EPA-600/8-80-0451
August 1980
Special Series
REVIEW OF CRITERIA
FOR VAPOR-PHASE
HYDROCARBONS
Environmental Criteria and Assessment Office
Office of Health and Environmental Assessment
Office of Research and Development
U.S. Environmental Protection Agency
Washington, D.C. 20460
-------
-------
EPA-600/8-80-045
August 1980
REVIEW OF CRITERIA
FOR
VAPOR-PHASE HYDROCARBONS
by
Beverly E. Tilton and Robert M. Bruce, Ph. D.
U.S. ENVIRONMENTAL PROTECTION AGENCY
Environmental Criteria and Assessment Office
Office of Health and Environmental Assessment
Office of Research and Development
Washington, D.C. 20460
-------
-------
FOREWORD
This document has been prepared pursuant to Section 108(c) of the
Clean Air Act, as amended, which requires that the Administrator from
time to time review and, as appropriate, modify and reissue criteria
issued pursuant to Section 108(a). Air quality criteria are required by
Section 108(a) to reflect accurately the latest scientific information
useful in indicating the kind and extent of all identifiable effects on
public health and welfare that may be expected from the presence of a
pollutant in the ambient air in varying quantities.
The original criteria document for hydrocarbons, AP-64, was issued
in 1970. Since that time, new information has been developed, and this
document represents the review and revision of pertinent air quality
criteria for hydrocarbons.
The regulatory purpose of this review is to serve as the basis for
national ambient air quality standards promulgated by the Administrator
under Section 109 of the Clean Air Act, as amended. Accordingly, as
provided by Section 109(d), the Administrator has reviewed the national
ambient air quality standards for hydrocarbons based on this review of
criteria and literature since 1970 and is proposing appropriate action
with respect to those standards concurrently with the issuance of this
document. :
The Agency is pleased to acknowledge the efforts and contributions
of all persons and groups who have contributed to this document as
participating authors or reviewers. In the last analysis, however, the
Environmental Protection Agency is responsible for its content.
DOUGLAS M. COSTLE
Administrator
U.S. Environmental Protection Agency
iii
-------
NOTICE
EPA Project Officer on this document is Dr. Robert M. Bruce. Correspondence
relating to the subject matter of the document should be addressed to:
Dr. Robert M. Bruce
Environmental Criteria and Assessment Office (MD-52)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
IV
-------
SCIENCE ADVISORY BOARD
CLEAN AIR SCIENTIFIC ADVISORY COMMITTEE
The substance of this document was reviewed by the Clean Air Scientific
Advisory Committee of the Science Advisory Board in public session.
Chairman:
Dr. Sheldon K. Fried!ander, Vice Chairman of Chemical Engineering,
Department of Chemical, Nuclear, and Thermal Engineering, School
of Engineering and Applied Science, University of California at
Los Angeles, California 90024
Members:
Dr. Mary 0. Amdur, Department of Nutrition and Food Science,
Room 16-339, Massachusetts Institute of Technology, Cambridge,
Massachusetts 02139
Dr. Robert Dorfman, Department of Economics, Harvard University,
325 Littauer, Cambridge, Massachusetts 02138
Mr. Harry H. Hovey, Jr. New York Department of Environmental
Conservation, 50 Wolf Road, Albany, New York 12233
Dr. Judy A. Bean, College of Medicine, Department of Preventive
Medicine and Environmental Health, University of Iowa, Iowa
City, Iowa 52242
Mr. Donald H. Pack, 1826 Opalocka Drive, McLean, Virginia 22101
Consultants
Dr. Herschel E. Griffin, Dean, Graduate School of Public Health,
Room A-629, Crabtree Hall, University of Pittsburgh, Pittsburgh,
Pennsylvania 15261
Dr. Michael Treshow, Department of Biology, University of Utah,
Salt Lake City, Utah 84112
SAB Staff Officer:
Mr. Terry F. Yosie, EPA Staff Officer, Science Advisory Board
(A-101), 401 M Street, SW, Washington, DC 20460
-------
ACKNOWLEDGMENTS .
Appreciation is extended to the following members of the Environmental
Criteria and Assessment Office Technical Services Staff, under the guidance of
Frances P. Bradow, for their contributions to the final production of this
document:
Mrs. Dela Bates
Mrs. Diane Chappell
Ms. Deborah Doerr
Mr. Douglas Fennel!
Mr. John Ferrell
Mrs. Bettie Haley
Mr. Allen G. Hoyt
Mrs. Constance van Oosten
Ms. Evelynne Rash
Mrs. Aimee Tattersall
Ms. Vickie Ellis, Environmental Protection Specialist, ECAO Scientific
Staff, also provided assistance in preparing this report.
The Audio-Visual Branch, under the direction of C. R. Rodriguez, and the
Printing Office, under the supervision of B. H. Poole, both within General
Services Division, Office of Administration, lent their respective support in
preparing figures and in printing this document.
vi
-------
AUTHORS AND REVIEWERS
The authors of this document were:
Ms. Beverly E. Til ton
Environmental Criteria and Assessment Office (MD-52)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Dr. Robert M. Bruce
Environmental Criteria and Assessment Office (MD-52)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
The following persons within EPA reviewed this document and
submitted comments:
Dr. Paul Altshuller .
Office of the Assistant Administrator for
Research and Development (MD-59)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Mr. Michael A. Berry
Environmental Criteria and Assessment Office (MD-52)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Mr. Francis M. Black
Environmental Sciences Research Laboratory (MD-46)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Dr. John Burchard
Industrial Environmental Research Laboratory (MD-60)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Mr. Alfred H. Campbell
Strategies and Air Standards Division,
Office of Air Quality Planning and Standards (MD-12)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Ms. Josephine Cooper
Office of the Assistant Administrator for
Research and Development (RD-672)
U.S. Environmental Protection Agency
Washington, DC 20460
-------
Dr. Basil Dimitriades
Environmental Sciences Research Laboratory (MD-59)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Dr. Harcia C. Dodge
Environmental Sciences Research Laboratory (MD-84)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Mr. Michael H. Jones
Strategies and Air Standards Division (MD-12)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Mr. Edward 0. Lillis
Monitoring and Data Analysis Division,
Office of Air Quality Planning and Standards (MD-14)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Dr. William A. McClenny
Environmental Sciences Research Laboratory (MD-47)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Mr. Frank McElroy
Environmental Monitoring Support Laboratory (MD-77)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Mr. John O'Connor
Strategies and Air Standards Division
Office of Air Quality Planning and Standards (MD-12)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Mr. Patrick K. O'Hare
Office of General Counsel (A-133)
U.S. Environmental Protection Agency
Washington, DC 20460
Mr. Joseph Padgett
Strategies and Air Standards Division
Office of Air Quality Planning and Standards (MD-12)
U.S. Environmental Protection Agency .
Research Triangle Park, NC 27711
Mr. Johnnie Pearson
Control Programs Development Division
Office of Air Quality Planning and Standards (MD-15)
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
viii
-------
Mr. Charles H. Ris III
Office of Health and Environmental
Assessment (RD-689)
Office of Research and Development
U.S. Environmental Protection Agency
Washington, DC 20460
Mr. Matthew Van Hook
Office of General Counsel (A-133)
U.S. Environmental Protection Agency
Washington, DC 20460
-------
-------
TABLE OF CONTENTS
Page
FIGURES xiii
TABLES ., xiv
ABSTRACT -.- xvi i
1. SUMMARY AND CONCLUSIONS ... ' 1-1
1.1 INTRODUCTION '... .. 1-1
1.2 BACKGROUND INFORMATION 1-3
1.3 CONTRIBUTION OF HYDROCARBONS TO OXIDANT FORMATION..... .. 1-8
1.4 MEASUREMENT METHODOLOGY.... . 1-9
1.5 SOURCES AND EMISSIONS 1-10
1.6 AMBIENT AIR CONCENTRATIONS 1-12
1.7 HYDROCARBON/OXIDANT RELATIONSHIP 1-15
1.8 HEALTH AND WELFARE EFFECTS OF HYDROCARBONS. ... 1-18
2. RECOMMENDATIONS .. 2-1
3. INTRODUCTION.... .'., 3-1
3.1 BASIS FOR AIR QUALITY CRITERIA AND NATIONAL AMBIENT AIR
QUALITY STANDARDS 3-1
3.2 BASIS AND NATURE OF PRESENT REVIEW OF CRITERIA FOR HYDRO-
CARBONS. 3-2
4. BACKGROUND INFORMATION 4-1
4.1 UNIQUE CHARACTERISTICS OF CRITERIA AND STANDARDS FOR
HYDROCARBONS 4-1
4.2 NEED FOR A CRITERIA DOCUMENT FOR HYDROCARBONS 4-3
5. SUMMARY OF 1970 CRITERIA FOR HYDROCARBONS 5-1
5.1 CONTRIBUTION OF HYDROCARBONS TO FORMATION OF PHOTOCHEMICAL
OXIDANTS 5-1
5.2 MEASUREMENT METHODOLOGY 5-4
5.3 SOURCES AND EMISSIONS OF HYDROCARBONS 5-7
5.4 AMBIENT AIR CONCENTRATIONS 5-8
5.5 HYDROCARBON/OXIDANT RELATIONSHIPS 5-17
5.6 HEALTH EFFECTS 5-23
5.6.1 Aliphatic Hydrocarbons 5-26
5.6.2 Alicyclic Hydrocarbons 5-26
5.6.3 Aromatic Hydrocarbons ; 5-30
5.6.4 Hydrocarbon Mixtures 5-32
5.7 WELFARE EFFECTS. 5-38
XI
-------
SCIENTIFIC DATA BASE ON HYDROCARBONS, 1970 THROUGH PRESENT.
6.1
6.2
6.3
6.4
6.5
6.6
6.7
CONTRIBUTION OF HYDROCARBONS TO OXIDANT FORMATION.
MEASUREMENT METHODOLOGY
SOURCES AND EMISSIONS OF HYDROCARBONS
6.3.1 General
6.3.2 Natural Sources and Emissions
6.3.3 Manmade Sources and Emissions
AMBIENT AIR CONCENTRATIONS
HYDROCARBON-OZONE RELATIONSHIP
6.5.1 Factors Affecting the Hydrocarbon-ozone Relationship.
6.5.2 Models for Determining Hydrocarbon Emission
Reductions
HEALTH EFFECTS :
6.6.1 Aliphatic Hydrocarbons
6.6.2 Alicyclic Hydrocarbons
6.6.3 Aromati c Hydrocarbons
6.6.4 Hydrocarbon Mixtures—Gasoline
6.6.5 Miscellaneous Hydrocarbons..
6.6.6 Health Effects Summary ,
WELFARE EFFECTS.
6.7.1 Welfare Effects Summary.
REFERENCES
6-1
6-1
6-8
6-19
6-19
6-31
6-39
6-94
6-131
6-135
6-144
6-149
6-150
6-153
6-159
6-163
6-167
6-170
6-176
6-182
7-1
xii
-------
LIST OF FIGURES
Figure Page
5-1 Nonmethane hydrocarbons by flame ionization analyzer, averaged
by hour of day over several months for four cities. 5-16
5-2 Hourly aldehyde concentrations at two Los Angeles sites,
October 22, 1968 5-19
5-3 Maximum daily oxidaht as a function of early morning nonmethane
hydrocarbons, 1966-1968, for CAMP stations; May through October
1967 for Los Angeles 5-21
5-4 Required hydrocarbon emission control as a function of
photochemical oxidant concentration „.. 5-24
6-1 Density of population in U.S. in 1975 by state, no. people/mi. .. 6-24
6-2 Density of total nonmethane hydrocarbon emissions in U.S. in
1975 by state, tons/mi. 6-25
6-3 Effects of fuel composition on THC and ethanol emission rates 6-80
6-4 Distribution of diesel hydrocarbon exhaust emissions, by molecular
weight, between gas-phase and particulate forms 6-84
6-5 Distribution of hydrocarbons in diesel fuel and lubricant
according to molecular weight 6-85
6-6 Nonmethane hydrocarbon trends in Los Angeles, 1963 through 1972.. 6-100
6-7 Sum of nonmethane hydrocarbon concentrations versus number of
samples for two ground sites used in the 1974 Midwest Oxidant
Transport Study 6-110
6-8 Occurrence of gas-phase hydrocarbons, by carbon number, in
gasoline, exhaust, and urban ambient air 6-132
6-9 Calculated gasoline threshold limit value, reflecting impact
of present vs. newly promulgated TLV standard as a function
of the liquid volume percent benzene in the gasoline 6-168
6-10 Geographic location of the grab sample collection sites 6-179
xiii
-------
LIST OF TABLES
Table , , page
5-1 Some hydrocarbons identified in ambient air 5-10
5-2 Average hydrocarbon compositon, 218 ambient air samples, Los
Angel es, 1965 5-12
5-3 Average and highest concentration measured for various aromatic
hydrocarbons in Los Angeles, 26 days, September through November,
1966 5-13
5-4 Range of yearly maximum 1-hour average concentrations of
aldehydes, Los Angeles County stations, 1951 through 1957 5-15
5-5 Average formaldehyde and acrolien concentrations by time of day
in Los Angeles, September 25 through November 15, 1961 5-18
5-6 Toxicity of saturated al iphatic hydrocarbons 5-27
5-7 Toxicity of unsaturated al iphatic hydrocarbons 5-29
5-8 Toxicity of alicyclic hydrocarbons: Comparative effects of
chronic and acute exposure in air 5-31
5-9 Toxicity of aromatic hydrocarbons: Comparative effects of acute
and chronic exposure in air , 5-33
5-10 Principal mixtures containing paraffin hydrocarbons 5-37
5-11 Human response to gasoline vapors distilling below 230°F 5-39
5-12 Comparative phytotoxicity of unsaturated carbon gases 5-41
5-13 Dosage-response relationships of various plants to ethylene 5-43
5-14 American Industrial Hygiene Association recommended standards
for ethyl ene 5-44
6-1 Possible atmospheric gas phase S0? oxidation processes 6-7
6-2 Percentage difference from known concentrations of nonmethane
hydrocarbons obtained by sixteen users 6-9
6-3 Summary of national estimates of volatile organic emissions,
1970-1977 6-21
6-4 National estimates of volatile organic emissions, 1977, by
source category 6-22
6-5 Comparison of selected Houston emission rate and ambient air data 6-34
6-6 Comparison of Houston and Tampa/St. Petersburg TNMHC emission
rates 6-35
6-7 Biogenic HC emission rates at constant environmental conditions.. 6-37
6-8 Comparison of monoterpene emissions from various plant species... 6-38
6-9 Emissions of volatile organic compounds from stationary external
combustion sources 6-41
6-10 Emissions of volatile organic compounds from stationary source
internal combustion engines 6-42
6-11 Emissions of volatile organic compounds from manufacture of
selected chemicals/products 6-44
6-12 Emissions of volatile organic compounds from sanitary landfills.. 6-45
6-13 Emissions of volatile organic compounds from general use of
domestic solvents 6-47
6-14 Emissions of volatile organic compounds from domestic and
commercial use of pesticides 6-48
xiv
-------
Table
Page
6-15 Emissions of volatile organic compounds from the use of
architectural surface coati ngs 6-50
6-16 Emissions of volatile organic compounds from forest fires and
from open burning of agricultural/landscape wastes 6-51
6-17 Hydrocarbon emissions from agricultural burning of field and
orchard crop residues 6-52
6-18 Nationwide estimates of VOC emissions from transportation sources,
1970 through 1977 6-54
6-19 Hydrocarbon exhaust emission factors for light-duty, gasoline-
powered vehicles for all areas except California and high-
altitude. . ". 6-56
6-20 Composite crankcase and evaporative emission factors for light-
duty, gasoline-powered vehicles for all areas except California
• and high-al titude 6-56
6-21 General Motors Corporation systems for control of hydrocarbon
and other pollutant emissions 6-57
6-22 Summary data on gasoline composition, reported as weight percent. 6-62
6-23 Average gasoline vapor composition,..... 6-63
6-24 Predominant hydrocarbons in exhaust emissions from gasoline-
fueled autos 6-65
6-25 Summary of exhaust emission data for uncontrolled and catalyst-
equipped gasoline-fueled cars by model year 6-67
6-26 Exhaust, evaporative, and aggregate emissions of total hydro-
carbons, benzene, and toluene from four passenger cars (1963
through 1979 model years) and four fuels 6-73
6-27 Increase in exhaust emission rates of aromatic aldehyde for
non-catalyst gasoline cars with increase in fuel aromaticity 6-74
6-28 Total hydrocarbon emission rates on (ECE test cycle) for car-
fueled with gasoline, methanol, or a 15% methanol-85% gasoline
mixture 6-75
6-29 Exhaust emissions from vehicle equipped with three-way conversion
catalyst and fueled with gasoline, ethanol-gasoline, or methanol-
gasoline 6-77
6-30 Test fuel specifications, 6-79
6-31 Emission rates for hydrocarbons in exhaust from two diesel .
vehi cl es 6-86
6-32 Aldehyde emissions from diesel vehicles operating on five
different fuels, 1975 FTP 6-89
6-33 Summary of reactivity of diesel exhaust hydrocarbon emissions
and percentage contributions of lower-molecular-weight hydro-
carbons and al dehydes 6-91
6-34 Summary of emission characteristics for autos fueled by gasoline
diesel and alcohol-gasoline or ether-gasoline blends 6-92
6-35 Ambient air concentrations of hydrocarbons reported by Graedel... 6-96
6-36 Frequency distributions for 6-to-9 a.m. nonmethane hydrocarbon
concentrations at CAMP sites, 1967-1972 6-99
6-37 Trends, in 6-to-9 a.m. concentrations of indicator hydrocarbons
at two sites in the California South Coast Air Basin,
1967 through 1977 6-102
xv
-------
Table
6-38 Maximum and average 6-to-9 a.m. nonmethane hydrocarbon
concentrations at four sites in California South Coast Air
Basins June through September 1975 6-103
6-39 Total nonmethane hydrocarbon concentrations in 2-hour samples
measured by gas chromatography at urban site in St. Louis for 6
days, September 13-25, 1972 6-105
6-40 Total nonmethane hydrocarbon concentrations in 2-hour samples
measured by gas chromatography at urban site in St. Louis for 4
days, June 21-July 8, 1973 6-106
6-41 24-hour average hydrocarbon concentrations measured by gas
chromatography at urban site in St. Louis, September 13, 1972 6-107
6-42 24-hour average individual hydrocarbon measurements at
Wilmington, Ohio, July 18, 1974, determined by gas
chromatography 6-111
6-43 Concentrations of total nonmethane hydrocarbons, methane,
and selected oxygenates in eleven 2-hour and two 1-hour
samples taken over 24 hours, July 18, 1974, Wilmington, Ohio 6-113
6-44 Concentrations of natural hydrocarbons in ambient air samples
taken within and outside a forest conopy 6-115
6-45 24-hour average individual hydrocarbon concentrations
measured by gas chromatography, Boston area, 1975 6-117
6-46 Total nonmethane hydrocarbon concentrations in Houston by
time of day for 23 days, July 3-25, 1976, determined by FID 6-120
6-47 Concentrations of hydrocarbons by class in air samples collected
at 19 sites in Houston, 1973 and 1974, determined by gas
chromatography 6-121
6-48 Concentrations of individual hydrocarbons in air samples
collected in Houston on September 11, 1973, determined by gas
chromatography 6-122
6-49 Average percentage composition of hydrocarbons in ambient air
at selected sites in Houston (1978) and St. Louis (1972
and 1977) 6-to-9 a.m. period 6-126
6-50 Effect of wind direction on hydrocarbon composition at site 03,
Houston 6-127
6-51 Average concentrations of benzene in ambient air at four
California sites, Summer 1975 6-130
6-52 Mean values and ranges in aromatic hydrocarbon concentrations
in West Germany, July 1976 6-131
6-53 Some factors that affect the photochemical formation of ozone
from precursors 6-136
6-54 Comparison of reactivities of different types of organics 6-141
6-55 Classification of organics with respect to oxidant-related
reactivity in urban atmospheres 6-143
6-56 Effects of al kane vapor pressure on humans 6-154
6-57 Effects of alkane vapor pressure on animals 6-156
6-58 Human experience: exposure to gasoline vapors 6-166
6-59 Suggested hygienic standard for various hydrocarbon solvent
mixtures based on inhalation toxicity studies of animals and
sensory response of human subjects 6-171
6-60 Houston study ethylene levels 6-178
xvi
-------
ABSTRACT
Information on vapor-phase hydrocarbons presented in this document covers
basic atmospheric chemistry relative to secondary products, especially ozone;
sources and emissions; ambient air concentrations; relationship of precursor
hydrocarbons to resultant ozone levels in ambient air; health effects; and
welfare effects.
The principal conclusions from this document are as follows. Vapor-phase
hydrocarbons, along with many other volatile organic compounds, contribute to
/
the formation of ozone and other photochemical oxidants. The relationship
between precursor hydrocarbons and consequent ambient air levels of ozone and
other oxidants is complex, however, and varies from site to site, city to
city, and year to year, depending on local precursor mixes, transport
considerations, and meteorological factors. Consequently, no single
quantitative relationship between hydrocarbons and ozone/oxidants can be
defined nationwide.
While specific vapor-phase hydrocarbon compounds can be of concern to
public health (benzene) and welfare (ethylene), as a class, this group of
compounds cannot be considered a hazard to human health or welfare at or even
well above those concentrations observed in the ambient air.
xvn
-------
-------
1. SUMMARY AND CONCLUSIONS
1.1 INTRODUCTION
Section 109(d)(l) of the Clean Air Act, added by the 1977 amendments,
requires that EPA thoroughly review each 5 years the criteria for those pollutants
listed under Section 108 and regulated by National Ambient Air Quality Standards
(NAAQS) under Section 109. According to this requirement, the first such
review of the criteria and standards for hydrocarbons must be completed by
December 31, 1980. Most reviews of the criteria for regulated pollutants have
led to revisions of the existing criteria documents. In the revision process,
thorough evaluation is made of comprehensive scientific information that will
form the basis for a decision on the need for and nature of any revisions to
the National Ambient Air Quality Standard (NAAQS) for the pollutant under
consideration. As EPA proceeded in its review of information on hydrocarbons,
it became apparent that revisions to the original criteria document were not
appropriate. Rather, EPA's review indicated that the unique nature of the
criteria and standards for hydrocarbons made the usefulness of revising
the original criteria document questionable, and indicated instead that a
sufficient basis for regulatory decisions relative to the hydrocarbon standard
could be established by means of a thorough review of current scientific data
followed by the identification of certain key facts, or pivotal criteria, for
hydrocarbons. Chapter 4 presents the rationale behind preparation of this
document and use of this approach in developing current criteria for hydrocarbons.
The title of the external review draft of this document (February 1980) was
not fully descriptive, since both that draft and this document develop and present
those criteria pertinent to a review of the NAAQS for hydrocarbons. The present
1-1
-------
document statisfies the requirements for the review of criteria specified in
Section 109(d)(l) of the Clean Air Act. It should be noted that the Clean Air
Act does not specify the form in which such reviews should be published or the
form in which resulting criteria should be issued.
The 1970 air quality criteria document covered only those organic compounds
that are composed solely of carbon and hydrogen and that occur in the atmosphere
in the vapor phase. These compounds, hundreds of which have been identified
as being emitted into the atmosphere, are collectively referred to as hydro-
carbons. Hydrocarbons are a specific class of organic compounds, so that all
hydrocarbons are organic compounds but not all organic compounds are hydro-
carbons. The 1970 criteria document excluded compounds such as substituted or.
derivatized hydrocarbons, e.g., halogenated hydrocarbons. It also excluded
all organic compounds, hydrocarbon or nonhydrocarbon, that occur in the
atmosphere in aerosol or particulate form. It included a brief treatment of
aldehydes to ensure coverage of this class of secondary pollutants formed from
hydrocarbons in the atmosphere. It should be noted, however, that no standard
was subsequently promulgated for aldehydes and that the EPA reference method
for hydrocarbons does not measure aldehydes.
Similarly, this issue paper covers only volatile hydrocarbons as defined
in the 1970 criteria document and, thus, as covered by the 1971 NAAQS for
hydrocarbons. Aldehydes are generally excluded from this paper since they are
not hydrocarbons, since they are not covered by the NAAQS for hydrocarbons,
and since the 1978 criteria document for ozone and other photochemical oxidants
included information on the photochemistry and health effects of aldehydes.
Furthermore, the National Academy of Sciences is presently conducting a review
of the atmospheric chemistry and effects of aldehyde air pollutants for EPA. In
1-2
-------
addition, EPA is conducting a thorough review o.f the health-related data on
formaldehyde. At the conclusion of the MAS review, EPA will determine whether
further assessment should be conducted as a basis for possible regulatory
action on aldehydes. The present EPA assessment of formaldehyde will determine
whether health-related regulatory action should be pursued for this specific
aldehyde. Also, EPA is currently preparing separate documents that assess the
health effects of a number of nonhydrocarbon volatile organic compounds, e.g.,
perchloroethylene, trichloroethylene, ethylene dichloride, acrylonitrile, and
vinylidene chloride.
1.2 BACKGROUND INFORMATION
The class of compounds known as hydrocarbons is unique among the seven
pollutants or classes of pollutants listed under Section 108 and regulated
under Section 109 of the Clean Air Act. All pollutants regulated by NAAQS
except for hydrocarbons are thought to have direct deleterious effects on
public health, and, in some instances, on public welfare as well. The criteria
developed in 1970 for hydrocarbons showed clearly, however, that hydrocarbons
do not directly produce deleterious effects on public health or welfare.
Instead, the criteria showed that hydrocarbons indirectly cause adverse health
and welfare effects through their contribution to the formation of photochemical
oxidants in general and of ozone in particular. Thus, hydrocarbons were the
only pollutant or pollutant class regulated under Section 109 solely for the
purpose of controlling their secondary atmospheric products.
This approach resulted in an NAAQS that is unique among the seven existing
NAAQS in the following respects: (1) it was not based on the health or welfare
effects of hydrocarbons, either singly or as a class; (2) it was accompanied
by a measurement method that is not as specific for the regulated pollutant as
the methods available for other criteria pollutants; (3) it was intended to
1-3
-------
serve solely as a guide in helping States determine hydrocarbon emission
reductions needed to attain the 1971 oxidant standard; and (4) it was not
intended to be an enforceable standard having the same regulatory weight and
function as the other standards. Because of its intended use as a guideline,
it has not been enforced, no State Implementation Plans for its attainment
have been required, and no routine monitoring of ambient air levels of
nonmethane hydrocarbons has been required under the hydrocarbon standard. In
keeping with its use in achieving the oxidant standard (which is now a standard
for ozone), the level selected for the 1971 NAAQS for hydrocarbons was fixed
in relation to the level selected for the oxidant standard, this relationship
having been determined through the application of an empirical model derived
from aerometric data in conjunction with the atmospheric dispersion-related
equation used in the linear rollback model.
The basis for the existing NAAQS for hydrocarbons, then, is the role of
hydrocarbons, along with other volatile organic compounds, as precursors to
ozone and other photochemical oxidants. From the review of current scientific
information on hydrocarbons that EPA conducted to prepare this document, two
facts were established, one qualitative and one .quantitative, concerning the
basis for the existing hydrocarbon NAAQS:
1. According to present knowledge, vapor-phase hydrocarbons—as a
class—contribute to the formation of ozone and other photochemical
oxidants in the ambient air.
2. According to present knowledge, no fixed level of nonmethane hydro-
carbons can be used nationwide to ensure the attainment and
maintenance of the ozone standard.
The first of these facts was clearly established by the 1970 criteria
document and confirmed when the criteria for ozone and other photochemical
1-4
-------
oxidants were revised in 1978. The concept embodied in the second fact above
is whether the quantitative relationship between ozone concentrations in
ambient air and hydrocarbon concentrations in ambient air is fixed and constant
so that an NAAQS for hydrocarbons can be set in relation to the ozone NAAQS.
