and xylene
CAA Clean Air Act
EPA Environmental Protection Agency
FIRE Factor Information Retrieval system
HAPs Hazardous Air Pollutants
L&E Locating and Estimating documents
IARC International Agency for Research on Cancer
LOAEL Lowest-Observed-Adverse-Effect Level
MACT Maximum Achievable Control Technology
PCBs Polychlorinated biphenyl compounds
PIC Products of Incomplete Combustion
POM Polycyclic Organic Matter
RfC Reference Concentrations
TEAM Total Exposure Assessment Methods
TSDFs Waste Treatment Storage and Disposal Facilities
VOCs Volatile Organic Compounds
VI
-------
Executive Summary
The Clean Air Act (CAA) Amendments of
1990 require the Environmental Protection Agen-
cy (EPA) to develop an "Area Source Program"
that includes both a Research Program and a
National Strategy to "substantially reduce the
public health risks posed by the release of hazard-
ous air pollutants from area sources ...." The
Research Program is to include three components:
(a) characterization of the sources of hazardous
air pollutants (HAPs), especially area sources, (b)
characterization of the concentrations of HAPs to
which people are exposed, and (c) consideration
of public health risks from the emitted and trans-
formed HAPs.
The Research Program is intended to support
development of the National Strategy. The Na-
tional Strategy must "identify not less than 30
hazardous air pollutants which, as the result of
emissions from area sources, present the greatest
threat to public health...." The National Strategy
must then propose a strategy to control the sourc-
es of the identified pollutants. The strategy must
also reduce the incidence of cancer attributable to
exposure to HAPs by 75% or more.
This report deals with the Research Program
and current research capability to characterize the
Emission Sources, the Exposure Concentrations,
and the Health Risks due to area source emissions
of HAPs. These three areas are discussed in terms
of the Environmental Health Paradigm. (See
Figure E-l.) This paradigm provides a conceptual
framework to describe both the three aspects of
the Research Program and the process of risk
assessment - risk management under the National
Strategy.
There are two primary activities hi the Envi-
ronmental Health Paradigm: exposure assessment
and effects assessment. Exposure Assessment
evaluates how likely people are to come into
contact with HAPs and determines how large their
exposure is likely to be. Effects Assessment iden-
tifies what health effects are likely to occur once
people are exposed to HAPs. In order to under-
stand environmental health issues, it is necessary
to have some knowledge about each component of
the paradigm.
The current status of information needed for
each of the components in the Environmental
Health Paradigm for HAPs is discussed. The
availability of data to assess the risks potentially
posed by each of the 189 HAPs listed in the
Clean Air Act was evaluated hi three broad cate-
gories: (1) characterization of area sources, (2)
characterization of exposure concentrations, and
(3) characterization of probable health effects.
The health effects data were characterized for
both non-cancer effects and cancer. In general, a
few HAPs in each category had a great deal of
data, while many chemicals had little or no data.
Twenty HAPs were found to have "Fair or
Better" data available in all three of the catego-
ries. (See Table E-l.) This list of chemicals does
not identify the 30 or more "worst" HAPs; rath-
er, the list simply identifies those HAPs with
sufficient data to begin a risk assessment of either
Vll
-------
Environmental Health Paradigm
EXPOSURE ASSESSMENT
Emission
Sources
Pollutant
Amount
Location
Environmental
Concentrations
Air
Water
Soil / Dust
Food
Human
Exposures
Route
Magnitude
Duration
Frequency
Internal
Dose
Absorbed
Dose
Target
Dose
Biomarkers
>
Health
Effect(s)
Cancer
Noncancer
- Symptoms
- Damage or
Disease
EFFECTS ASSESSMENT
Figure E-l. The components of the Environmental Health Paradigm.
the cancer or noncancer effects due to exposure to
that chemical. Another 20 HAPs are rated "Fair
or Better" in two of the three required areas.
Targeted research on this second group of HAPs
could readily provide sufficient data to allow a
risk assessment to be initiated. The 40 HAPs with
the most complete available data are listed in the
Table. The remaining 149 HAPs lacked important
data in two or more of the categories. In addition
to the 189 listed HAPs, other chemicals, such as
those produced by atmospheric transformation,
may also be of concern.
As a consequence of these data limitations,
risk estimates for many of the chemicals known to
be present hi urban environments will be very
uncertain. Research to overcome or address these
data limitations will likely be both expensive and
time-consuming. Data for selected chemicals,
however, appear sufficient to assess risks and to
develop control strategies as warranted.
Vlll
-------
Table E-l. The HAPs with the most extensive available data needed for evaluation of the Environmen-
tal Health Paradigm.
HAPs with data rated "Fair or Better" in
the three areas:
Source Emissions
Ambient Concentrations
and
Health Effects (Cancer or Noncancer)
HAPs with data rated "Fan- or Better" in
two of the following three areas:
Source Emissions
Ambient Concentrations
and
Health Effects (Cancer or Noncancer)
Benzene
1,3-Butadiene
Carbon tetrachloride
Chloroform
Ethylene dibromide
Ethylene dichloride
Formaldehyde
Methylene chloride
Styrene
Tetrachloroethylene
Toluene
Trichloroethylene
Vinyl chloride
Arsenic compounds
Chromium compounds
Lead compounds
Manganese compounds
Mercury compounds
Nickel compounds
Selenium compounds
Acetaldehyde
DDE (p,p'-dichlorodiphenyldichloro-
ethylene)
1,4-Dichlorobenzene
Ethylbenzene
Ethylene oxide
Hexachlorobenzene
Hexane
Methyl bromide
Methyl chloroform
Pentachlorophenol
Polychlorinated biphenyls
Propylene dichloride
2,3,7,8-Tetrachlorodibenzo-p-dioxin
2,4,6-Trichlorophenol
Vinylidene chloride
Xylenes (mixed isomers)
Antimony compounds
Beryllium compounds
Cadmium compounds
Polycyclic Organic Matter
IX
-------
-------
Acknowledgements
The authors would like to acknowledge and
recognize the efforts of a number of other contri-
butors to this report. Jeanette Wiltse (OHEA,
ORD) and Robert Fegley (OSPRE, ORD) played
key roles hi designing, drafting, and reviewing
this report. Cheryl Scott (OHEA, ORD) provided
important data on the health effects of the HAPs.
Anne Pope (OAQPS) provided critical information
about emission inventories and the FIRE system.
The authors are grateful to the individuals listed
below who provided reviews of the report. We
would especially like to recognize Blair Martin
for his insightful comments and suggestions.
Reviewers
G. Blair Martin, AEERL, ORD
Hal Zenick, HERL, ORD
Dale Pahl, AREAL, ORD
John Vandenberg, HERL, ORD
Susan Perlin, OHR, ORD
David Klefffman, OHR, ORD
David Guinnup, OAQPS
Don Theiler, STAPPA/ALAPCO
Benjamin Shaw, South Coast Air Quality Manage-
ment District
Jeffrey Myers, Department of Natural Resources,
State of Wisconsin
Naydene Maykut, Puget Sound Air Pollution
Control Agency
XI
-------
Xll
-------
Section 1
introduction
The purpose of this report is to summarize
what is currently known about exposures to and
risks from hazardous air pollutants (HAPs) that
are emitted by "area" sources. The Clean Air Act
(CAA) Amendments of 1990 require the Environ-
mental Protection Agency (EPA) to develop an
"Area Source Program" that includes both a
National Strategy and a research program. The
law also requires EPA to report the results of its
preliminary research efforts. This report describes
those preliminary research findings on area source
emissions.
Section 112(k) of the CAA1 mandates that
EPA conduct an area source research program
"after consultation with state and local air pollu-
tion control officials." The law specifies that the
research program should contain at least three
elements: (1) "ambient monitoring for a broad
range of hazardous air pollutants ... in a represen-
tative number of urban locations;" (2) "analysis to
characterize the sources" of hazardous air pollut-
ants (HAPs), with a focus on area sources and
then* public health risks; and (3) "consideration of
atmospheric transformation ... which can elevate
public health risks."
The mandated research program is intended to
provide the scientific basis for development of a
comprehensive National Strategy to control emis-
sions of HAPs from area sources. The National
Strategy must be published by November, 1995,
in a report to Congress. It must "identify not less
than 30" HAPs that "present the greatest threat to
public health in the largest number of urban ar-
eas." The strategy is to be fully implemented by
the year 2000 and must provide guidelines for
controlling the area source emissions of the 30 or
more identified HAPs, while simultaneously
ensuring a reduction of at least 75% in the "inci-
dence of cancer attributable to exposure to haz-
ardous air pollutants emitted by stationary sources
..., considering control of emissions of hazardous
air pollutants from all stationary sources and
resulting from measures implemented ... under
[the CAA] or other laws."
The area source National Strategy is a key
component of the Agency's overall approach to
reducing exposure to and risk from HAPs. It is
especially important because of the variety and
number of sources that might be controlled under
this strategy.
Traditionally, scientists and engineers have
associated "area sources" with small, but numer-
ous, sources that are likely to be found in any
urban area sources like gas stations, dry clean-
ers, auto repair shops, and even emissions from
cars and trucks. However, the definition of an
area source of HAPs in the CAA is different from
the traditional meaning of the term. The CAA de-
fines an "area source" as "any stationary source
of hazardous air pollutants that is not a major
source." In the CAA, a "major" source of HAPs
is "any stationary source ... that emits or has the
potential to emit considering controls, in the
aggregate, 10 tons per year or more of any haz-
-------
ardous air pollutant or 25 tons per year or more
of any combination of hazardous air pollutants."3
An "area source" of HAPs, as defined in the
CAA, therefore, is any stationary source of HAPs
that emits less than 10 tons per year of any single
HAP and less than 25 tons per year of all of the
HAPs emitted by that source.
Clearly, the definition of an "area source" of
HAPs in the CAA is somewhat different from the
traditional definition. Specifically, the definition
in the legislation excludes motor vehicles and
nonroad mobile sources (which are regulated else-
where in the Act), while it does include small
stationary sources, even though they may not be
"numerous" in an urban area.
The National Strategy must address area
sources as they are defined hi the CAA, rather
than the traditional definition. Throughout the
remainder of this document, the term "area
source" refers to the definition found in the CAA.
Other documents, some of which are cited in this
report, however, may use the traditional defini-
tion. Because the term "area source" may nave
different meanings hi different documents (espe-
cially those that date from prior to the CAA
Amendments of 1990), readers must be careful to
understand what is included as an area source
when evaluating other sources of information.
a Also note that the CAA defines a "major" source differently
when dealing with volatile organic compounds (VOCs),
pollutants that help produce ozone pollution. Throughout this
document, the term "major source" refers to a major source of
HAPs.
-------
Section 2
Hazardous Air Pollutant Assessment
2.1 Overview
Ambient air pollution can contribute to the
occurrence and/or aggravation of disease in urban
and/or industrialized areas. Diseases associated
with air pollution include respiratory diseases
(e.g., asthma, bronchitis, and emphysema) and
cancer.2-3> 4 EPA has conducted a number of
"screening" studies to begin to define the contri-
bution of HAPs to this problem in the U.S.b The
"screening" studies, which are discussed in Sec-
tion 3, were intended to make broad comparisons
of risks for program planning purposes. Such
studies typically attempted to define exposures
and risks from as many pollutants and sources as
possible, although most studies included only 10
or fewer of the HAPs listed in the CAA. Because
many assumptions about emissions, exposures,
and health effects were commonly made in these
studies, the results are generally viewed, at best,
as crude approximations of the comparative risks
posed to individuals and populations. While the
results, typically expressed in terms of cancer
risks or potential noncancer effects, are not
viewed as representing absolute risks, they pro-
vide the best available estimates of the potential
magnitude of the broad air toxics problem. Con-
gress clearly considered the results of such
screening studies to be relevant when legislating
the Section 112(k) area source program, as evi-
b Such studies have been conducted in Philadelphia, Baltimore,
Kanawha Valley (WV). Los Angeles, Chicago, Santa Clara
(CA), Baton Rouge, Phoenix, and a few other locations.
denced by the extensive citations from various
House and Senate Committee Reports containing
the legislative history of the Clean Air Act
Amendments.5
2.2 Environmental Health Paradigm
In order to assess the risks of HAPs, and to
manage or control those risks, it is often helpful
to consider the interrelated processes of exposure
and effects assessment in a conceptual framework,
or paradigm. Figure 2-1 illustrates one such para-
digm that is especially useful for describing what
is known about HAPs in urban air.6
Evaluation of potential health risks from expo-
sure to environmental pollutants is composed of
two primary activities that make up the Environ-
mental Health Paradigm: exposure assessment and
effects assessment. Exposure Assessment evaluates
how likely people are to come into contact with
HAPs and determines how large their exposure is
likely to be. Effects Assessment identifies what
health effects are likely to occur once people are
exposed to HAPs. In order to understand environ-
mental health issues, it is necessary to have some
knowledge about each component of the paradigm
from Emission Sources through Health Effects.
Never will EPA have perfect and complete
data about all aspects of the paradigm, yet critical
decisions about the National Strategy must be
-------
Environmental Health Paradigm
EXPOSURE ASSESSMENT
Emission
Sources
Pollutant
Amount
Location
Environmental
Concentrations
Air
Water
Soil / Dust
Food
4
>
Human
Exposures
Route
Magnitude
Duration
Frequency
»
Internal
Dose
Absorbed
Dose
Target
Dose
*
Biomarkers
Health
Effect(s)
» Cancer
» Noncancer
- Symptoms
- Damage or
Disease
EFFECTS ASSESSMENT
Figure 2-1. The components of the Environmental Health Paradigm and their relationship to Exposure
Assessment and Effects Assessment.
made. Often, assumptions about one or more
aspects of the Environmental Health Paradigm
must be made in order to fill in the data gaps. In
some situations, simplifying assumptions might
not significantly affect the risk assessment. For
other chemicals or locations, the need to make
such assumptions might introduce large uncer-
tainties into the assessment. The amount and
quality of information needed to evaluate properly
each component of the Environmental Health
Paradigm will vary from case to case and chemi-
cal to chemical.
To assess exposure thoroughly, one must
characterize the Emission Sources, Environmental
Concentrations, and Human Exposure factors.
Knowledge of Emission Sources is needed to
determine where, how much, and when HAPs are
emitted. Critical information includes the types
and amounts of pollutants released and the loca-
tions of the sources. Once the HAPs are emitted
into the air, they are transported and transformed
until some of them come into contact with hu-
mans. Information about Environmental Concen-
trations is necessary to determine the pollution
levels to which people might be exposed. For a
-------
comprehensive assessment, data are needed for all
media through which exposure might occur,
including air, water, soil, or food. The Human
Exposure factors consider how people and pollut-
ants come into contact with each other. The goal
of the human exposure factor is to define the
route, magnitude, duration, and frequency of the
contact between humans and HAPs. Exposure is
measured as the product of the pollutant concen-
tration and the time during which people are ex-
posed.
Human exposures to HAPs can occur through
a variety of routes, in addition to the air that
people breathe. Total exposure assessments in-
clude estimates for each route. HAPs can deposit
out of the air to a variety of surfaces, eventually
polluting water, soil, food, and objects around us.