While the concept itself was not explicity addressed in the 1978 ozone criteria
document, it was implicit in that document and in the promulgation by the
Administrator of EPA in 1979 of an ozone standard without the concomitant
revision of the hydrocarbon standard. This more general concept relative to
the quantitative aspect of the basis of the current standard is addressed in
this document. The replacement by EPA in 1979 of the Appendix J model by
other models clearly established the deficiencies of the specific method used
in 1970 to quantify the relationship between precursor hydrocarbons and ozone
and other photochemical oxidants.
The review of literature undertaken in preparation of this document
revealed no evidence that would alter the findings of the ozone document
relative to oxidant-precursor relationships. This document fully corroborates
that the qualitative aspect of the basis for the NAAQS for hydrocarbons
remains valid. Hydrocarbons in ambient air are major precursors to ozone and
other photochemical oxidants in ambient air and must be controlled in order to
reduce ambient air levels of ozone and othe photochemical oxidants. However,
the quantitative aspect of the basis of the present NAAQS for hydrocarbons
was not substantiated in the 1978 ozone criteria document, a finding that is
also fully corroborated by this document.
The two basic facts cited above, which are discussed in Sections 5 and 6,
are key criteria required for regulatory analyses and decisions regarding the
basis for the current hydrocarbon NAAQS. However, no review of criteria for
1-5
-------
hydrocarbons could be considered complete without the inclusion of the answer
to a crucial question relating to the possible need for a hydrocarbon class
standard on a new basis. That question is:
3. According to present knowledge, do vapor-phase hydrocarbons—as a
class—cause adverse effects on public health or welfare at or near
levels found in the ambient air?
The answer to this question provides the third key criterion needed for a
regulatory decision relative to an NAAQS governing hydrocarbons as a class.
The data base from 1970 to the present has been thoroughly reviewed to ascertain
the answer to this crucial question.
The following additional questions, pertinent but subsidiary, are also
addressed, using the same data base:
4. What analytical methods are available for ambient air hydrocarbons?
How accurate and specific are they?
5. What is the relationship between measurements of hydrocarbon mass in
the ambient air and consequent oxidant pollution? That is, do
monitoring methods for hydrocarbons reflect the oxidant-forming
potential of hydrocarbon mixtures in the ambient air?
6. What are the major sources of hydrocarbon emissions to the atmosphere?
7. What kinds of hydrocarbons are emitted from these sources?
8. v What are the levels of emissions from the major sources?
9. What are the ambient air levels of hydrocarbons as a class and of
certain individual hydrocarbons or subclasses of hydrocarbons?
10. Have any subclasses of vapor-phase hydrocarbons (aliphatics, alicyclics,
aromatics, or mixtures) been demonstrated to have adverse health or
welfare effects—as a subclass? If so, what kind and at what levels?
1-6
-------
11. Have any individual hydrocarbons been shown to have adverse health
or welfare effects? If so, what kind and at what levels?
The answers to these questions have been presented in this document
(Chapters 5 and 6) because of their bearing on the three key criteria addressed.
Ambient air data in the form of concentrations of total nonmethane hydrocarbons;
of total paraffins, olefins, and aromatics; and of individual species of
hydrocarbons have been presented mainly for contrast with the levels reported
in the health and welfare effects sections. EPA must be able to relate health
and welfare effects data with data on the levels and composition of atmospheric
burdens of hydrocarbons.
In order to assess properly the validity and relevance of the ambient air
data, at least some information has to be presented on measurement methodology,
since any errors or problems in methodology must be taken into account when
relating ambient air data to effects information. Thus, a review of the basic
measurement methods and the relative merits of those methods has been presented.
A fairly lengthy section on sources and emissions of hydrocarbons, including
some nonhydrocarbon organic compounds, has been included in this issue paper.
It is helpful for the EPA regulatory office to have a summary of sources and
their emissions and to be aware of possible trends in technology that could
result in changes in the composition or levels of ambient air hydrocarbons.
Fuel conversion in stationary combustion sources, changes in auto emission
controls, possible fuel conversion for mobile sources, or an increase in the
proportion of diesel vehicles in operation are examples of the kind of technologic
changes that would influence the composition or levels of hydrocarbons emitted.
Consequently, a review of what is presently known on sources, levels, and
kinds of emissions of hydrocarbons has been presented in this document.
1-7
-------
Health effects information has been presented according to subclasses of
hydrocarbons to permit identification of any associations between compound
subclass and any possible adverse health effects.
The sections that follow summarize the information given in the body of
this report and are presented in accordance with the organization of Chapters
5 and 6.
1.3 CONTRIBUTION OF HYDROCARBONS TO OXIDANT FORMATION
The information presented in Sections 5.1 and 6.1 substantiates that
most, if not all, vapor-phase hydrocarbons participate in a complex series of
photochemical reactions that give rise to the formation of ozone and other
photochemical oxidants in ambient air. Hydrocarbons are involved in the
formation of ozone and other photochemical oxidants through, the generation,
from hydrocarbons, of alkylperoxy and hydroperoxy radicals that react with
nitric oxide to form nitrogen dioxide (NOp). The increase in production of
NOg brought about by the radicals generated from hydrocarbon oxidation results
in a greater abundance of atomic oxygen that subsequently reacts with molecular
oxygen, in the presence of a third body, to form ozone (Sections 5.1 and 6.1).
Evidence accumulated from both experimental and theoretical research over
the past decade strongly supports the position that reducing the emissions of
hydrocarbons will reduce ambient air concentrations of ozone. Research results
continue to indicate, however, that the hydrocarbon-ozone relationship is
exceedingly complex and that it is dependent upon many factors, including the
ratio of nonmethane hydrocarbons to nitrogen oxides. Research has increasingly
corroborated over the past decade that volatile organic compounds other than
hydrocarbons are also photochemically reactive in the production of ozone and
other photochemical oxidants. The first fact corroborated in this document is
1-8
-------
that vapor-phase hydrocarbons, along with other compounds, do indeed serve as
precursors to ozone and other photochemical oxidants in the ambient air.
Hydrocarbons are also precursors to other secondary products of atmos-
pheric reactions, namely, aerosols and acid precipitation.
Evidence to date indicates that unsaturated hydrocarbons, specifically
alkadienes and cyclic alkenes, are important precursors to secondary aerosols
(aerosols formed in the atmosphere and not emitted as aerosols). Data show,
however, that only a small fraction of the vapor-phase hydrocarbons in ambient
air is converted into particulate matter. Though specific hydrocarbons differ
in their capacity to form aerosols (Section 6.5), control of hydrocarbons for
the sake of ozone/oxidant control will suppress the ambient air levels of
those specific hydrocarbons leading to the formation of secondary aerosols.
Hydrocarbons participate in the acidification of rain through their role
as precursors to hydroperoxy radicals (H02*) and, thus, hydrogen peroxide,
which under certain atmospheric conditions oxidizes sulfur dioxide to sulfur
trioxide. In the presence of moisture in the atmosphere, sulfur trioxide
forms sulfuric acid.
1.4 MEASUREMENT METHODOLOGY
Routine measurements of total and nonmethane hydrocarbons in .ambient air
are made using continuous monitoring analyzers that employ a flame-ionization
detector (FID) as the sensing element. This method is the EPA-adopted method
promulgated when the 1971 NAAQS for HC was promulgated. Nonmethane hydro-
carbon concentrations are determined by separating methane chromatographically
from the remaining hydrocarbons in the ambient air sample and determining the
methane concentration separately. Subtraction of the value for methane yields
the nonmethane hydrocarbon concentration. The methane concentration is of no
particular interest since methane is photochemically nonreactive.
1-9
-------
The FID is sensitive to a fraction of a part-per-million (ppm) carbon,
expressed as methane. The FID is not specific for hydrocarbons alone but
responds to any C-H or C-C bonds. Thus, though it is often referred to as a
nonmethane hydrocarbon (NMHC) method, the FID method is more appropriately
called a nonmethane organic compound (NMOC) method. The values obtained by
this method are a measurement of the mass of the hydrocarbons in ambient air,
Q
expressed as ppm carbon or as micrograms of carbon per cubic meter (ug/m ).
The values give no indication of the kind of hydrocarbons measured or the mole
concentration of total hydrocarbons measured.
A collaborative study in 1974 by 16 different users of the EPA reference
method (FID) showed that at the level of the NAAQS for HC, 160 pg/m3 (0.24
ppm) C, the majority of the measurements were in error by 50 to 100 percent,
with no consistent bias. At higher hydrocarbon concentrations, the measurement
error was considerably less, only 0 to 20 percent.
Gas chromatography (GC) is a more specific method than the FID analyzer
and is the only method available for measurement of individual hydrocarbons.
It is more specific because individual compounds can be identified and peaks
resulting from the presence of orgarvic compounds other than hydrocarbons can
be excluded from the determination of total nonmethane hydrocarbons. GC
techniques are sensitive to the partner-billion (ppb) level. GC techniques
are more sophisticated and time-consuming, however, than the FID analyzer and
do not lend themselves to continuous monitoring or to long-term analysis of
hydrocarbons in ambient air.
1.5 SOURCES AND EMISSIONS
As noted in the preceding section (1.4), FID measurement methods are not
specific for hydrocarbons but may give a positive response to the presence of
1-10
-------
many other organic compounds, even though that response may be a reduced
response. Also, as noted in Section 1.3, many volatile organic compounds
besides hydrocarbons participate in the atmospheric photochemical reactions
that give rise to ozone and other photochemical oxidants. The Environmental
Protection Agency in recent years has for these reasons included hydrocarbon
source and emissions data under the broader category of volatile organic
compounds (VOC).
Volatile organic compounds are emitted from both natural and manmade
sources. Methane emissions from natural sources in the United States have
been estimated at about 4.5 x 10 metric tons/year (5 x 10 tons/year).
Though its role in photochemical smog formation is under renewed investigation,
evidence to date indicates that methane plays no significant role in the
photochemical formation of oxidants and other smog components. Emissions of
isoprene and terpenes from vegetation in the United States have been estimated
to be about 6.4 x 10 metric tons/year. Isoprene and terpene emission densities
are not temporally or spatially uniform in the United States. Rather, about
43 percent of such emissions occur in the summer months and about 45 percent
of the annual isoprene and terpene emissions occur in the Southern United States.
The role of natural (vegetational) hydrocarbons in the formation of ozone is
controversial at present. Detailed discussions of this controversy are not
included in this document but are available in the open literature. The issues
in question were addressed in the 1978 criteria document for ozone and will again
be treated when that document is revised again.
On a nationwide basis, manmade emissions of volatile organic compounds
from all sources were estimated to be 28.3 x 10. metric tons/year in 1977, a
decrease of about 3 percent since 1970. In 1977, the major source of volatile
organic emissions was transportation (41 percent), followed by industrial
1-11
-------
processes (36 percent), and miscellaneous organic solvent use (3.7 percent).
Emissions due to evaporative losses and industrial processes increased from
1970 through 1977, but emissions from highway vehicles decreased by about 7
percent over the same period as a result of the use of control devices. The
decrease in vehicle emissions occurred in spite of an estimated 30 percent
increase in motor vehicle miles traveled during that period and reflect
intensive effort on the part of the auto industry to reduce hydrocarbon and
other pollutant emissions from autos and other highway vehicles.
Within the categories of industrial processes and miscellaneous sources
of VOC emissions, chemical manufacturing, industrial solvent use, and miscellaneous
solvent use (domestic and commercial) together resulted in VOC emissions of
about 9.1 x 106 metric tons in 1977, nearly the total of VOC.emissions produced
by highway vehicles (9.9 x 10 metric/tons year). As stated earlier, hydrocarbons
are just one class of those compounds categorized by EPA as volatile organic
compounds. A rough estimate of the percentage of VOC emissions that are
actually hydrocarbon emissions may be obtained by assuming that emissions from
chemical manufacturing and from industrial and miscellaneous organic solvent
use are 100 percent non-hydrocarbon VOC. While this assumption is not wholly
accurate, it is also true that exceptions to it are more than likely offset by
non-hydrocarbon VOC emissions such as oxygenated or even halogenated hydrocarbons
from highway vehicles, stationary source fuel combustion, oil and gas production
and marketing, and miscellaneous combustion sources. Given the basic assumption,
non-hydrocarbon emissions may represent as much as 32 percent of the total VOC
emissions estimated for 1977 (9.1 x 106 MT/yr out of 28.3 x 106 MT/yr).
1.6 AMBIENT AIR CONCENTRATIONS
The monitoring of hydrocarbons that is conducted by the states is for the
purpose of determining reductions in HC emissions needed to control oxidant
1-12
-------
levels and is. not required for the purpose of enforcing the HC NAAQS. Hydro-
carbon monitoring by the states is conducted mainly to obtain ambient air data
that can be used as input to the respective models used to determine the
percentage reductions in HC emissions that are needed to attain and maintain
the NAAQS for ozone.
Since routine monitoring has not been required, most of the data available
have been obtained from field studies conducted by EPA. Data presented in
this document were obtained from these field studies as well as from the open
literature. Ambient air data are generally reported as parts per million or
parts per billion carbon (ppm or ppb C), whereas data reported by physicians
and toxicologists in animal and human exposure studies are generally expressed
as ppm or ppb compound.
In a recent compendium of ambient air data on gas-phase hydrocarbons by
Graedel (Section 6.4), the detection of over 1000 organic compounds is documented
according to class of hydrocarbon and source. The most abundant hydrocarbon
by far is methane, with a concentration range of 1.3 to 4 ppm; ethylene and
acetylene are second and third, respectively, with concentrations in the ppb
range. The most abundant of the alkanes are ethane, propane, butane, pentane,
and isopentane, with upper limits all above 50 ppb but less than 200 ppb. The
most abundant unsaturated alkenes and alkynes are ethylene and acetylene,
which are typically present at concentrations less than about 50 ppb but have
been detected at concentrations as high as 227 to 700 ppb. In comparison to
all the other hydrocarbon classes, the alicyclic compounds are very limited in
number and their combined concentrations are quite low. Excluding benzene,
toluene, and the xylenes, the upper limit of the concentrations of individual
aromatics is less than 22 ppb. Benzene concentrations range from 0.025 to 57 ppb;
toluene ranges from 0.005 to 129 ppb; and the xylenes range from 0.5 to 61 ppb.
1-13
-------
Many of the published detailed studies of organic compounds in the urban
atmosphere were conducted in the 1960s. Results of those studies were summarized
and discussed in the 1970 criteria document for hydrocarbons. Published data
on total NMHC levels in urban atmosphere from 1970 through the present are
relatively abundant, but most of these data suffer from considerable measurement
error since they were obtained with FID instrumentation. Data obtained by EPA
with gas chromatography, as well as FID, in field studies in St. Louis and
Houston are representative of urban atmospheres; however, the atmospheric
hydrocarbon burdens of these two cities are influenced by different sources.
St. Louis is dominated by automotive-related sources whereas Houston is influenced
by both automotive and industrial emissions. Data reported in this document .
for the greater Boston area are representative of HC concentrations found in
suburban areas and data reported for Wilmington, Ohio, are representative of
levels found in rural areas. The Houston area has a greater number of
hydrocarbon-polluting industries than St. Louis or the Boston area (suburban
sites), and certainly more than the rural area represented by Wilmington,
Ohio, even though local influences there include a refinery.
In general, the individual hydrocarbon concentrations observed in the
field studies conducted by EPA were less than the upper limits quoted by
Graedel (Section 6.4), though in a few cases the values were higher. Typical
concentrations of total nonmethane hydrocarbons (EPA studies) in urban
atmospheres are in the range of 0.1 to 3 ppm C. Concentrations of total
nonmethane hydrocarbons in rural areas range from 0.04 to 0.2 ppm C.
As the data presented in Section 6.4 show, natural hydrocarbons are not
detected in ambient air in urban areas nor even at most sites outside of
forest canopies in rural areas. Data given in Sections 6.3 and 6.4 show that
1-14
-------
concentrations of natural hydrocarbons fall off rapidly with distance from the
vegetation that emits them. Measurements by Westberg, for example, show no
detectable levels of natural hydrocarbons outside the forest canopy, whereas
levels of hydrocarbons associated with manmade sources (benzene and toluene)
are virtually the same within and outside the canopy (Section 6.4). Westberg
reported average natural hydrocarbon concentrations of 0.2 ppb C in air sampled
within an Idaho conifer forest. At rural sites in the east, terpene concentrations
of 0.5 to 2 ppb C and isoprene concentrations of about 2 ppb C have been
reported, with total NMHC concentrations of about 100 ppb C at these sites.
Other measurements, made at various rural and remote areas in the southeast,
show ambient air concentrations of total natural hydrocarbons (isoprene and/or
terpenes) to range from < 4 to 10 ppb C, with total NMHC averaging 40 to 100
ppb C. All of the measurements showing detectable levels of natural hydrocarbons
in rural and remote areas have been made within canopies or very short distances
from forests or large stands of trees. In Wilmington, Ohio, measurements made
at a site in an open field showed no detectable levels of natural hydrocarbons.
1.7 HYDROCARBON/OXIDANT RELATIONSHIP
The qualitative aspects of hydrocarbon-oxidant relationships are better
understood now than in 1970 when the hydrocarbon criteria document was published.
Factors affecting the atmospheric photochemical processes were treated in
detail in the 1978 criteria document for ozone and other photochemical oxidants.
That document also gave in detail other qualitative and quantitative information
on the precursor-oxidant relationship, including information on the photochemical
products to be expected from specific organic precursors and the reaction
rates, where known, for the formation of those products. Such information has
therefore been discussed only briefly in this paper. Among those physical
1-15
-------
factors having important, even profound, influence on the photochemical formation
of ozone and other photochemical oxidants are: temperature, sunlight intensity,
spectral distribution of sunlight, occurrence and persistence of inversions,
and other meteorologic, geographic, or climatologic factors. Among the chemical
factors that are important influences are the initial precursor concentrations
(including background levels); composition of the NMHC mix in ambient air and
the relative reactivities of the species present; and the NMHC/NOV ratio. (In
f\
this context, NMHC actually refers to nonmethane organic compounds, NMOC.)
Data on precursor-oxidant relationships, including both theoretical and empirical
data, corroborate that a reduction in NMOC will result in reductions,in ambient
air concentrations of ozone and other photochemical oxidants. Available
reactivity data indicate that all hydrocarbons that participate in the formation
of peroxyacyl nitrates also participate in the formation of ozone, even though
reactivities of the various hydrocarbons may differ with respect to the different
oxidant products.
It is with respect to quantifying the precursor-oxidant relationship that
the relative disparity between the 1970 and 1979 data bases becomes apparent.
The term "relative disparity" is used here because the Administrator acknowledged
in the 1970 criteria document and in the notice of promulgation of the 1971 standard
the limitations of models available then for describing the relationship between
hydrocarbons and oxidants. The Appendix J model was the best one available in
1970 when the air quality criteria document was written; and in using it in
its 1971 standard-setting process, EPA made the most effective use it could of
available data describing the hydrocarbon-oxidant relationship. An attempt
has been made in Sections 5 and 6 of this paper to note key characteristics
and shortcomings of the Appendix J method for determining reductions in
1-16
-------
hydrocarbon emissions needed to achieve the 1971 NAAQS for oxidant. It must
be emphasized, however, that EPA has since replaced this method by the promulga-
tion of other models because data obtained since 1970 have substantiated the
theoretical inadequacies of the Appendix J method.
The characteristics of newer models (AQSM and EKMA) for determining
necessary reductions in hydrocarbon emissions have been briefly summarized in
Section 6. It is apparent that the models give different results, as stated
by EPA when it promulgated the 1979 NAAQS for ozone. It is also apparent that
no one model can serve all localities. EPA is engaged in a continuing process
of issuing additional information on precursor/oxidant models to the states to
be used in preparing revisions to State Implementation Plans.
No model is available that can quantify the relationship between HC
emissions and ozone across the entire nation. The relationship between HC
precursor emissions and oxidant air quality is such that the attainment and
maintenance of a nationally applied ozone standard depend upon HC emission
reductions specifically tailored in accordance with the HC mix (profile), HC
levels, NO levels, HC/NO ratio, and meteorological parameters of local or,
f\ S\
at most, regional areas. The most fundamental and sophisticated of the models,
the photochemical diffusion models (air quality simulation models,.AQSM),
relate hydrocarbon emissions directly to ozone air quality by means of chemical
reaction mechanisms coupled with meteorological factors. These AQSMs require
meteorological and emissions data, disaggregated spatially and temporally, for
the specific area or region in which ozone levels are to be controlled.
Irrespective of which of the EPA-prescribed models may be used, however,
an important point to be made is that all of the models are used to determine
needed reductions in hydrocarbon emissions rather than reductions in ambient
1-17
-------
air levels of hydrocarbons as a means of achieving the ozone standard. To
link ozone concentrations to a fixed hydrocarbon concentration in air nation-
wide would not be consonant with v/hat is now known about the atmospheric
reactions and processes that result in the formation of ozone, many of which
are dependent upon meteorological and source characteristics that are by
nature local, or perhaps .regional, but not national. Thus, according to
present knowledge, the relationship between hydrocarbon precursors and resulting
ambient air levels of ozone and other oxidants is not uniform or monotonic
across the nation and cannot be described on a nationwide basis. Rather, it
can be quantified with any degree of accuracy only on a local, area-specific
basis. The scientific evidence accumulated since 1971 has invalidated the
concept that any one ambient air level of hydrocarbons can be derived that
describes nationwide the relationship between hydrocarbon emissions and result-
ing ozone/oxidant ambient air levels or that in any way numerically controls
ambient air levels of ozone/oxidant nationwide.
1.8 HEALTH AND WELFARE EFFECTS OF HYDROCARBONS
The criteria developed by EPA in 1970 for hydrocarbons provided no health
data directly related to ambient levels of gaseous or volatile hydrocarbons as
a pollutant class. The only direct adverse effect attributable at that time
to gas-phase hydrocarbons at levels approximating ambient air concentrations
was the vegetation damage from ethylene; however, this effect was limited to
certain species of plants and certain areas of the country and was by no means
a problem nationwide. The sole purpose of prescribing a nonmethane hydrocarbon
standard was to control the ambient levels of specific smog components such
that the oxidant standard would be met.
1-18
-------
Since the original data base was developed, the criteria document published,
and the standards promulgated for hydrocarbons, nearly 10 years have passed.
A review of the literature since 1970 reveals once again that hydrocarbons, as
a class, do not present a significant potential for adverse health or welfare
effects at the present detectable levels but that hydrocarbons should be
controlled or restricted on the basis of their contribution to photochemical
smog and the resultant health and welfare effects of the smog products.
In order for any direct effects on health to be observed from exposure to
hydrocarbons, as a class, in the ambient air, the present levels of vapor-phase
hydrocarbons would have to be increased by hundreds to thousand of times. One
member of this class, however, is present in ambient air at levels that represent
a potential for adverse health effects.
This compound, benzene, is a unique aromatic hydrocarbon, and has been
implicated in three pathological conditions; namely, leukemia, pancytopenia,
and chromosomal aberrations. The concern over benzene as a leukemogen and as
the cause of other severe systemic toxicities at low exposure levels has been
widely recognized, as evidenced by the facts that EPA has listed it as a
hazardous pollutant under Section 112 of the Clean Air Act, that NIOSH recently
recommended a much more stringent standard for occupational exposure, and that
Runion (Section 6) has reexamined the practical impact of such a benzene
standard in relation to gasoline exposure as expressed by TLV values. The
proposed OSHA standard for benzene that was based on the NIOSH recommendation,
however, has recently been overturned by the Supreme Court, based partly on the
absence of cost/benefit analysis in the formulation of the standard.
The leukemogenic nature of benzene is specific to the structure of this
aromatic hydrocarbon. The insertion of an alkyl substituent into the benzene
1-19
-------
nucleus to give, e.g., methyl benzene (toluene), completely changes the route
by which the substituted compound is metabolized and changes the resulting
chronic toxicity to the point that toluene lacks the potential to cause
hematological damage., The toxicity of the xylenes is even greater than that
of benzene or toluene in acute exposures, but less than;that of benzene and
toluene in chronic exposures.
Relative to benzene, the remaining hydrocarbons, whether they occur as
mixtures (Carpenter's solvents, Section 6.6) or as individual hydrocarbons,
can in general be tolerated at elevated concentration levels. In the,case of
the aliphatic hydrocarbons, no threshold limit values have been assigned to
the collective class of hydrocarbons known as the "simple asphyxiants" since
their tolerable concentrations in air are limited only by the percentage (18
percent) of available oxygen. Until recently, the C5-Cg alkanes were considered
innocuous as evidenced by TLV's of 500 ppm or more; however, NIOSH has now
recommended that a TLV limit of 350 mg/m be established for total C5-Cg
alkanes—which includes pentane, hexane, heptane, and octane—based upon
chronic neurological disorders. On a volume-to-volume basis, these concentra-
tions are equal to about 120 ppm pentane, 100 ppm hexane, 85 ppm heptane, and
75 ppm octane. No studies suggest that any of the gas-phase aliphatics are
carcinogenic, mutagenic, or teratogenic in humans or experimental animals; nor
is there any reason to suspect that they will be found to produce such effects
based on their chemical structures relative to known compounds that exhibit
these effects. Alkanes and alkenes, based on limited bioassay data, have been
classified as noncarcinogenic; however, some of the long-chain (C^ °r greater)
aliphatic hydrocarbons have been implicated as cocarcinogens or tumor promoters,
based on mouse skin experiments. These long-chain aliphatics are not generally
1-20
-------
emitted as gases and are not generally encountered in the atmosphere in the
gas phase.
Toxicologically, the alicyclic compounds, as a class, resemble the pre-
viously discussed aliphatic hydrocarbons, in that they act as general anesthe-
tics and central nervous system depressants with a relatively low order of
acute toxicity. No occupational or epidemiological evidence was found to
indicate that naphthenes (alicyclic components of gasoline and other hydro-
carbon mixtures) give rise to systemic effects nor are they implicated as
hematopoietic toxicants. Unlike the aliphatic hydrocarbons, the degree of
toxicity of the alicyclic hydrocarbons does not correlate with structural
characteristics. Naphthene vapors at high concentrations cause irritation of
the mucous membranes, with the saturated hydrocarbons generally causing less
irritation than the corresponding unsaturated. No studies were found regarding
the carcinogenic, mutagenic, or teratogenic potential of any alicyclic hydro-
carbons.
Studies by Carpenter (Section 6.6) on miscellaneous hydrocarbon mixtures
lend additional support to the concept that the majority of aliphatic and
aromatic hydrocarbons, with the exception of benzene, appear to be relatively
nontoxic, even when encountered as mixtures. The lowest suggested hygienic
standard for these solvent mixtures was 90 ppm based upon inhalation studies
with animals in a lethal atmosphere and based upon the sensory response of human
subjects. (As Section 6.6 shows, typical concentrations of total nonmethane
hydrocarbons in urban atmospheres are in the range of 0.1 to 3 ppm C. Concen-
trations in rural areas range from 0.04 to 0.2 ppm C.)
No welfare effects are produced by hydrocarbons as a class at or near
ambient levels. The effects of a specific hydrocarbon, ethylene, on vegetation
have been well documented. In specific areas of the country, due to a
combination of meteorological conditions and vehicle exhaust emissions, the
1-21
-------
levels of ethylene in the ambient air have resulted in damage to ornamental
plant species. However, ethylene does not appear to be a problem nationwide,
partly because the susceptible ornamentals are grown only in certain localities.
Furthermore, the use of catalytic converters on autos, especially improved
ones expected in the future, may alleviate the problem, if not eliminate it.
Now, as in 1970, there are certain areas of the country that have air standards
for dealing with local problems from ethylene. California has an air quality
standard for ethylene, for example, which was adopted in 1962 prior to the
promulgation of the air quality standard addressing hydrocarbons.
The deleterious effects on public welfare of acid precipitation, to which
hydrocarbons are precursors via the formation of hydroperoxy radicals that
oxidize S02 to sulfur trioxide, have been documented in recent EPA publica-
tions and are only.briefly summarized in this document (Section 6.7).
In summary, with the exception of benzene hydrocarbons do not appear to
cause direct adverse health effects and except for ethylene, no direct adverse
welfare effects.
The majority of studies used ,to support the present health effects position
on hydrocarbons are studies of acute exposures. Thus, it may be advantageous
to encourage additional research studies under conditions of chronic exposure
to low levels of individual hydrocarbons and mixtures to ensure against any
long-term effects that are not obvious under acute exposure conditions. As
more information develops in the future, the health and welfare effects of
individual hydrocarbons may have to be determined and documented in separate
assessment reports.
In conclusion, there is no scientific evidence to indicate the need for
preparing a more detailed document on hydrocarbons, since now, as in 1970, an
air quality standard for hydrocarbons cannot be supported based on the health
and welfare effects of hydrocarbons as a class.
1-22
-------
2. RECOMMENDATIONS
Based upon its review of the literature and the scientific evidence
accumulated, ECAO recommends that the appropriate offices and laboratories of
EPA do the following:
1. Make a regulatory decision on the national ambient air quality
standard for hydrocarbons, without further documentation on
hydrocarbons. In the future, address hydrocarbons as precursors to
photochemical oxidants in the ozone/oxidant criteria document.
2. Study and document the effects of ethylene on plants, but as a
low-priority item. The effects of ethylene on plants are
nonspecific and resemble the effects of '.certain other pollutants, of
other environmental factors, and of some plant diseases. Since
gasoline-powered vehicles are its primary source, ethylene is a
ubiquitous pollutant. Ethylene damage to plants is not an acute
problem nationwide, however, because ethylene-sensitive species—
mostly ornamentals—are grown only in certain localities. Also, the
use of catalytic converters on autos and other highway vehicles,
especially improved converters expected in the future, may alleviate
the problem if not eliminate it.
3. Investigate the need for documenting the health effects of
aldehydes. Secondary aldehydes formed in the atmosphere from the
oxidation of hydrocarbons will be controlled to the extent that
hydrocarbon precursors to oxidants are controlled. Aldehydes are
also primary pollutants, however, and as such are emitted directly
2-1
-------
into the atmosphere from a variety of sources. Some of the aldehydes
(e.g., formaldehyde) are photochemically reactive and may also be
deleterious to human health. The decision to document the effects of
aldehydes should probably wait on the completion by NAS of its document
on aldehydes; and should wait on the preparation by EPA of its assessment
of formaldehyde, already in .progress.
4. Inasmuch as chronic exposure data are lacking, consider initiation
by EPA of research on the long-term health effects -of low levels of
individual hydrocarbons selected on the basis of toxicologic effects
coupled with ambient air concentrations. If evidence for adverse
health effects accrues from such research or in the open literature,
document the health effects of individual hydrocarbons on a case-by-case
basis to determine the need for their individual regulation.
5. Available data indicate that volatile organic emissions, including
those of hydrocarbons, from some stationary sources, are poorly
characterized with respect to the magnitude of emissions and, in
some instances, even species. Efforts to identify magnitude and
species of HC emissions from stationary sources should be intensified,
so that any changes in qualitative profiles of emissions resulting
from technologic changes, fuel conversion, etc. s can be detected
before such changes are pervasive enough to affect profiles of
ambient air HC. As species and the magnitude of emissions are
identified, study on a continuing basis the need for preparing
health assessment documents on individual compounds.
2-2
-------
6. Continue development of an accurate and reliable field instrument
for measurement of total nonmethane organic compounds (NMOC) in
ambient air. Develop instrumentation for source emission measure-
ments of NMOC, possibly including instrumentation for measuring
fluxes in organic vapors from evaporation losses. Continue to
refine methodology and instrumentation specific for measurements
of individual nonmethane hydrocarbons in ambient air.
2-3
-------
-------
3. INTRODUCTION
3.1 BASIS FOR AIR QUALITY CRITERIA AND NATIONAL AMBIENT AIR QUALITY STANDARDS
Passage of the Clean Air Act Amendments of 1970 marked the initiation of
the present role of the Federal government in air quality management by providing
authority for establishing national ambient air quality standards. The 1977
amendments to the Clean Air Act requires the Administrator to publish, and
revise from time to time a list that includes each air pollutant—
(A) emissions of which, in his judgment, cause or contribute ,
to air pollution which may reasonably be anticipated to endanger
public health or welfare;
(B) the presence of which in the ambient air results from
numerous or diverse mobile or stationary sources; and
(C) for which air quality criteria had not been issued before
the date of enactment of the Clean Air Amendments of 1970, but for
which he plans to issue air quality criteria under this section.
These amendments further require the Administrator to develop criteria for all
pollutants listed. The criteria are to reflect accurately the latest scienti-
fic knowledge useful in indicating the kind and extent of all identifiable
effects on public health or welfare that result from exposure to each of the
pollutants listed! The criteria are to include information on (1) known or
anticipated adverse effects of the pollutants on welfare; (2) interactions
between the criteria pollutant and other pollutants in the atmosphere if those
interactions will result in adverse effects on public health or welfare; and
(3) variables, including atmospheric conditions, that may alter the effects of
the listed pollutant on public health or welfare. Though criteria have his-
torically been issued in the form of "criteria documents," the Clean Air Act
does not specify the form in which criteria must be issued.
3-1
-------
Simultaneously with the issuance of criteria, the Administrator is to.
provide to the States and appropriate air pollution control agencies informa-
tion on techniques for controlling the criteria pollutant. In addition, the
Administrator must simultaneously propose primary and secondary national
ambient air quality standards that specify levels of air quality that, based
on the criteria issued and according to the Administrator's judgment, should
not be exceeded if public health and welfare, respectively, are to be pro-
tected. After a suitable period, of public comment and review (<90 days), the
Administrator is required to promulgate primary and secondary national air
quality standards, with such modifications as he deems appropriate.
3.2 BASIS AND NATURE OF PRESENT REVIEW OF CRITERIA FOR HYDROCARBONS
Section 109(d)(l), added to the Clean Air Act in 1977,.requires that EPA
review each 5 years, or more often if evidence warrants it, the criteria for
those pollutants listed under Section 108 and regulated by National Ambient
Air Quality Standards (NAAQS) under Section 109. According to this require-
ment, the first such review of the criteria and standards for hydrocarbons
must be completed by December 31, 1980.
The Environmental Protection Agency (EPA) therefore initiated earlier this
year the appropriate literature searches and review procedures, and confirmed
early in the review that current data on hydrocarbons do not substantially
alter, on two of three major issues, the scientific position on hydrocarbons
p
that prevailed when the first criteria document for hydrocarbons was written
in 1970. Furthermore, preliminary review by EPA of the criteria developed in
1970, and of the data base on hydrocarbons compiled since then, called into
question the fruitful ness of revising the original criteria document for
hydrocarbons. Instead, that review indicated the preferability of preparing
3-2
-------
a document that would serve as a basis for an appropriate regulatory decision
by the Agency by means of a thorough review of current scientific data followed
by the identification of certain key facts, or pivotal criteria, for hydrocarbons.
To fulfill its purpose, this document is essentially threefold in nature:
(1) it presents those aspects of hydrocarbons and of the NAAQS for hydrocarbons
that are unique among the criteria pollutants and the standards regulating
them; (2) it presents a brief review of the key criteria developed in 1970 and
used in the standard-setting process of 1971; and (3) it examines the 1970-1980
data on hydrocarbons that are pertinent to those key criteria. This document
is based on a comprehensive survey of available health and welfare effects
information, but the document itself is not intended to be an exhaustive
review of all aspects of hydrocarbon pollution. Rather, in areas other than
health and welfare effects, an attempt has been made to review those aspects
of hydrocarbon pollution pertinent to the criteria developed in 1970 and,
relative to those criteria, to present only those established facts that
represent a scientific consensus of present knowledge concerning hydrocarbons.
The health and welfare effects information presented is the result of a thorough
screening of the literature but is given in as concise a form as possible.
A number of issues concerning the role of hydrocarbons in the. formation
of ozone and other photochemical oxidants are important in the context of
formulating and executing effective control strategies for ozone. These
issues include the contribution of transported precursors to ozone levels in
both urban and rural ambient air; the role of natural hydrocarbons in the
formation of ozone in rural areas; and the relative contributions of oxidized
hydrocarbons versus their parent compounds in ozone formation. While these
3-3
-------
are exceedingly important issues with respect to the attainment and maintenance
of the ozone standard, they are not germane to the facts and issues concerning
the existing hydrocarbon standard. Thus the focus of this document is those criteri;
needed for regulatory analyses and decisions regarding the hydrocarbon standard
itself. Consequently, those issues relating to attainment and maintenance of
the ozpne standard but not to the necessity for or usefulness of a hydrocarbon
standard have bseen omitted frojn this document. The air quality management
information needed for implementation of the ozone standard was treated in
the 197,8 ozone criteria dpcujnent and will be addressed again when the ozone
criteria are next revised.
It must be emphasized that the 1970 criteria document for hydrocarbons
covered only those organic compounds that are composed solely of carbon and
hydrogen and that occur in the atmosphere in the gas phase. It excluded other
gas-phase organics such as substituted or derivatized hydrocarbons; e.g.,
halogenated hydrocarbons. It also excluded all organic compounds, whether
hydrocarbon or nonhydrocarbon, that occur in the atmosphere in aerosol or
particulate form. It included a brief treatment of aldehydes to ensure coverage
of this class of secondary pollutants formed from hydrocarbons in the atmosphere.
It did not cover the oxidants, since that class of secondary pollutants formed
from hydrocarbons was treated in a separate criteria document for photochemical
oxidants.
Similarly, this issue paper covers' only gas-phase hydrocarbons as defined
in the 1970 criteria document; that is, compounds of carbon and hydrogen
ranging in carbon number from 1 to about 12 (C-, - C^)- Aldehydes are treated
quite briefly, since the 1978 criteria document for ozone and other photochemical
oxidants included information on the photochemistry and health effects of aldehydes.
3-4
-------
In addition, the National Academy of Sciences is presently conducting a review
of aldehydes for EPA, and EPA is conducting its own assessment of possible
health effects of formaldehyde. At the conclusion of these reviews, EPA will
determine whether regulatory action on aldehydes should be pursued. Other
volatile organic compounds are excluded from this issue paper because they are
not hydrocarbons; because they were not covered by the 1970 criteria document
or by the 1971 NAAQS for hydrocarbons; and because EPA is currently preparing
separate documents that assess the health effects of a number of volatile
organic compounds, e.g., perchloroethylene, trichloroethylene, ethylene
dichloride, acrylonitrile, and vinylidene chloride.
3-5
-------
-------
4. BACKGROUND INFORMATION
4.1 UNIQUE CHARACTERISTICS OF CRITERIA AND STANDARDS FOR HYDROCARBONS
Originally listed as a criteria pollutant pursuant to the Clean Air Act
3
of 1970, hydrocarbons are regulated by a National Ambient Air Quality Standard
that specifies that nonmethane hydrocarbons in the ambient air shall not
3
exceed 0.24 ppm C (160 jjg/m ), maximum 3-hour average concentration (6-to-9
4
A
a.m.), more than once per year.
The NAAQS for hydrocarbons is one of six NAAQS promulgated in April 1971
A
pursuant to Section 109. Of the six NAAQS promulgated in 1971, three governed
individual or narrow chemical classes of pollutants—carbon monoxide, nitrogen
dioxide, and sulfur oxides; and three governed broad classes of pollutants—
photochemical oxidants, total suspended particulates, and gas-phase hydro-
carbons. Of the pollutants regulated as classes, hydrocarbons constitute the
most diverse and heterogeneous group in terms of (1) their health effects at
high occupational levels, (2) the secondary pollutants they give rise to, and
(3) the rates at which they react in the atmosphere. Hydrocarbons constitute
the largest and most chemically heterogeneous class of pollutants regulated by
one standard.
Hydrocarbons are unique among the seven pollutants or classes of pollutants
listed under Section 108 and regulated under Section 109 of the Clean Air Act.
All pollutants regulated by NAAQS except for hydrocarbons are thought to have
direct deleterious effects on public health, and, in some instances, on public
4
welfare as well. The criteria developed in 1970 for hydrocarbons showed
4-1
-------
clearly, however, that hydrocarbons do not directly produce deleterious effects
p
on public health or welfare. The criteria showed instead that hydrocarbons
indirectly cause adverse health and welfare effects through their contribution.
to the formation of photochemical oxidants in general and of ozone in particular.
Thus, hydrocarbons are the only pollutant or pollutant class regulated under
Section 109 for the sole purpose of controlling their secondary atmospheric
products. Hydrocarbons are the only regulated pollutants that do not produce
direct adverse effects in a human receptor but which instead produce direct
effects on receptors that are other chemical species in the ambient air. The
ambient air may be considered the receptor for which a hydrocarbon dose-
response relationship must be determined.
These unique characteristics of hydrocarbons and of the criteria for
hydrocarbons have combined to result in an NAAQS for hydrocarbons (HC) that
differs from the other NAAQS. The HC NAAQS is unique among the seven existing
NAAQS for the principal reasons listed below:
1. The NAAQS for HC was not based upon demonstration of adverse effects
45
of HC on human health or welfare. '
2. The NAAQS for HC was based solely on the role that HC play in the
atmospheric photochemical reactions that produce ozone and other
oxidants. HC were regulated in 1971 strictly on the basis that they
4 5
are precursors to photochemical oxidants. '
3. The NAAQS for HC was prescribed solely "to provide guidance in
formulating emission control strategies for attainment and maintenance
of the national standard for photochemical oxidants." It was not
intended to be an enforceable standard comparable to the other
NAAQS, but was intended to serve as a point of departure to be used
4-2
-------
in determining the reductions in HC emissions needed for achievement
and maintenance of the oxidant standard (now an ozone standard,
since a standard for ozone was promulgated in place of the oxidant
standard in February 1979).
All other pollutants or classes of pollutants for which NAAQS exist are
regulated on the basis of their direct adverse effects on health or on both
health and welfare. Except for hydrocarbons, no other precursor to a regulated
pollutant is regulated by its own separate NAAQS.
4.2 NEED FOR A DETAILED CRITERIA DOCUMENT FOR HYDROCARBONS
Air quality criteria are descriptors of the relationship that exists
between a pollutant in the ambient air and the effects caused by that pollutant
in a specific receptor or receptor population. Criteria, as described by the
Clean Air Act, must reflect the latest scientific knowledge available and
must identify the effects of a given pollutant expected from the presence of
that pollutant in the air. Where possible, criteria are expressed as a dose-
response relationship. Since the 1971 NAAQS for hydrocarbons was based on the
contribution of hydrocarbons to the formation in ambient air of photochemical
oxidants, the dose-response function that was derived was a quantitative
relationship between hydrocarbon emissions and the resulting levels of photo-
chemical oxidants formed.
Consequently, the first question asked in EPA's current review of the
1970 criteria and of the 1970 to 1980 data base was whether present knowledge
confirms that hydrocarbons in ambient air contribute to or result in the
formation of ozone and other photochemical oxidants. The second question
asked, as the logical outgrowth of the first, was whether the relationship
between precursor hydrocarbons and the resulting oxidant pollution formed was
known to be the same as that expressed in the 1970 criteria document and
4-3
-------
promulgated in 1971, in a different form, as the relationship to be used by
the States in attaining and maintaining the oxidant standard. The revision of
the air quality criteria for ozone and other photochemical oxidants and the
subsequent promulgation in 1979 of the current ozone standard provided answers
to both these questions that were confirmed in the present review. Though
both these questions are basic and crucial to any regulatory decision on
hydrocarbons, they are not sufficient by themselves. The third basic and
crucial question that must be asked is whether hydrocarbons, as a class,
produce any direct adverse health or welfare effects.
A comprehensive review of all aspects of hydrocarbon pollution in ambient
air would be counterproductive since the three questions or issues delineated
above are the pivotal issues to be resolved. First, such a.review would be
precluded on the basis pf the volume of information alone, given the number of
hydrocarbons that have been detected in ambient air. Second, and most important,
there is no known gas-phase hydrocarbon that can serve as a surrogate for the
entire class of compounds, as an index of the ambient air level of the entire
class, as an index of the oxidant-forming potential of hydrocarbon mixtures
in ambient air, or as a control target for the whole class. Consequently,
even if a dose-response relationship were derived for a single hydrocarbon,
this criterion would be of no value in a review of the existing class standard.
It would be useful only for regulatory decisions and action with respect to
that individual compound.
In view of these facts, EPA believes that a comprehensive scientific
treatise on hydrocarbons is not appropriate as a review of hydrocarbon criteria.
This issue paper offers and documents answers to the three basic questions
posed above, answers that constitute the pivotal criteria needed to review the
NAAQS for hydrocarbons. If review of the standards should prove the scientific
4-4
-------
answers given to be inadequate, or should raise other key questions that must
be answered, EPA will undertake the necessary documentation, either in a
revised issue paper or in a more comprehensive document.
4-5
-------
-------
5. SUMMARY OF 1970 CRITERIA FOR HYDROCARBONS
This section sets forth in summary form the criteria for hydrocarbons
that were published in 1970 and upon which the NAAQS for hydrocarbons promulgated
in 1971 was based. This review of the pertinent findings of the 1970 criteria
document is presented here to provide background for understanding the basis
of the present NAAQS for hydrocarbons and to facilitate comparison of that
basis with the current data base on hydrocarbons.
5.1 CONTRIBUTION OF HYDROCARBONS TO FORMATION OF PHOTOCHEMICAL OXIDANTS
2
As stated in the 1970 document, the ultimate products of the photooxidation
of HC in ambient atmospheres would probably be carbon dioxide and water vapor,
if a parcel of air were irradiated by sunlight long enough. Complete oxidation
of hydrocarbons in an air parcel does not occur, however, because of shortened
irradiation time, transport, diffusion, the infusion of "fresh" emissions,
etc. At any given time, then, the photooxidation products of hydrocarbons
that reside in the ambient air consist of various intermediates in the
oxidation process. These intermediates are all capable of further reaction
and degradation.
Hydrocarbons become involved in the production of the photochemical air
pollution complex not because of their exposure to sunlight, however, but
2
because of their reaction with other components of the atmosphere. The
processes by which hydrocarbons participate in the formation of ozone and
other photochemical oxidants have been reviewed in detail in two recent EPA
documents, Air Quality Criteria for Ozone and Other Photochemical Oxidants
(1978),7 and Air Quality Criteria for Nitrogen Oxides (1979).8 In additior
the chemistry of oxidant formation and of the role of organic compounds,
5-1
-------
including hydrocarbons, in those processes has been reviewed in two National
9 10
Academy of Sciences documents. ' The description given below is a highly
simplified version of the complex chemical reactions that actually occur in
the atmosphere. It is presented merely to underscore the fact that gas-phase
hydrocarbons do contribute to the formation of ozone and other photochemical
oxidants.
The photochemical oxidants observed in the atmosphere are ozone (0-);
nitrogen dioxide (N02); and peroxyacetyl nitrate (PAN).9 Several other
substances, such as hydrogen peroxide (H202), may be classified as
photochemical oxidants, but their common presence in smog is not well
Q
established. Of these oxidants— all of which are secondary pollutants formed
as a result of chemical reactions in the atmosphere — ozone occurs in the
highest concentrations,
Ozone formation begins with the photodissociation of nitrogen dioxide (N02),
as shown in Equation (5-1):
N02 + hv -»• 0 + NO (5-1)
The oxygen atoms thus formed react with molecular oxygen in the presence of a
third body (M):
0 + 02 + M
0
M
3 (5-2)
These two reactions together form a mechanism for ozone formation in the
atmosphere, a mechanism that is completed by an additional reaction in which
N02 is regenerated:
NO + 03 -> N02 + 02 (5-3)
The role of hydrocarbons in promoting reactions 5-1 through 5-3 is complex,
and at the time of the 1970 document was not very well understood. Although
5-2
-------
of central importance in smog chemistry, reactions (5-1) through (5-3) by
g
themselves do not explain the atmospheric buildup of ozone. If only these
three reactions were important, the photodissociation of nitrogen dioxide
would rapidly establish a small, steady-state concentration of ozone expressed
by the equation:
C03] =
6
k3 [NO] -
(5-4)
The photolysis of nitrogen dioxide alone then, does not explain the atmospheric
q
accumulation of ozone, even though it is the mechanism of ozone formation.
The dominant factor in this system is the ratio of NOp to NO. Once the conversion
of NO to NOp is explained, the ozone concentration follows the ratio of NOp to
Q
NO concentrations given in equation (5-4). What was inexplicable until about
1970 was the rapidity of oxidation of NO to NOp and the continuing conversion
of NO into NOp during the buildup of ozone, because this defies the relatively
simple theoretical relationship described in equation (5-4).
Early laboratory studies had implicated hydrocarbons, especially olefins,
in the formation of ozone by showing that both ozone and ground-state atomic
oxygen would attack reactive hydrocarbons. Modeling and smog-chamber
simulations have since shown that significant oxidant formation occurs with
9
nitrogen oxides (NO ) plus aldehydes, or with NO plus nonmethane hydrocarbons.
f\ /\
Though the mechanism had barely been suggested when the 1970 criteria document
was written, it is now known that hydrocarbons—along with the other oxidizable
pollutants such as aldehydes and CO—serve the function of regenerating free
radicals that react with oxygen in the air to form alkylperoxy and hydroperoxy
5-3
-------
radicals. These radicals react with NO to form N02:
or
H02 + NO -» OH + N0£
R02 + NO -> OH + N02
(5-5)
Thus, these oxidizable pollutants can be thought of as pumping the nitric
oxide to nitrogen dioxide. In the process, hydrocarbons become degraded to
other compounds, some of which are also photochemically reactive, such as
formaldehyde. The amount of "pumping" that can be done, and thus the amount
of oxidant formed, depends in a nonlinear manner on both the reactivity of the
oxidizable pollutant—in this case, hydrocarbons—and its concentration.
Advances made since 1970 in understanding the reaction mechanisms involved are
summarized in Section 6.
5.2 MEASUREMENT METHODOLOGY
As stated in the 1970 air quality criteria document for hydrocarbons, the
complexity of the mixtures of hydrocarbons present in ambient air demands
sophisticated instrumentation and techniques for measurement. In 1970 a
technique was already available for the measurement of the concentration of
total hydrocarbons in ambient air. Originally developed as a detector for gas
chromatography, this flame ionization technique (FID) was adapted for total
hydrocarbon analysis around 1960. In the technique, a sensitive electrometer
detects the increase in ion intensity resulting from the introduction into a
hydrogen flame of a sample of air containing any organic compound (e.g.,
hydrocarbons, aldehydes, alcohols). The response is approximately in proportion
to the number of organically bound carbon atoms in the sample, but its response
5-4
-------
to carbon atoms in different compounds is nonlinear. That is, the response of
the detector to different classes of hydrocarbons is variable and the response
12
appears to fall off rapidly with increasing carbon number. FID data are
usually expressed as the calibration gas used: for example, parts per million
(ppm) of carbon as methane. The fact that FID response may be, and often is,
related to different reference compounds (methane, propane, butane, etc.)
further complicates the comparability, interpretation, and usefulness of FID
' 12
data. Carbon atoms bound to oxygen, nitrogen, or halogens give reduced or
no response; hence, the method may detect oxygenated hydrocarbons such as
aldehydes but does not actually measure their concentrations. The instrument
responds to hydrocarbon derivatives approximately according to the proportion
of carbon atoms bound to carbon or hydrogen. There is no response to nitrogen,
carbon monoxide, carbon dioxide, or water vapor; but there is an oxygen effect,
which can be minimized by appropriate operating conditions.
The response of the FID is rapid and, with careful calibration, is sensitive
to a fraction of a ppm carbon as methane. Variations in FID response to various
hydrocarbons and derivatives detract from the comparability and usefulness
13
of FID data. Practically all continuous hydrocarbon analyzers in use in
1970 utilized the flame ionization detector as the sensing element.
The value of the data obtained from measuring total hydrocarbon
concentrations is limited. The high background level of naturally occurring
methane obscures variations in levels of other hydrocarbon species which,
unlike methane, participate in photochemical reactions and are therefore of
principal interest. Consequently, methods for measuring methane separately
and determining nonmethane hydrocarbons by subtraction of the methane values
5-5
-------
had been attempted by 1970 and had met with some success. Many of the
nonmethane hydrocarbon data obtained prior to 1970 were data obtained by use
~\9
of a relatively crude carbon absorber to separate out the methane. At the
*
time the NAAQS for hydrocarbons was promulgated, EPA promulgated as the
Federal Reference Method an FID method for determining nonmethane
hydrocarbons. In the EPA method, which was developed around 1970 and which
represented a major improvement over the carbon absorber, the air sample
containing methane and carbon monoxide is passed quantitatively through a
stripper column to a gas chromatographic column that separates the two gases.
The methane is eluted first and is passed unchanged into the FID. (The CO is
eluted into a catalytic reduction tube where it is reduced to methane and is
subsequently measured'by the FID. The CO measurement is not needed in
measuring NMHC.) The NMHC concentration is obtained by subtracting the
methane value from the total hydrocarbon concentration. The method is
applicable to the semicontinuous measurement of hydrocarbons corrected for
methane in the ambient air. This method has been incorrectly called a "total
hydrocarbon (THC)" measurement method, when in fact the FID also responds to
oxygenated compounds such as alcohols and aldehydes having >2 carbon atoms,
as well as to other volatile organic compounds (VOC). Nevertheless, in the
absence of a suitable method specific for hydrocarbons, EPA promulgated this
method in 1971 as the best available method, and thus the method of choice,
2 4
for the measurement of total hydrocarbons and of nonmethane hydrocarbons,. '
Other methods available in 1970 for the measurement of hydrocarbons
included spectrometric and gas chromatographic techniques. Available spectro-
metric methods for total and specific analyses were complex and generally ••
5-6
-------
insensitive. Gas chromatographic techniques, while providing the requisite
sensitivity and specificity for the measurement of individual hydrocarbons, '
required a high degree of operator skill and experience. Furthermore, they
were applicable only to short-term analyses because of the time, effort, and
2
skill involved in data reduction.
5.3 SOURCES AND EMISSIONS OF HYDROCARBONS
Hydrocarbons in the ambient air arise from both natural and man-made
(technologic) sources. The presence of hydrocarbons from natural sources was
established in 1948 with the measurement of background methane concentrations.
Although most natural hydrocarbons arise from biological sources, small and
highly localized quantities of methane and a few other lower-molecular-weight
hydrocarbons are emitted by geothermal areas, coal fields, natural gas and
2
petroleum fields, and natural fires.
Some estimates of natural methane and of terpenes and isoprene from
vegetation were available in 1970. Koyama, in 1963, conservatively estimated
ft
the worldwide emission of natural methane to be 2.73 x 10 metric tons per year
8 18
(3 x 10 tons per year.) Worldwide emissions of volatile terpenes and
isoprenes from vegetation were estimated in 1965 by Rasmussen and Went at 4.0
O Q -j Q
x 10 metric tons per year (4.4 x 10 tons per year.) No estimate of natural
9
emissions in the United States alone was'given in the 1970 document.
Total nationwide manmade emissions of hydrocarbons and related organic
compounds were estimated at 29.1 x 10 metric tons (32 x 10 tons) for 1968,
the base year for which emissions data were available for use in the 1970
document. Transportation (mobile sources) accounted for 52 percent of this
estimate. Miscellaneous sources, principally solvent evaporation, constituted
the second largest category of sources and contributed 27 percent of these
5-7
-------
emissions. Industrial processes were third, with 14 percent; solid waste
disposal was fourth, with 5 percent; and fuel combustion in stationary sources
was fifth, with 2 percent.2
Local emissions for 22 metropolitan areas, as opposed to total nationwide
emissions, were estimated to range from about 0.04 to 1.2 million metric
tons per year (about 0.05 to 1.3 million tons per year), depending on the area.