Indirect exposures to HAPs can also occur from
the food and water people consume, and from the
objects that humans touch. Although such indirect
exposures can be extremely important in some
cases,0 this report will consider primarily expo-
sures through the air people breathe.
The intent of the final components of the
Environmental Health Paradigm is to identify the
health hazards associated with HAPs and to define
the relationships between exposure, target dose
(the dose to the affected organs or biological
systems), and health in human populations. This
is also known as the exposure-response relation-
c Other routes of exposure may be very important in many
cases. The Great Waters Program was authorized under
Section 112(m) of the Clean Air Act because of deposition of
toxic air pollutants to lakes and other bodies of water with
subsequent entry into the food chain or drinking water and
human exposure by ingestion. Recent National Academy of
Science reports on lead discuss human exposure by ingestion
of lead-containing particles deposited on food, as well as child
ingestion of lead-containing dust. (See, for example, National
Academy of Sciences, Measuring Lead Exposure in Infants,
Children, and Other Sensitive Populations, National Academy
Press, Washington, DC, 1993.)
ship. The overlap between Exposure Assessment
and Effects Assessment, as shown in Figure 2-1,
reflects the interrelationship of these two assess-
ment activities.
For a health effect to occur, HAPs in ambient
air first must actually get into the body. Internal
Dose defines how much of the HAPs that one
breathes (or ingests or contacts) actually gets into
the body (absorbed dose), and how much gets to
the specific organ(s) where they might cause
damage (target dose). Significant biologic events
resulting from this target dose can be used as
measures of internal dose (biomarkers). Absorbed
dose, target dose and resulting biomarkers are all
critical links between human exposure and conse-
quent health effects. Improving measures of these
links improves the estimates of risks posed by
HAPs.
Health Effects are often categorized into can-
cer and noncancer health effects. Historically, one
basis for this categorization of health effects is the
dichotomous nature of cancer (that is, either you
have it or you don't) versus the wider variety of
symptoms, damage, or disease associated with
noncancer effects. For example, respiratory disor-
ders resulting from exposure to HAPs can range
from itching noses, coughing, shortness of breath,
decreased capacity to inhale or exhale, bronchitis,
increased asthma attacks, emphysema, pulmonary
edema and death. More than one effect, like those
listed, can often appear together, in varying de-
grees of severity. Effects in different organs or
biological systems also can occur simultaneously.
Consequently, Effects Assessment must often
evaluate a complex set of health effects, with
different patterns of affected organ systems and
with widely different severity of effects. These
patterns are often chemical-specific and change
with exposure concentrations, durations, frequen-
cy of exposure, and with characteristics unique to
the population that is exposed (for example, ge-
netic or gender or age-related characteristics).
-------
Based primarily on laboratory animal studies and
occupational observations, the health effects most
commonly associated with HAPs exposures are
cancer, developmental and reproductive disorders
(for example, retarded development in children or
birth defects), neurotoxicity, and short-term and
long-term pulmonary disorders.7'8
The components of the Environmental Health
Paradigm also provide a reasonable way to sum-
marize the current understanding of HAPs in
urban air. The following discussions will focus on
the Exposure Assessment and the Effects Assess-
ment.
2.2.1 Exposure Assessment
The first component of the Environmental
Health Paradigm is Exposure Assessment. In this
section, we consider each of the components of
Exposure Assessment:
Emission Sources
Environmental Concentrations
Human Exposure.
2.2.1.1 Emission Sources
Reliable data on emissions of HAPs from area
sources are limited. Most previous studies of
emissions in urban areas have focused primarily
on criteria pollutants or their precursors, such as
volatile organic compounds (VOCs), paniculate
matter, sulfur oxides, and nitrogen oxides, not on
the 189 chemicals listed as HAPs. Furthermore,
previous studies focused primarily on all types of
sources (major point sources, mobile sources, and
area sources), not just area sources. Emissions
(e.g., tons of pollutant per year) from area sourc-
es may have not been included in such studies,
and even if they were included, the data may not
allow a complete and accurate emissions invento-
ry to be assembled.
Deficiencies in emissions data might involve
any of the various aspects of Emission Source
characterization, including describing the type of
pollutants, quantify ing how much of the HAP is
released, or locating the sources geographically.
Data available under the Toxic Release Inventory
are very useful in locating potential releases of
HAPs from many major sources, but similar data
are not available for the smaller area sources.
Nonetheless, some area sources of HAPs are well
defined, and a great deal of data are available for
area sources like residential wood combustion,
dry cleaners, and publicly-owned treatment
works. Aside from such sources, however, emis-
sion inventories have traditionally focused mostly
on major sources of VOC emissions (some of
which are also HAPs) or on sources of criteria
pollutants (for example, sulfur oxides, paniculate
matter, and nitrogen oxides). In many cases the
exact HAPs and the concentrations that are emit-
ted from small sources are not well known. In
many inventories, emissions from small area
sources are not located or measured precisely, but
are estimated from indirect measures like the
number of people in an area, the number of cars,
and the quantity of solvent sold.
Efforts are underway to reduce the uncertain-
ties in emissions inventories for a number of
important HAPs. EPA is continuing to develop
improved tools for use in developing HAP emis-
sion inventories. "Locating and Estimating"
("L&E") reports are available for more than 30
HAPs. These reports contain pollutant-specific
information on industrial processes, emission
factors (e.g., pounds of pollutant emitted per ton
of fuel burned), source test methods, and in the
recently updated reports, national inventories,
including emission estimates for point, area and
mobile sources. Thirteen "L&E" reports were
developed or upgraded in fiscal year 1993, and
seven additional updated reports are anticipated
for 1994. In addition, the Factor Information
REtrieval system (FIRE) contains evaluated emis-
-------
sion factors for both criteria pollutants and HAPs.
FIRE is updated periodically and now contains
9700 rated emission factors, of which approxi-
mately 4000 factors are available for 29 of the
listed HAPs.
Even when data on emissions are available,
there are still uncertainties involved in extrapolat-
ing the data to other locations or to other opera-
tional conditions. To ensure the development of
reliable emission factors, one must measure the
emissions at a variety of sources in a specific
category and must collect sufficient data on plant
operations, processes, and conditions. Obtaining
reliable emission factors is expensive even when
the source is not difficult to test and reliable
measurement techniques are available. Under
California's Assembly Bill 2588 program (the
"Hot Spots" Program), many producers of HAPs
are required to conduct such tests at their own
expense. EPA has used the data from California
to extract more than 1500 HAP emission factors,
and is implementing a project to obtain source test
data from other state and local agencies.
A number of state or local air pollution con-
trol agencies have voluntarily developed invento-
ries9 of HAP emission sources over the last de-
cade, despite the lack of a federal requirement for
HAP emission inventories. Without specific guid-
ance about such inventories, the state and local
agencies have chosen to include different HAPs hi
their inventories and to use a wide variety of
methods to estimate emissions. Consequently,
there is often very little consistency between the
available inventories. Nonetheless, efforts are
underway both at the federal and state levels to
overcome some of the shortcomings found in the
inventories and to reduce the inconsistencies. As
mentioned above, the California "Hot Spots" Pro-
gram has proven to be very productive in provid-
ing better emissions data. In addition, the eight
member states of the Great Lakes Commission are
working together to develop a regional emissions
inventory for mobile, area, and point source
emissions of 49 HAPs. Additional data on source
emissions should become available as states im-
plement the permit programs as required by the
Clean Air Act Amendments of 1990.
On a national scale, EPA has also supported
national HAP emission inventories for fourteen
HAPs (and related species) in 1993. These HAPs
included mercury, alkylated lead, hexachloroben-
zene, POM, polychlorinated biphenyls (PCBs),
tetrachlorodibenzodioxin, tetrachlorodibenzofuran,
benzene, 1,3-butadiene, carbon tetrachloride,
tetrachloroethylene, trichloroethylene, methylene
chloride, and formaldehyde. These national inven-
tories include estimates for mobile, area and point
sources and are allocated to the county level.
Although such inventories do not precisely locate
all sources of HAPs, they can still provide valu-
able information for estimating urban emissions of
HAPs.
Efforts to assemble emission inventories have
been identified for a variety of urban sources,
including area sources, for more than 60 HAPs,
but fewer than 20 of the HAPs appear with regu-
larity (that is, in 50% or more of studies) hi the
detailed inventories compiled by state and local
agencies.10 Emissions of other HAPs can be
estimated on the basis of national inventories, or
might be computed from available emission fac-
tors. Figure 2-2 illustrates the availability of emis-
sions data for the 189 listed HAPs. Forty-two
HAPs (seventeen HAPs that appear in 50% or
more of the state and local inventories, together
with an additional twenty-five HAPs that appear
in the FIRE data base or that are included in
national inventories) are categorized as "Fair or
Better." HAPs that appear infrequently (less than
50% of the time) in detailed inventories are listed
as "Occasionally Found." There are little or no
emissions data for more than 120 HAPs.
-------
Availability of Source Emissions Data
For the 189 Listed HAPs**
Little or No Information
66.7%
Fair or Better Data
22.2%
Occasionally Found
11.1%
** Based on frequency of inclusion in state or local inventories, data availability in the FIRE
data base, or the availability of a national inventory.
Figure 2-2. Summary of the available data on emissions of HAPs from all source types. (Table A-l,
Appendix A, categorizes the data for each of the 189 HAPs.)
Two major approaches can be used to identify
how much of urban pollution comes from the area
sources: dispersion modeling and source appor-
tionment.
If the emissions from all sources are well
known, the contribution from area sources to
ambient concentrations of HAPs can be estimated.
The estimates for area sources may then be com-
pared with the contributions from all other types
of sources, through dispersion modeling. Dis-
persion models describe how the emissions mix in
the atmosphere and are distributed throughout the
urban area. However, there are serious short-
comings in the current emission inventories for
urban areas with regard to area source emissions
of HAPs, as previously noted. These shortcom-
ings bring into question the reliability and accu-
racy of the dispersion modeling approach.
The second approach, source apportionment,
uses ambient monitoring data to estimate how
much of the pollution came from each source.
This approach works best when each source (or
source category) contributes substantially to the
total pollution in a unique and distinctive way.
8
-------
Such is not the case, however, for many sources.
For example, benzene, toluene, and xylene (often
referred to jointly as BTX) are frequently the
HAPs with the highest concentrations in urban
air. It would be very useful to know what fraction
of BTX in air was due to area sources. Two
recent apportionment studies11'12 found that
85-95% of the BTX in urban air came from
mobile sources. With such a large and dominant
source, apportionment of the small remaining
fraction of BTX from area sources will prove
very difficult to assign to specific area sources or
source categories.
While both dispersion modeling and source
apportionment methods have their limitations,
they can be used together to complement the
strengths and weaknesses of each approach.
It is important to understand the impact of
area sources on human exposure and risk, even if
their emissions are small compared to the total
quantities emitted by all sources. Exposure and
risk is not necessarily proportional to the magni-
tude of the emissions. This is especially true if
the area sources (and other sources, like indoor
sources or sources from personal habits) are much
more closely linked with human activities, be-
cause such sources could still dominate the result-
ing risk since they could contribute disproportion-
ally to human exposure.
2.2.1.2 Environmental Concentrations
The availability of data on ambient outdoor
concentrations of the 189 HAPs is highly uneven
(Figure 2-3). The ambient outdoor concentrations
result from emissions from all types of sources,
including point, area, and mobile sources. The
figure plots the total number of HAPs that have
been measured versus the number of times they
have been measured in outdoor air in populated
areas.13 There are little or no ambient measure-
ment data (fewer than 100 observations) for near-
ly two-thirds (112) of the HAPs, while a few
chemicals notably benzene, toluene, and the
three xylene isomers have each been measured
many thousands of times. For 71 of the 189
HAPs (38%). there are no ambient measurements
at all. The 43 HAPs with "Fair or Better" data all
have more than 1000 observations. (An "observa-
tion" is one or more measurements at the same
location within a 24 hour period.) Clearly, there
is little or no information about a large number of
HAPs, but a great deal of information about a
smaller number of HAPs.
The same conclusion (very little data for most
HAPs; considerable data for some HAPs) also
extends to the number of cities for which ambient
outdoor concentration data are available. Nearly
two-thirds of the listed HAPs have been measured
at fewer than 5 cities or towns, while BTX data
are available for more than one hundred cities.
Figure 2-4 illustrates just how few HAPs have
been measured at an adequate number of cities.
Additionally, the data are often available only for
short periods of time a few days or weeks
while special studies were underway. Long-term
collection of data on HAPs is available for only a
very few cities.
When two-thirds of the designated HAPs have
been measured only a few times and at only a few
cities, the "representativeness" of the ambient
outdoor data becomes an important issue. Even
the data that are available are of inconsistent
quality and duration. When large data gaps exist,
either in space or time, it is very difficult to
estimate human exposures and potential health
effects reliably, or to identify trends in order to
characterize the impacts of regulatory programs.
Table 2-1 lists typical outdoor concentrations
of a few HAPs13 that are among the best-studied
in terms of health effects. As discussed earlier,
the actual ambient outdoor measurements are
often variable; nevertheless, these concentrations
-------
Availability of Ambient Outdoor
Concentration Data For the 189 Listed HAPs
No Data
37.6%
Little Data
21.7%
Occasionally Observed
18.0%
Fair or Better Data
22.8%
The categories are based on the number of reported observations, as described in the text
HAPs with No data have 0 observations; HAPs with "Little Data" have <100 observations. The
"occasionally observed" HAPs have 100-1000 observations, and HAPs with "Fair or Better Data"
have been observed more than 1000 times.
Figure 2-3. Summary of available data on ambient outdoor concentrations of HAPs. (Table A-l,
Appendix A, categorizes the available data for each of the 189 HAPs.)
are typical of the reported data. Median concen-
trations, in micrograms per cubic meter (/ig/m3),
are listed in the table. The median is the middle
of the distribution of observed concentrations: half
of the time, the measured concentrations were
larger than those listed, and half of die time, die
concentrations were reported to be smaller. "Av-
erage" concentrations are not given since an
arithmetic average can sometimes be misleading,
especially if there are a few very large concentra-
tion measurements or if there are many observa-
tions with concentrations too small to measure
accurately. There are major differences between
the number of times and number of locations in
which the various chemicals have been measured.
2.2.1.3 Human Exposures
To develop the National Strategy to minimize
adverse health effects from area source emissions
of HAPs, it is necessary to consider the actual hu-
man exposure to die HAPs, not merely the ambi-
ent concentrations. The following text describes
what is currently known about the distribution of
HAPs across urban areas and about the impact of
outdoor air on indoor air and personal exposures.
10
-------
Cities and Towns with Ambient Outdoor Data
For the 189 Listed HAPs
Fewer than 5 Cities
60.3%
More than 50 Cities
12.7%
5 to 50 Cities
27.0%
Figure 2-4. Summary of the number of HAPs that have been measured in a variety of U.S. cities or
towns.