Transportation sources accounted for 37 to about 99 percent of local emissions,
and process losses—solvent evaporation and industrial processes combined--
accounted for 1 to 63 percent. Thus, total hydrocarbon emissions in 1968, the
base year used in the 1970 criteria document, originated primarily from the
inefficient combustion of volatile fuel in mobile and stationary sources and
on
from the use of hydrocarbons as process raw materials.
•s
Conventional automobiles (gasoline-powered, internal combustion engines)
potentially emit gas-phase hydrocarbons from four sources: engine exhaust,
crankcase blow-by, carburetor evaporation, and fuel tank evaporation.2 Positive
crankcase ventilation (PCV) systems were introduced in 1963 that virtually
9]
eliminated crankcase blow-by emissions. Emissions from the other three
sources were not yet controlled to any appreciable extent in 1968, so that
uncontrolled conventional automobiles constituted most of the aggregate of
passenger cars and other light-duty vehicles at that time.
5.4 AMBIENT AIR CONCENTRATIONS
Methane concentrations were reported in the 1970 document to range generally
3
from about 0.7 to 1.0 mg/m (1.0 to 1.5 ppm), but to occur often at levels as
3
high as 4 mg/m (6 ppm) in populous areas. Yearly averages of monthly maximum
1-hour average total hydrocarbon concentrations, which included methane,
5-8
-------
ranged from 8 to 17 ppm (as carbon) at stations of the Continuous Air Monitor-
2
ing Projects (CAMP) network for 1962 through 1967. Few data were available
in 1970 for comparing nonmethane hydrocarbon (NMHC) concentrations at various
2
geographic locations. From the data available in 1968 from the CAMP network,
24-hour average NMHC concentrations, averaged over several months, were about
0.5 to 0.8 ppm C in Washington, D.C.; about 0.6 to 1.5 ppm C in Denver; about
22
0.8 to 1.3 ppm C in St. Louis; and about 0.6 to 1.3 ppm C in Chicago. In
urban areas, the ratio of nonmethane hydrocarbons to methane was estimated at
0.6 in Cincinnati and at 1.9 in Los Angeles in 1964. The higher ratio in Los
Angeles may have been a reflection of the greater traffic density in that
city.2
In a series of 200 samples taken in one urban location, average con-
centrations of the most abundant hydrocarbons were (in ppm as C): methane,
3.22; toluene, 0.37; n-butane, 0.26; isopentane, 0.21; ethane, 0.20; benzene,
0.19; n-pentane, 0.18; propane, 0.15; and ethylene, 0.12. Among classes of
hydrocarbons, alkanes predominated even when methane levels were excluded.
They were followed, in order, by aromatics, olefins, acetylene, and alicyclics.
Detailed hydrocarbon data for samples taken in Los Angeles in 1965, as
23-27
determined by gas chromatography, are shown in Tables 5-1 through 5-3.
The hydrocarbons that appear on these tables are typically found in most urban
areas. The quantitative profiles differ from one urban area to the next;
however, the qualitative profiles do not vary widely in urban areas influenced
predominantly by automobile emissions. Examination of these qualitative
profiles and comparison with whole gasoline and gasoline vapor constituents
5-9
-------
TABLE 5-1. SOME HYDROCARBONS IDENTIFIED IN AMBIENT AIR
23-25
Carbon
number
1
2
2
2
3
3
3
3
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
5
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
7
Class
Al kane
Al kane
Al kene
Al kyne
Al kane
Al kene
Al kene
Al kyne
Alkane
Al kane
Al kene
Al kene
Al kene
Al kene
Al kene
Al kane
Al kane
Al kene
Al kene
Al kene
Al kene
Al kene
Al kene
Al kene
Cycloalkane
Cycloalkene
Al kane
Al kane
Al kane
Al kane
Al kane
Al kene
Alkene
Al kene
Al kene
Al kene
Alkene
Al kene
Al kene
Aromatic
Cycloalkane
Cycloalkane
Al kane
Compound
Methane
Ethane
Ethyl ene
Acetylene
Propane
Propylene
Propadiene
Methyl acetylene
Butane
Isobutane •
1-Butene
cis-2-Butene
trans-2-Bu.tene
Isobutene
1,3-Butadiene
Pentane
Isopentane
1-Pentene
cis-2-Pentene
trans-2-Pentene
2-Methyl-l-butene
2-Methyl-2-butene
3-Methyl-l-butene
2-Methy 1 -1, 3-butadi ene
Cyclopentane
Cyclopentene
Hexane
2-Methy Ipentane
3-Methylpentane .
2, 2-Dimethyl butane
2 , 3-Dimethyl butane
1-Hexene
cis-2-Hexene
trans-2-Hexene
cis-3-Hexene
trans-3-Hexene
2-Methyl -1-pentene
4-Methy 1 -1-pentene
4-Methyl -2-pentene
Benzene
Cyclohexane
Methyl cycl opentane
2-Methy Ihexane
5-10
-------
TABLE 5-1 (continued)
Carbon
number
Class
Compound
7
7
7
7
8
8
10
Al kane 3-Methylhexane
Alkane 2,3-Dimethylpentane
Alkane 2,4-Dimethylpentane
Aromatic Toluene
Alkane 2j2,4-Trimethylpentane
Aromatic o-Xylene
m-Xylene
£-Xylene
Aromatic m-Ethyl toluene
S-Ethyl toluene
52,4-Trimethyl benzene
1,3,5-Trimethylbenzene
Aromatic sec-Butyl benzene
5-11
-------
TABLE 5-2. AVERAGE HYDROCARBON COMPOSITION,
218 AMBIENT AIR SAMPLES,
LOS ANGELES, 196526
Concentration
Compound
Methane
Ethane
Propane
Isobutane
n-Butane
Tsopentahe
rrPentane
2 , 2-Di methyl butane
2-Methylpentane
2, 3-Dimethyl butane
Cycl opentane
3-Methylpentane
ji-Hexane
Total alkanes (excluding methane)
Ethyl ene
Propene
1-Butene + Isobutylene
trans-2-Butene
cis-2-Butene
1-Pentene
2-Methyl-l-Butene
trans-2-Pentene
cis-2-Pentene
2-Methyl-2-Butene
Propadiene
1,3-Butadiene
Total alkenes
Acetyl ene
Methyl acetyl ene
Total acetylene
Benzene
Toluene
Total aromatics
Total
ppm
3.22
0.098
0.049
0.013
0.064
0.043
0.035
0.0012
0.014
0.004
0.008
0.012
0.3412
0.060
0.018
0.007
0.0014
0.0012
0.002
0.002
0.003
0.0013
0.004
0.0001
0.002
0.1020
- 0.039
0.0014
0.0404
0.032
0.053
0.0850
3.7886
ppm (as carbon)9
3.22
0.20
0.15
0.05
0.26
0.21
0.18
0.01
0.08
0.02
0.05
0.07
1.28
0.12
0.05
0.03
0.01
Negligible
0.01
0.01
0.02
0.01
0.02
Negligible
0.01
0.29
0.08
Negligible
0.08
0.19
0.37
0.56
5.43
Determined by multiplying compound concentration by number of carbon atoms
in the compound.
5-12
-------
TABLE 5-3. AVERAGE AND HIGHEST CONCENTRATION MEASURED
FOR VARIOUS AROMATIC HYDROCARBONS IN LOS ANGELES,
26 DAYS, SEPTEMBER THROUGH NOVEMBER, 196627
Aromatic hydrocarbon
Average
concentration,
ppm
Highest measured
concentration,
ppm
Benzene
Toluene
Ethyl benzene
p_-Xylene
m-Xylene
o-Xylene
T-Propyl benzene
n-Propyl benzene
3-and 4- Ethyl toluene
1, 3, 5-Tri methyl benzene
1,2,4-Trimethylbenzene, and
i -Butyl -and sec-Butyl benzene
tert-Butyl benzene
0.015
0.037
0.006
0.006
0.016
0.008
0.003
0.002
0.008
0.003
0.009
0.002
0.057
0.129
0.022
0.025
0.061
0.033
0.012
0.006
0.027
0.011
0.030
0.006
Total aromatics
0.106
0.330
5-13
-------
(see Section 6) reveal the important influence of mobile source emissions .on
hydrocarbon mixtures in urban atmospheres.
Among the hydrocarbon oxidation products for which ambient air data were
available in 1970, aldehydes were by far the predominating compounds. Total
aldehyde concentrations for Los Angeles for 1951 through 1957, as determined
28
by manual, wet-chemical methods, are shown in Table 5-4. Further study in
1960 in the Los Angeles area showed that concentrations of total aldehydes
ranged up to 0.36 ppm for a 10-minute sample, with formaldehyde not exceeding
q
130 ug/m (0.10 ppm). The maximum acrolein (acrylic aldehyde) value was 25.2
o
ug/m (0.011 ppm); but most values were less than half that level. Typical
29
aldehyde concentrations were around 0.10 ppm on many days. Yearly maximum
1-hour average aldehyde concentrations in Los Angeles ranged from 0.20 to 1.30
ppm.
The 1970 report documented the diurnal patterns of nonmethane hydrocarbons
for several cities, showing that in most of the urban areas sampled maximum HC
concentrations coincided with the peak traffic period of about 6- to 9-a.m.
po
These diurnal patterns, illustrated in Figure 5-1, do not parallel the
diurnal variations in secondary contaminants, particularly ozone. Instead,
2
secondary pollutants typically show a pronounced maximum in the afternoon.
Insufficient data were available in 1970 to determine seasonal variations in
hydrocarbon concentrations nationwide. Data available then for 17 California
cities showed, however, that in 14 of those cities maximum HC Concentrations
(averages of maximum hourly averages) occurred in October or November,
presumably as a consequence of the similar meteorological conditions that
2
prevail along the California coast.
5-14
-------
TABLE 5-4. RANGE OF YEARLY MAXIMUM 1-HOUR AVERAGE CONCENTRATIONS
OF ALDEHYDES, LOS ANGELES COUNTY STATIONS, 1951 THROUGH 195728
Year
Concentration range, ppm
Formaldehyde
Total aldehydes
1951
1952
1953
1954
1955
1956
1957
0.05 - 0.12
0.26
0.20
0.25
0.39
0.47
0.51
0.27
- 0.67
- 0.27
- 1.20
- 0.80
- 1
28
1.30
0.47
5-15
-------
1.5
1.0
0.5
o
\ \ 1 II 1
CHICAGO *"**^
X / "* -
X._ /
(MAY THROUGH AUGUST AND
1 1 1 1 1 1
1 1 1 1
.^" "S _jr- m
** **..—' ""*"•«
__
OCTOBER. 1968)
1 1 1 1 1
1.5
O
| 1.0
§ 0.5
U
ST. LOUIS
s-^.
..,.--' *.,'
(MAY THROUGH JULY, SEPTEMBER, AND OCTOBER, 1968)
£ 2.0
UJ
I 1.S
111
O
t.o
0.5
1.0
o.s
1 I
DENVER
1 T
(JANUARY THROUGH MARCH, MAY, SEPTEMBER, AND OCTOBER, 1968)
i i i I I I I L I I I
—i—n—r
WASHINGTON, D.C.*
O1—
12
(JANUARY THROUGH APRIL AND AUGUST THROUGH OCTOBER, 1968)
I I I I I I I I I J L_
6
-a.m.
12
6
-p*m*
LOCAL TIME
Figure 5-1. Nonmethane hydrocarbons by flame ionization analyzer, averaged by
hour of day over several months for four cities.29
5-16
-------
Diurnal variations in aldehyde concentrations display an early rise, a
broad plateau or maximum through most of the day, and a decrease in the
afternoon. This pattern is shown for formaldehyde and acrolein concentrations
•an 31
in Los Angeles in Table 5-5, for 1961, and in Figure 5-2, x for 1968.
5.5 HYDROCARBON/OXIDANT RELATIONSHIPS
National ambient air quality standards regulate the concentrations of
criteria pollutants in the air. They do not directly regulate emissions,
though emissions are controlled in order to meet NAAQS. In the case of measure-
ments of ambient air concentrations of hydrocarbons, such measurements do not
necessarily reflect emissions of hydrocarbons. At any given time that an
urban atmosphere is being measured, the hydrocarbon concentrations measured
may represent just one point in time—and space—during a dynamic process that
involves the mixing of fresh hydrocarbon emissions (1) with already-reacted
(and thus chemically transformed) hydrocarbons, (2) with relatively unreactive
or slowly reactive hydrocarbons, and (3) with many other pollutants under
various changing meteorologic conditions. Consequently, the major problem
facing atmospheric chemists in 1970, including the authors of the air quality
criteria document, was that of relating hydrocarbon emissions to resulting
oxidant air quality by means of relationships based on ambient air concentra-
2
tions of hydrocarbons. The criteria document stated:
The development of a model to relate emission rates of hydrocarbons
to ambient air quality and then to the secondary products of photochemical
reactions has proved to be an elusive problem. Because of this lack
of an appropriate model, the relationship between hydrocarbon emissions
and subsequent maximum daily oxidant levels must be approached empirically.
The empirical approach adopted is a comparison of 6 to 9 a.m. average
hydrocarbon values with hourly maximum oxidant values attained later
5-17
-------
TABLE 5-5. AVERAGE FORMALDEHYDE AND ACROLEIN CONCENTRATIONS BY TIME OF DAY IN
LOS ANGELES, SEPTEMBER 25 THROUGH NOVEMBER 15, 196130
Formaldehyde
Acrolein
Sampling
time
Number of
days
Average
concentration, ppm
Number of Average
days concentration, ppm
7:00 a.m
8:00 a.m
9:00 a.m
10:00 a.m
11:00 a,m
Noon
1:00 p.m
2:00 p.m
3:00 p.m
4:00 p.m.
7
18
21
28
27
23
25
27
25
15
0.041
0.043
0.045
0.044
0.051
0.044
0.041
0.034
0.026
0.019
2
3
3
5
5
3
7
5
4
5
0.007
0.009
0.009
0.008
0.008
0.005
0.008
0.007
0.004
0.004
5-18
-------
0.18
0.16
UJ
>- *
£ 0.12
UJ
J 0.10
u 1 °-08
I*™ *
|| 0.04
^ ^ 0.02
_ o: _
O (_ 0
<'S 0.14
ILIU
O Z 0.12
>o
^1 1 "
5 0.08
^
-g 0.06
O 0.04
U.
O.02
n
'III
— ,
— .,** HUNTINGTON —
xx \
- / X \\
- / x /%'N
/ • / \>\
PARK _
—
—
_X /..*** ALIPHATIC —
^ ^-* """"ALDEHYDES _
~^»* . .. FORMALDEHYDE
~ ...... ACROLEIN ~
~ y
- _
1 m
— ^X*^^ ./j*
'^ ^^ »*
""•• ^J1*
— . if****'
— •»•*
III
± EL MONTE__
* _L.-.~I
•^•1 \
...
I 1
U.U ID
0.016
0.014
0.012
0.010 £
0.008 g
0.006 i=
0.004 ^
0.002 £
U
O z
0
0.014 U
0.012 Z
O.010 Q
n;
0.008 o
0.006 "*
O.004
0.002
0
LOCAL TIME
-p.m.
Figure 5-2. Hourly aldehyde concentrations at two
Los Angeles sites, October 22,1968.31
5-19
-------
in the day. This approach has validity only because of the dominating
influence of the macro-meteorological variables on both the concentra-
tions of precursors and photochemical products. Furthermore, this
approach can yield useful information only when a large number of
days are considered; this guarantees the inclusion of all possible
combinations o^ emission rates, meteorological dilution and dispersion
variables, sunlight intensity, and ratios of precursor emissions.
When maximum daily oxidant values from such an unrestricted data base
are plotted as a function of the early morning hydrocarbons, a complete
range of oxidant values—starting near zero and ranging up to finite
and limiting values—is observed. Given data for a sufficient number
of days, it becomes apparent that the maximum values of attainable
oxidant are a direct function of the early morning hydrocarbon con-
centration. This upper limit of the maximum daily oxidant concentra-
tion is dependent on the metropolitan geographical area only to the
extent, that differences in meteorological variables exist between
these areas. Thus the data from all cities can be plotted on one
graph when defining the oxidant upper limit as a function of early
morning hydrocarbon.
As the document further noted, all available data relating directly
measured nonmethane hydrocarbon concentrations to maximum daily oxidant
o
concentrations were used. For Los Angeles, however, no nonmethane data were
available; and for some stations in each of the other cities nonmethane hydro-
carbon concentrations were derived by applying an empirically determined ratio
32 33
between total and nonmethane hydrocarbons. ' Figure 5-3 shows maximum
oxidant concentrations as a function of 6-to-9-a.m. average total hydrocarbon
concentrations, developed from 326 days of data obtained from five cities.
Direct examination of the data used to develop this plot shows that, at any
given hydrocarbon level, there appears to be a limit to the amount of oxidant
32
that can be generated. Specifically," the data appear to indicate that an
average 6-to-9-a.m. concentration of 0.3 ppm nonmethane hydrocarbon (expressed
as carbon) can be expected to produce a maximum hourly average oxidant con-
32
centration of up to 0.1 ppm. It must be borne in mind, however, that the
aerometric data base underlying this model was obtained by FID measurements
that were subject to all the errors associated with this method.
5-20
-------
0.30>
0.25''
< 6.20
Q
X
o
UJ
o
0.15
Of
O
O
'•& 0.10
:a
X
O.05
LOS ANGELES*
LOS ANGELES^.-^* DENVER
WASHINGTON* ^* ~
. •x° A LOS ANGELES
»^A A PHILADELPHIA
LOS ANGELES
PHILADELPHIA
PHILADELPHIA^X
WASHINGTON A'*
WASHINGTON/ * *
>' ' v.-.
WASHINGTON ^
« A A* A A A* A
- $ »** A * * A *f
A* A t A lA Al Ai A* AA
***** \i & * * ** * **
A A A A A A
A * * *
A A
A
A*
0.5
1.0
1.5
2.0
6-9 a.m. AVERAGE NONMETHANE HYDROCARBON
CONCENTRATION, ppm C
Figure 5-3. Maximum daily oxidant as a function of early morning nonmethane hydrocarbons,
1966-1968 for CAMP stations; May through October 1967 for Los Angeles.2
5-21
-------
A number of caveats for this method of relating hydrocarbon concentra-
2
tions to oxidant concentrations were given in the 1970 hydrocarbon document,
33
in the 1971 air quality criteria document for nitrogen oxides, and in the
journal publication that described the development of the upper-limit curve of
Figure 5-3.32 Schuck et al.32 clearly stated that the upper limit curve is
descriptive rather than predictive. It defines the limit on oxidant associated
with certain levels of ambient hydrocarbons but cannot predict or explain this
observed relationship.32 The nitrogen oxides document of 1971, which discussed
the upper-limit curve in detail, noted that the derivation of an upper limit
curve requires abundant measurements. "Since the upper limit is attained on
only about 1 percent of the days in a year, many days' measurements are needed
in order to provide reasonable assurance that an upper-limit point has actually
been observed."33 The relative paucity of data at the actual curve in Figure 5-3
demonstrates this point.
Nevertheless, since the hydrocarbon standard was promulgated for the
purpose of ensuring the attainment and maintenance of the oxidant standard,
EPA had to prescribe a method for determining the relationship between hydro-
carbon emissions and resulting ambient air concentrations of oxidant. In
response to that need, EPA scientists developed a method that used the
upper-limit curves in conjunction with the dispersion-related equations used
in linear rollback models.
Rollback models embody the principle that reductions in emissions of a
pollutant are reflected by improvements in air quality, as may be shown by a
straight line, a curved line, or a complex surface that expresses some pro-
portionality relationship.9 Linear rollback, though it would result in
improved oxidant air quality, does not accurately describe the relationship
5-22
-------
between hydrocarbons and oxidant air quality; it is best applied to pollutants
that, unlike hydrocarbons, do not undergo chemical transformations in the
Q
atmosphere. Consequently, EPA used the upper-limit curve to derive a type of
modified rollback model for relating maximum 1-hour photochemical oxidant
concentrations to the percentage reduction needed in hydrocarbon emissions in
order to achieve the oxidant standard. The resulting model or method was
34
promulgated in 1971 as Appendix J. It represented EPA's best attempt, given
the data available and the state of the art of modeling in 1971, to go beyond
relating ambient air HC levels to oxidant levels and to attempt the more
difficult task of relating hydrocarbon emissions to oxidant levels. The
33 34
Appendix J curve is shown as Figure 5-4. V As discussed in Section 6,
revision of the NAAQS in 1978 for. oxidants and promulgation of
v
an NAAQS for ozone in 1979 led to replacement of the Appendix J model by four
other modeling techniques for use by the States in determining hydrocarbon
reductions necessary to attain the NAAQS for ozone.
5.6 HEALTH EFFECTS
2
The document entitled, Air Quality Criteria for Hydrocarbons, which was
published in 1970, provided no health effects data directly related to ambient
concentrations of gaseous or volatile hydrocarbons as a pollutant class.
Excluded from this health assessment were hydrocarbons (>C12) and other
organics (e.g., polycyclic organic matter) associated only with suspended
particles in the atmosphere. The only direct adverse effect at that time
attributable to known ambient concentrations of hydrocarbons was the
vegetation damage from ethylene; however, these effects were highly dependent
on the locality and the species of plants grown for retail sales. The
sole purpose of prescribing a nonmethane hydrocarbon air quality standard
5-23
-------
'1NVOIXO IVOIIrtlSHOOlOHd UOd OUVONVIS
IVNOIJ.VN 3A3iHOVOiaaainD3U SNOISSIW3 NoauvoouaAH NI NOiionaau
5-24
-------
o
of 160 ug/m (0.24 ppm), maximum 3-hr average concentration (6-to-9-a.m.) not
to be exceeded more than once per year, was to control the ambient levels of
«3
specific smog components such that the oxidant standard of 160 ug/m (0.08
ppm), maximum 1-hour concentration not to be exceeded more than once a year,
would be met. Inherent in the setting of this hydrocarbon standard was the
fact that these specific smog components known as "oxidants" are produced from
hydrocarbons by gas-phase photochemical reactions involving oxides of nitrogen
(NO ), oxygen, and sunlight. To date, the only health effects reported from
/\
exposure to smog products derived directly from hydrocarbons, as opposed to
indirectly derived products such as ozone, at levels approximating those found
in the ambient air, are the irritating effects of aldehydes and peroxyacyl-
nitrates on mucous membranes of the eyes, nose, and throat. The aldehydes
that have been identified as the most effective irritants are acrolein and
formaldehyde, while the peroxyacylnitrates identified as such are the acetyl
and benzoyl derivatives. These smog components (aldehydes, peroxyacylnitrates)
are governed by the control strategy cited in the revisions to the NAAQS for
photochemical oxidants.
The lack in 1970 of a health effects data base directly related to hydro-
carbons at ambient concentrations, either as individual compounds or as a
class, is readily observed from the toxicological and clinical studies discussed
below. These studies were those reviewed in the preparation of the 1970 criteria
o
document for hydrocarbons that served as the basis for the promulgation of
the standards.
The effects observed up to 1970 for hydrocarbons either as a class or as
individual compounds are as follows:
5-25
-------
5.6.1 Aliphatic Hydrocarbons
In general, members of this hydrocarbon series are virtually inert and
produce no demonstrable pathological effects. The first two members of these
alkane hydrocarbons, methane and ethane, are pharmacologically "inert,"
belonging to a group of gases called "simple asphyxiants." These gases can be
tolerated in high concentrations in inspired air; however, above concentrations
*
of 50,000 and 100,000 ppm, one observes systemic effects of ethane and methane,
respectively, due to oxygen deprivation or asphyxia. Pharmacologically, the
aliphatic hydrocarbons above ethane, both saturated and uhsaturated, can be
grouped with the general anesthetics in a large class known as the central
nervous system depressants. This is particularly true of ethylene, propylene,
s
and acetylene, which have all been used as anesthetics at high concentrations.
The vapors of the alkane hydrocarbons are mildly irritating to mucous membranes,
the irritation increasing in intensity from pentane to octane. Generally,
alkanes from the pentanes (Cg) through the octanes (Cg) show increasingly
strong narcotic properties which have been correlated with increased lipophilia.
Narcotip effects may be accompanied by exhilaration, vertigo, headache, anorexia,
incoordination, and nausea. These compounds in general are relatively inactive
toxicologically at concentrations of hundreds to thousands of times above
those levels found in the atmosphere. No effects were observed at levels
below 500 ppm for the alkanes and 1,000 ppm for the alkenes. These observed
jf
35
effects are summarized in Tables 5-6 and 5-7.
5.6.2 Alicyclic Hydrocarbons
Toxicologically, the alicyclic hydrocarbons, both saturated and unsaturated,
are similar to the aliphatic hydrocarbons in that they act as general anesthetics
5-26
-------
in
CO
z
0
CO
^
U
o
Of
Q
U
HH
1—
Q.
_J
^
S
s
o:
1
u.
o
i—
0
X
o
1—
CO
1
in
LU
L_
in
O
•r-
ua
4*0
c «
OHO
U JO
C 0.
O CL
U
[ *
c
01
•r—
JD
U
Ol
JO*
3
CO
s-
'5
-Q O
i< 01
,s
o
E
Q.
Q.
C
O
1 *
id
C
01
o
CJ
in
o
0)
tf-
<*-
01
5
r—
ID
Ol
(0
3
O
u.
c
•o
s.
a
u
2
•u
^
o
0
o
i
o
o
CO
rH
O
1— 1
§
o
o
s
V
in
tJ
Ol
4-
01
U
-P
O
z
•*•
z
o
Ol
c
CO
JC
+J
3EZ
in
in
o
o
c
R)
E
in
o
o
a
S
V
in
+J
U
01
M-
Ol
U
•*—
ai
in
3?
o
z
10
CJ
0)
c
10
j=-
+j
LU
on
i
CM
t-l
C C C
E E E
M CM O
§ § 1
O O -
- o
o o o
i-l CM rH
V V V
I- T3
ai • ai
+J Ol 4-> (0
*- S- O
(0 3 0) C C •
in 4J o *t— in
in o ai -i- j-» ai
E Q. *O 4^ ^ in -P
ox ra en in 3
+J Ol +J +J •!- 01 C
tO •!- r— C •!-
i- c s- in T- E
>> ai s- N
in -i- &. •!-•!•> N x
s- o 3 -i- ai
O J3 T3 O -Q T3 t-
z o z
00
u
ai
c
to
a.
o
s.
CL
CM
1
.-t
0
C3
C
gi
in
o
o
0
o
in"
V
T3
U
Ol
01
O
C
1
o
o
o
ai
c
a
3
BO
C
gi
O
r-5
o
o
0
A
O
c.
'E
S
CL
ai
u •
u x in
•f- Ol
E Ol
ai in c
+» +J •!-
in u in
>> 01 S
in
O
§ §
o
in
o
@l CM
C O
•i §
s g"
,
c
'i
s-
01 0
Ol JC IO
c +->
10 0 0
i- +->
ja s.
& o in
EC C
O • v
O) *r- U)
3 +J E U>
O IO O -i-
o +J +J in
3 •!- O. O
st§, e
o -i- in to
z z
r-.
CM
1
«^
01
c c c u
E E E E
'
in
o CM in o
O 0 0 CO
c
'i
o
r-t
«"
00 0
o o o
o o o
in CM in co
E T3 t-
•i- . CO
Jc in
in o in •
in -i- j *r- in *O O
0) in N c 73 u
&- N Ol «r- £.
jc o -r- in O) 10
i— z a z
i-4
u
ai
c
10
X
01
5-27
-------
q>
C
I
U
ID
tn
Uf
P:
in
o
•a
"§
^» m
o
s.