Distribution
To estimate human exposure to HAPs one
must know how widespread are the concentrations
of urban air pollutants. If area sources are uni-
formly and widely distributed across an urban
area, one would expect the concentrations of the
emissions to be relatively consistent across the
community, although this is not always true. Only
a few studies have included simultaneous mea-
surements of pollutants at different sites across an
urban area. One recent study, focusing on the
particle-bound pollutants from residential wood
burning and from automobiles, found the concen-
tration of fine particles from these sources to be
relatively consistent across an urban area.14 Oth-
er studies measuring gaseous HAPs and other -
vapor-phase pollutants have found that a few gas-
eous pollutants appear to have relatively constant
concentrations15 across distances as large as 10
km, implying that the sources of those pollutants
are widely and uniformly distributed throughout
the community.11 However, in the same study, a
larger group of gaseous pollutants were reason-
11
-------
Table 2-1. Typical median ambient outdoor concentrations of some of the 189 listed HAPS.
concentrations are in micrograms per cubic meter fc/g/m ) of air.
Chemical
Abstract
System
Number
71-43-2
108-88-8
95-47-6
108-38-3
106-42-3
75-09-2
67-66-3
106-99-0
71-55-6
56-23-5
50-00-0
Not applicable
57-74-9
Chemical Name
Benzene
Toluene
xylenes (o-, m-, p- Isomers)
Methyiene chloride
Chloroform
1,3-Butadiene
Methyl chloroform
Carbon Tetrachlorlde
Formaldehyde
Chromium compounds
Chlordane
Median
Concentration
fjg/ml
5
9
2 to 4
0.5
0.2
0.4
2
0.8
3
0.003
0.02
Number
of
Observations'
>8600
>6500
>5700
> 3400
> 4900
> 1900
> 4900
> 6300
> 2500
> 1800
> 345
Number of
Cities
with
Ambient
Data
172
159
> 130
86
135
66
155
149
75
> 192
> 8
An observation is one or more measurements taken within the same 24-hour day.
ably constant only across distances of about 1 km,
while still other pollutants were even more vari-
able.
Impact of Outdoor Air
People typically spend more than 80% of their
time indoors,16 so any analysis of the health ef-
fects from exposure to area sources must assess
the penetration of the area-source pollutants from
outdoors to indoors. Many of the volatile HAPs
are stable chemicals that do not react o^iickly with
other chemicals in the environment. Such stable
gaseous pollutants can easily penetrate indoors
with little or no loss of concentration. The instan-
taneous indoor and outdoor concentrations can be
different, however, due to delays caused by the
rate at which outdoor air enters the building
the air exchange rate.d
For time periods longer than a few hours, the
average indoor concentration of stable gaseous
pollutants generated by outdoor sources (including
area sources) is identical to the outdoor concentra-
tion adjacent to the house (for example, "on the
front porch.")17
Some HAPs (for example, most POMs) are
not volatile vapors; instead, these HAPs are at-
tached to small particles in the air that people
breathe. Non-volatile HAPs that are emitted by
chemical or combustion processes are often bound
d Conversely, reactive pollutants, for example ozone, are
readily destroyed as they penetrate indoors resulting in indoor
concentrations that are generally less than outdoor concentra-
tions. Few of titie listed HAPs are expected to be so reactive.
12
-------
to "fine" particles (less than 2.5 micrometers in
diameter). Such particles are only partially re-
moved as the air penetrates indoors. The number
of particles that successfully penetrate indoors is
roughly proportional to the air exchange rate.
Thus, the more air brought indoors, the more the
concentration of particles in the indoor air is like
the concentration of particles outdoors. For many
buildings, the air exchange rate is large enough to
permit about 50%-90% of the outdoor fine parti-
cles to penetrate indoors successfully. Non-vola-
tile HAPs found on particles that are generated by
mechanical processes (like dust kicked-up by
automotive traffic, wind-blown dust, and con-
struction projects) are usually bound to larger
particles that are much less likely to penetrate in-
doors.17
Finally, indoor sources, workplace sources,
and personal activities can provide additional
exposures to HAPs, beyond those due to the out-
door sources. The outdoor sources provide a
baseline of exposure to HAPs, on top of which
indoor sources, workplace sources, and personal
activities add additional exposures. If such indoor,
workplace, or personal sources are large, they can
dominate the total exposure calculation for those
exposed individuals. These sources must be taken
into account when determining the total human
exposure to HAPs.
2.2.1.4 Complicating Factors
There are a number of factors that make
Exposure Assessment a difficult and complex
task. Two factors that make identifying and char-
acterizing the urban area sources of HAPs diffi-
cult are: the complexity of urban air pollution,
and uncertainty in defining area sources.
Urban air is a complex mixture of thousands
of chemicals. These chemicals come from a wide
variety of sources, including major point sources,
area sources, mobile sources, and natural sources.
Examples of natural sources of HAPs include
forest fires, plant decay, and weathering of miner-
als containing heavy metals. The objective of the
urban area source research program is to charac-
terize the exposures and health risks due to area
sources in support of the mandated National
Strategy. But once the pollutants from the area
sources have mixed with those from major point
sources, mobile sources, and natural sources, it is
extremely difficult to identify how much of a
specific pollutant came from just the area sources.
Even the definition of an area source under
Section 112 adds a complicating factor. For pur-
poses of the HAP National Strategy, area sources
also include point sources that do not meet the
requirements to be classified as major sources.
These "non-major point" sources have not tradi-
tionally been considered as area sources, and
were not previously included in efforts to charac-
terize area sources. "Major" sources are defined
as part of the Maximum Achievable Control
Technology (MACT) standard setting process
under Section 112(d): sources that do not meet the
requirements for MACT standards are by default
"non-major point" sources, or area sources.
(Some source categories that include individual
sources that are likely not to meet the definition
of a major source are: bulk liquid (e.g., gasoline)
terminals, electric arc furnaces/stainless steel
mini-mills, wood furniture manufacturing, second-
ary lead smelters, etc.) The final MACT stan-
dards are not scheduled for promulgation until
November 2000. Additional area sources might be
added for consideration, long after the National
Strategy has had to go into effect.
Other factors make characterization of ambient
outdoor concentrations of HAPs a difficult under-
taking. For example, measurement methods are
not available for many HAPs, and natural reac-
tions in the atmosphere can either destroy or
produce HAPs.
13
-------
Measurement Methods
One reason for the lack of data on both emis-
sions and environmental concentrations of many
of the HAPs is that there are often no reliable
methods to collect and measure these chemicals.
Measurements at the source and in ambient air are
often made under distinctive conditions that make
such measurements difficult. For example, source
measurements often have high concentrations of
contaminants and harsh conditions that make
sampling and analysis difficult: ambient samples
contain very small amounts of the species of
interest and must be concentrated to be detected
reliably. Validated source sampling methods exist
for only 87 of the HAPs. In ambient air, there is
one group of HAPs where there is a particularly
noteworthy lack of data. These compounds, nitro-
genated or oxygenated organics, are often referred
to as "polar" organics, and they comprise 89 of
the 189 HAPs. Only about one-third of these
polar organics have actually been measured in
ambient ah-.
Atmospheric Transformation
Another difficulty with evaluating HAPs in
urban air is atmospheric transformation. Natural
atmospheric events cause chemical reactions that
can both destroy and create HAPs. These trans-
formation processes will eventually break down
and remove some of the HAPs from the ah".
Conversely, transformation processes might con-
vert non-hazardous pollutants into dangerous
products (or even transform HAPs into products
that are more hazardous than the original HAPs.)
The HAPs formaldehyde, acetaldehyde, acetone,
and acrolein, for example, are all produced hi
significant quantities in urban air18 by the atmo-
spheric transformation of many organic com-
pounds including many compounds not on the
list of HAPs. In other words, transformation pro-
cesses can produce a HAP even when one was not
emitted. This is similar to the situation with
ground-level ozone, which is produced primarily
through transformation of other pollutants, even
though it is not directly emitted.
Many of the most important of these atmo-
spheric transformation processes involve sunlight.
Sunlight, shining on polluted urban air, sets into
motion a complex series of chemical reactions
that convert the directly emitted pollutants into an
even more complex "soup." It is not possible to
identify all of the chemicals in the resulting prod-
uct mixture, but studies over the last decade
suggest that the sunlight-transformed mixture
might be even more hazardous than the originally
emitted pollutants. As an indicator of this poten-
tial for increased hazard, the bacterial mutagenici-
ty the ability to cause changes hi the genetic
material of bacteria of the transformed mixture
is often much greater than that of the original
pollutants. This increase in mutagenicity is espe-
cially true for the gaseous products, which are
likely to be the partially-oxygenated or -nitrogen-
ated transformation products of the emitted chemi-
cals. Figure 2-5 shows the dramatic increases in
bacterial mutagenicity brought about by sunlight
in two complex pollutant mixtures that are often
found hi urban air, namely wood smoke and auto-
mobile exhaust.19'20
The data hi Figure 2-5 are from smog cham-
ber simulations of atmospheric reactions, but at
concentrations higher than those normally found
in the environment: such simulations are neces-
sary, since the mutagenicity tests are not suffi-
ciently sensitive to measure such changes in actual
urban air. Indirect evidence, however, suggests
such transformation effects do occur in ambient
outdoor urban air. A variety of simulations by re-
searchers around the world, involving many of
the pollutants commonly found in urban air, have
demonstrated several important facts about the
mutagenic products of atmospheric transforma-
tion:
14
-------
o
5
UJ
o
o
o
Effect of Atmospheric Transformation
Increases in Bacterial Mutagenicity During Chamber Studies
b 20000
o
LU
<3
Legend
After Transformation
Before Transformation
Gas Phase Particulate Phase
Wood Smoke
Gas Phase Particulate Phase
Auto Exhaust
Figure 2-5. Effects of photochemical reactions on the mutagenicity of wood smoke and auto exhaust,
two common pollutant sources in populated areas. (Mutagenicity was measured using two different
bacterial reversion assays.)
Sunlight transforms many, but not all, urban
pollutants into both gaseous and particle-bound
mutagenic' products.
The gaseous mutagenic transformation prod-
ucts are persistent: in the laboratory simula-
tions, they are stable in the air for hours after
they are produced. If they are produced and
are stable under ambient conditions, then
exposures can occur over large areas and for
long times.
About 90% (by mass) of organic chemicals in
urban air are gaseous, with only about 10%
bound to particles. In the laboratory simula-
tions, the total mutagenicity of the gaseous
transformation products in the air greatly
exceeded the total mutagenicity of the particle-
bound products in the same volume of air.
The relative risk from gaseous mutagens
versus particle-bound mutagens is unknown.
These data on mutagenicity taken together
cause concern about the potential impact of atmo-
15
-------
spheric transformation on cancer risks in urban
areas. If transformation of non-hazardous air
pollutants can cause a substantial cancer risk hi
urban areas, it will make it difficult to develop a
National Strategy that can reduce cancer risks by
75%, as required by law.
Exposure Variabilities
One factor that complicates efforts to estimate
human exposure is the fact that people and air
pollutants move around throughout the day. What
people do, and where they are, and when they are
at a specific location all affect their exposure.
Available exposure or concentration data often do
not describe well the extremes in exposure, either
very large or very small exposures. People who
live very close to a source (for example, in an
apartment above a dry cleaning business, or near
industrial or gasoline-handling facilities) can be
exposed to abnormally high concentrations of
specific HAPs. In addition, both people and air
pollutants move about during the day. As people
move in and out of polluted areas, their exposures
can change. Time-activity patterns are descrip-
tions of: 1) where people are throughout the day,
2) how long they remain in each location, and 3)
what activities they are doing that can influence
exposure (for example, jogging in a park will
cause a person to inhale more air and more pol-
lutants than will reading a book on a bench in the
same park). Clearly, where a person is during the
day and what be or she is doing can significantly
affect that person's exposure to HAPs. Only with
information on the time-activity patterns of the
population relative to the sources of HAPs is it
possible to characterize accurately the exposures
of people at the high end of the range of expo-
sures the very people who are most likely to be
at risk. Some studies, like the Total Exposure
Assessment Methodology (TEAM) studies21-22
or the planned National Human Exposure Assess-
ment Survey (NHEXAS),23 have a statistical
approach that is designed to measure a wide range
of exposures. Such studies are extending the
understanding of the range of potential human
exposures, but such statistically based studies are
very expensive to conduct and difficult to analyze.
2.2.2 Effects Assessment
The second major aspect of the Environmental
Health Paradigm is Effects Assessment. Effects
Assessment is concerned with what happens to
human health once someone is exposed to HAPs.
There are three components of Effects Assess-
ment: Human Exposure, Internal Dose, and
Health Effects. Since Human Exposure is also a
part of Exposure Assessment and was described
earlier, the following describes the remaining two
components of Effects Assessment:
Internal Dose
Health Effect(s).
2.2.2.1 internal Dose
The term "Internal Dose" is often used to
convey a variety of concepts. In the current con-
text it means the estimation of the amount of HAP
that enters the body and reaches an organ or
system where it might cause damage to human
health. Ambient air concentrations of HAPs have
often been used as surrogates for Internal Dose.
However, this practice can result in either over-
or tinder-estimations of risk. Ambient concentra-
tions are not always reliable indicators of internal
dose because biological and biochemical process-
es, such as absorption into the body, distribution
in the body, metabolism, and excretion, all affect
how much of the HAP concentration in the air
actually reaches the organs or physiological sys-
tems where the pollutants might cause damage.
For particle-bound HAPs, even the physical char-
acteristics of the pollution may be important.
Particle size and the nature of the particles on
which the HAPs are carried may strongly influ-
ence the location in the body where the HAPs are
deposited, the mechanism by which adverse ef-
16
-------
fects may occur, the distribution of the pollutant
within the body, and the internal persistence of
the pollutants. It is important, therefore, to esti-
mate Internal Dose as precisely as possible. The
more accurate this estimation, the more accurate
will be the assessment of potential HAP risks.
The use of Internal Dose is particularly valu-
able when human risk estimates are derived from
animal laboratory experiments or occupational
studies. (HAP risk assessments are almost always
derived from these types of data [see discussion
of extrapolation of health effects data in the dis-
cussion of complicating factors that follows]).
New techniques are now being developed that
allow for better estimates of Internal Dose. Some
of these techniques are: measurements of biologi-
cal and biochemical processes (pharmacokinetics);
use of alternative and more relevant surrogates
(biomarkers) of Internal Dose; and actual mea-
surement of the HAPs at the affected tissue (mo-
lecular dosimetry). Scientific groups such as the
National Academy of Sciences and EPA's Science
Advisory Board have encouraged the use of im-
proved estimates of Internal Dose hi risk assess-
ments. Unfortunately, reliable information on
Internal Dose is currently available for only a few
HAPs, and development of such information is
currently expensive, slow, and laborious. Through
experience with available methods, and through
research to improve methodology, the costs to
obtain better estimates of Internal Dose will,
undoubtedly, decline over tune, and unproved
estimates will become more and more available.