••
|
ncentra
o
u
in
u
"TO
ei
as
i—
S
o
LL.
c
1
ro
u
g
•0
•^
01
u
•r-
O
1
in
CO
g
o
T
g
o
in* •
ro
•a
•a in
c o>
ra s.
in a)
c >
lilt
C 09 b CT^
o *D 4- at
^
?
o
Jc c c
xx: £
rH CM in
0 OO O
c c G
"5*5 li
to<- «d-
ra o o o
in r-ftsi in ,
g
ra
c
f- o cT-a-S
• CD O> C £-
""" u s- •> u
XI 01 A) >>
f— > > -P <*-
O *f~ O
x: -p -at.
to .e cv ra ^
£a> j^ i— u
•r- t-'t- IO
j: r- ra j=i—
l-oo s:
S
oT
o
01
c
ra
a.
3j;
01
c u
£ £
in
in i-i
0 rH
c
•r- O
g 0
0 •
in - c
-• S'l
« 00
g g?
o «o
- , . O CO
in rH
• -a;
en a! c: -C
c t.^- o
f- 3 ro-i-
•a *> in r— F-
e tn o a. in
>, ro ro Q.E
•ff r- X 0 -
•i- in a) u at 01 ai
S- i— !- -P +-> C
ra ra o s- in T- in -i-
>>i— 3 Q. ai i— -P ra f—
^•r- -a 3 *> ra oi -P o
.C •!- +> *- 3 a. tn
*O >tnco*OO.rara
a) *o »f— f- ra a>
c N 01 -o ro • > T>
Ol-l— C C-r-t-CO)
•r- S- i— "r- 0> 1- "O O IO C
+> 01 0 S- E C ^'r-
u u -P E ^= o in to ja -i-
Xkoino in r— tn ai o
CU3*f— «r-<#-F—OCS- rO
H- * Z
*!^
co
o
0
8. =
o
in
CM O
i •
in o
r-l
(-
in
c
0
convulsl
S
s
o
-P
a.
01
t-
1
1
1
c
ra
c
ro
01
a.
a.
ra
in
0
4J
ra
t-
1 <
c
§
-------
in
co
U)
o
ca
oe
o
o
ce
1
o
1—
Q»
M
<
Q
UJ
^
«
z
u.
o
£
HH
O
2^
o
t—
^
1
UJ
H-
U)
c
o
n)
uo
c -
am
u .a
c a
o a
u
c
0)
.s
^
u
a>
•»-3
in
s~
ns
&« o>
E
"c
1
C
C
c
mcentr
w
O
V)
4:
a
4-
0)
r*
4-
n
0
T"
fl
~
£
O
it
C
c
Hvdrocarl
o
o
1
o
c c
id IB
s: E:
H in
o o
§§
o in
r-» m
JO U
^j f^
i^ ^-,
CM
CM
O
Ethyl ene
en r»-
CM .CM
in vo en CM CM
iii i i
rH rH rH CM CM
O O
u>
J3 0) 0)
C C C C J2 U U C C C C C
« « « « « £ .- m ra « « m
o
in
* * rt co CM •* m
.... i CM i • o o o in
OOOOO inOi-HCMfOCO
CM
CO
0 I.
o o
0> , s. -
CM T3 Id O O
1 T3 4- 1-
® O ro i o o o o
OrH OOOOOO
OOOO&-OOICM C3S-O O O O O
OOOO.C«UI O.COO- ...
caoooocin - -oo o o o
»J^«^rHOOCM CMPOinrH CMCOCO r~
•O rH
£-5
Ol U 0) i— N Q. >c;"£3E4J*-£- c o 5
rdS-ldt*-U)T-0>O-Q~O mr~ *r- Id
3SOIOCJC 0} 4^4^ t-
Q. id 4J ,+-> n> 18 • « oi
>>IOCU)X I/1V1U)O U^UCUl 4^
O) i- •»- 3 Ol O) O> O ••— C »r- *(- O Ol 'r-
•i- O J: C r— O • X X-r-C r—
4JQ.<='i-0)4->tdZOIU» 0 04JOJ ^_.
cn a) •-- 01 01 - >i o>u> c ecocf 01
.^ &• 4^ 4J t™ £„ ^— rj) . J o 'r* . 'I" *r~ . *^ .^ 4») ^
r— ID OJ +J O) -P C X U 4-> "OUE -Ol
oi*f-a>EoiD)3i — O) (d x:oj o> o cm *r* 3
"^j 'Ij s^> ^"tti. £ o '^- -c a) "ro —i-i-M-s-uoc ai'idi—
a. a. !C3«r-oM— x:uiu)t-t. ol 0.1 — o> id c c-r- c: >^3
ZZI — O Z -Zl Zt/> Z*-*^ "i— T-
t-4JU>
id •!- w
Ol E -i-
a. -r- E
UJOOIO OOCM Q.I— S-
z i z z x id 01
co «*• «a- in CM -a a.
o o o o o 01 i —
C O E
O) 1 O £ 3
c co «r- in E
HI * n ,*** 4J O) *r*
•r- rH 01 01 Id (. X
o>~a icca) s-jzid
COIId I— O)OIC 4J|— Z
oi c +-> >>•>- s. at • c
r=- 5 3 -f?T! 0. r- 01 1 1
>> +> • ea 4->ido>» o
£3 0)4JU>4J C > — 1
OOQPO Z3T-OI O — 1 O.
t- | 1 J3 •— x U O 1— Z
Q-rHrH CM< idjao
5-29
-------
and central nervous system depressants having a relatively low order of acute
toxicity. Cumulative toxicity from repeated exposure to low atmospheric
concentrations is unlikely because the alicyclics do not tend to accumulate in
body tissues. Massive acute exposure that results in prolonged unconscious-
ness, anoxia, and convulsions may cause central nervous system sequelae similar
to those described as occurring after exposure to volatile aliphatic hydro-
carbons. Like the aliphatic compounds, the alicyclic compounds are generally
biochemically inert, although not biologically inert and are only reactive at
concentrations of 100-1000 higher than those levels found in the ambient air.
The vapors in sufficient concentrations will cause irritation of the mucous
membranes. In general, the saturated hydrocarbons are less irritating than
the corresponding unsaturated compounds. No evidence was'found to indicate
that alicyclic hydrocarbons are specific hematopoietic toxicants. Like the
aliphatic compounds, the alicyclic compounds are generally biochemically
inert, though not biologically inert and are only reactive at concentrations
of 100 to 1000 higher than those levels found in the ambient air. Data con-
cerning the toxicity of cyclohexane and methylcyclohexane are summarized in
Table 5-8.37
5.6.3 Aromatic Hydrocarbons
The aromatic hydrocarbons are biochemically and biologically active. The
vapors are more irritating to the mucous membranes than equivalent concentra-
tions of the aliphatic or alicyclic hydrocarbons. Systemic injury can result
from the inhalation of vapors of the aromatic compounds; no effects, however,
have been reported below 25 ppm. This value is 1000 times greater than the
average benzene concentration reported in California (Los Angeles) during
1966.
5-30
-------
CO
ae.
t-H
<:
•z.
• • H- 1
CO
Z LU
§^3
CO
<£ O
U Q-
O X
ae * * *
a
3
U O
o a
>-z
o - CJ
SLL.
0 O
t— I
9< CO
O (—
1—0
LU
LU
. LU
CO LU
I
in LU
LU I— 1
TABL
COMPARAT
CO
CO
XI
a.
a.
Ul
C 0
•S *3
XI 10
c
01
u
0
u
ll
15
s
c
Ul
" «"
3 -o
Ul
o
a
Ixl t-
01
E
3
O T-
> ><<-
0 X? 0
s*
E-
•*•> w
C E
0) \
0 0,
o
o
i.
Q.
Ul
u
Ol
id
O 01
z o
r~
CO
u u u
c -c c
o o o
i- i. S-
o o o
Ul Ul Ul
XI XI XI
to to to
as.ce. an
0 O O
r*» p^ r^
«*••*«*•
CM CM CM
O rH 'CM
CD 0 0
co cn co
cn in CM
cn co CM
rH
rHCM 0
•>*• CO O
CM rH CO
rH CM
• • •
+» +•> 4->
U U U
oi ai ai
at a> a>
o o o
z z z
CO
3:
o
rH
^C
CO
O
1
o
u
U Ol
*>> <0
X? X
4_j a,
0) -C
u
c
o
i.
o
Ul
ae
o
r^«
cn
CO
CO
o
CO
CO
CM
CO
rH
O
0
CO
CO
ai
u
Ol
-a
'i
£.
•E
?>,
id s-
>>£>
ai c
C -i-
5 s-
*u* Ol
O r—
U
O
Ul
•r-
XI
Id
ae
CO
CM
CO
CO
in
o
CO
in
CM
CM-
o
o
CO
in
..S
Ul
Ol C
4J
"id s-
21
ai
O r—
z
u u
C C 01
O O 4->
j- t- a
x: x: u
O CJ «S
Ul U) Ul
•r- T- -r-
(O tO (O
ce of oe
*»••*
rH rH
C
E
CO CO O
^
CO
I~. o in
C3 rH r-i
CO O O
r-» o in
CM rH rH
en o o
CM «s- co
ss§
CO O O
o in
rH rH
• ui ui
Ul Ol Ol
CJ) *r- *r-
•1- *J +J
•r- I— ^
r— flj tO
4J 10 to
id i— H—
^,22
o
0 01
ui x
•t— Ol
lohexane
hylcycloll
U 4->
?!
0°
Ol Ol
3 3
id to
> >
+* +^
'i 'E
_j _i
•a TJ
o o
x: x:
VI Ul
Ol Ol
x: x:
! — r—
ai ai
x: x:
^— j—
id xi
5-31
-------
It is well established that benzene is an insidious toxicant which has a
destructive effect on the hematopoietic system. Experimental evidence indicates,
however, that the alky! derivatives of benzene are not capable of inducing
these effects and that benzene is unique among the aromatic hydrocarbons as a
myelotoxicant. This difference in toxicity between benzene and its alky!
derivatives has been primarily attributed to detoxification of the derivatives
via a different metabolic route. Pharmacologically, the alkyl benzenes or
phenylalkanes can be classified with the central nervous system depressants.
Toluene is a more potent narcotic and acute toxicant than benzene, but it
* 3ft
does not affect the hematological system. Xylenes are more acutely toxic
than benzene or toluene, but, again, they do not affect the hematopoietic
system. The comparative effects of acute and chronic exposure to aromatic
hydrocarbon vapors in air are shown in Table 5-9.39
5.6.4 Hydrocarbon Mixtures
Hydrocarbon mixtures produced and consumed in this country include natural
gas, liquid propane gas, petroleum ether, petroleum benzene, petroleum naphtha,
gasoline, mineral spirits, kerosene, jet and turbo fuels, and lubricating
oils.
Physiological responses resulting from exposure to these hydrocarbon
mixtures can be judged from the composition of the mixtures (Table 5-10) and
the information given above for physiological responses to the specific hydro-
carbons, methane through octane.
Of all the hydrocarbon mixtures, gasoline is undoubtedly the most extensively
used and represents a major, nearly homogeneous hydrocarbon mixture that is emitted
5-32
-------
on
CO
OL
«
Z
..M
CO LU
z a:
S3
CO
ce o
32=
oui
a u
1 z
o ae
HH IE
l—o
£ a
S *
LU
LL. |—
0=>
o
l-H LL.
O O
1— <
X CO
pl-
r- O
LU
• U.
0> LU
in LU
LU l-H
— * r-
§2
r- «t
a.
o
o
CO
a.
a.
P C
c o
01 -r-
S|
C
01
u
0
u
*J
5
CO
in
0) >>
S- R)
3TJ
U>
0
a
X
LU i-
01
E
r— t-
O •!-
> R>
§.§••5
•r-
RJ
s-
•P
c to
a> E
c "Si
O E
0
E
a.
a.
tn
P
u
t-
01
x:
+f
R)
01
3:
R)
0
LL.
C
o
ID
Q
•a
:="
^
in
i
LO
CM
O
O
1
f
1
1
01
u
LO
CM
0
o
0
o
00
LO.
CM
3
t-
^
o
01
01
IM
c
01
CO
i
1
1
01
3
U
rH
O
O
Cl
rH
CO
O
O
rH
01
C
R)
S- .
01 •!-
S -P
R)
in P
3 •!-
0 (.
u c-
3 •!-
Z
01
u
•r-
1
1
0)
JJ
3
U
,<•-£
01
to f— o
3 .O .
0 R) rH
> S- 1
i- 3 LO
0) T3 .
CCO
LU
01
U
sE
(
i
i
Q}
3
f«*^
«*
O
CO
S
^i
rH
O
O
c
o
^J
Q
i.
P
tn
£
Q.
0)
U
IE
(
i
Ol
3
£
*
o
o
CO
CO
CM
|
,
O
LO.
O
— 1
•
c
10
(
1
1
0)
3
2
LO
o
LO
CM
en
co
CM
0
o
LO
i.
O)
M- .
R> S-
•a-
0
S
«*
rH
O
O
i-H
O
—I
u>
1
1
1
01
3
U
-------
•a
I
+j
c
0
u
en
i
tn
3
€
to
CO
.n
Q.
a.
-p uT
c c
0) O
S *j
^J;
8)
U
0
u
u
O)
1
in
It
Ul
o
a
Ul t-
£
jk
C *>•
o >> o
.r-S"
jj
c a
u S
o m
o
1
i!
u
i
o>
a
*=.
« in in
at 10
3 •»•«••
O3 C in
5.2 i!
a in *>
I- 3 in
c +> *>
to o. c
I- S3)
a. >> e
v> t/>
c
ro
rH
CO
s
o
en
a
CM
O
O
to
in
U)
. -2
SB
S'-S
c
W 01
3 JZ
r— O
Q. (O
•a
0 10
in a)
5*
c
rH
Op
to
o
<=3
en
in
CM
CM"
0
o
to
c
a><->
o n> (0
r- C 0>i—
•r" C **•
in -Of- T3
3 C- I.
r— o ai in
a. o CTI—
U O)-r-
o 10 a.
in "*- +> 3
r- 0 W Q.
c
rH
00
§
O
CM
g
CO
§
CO
•a
01
N
i- a>
to 01 +»
« -U 0
*- 10
u> aj s-
O J= J=
+> +> o
fi-n
Ǥs
15t
ro ID HI
«rt
'
in ~a
in c
ai n)
in -
3 0)
O 3
£.?
0) +J
c to
by severe
muscular f
insomnia.
5-34
-------
•a
01
3
C
O
U
cn
tn
LU
>
3 -a
to
0
Ul t-
.c
i
3
r— S-
O 'f
> 10
O -Q O
10
i.
-P
ai
u n
C E
O ****
0 01
E
a.
a.
in
u
at
01
Si
s
z
^^
*"" _1
r|t tj
-CO) 4-
Ul 3 o
p- -x.
CM
o
^
u
in
to
o
OJ
ai
t-
ff
tn
D)
"Q.
+j 10
•i- at
ja c
*> -Q T-
10 c
O i. -r-
• » U> 4J * Ul
ui a) ui c ui
at at o ai
•i- S- -•>- C
•P D> Ul -P Ul
*r- O> Ul 10 3
i— -a o» c o
ro c •>- -r-
-p O)-^ T3 U •
10 C 10 i- Ul <"-N
, 3 O O -C
JC o o
10 4- C C O
> O •!- 3 rH
O>
"5.-!J
n s.
ai
'5 10
rH
rH
B
in
o
m
CO
o
CM
0
o
o
in
c
•r*
in
in
at
c
in
3
O
»r»
U
Ul *
C S-
8-=
C rH
=>
D)
'S.'P
10 (-
01
c -a
•r- C
3 a>
o
-
rH
CM
1-1
O
rH
0
in
r-T •
*
o
8
o
rH
Ul
m
in ai
to "O •
01 C
C •-•!-
= = &
O -i- O
•r- E IO
U 1
Ul O O
C rH ro
O
2 "
3
5-35
-------
1
c
+»
c
0
en
i
in
a
8
s
ja
It
IA
+* c
42 *»
3+»
S)
§
u
^
u
>O
f-
+» 0
I -1
1 B
i.
OL
irt
*»
a>
o
1
id
r—
O
U.
o
JQ
t.
a
t-
•a
en
rH
iH
•h
CM
O
u
•^"
X
1
1
1
u
0 *» 3
f U 0
to 01 >
ai 4- i-
t. H- O)
j= nj c
CM
?
^f
U9
*~s
•04*
raS
i 152
x ce
1 O
t CO
i i-i
1 "
en
CM ID
,O 0
0 0
co in
ID en
co en
CM
o en
CM ID
+i O>
*5 • *•
T-/-V OJ
f > O
•ot- *S '.
•— »-» 4-> U>
0
Oil— 4=1*-
i. at «-
j= > o at
1— Z
*>
S 01 01
J3 U U
a -^- -r-
ac XX
in i i
in ii
i i
a) O)
+> ••->
00 33
£ £
in- en o
t-4 ID CM
IH * en
o o CD
S g 03
O» Ol CM
cn co o^
<^ o ,— t-
U IA 01
Ol— 4->
J£ r~ 10
3 Oil—
Ol U Q. •
f C
•o-oo
•o o 011-
O O IA +»
IA i— CO 18
i-
t. *O U IA O
u a> c o in
aj t. i- t- o
a a. _i
o
0 4*
f to
X ee.
i i
i i
i i
1 1
u u
ID CM
CM r^
r4 t-5
i-i tn
O ID
en co
•f v?
in r»
0 O
in in
ID CM
CM f^
iH r-l
0 0
o3 cT
±1 _l
5-36
-------
4fl
TABLE 5-10. PRINCIPAL MIXTURES CONTAINING PARAFFIN HYDROCARBONS™
Mixture
Boiling range, °C
Principal paraffins
Natural gas
LPG ("bottled gas")
Petroleum ether
Petroleum bezin
Petroleum naphtha
Gasoline
Mineral spirits
Kerosene (coal oil)
Jet and turbo fuels
Lubricating oils
Gas at room temp.
Gas at room temp.
20-60
40-90
65-120
36-210
150-210
170-300
40-300
300-700
P P
Ll» L2
c3, c4
C4 to C
C5 to C
C6 to C
C5 to C
C? tO C
Cg tO C
to C16
and up
5-37
-------
nationwide, whether due to evaporative loss or combustion. Gasoline is a
mixture of C^ to C12 hydrocarbons, including paraffins, olefins, naphthenes,
and aromatics over a boiling range of 26°C to 204°C (80°F to 400°F). Alkanes
and aromatics generally constitute the largest fraction, but olefins and
naphthenes are also present. In addition to the 200 hydrocarbons generally
contained in gasoline, nonhydrocarbons (additives) are added to improve engine
performance. Variations in its hydrocarbon composition do not significantly
alter the basic pharmacology and toxicology of gasoline. An acute exposure
to gasoline vapors will elicit the symptoms and signs of intoxication described
for exposure to heptane. The atmospheric concentrations and the duration of
exposure required to elicit these responses will differ, however, with the
composition of the gasoline. The concentration of gasoline.hydrocarbons that
causes mucous membrane irritation will vary with the degree of branching of
the paraffins and the content of alky! derivatives of benzene and olefins.
The foregoing does not apply to gasolines that contain a significant concen-
tration of benzene, which appears to be unique among hydrocarbons in its
effect on blood-forming tissues. In light of the variability of the composi-
tion of gasolines (blends), a single threshold limit value is not applicable;
however, as a general rule the contents of the aromatic hydrocarbons and, to a
lesser degree, of the additives are used to arrive at the appropriate TLV
42 43
values. Based on Machle's summary (Table 5-11) of man's response to air
concentrations of the volatile fraction (<230°F) of unleaded straight-run
gasoline, Gerarde recommended 500 ppm as a TLV for gasoline in 1962.
5.7 WELFARE EFFECTS
Hydrocarbons were recognized as phytotoxic air pollutants at least as
early as 1871 as a result of injury to green house plants from illuminating
5-38
-------
TABLE 5-11. HUMAN RESPONSE TO GASOLINE VAPORS DISTILLING BELOW 230°F
43
Concentration,
ppm
Exposure time
Response
550
900
2,000
10,000
1 hr No effects.
1 hr Slight dizziness and irritation of eyes,
nose, and throat.
1 hr Dizziness, mucous membrane irritation,
and anesthesia.
10 min Nose and throat irritation in 2 min;
dizziness in 4 min; signs of
intoxication in 4 to 10 min.
5-39
-------
gas. Renewed interest in hydrocarbons, and ethylene in particular, occurred
in the mid-1950s when ethylene was found to be one of the primary pollutants
in the photochemical reaction complex. Although additional combustion sources
contribute to ethylene in the atmosphere, the internal combustion engine is
the major source of ethylene pollution in metropolitan areas.
The levels of ethylene found in the ambient air vary according to location,
time of day, and weather. Low levels in the range of 3 to 5 ppb have been
reported for desert areas and Davis, California. ' In contrast, Saii Francisco
air has been reported to contain peaks of 100 ppb for an hour or more on
2 45-47
several occasions; 5 however, the average for that city was 50 ppb in the
2 45 48 49
mid-1960s. ' » The Pasadena area has, in the past, reported an unusually
high value of 500 ppb, while surrounding areas, such as the San Gabriel Valley,
have had lower peak values, i.e., 30 ppb.
2,47
Similarily, on the east coast,
Washington, D.C., reported average values of 10 to 20 ppb ethylene with maximum
values near 60 ppb. In Frederick, Maryland (50 miles west of Washington, D.C
51
a rural area at the time, ethylene levels ranged from 1 to 5 ppb.
Research on several unsaturated and saturated hydrocarbons has shown that
only ethylene has adverse effects on vegetation at atmospheric concentrations
of 1 ppm or less. Concentrations as low as 0.05 ppm have caused.malformations
in certain plants and injuries to flowers. This concentration is well within
the range measured in ambient atmosphere. Other unsaturated hydrocarbons
52
produce similar effects, but at much higher concentrations (Table 5-12).
Such concentrations are orders of magnitude higher than the concentrations of
ethylene found in the atmosphere.
Research has demonstrated that ethylene is produced naturally within
tissues of plants and serves as a hormone in regulating growth and development
and other processes such as the ripening of fruit. Thus, ethylene is unique
5-40
-------
TABLE 5-12. COMPARATIVE PHYTOTOXICITY OF UNSATURATED CARBON GASES
52
Minimum concentration of
gas producing response, ppm
Gas
Ethyl ene
Acetylene
Propylene
Carbon monoxide
Butyl ene
Sweet peaa
0.2
250
1,000
5,000
-r * b
Tomato
0.1
50
50
500
50,000
Effect noted: declination in sweet pea seedlings (3-day exposures).
^Effect noted: epinasty in tomato petiole (2-day exposures).
5-41
-------
in being an endogenous plant-growth regulator and a serious phytotoxic air
pollutant. A delicate balance between auxin and ethylene in the petiole is
required for normal defoliation; however, an artificial increase in ethylene
through pollution would upset this balance and cause early abscission or
defoliation without noticeable leaf injury. Unfortunately, this plant-growth
inhibitory effect does not characterize ethylene, because other pollutants at
sublethal dosages, as well as some diseases and environmental factors, may .
also inhibit growth. The various effects corresponding to different levels of
2
ethylene are illustrated in Table 5-13.
It should be emphasized that data on the phytotoxicity of ethylene were
not used in promulgating a secondary or welfare standard for hydrocarbons, but,
like the primary standard, the secondary standard was based, on the specifi-
cations of the oxidant standards, which automatically fixed the hydrocarbon
standards. Prior to 1970, various standards for acceptable levels of ethylene
were proposed or adopted. In 1962, the State of California adopted ambient air
3
quality standards for ethylene of 500 ppb (575 ug/m ) for 1 hour or 100 ppb
o
(115 ug/m ) for 8 hours. These are considered "adverse" levels, based on the
53-55
damage to vegetation. The American Industrial Hygiene Association (ACGIH)
cc
has also proposed air quality standards for ethylene. Based on insufficient
data at that time, these latter standards were never unequivocally adopted and
served only as guidelines. The recommended standards are given in Table 5-14.
56
5-42
-------
TABLE 5-13. DOSAGE-RESPONSE RELATIONSHIPS OF VARIOUS PLANTS TO ETHYLENE'
Dosage
Response
Abscission
Cotton leaves, square
Cotton leaves
Pepper and tomato
flower buds
Rose leaves
Snapdragon petals
Chlorosis on leaves
Cotton (slight)
Cowpea
Rose
Death of plant
Cowpea
Dry sepal injury
Orchids (severe)
Orchids (typical)
Orchids (slight)
Epi nasty
. African marigold
Various plants
Flowers do not open
Carnation
Orchid
Flowers close
Carnation
Growth inhibition
Cotton
Lily family
Various plants
Loss of apical dominance
Cotton
(jg/m
46-3,435
685
115
345
11,450
575
685
2,290
1,145
2,290
115
46
23
5.75
345
57.5
11.5
1.15
345
3,435
2,290
115
575-1,145
115
575
685
46-3,435
860
2,390
46-3,435
ppm
0.04-3.0
0.6
0.1
0.3
10.0
0.5
0.6
2.0
1.0
2.0
0.1
0.04
0.02
0.005
0.3
0.05
0.01
0.001
0.3
3.0
2.0
0.1
0.5-1.0
0.1
0.5
0.6
0.4-3.0
0.75
2.0
0.04-3.0
Time
Not stated
1 month
Less than
8 hr
120 hr
24 hr
1 hr
1. month
1 day
5 days
10 days
8 hr
8 hr
24 hr
24 hr
1 hr
6 hr
24 hr
20 hr
24 hr
3 hr
10 days
3 days
20 hr
6 hr
12 hr
1 mo
Not stated
7 days
10 days
Not stated
5-43
-------
TABLE 5-14. AMERICAN INDUSTRIAL HYGIENE ASSOCIATION
RECOMMENDED STANDARDS FOR ETHYLENE56
Rural
Residential
Commercial
' Industrial
1-hr
ug/m
287.5
575
862.5
1,150
max3
ppm
0.25
0.50
0.75
1.00
8-hr
H9/m
57.5
115
172.5
230
max3
ppm
0.05
0.10
0.15
0.20
The rural concentrations should not produce adverse effects in the most
sensitive plants. The residential concentrations should produce only slight
injury to the most sensitive plants. - !
5-44
-------
6. SCIENTIFIC DATA BASE ON HYDROCARBONS,
1970 THROUGH PRESENT
6.1 CONTRIBUTION OF HYDROCARBONS TO OXIDANT FORMATION
A brief and highly simplified summary of the role of hydrocarbons in the
formation of photochemical oxidants was presented in Section 5.1. Since
publication of the 1970 criteria document on hydrocarbons, research has produced
new findings in the area of reaction mechanisms but has consistently substantiated
the basic chemistry of oxidant formation as presented in the 1970 document.
The new findings and their significance have been discussed in.detail,in three
recent reviews, Ozone and Other Photochemical Oxidants, Air Quality Criteria
for Ozone and Other Photochemical Oxidants, and Air Quality Criteria for
Nitrogen Oxides. Consequently, only brief mention is made here of the
significant research findings of the past decade.
A major accomplishment in recent years is the development of computer
r-7 ro
techniques for simulating the atmospheric smog-forming process. ' Such
techniques have provided a useful tool for identifying and studying those
reactions that play a key mechanistic role and for predicting potentially
important but as yet unidentified reaction products in the ambient atmosphere.