2.2.2.2 Health Effects
There are some toxicity data available for
each of the 189 HAPs. In almost no case, howev-
er, are data available on all of the most important
health effects: cancer, developmental and repro-
ductive disorders (birth defects), neurotoxicity,
and acute (short-term) and chronic (long-term)
pulmonary effects. Moreover, the quality of the
available studies is highly variable. Some studies
are barely adequate, others are excellent. Another
problem is the lack of data on toxicity associated
with exposure by inhalation. Much of data on
health effects comes from tests involving only
ingestion of the HAP (commonly called oral
exposure). However, it is known that differences
hi the route of exposure can produce major differ-
ences hi the character and extent of toxicity.
Relying on only ingestion data alone generally
results hi large uncertainties hi the prediction of
health effects.
Cancer
A serious possible health effect of HAPs is
their potential to cause cancer. More than 100 of
the 189 HAPs have sufficient data to assess their
ability to cause cancer qualitatively:24 even for
these chemicals, however, a quantitative estimate
of the dose-response relationship (potency) is not
always possible. Chemicals are classified based on
a variety of factors such as the quality of the
studies, the number of studies, and the species
reported to have chemically induced cancer. Both
human and animal data are considered. Eighty-
three of the listed HAPs are considered to be
"probable" or known human carcinogens. An-
other 25 HAPs are considered "possible" human
carcinogens. (See the classification definitions in
the glossary. N.B., EPA is revising its guidelines
for carcinogen risk assessment and the definitions
are expected to change.) Twenty-two of the HAPs
lack sufficient data for a classification, while the
remaining 59 of the HAPs have not been evaluat-
ed for carcinogenicity.e The carcinogenicity data
are illustrated hi Figure 2-6.
Noncancer Effects
For noncancer health effects, a Reference
Concentration (RfC) is used to estimate an expo-
sure concentration that is not harmful. An RfC is
an estimate, based on a single critical effect, of
17
-------
Availability of Carcinogenicity Data
For the 189 Listed HAPs
Unknown Carcinogenicity
11.6%
Possible Carcinogen
13.2%
Unevaluated HAPs
31.2%
Probable or Known Carcinogen
43.9%
(ARC carcinogens and Class A & B carcinogens are listed as "Probable or Known." Class C
chemicals are listed as "Possible." Class D chemicals are classified as "Unknown." The remaining
chemicals have not been evaluated.
Figure 2-6. Evidence of Carcinogenicity of the HAPs. (Table A-l, Appendix A, categorizes the data
for each of the 189 HAPs.)
the concentration (with uncertainty spanning a
factor of 10) that could be inhaled for a lifetime
c Some of the listed HAPs are actually groups of compounds,
and data may exist on several different chemicals within a
single HAP definition. For example, nickel compounds are
generally considered to be "possible" carcinogens, but nickel
subsulfide and nickel refinery dust are "known" human carcin-
ogens. In such case, the HAP (nickel compounds) was classi-
fied at the higher risk level ("Probable or Known") for tabula-
tion in this report. Similarly, Polycyclic Organic Matter
(POM) was classified as a "probable" carcinogen on the basis
of some specific compounds (for example, benzo(a)pyrene)
that often occur in POM.
with no adverse health effects. Only 40 of the 189
HAPs have sufficient data to support estimation of
an RfC. Confidence levels for an RfC vary from
unknown to low to high as illustrated in Figure
2-7. Only five of the RfCs have "high" confi-
dence. Most of the 149 HAPs without a validated
RfC have not been studied for chronic inhalation
effects at all.
Some of the chemicals on the list of 189
HAPs are also of concern to EPA because of their
potential to cause serious, immediate health ef-
fects if people are exposed to very large concen-
18
-------
Availability of Noncancer Effects Data
For the 189 Listed HAPs
No validated RfC
78.8%
Moderate or High Confidence in RfC
14.3%
Low Confidence in RfC
6.9%
Based on the level of confidence placed on the calculated Reference Concentration (RfC), the
concentration that could safely be inhaled for a lifetime with no harmful noncancer health effect
Figure 2-7. Availability of validated Reference Concentrations (RfCs) for the 189 listed HAPs. RfCs
are available for more chemicals, but several are grouped under a single listed HAP. (See Table A-l
for data on each chemical.)
trations. Exposures to large concentrations might
occur, for example, after an industrial accident
that releases large quantities of a chemical. An
accepted method for describing the relationship
between dose of pollutants and the biological
effects (the dose-response) due to large, short-
term exposures to these chemicals is currently
under development by EPA. Data on short-term
(acute) effects are critical to EPA's Accidental
Release Program which is also mandated in the
Clean Air Act.
2.2.2.3 Complicating Factors
Several factors make Effects Assessment
evaluation of health effects from exposures to
HAPs in urban air a very difficult task. Three
complicating factors are discussed below: (1)
extrapolation of health effects data, (2) exposure
to complex mixtures of environmental pollutants,
and (3) the fact that chemicals other than those on
the list of 189 HAPs can pose hazards to human
health.
19
-------
Extrapolation
Most of the HAPs data available must be
interpreted in some manner in order to assess
public health risks. The most common types of
interpretations involve the following: 1) using
animal effects data to predict effects that might
occur in humans; 2) using effects data collected at
relatively high exposure concentrations to predict
effects that might occur at lower exposure concen-
trations; and 3) using effects data collected with
certain exposure durations and patterns to predict
the effects that might occur with different expo-
sure durations and patterns. These interpretations
are often called animal-to-human, high-to-low
dose, and across-exposure-scenario extrapolations.
These types of extrapolations are often difficult to
perform with a great degree of certainty. Limited
data on which to base extrapolations increase the
uncertainty.
Biological or biochemical processes might
differ between laboratory animals and humans.
Consequently, responses to the same ambient
exposures can also differ. Similarly, biological
and biochemical processes in healthy adult male
workers can differ from important segments of
the general population, such as children and the
elderly. Also, exposure concentrations in animal
experiments and occupational studies are likely to
be higher than environmental exposures. Exposure
durations and patterns are also often different.
These disparities can differentially affect biolog-
ical and biochemical processes, and consequently,
Internal Dose. With careful study and estimation
of Internal Dose, many of these differences can
be understood and quantified.
Many of the uncertainties in risk assessment
are unavoidable, given the current state of knowl-
edge and the need to assess public health risks
from HAPs. Particular types of scientific informa-
tion, however, can improve analyses and reduce
some uncertainties. In particular, reduction in
uncertainties can occur via better estimation of
dose to the affected organ (through such methods
as evaluation of pharmacokinetics, biomarkers,
and molecular dosimetry), and understanding what
causes HAPs to have a toxic effect (the mecha-
nisms of action). The size of the effort that will
be required to gather these types of data, for even
just the most important HAPs, is substantial.
Complex Mixtures
Complex mixtures confound the evaluation of
the HAP problem in urban air. Urban air is a
mixture of many pollutants, and little is known
about the effects of exposure to mixtures of chem-
icals. Usually, effects assessments deal with only
one chemical at a time. Sometimes, however, the
effects of simple mixtures are assessed by adding
together the anticipated effects from exposure to
each individual compound. This additivity ap-
proach is normally only used when the anticipated
effects are similar for the various chemicals in the
mixture. When dealing with complex mixtures
(like those found hi urban air) and with many
different potential health effects, scientists are
reluctant simply to add together all of the antici-
pated individual health effects. They are reluctant
because the interactions of mixtures on health are
not well understood. Because of the complexity of
the interactions, the total effect of the mixture
might be very different than the simple sum of the
individual effects. Additional research is currently
being conducted to develop methods that will
allow assessment of the effects of exposure to
complex mixtures.
Chemicals Not on the List of 189 HAPs
Another important uncertainty in evaluating
urban air is that chemicals, other than the 189
listed HAPS, might be shown to be more impor-
tant air pollutants in the future. Thousands of
individual chemicals, representing almost every
known chemical class, are expected to be present
20
-------
in urban air. There is recent evidence from stud-
ies of complex mixtures of urban air particles and
gases that the major contributors to the mutage-
nicity of urban air are chemicals that have not yet
been identified. These as yet unidentified chemi-
cals might be produced by atmospheric transfor-
mation of organic pollutants emitted by a variety
of sources. Specifically, new bioassay-directed
chemical identification techniques have identified
polar organic chemicals (for example, hydroxyla-
ted- and nitrated-aromatic hydrocarbons) in urban
air that appear to arise from atmospheric transfor-
consider the wide range of toxic effects needed
for a full evaluation of the potential hazard. The
development of data on this broad range of sub-
stances is almost certainly not warranted. Further
analysis is needed to target specific chemicals for
further evaluation.
mation
25
Table 2-2 categorizes the almost 3000 chemi-
cals that have been detected in ambient air: it also
notes the number of chemicals in each category
that have been evaluated in cancer biological
assays and the number that have been found to be
carcinogenic.26 Several important points can be
seen from this table: 1) only 10% of the chemi-
cals detected hi ah" have been screened in short-
term genotoxic tests for their ability to cause
cancer; 2) of the approximately 300 chemicals
that have been screened, roughly 22% were found
to be carcinogenic hi the laboratory animal stud-
ies^ and 3) most evaluation has been focused on a
few pollutant categories.8 Consequently the
contribution of many categories of chemicals as
airborne carcinogens cannot be estimated. Fur-
ther, it should be noted that this analysis does not
f This percentage(of positive cancer results must be interpreted
with caution. Candidates for carcinogenicity testing often can
be identified based on short-term mutagenic assays or other
assays that detect genetic changes. Consequently, the chemicals
selected for long-term cancer bioassays are more likely to be
positive than randomly selected chemicals.
8 Some categories of chemicals (for example, hydrocarbons,
nitrogen-containing organics and halogenated organics) are
relatively well tested. Other categories of chemicals, like
ketones and carboxylic acids and their derivatives, are com-
monly detected in ambient air but have not been extensively
evaluated.
21
-------
Table 2-2. occurrence and biological test results indicating carcinogenicity of airborne
for the 2,827 chemicals that have been reported to exist in the air.
Category
inorganics
Hydrocarbons
Ethers
Alcohols
Ketones
Aldehydes
Carboxylic Acid Derivatives
Carboxyllc Acids
Heterocycllc Oxygen compounds
Nitrogen-containing Organics
Sulfur-
Containing Organics
Halogen-Containing Organics
Organometallic Compounds
GRAND TOTALS
Number of
Air Pollutants
identified
in Each
Category
260
729
44
233
227
108
219
174
93
384
99
216
41
2,827
Number of
Pollutants
that have
Been Screened
For Genotoxic
Effects3
30
51
3
28
11
6
6
5
16
59
4
71
13
503
Number of
Pollutants that
have been
Found Positive
in Genotoxic
Tests
5
12
1
1
0
4
0
0
4
22
1
16
6
72
chemicals,
Number of
Chemicals
Found to
Cause
Cancer In
Laboratory
Animals 6
4
19
0
0
0
1
2
0
7
12
1
21
0
67
a Short-term mutagenic or other genotoxic tests.
b Does not Include all human carcinogens.
Data are compiled from craedel, Hawkins and Claxton. Atmospheric Chemical
Compounds: Sources. Occurrence, and Bioassay, Academic Press, inc.. Orlando,
FL 1986.
22
-------
Section 3
Previous Assessments
The results from many previous screening
studies have been compiled and presented in a
1990 report entitled Cancer Risk From Outdoor
Exposure to Air Toxics27. This report provides a
"snapshot" of the current understanding of the air
toxics problem. It included emissions from all
source types, not just area sources, including
motor vehicle emissions. The report estimates that
exposure to hazardous air pollutants from all
source types accounts for as many as 1000-3000
cancer deaths each year in the U.S.h
Figures 3-1 and 3-2, adapted from the Cancer
Risk From Outdoor Exposure to Air Toxics report,
show the HAPs and source categories associated
with the estimated HAPs-related cancer risks,
respectively. Figure 3-1 shows that products of
incomplete combustion (PIC), 1,3-butadiene,
hexavalent chromium, and benzene account for
more than half of the overall cancer risk among
the pollutants evaluated.'28 ("PIC" refers to a
group of chemicals generated when fuels are only
partially burned. PIC includes the HAP listed as
polycyclic organic matter, or POM.) Results from
such screening studies suggest that a handful of
source categories such as motor vehicles,
chrome elecrroplaters, waste treatment storage
b Please note the use of the term "as many as." The risk
factors used to derive the estimates of possible cancer deaths in
the cited report are "upper bound estimates." Such estimates
are highly uncertain. The actual human risks are not known
and are expected to be lower than the "upper bound" estimates
used in the report.
and disposal facilities (TSDFs), woodstoves and
fireplaces, asbestos demolition, and gasoline mar-
keting account for a majority of HAPs-related
cancer risks (see Figure 3-2) in these screening
studies. Lifetime cancer risk to individuals living
in urban areas, aside from those risks obviously
associated with major sources of HAPs, typically
range from 1 in 100,000 (10'5) to 1 in 1000
(10"3). These figures demonstrate the relative
importance of controlling non-major sources of
HAPs in urban areas.
Other cancer screening studies not covered hi
Cancer Risk From Outdoor Exposure to Air Toxics
generally suggest similar results; while data are
available for only a small number of sources and
pollutants, a relatively small subset of these gen-
erally account for most of the currently estimated
HAPs-related cancer risk. The comparative rank-
ings of sources and pollutants in each study vary,
depending on what cities, sources, and pollutants
are included in the analysis, and on methodologi-
cal differences in the risk assessments.
1 The actual risk estimates will change as new and better data
are obtained. Indeed, a recent update (Motor Vehicle-Related
Air Toxics Study, EPA 420-R-93-005, April, 1993) of mobile
source risks suggests that the relative roles of PIC and 1,3-
butadiene may be reversed. This assessment found the risk
from PIC from all urban sources may be less than that shown
and the risk from 1,3-butadiene may be greater than that found
in the Cancer Risk From Outdoor Exposure to Air Toxics
report and discussed in this section.
23
-------
Relative Contribution by Pollutant
To Total Nationwide Cancer Cases
I Others
[Vinyl chloride
Carbon Tetrachloride
Ethylene dichloride
[Gasoline vapors
Dioxins
Ethylene dibromide
Arsenic
Asbestos
I Chloroform
Formaldehyde
Benzene
Chromium, hexavalent
1,3-Butadiene
15 20 25
Percent Contribution (%)
Other Pollutants
Acrylonitrile
Cadmium
Vinylidene chloride
Hexachlorobutadiene
Trichloroethylene
Coke oven emissions
Perchloroethylene
Hydrazine
Ethylene oxide
Methylene chloride
Radon
Other Radionuclides
56 other pollutants
PIC
30
35
40
Figure 3-1. Relative contribution of various hazardous air pollutants to the estimate of nationwide
cancer cases (from Cancer Risk From Outdoor Exposure to Air Toxics).
Very few screening studies have examined
health effects other than cancer. One such effort,
however, found that noncancer effects8'29 would
likely be expected to occur in exposed urban
populations.
The study attempted to estimate the potential
noncancer effects of urban air pollutants, not just
the listed HAPs. The study considered pollutants
from all types of sources (not just area sources).
Outdoor air monitoring data or computer-modeled
estimates of ambient outdoor concentrations were
used to examine potential exposures to air pollut-
ants. Monitoring or modeling estimates of ambi-
ent outdoor concentrations were available for 334
air pollutants.J
j The average annual concentrations of 40 chemicals were
modeled, based on estimated emissions data that were provided
by more man 3500 individual commercial and industrial facili-
ties across the U.S. Measured outdoor concentrations, of
varying reliability and completeness, were available for more
man 300 volatile organic chemicals at more than 1000 sites in
310 cities, and for 6 trace metals in more man two million
samples from more than 1500 U.S. cities.
-------
Relative Contribution by Source Categories
To Total Estimated Nationwide Cancer Cases
Treatment, Storage & Disposal Facilities
Woodsmoke
Asbestos, demolition
Solvent use/degreasing
Gasoline marketing-
Other area sources
Secondary formaldehyde, area-
Secondary formaldehyde, point
Chemical users/producers
Iron and steel
Cooling towers
Coal & oil combustion
Electroplating
Motor vehicles
Other point sources
Figure 3-2. Relative contribution by source to the estimate of nationwide cancer cases per year caused
by all sources, as reported in Cancer Risk From Outdoor Exposure to Air Toxics.
Of the 334 air pollutants with estimated ambi-
ent outdoor concentrations, information on poten-
tial noncancer health effects were available for
143 chemicals. For these pollutants, the estimated
outdoor concentrations were compared to the
lowest-observed-adverse-effect level (LOAEL)
and to a health reference level.k Concentrations of
54 of the 143 pollutants exceeded the health refer-
ence level at one or more sites; more than 20
pollutants exceeded the health reference levels at
more than 25% of the sites studied. Figure 3-3
shows the number of chemicals that exceeded
these levels. The data are grouped according to
whether the noncancer health effect was acute or
chronic, and whether the estimated concentration
k A LOAEL is the lowest dose or exposure level at which an
adverse effect has been reported in the health literature,
typically from studies conducted in laboratory animals A
health reference level is die LOAEL divided by appropriate
uncertainty factors to account for intra- and inter-species
variability. The goal is to establish an exposure level below
which die population is not expected to be affected at some
unspecified level of frequency (risk). Health reference levels
differ from Reference Concentrations (RfCs), which will be
discussed later in this document, in that health reference levels
receive much less review and validation than do RfCs.
25
-------
Nationwide Screening Study
Number of Individual Air Pollutants Exceeding Noncancer Health Levels
Chronic Modeling
(Includes thirty one
chemicals with chronic
health effects data and
modeled concentrations.)
Chronic Monitoring
(Includes seventy five
chemicals with chronic
health effects data,
combined with median
values of the monitored
concentrations.)
Acute Monitoring
(Includes one hundred
nine chemicals with
acute health effects data,
combined with
maritniim values of the
monitored
Above HRL & LOAEL
Above Health Reference Level (HRL)
Below Health Reference Levels
60
Number of Air Pollutants
Figure 3-3. Results of a screening study to identify air pollutants with potential noncancer health
effects.
was modeled or measured. An estimated 50 mil-
lion persons lived within 10 km of monitored sites
or within 2 km of facilities where modeled con-
centrations of one or more chemicals exceeded the
health reference level. For the LOAEL, the com-
parable population estimate was 19 million per-
sons. The data in Figure 3-3 are for individual air
pollutants: typically, however, several pollutants
were present in each area studied, but the effects
of simultaneous exposure to multiple pollutants
were not considered. This screening study con-
cluded that exposure to air pollutants may pose
risks of respiratory, neurologic, and reproductive
systems effects and a risk for adverse develop-
mental effects, for both individual chemicals, and
chemical mixtures.
None of the screening studies performed to
date claim to demonstrate actual cause-and-effect
relationships between routine emissions of HAPs
(or their resulting exposures) and an observed
disease or other health effect. As noted previ-
ously, the risk factors used in such screening
studies are "upper bound estimates" and are
highly uncertain. The actual human risks are not
known and are expected to be lower than the
26
-------
"upper bound" estimates derived in such screen-
ing studies. The screening studies are useful,
however, for comparing the relative ranking of
the potential risks due to different pollutants and
sources.
27
-------
28
-------
Section 4
Research Needs
The limited information currently available
suggests that area sources contribute to air pollu-
tion that can potentially damage public health. In
addition, atmospheric transformation products
formed from area source emissions might also
contribute to health risks. Only a limited amount
of credible data is available with which to charac-
terize the risks posed by exposures to urban air.
Determining exposures, identifying the sources of
those exposures, estimating the likely resulting
health impacts, and identifying needed controls
are very complex tasks. The scope and the com-
plexity of these tasks make it necessary to identify
the most critical research needs. Identification of
the critical research needs provides a framework
for systematically gathering information about
urban HAPs over the coming decade to support
development and implementation of the National
Strategy for area sources.
To assess and manage efficiently the risks
associated with HAPs from area sources, data
from each compartment of the Environmental
Health Paradigm are needed. Consequently, the
research needs are organized and presented using
this paradigm.
4.1 Research on Exposure Assess-
ment
The discussion of Exposure Assessment re-
search needs will address the key research ques-
tions related to Emission Sources, Environmental
Concentrations, and Human Exposures.
Emission Sources
The key research needs for characterizing
emission sources of HAPs are organized around
the following questions:
Which area sources emit HAPs, and how
much do they emit?
What are the most important sources and
pollutants for which detailed emissions
data must be developed?
What are the most reasonable approaches
for reducing emissions?
To assist with implementation of the National
Strategy, data on the feasibility of pollution pre-
vention or of adding emission controls must also
be addressed.
Which area sources emit HAPs, and how much
do they emit?
Given the limited availability of high-quality
data on emissions of HAPs from many area
sources, research is needed to identify the
specific types of stationary sources that meet
the definition of an "area source" and to char-
acterize which HAPs they emit and in what
quantities. Such data are critical to identifying
the 30 or more "worst" HAPs, as required by
the CAA. Research into methods to measure
the emitted HAPs is fundamental to increasing
our knowledge of area source emissions.
29
-------
What are the most important sources and pollut-
ants for which detailed emissions data must be
developed?
The 30 or more "worst" HAPs have not yet
been identified. Once the chemicals are speci-
fied, the task of identify ing the area sources
accounting for 90% of the area source emis-
sions of each of the identified compounds
becomes critical. In order to identify those
area sources, detailed emission factors and
emission estimation techniques will need to be
developed.
What are the most logical approaches for reducing
emissions from area sources, in terms of potential
benefit, technical feasibility, costs, and impacts of
other control programs?
Currently, the best approaches have not been
determined. Pollution prevention approaches
must be explored, while taking into account
current data to define the achievable level of
control and the costs of control. Other emis-
sion control programs, notably efforts to limit
precursors of ozone (some of which are also
HAPs; others of which might produce HAPs
through transformation processes, etc.), can
indirectly benefit the National Strategy, and
the benefits from those programs must also be
considered.
Environmental Concentrations
The key research issues under the Environ-
mental Concentrations component deal with col-
lecting the ambient data, considering the impacts
of atmospheric transformation, and developing
methods to make use of the ambient monitoring
data. The key research questions are:
What are the concentrations of HAPs from
area sources?
How does atmospheric transformation
increase public risks?
How can monitoring and modeling best be
used to assess the effectiveness of the
National Strategy?
What are the concentrations of HAPs from area
sources, both from direct emissions and as sec-
ondary products, to which people are exposed?
Research is needed to develop methods to
measure not only the listed HAPs, but the
myriad potentially harmful chemicals present
in urban air. Data are also needed to assess
just how much monitoring is needed (for
example, number of cities needed to provide a
"representative" sample, the number of sites
per city and the distances between sites, and
the frequency of sample collection) to charac-
terize the urban levels to which people are
exposed.
How does atmospheric transformation increase
public risks?
Research is needed to determine if the muta-
genic transformation products formed in urban
air are actually a hazard to human health, and
if so, to identify the specific transformation
products and any other necessary precursors
that are responsible for the potential elevated
risks. Only then can reasonable steps be taken
to mitigate or prevent the exposure to and risk
from these transformation products.
How can ambient monitoring best be used with
available modeling methods (including emissions
modeling, dispersion modeling, and source appor-
tionment modeling) to demonstrate the effective-
ness of the National Strategy (as required in the
CAA)?
Critical components of this research are: 1)
defining how to use ambient outdoor monitor-
ing data to establish a "baseline" (the concen-
trations existing before the National Strategy
30
-------
is implemented) and to determine concentra-
tion trends to measure the effectiveness of the
National Strategy; 2) identify ing the other
factors (like wind speed, wind direction and
the mixing depth, source emission profiles,
and the distribution of sources throughout the
urban area) that must be measured in order to
derive an estimate of total area source emis-
sions from the measured ambient outdoor
concentrations; 3) developing data analysis
methods to allow the trend in area source
emissions to be determined despite "noise"
from natural variations (like those caused by
year to year changes hi weather) and from the
trends of point sources and mobile sources;
and 4) determining if ambient outdoor data
indicate that all area sources of the controlled
HAPs have been recognized (that is, do the
ambient concentrations reconcile with EPA's
understanding of the emission sources?)
Human Exposures
The key research questions for Human Expo-
sures are:
What are the human exposures to HAPs?
What are the routes of exposure?
What is the distribution of human exposures to
the various HAPs? By what route, and how effec-
tively, do the HAPs reach humans?
Data are needed to define how people's activi-
ties and the concentration of the HAPs vary
with time and to characterize how that varia-
tion will affect the distribution of exposures.
Research is also needed to define those cir-
cumstances that will lead to high exposures
and high potential risks, including research to
identify the chemicals and circumstances that
make indirect exposures important.
4.2 Research on Effects Assessment
As with Exposure Assessment, there is a need
for more research into Effects Assessment. Two
areas that need additional research are Internal
Dose and Health Effects.
internal Dose and Health Effects
Critical issues facing health effects researchers
in trying to define the potential human health
effects of hazardous air pollutant emissions from
area sources are:
How can the most substantial hazards from
HAPs be identified?
How can health risks be estimated reli-
ably?
How can the most substantial hazards from HAPs
be identified?
Hazard identification research is needed to
develop, refine, and validate methods for
identifying chemicals and agents that pose
potential human hazards. Faster, more accu-
rate, less expensive, and more reliable tech-
niques are needed to determine cause and
effect relationships between environmental
pollutants and adverse health outcomes than
the methods that are currently available. Bat-
teries of test methods designed to evaluate
potential hazards comprehensively also need to
be validated. A comprehensive program to
collect toxicity data also is needed. Efforts
should include evaluation of realistic scenarios
for concentrations and exposures.
Additionally, field studies that evaluate the
biological effects of exposure to urban air
pollution are needed. These field studies
should combine short-term methods developed
in the laboratory to screen for problem chemi-
cals, mixtures, and/or sources, and longer-
31
-------
term studies to describe in more detail the
hazards of urban air pollutant exposures.
How can health risks be estimated reliably?
Improved methods are needed to link ambient
exposures to internal dose. Efforts in this area
should include development and validation of
biological markers for exposure, effects, and
susceptibility in human populations; and im-
provement hi pharmacokinetic models. These
models use physiological and biochemical data
to estimate internal doses resulting from exter-
nal exposures. These efforts improve the
confidence in extrapolation of animal data to
humans and from the high doses used in labo-
ratory studies to the lower doses more typical
of human exposures.
Dose-response research is needed to develop
biologically based dose-response models that
elucidate: 1) the relationship between exposure
concentration (or, the applied dose) and the
dose at the site of toxics action (that is, the
target dose) and 2) the basic biological mecha-
nisms responsible for the observed effects.
Understanding of underlying biological mecha-
nisms is crucial to the accurate extrapolation
of research results (for example, extrapolation
of results from animals to humans, from high-
to low-dose, and from "across-exposure sce-
nario" effects.) These models estimate the
type and extent of biological damage resulting
from doses to the affected tissues, which,
when coupled with exposure data, provides
estimates of public health risks.
In addition, because HAPs in the environment
never occur alone, predictive models for risk
assessment of complex mixtures of HAPs are
needed: the most urgent needs include tech-
niques to compare potencies of various mix-
tures, to understand the mechanisms of chemi-
cal interactions hi complex mixtures, to identi-
fy the most critical components leading to
biological activity in complex mixtures, to
determine the quantity of the biologically
active components that reach susceptible or-
gans or tissues hi exposed people, and to
develop biological markers of exposure and
effects. These efforts are necessary to enable
evaluation of the risks from environmental
mixtures of pollutants.
Lastly, research hi environmental epidemiolo-
gy is needed to assess the impact of exposure
to HAPs on the general population and to
establish the link between environmental
exposures and human health effects. Identifi-
cation of appropriate biomarkers of exposure
and effects are likely to be necessary to make
such studies feasible for many pollutants.
32
-------
section 5
Summary of Preliminary Findings
Much of the discussion in this report has been
framed around the Environmental Health Para-
digm. Without at least some understanding of
each component hi the paradigm, it is impossible
to develop reliable risk assessments. Adequate
data exist only for a few HAPs. Screening stud-
ies, like those referred to in Section 3, are helpful
for outlining the potential dimensions of the urban
HAP problem, but these studies are often based
on incomplete, inadequate, and unreliable data.
From a strictly scientific perspective, such studies
are suggestive; however, they might not be suffi-
ciently comprehensive or reliable to use for identi-
fying the "worst" HAPs from area sources, or to
use as the basis for the National Strategy. In the
following discussion, the summary of preliminary
findings on what is currently known about HAPs
from area sources in urban areas is organized
according to the components of the Environmental
Health Paradigm.
Emission Sources
A total of 42 HAPs appear to have "Fair or
Better" emissions data for all (not just area)
sources. (Seventeen HAPs are regularly in-
cluded in the available urban area emission
inventories; an additional twenty-five HAPs
either have national inventories or have vali-
dated emission factors in the FIRE data base.)
Detailed area source information in most
urban area HAP emission inventories is limit-
ed. Much of the data (including data available
under Title m of the Superfund Amendments
and Reauthorization Act) is considered to be
incomplete, out-of-date, or limited in scope
and application. More than 120 HAPs have
little or no validated source emissions data.
Emission factors, source activity data, and
other emission estimation techniques are of
questionable quality or are currently unavail-
able for a number of area sources of HAPs.
Environmental Concentrations
There are no measurements of the air concen-
trations of almost 40% of the listed HAPs.
Another 20% of the HAPs have very little
monitoring data. For a few compounds, there
are considerable monitoring data collected at a
variety of locations. The ability to measure the
HAPs is severely limited by the lack of meth-
ods to collect and analyze many of the listed
chemicals.
Atmospheric transformations complicate expo-
sure assessment because they can increase or
decrease the environmental concentrations of
the listed HAPs. In addition, sunlight causes
reactions among pollutants in urban air that
can produce a variety of products, some of
which are potentially even more harmful than
the original pollutants. HAPs might be formed
from non-hazardous precursors, some of
which are emitted in large amounts into urban
air.