The most important mechanistic finding of recent years is that the
hydroxyl radical (OH*) is responsible for much of the oxidation of hydrocarbons
and aldehydes in the atmosphere.59'60 Hydroperoxy (H02') and alkylperoxy
(R02*) radicals have also been identified as having major roles in the
oxidation of NO to NOp, and some sources of these radicals in the atmosphere
have been identified. In addition, more complete information is now available
to account for the degradation of organic oxidant, the oxidation of NO to N02,
6-1
-------
and the formation of organic products in the ambient atmosphere, for paraffinic
and olefinic reactants though not for aromatics. Another noteworthy mechanistic
finding is that some nitrogenated components of the photochemical smog system,
e.g., peroxyacetyl nitrate (PAN), may have greater roles in atmospheric photo-
chemical processes than thought earlier.
In addition to new information on reaction mechanisms involving hydrocarbons,
new reaction products of photochemical processes have been identified or
postulated, including organic compounds and a number of inorganic nitrogenous
acids.
Three additional areas of hydrocarbon air chemistry have been the subjects
of rather extensive research over the past decade: (1) the role of gas-phase
natural hydrocarbons, in photochemical processes; (2) the role of gas-phase
hydrocarbons in the formation of secondary organic aerosols; and (3) the con-
tribution of gas-phase hydrocarbons to the oxidation of sulfur dioxide and the
consequent production of acidic precipitation.
The role of natural hydrocarbon emissions in photochemical processes and
reactions, which is a topic of major importance in the context of ozone control
strategies, was treated in detail in Air Quality Criteria for Ozone and Other
Photochemical Oxidants and will again be addressed when that document is
revised. The question of the role of gas-phase hydrocarbons in the formation
of secondary organic aerosols is treated at length in the NAS document, Ozone
and Other Photochemical Oxidants. and in some detail in Air Quality Criteria
for Sulfur Oxides and Particulate Matter;61 but will be briefly summarized
here to demonstrate the diversity and divergent chemical behavior of individual
species within the class of compounds known as hydrocarbons. Likewise, the .
formation and effects of acidic precipitation are discussed in Air Quality
6-2
-------
o
Criteria for Nitrogen Oxides, but the role of hydrocarbons in contributing
free radicals that oxidize sulfur dioxide (S02) to sulfur trioxide (S03), from
which sulfuric acid is formed in the atmosphere, is not discussed in that
document and thus will also be summarized here.
Secondary organic aerosols are aerosols containing organic compounds that
are formed in the atmosphere by means of gas-phase photochemical reactions in-
volving the primary pollutants, hydrocarbons and nitrogen oxides, and a secondary
pollutant, ozone. Using a chemical-element balance technique that he developed,
CO
Fried!ander estimated that secondary organic aerosols accounted for 76
percent and 82 percent of the aerosol carbon in 2-hour and 24-hour samples,
respectively, in Pasadena, California, in September 1973. During a severe
photochemical episode in which the ozone maximum was 0.67 ppm, secondary
organics reached as high as 95 percent of the total ambient air organics.62
Investigations into the chemical composition of secondary organic aerosols,
which necessitate the use of sophisticated techniques and instrumentation,
have shown that carboxylic acids and dicarboxylic acids abound, along with
lesser amounts of other aliphatic di- or multifunctional oxygenated compounds,
aromatic monofunctional oxygenated compounds, and terpene-derived oxygenates.61"67
Most of the compounds are difunctional compounds that could arise through the
photochemical oxidation of cyclic alkenes and alkadienes.9 Ambient air data
on gas-phase organics and on aerosols reveal that organics always account for
an important fraction of urban aerosols, but that only a small fraction of the
q cp
gas-phase organics in ambient air is converted into particulate matter. '
In view of these two facts, researchers have sought to identify those
specific hydrocarbons and other organics that generate aerosols. Smog chamber
studies have been conducted using mixtures of hydrocarbons irradiated in the
6-3
-------
presence of nitric oxide or nitrogen dioxide or both in "clean" air, in ambient
air, and with and without the addition of sulfur dioxide, water, and ammonia.
These smog chamber studies have shown that, in the absence of sulfur dioxide
(the presence of which would result in inorganic aerosols), aerosol formation
depends strongly on the type of precursor hydrocarbons irradiated. The trends
observed from smog chamber data and from examination of vapor-pressure/volatility
9
data are enumerated below.
1. A threshold concentration exists for each hydrocarbon precursor, below
which no organic aerosol is formed.
2. Most paraffinic hydrocarbons, even at high concentrations, do not generate
aerosols when irradiated. However, the more reactive, higher branched-chain
paraffins (>6 carbon atoms) will generate aerosols if irradiated for a
long time. ,
3. Acetylenics do not form aerosols.
4. Carbonyl compounds (ketones, 0,-Cy aldehydes, dialdehydes) do not generate
aerosols.
5, The role of aromatic hydrocarbons in generating aerosols remains uncertain
because conflicting data have been reported. Threshold concentrations
for any aromatics that may form aerosols remain unknown as well.
6. At high concentrations, all unsaturated hydrocarbons can form organic
aerosols when reacted with ozone.
7. At parts-per-million concentrations, alkenes having fewer than six carbon
atoms do not form aerosols. Aerosol data obtained with higher concentrations
of these low-molecular-weight alkenes cannot be extrapolated to atmospheric
concentrations.
8. At lower concentrations (1 to 10 ppm), the chain length of alkenes produces
a marked effect on aerosol formation. Alkenes with more than six carbon
6-4
-------
atoms form aerosols provided scission of the double bond(s) yields a
fragment of five or more carbon atoms.
9. Threshold concentrations for >C7 alkenes are in the parts-per-million to
parts-per-hundred-million range; arid determination of their gas-phase
concentrations is necessary for any assessment of their contributions to
secondary organic aerosols.
10. Cyclic olefins and diolefins form aerosol even when present at low con-
centrations. Cyclic olefins and diolefins from much more aerosol per
unit of precursor than 1-alkenes that have the same carbon number. The
same effect of chain length and double-bond position is observed for
cyclic diolefins as for alkenes.
These data on aerosol precursors are summarized in tabular form in Section
6.5.1 (Table 6-27). Ambient air concentrations of those compounds that can
generate aerosols are given in Section 6.4, along with the ambient air concen-
trations of other gas-phase hydrocarbons.
Hydrocarbons have not only been implicated in the production of secondary
organic aerosol but also in the production of urban sulfate aerosol and ulti-
mately in the acidification of rain. The acidification of rain and snow is
directly caused by acids, namely, hydrochloric, sulfuric, and nitric. Hydro-
chloric acid may be directly emitted from coal fired power plants; however,
the latter two acids are initially emitted as the respective anhydrides (S02
and NCO and are transformed to sulfuric and nitric acids during atmospheric
transport by a complex series of chemical reactions.
Oxidation of S02 in the atmosphere can either take place in the gas
phase or in solution. Both processes contribute to the presence of sulfate
in cloud water and rain water. The acidification of rain is principally due
6-5
-------
to oxidation of S02 in cloud droplets and rain drops, the absorption of free
gaseous acids (e.g., S02) by droplets and raindrops, or the scavenging of acid
aerosol by the falling raindrops. A series of potential gas-phase reactions
by which SCL may possibly be oxidized in the atmosphere is depicted in Table
71 72
6-1. Results of recent chemical rate studies of S02 oxidation in water
were extrapolated to sulfate formation in water droplets under atmospheric
conditions using oxygen, ozone, and hydrogen peroxide (H202) as the oxidants.
Extrapolation of these laboratory data.indicates that in a relatively clean
atmosphere (5 ppb S02, 50 ppb 03, 1 ppb H202) the favored oxidizing agent is
H202 rather than 03 or 02, provided the pH is less than 5.5. Overall, both of
these reactions probably contribute to worldwide urban sulfate production on a
large scale, and ultimately, to the acidification of rain. •
This information bears on the broader problem of the potential welfare
effects of acidic precipitation, which are discussed briefly in Section 6.7.
It is apparent from the examples cited above that information generated
since the 1970 criteria document was issued has provided a much more complete
view of probable atmospheric oxidant formation processes and the role of
organic compounds in those and'other atmospheric processes, though much of the
evidence for reaction mechanisms has been obtained from laboratory studies and
remains to be verified in the ambient atmosphere. The elucidation of atmospheric
reaction mechanisms, along with other research of the past decade, has confirmed
that photochemical oxidants in the ambient air are a function of the presence
of hydrocarbons, as well as other organic compounds and nitrogen oxides, in
7-9
the atmosphere.
6.2 MEASUREMENT METHODOLOGY
The essential features of the 1971 Federal Reference Method for the
measurement of nonmethane hydrocarbons were described in Section 5.2.
6-6
-------
TABLE 6-1. POSSIBLE ATMOSPHERIC GAS PHASE S02 OXIDATION PROCESSES
71
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(ID
(12)
(13)
(14)
(15)
(16) ..
(17)
Initial Steps
0
S02 + hv(2400-3400A)
so2 + o2
S02 + 0 + M
so, + o3
S02 + N03
S02 + N205
SO, + CH,0.>
&. O £.
S02 + H02
S02 + OH + M
Possible Subsequent
S03 + H20
S03* + H
HS03' +
HS03* +
HS03* + HSO
HS03- + H
HS03' -H
HS04* + H
-> so2
•* (so4)
^ S03 + M
•* SO, + 09
O £-
-. S03 + N02
' * S03 H- N204
-»• SO, + CH- 0
o *3
-»• S03 + OH
-> HS03 + M
Reactions
'•»• S03 *H20 -> H2S04
20 •» nucleus
OH -»• H£S04
OH -»• H90 + SO,
- b O
*
3 "* H2S2°6
20 •* HS03* (H20)
0, -»• HSO,
c, O
I20 * HS05* (H20)
6-7
-------
Instrumentation for total and nonmethane hydrocarbon measurements by
flame-ionization detection has not changed appreciably since 1971. For that
reason, this section mainly presents a critique of the method, based largely
on work done since 1971; and presents some of the problems associated with the
application of mass measurements of hydrocarbons. In addition, information is
presented on the use of gas chromatographic techniques for hydrocarbon
measurement and on developments since 1971 in FID measurement methods.
The EPA reference method for nonmethane hydrocarbons (NMHC) involves the
gas chromatographic separation of methane from total hydrocarbons in an air
sample. This is followed by FID measurement of the separated methane and of
total hydrocarbons containing methane, in aliquots of the same air sample.
The concentration of NMHC is obtained by subtraction of the methane
measurement value. Presently available commercial analyzers are still based
on these principles and provide a measurement of total "nonmethane hydrocarbons"
by subtraction of separate measurements of total hydrocarbons (THC) and methane
in an FID. Methane is measured apart from THC in one of twp ways, either by
separation in a chromatographic column or by selective oxidation in a .heated
oxidizer. 5 The chromatographic separation of methane that is used in the
originally promulgated Reference Method represents a significant improvement
in methodology over the separation technique used prior to 1970. The catalytic,
selective oxidation of nonmethane hydrocarbons to carbon monoxide in a heated
oxidizer makes it possible to determine methane separately. By subtracting
the FID value for methane from the measurement of THC, the NMHC concentration
is obtained. * In recognition of the fact that the FID responds to
organically bound carbon in compounds other than hydrocarbons, the measurement
6-8
-------
of NMHC is now referred to as nonmethane organic compound (NMOC) measurement.
This nomenclature is more accurate than the previous nomenclature and will be
used where appropriate in the remainder of this section.
Studies of the FID Federal Reference Method, using commercial analyzers,
have been reviewed in a recent EPA draft report and two such studies are cited
73
here from that report.
The first of the two studies involved the analysis of known synthetic
mixtures of NMHC in compressed air cylinders by 16 different users of the
reference method.74 The NMHC concentrations tested were 0.23 and 2.90 ppm C.
As shown in Table 6-2, the majority of the measurements of the lower concentration,
which was near the HC NAAQS level of 0.24 ppm C, were in error by 50 to over
100 percent. At 2.90 ppm C, the majority of the measurements were in error by
only 0 to 20 percent. Thus, the higher the concentrations measured, the
better the accuracy. The major causes of error were the inability of the
instrument to measure low levels of NMHC; the complexity of the instrumentation;
and the inability of the average user to identify and correct problems.
Table 6-2 PERCENTAGE DIFFERENCE FROM KNOWN CONCENTRATIONS
OF NONMETHANE HYDROCARBONS OBTAINED BY SIXTEEN USERS74
Known NMHC
concentration
0.23 ppm
2.90 ppm
Percent
>100
6
2
50-100
4
— —
20-50
3 •
3
10-20
2
2
0-10
1
9
In the second study, aliquots of the same ambient air sample were measured
75
with five different commercial FID instruments operated by skilled personnel.
Results from different analyzer pairs agreed within 0.1 to 0.'5 ppm C. Although
6-9
-------
these differences represent only 1 to 5 percent of the full scale of the
instrument (0 to 10 ppm, the range necessary to include all ambient air HC
values), and are normal errors for ambient air monitoring, they are quite
large relative to the NAAQS of 0.24 ppm C. As these two studies show, even
under optimum operating conditions, agreement between instruments and measure-
•70
ments is poor at low concentrations.
In a fairly recent study conducted for EPA, the reference method was
subjected to a comprehensive evaluation that included testing of six commercial
FID instruments. Generally, results showed poor performance; the major problems
were wide differences in response to different NMHC species and discrepancies
that were apparently related to variations in ambient relative humidity.76
In its assessment of the flame ionization detection method and instruments
based on that method, EPA summarized the problems as follows:14
1. The FID responds non-uniformly to various organic compounds.
2. The FID response varies with design and operating conditions.
3. The normal FID response is roughly proportional to the "carbon
number" of the compound, but otherwise has little or no relationship
to the photochemical reactivity of the compound.
4. Normal instrumental variability errors are substantially increased
by the need to subtract two separate and independent measurements
and by the need for higher scale ranges to accommodate the THC and
methane concentrations, which can be much higher than the NMOC
difference.
5. The FID is susceptible to zero drift and possible interference from
water vapor.
6-10
-------
6. Calibration procedures have not been standardized and calibration
may be based on methane, propane, or other compounds depending on
analyzer design or operator preference.
7. The FID requires hydrogen, which presents a degree of operational
hazard.
8. Analyzers using chromatographic separation of methane are operationally
complex and require higher levels of operator training and effort to
maintain than most other types of ambient monitors. These problems
appear to limit the useful sensitivity of currently available analyzers
to a few tenths of 1 ppm (methane equivalent). .
The major limitations of the FID measurement method for hydrocarbons
result in some very practical and pertinent consequences that must be mentioned,
if only briefly. First, the high measurement error associated with the method
results in erroneous values for hydrocarbon concentrations in ambient air.
This is particularly true for concentrations in the range of the present NAAQS
for hydrocarbons. This measurement error must be kept in mind when NMHC
ambient air data are compared with exposure data from human or animal toxicity
studies. In addition, the errors associated with FID measurements limit the
confidence that can be placed in the use of the FID method for routine monitoring
and in the use of FID data for predictive purposes or for verification of air
quality trends of emission inventory accuracy. It should be noted, however,
that at higher NMOC levels, e.g., those required for predictive models such as
EKMA (Section 6.5), the errors in NMOC measurements made by current methods
are commensurate with the variability of the models, because the models themselves
have limited precision arising from necessary assumptions, estimates, and
14
simplifications.
used to develop or establish the model.
In many cases, the same NMOC methods have been or will be
14
6-11
-------
Second, the FID technique produces measurements of total mass of NMOC in
ambient air. Measurement of the mass of total NMOC in ambient air, however,
is of somewhat limited utility in the overall control of ozone air quality.
This limited utility is related to four factors: (1) Other volatile organic
compounds besides hydrocarbons are photochemically reactive but, even if de-
tected, their mass is not quantitatively measured by the FID method. (2) The
measurement of hydrocarbon mass alone does not give any indication of the
reactivity with respect to ozone formation of the hydrocarbon mixture measured.
(3) Mass measurements cannot be related to the molar concentration of total hydro-
carbons in the air sampled. (4) Total hydrocarbon mass, without identification of
individual hydrocarbons, does not permit identification of those sources that
contribute most heavily to ozone formation. As stated in the 1970 criteria
document for hydroparbons:
As a consequence of the different reactivities of individual
hydrocarbons, it is impossible to predict accurately the rate of
consumption of hydrocarbons in photochemical air pollution unless
the detailed composition of the hydrocarbon Component is.known or
can be estimated. Knowledge of the total concentration of hydrocarbons
is insufficient, since two atmospheres .having the same total hydrocarbon
measurement may contain individual hydrocarbons of very different
reactivity and thus exhibit very different rates of hydrocarbon
consumption and photochemical air pollution development.
An EPA study on reactivity of hydrocarbons from mobile sources concluded that
since "hydrocarbons do not participate to the same extent in atmospheric
reactions leading to oxidant formation,, methods for measurement of more than
mass of emissions are necessary to project adequately the impact of mobile
sources on photochemical pollution." The same is true with respect to the
impact of HC emissions from other sources. It is apparent that FID measurements
are inadequate for the complete characterization of hydrocarbons in ambient
air.
6-12
-------
There are applications, howeverj for which mass measurements of NMOC must
be obtained and for which total NMOC concentrations are the best available measure
of the oxidant-forming potential of organic compounds in the ambient air.
Studies are under way in EPA to develop better methodology for NMOC measurements
7ft
for such applications.
Consequently, modifications of the FID method as well as the development
of other NMOC methods are in progress, though these methods and techniques
will probably not be available for routine use for several years. These
12 14 78
developments are briefly summarized below. ' '
In a variation of the present method for chromatographic separation of
methane from nonmethane hydrocarbons, chromatographically separated nonmethane
organics are backflushed and are converted to C02 rather than methane. The
COp is then measured by conventional infrared methods. This technique is
presently used to measure source concentration levels. It is a direct measure-
ment of NMOC, avoids the FID response problems, and does not require hydrogen.
It has, however, the operational problems of a chromatograph as well as the
sensitivity and response-time problems associated with low-level infrared C02
measurement. '
Another method for measuring methane is gas filter correlation, an optical
technique which can detect methane specifically without prior separation from
other ambient gases. The total hydrocarbon concentration is measured conven-
tionally by FID, and the NMOC concentration is obtained by difference. This
technique may prove to be a useful alternative to the currently available
selective oxidation technique, but otherwise promises no substantial advantages.
In another approach, which is currently under consideration by EPA, an
ambient air sample is passed through a scrubber which has been shown to eliminate
14
6-13
-------
>99.98% (330 ppm is reduced to <60 ppb) of the carbon dioxide (C0£) in the
79
sample. The air stream is then split quantitatively into two streams, one
of which passes through a converter that converts both the CO and the NMOC in
that half of the air sample to C02- The second stream bypasses the converter.
Downstream of the converter, the two streams are rejoined and pass through a
chamber where optical absorption measurements are made by gas filter correlation
of both the CO and the C02- Since the stream passing through the converter
contains one-half the CO concentration and one-half the NMOC concentration in
the original sample, and since the concentration of CO in the stream bypassing
the converter is one-half that of the original sample, subtraction of CO from
C02 gives a quantitative measure, when multiplied by 2, of the NMOC concentration
in the sample. This approach appears highly advantageous, in that it would
avoid all the problems associated with conventional NMOC methods.*4'78'79
EPA is also investigating non-continuous gas chromatographic techniques
in which samples are collected in inert bags and analyzed in a central laboratory
using a cryogenic preconcentration procedure and a gas chromatograph with a
special column designed to measure specific organic compounds. The chromatograph
can be calibrated with specific individual compounds or a single compound, and
total hydrocarbons can then be estimated as CH- by summing the response for
all compounds of interest. In this approach, samples can be collected by less
experienced personnel, stored, and later brought to the central laboratory for
analysis. This technique is a simplified variation of the GC-mass spectrometry
techniques often used to identify and measure specific organic compounds for
developing and verifying some types of photochemical models.14 Recent work
has indicated the possibility of further simplifying the technique by eliminating
the gas chromatographic column, selectively trapping the NMOC cryogenically,
6-14
-------
and passing the preconcentrated NMOC through the detector of the GC, so that a
single response is obtained that is proportional to the NMOC concentration of
79
the sample. These methods would have the disadvantages of problems associated
with storage of ambient air samples as described later in this section.
In addition to the above methods, more unique approaches for measuring
NMOC have been proposed, including the chemiluminescence method for reactive
hydrocarbons noted above and the use of chlorine atoms to abstract hydrogen
atoms from organic molecules to form CO and C02- These approaches would
require much further research, a redefinition of non-methane hydrocarbons, a
determination of the relationship of these measurements to the photochemical
formation of ozone, and possible modification or adaptation of the photo-
1-2 14
chemical models. '
Though the more promising of the above methods are under development now,
the research in progress represents a rather recent effort to refine and
improve NMOC measurement methodology. The relatively slow development of NMOC
methods can be attributed partly to the fact that the HC NAAQS, in keeping
with its guideline nature, has never been enforced and compliance monitoring
is not required under EPA regulations . In part, however, the relatively slow
development of NMOC methodology can be attributed to a lack of scientific
12
consensus on how NMOC measurements should be defined or expressed. Should
they, for example, be related (1) to molar volume; (2) to mass; (3) to the
product of carbon number and molar volume; or (4) to the product of molar
12
volume and some reactivity factor? Research on NMOC methodology will
continue since there exists a demonstrated need for reliable and accurate, as
well as inexpensive, field instrumentation for measuring total NMOC in ambient
80 81
air and in source emissions. ' Measurements of NMOC in ambient air are
needed for diverse applications such as (1) analysis of trends in ozone air
6-15
-------
quality versus precursor ambient air levels, to determine the efficiency of
control strategies; (2) verification of emission inventories by comparison
with empirical data; (3) acquisition of input data for models used to determine
reductions in hydrocarbon or volatile organic emissions needed to achieve and
maintain the ozone standard; (4) acquisition of data required for the prediction
of the impact of changes in the spatial configuration of sources, of the
addition of major new sources, or of changes in temporal patterns of emissions;
and (5) possible development of new or revised empirical models, since some
empirical models may have been developed from NMOC measurements of doubtful
quality.
Although there are numerous applications for which total NMOC are useful,
there are also applications for which the detailed hydrocarbon composition,
qualitative and quantitative, of ambient air must be known. Now, as in 1970,
the only method available for compositional analysis of hydrocarbons in ambient
air is gas chromatography (GC). ' It is an accurate and reliable method,
commercially available, for the measurement of individual hydrocarbons,.NMHC,
73
and total hydrocarbons. With flame ionization detection, it is sensitive in
the ppb range. With appropriate columns and temperatures, almost any desired
73
separation of components can be achieved. A number of factors, however,
preclude the use of gas chromatography for the long-term, routine analysis of
hydrocarbon mixtures in ambient air. First, the instrumentation and its
operation are expensive. Second, the instrumentation is sophisticated and
requires operator skill and experience. Third, data reduction is tedious and
time-consuming. This problem, especially acute in 1971 when the NMHC Federal
Reference Method was promulgated, has been alleviated by the development of
automated systems that reduce, and often eliminate, the need for manual
identification and calculation of the chromatogram peaks. These systems also
6-16
-------
permit automatic summation of individual hydrocarbon concentrations. In spite
of these advances, however, the effort and personnel still needed for data
reduction make this step expensive and time-consuming. Fourth, dynamic
calibration with standards of individual hydrocarbons or hydrocarbon mixtures
is required. When unknowns are encountered, or when resolution of peaks is
poor, GC analysis must be supplemented with infrared or mass spectrometry for
final identification.73
Though calibration and the need for supplementation by spectrometry are
hindrances to the routine use of GC techniques, they also constitute part of
the overall specificity of GC methods that is advantageous in detailed
hydrocarbon analysis. Dynamic calibration with known compounds or mixtures
means that quantitative evaluation of individual chromatogram peaks can be
made without standardization of each compound peak. The concentration of each
component can be computed by using an average per-carbon factor determined
.from measurements of known concentrations of the identified components. The
fact that GC techniques can be coupled with spectrometric techniques ensures
the confirmation of compound identity; spectrometric techniques in and of
themselves are imprecise for the determination of compounds in a mixture, but
coupled with GC they become valuable in the final identification of organic
compounds. An additional advantage of GC techniques is their sensitivity at
the ppb level. In urban atmospheres, hydrocarbons generally exist at levels
high enough to permit the collection and analysis of samples without pre-
concentration. For analysis for rural air samples, in which hydrocarbons are
typically present in parts-per-billion quantities or fractions therof, cryogenic
trapping techniques are available for the pre-concentration of large samples.
An additional problem exists that complicates the use of gas chromato-
graphy—sampling procedures even for short-term studies. As yet, no material
6-17
-------
has been found that is ideal for use in fabricating air sample containers.
Flexible bags made of Tedlar and Teflon are often used for sample collection
and storage, but these materials result in some hydrocarbon losses and some
contamination by bag emissions. Tedlar bags have been shown to be adequate
for the storage of paraffinic and olefinic hydrocarbons (0.0 to 0.3 ppm C) for
up to 191 hours. Aromatic hydrocarbons showed a 5 to 15 percent decay over
83
this period. Storage of zero-hydrocarbon air in Tedlar bags for 4 hours,
however, resulted in production of up to 40 ppb C. Longer storage~24 to 48
hours—resulted in acetaldehyde and acetone in excess of that level. Irradia-
83
tion prior to use increases emissions from Tedlar bags. When zero air was
84
stored, in another study, in 5 mil Teflon bags, contamination of 1 ppm C
occurred after 72 hours. The major contaminants of samples- stored in Teflon
have been identified as C2~C» fluorinated hydrocarbons. Pre-irradiation for
12 hours and vacuum heating reduce contamination from Teflon outgassing but
also reduce bag life. Untreated 2 mil Teflon produces no contamination by
Cg-C, fluorinated hydrocarbons but does produce fluorinated aromatics with
time.84
Stainless steel canisters have also been used in collection and storage
of air samples for subsequent GC analysis. While they do not produce outgassing,
canisters must be thoroughly cleaned after storage of each sample to prevent
cross-contamination. Cleaning, however, often results in rust. Contamination
from stainless steel canisters appears to be less severe than from Teflon or
Tedlar bags. Because of the bulk, weight, and need for cleaning, canisters
QC
have not been favored as storage vessels by researchers in the field.
Despite these sample handling and storage problems, gas chromatography
can be an accurate method of analysis of individual hydrocarbons. Data presented
later in this section, for example, were obtained by deploying mobile laboratories
6-18
-------
outfitted with gas chromatographs so that samples could be analyzed soon after
they were obtained. Care was taken in these studies to identify any chromatogram
peak artifacts so that they could be assigned to bag emissions rather than
84
ambient air compounds. Furthermore, analysis of hydrocarbon mixtures in
ambient air must be performed by GC if the contributions of those mixtures and
their individual components to photochemical oxidant formation are to be
determined, and if relative contributions of various souces are to be
QC
evaluated. Thus, for these and other applications in which compositional
analysis is required, GC techniques must be used. Likewise, for short-term
studies, such as may be required to provide data for use in a model or to
validate a model over a short but intensive period of data collection, GC may
not be too expensive and time-consuming but may, in fact, prove to be the
method of choice.
•
6.3 SOURCES AND EMISSIONS OF HYDROCARBONS
6.3.1 General
Data on sources and emissions of hydrocarbons that appeared in the 1970
criteria document were for 1968. The most recent compilation of emissions
data, prepared by EPA, covers 1977 but also includes summaries of data for
1970 through 1977. The data appearing in the 1970 document, as well as the
more recent data, are not specific for hydrocarbons but include emissions of
all volatile organic compounds (VOC). Consequently, most manmade emissions
reported in this section will be referred to as VOC. Generally, emissions of
methane, ethane, methyl chloroform, and Freon 112 are considered to have
73
negligible photochemical reactivity and are not included. For many stationary
fuel combustion sources, however, sufficient information is not available to
permit exclusion of these compounds, and current emission factors that include
6-19
-------
methane and ethane have been used to estimate these emissions. Highway vehicle
emissions, which are reported as nonmethane hydrocarbons, are derived by
87
subtracting methane from total hydrocarbon emissions.
A summary of national emission estimates for VOC for 1970 through 1977 is
presented in Table 6-3, and emissions of VOC by source category for 1977 are
88
presented in Table 6-4. It is clear from Table 6-4 that transportation is a
major source of emissions. Other major sources of VOC emissions are industrial
processes and miscellaneous sources. Within these two categories, chemical
manufacturing (listed as chemicals), industrial organic solvent use, and
miscellaneous organic solvent use (domestic and commercial) together total 9.1
x 10 MT/yr, nearly the total of VOC emissions produced by highway vehicles
(9.9 x 106 MT/yr). ^
A very rough estimate of the percentage of VOC which are actually hydro-
carbons may be obtained from Table 6-4 by assuming that emissions from chemical
manufacturing and from industrial and miscellaneous organic solvent use are
100 percent non-hydrocarbon VOC. While this assumption is not wholly accurate,
it is also true that exceptions to it are more than likely offset by non-hydro-
carbon VOC emissions such as oxygenated or possibly even halogenated HC from
highway vehicles, stationary source fuel combustion, oil and gas production
and marketing, and miscellaneous combustion sources.
Evaporative VOC emissions result primarily from the transportation and
storage of crude oil and volatile petroleum products such as gasoline and
distillate fuel oil, and from the use of organic solvents for surface coating,
degreasing, and other applications. Evaporative emissions from stationary
sources have shown a steady increase over the years because of increased
89
demands for motor vehicle fuels and organic solvents. ,As the result of
6-20
-------
TABLE 6-3. SUMMARY OF NATIONAL ESTIMATES
OF VOLATILE ORGANIC EMISSIONS, 1970-197767
(106 MT/yr)
YEAR
1970
1971
1972
1973
1974
1975
1976
1977
VOC
29.5
29.1
29.6
29.7
28.6
26.9
28.7
28.3
6-21
-------
TABLE 6-4. NATIONAL ESTIMATES OF
VOLATILE ORGANIC EMISSIONS, 1977,
BY SOURCE CATEGORY67
(106 MT/yr)
Source category
Transportation
Highway vehicles
Nonhighway vehicles
Stationary fuel combustion
Electric utilities
Industrial
Residential, commercial,
and institutional
Industrial processes
Chemicals
Petroleum refining
Metals
Mineral products
Oil and gas production
and marketing
Industrial organic
solvent use
Other processes
Solid waste
Miscellaneous
Forest wildfires
Agricultural burning
Coal refuse burning
Structural fires
Miscellaneous organic
solvent use
VOC
11.5
(9.9)
(1.6)
1.5
(0.1)
(1.3)
(0.1)
10.1
(2.7)
(1.1)
(0.1)
(0.1)
(3.1)
(2.7)
(0.3)
0.7
4.6
(0.7)
(0.1)
(O)5
(0)
(3.7)
28.3
of Total
41
36
2
16
100
Zero indicates emissions <50,000 MT/yr.
6-22
-------
industrial growth, industrial process emissions have increased by about 17
percent. This increase has been offset by decreases in emissions from other
categories. Emissions from highway vehicles have decreased as a result of
emission control efforts by auto manufacturers in response to a Federal mandate,
implemented under the Federal Motor Vehicle Control Program, for meeting
requirements of Title II provisions of the Clean Air Act. This decrease of
about 7 percent from 1970 through 1977 occurred in spite of an estimated 30
percent increase in motor vehicle travel during that period (see Section
6.3.3.2). Less solid waste was burned during that time, resulting in fewer
emissions. Emissions from miscellaneous organic solvent use decreased as
water-based emulsified asphalts were introduced in place of paving asphalts
liquefied with petroleum distillates that evaporate when the asphalt cures.
Overall, emissions of VOC decreased slightly from 1970 through 1977.
Emissions of VOC are not uniformly distributed throughout the United
States. Figures 6-1 and 6-2, which are population density and VOC emission
density maps prepared from 1975 data,90'91 are presented to illustrate graphi-
cally the correspondence between population and pollution. The areas with the
greatest population and greatest VOC emission densities are (1) the northeast
corridor extending from Washington, D.C., to Boston; (2) certain industrial
states (Illinois, Indiana, Ohio, Pennsylvania, and New York); and (3) states
along the Gulf Coast. Maps showing VOC emission densities by county rather
than state have better resolution.88 These show that high emission densities
also exist in the Los Angeles and San Diego, California, areas; and in portions
of Texas and the Gulf Coast states.
Total emission inventories for the entire nation give no indication of
the contributions of specific sources to local or regional ambient air
6-23
-------
01
o*
O)
u>
CO
(O
0
I
6-24
-------
h if- if —I! ^^^^^9^^^
If i| (I i $j$;&3%:^^
f ll i i ^S::SxS:!:
01
o
01
a
a*
en
_c
CO
c
i
•s
CO
i
•a
o
0)
O
OJ
(6
6-25
-------
concentrations of hydrocarbons. Local and statewide emission inventories give
approximate indications of the contributions of respective source types, or
categories, to the hydrocarbon loading in ambient air. The development of
accurate and comprehensive emission inventories, however, is complicated by a
number of factors, including shortcomings in emission measurement methodology
(for determination of both composition of emissions and emission rates); lack.
of data on process emissions; inability to quantify manmade fugitive emissions;
difficulties in identifying and quantifying natural emissions; and difficulties
in estimating actual vehicle density, speeds, and miles traveled for mobile
sources. Furthermore, atmospheric chemical reactions involving hydrocarbons
and meteorological factors affecting their dispersion are quite complex, so
that significant differences can exist between emission profiles and ambient
air profiles, as demonstrated by Gordon et al., even if emission inventories
are reasonably comprehensive and accurate. As a result, the contributions of
respective sources to the ambient air concentrations of nonmethane hydrocarbons
are not readily apparent nor easily determined.
A number of researchers have pioneered in the development of numerical
techniques that can be used in conjunction with ambient air measurements to
determine contributions from major source categories. In the case of aerosols,
go
for example, the chemical element balance (CEB) technique of Friedlander,
which is based on material balance principles and on the fact that air pollution
sources of a given type emit chemical elements in a characteristic, reasonably
fixed proportion, can be used to determine the relative contributions of
respective sources to ambient air aerosol mass. In this technique, if the
emission sources in a given region are known, contributions to ambient air
concentrations of particulate matter from those sources can be estimated by
6-26
-------
measuring the elemental concentrations at a given point and then solving a set
of simultaneous linear equations. This technique has been applied fairly
92~95
extensively to the estimation of source contributions to particulate matter.
Similar numerical techniques can be used to calculate estimates of source
contributions to the concentrations of vapor-phase hydrocarbons in ambient
air. In this case, composition profiles, both qualitative and quantitative,
of nonmethane hydrocarbons from major emission sources have been compiled and
compared with ambient air profiles of NMHC by many investigators.
Neligan analyzed air samples in Los Angeles for detailed hydrocarbon
composition in 1962. His comparison of ambient air profiles with the hydrocarbon
composition of auto exhaust led to the identification of other sources of
paraffinic hydrocarbons; namely, natural gas leaks, gasoline evaporation, and
industrial emissions. From their study of Los Angeles air samples, Altshuller
44
and Bellar reported ambient air concentrations of ethane, propane, and
isobutane that were five to ten times greater than the concentrations of these
same compounds in auto exhaust given off during dynamometer studies. A
97
comparison by Stephens and Burleson of C^-Cg aliphatic hydrocarbons in
ambient air in Riverside, California, to NMHC profiles for auto exhaust,
liquid gasoline, gasoline vapor, and natural gas showed that propane probably
originated from a combination of vehicular emissions, gasoline evaporation,
and losses from natural gas and oil fields. High concentrations of ethane
were found and attributed to natural gas sources. The results of that study
also confirmed the finding of Altshuller and Bellar that large amounts of
propane occur in Los Angeles ambient air relative to the negligible amounts
found in auto exhaust. It also confirmed the earlier work of Gordon et al.,
who had demonstrated that significant differences existed between ambient air
hydrocarbon profiles and those of auto exhaust.
47
6-27
-------
98
Lonneman et al. studied auto emissions in the Lincoln Tunnel (New
York City) in order to determine typical hydrocarbon profiles from autos under
actual roadway, as opposed to dynamometer, operation. In this study, acetylene
was used, because of its virtual nonreactivity, to normalize concentrations of
all other nonmethane hydrocarbons. Ratios of ethylene, isobutane, n-butane,
isopentane, and n-pentane to acetylene, as well as other NMHC/acetylene ratios,
were derived for vehicular emissions. These ratios have since been confirmed,
qq
within experimental error, in a tunnel study in Houston. The work of
Kopczynski et al. in St. Louis added to the species and ratios known to be
indicative of automative-related sources. These workers found that o-xylene,
ethylene, and 2-methylpentane, along with carbon monoxide and acetylene, occur
in consistent proportions to other compounds. These compounds are all more
reactive than acetylene, however, and losses will occur from photochemical
reactions. Using the NMHC/acetylene ratio for vehicular emissions, Kopczynski
et al. estimated that less than 50 percent of the paraffinic and aromatic
hydrocarbons in ambient air in St. Louis were attributable to automotive-related
emissions; whereas most of the olefins in evening and early morning samples
were automotive-related.
In the mid-1970s, Mayrsohn and Crabtree developed a procedure, which
they termed "source reconciliation," for estimating the relative contributions
of respective sources to ambient air concentrations of hydrocarbons. Their
procedure is a multivariate regression analysis which employs an algorithm,
the logical and mathematical bases of which are too detailed to present here,
to find normalization constants which correspond to the most accurate source
configuration relative to an atmospheric hydrocarbon profile. The optimum
source configuration results when the error between the normalized source
6-28
-------
component sums and the corresponding profile components is minimized. Source
profiles were developed for major sources of atmospheric NMHC in the Los Angeles
area and atmospheric profiles were determined from compositional analyses on
ambient air samples collected at three sites in the area during the summer of
1973. The average distribution of NMHC according to source was estimated to
be: automotive exhaust, 47 percent; gasoline (including liquid and vapor), 31
percent; commercial natural gas, 8 percent; and geogenic natural gas, 14
percent. If exhaust and gasoline are combined, automotive-related sources
contributed about 78 percent of the C2-C10 NMHC in ambient air in the Los
Angeles area in 1973.
In a subsequent study covering June through September 1974, Mayrsohn
102
et al. applied their source reconciliation procedures to 900 NMHC analyses
of samples obtained at eight sampling sites, seven of which were located on an
east-west line extending from Los Angeles to Palm Springs, a distance of about
.110 km (about 69 miles). The results of that study showed the estimated
source contributions to ambient air NMHC (C2-C1Q) to be: automotive exhaust,
53 percent; gasoline, 12 percent; gasoline vapor, 10 percent; commercial
natural gas, 5 percent; geogenic natural gas, 19 percent; and liquified
petroleum gas, 1 percent. Again, when exhaust, gasoline, and gasoline vapor
are combined, automotive-related sources contributed about 75 percent of the
C2~C10 NMHC 1'n ambl'ent al'r in tne Los Angeles area in 1974. According to
102
Mayrsohn et al. the estimated source distribution obtained by their procedure
differed substantially from the emissions inventory compiled for Los Angeles
County, in which automotive exhaust was reported to account for about 90
percent of the mass of "reactive" hydrocarbons, a percentage that was determined
by assuming a set of source composition and emission rate factors for sources
of photochemically reactive hydrocarbons only.
6-29
-------
From a preliminary source reconciliation study in Houston, Siddiqi and
99
Worley concluded that both automotive and industrial sources are important
contributors to ambient air NMHC in Houston but that in downtown Houston
automotive sources play the more important role. These researchers normalized
concentrations of NMHC with the sum of acetylene plus propane, instead of
98 99
acetylene only, as used by previous investigators, ' because both are
unreactive and the two could not be resolved on the gas chromatographic (GC)
103
column used. Crabtree and Mayrsohn later noted that if the unresolved
acetylene-propane GC peak was due mostly to propane, instead of acetylene as
99
assumed by Siddiqi and Worley, then geogenic natural gas could well be the
greater contributor to the NMHC burden in downtown Houston. Adding to this
interpretation is the fact that Siddiqi and Worley found negligible hydro-
carbons in the Cg-C-,Q range and nearly half of the mass of auto exhaust
98
hydrocarbons is in this range.
104
Seila studied NMHC concentrations, composition, and source contributions
in ambient air samples taken in January 1978 in Jones State Forest, about 61
km (38 miles) north of Houston. He found that automotive sources accounted
for 35 percent and non-automotive sources for 54 percent of the NMHC burden in
Jones State Forest samples. Of the non-automotive NMHC contribution, 41
percent was found to originate from sources in Houston and 59 percent from
sources north of Houston. Most of the non-automotive NMHC contribution was
attributed to refinery and natural gas emissions. Natural (vegetative)
hydrocarbon emissions in these samples, all of which were taken within the
forest, were reported to account for 2 percent of the NMHC loading.
104
6-30
-------
6.3.2 Natural Sources and Emissions
The estimates for VOC presented in the previous section included some
natural VOC emissions, those given off by forest wildfires. Other natural
sources of hydrocarbons (not VOC) have already been described in Section 5.
-1 Q
The 1970 document cited Koyama's estimate of natural methane emissions of
Q Q
2.73 x 10 metric tons/yr (3 x 10 tons/yr). That estimate did not include
methane emissions generated in swamps and tropical areas, and thus can be
considered a conservative estimate. Robinson and Robbins added production
from swamps and humid tropical areas to Koyama's estimate for an estimated
natural production of methane of 1.45 x 10 metric tons/yr (1.6 x 109 tons/yr).
Assuming a methane emission rate per unit of land area that is half the world
average—a reasonable assumption, given our temperate climate—natural methane
emissions in the United States would be about 4.5 x 107 metric tons/yr (5 x
7 73
10 tons/year). Though its role in photochemical smog systems is under
renewed investigation, ' methane is still considered to play little, if
any, role in the photochemical formation of oxidants and other smog components.
In addition to methane, other hydrocarbons are emitted from natural
-i no
sources. A recent report by West German researchers raises the possibility
that natural benthic sources may be responsible for the occurrence of higher-
molecular-weight, vapor-phase n-alkanes (> Cg) in "clean" ground-level air.
They sampled air of three types in order to sort out the influence of manmade
sources: (1) air influenced by passage over the European continent (surface
winds and easterly trajectories); (2) clean marine air (no known recent contact
with land masses as determined from trajectories); and (3) marine air influenced
by passage over land areas, but predominantly influenced by passage over
bodies of sea water. They also sampled sea.water and surface films from sea
6-31
-------
water. They estimated the global oceanic strength of > Cg vapor-phase n-alkanes
Q Q -
to be 2 to 20 Tg/year (2 x 10 to 20 x 10 metric tons/yr). They concluded
that it is unclear whether the bulk of ri-alkanes observed in marine air originates
from the sea surface or from the continents, but noted that the calculated
residence times for the n-alkanes, which appear to be controlled by their
reactions with OH", are sufficiently short that only small quantities of
higher continental ]i-alkanes should reach the open ocean. They did not rule
out manmade fossil fuels, from spills and seepage, as the ultimate source, but
108
did cite evidence for the production of ji-alkanes by marine microorganisms.
Vegetation is a well-known source of natural hydrocarbon emissions. It
has been shown unequivocally that monoterpenes, CIQ compounds with one to
three double bonds, are emitted from evergreen foliage and that isoprene, a C,-
henviterpene, a diolefin, is emitted from deciduous foliage. '
19
Rasmussen and Went estimated a global production rate for volatile
Q C
organics from vegetation of about 3.98 x 10 metric tons/yr (438 x 10 tons/yr).
In more recent work,, done since publication of the 1970 criteria document,
Zimmerman estimated net worldwide emissions of all natural hydrocarbons,
8 8 109
methane plus vegetational, at 7.54 x 10 metric tons/yr (8.3 x 10 tons/yr).
Zimmerman derived an estimate of isoprene plus terpene emissions of 6.36 x 10
metric tons/yr (70 x 10 tons/yr) in the United States. The complexity of
the factors involved, coupled with some methodological uncertainties, makes
this estimate somewhat less than firm, but it appears to be the best estimate
available for emissions from vegetation. The distribution of terpene and
isoprene emissions is not as uniform in the United States as the methane
emissions, based on background concentrations of methane measured at various
sites. Zimmerman's emission inventory indicates that 43 percent of the total
6-32
-------
natural HC emissions occur during the summer months (June, July, and August);
that 45 percent of the annual emissions occur in the southern United States,
115
and that isoprene constitutes 34 percent of this portion.
The role of natural hydrocarbon emissions in photochemical oxidant
pollution is controversial, the controversy arising in large measure from
apparent disparities between observed emission rates and aerometric data.
Discussion of the issues is beyond the scope of this document, but detailed
discussions are available in the open literature, including three recent EPA
publications.7'73'107'116'118 The role of natural hydrocarbons will again be
addressed in detail in the next revision of Air Quality Criteria for Ozone and
Other Photochemical Oxidants. This section merely presents a summary of the
emission sources and rates for natural, vegetational hydrocarbons reported by
various researchers. Aerometric measurements of natural hydrocarbons are
presented in Section 6.4.
Using essentially the same techniques as those he used in making estimates
119
of nationwide emissions of natural hydrocarbons, Zimmerman conducted a
study of natural hydrocarbons in Houston in 1978. He measured both emission
rates and ambient air concentrations of isoprene and the monoterpenes at a
large number of sites in the Houston area typical of various kinds of vegetation
and typical of varying degrees of influence from manmade emissions. A summary
of emission rates and ambient air measurements from this study is presented as
1 1Q
Table 6-5. Table 6-6 presents a comparison of emission rate data for
119
Houston and the Tampa-St. Petersburg area, as reported by Zimmerman.
6-33
-------
TABLE 6-5. COMPARISON OF SELECTED HOUSTON EMISSION
RATE AND AMBIENT AIR DATA119
No. samples/ Emission rate, Ambient concentration,
Sample type no. sites jjg/g/hr HS/m3
Live oak 12/2
TNMHC
Isoprene
Slash pine 6/1
TNMHC
crPinene
p-Pinene
d-Limonene
Slash pine 6/1
TNMHC
crPinene
p-Pinene
Short leaf pine 1/1
TNMHC
crPinene
p-Pinene
d-Limonene
Common fig 1/1
TNMHC
crPinene
p-Pinene
Red bay 1/1
TNMHC
crPinene
p-Pinene
29.3
29.3
20
5
4
7
0.5
0.2
0.2
20
5
4
7
1
0.1
0.1
0.4
0.1
0.3
79.8
2.4
27.1
1.5
0
0
43 ,
0
0
27
1.5
0
0
259
1.3
2.4
91
4
0
Distance from
branch, in.
12
6
6
6
6
6
a
.Emission rate in ng/g of leaf biomass/hr.
Distance at which grab samples were taken for ambient air measurements.
6-34
-------
TABLE 6-6. COMPARISON OF HOUSTON AND TAMPA/ST. PETERSBURG
TNMHC EMISSION RATES119
Type
Oaks (daytime),
Mg/g/hr
Pines, ijg/g/hr
NC-NI,3 |jg/g/hr
Soil , |jg/m2/hr
H20, ug/m2/hrb
N
7
7
11
8
12
Houston
X
26
5
0.2
195
48
sd
23
7
0.7
134
38
Tampa/St. Petersburg
N X sd
47
70
143
101
152
23
9
4
324
231
23
11
6
177
208
NC-NI includes non-conifer trees and shrubs for which less than 50% of
their TNMHC emission rate is isoprene (Non-Conifer, Non-Isoprene).
3Surface water measurements. Houston samples were taken from fresh, swamp,
and ship channel surface waters.
6-35
-------
The methods by which samples are taken for determination of natural HC
emission rates are the subject of much of the current controversy over the
role of natural HC in the atmosphere. Again, discussion of the methodologic
assumptions and uncertainties is too detailed to be presented in this document.
Basically, however, the Zimmerman method involves enclosing a branch of a tree
or portion of other vegetation in a flexible plastic enclosure chamber in
order to capture the vapor-phase organics emitted by the vegetation. Other
methods have been used by other investigators to determine natural HC emission
120
rates. Tingey, for example, has used a glass chamber that encloses the
entire plant. In the case of trees, seedlings are used.
Results obtained by Tingey for four isoprene producers and one terpene
120
producer are shown in Table 6-7. A comparison of emission rates for mono-
terpenes from various plant species is presented in Table 6-8, also taken from
Tingey.
Though technical details and arguments concerning the relative abundance
of natural HC sources and emissions are not being presented in this document,
it should be pointed out that a number of scientific techniques, facts, and
questions are bound up in the present controversy over the contribution of
natural HC to ambient air levels of total NMHC and to ozone formation. Un-
certainties and room for error inhere in methods for determining emission
rates; in methods for determining leaf biomass over areas even as large as a
few square kilometers, not to mention entire areas of the country or the
121
country as a whole; in ambient air measurements—both sampling and analytical
techniques—of natural HC; in models used to reconcile emissions and ambient
air measurements; and in methods for establishing accurate inventories of
manmade emissions of HC, such that the component of total NMHC in ambient air
6-36
-------
that is attributable to manmade emissions can be known with a verifiable
degree of certainty. For further discussion of the various factors that are
pertinent to the abundance and role of natural hydrocarbons, the reader is
referred to a recent summary by Dimitriades of issues regarding natural
70
hydrocarbons and to a recent EPA publication.
TABLE 6-7. BIOGENIC HC EMISSION RATES AT CONSTANT ENVIRONMENTAL CONDITIONS3'120
Emission rate
Plant species
ug C/dm2/hrb
C/g/hrc
Isoprene
American sycamore
Live oak
Soybean
Bush bean
Total monoterpenes
Slash pine
49.4
41.2
0.007
0.010
34.2
6.4
a • 2
Leaf temperature was 30°C and light intensity was 800 uE/m /sec.
Generally calculated as emissions per unit of leaf surface of
deciduous foilage; calculated as emissions per g of biomass for
conifer needles.
6-37
-------
TABLE 6-8. COMPARISON OF MONOTERPENE EMISSIONS
FROM VARIOUS PLANT SPECIES120
Emission rate
Plant species jjg/g/hr ng/m2/min
Slash pine
Total monoterpenes 6.4
orPinene 3.2 35.0
Slash pine
Total monoterpenes 2.6
Loblobby pine
crPinene - 46.1
Longleaf pine
Total monoterpenes 5.6
Cryptomeria (Japan cedar)
Total monoterpenes 3.0
Sal via (sage)
Total monoterpenes - 18.1
Reference3
Ti ngey et al . ,
1980
Zimmerman, 1979
Arnts et al . , 1978
Zimmerman, 1979
Kamiyama et al . ,
1978
Tyson et al . , 1974
aComplete citations given in Reference.
6-38
-------
6.3.3 Manmade Sources and Emissions
Several comprehensive descriptions of manmade sources and emissions of
in i?? i?3
VOC are available. '' EPA has published and periodically updates detailed,
comprehensive data on VOC emission rates in Compilation £f Emission Factors
(AP-42), but because that document presents emission rates for source categories
and subcategories in terms of total VOC rather than species of VOC or HC, the
information given here was derived from two other EPA documents122'123 and from
Vapor-Phase Organic Pollutants.10 published by the National Academy of Sciences.
6.3.3.1 Stationary Source Emissions—The material presented in this section
was taken directly from or derived from Volatile Organic Compound (VOC) Species
123
Data Manual. a report prepared for EPA in 1978. The manual summarizes and
documents the available information on species of VOC emitted from stationary
and mobile sources in a format usable by those preparing emission inventories
for photochemical models. Most of the information in the manual was obtained
from material prepared for the California Air Resources Board or from reports
supplied by EPA. Individual references will not be cited in the summary
presented here, but are available in Reference 123.
The information summarized here was chosen to represent the kinds of
emissions and sources characteristic of the major stationary source categories
presented in Table 6-4: (1) stationary fuel combustion, (2) industrial
processes, (3) solid waste disposal, and (4) miscellaneous sources.
Sources of stationary fuel combustion include both external combustion
sources and internal combustion engines. The major fuels burned by external
combustion sources are coal, oil, and natural gas. External combustion units
include utility, industrial, commercial, and institutional boilers; commercial
6-39
-------
and domestic combustion units; process heaters; furnaces; kilns; etc. If
units are properly operated and maintained, emissions from these sources are
small. For example, a 1000 megawatt power plant is estimated to emit about 91
to 182 metric tons per year (100 to 200 tons per year) of volatile organic
124
compounds. This low concentration of organics in source effluents makes
their analysis difficult and readily accounts for the fact that newer infor-
122 123
mation, ' based on newer and more sophisticated technologies, may
disagree with emission rates and species presented in the NAS document,10
which were based on 1968 data. Table 6-9 presents a profile of the VOC
emitted from external combustion sources, categorized by fuel burned.
Internal combustion engines are used in applications similar to those
employing external combustion units. Internal combustion engines may be
turbines or reciprocating engines, and may be fueled by the usual fossil fuels
such as diesel, oil, or natural gas. They are' used in electrial power
generation, in gas pipeline pump and compressor drives, and in various in-
dustrial processes. The majority of gas turbines are used in electrical
generation for continuous, peaking, or stand-by power. Table 6-10 tabulates
emissions of VOC from internal combustion engines used in electrical power
generation and in industrial processes. Note the differing emission profiles
for larger versus smaller (30 hp) reciprocating engines. Smaller engines also
have higher emission rates (emissions per unit of fuel burned) than larger
engines, though the profiles presented here do not include comparisons of
emission rates.
6-40
-------
TABLE 6-9. EMISSIONS OF VOLATILE ORGANIC COMPOUNDS FROM
STATIONARY EXTERNAL COMBUSTION SOURCES123
(wt %)
Compound
emitted
Methane
Formaldehyde
Ethane
Ethyl ene
Acetylene
Acetone
Propane
Propylene
ri-Butane
Isobutane
ji-Pentane
Isomers of pentahe
ji-Hexane
Isomers of hexane
rrHeptane
Isomers of heptane
n-Octane
Isomers of octane
Cyclohexane
Benzene
Toluene
Residual Distillate
oil oil
11.0
42.0 48.7
.
-
_
28.0
1.2
-
14.0 12.2
4.1
4.7
5.5
5.0 10.8
5.2
0.3
2.6
.. -
4.7
. -..-..-.
-
-
Fuel
Natural Refinery
gas oil
56.0 7.6
8.0 7.6
20.9
-
-
- -
4.0 18.9
17.5
9.0 23.1
4.4
6.0
9.0
-
1.0
-
-
-
-
1.0
4.0
2.0
Coke oven
gas
82.8
• " -
2.5
11.7
0.8
-
-
0.3
. -
•
-
-
-
-
-
-
-
.
1.9
-
6-41
-------
TABLE 6-10. EMISSIONS OF VOLATILE ORGANIC COMPOUNDS FROM
STATIONARY SOURCE INTERNAL COMBUSTION ENGINES74
(wt %)
Compound
emitted
Methane
Formal dehyde
Ethane
Ethyl ene
Acetyl ene
Propane
Propyl ene
rv-Butane
Isobutane
Butene
1,3-Butadiene
Benzene
Electrical generation
Natural Diesel,
gas, turbine recip.
70.0 11.6
30.0
2.8
28.7
11. 3
->
17.3
-
-
13.4
7.0
7.9
Dist.
oil, recip.
11.6
-
2.8
28.7
11.3
-
17.3
-
-
13.4
7.0
7.9
Industrial
Natural
gas, recip.