33
-------
Human Exposures
Outdoor sources of HAPs form the baseline
for human exposure, on top of which HAPs
from indoor, workplace, and personal use
sources add additional exposures. For some of
the HAPs, such interior sources may be very
commonplace and may frequently increase
interior concentrations substantially above
outdoor concentrations. For many of the
gaseous HAPs, the indoor concentrations due
to outdoor sources are equal to the outdoor
concentrations. For other HAPs, the indoor
concentrations attributable to outdoor sources
are expected to be somewhat less because of
physical or chemical losses as the HAPs are
transported indoors. For HAPs attached to
fine particles hi the ah", the indoor concentra-
tions from outdoor sources are expected to be
50-90% of the outdoor concentrations.
Available human exposure data often do not
describe well those situations that can lead to
very high exposures to area source emissions
(for example, living above a dry cleaning
establishment or adjacent to a gas station).
Internal Dose
Estimating the amount of HAP that reaches
affected or susceptible organ(s) and causes
damage to health is important in understanding
the relationship between exposures to HAPs
and the nature and magnitude of potential
public health effects. This is particularly true
when risk estimates are based on extrapolated
information. Current methods and data for
estimating Internal dose are often crude. Good
information exists only for a few HAPs.
Health Effects
There are some health effects data available
for each of the 189 HAPs. In almost no case,
however, are there data available on all of the
most important health effects: cancer, devel-
opmental and reproductive effects, neurotoxic-
ity, and short-term and long-term pulmonary
effects. The quality of the available data var-
ies, ranging from inadequate to excellent.
The evaluation of the cancer-causing potential
of the HAPs is more complete than for other
health effects. Also, reference concentrations
(RfCs) for noncancer health effects have been
developed for 40 of the listed HAPs. Values
of the cancer risk estimates and of the RfCs
are likely to change as new and better data be-
come available.
Only 10% of nearly 3,000 chemicals that can
exist as air pollutants have been tested for
genotoxicity or carcinogenicity. The number
of chemicals tested for noncancer effects is
even smaller. The development of data on this
broad range of substances is almost certainly
not warranted. Further analysis is needed to
target specific chemicals for further evalua-
tion.
People are exposed to mixtures of many pol-
lutants simultaneously, not just one pollutant
at a time. Yet, how these mixtures of pollut-
ants interact to affect human health is only
poorly understood.
Synopsis
The availability of data on the 189 HAPs that
are needed to do a complete environmental health
assessment is illustrated in Figure 5-1. It reveals
that very little is known about many of the HAPs,
while significant amounts of information exist for
a few chemicals. The same data are given for
each of the 189 listed HAPs in Table A-l, found
in the Appendix. A review of Table A-l reveals
that 20 chemicals have enough data to merit "Fair
or Better" classifications in Source Emissions
34
-------
Availability of Data for Various Categories
For the 189 Listed HAPs
Source Emissions Data
Uttie or No Information
Ambient Concentration Data
No Data
Little Data
Fair or Better Data ^f^' Fair or Better Data
Occasionally Found Occasionally Observed
Noncancer Effects Data
No validated RfC
Carcinogenicity Data
Unknown Carcinogenicity ^t > Devaluated HAPs
59
ASsKV
Possible Carcinogen
Moderate or High Confidence in RfC
Low Confidence in RfC
Probable or Known Carcinogen
Figure 5-1. Summary of the available data on the 189 listed HAPs. (Table A-l, Appendix A,
categorizes the data for each of the 189 HAPs.)
Data, Ambient Concentration Data, and in one of
the Health Effects areas, either Noncancer Health
Effects or Cancer Health Effects. This list of
chemicals does not identify the 30 or more
"worst" HAPs; rather, the list simply identifies
those HAPs with sufficient data to begin a risk
assessment of either the cancer or noncancer
effects due to exposure to that chemical. Another
20 HAPs are rated "Fair or Better" in two of the
three required areas. Targeted research on this
second group of HAPs could readily provide
sufficient data to allow a risk assessment to be
initiated. The 40 HAPs with the most complete
available data are listed in Table 5-1.
Continuing research will undoubtedly improve
the scientific understanding of human exposures
and health effects from increasing numbers of
HAPs.
35
-------
Table 5-1. The HAPs with the most extensive available data needed for a risk assessment.
HAPs with data rated "Fair or Better" in
the three areas:
Source Emissions
Ambient Concentrations
and
Health Effects (Cancer or Noncancer)
HAPs with data rated "Fair or Better" in
two of the following three areas:
Source Emissions
Ambient Concentrations
and
Health Effects (Cancer or Noncancer)
Benzene
1,3-Butadiene
Carbon tetrachloride
Chloroform
Ethylene dibromide
Ethylene dichloride
Formaldehyde
Methylene chloride
Styrene
Tetrachloroethylene
Toluene
Trichloroethylene
Vinyl chloride
Arsenic compounds
Chromium compounds
Lead compounds
Manganese compounds
Mercury compounds
Nickel compounds
Selenium compounds
Acetaldehyde
DDE (p,p'-dichlorodiphenyldichloro-
ethylene)
1,4-Dichlorobenzene
Ethylbenzene
Ethylene oxide
Hexachlorobenzene
Hexane
Methyl bromide
Methyl chloroform
Pentachlorophenol
Polychlorinated biphenyls
Propylene dichloride
2,3,7,8-Tetrachlorodibenzo-p-dioxin
2,4,6-Trichlorophenol
Vinylidene chloride
Xylenes (mixed isomers)
Antimony compounds
Beryllium compounds
Cadmium compounds
Polycyclic Organic Matter
36
-------
References
37
-------
1. The Clean Air Act (42 U.S.C. 7401-7626) consists of Public Law 159 (July
14, 1955; 69 Stat. 322) and the amendments made by subsequent enactments,
including'Public Law 101-549 (The Clean Air Act Amendments of 1990)
approved November 15, 1990.
2. Andur, M. O. "Air Pollutants" In: Toxicology: Basic Science of Poisons. Eds.
M. O. Andur, J. Doull, C. Klaassen, Pergamon Press, NY, NY. 1994, p.
856.
3. World Health Organization and United Nations Environment Programme.
Urban Air Pollution in Megacities of the World. Blackwell Publishers, Oxford,
UK. p. 7, 1992.
4. World Health Organization. Air Quality Guidelines for Europe. WHO Regional
Publishing, Copenhagen, DK. 1987.
5. See, for example: A Legislative History of the Clean Air Act Amendments of
1990, Volume V, S.Prt. 103-38, 103rd Congress, November 1993, pp. 8468-
8471 and pp. 8489-8492.
6. Sexton, K., S.G. Selevan, D.K. Wagener, and J.A. Lybarger, Estimating
Human Exposures to Environmental Pollutants: Availability and Utility of
Existing Databases. Archives of Environmental Health. 47: 398, 1992.
7. Cote, I. L. and J. J. Vandenberg, "Overview of Health Effects and Risk
Assessment Issues Associated with Air Pollution," in The Vulnerable Brain,
Vol. Ill, Isaacson and Jensen, editors, Plenum Press, in press.
8. U.S. Environmental Protection Agency, Toxic Air Pollutants and Noncancer
Risks: Screening Studies, External Review Draft, September, 1990.
9. Jones, J. W., D. Campbell, D.L. Jones, S. Kersteter and M. Saeger,
"AEERL's Hazardous Air Pollutant Emissions Research Under JEIOG: A
Status Report" Proceedings VIP-27, Emission Inventory Issues, Specialty
Conference, Durham, NC. October 19, 1992.
10. Sullivan, D., T. Lahre and M. Alford, Assessing Multiple Pollutant Multiple
Source Cancer Risks from Urban Air Toxics, Office of Air Quality Planning
and Standards, Research Triangle Park, NC. EPA-450/2-89-010. (NTIS PB89-
197222), April, 1989.
11. Ramamurthi, M. T.J. Kelly and C. W. Spicer, Temporal and Spatial Variabil-
ity of Toxic VOC Sources in Columbus, Ohio, Measurement of Toxic and
Related Air Pollutants: Proceedings of the 1993 EPA/A&WMA International
Specialty Conference, Durham, NC. May, 1993. EPA/600/A93/024.
38
-------
12. Lewis, C. L., T.L. Conner, R.K. Stevens, J.F. Collins and R.C. Henry,
"Receptor Modeling of Volatile Hydrocarbons Measured in the 1990 Atlanta
Ozone Precursor Study," Proceedings of the 86th Annual Meeting of the Air &
Waste Management Association, Volume II, 93-TP-58.04, Denver, CO. June,
1993.
13. Kelly, T.J., M. Ramamurthi, A.J. Pollack, C.W. Spicer, J. Shah, D. Joseph
and L.T. Cupitt, "Surveys of the 189 CAAA Hazardous Air Pollutants: I
Atmospheric Concentrations in the U.S.", Measurement of Toxic and Related
Air Pollutants: Proceedings of the 1993 EPA/A&WMA International Specialty
Conference, Durham, NC, May, 1993. EPA/600/A93/024.
14. U.S. Environmental Protection Agency, Integrated Air Cancer Project:
Summary Report, Office of Research and Development, Research Triangle
Park, NC. 1993.
15. Buxton, B.E. and A.D. Pate, "Joint Temporal-Spatial Modeling of Concentra-
tions of Hazardous Pollutants in Urban Air," presented at Forum on Geo-
statistics for the Next Century, Battelle, Columbus, OH. 1993.
16. Robinson, J. and J. Holland, "Trends in Time Use," in Technology and the
American Economic Transition, OTA-TET-283, Office of Technology Assess-
ment, 1988.
17. Lewis, C. L. "Sources of Air Pollutants Indoors: VOC and Fine Paniculate
Species," J. Exposure Analysis and Environ., 1:31-44, 1991.
18. Harley, R. A. and G. R. Cass, "Modeling the Concentrations of Gas-Phase
Toxic Organic Air Pollutants: Direct Emissions and Atmospheric Formation,"
Environ. Sci. Technol., 28, 88-98, 1994.
19. Kleindienst, T.E., P.B. Shepson, E.G. Edney, L.D. Claxton, and L.T. Cupitt,
"Wood smoke: measurement of the mutagenic activities of its gas- and parti-
cle-phase photooxidation products," Environmental Science & Technology,
(20), 493-501, 1986.
20. Cupitt, L. T., L. D. Claxton, T. E. Kleindienst, D. F. Smith and P. B.
Shepson, "Transformation of Boise Sources: The Production and Distribution
of Mutagenic Compounds in Wood Smoke and Auto Exhaust," Proceedings of
the 1988 EPA/APCA International Symposium on Measurement of Toxic and
Related Air Pollutants, EPA-600/9-88-015 (NTIS PB90-225863), pp. 885-889,
1988.
39
-------
21. Ott, W. R. "Total Human Exposure: Basic Concepts, EPA Field Studies, and
Future Research Needs," Journal of the Air & Waste Management Association,
Vol. 40, No. 7, 966-975 (1990).
22. Wallace, L. "The Team Studies," EPA Journal, Volume 19, Number 4, EPA
175-N-93-027, October-December 1993.
23. See, for example: Sexton, K. "National human exposure assessment survey,"
U.S. EPA Position Paper, Office of Health Research, US EPA, Washington,
DC, 1991. or
Clickner, R., G. Kalton and A. Chu, "Statistical Design Issues in Human
Exposure Assessment Surveys: Sample Design Issues and Options", Delivery
Order 23 Report, EPA Contract 68-W1-0019. Westat, Inc., Rockville, MD,
July, 1993.
24. U.S. Environmental Protection Agency. Guidelines for Carcinogen Risk
Assessment. Carcinogen Assessment Group, Office of Health and Environmen-
tal Assessment. Washington, D.C. 51 FR 33992. September 24, 1986.
25. Schuetzle D, and J. Lewtas. "Bioassay-directed Chemical Analysis in Envi-
ronmental Research," Anal. Chem. (58):1060A-1075A, 1986.
26. Graedel, T.E., D. T. Hawkins and L.D. Claxton, Atmospheric Chemical Com-
pounds: Sources, Occurrence, andBioassay, Academic Press, Inc., Orlando,
FL. 1986.
27. Meardon, K., Cancer Risk from Outdoor Exposure to Air Toxics, Vols. 1 and
2, Office of Air Quality Planning and Standards, Research Triangle Park, NC.
EPA-450/l-90-004a,b (NTIS PB91-159624, -159632), September, 1990.
28. U.S. Environmental Protection Agency, Motor Vehicle-Related Air Toxics
Study, Office of Mobile Sources, Ann Arbor, MI. EPA 420-R-93-005. April,
1993.
29. Cote, I., L. T. Cupitt and B. M. Hassett, "Toxic Air Pollutants and Noncan-
cer Health Risks" in Risk Analysis: Prospects and Opportunities, C. Zervos,
editor, Plenum Press, New York, NY, pp. 697-706, 1991.
40
-------
Appendix
41
-------
Table A-l provides a listing of the available data for each of the 189 listed
HAPs. The characterizations are consistent with those used in the text of the report.
The shaded chemicals are the 20 HAPs identified in Table 5-1 as having "Fair or
Better" data in three categories: Emission Sources, Environmental Concentrations,
and Health Effects (either Cancer Effects or Noncancer Effects).
Meaning of the Symbols Used hi the Table:
Source Emissions Data
Blank
/
Ambient Concentration Data
Seldom included in emissions in-
ventories (Little or no data)
Occasionally included in emissions
inventories (i.e., included in 10%
or more, but less than half, of
emission inventories studied)
Routinely included in emissions
inventories (i.e., in 50% or more
of the case studies), or national
inventory is available, or emission
factors included in FIRE data base
Blank Little or no (71 HAPs) ambient
data available
/ Between 100 and 1000 observations
More than 1000 observations
Noncancer Health Effects Data Blank No validated RfC available
/
Cancer Health Effects Data
RfC available, but "low" confi-
dence
RfC available, "moderate" or
"high" confidence
Blank Unclassified (59 HAPs) or Class D
(22 HAPs)
/ Class C carcinogen or LARC Class
2B chemical
Class A or B carcinogen or IARC
Class 1 or Class 2A chemical
42
-------
Table A-l. Availability of data on the 189 listed HAPs.
NO.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
/*
16
17
18
19
20
21
Chemical Name
Acetaldehyde
Acetamide
Acetonltrlle
Acetophenone
2-Acetylamlnofluorene
Acroleln
Acrylamide
Acrylic acid
Acrylonltrlle
Ally! chloride
4-Amlnobiphenyl
Aniline
o-Anlsldlne
Asbestos
- ; , - x - - - \ .:!,; ' - "
Benzldine
Benzotrichlorlde
Benzyl chloride
Blphenyl
Bls(2-ethylhexyl)phthalate
Bls(chloromethyOether
Frequency of Occur-
rence
Source
Emis-
sions
Data
/
/
/
/
'
/
Ambient
Concen-
tration
Data
ss
s
s
'
f f
^
Health Effects Data
Noncan-
cer
s
SS
ss
ss
s
s
; - -I
Cancer
ss
s
ss
/
//
ss
s
ss
ss
s
ss
-'**'
ss
SJ
ss
ss
ss
43
-------
No.
22
23
24
25
26
27
28
20
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
Chemical Name
Bromoform
1T5«Butatfieft&
Calcium cyanamlde
Caprolactam
Captan
Carbaryl
carbon disulflde
carboB tetrachiorwe
Carbonyl sulflde
Catechol
Chloramben
Chlordane
Chlorine
Chloroacetlc acid
2-Chloroacetophenone
Chlorobenzene
Chlorobenzllate
ctttomfdrm
Chloromethyl methyl ether
Chloroprene
Cresols (Isomers and mixture)
o-Cresol
m-Cresol
p-Cresol
Frequency of Occur-
rence
Source
Emis-
sions
Data
/
vV
SS
JJ
Ambient
Concen-
tration
Data
/
//
/
/
ss
s
ss
ss
/
Health Effects Data
Noncan-
cer
s
SS
Cancer
//
SS
ss
//
//
ss
//
/
/
/
/
44
-------
NO.
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
Chemical Name
Cumene
2,4-D, salts & esters
DDE (72-55-9: p.p'-dlchlorodiphenyldichloroethylene)
Diazomethane
Dibenzofuran
1,2-Dlbromo-3-chloropropane
Dlbutyl phthalate
1,4-Dlchlorobenzene
3,3'-Dlchlorobenzldlne
Dlchloroethyl ether
1,3-Dlchloropropene
Dichlorvos
Dlethanolamlne
N,N-Dimethyl aniline (also, dlethyl)
Dlethyl sulfate
3.3'-Dlmethoxybenzldlne
4-Dlmethyiamlnoazobenzene
3,3'-Dlmethylbenzldlne
Dlmethylcarbamoyl chloride
Dlmethylformamlde
1,1-Dlmethylhydrazlne
Dimethyl phthalate
Dimethyl sulfate
Frequency of Occur-
rence
Source
Emis-
sions
Data
//
//
/
/
Ambient
concen-
tration
Data
//
//
//
//
/
/
Health Effects Data
Noncan-
cer
SS
ss
ss
ss
ss
Cancer
ss
ss
ss
ss
ss
ss
ss
ss
ss
ss
ss
ss
s
ss
ss
45
-------
NO
69
70
71
72
73
74
75
76
77
78
79
80
»1
82
83
84
85
86
,«M
88
89
90
91
Chemical Name
4,6-Dinltro-o-cresol & salts
2,4-Dinltrophenol
2,4-Dlnltrotoluene
1,4-Dloxane
1 ,2-Dlphenylhydrazlne
Eplchlorohydrln
1,2-Epoxybutane
Ethyl acrylate
Ethylbenzene
Ethyl carbamate
Ethyl chloride
£awleaedl&ram&i»
Sttiylene dlchtoridsi
Ethylene glycol
Ethylenimlne
Ethylene oxide
Ethylenethlourea
Ethylldene dlchlorlde
i^m»sktei*y
-------
NO.
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
Chemical Name
Hexachloroethane
Hexamethylene-l.e-dilsocyanate
Hexamethylphosphoramlde
Hexane
Hydrazine
Hydrochloric acid
Hydrogen fluoride
Hydroguinone
isophorone
Llndane
Malelc anhydride
Methanol
Methoxychlor
Methyl bromide
Methyl chloride
Methyl chloroform
Methyl ethyl ketone
Methyl hydrazlne
Methyl iodide
Methyl Isobutyl ketone
Methyl isocyanate
Methyl methacrylate
Methyl tert-butyi ether
4^'-Methylenebis(2-chloroanlllne)
Frequency of Occur-
rence
Source
Emis-
sions
Data
SS
ss
ss
/
/
//
//
Ambient
Concen-
tration
Data
//
/
^
SS
ss
ss
/
/
^
Health Effects Data
Noncan-
cer
^/
/
//
/
SS
s
ss
J
ss
Cancer
/
/
^/
/
SS
V
ss
/
ss
47
-------
No.
-m
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
Chemical Name
»ea»$et»<&lortflfr
Methylene dipnenyl dilsocyanate
4.4'-Methylenedianillne
Naphthalene
Nitrobenzene
4-Nltrobiphenyl
4-Nitrophenol
2-Nltropropane
N-Nltroso-N-methylurea
N-Nltrosodlmethylamlne
N-Nltrosomorpholine
Parathlon
Pentachloronltrobenzene
Pentachlorophenol
Phenol
p-Phenylenedlamine
Phosgene
Phosphlne
Phosphorus
Phthallc anhydride
Polychlorinated blphenyls
1,3-Propane sultone
beta-Propiolactone
Proplonaldehyde
Frequency of Occur-
rence
Source
Emis-
sions
Data
SS
/
/
/
J
ss
ss
ss
Ambient
Concen-
tration
Data
SS
s
/"
j
s
Health Effects Data
Noncan-
cer
SS
j
Cancer
//
//
^/
//
/
s
s
SS
ss
SS
SS
48
-------
NO.
140
141
142
143
144
145
*4$
147
148
149
m*
151
*&"
153
154
155
156
157
158
'JsVSV
160
161
162
163
Chemical Name
Propoxur
Propylene dichlorlde
Propylene oxide
1,2-Propylenimlne
Qulnollne
Qulnone
$tY**n*
Styrene oxide
2,3,7,8-Tetrachlorodlbenzo-p-dloxln
1 ,1 ,2,2-Tetrachloroethane
t^*s^|j?ro0toy(^i$ - ' ^ : -- - *% ,
Titanium tetrachlorlde
Yfr(uenVx - m - ,/ "" ^'-;- c"- -
2,4-Toluenedlamlne
Toluene-2,4-dllsocyanate
o-Toluldlne
Toxaphene
1.2,4-Trlchlorobenzene
1,1,2-Trlchloroethane
\t,--^ X- \ y ' 0x*xV ° «0 ; , - >^oN% I - '- - ' *
*3$ftto&$fr\&t& * <^1V ,;H; i 4\\ ^ -,- '
2,4,5-Trlchlorophenol
2,4,6-Trichlorophenol
Trlethylamlne
Trifluralln
Frequency of Occur-
rence
Source
Emis-
sions
Data
/
/
M
ss
JS
\.-ff"
JS
ss
ss
Ambient
Concen-
tration
Data
/
//
ss
/
//
Jf
JS
/
/
//
"-v;r;:'l
/
Health Effects Data
Noncan-
cer
//
//
/y
.
/y
zff^sj
<, \'f^\-
{', ,,,\ ' * '
s
Cancer
^/
^/
SS
s
ss
ss
/
ss
ss
ss
ss
/
''-"ss -
ss
/
49
-------
NO.
164
165
166
«7
168
169
170
171
172
173
174
175
176
177
178
179
180,
181
^182:
18$
184'
185
xW
Chemical Name
2,2,4-Trlmethylpentane
Vinyl acetate
Vinyl bromide
Vlrtyl tfcloritie
Vinylldene chloride
xylenes (mixed isomers)
o-Xylene
m-xylene
p-Xylene
Antimony compounds
Arsftrtfp OOmpOUntfe /'.
*&$'^
50
-------
NO.
187
188
m
Chemical Name
Polycycllc organic Matter (various PAHs)
Radlonuclldes
Selenium compounds butf fde, dfeuif Itte ar otfw
Frequency of Occur-
rence
source
Emis-
sions
Data
SS
//
Ambient
Concen-
tration
Data
/
vV
Health Effects Data
Noncan-
cer
Cancer
^S^^^^^^^^^^^E
//
51
-------
Table A-2 identifies the types of health effects, other than cancer (referred to
as noncancer effects), that have been reported for the listed HAPs. The table presents
data for only those HAPs that have produced effects in humans or animals by
inhalation exposure.
52
-------
SYSTEM
or HEALTH EFFECT
Bone
Cardiovascular
Death
Dermal
Reproductive/
Developmental
Endocrlne/Exocrlne
Ocular
Gastrointestinal
Hematopoletlc
Hepatic
Immunologic
Multiple
Neurologic/Behavioral
Olfactory
Pancreatic
Renal
Respiratory
spleen
Systemic
LD50
Total Number of HAPs
Showing an Effect
EXPOSURE DURATION
ACUTE
0
52
56
33
7
24
96
63
38
51
14
1
107
8
1
47
114
3
48
65
142
SUBCHRONIC
6
30
15
21
54
24
44
31
49
69
24
0
74
15
0
49
- 78
18
71
0
122
CHRONIC
2
6
1
3
13
8
7
7
16
27
5
1
20
6
1
23
30
11
27
0
57
Table A-2. Number of hazardous air pollutants that have been reported to produce health effects In humans
or animals by Inhalation exposure.
53
-------
54
-------
Glossary
55
-------
Term
Definition
Accidental re-
lease
Accuracy
Acute effects
Adverse health
effects
Air quality mod-
eling
Air toxics
Ambient air
Ambient concen-
tration
Ambient mea-
surement
Ambient monitor-
ing
Emissions resulting from an unpredicted failure of a system due
to which some harm results.
The quality of being free from error. The degree of accuracy is a
measure of the uncertainty in identifying the true measure of a
quantity at the level of precision of the scale used for quantity.
Toxic effects of a substance which become manifest after only a
short period of exposure of a duration measured in minutes,
hours, or days.
An undesirable antagonistic consequence to human health due to
some causative agent.
A mathematical representation of pollutant concentrations and
their distribution in the atmosphere based upon assumptions or
simulations of pollutant emissions, meteorological dispersion and
transport, chemical and physical reactions, etc.
An expression commonly used to refer to hazardous air pollutants
often used interchangeably with "hazardous air pollutants."
Any air pollutant (excluding those pollutants for which ambient
criteria do exist, namely ozone, sulfur dioxide, carbon monoxide,
nitrogen oxides, lead, and particulate matter) that may cause any
of a wide range of potential harmful effects.
The surrounding or encompassing atmosphere. In the context of
pollution monitoring, ambient air is often erroneously used to
refer only to "outdoor" air, even though indoor air is "ambient"
to a person who is indoors.
The concentration of a chemical (usually, a pollutant) in the
atmosphere surrounding humans or other potentially affected
receptors.
Measurement of a chemical (pollutant) found in the atmosphere
surrounding humans or other receptors, any potentially affected
species or ecosystem.
Measuring the concentrations of pollutants or other species in
ambient air.
56
-------
Term
Area source
of hazardous air
pollutants
Area Source Na-
tional Strategy
Atmospheric
transformation
Bacterial mutage-
nicity
Bioassay
Definition
A stationary source which annually releases to the atmosphere, or
has the potential to release considering controls, less than 10 tons
of a single hazardous air pollutant listed hi the Clean Air Act or
less than 25 tons of a mixture of these pollutants. The term "area
source" shall not include motor vehicles or nonroad vehicles
subject to regulation under Title II of the Clean Air Act.
The National Strategy mandated hi Section 112(k) of the Clean
Air Act. By November 1995, EPA must "prepare and transmit to
Congress a comprehensive strategy to control emissions of haz-
ardous air pollutants from area sources hi urban areas." The
strategy shall "identify not less than 30 hazardous air pollutants
which, as the result of emissions from area sources, present the
greatest threat to public health hi the largest number of urban
areas." The strategy also shall "identify the source categories or
subcategories" emitting the 30 or more hazardous air pollutants
and "shall assure that sources accounting for 90 per centum or
more of the aggregate emissions of each of the 30 identified
hazardous air pollutants are subject to [emission] standards."
"The strategy shall achieve a reduction hi the incidence of cancer
attributable to exposure to hazardous ah- pollutants emitted by
stationary sources of not less than 75 per centum, considering
control of emissions of hazardous air pollutants from all station-
ary sources and resulting from measures implemented ... under
[the Clean Air Act] or other laws."
The chemical reactions hi the atmosphere, many of which occur
naturally and are unavoidable, that change (transform) one sub-
stance hi the air into a different chemical or chemicals; or the
physical processes (like washout into rain water or adsorption
onto particles) that change the form of the chemical hi the atmo-
sphere and affect its distribution hi the environment.
Refers to the use of bacteria to assess the mutagenic potential of
pollutants.
Determination of the biological activity or potency of a substance
by testing its effect on an organism. As used in this report, a test
for carcinogenicity in laboratory animals (generally, rats and
mice) that includes near-lifelong exposure to the agent (pollutant)
under test. The term is used interchangeably with "animal test."
57
-------
Term
Definition
Bioassay-directed
chemical identifi-
cation techniques
Biomarkers
Cancer risk
Carcinogen
Carcinogenicity
Characterizing
emission sources
Chronic effects
Data base
Developmental
disorders or
effects
Direct emissions
Distribution of
human exposures
Dose
Dose-response
relationship
A combination of chemical and physical separation and identifi-
cation methods with short-term biological tests in order to identi-
fy the chemicals in a complex mixture of pollutants that have a
potential biological effect.
Surrogates or indicators of biological exposure, dose, or effect.
Part of the considerations under the Internal Dose component of
the Environmental Health Paradigm.
The risk of developing cancer.
A substance or agent that tends to produce cancer in living
organisms.
The ability of a substance or agent to produce cancer.
Describing the pollutants emitted by a source, including the
chemical composition, the quantity emitted as a function of time,
and the location and relevant operational parameters of the
source.
Toxic effects of a substance that become manifest after prolonged
or repeated exposures of a duration measured in weeks, months,
or years.
Available, relevant raw information about the subject of concern.
One type of noncancer health effects of concern; impairment of
the normal development of a fetus, infant, or child, including
developmental retardation and birth defects.
Emissions of a pollutant that come directly from the source,
without having to be produced by transformations.
A mathematical representation or other characterization of the
range of exposures that people have to a pollutant.
The amount of a substance administered to an animal or human,
usually measured in mg/kg of body weight, mg/m2 of body
surface area, or parts per million in the food, drinking water, or
inhaled air. Dose, or target dose, is often used to refer to the
quantity of the agent that reaches an affected organ of interest.
The functional relationship between the amount of a substance at
the affected organ and the lethality, morbidity, or level of health
effect produced.
58
-------
Term
Effects Assess-
ment
Emission
Emission estima-
tion techniques
Emission factor
Definition
Identification of the health effects that are likely to occur once
humans (or ecosystems) are exposed to HAPs (or other pollut-
ants).
The releasing of pollutant(s) to the atmosphere by a source or
source category.
A method of estimating pollutant emissions from a particular
source or category of sources. Such methods include the use of
emission factors and activity data for the source or source catego-
ry, as well as statistical approaches using surrogate data (e.g.,
census information) to estimate emissions hi a specific geographic
area.
An emission factor is a measure of the quantity of HAP that is
emitted per unit quantity of a source activity (for example,
pounds of HAPs per barrel of crude oil processed). Ideally, the
source "activity" will represent the operations that lead to emis-
sions (for example, how many barrels of crude oil are processed
in a day). The product of the emission factor and the source
activity is used to estimate the mass of HAP emitted. An emis-
sion factor is an average value which relates the quantity of a
pollutant released to the atmosphere by a source (e.g., chemical
process, fuel combustion) to the activity associated with release
of that pollutant. It is usually expressed as the weight of pollutant
per unit weight, volume, distance, or duration of the activity that
emits the pollutant (e.g., kg of paniculate matter per Mg of coal
burned). To estimate emissions of a pollutant from a source, the
emission factor for that source/pollutant is typically multiplied by
the corresponding source activity level.
A commercial, individual/residential, industrial, or institutional
activity or process that releases pollutants to the atmosphere.
These can be stationary (at a fixed geographic location) or mobile
(e.g., automobiles).
Emission Sources One of the components of the Environmental Health Paradigm: it
includes evaluation of the pollutants emitted by the sources,
including identification of the chemical emitted, the amount
emitted, and the location of the source and the emission points
and their characteristics.
Emission source
59
-------
Term
Definition
Environmental
Concentrations
Environmental
fate
Environmental
Health Paradigm
Epidemiology
Estimation
Exposure
Exposure Assess-
ment
Extrapolation
(e.g., across-
exposure-scenar-
io, animal-to-
human, high-to-
low dose)
A component of the Environmental Health Paradigm: it includes
evaluation of the concentrations of the pollutants in all environ-
mental compartments and media, as appropriate, including indoor
and outdoor air, water, soil, and food.
The disposition of substance hi the environment, including a
description of the distribution between various media (air, water,
soil).
A conceptual framework with which to organize and relate all of
the aspects or considerations needed to characterize how pollut-
ants from a source reach a human (or other receptor) and cause
an effect. Understanding the linkages between the components of
the paradigm also helps with evaluation of environmental man-
agement options. The paradigm includes evaluation of Emission
Sources, Environmental Concentrations, Human Exposures,
Internal Dose, and Health Effect(s).
The study of the causes of diseases by identifying personal and
environmental characteristics common to those contracting the
disease. The sum of the factors controlling the presence or
absence of a disease or pathogen.
The assignment or derivation of outcome values and/or probabili-
ty measures to a postulated event; a rough or approximate calcu-
lation; a numerical value obtained from a statistical sample and
assigned to a population parameter.
The coming into contact of humans (or ecosystems) with pollut-
ants; exposure is measured as the product of concentration of the
pollutant and the time of the exposure.
Evaluation of how people are likely to come into contact with
HAPs (or other pollutants) and the determination of how large
the exposure is likely to be; the measurement or estimation of the
magnitude, frequency, duration, and route of exposure to a haz-
ardous substance or situation, and the size, nature, and classes of
the exposed population.
To project, extend, or expand known or observed data to an area
not known or observed. In the context of hazardous air pollut-
ants, extrapolation is used to predict the following: responses in
humans from animal data; low-dose responses from high-dose
responses; and responses from one specific hazardous air pollut-
ant exposure scenario to another different exposure scenario.
60
-------
Term
Definition
Factor Informa-
tion Retrieval
(FIRE)
Genotoxic
Great Waters
Program
Hazard
Hazardous air
pollutant
Health Effect(s)
Human Carcino-
gen
Human Expo-
sures
Incidence
An EPA supported and published data base of information on
emission factors of various sources.
Possessing the ability to produce harmful effects in the genetic
makeup of an organism.
A research and assessment program being conducted by EPA hi
response to Section 112(m) of the Clean Air Act, entitled "Atmo-
spheric Deposition to Great Lakes and Coastal Waters."
A source of risk (danger, peril, threat) that does not necessarily
imply potential for occurrence. A hazard produces risk only if an
exposure pathway exists and if exposures create the possibility of
adverse consequences.
An airborne substance whose effect on man or animals is poten-
tially large but undefined since an exposure pathway may or may
not exist; the 189 chemicals, or groups of chemicals, in the
initial list of hazardous air pollutants found in Section 112(b) of
the Clean Air Act; an air toxic.
One of the components of the Environmental Health Paradigm: it
includes characterization of the potential health effects due to
exposure, including cancer effects, noncancer effects, any ob-
servable damage of disease or symptoms of adverse effects.
A classification given to a chemical when there is sufficient
evidence from epidemiologic studies to support a causal associa-
tion between exposure to the agents and cancer.
One of the components of the Environmental Health Paradigm: it
involves evaluation of the route, magnitude, duration and fre-
quency of exposure; the interaction of humans with a pollutant or
other physical parameter. Exposure is measured as the product of
concentration and time.
The number of new cases of a disease, usually expressed as a
rate; typically, the number of new cases of a disease occurring in
a population during a specified period of time divided by the
number of persons exposed to risk of developing the disease
during that period of time. The incidence rate is a direct estimate
of the probability of developing a disease during a specified
period of time.
61
-------
Term
Internal dose
LD50
(lethal dose for a
50% death rate)
Locating and
Estimating Re-
ports
Major source
of hazardous air
pollutants
Margin of safety
Mechanism(s) of
action
Mechanistic data
Median
Metabolism
Mobile sources
Definition
One of the components of the Environmental Health Paradigm: it
involves identification of the quantity (dose) of pollutant that is
absorbed (the absorbed dose), the quantity that reaches the affect-
ed organ where it may have an effect (the target dose), and
biological indicators (biomarkers) of exposure and effects.
A calculated dose of a substance that is expected to cause the
death of 50% of an entire defined experimental population within
a specified length of time.
A series of documents issued by EPA to compile available infor-
mation on sources and emissions of substances which may be
toxic at certain concentrations in the ambient air.
A stationary source or group of stationary sources located within
a contiguous area and under common control that annually releas-
es to the atmosphere, or has the potential to release considering
controls, 10 tons or more of a single hazardous ah- pollutant
listed hi the Clean Air Act or 25 tons or more of a mixture of
these pollutants.
A factor added to an estimated risk level for purposes of increas-
ing the probability that a standard based on the resultant level
will provide increased protection to the general population and
individual members from harmful effects of a given substance.
The underlying cause of disorder or disease; the specific physi-
cal, chemical, and/or biological events caused by HAP exposure
that are necessary for development of the resulting symptoms,
disorder, or disease.
Data describing or pertaining to mechanisms of action.
The value in an ordered set of values (that is, ambient concentra-
tion measurements arranged from lowest to highest) hi the mid-
dle, with the number of values (measurements) that are larger
than the median being equal to the number of values (measure-
ments) that are smaller than the median.
The sum of the physical and chemical process hi an organism by
which its material substance is produced, maintained, and de-
stroyed, and by which energy is made available.
Sources of emissions that can move, like automobiles, trucks,
planes, boats, and trams.
62
-------
Term
Definition
Model
Molecular dosim-
etry
Monitoring
(of pollutants)
Mutagen
Mutagenic
Mutagenicity
Mutagenic prod-
ucts
Mutation
A simplified representation of a system or phenomenon, as in the
sciences or economics, with any hypotheses required to describe
the system or explain the phenomenon, often mathematically; a
system of postulates, data, and inferences presented as a mathe-
matical description of an entity or state of affairs; a represen-
tation of reality; a description or analogy used to help visualize
something (e.g., air pollution patterns across a city) that can not
be directly observed.
Characterization of the quantity of a chemical reaching an affect-
ed organ at a molecular level.
Periodic or continuous sampling and analysis to determine the
level of pollution or other characteristics.
A substance possessing the ability to induce heritable mutations
in living organisms.
Having the characteristic of being a mutagen.
The quality of being mutagenic. Mutagenicity is often measured
using short-term bioassays in which changes to the genetic code
of bacteria are identified.
Products of atmospheric transformation that are mutagenic.
A departure from being like the parent in one or more heritable
characteristics, due to a change in a gene or chromosome.
The degree to which a substance is toxic to nerve tissues; one of
the noncancer health effects of concern in development of the
Area Source National Strategy.
A health effect other than the development of cancer. Section
112(k) of the Clean Air Act lists a number of noncancer health
effects to be considered under the Area Source program, includ-
ing "mutagenicity, teratogenicity, neurotoxicity, reproductive
dysfunction and other acute and chronic effects including the role
of such pollutants as precursors of ozone or acid aerosol forma-
tion."
Noncancer risks The risk of developing a noncancer health effect.
Neurotoxicity
Noncancer health
effect
Oral exposure
data
Data on health effects developed from animal tests in which the
exposure to the pollutant is through ingestion.
63
-------
Term
Definition
Particulate matter Solid or liquid particles suspended in the atmosphere; a form of
pollution for which maximum allowable concentrations m the air
have been established through legislation and regulation.
Models that describe the fate of pharmacological substances hi
the body, including absorption, distribution, metabolism, and
elimination; dose-response models based on the principle that
biological effects are the result of biochemical interaction be-
tween foreign substances or their metabolites and parts of the
body.
Pharmacokinetic
models
Photochemical
process
Point source
Possible human
carcinogen
Potency
Probable human
carcinogen
Products of
incomplete com-
bustion (PIC)
Pulmonary ef-
fects, acute and
chronic
Chemical reactions initiated by the absorption of light. Formation
of ozone and other manifestations of "smog" are the result of a
long series of atmospheric reactions that are started by the ab-
sorption of light by chemicals in the air and the resultant produc-
tion of highly reactive molecular fragments.
A stationary source of pollutants, where the location of the
source and its emissions of pollutants can be specified.
A classification given to a chemical when there is limited evi-
dence of carcinogenicity in animals in the absence of human
data.
The efficacy, effectiveness, or strength of a chemical to cause a
toxicologic response.
A classification given to a chemical when there is limited evi-
dence of human carcinogenicity based on epidemiologic studies
or sufficient evidence of carcinogenicity based on annual studies.
All of the products other than water and carbon dioxide that are
produced when an organic fuel, like gasoline, fuel oil, or wood,
is burned; commonly, PIC is used to refer to a complex mixture
of non-volatile and semi-volatile organic chemicals, many of
which are polycyclic organic compounds, associated with panicu-
late emissions that occur whenever a fuel is burned incompletely.
Adverse health effects involving the lungs and due to short-term
exposures to high concentrations of pollutants (acute) or to long-
term exposures to lower concentrations of pollutants (chronic).
One of the noncancer health effects listed in Section 112(k) of the
Clean Air Act.
64
-------
Term
Quantification
Range of values
Reference con-
centrations
(RfCs)
Reliability
Revertants
Risk
Risk, absolute
Risk, compara-
tive
Risk assessment
Risk assessment
method
Risk estimation
Superfund
Amendments and
Reauthorization
Act (SARA) Title
m
Definition
The assignment of a number to an entity; a method for determin-
ing a number to be assigned to an entity; the act of determining,
indicating, or expressing the quantity of an item.
Evaluation of an uncertain outcome by estimation of maximal and
minimal values.
An estimate, with uncertainty spanning a factor of 10, of the con-
centration that could be inhaled for a lifetime with no adverse
health effects.
The probability that a system will perform its required functions
under conditions for a specified operating time.
A measure of mutagenicity in a short-term bioassay using bacte-
ria. Specifically, mutant strains of bacteria are exposed to pollut-
ants, and only those bacteria that mutate back, or "revert," to
their original genetic coding are able to survive and produce
colonies.
The probability of uncertain, undesirable consequences or out-
comes; having a chance of injury or loss.
A quantifiable estimate of a risk, based upon measurable and
observable data or statistics, without major assumptions or upper-
limit estimates.
An evaluation of the ranking of risks from a variety of causes in
relationship to each other.
The process of quantifying the level of risk associated with some
situation or action.
A systematic procedure or mode of inquiry that may be employed
as part of a risk assessment.
The process of characterizing uncertainty (i.e., quantification of
probabilities) and consequence values for risk.
Title m of the Superfund Amendments and Reauthorization Act
of 1986, also known as the "Emergency Planning and Com-
munity Right-to-Know Act of 1986," (EPCRA) which requires a
periodic (annual) inventory of toxic chemicals used, manufac-
tured, or processed in quantities above specified threshold
amounts at facilities in the U.S. [See Toxic Release Inventory.]
65
-------
Term
Definition
Screening (stud-
ies or hazards)
Secondary prod-
ucts
Short-term test
Smog
(photochemical
smog)
Smog chamber
Source category
Stationary source
Target dose
Technology
Teratogenicity
A preliminary process of hazard identification whereby a stan-
dardized procedure is applied to classify products, processes,
phenomena, or persons with respect to their hazard potential.
The products that are produced from the first, or primary, prod-
ucts; specifically, photochemical reactions in polluted air produce
primary products which themselves react further to produce the
secondary products, many of which are the more stable products
normally associated with smog.
Tests that take less time to complete than do other types of
bioassays. Many short-term tests measure the biological interac-
tions between the agent under test and deoxyribonucleic acid
(DNA). Agents that have effects hi short-term tests are generally
considered more likely to be health hazards than those that have
no effect.
Air pollution containing ozone and other reactive compounds
formed by the action of sunlight on nitrogen oxides and hydro-
carbons (or other organic precursors).
An experimental apparatus used to simulate the production of
photochemical "smog"; often a large, Teflon-lined enclosure or
bag, surrounded by lights that represent the sun's radiation or
open to natural sunlight, with connections for inserting and
withdrawing samples of pollutants.
A grouping of individual sources for consideration together
because of similarities in emissions, manufacturing processes, or
other factors.
A source of pollutants hi a fixed position, the location of which
can be specified.
The amount of HAP that directly impinges on tissues or organs
and induces a significant or toxic effect. Part of the consider-
ations under the Internal Dose component of the Environmental
Health Paradigm.
The tangible products of the application of scientific knowledge.
Production or induction of malformations or monstrosities,
especially of a developing embryo or fetus. One of the noncancer
health effects listed in Section 112(k) of the Clean Air Act.
66
-------
Term
Threshold
Toxicity
Toxic Release
Inventory (TRI)
Toxic substance
Uncertainty
Urban areas
Volatile organic
compounds
Definition
A discontinuous change of state of a parameter as its measure
increases. One condition exists below the discontinuity, and a
different one above it. In context of toxicity and this report,
exposures above a threshold produce effects, whereas exposures
below threshold do not produce effects.
Inherent ability of a substance to adversely affect living organ-
isms.
The TRI is an inventory of releases to air, water, and soil, or
transfers to treatment facilities of 322 toxic chemicals. The TRI
was mandated by the Emergency Planning and Community Right-
to-Know Act (EPCRA) of 1986 (also known as Title HI of the
Superfund Amendment and Reauthorization Act). Manufacturing
facilities that produce, import, or process 50,000 pounds or more
per year, or facilities that use 10,000 pounds or more per year of
the 322 chemicals specified in the EPCRA must report their
emissions annually to EPA. [See SARA Title IE.]
A substance of which exposure to humans or animals results in
deleterious effects.
A situation where there are a number of possible outcomes and
one does not know which of them has occurred or will occur;
indeterminacy; unpredictability; indefiniteness.
Areas in a city or town; areas that are city-like.
Two major definitions are common: (1) Under the regulatory
control program to limit production of ozone pollution, VOCs are
organic chemicals, usually hydrocarbons, that produce ozone at a
rate greater than ethane; (2) In a scientific sense, VOCs are
chemicals containing carbon that evaporate so readily that they
exist in the air as vapors.
67
------- |