76.0
1.0
10.0
1.0
-
10, 0
-
1.0
1.0
. -
- '
-
Natural gas,
recip. (30 hp)
93.5
-
2.6
2.5
0.9
0.4
0.1
-
-
-
-
-
6-42
-------
The manufacture of chemicals and chemical-based products contributes
about 27 percent of the total VOC emissions attributable to industrial processes
(Table 6-4). Emissions from the manufacture of one chemical and two chemical-
based products are shown in Table 6-11. As demonstrated in this table and
from comparison of this table with the preceding tables, emissions from chemical
manufacturing show wide qualitative variations. No one organic compound or
group of compounds dominates emissions.
In the case of varnish manufacture, cooking and thinning operations are
the major sources of VOC emissions. Incineration and catalytic oxidation have
been used to control VOC from varnish-making; 99 percent control is possible
with incineration. Uncontrolled emissions of VOC are about 185 kg/MT of
product, but controlled emissions are about 1.85 kg/MT.
Phthalic anhydride is produced by the vapor-phase oxidation of naph-
thalene or o-xylene with excess air or catalysis. Reactor and condenser
effluents are the main sources of VOC emissions, but other, minor sources
are present. The emissions profile in Table 6-11 for phthalic anhydride
production is a composite of all sources.
Printing inks consist of a fine dispersion of pigments or dyes in a
vehicle that consists of a drying oil with or without resins and added driers
or thinners. The preparation of the vehicle, by heating, is the largest
source of VOC from ink manufacturing. Composite controlled emissions from
ink manufacturing, reduced by about 90 percent of uncontrolled by means of
scrubbers or condensers followed by an afterburner, are about 21 kg/MT of
product.
Though the disposal of solid wastes contributes only 2 to 3 percent of
all VOC emissions in the United States, Table 6-12, a profile of VOC from
6-43
-------
TABLE 6-11. EMISSIONS OF VOLATILE ORGANIC COMPOUNDS
FROM MANUFACTURE OF SELECTED CHEMICALS/PRODUCTS
(wt %)
Compound
emitted
Varnish
Phthalic
anhydride
Printing
inks
Methane
Methyl alcohol
Ethane
Ethyl alcohol
Acetone 38.7
Glycol ether 3.0
Acetylene
Propane
Isopropyl alcohol
Methyl ethyl ketone 41.6
Propylehe
jv-Butane
Isobutane
n-Butyl alcohol
n-Butyl acetate
Methyl isobutyl ketone 16.7
Isomers of pentane
rrHexane
Cyclohexane
Ethyl benzene
Isomers of diethyl benzene
Mineral spirits (paraffins)
80.0
0.4
8.6
4.4
11.7
3.1
36.1
0.4
22.6
21.4
11.3
5.0
2.5
5.5
38.0
5.0
3.0
4.0
5.0
3.0
3.5
25.5
Composite emissions from chemical wastes.
6-44
-------
TABLE 6-12. EMISSIONS OF VOLATILE ORGANIC COMPOUNDS
FROM SANITARY LANDFILLS
Compound
emitted
Wt
Methane
Ethane
Perchloroethylene
Propane
ri-Butane
Isobutane
ri-Pentane
Cyclopentane
Terpenes
Toluene
Isomers of xylene
98.6
0.1
0.3
0.1
0.2
0.1
0.1
0.2
0.1
0.1
0.1
' 6-45
-------
sanitary landfills, is included here as an indication of the kinds of
emissions that are given off from solid wastes. The use of sanitary land-
fills is a long-accepted method for the disposal of residential and in-
dustrial wastes, most of which consist of refuse, domestic garbage, and inert
construction materials. Waste is generally layered, compacted, and covered
by thin layers of silt. Consequently, anaerobiosis occurs, resulting in the
generation of appreciable amounts of methane-rich gas. Methane and carbon
dioxide together constitute about 99 percent of gaseous emissions from
landfills.
Next to emissions from highway vehicles, miscellaneous use of organic
solvents contributes the largest percentage of total VOC emissions in the
United States, 13 percent. Three examples of area source emissions of VOC in
this category are presented in Tables 6-13 through 6-15. Domestic solvents
are those found in products used around the house, garage, or yard. Emissions
of VOC from domestic chemicals or chemical-based products result from the
vaporization of low-boiling-point solvents in the product, the quantity and
species depending on the product. Composite estimates of the percentage by
weight of organic species emitted from domestic solvents are given in Table
6-13; they were calculated on the basis of annual sales of products containing
these solvents. Organic emissions from domestic chemical use are estimated at
1.0 metric tons/1000 people/yr (1.1 tons/1000 people/yr).
Pesticides are commonly available in the form of liquids, aerosols, or
powders and are applied by spraying, or dusting, or both. The VOC emitted
from pesticide use arise from vaporization of the pesticide itself or from the
solvent(s) used as the pesticide vehicle. Table 6-14 presents a profile of
VOC emissions resulting from the domestic and commercial use of pesticides.
6-46
-------
TABLE 6-13. EMISSIONS OF VOLATILE ORGANIC COMPOUNDS
FROM GENERAL USE OF DOMESTIC SOLVENTS a
Compound
Formaldehyde
Ethyl alcohol
Acetone
Glycol ether
Propylene glycol
Isopropyl alcohol
Isobutane
n- Butyl acetate
Naphtha
Wt %
0.6
36.9
1.4
8.3
,3.2
38.5
5.3
1.3
4.5
Estimated at 1.0 metric tons/1000 people/yr (1.1 tons/1000 people/yr). Many
common domestic products contain solvents: furniture polish, shoe polish,
shaving soap, perfumes, cosmetics, shampoo, hair spray, hand lotion, rubbing
alcohol, nail polish remover, etc.
6-47
-------
TABLE 6-14. EMISSIONS OF VOLATILE ORGANIC COMPOUNDS FROM DOMESTIC
AND COMMERCIAL USE OF PESTICIDES3'123
Compound
Wt %
Methylene bromide 10.0
Propane 1.8
n-Butane 4.4
Isobutane 1.4
n-Pentane 3.2
Isomers of pentane 3.1
jV-Hexane 3.7
Isomers of hexane 8.1
C7 cycloparaffins 15.4
Cg cycloparaffins 1.6
Mineral spirits (paraffins) 15.0
Benzene 12.3
Toluene 5.0
Isomers of xylene 15.0
Estimated at 8.2 metric tons/100,000 people/yr (9 tons/100,000 people/yr).
6-48
-------
Total VOC emissions from pesticide use have been estimated at 8.2 metric
tons/1000,000 people/yr (9 tons/100,000 people/yr).
Table 6-15 presents a profile of VOC emissions resulting from the use of
architectural coatings, which are those paints and other coatings applied to
stationary surfaces such as pavements and curbs, and structures and their
appurtenances. Industrial coatings are excluded from this table. When paints
and other coatings are applied, the solvents in the coating materials must
evaporate so that the coating can form a film or barrier. The evaporation of
these solvents, along with that of the solvents used to thin coatings or to
clean up, generate emissions that represent a significant area source of VOC
emissions. Emissions of VOC from the use of paints and other coatings in
Southern California were estimated to be 3.2 metric tons/1000 people (3.5
tons/1000 people) in 1976. Because solvent emissions are controlled in
California (Rule 442, formerly Rule 66), this estimate is probably low
relative to the rest of the country.
Finally, profiles are given for VOC emissions given off by forest fires
and from the open burning of agricultural and landscape wastes such as "spent
field crops, weeds, prunings, etc. (Table 6-16). Ground-level open burning is
affected by many variables, including wind, ambient temperature, composition
and moisture content of the wastes, and compactness of the burning refuse.
Likewise, emissions differ according to burning pattern, i.e., head fires or
back fires. Actual emission rates are difficult to calculate because of the
important influence of these variables and none were given for the profiles
shown in Table 6-16. In a separate study on emissions from agricultural
125
burning, Darley determined speciation and emission factors for hydrocarbons
and some aldehydes from the burning of residues of eight selected field and
orchard crops: barley, corn, rice, sorghum, and wheat; and almond, grape, and
peach (Table 6-17).
6-49
-------
TABLE 6-15. EMISSIONS OF VOLATILE ORGANIC COMPOUNDS FROM THE
a ~
USE OF ARCHITECTURAL SURFACE COATINGS*'•
Compound
Wt %
Acetone
Methyl alcohol
Ethyl alcohol
Ethylene glycol
2-Ethoxyethyl acetate
Di methylformami de
Isopropyl alcohol
Methyl ethyl ketone
Propylene glycol
n-Butyl alcohol
rrButyl acetate
Isobutyl alcohol
Isobutyl acetate
Methyl n-butyl ketone
Methyl isobutyl ketone
Isobutyl isobutyrate
n-Hexane
Cyclohexane
Toluene
Isomers of xylene
Ethylbenzene
3.2
3.9
0.6
0.6
1.3
0.5
16,4
5.6
0.8
1.6
2.5
0.6
1.5
0.7
0.6
6.1
20.7
20.7
5.2
2.6
4.3
Estimated at 3.2 metric tons/1000 people/yr (3.5 tons/1000 people/yr),
but since estimates were derived from Southern California, which is
subject to control of solvent vapors, this estimate is probably low.
6-50
-------
TABLE 6-16. EMISSIONS OF VOLATILE ORGANIC COMPOUNDS FROM FOREST FIRES
AND FROM OPEN BURNING OF AGRICULTURAL/LANDSCAPE WASTES
(wt %)
Compound
emitted
Methane
Ethane
Ethyl ene
Acetyl ene
Methyl acetylene
Propane
Propylene
ji-Butane
Isobutane
Butene
Isomers of butene
1,3-Butadiene
3-Methyl -1-butene
n-Pentane
Isomers. of pentane
1-Pentene
Isomers of pentene
n-Hexane
n-Heptane
ji-Octane
Unidentified HCs
Forest fires
9.8
10.5
19.1
8.4
0.4
0.4
3.9
0.2
0.1
-
0.8
0.5
0.2
-
0.2
V
-
-
-
-
44.6
Agri cul tural /I andscape
burning
-
-
19.4
1.9
-
1.9
-
1.9
1.9
5.9
-
' -
.
1.9
11.8
11.8
13.9
13.9
13.8
-
6-51
-------
TABLE 6-17. HYDROCARBON EMISSIONS FROM AGRICULTURAL
1
BURNING OF FIELD AND ORCHARD CROP RESIDUES
Type of
Crop f i re
Barley Head
Back
Corn Head
Back
Rice Head
Back
Sorghum Head
Back
Wheat Head
Back
Almond Pile
Grape Pile
Peach Pile
Average values
Field crops
Orchard crops
Avg. emissions,
Ib THC/ton fuel
18.8
12.5
5.7
7.6
2.8
2.8
2.1
3.5
18.1
9.4
4.5
4.9
3.2
8.3
4.2
Methane
49.3
53.3
55.9
52. 9
54.2
50.1
62.9
55.1
47.7
49.5
60.5
54.0
61.7
53.1
58.7
HC in air
Other
saturates
14.1
5.7
10.1
9.4
6.5
6.5
6.9
7.3
7.S
5.2
8.8
9.6
6.7
8.0
8.4
sample, %
Olefins
33.0
34.2
29.6
31.4
33.0
35.8
25.9
31.6
37.0
35.5
24.9
27.9
22.0
\
32.7
24.9
Acetylene
3.7
6.8
4.5
6.3
6.4
6.9
4.3
6.0
7.6
9.8
5.8
8.5
9.6
6.2
24.9
6-52
-------
Combustion and sampling took place in an outdoor, instrumented burning tower
(an inverted funnel with a stack) that simulated open burning but channeled
the combustion products so that representative samples of gases could be
taken. Measurements for hydrocarbons in isokinetically sampled air were made
with gas chromatography coupled with mass spectrometry. Hydrocarbons > Cg
were first absorbed in a charcoal canister, then desorbed and chromatographed.
Aldehydes were determined as the respective 2,4-dinitrophenylhydrazones.
Field crops were burned as head and back fires and orchard crops were burned
in piles.
Most of the samples contained compounds up through cis-2-butene (£ C«);
occasionally compounds through 1-pentene (< C5) were found. Analytical
problems were encountered in the determination of aromatics (> Cg), but
benzene and toluene were detected from field crop combustion, by both
laboratories doing the analyses, in both head and back fires. Benzene
emissions were reported to be highest from barley head fires (0.364 kg/metric
ton of fuel) [0.728 Ib/ton of fuel] and lowest from sorghum head fires (0.014
kg/metric ton of fuel) [0.028 Ib/ton of fuel]. Toluene emissions were highest
from barley head fires [0.297 Ib/ton of fuel] and lowest from sorghum (head
fires, 0.005-0.01 kg/metric ton of fuel; back fires, 0.0015 kg/metric ton of
fuel) [head fires, 0.001-0.020 Ib/ton of fuel; back fires, 0.003 Ib/ton of
fuel].
Of the hydrocarbons < Cg, methane constituted more than half of the
emissions from both field and orchard crops. 01efins constituted 32.7 and
24.9 percent of hydrocarbon emissions from field and orchard crops,
respectively. Ethylene was the most abundant plefin, constituting 69.1
percent of olefinic emissions from both field and orchard crops.
125
6-53
-------
This survey of VOC species and their emissions demonstrates the diversity
of compounds arising from various types of sources. It also serves to
corroborate the estimate made earlier in this section that as much as 30
percent or more of VOC emissions may be non-hydrocarbons.
6.3.3.2 Mobile Source Emissions—Nationwide estimates of VOC from
88
transportation sources are shown for 1970 through 1977 in Table 6-18. As
that table shows, emissions from highway vehicles decreased nearly 7 percent
QQ
from 1970 through 1977.• This decrease was achieved in the face of an
increase during that period of about 30 percent in vehicle miles traveled.
The decrease resulted mainly from the application by auto manufacturers of a
variety of control devices. Some decrease was undoubtedly effected, as well,
through improved fuel economy such that the increase in vehicle miles traveled
was not accompanied by an equivalent increase in fuel consumption and, hence,
TABLE 6-18. NATIONWIDE ESTIMATES OF VOC
EMISSIONS FROM TRANSPORTATION SOURCES, 1970 THROUGH 1977
,88
Year
1970
1971
1972
1973
1974
1975
1976
1977
Total
transportation
VOC emissions,
10° MT/yr
12.2
12.2
12.5
12.3 •
11.5
11.3
11.6
11.5
Highway VOC,
10° MT/yr
10.6
10.6
10.9
10.7
10.0
9.8
10.0
9.9
% of
total
transpor-
tati on
emissions
86.9
86.9
87.2
87.0
87.0
86.7
86.2
86.1
Nonhighway
fivoc,
10° MT/yr
1.6
1.6
1.6
1.6
1.5
1.5
1.6
1.6
% of
total
transpor-
tation
emissions
13.1
13.1
12.8
13.0
13.0
13.3
13.8
13.9
6-54
-------
in exhaust emissions. As an example of the decreases in hydrocarbon emissions
accomplished by vehicle manufacturers through the use of pollution control
devices, exhaust hydrocarbon emission factors for gasoline-powered light-duty
vehicles are shown in Table 6-19 for 1968 through 1980 model-year vehicles, by
1 pc
calendar year of operation. Composite crankcase and evaporative hydrocarbon
i y-j
emission factors for the same period are given in Table 6-20. Crankcase
and evaporative emissions are, besides exhaust emissions, the main source of
hydrocarbons from vehicles. Since crankcase emissions from post-1963 vehicles
v
are negligible, emission factors are determined for crankcase and evaporative
127
emissions together. The main sources of evaporative emissions, as noted
earlier in Section 5.3, are the fuel tank and the carburetor system. Crankcase
and evaporative emissions may increase over the lifetime of the vehicle, but
properly maintained control devices for these emissions are assumed not to
127
deteriorate with age. Representative pollution control devices developed
by the auto industry to achieve hydrocarbon emission reductions are shown in
101
Table 6-21; while these are emission control systems used by General Motors,
they indicate the basic kinds of systems employed by all U.S. auto manufacturers
to meet Federally mandated emission standards.
A discussion of the calculation of emission estimates such as those
presented in Table 6-18 is beyond the scope of this report. It should be
noted, however, that the data base used and assumptions made in making those
calculations render them complex. For example, the estimate of total VOC
emissions attributable to transportation sources (Table 6-18) is derived from
mobile source emission factors and vehicle miles traveled, both of which, in
turn, are based on additional data such as factors for the deterioration with
age of pollution control devices, the "mix" of kind and age of vehicles in
6-55
-------
TABLE 6-19. HYDROCARBON EXHAUST EMISSION FACTORS
FOR LIGHT-DUTY, GASOLINE-POWERED VEHICLES FOR7
ALL AREAS EXCEPT CALIFORNIA AND HIGH-ALTITUDE^7
a,b
(g/mi)
Mnrlnl
Year 1970
1968 4.3
1969 3.5
1970 2.7
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
July, calendar
1971 1972
5.0 5.
4.3 5.
3.5 4.
2.7 3.
2.
7
0
3
5
7
1973
6.
5.
5.
4.
3.
2.
3
7
0
3
5
7
1974
6.9
6.3
5.7
5.0
4.3
3.5
2.7
year of operation
1975
7.4
6.9
6.3
5.7
5.0
4.3
3.5
1.3
1976
7.9
7.4
6.9
6.3
5.7
5.0
4.3
1.6
1.3
1977
8.3
7.9
7.4
6.9
6.3
5.7
5.0
1.9
1.6
1.3
1978
8.7
8.3
7.9
7.4
6.9
6.3
5.7
2.2
1.9
1.6
1.3
1979
9.1
8.7
8.3
7.9
7.4
6.9
6.3
2.5
2.2
1.9
1.6
1.3
1980
9.4
9.1
8.7
8.3
7.9
7.4
6.9
2.8
2.5
2.2
1.9
1.6
0.3
Emission factors for vehicles through model year 1975 and through calendar year
1975 are based on actual surveillance tests of in-use vehicles. Post 1975 calendar
year factors for all model-year vehicles are projected. Deterioration factors
used are: pre-1968, 0.58; 1968-1974,0.53; 1975-1979, 0.23; and > 1980, 0.23 ~
all in g/mi per 10,000 miles of travel. For complete information on the calculation
of emission factors, see reference 127.
To convert g/mi to g/km divide g/mi by 1.609.
TABLE 6-20. COMPOSITE CRANKCASE AND EVAPORATIVE
EMISSION FACTORS FOR LIGHT-DUTY, GASOLINE-POWERED
VEHICLES FOR ALL AREAS EXCEPI7CALIFORNIA AND
HIGH-ALTITUDE1^7>a
(g/mi)
Model year
1968-1970
1971-1977
1978-1979
1980
Emissions
2.53 .
1.76
0.60
0.15
To convert g/mi to g/km divide g/mi by 1.609.
6-56
-------
CO
CM
T-i
to
O
tO r-H
SL CO
LU CO
r- 1— i
£§
co
z z
0 <*
z
•a o
0
•p
e
•r- 10
•r—
S- C
fO S-
co o
•I-
fO
o
e
cu
in
in
P_
o
i %
c
o
o
m
s_
i — CO O
o in Q.
5- tO fO
-P U >
o c o
o to ai
.« O CO
in w
c .c to
O Cn (J
•i- 3 ^.
in o c
in s- to
•i- .C S-
E -P U
co
S_ +J
0 T- tO
3Z to -E:
0
c "to c s-
O > -i- 10
o
r-i
CO
to
en
r-l
r-l
tfcQ
€7)
r-l
c
o
-P
to
l_
•r—
-P
c:
CO
at
in
to
t_>
c:
to
(j
O)
,r-
•p /">
•r- O
in >
o a.
Qu *~s
r-5
0 T3
•P C.
fO
E
CO r—
-P CO
in 3
in
T3
•P CO
in c:
3 S-
tO 3
<-• f^
X C
CO 3
CO M-
C O
En £=
C 0 .
CO T- CO
-p -a
0 tO v-
-P -O X
C T- O
•r- X C
0 0
•r- CO
to -p c
o o
in E .a
205-
fO E- fO
s- a. o
a
CM
to
^O
O*t
r-l
C
o
"43
u
10
CO
S-
C T-
o to
•i— in
-p 1—
O 3
co o_
"•— 3
c ».
1 — 1
s- 'r— EO53S-
^3 UtOS— CCCOfOfO r— fO C -P r™ -P fO S— . TD CO
CO _Q'i— -I— -i—S- fO fOlO fOtOElOCOS
o^M- mtn s-^^ uc *3 "co-PU r—
• #1 -r" o s~ -P -P in 3 to S— to *— o • in S— co *—
S_v^/ OCC4->-P-P tO-PC35S-CT. CO i— -P U 10
•r- C-PCOCOr—XC ^C 3-r-C inCCCOT-3
lOi — OCOEE3T-T- Oi — OO-r- 3 T- •!— i — .C-C
inCO-r-S-COCOinE COS-COC C2-P
r— 3 4J 3 > > CO 4- 3JCS-S- S ~ .3 *""" jj ..
Q. N S- S- S- CO W D)J2 -P tO -£Z C CO
>*•»— fO O OL. -P 3 O5 S- E C £! in O5 " -P CD S-
S-JDEUEEC04-C OO2'r-S- >, CO-r-33
O T- •r-'r-Z'^,T- O-S-fOCO W -P S- 2 •— +J
-O4Jit- S-E tO't-S-CM— W3 -I— tO
o; co Q. o / — "O3 ' oisin'O'as-
I — i 4^ O CO "^J" • tO fO in 5- CD O tO in CO CO
v_ ' in 2 os-S- E LU-ceoxino.
•r- tn r— t- O) OC O T- W •!- X S- «r- tO E •
OE/— S3--TDOOS- O-lO-r-CT CUQ.E COW
4JO)rH r— C S- O-r— 10 C • 20. W4-JC
> < s~^ Ot04-'C a)>-"CUCO XCO O W 4-> O
COW CMS- COT3 3 i-C O-PfO'i-UC-r-
2> it".* ^^^ .I ? • *i Q *n~ d M~~ C__3 ^3 •!•• ^5 r<^ ^3 -^2 ft3 ^3 V)
•r-C.. CWU-PtO 3I4-J054J i— +J T-W
^fO-C'-Or— tO W O)C W 3 -P «r- -P -P T-
tO-PC35COUOO)S-D) i — XJS-COW i — UW WWE
C33C S-C^IC OS-3 S- OS-3C033CO
s- r— o T- co 4-> •!-•!- T- s-o^jco s-T-fOs-toja
cor— s-a5S-cEr— E 4->ws-coa. +Jux:coi:E X
i— Q.4-> CO -P O -P O +3 OfOU2> OS-C02LUOZ
<: oo
CO ^ LO
CO r-l CO
vo r**^ r*^
o^ en o^
rH tH rH
^D O CSJ
1^3 r^» ^^
en en en
r-l r-l r-l
E
CO
-P _
•w c
^*> *^
W . • T-
-p
c re
O W i —
.r- r— ' 3
4-> O U
W S- S-
3 -P "-
jn c: u
E 0 CO
o o , J-
u
co w
T3 > fO
CO •"— D3
r— -P
r— tO ^-> /^N
0 /~s S- W Qi
S- CO 0 30
+j o a. to LU
CO tO. J=^
0*~< > X
O LU LU
CO «5f- LO
6-57
-------
-p -a .—
05 C tO
=J fO -P
03 O
.coo:
X 1C
D)-
£= S-
O =3
CM
O -P
Cd W CD
-a o
c cu o
tO •" O -r-
•P X <*-
O 10 CU 4-
= ,>?, "
<»- "io j
a)
C O T3
O CU
•i- S- N •
•P CU «i- O
to > ~a CM
T3 O T-. HZ
*x 10 o -a .
o to c c
to cu to o
10 QLJQ «r-
CU CMP
+J 10 O O U
w
o
a)
o
o to o o co
S- D)U -P S-
CU
O
CM
CO
cu
.p
«o
•p
in
tn
r~.
en
co
cr>
cu
u
cu
•P
to
to
cu
in
i--
CT>
00
CD
to
o
cu
u
O)
•p
10
10
cu
o
(J
to
•p
to
a>
to
-p
to
cu
-P O
u s-
I CU
S- 4J
cu s-
-P CU
3 >
'a. c
e o
o u
o
10
6-58
-------
use, vehicle registration, fuel consumption, etc.128 Thus, the stated decrease
in transportation-related VOC emissions of about 7 percent from 1970 through
1977 may be greater if EPA has overestimated the deterioration with age of
pollution control devices, and thus the VOC emissions from controlled cars, as
has been claimed by the Motor Vehicle Manufacturers Association. For detailed
discussion on that point, the reader is referred to the proceedings of an
emission inventory/factor workshop sponsored by EPA in 1977,129>13° and to
documentation related to requests by auto manufacturers for waiver of the 1981 .
Federal emission standard for carbon monoxide from mobile sources.131'132
Though VOC emissions from highway vehicles decreased about 7 percent from 1970
through 1977 (Section 6.2.1), motor vehicles remain the largest single category
of contributors to atmospheric HC burdens, and are the most influential widespread
source of VOC emissions in general and of hydrocarbon emissions in particular.
Stationary sources have a major impact in most .urban areas and in some urban
areas, such as Houston and Chicago, they may have greater impact than mobile
133
sources. Nevertheless, HC emissions from light-duty gasoline vehicles come
the closest to being ubiquitous and consistent in composition of any hydrocarbon
mixtures from any major sources. This section is limited mainly to a discussion
of hydrocarbon and oxygenated hydrocarbon emissions from diesel- and gasoline-
powered automobiles, since they are the most important mobile sources of
QO
gas-phase hydrocarbons. According to EPA data, highway vehicles account for
86 to 87 percent of all transportation-related VOC emissions; and aircraft,
railroads, marine vessels, and other non-highway vehicles together account for
13 to 14 percent.
6-59
-------
The 1976 NAS document on vapor-phase organic pollutants states that
gasoline-powered vehicles account for about 99 percent of all vehicular hydro-
carbon emissions. While this estimate is probably high now because of a
growing trend toward use of diesel engines, it is still true that most cars
and light-duty trucks are gasoline-powered.
As noted in Section 5.3, there are potentially four sources of volatile
hydrocarbons and other organic compounds from gasoline-powered vehicles:
crankcase blowby emissions, carburetor evaporative emissions, fuel tank
evaporative emissions, and exhaust emissions. Diesel vehicles emit volatile
hydrocarbons and other organic compounds almost exclusively from the exhaust;
evaporative emissions are practically nonexistent because of low fuel
volatility.134'135
Sources of vehicle emissions have come under increasing control since
1963 when the introduction of positive crankcase ventilation systems virtually
eliminated blowby emissions from gasoline-powered vehicles. The introduction
of absorption-regeneration systems in 1970, which reduce evaporative emissions,
has resulted in a decline of more than 80 percent since that time in atmospheric
emissions from carburetor and fuel tank evaporation.136 Likewise, exhaust HC
emissions have been reduced by about 90 percent in new vehicles since 1968 by
means of a variety of engine modifications, including, most recently (about
1975), catalytic oxidation systems. The actual reduction in HC emissions
effected by these control systems is less than the percentages cited above
73
because of the inevitable deterioration of these control devices. Never-
theless, with the implementation of these emission controls in highway vehicles,
emissions from mobile sources have been reduced in spite of continuing increases
6-60
-------
in motor vehicle travel. As cars are replaced, both current control systems
and more advanced controls anticipated for the future are expected to reduce
HC emissions from motor vehicles further, at least up through the mid-to-late
73
1980s. However, it is projected that the automobile will continue to be a
significant source of air pollution, partly because of the expected continued
T37
concentration of automobiles in congested urban areas.
For late-model cars equipped with catalytic oxidation systems, evaporative
emissions account for nearly one-third to one-half of THC emissions; and
138
exhaust emissions account for the rest. Deterioration of exhaust control
systems results in a change in those proportions with time, however, so that
exhaust emissions from a car 4 to 5 years of age can be expected at that time
to contribute the major fraction of vehicle hydrocarbon emissions.
Because of the contribution of evaporative emissions, it is worthwhile to
look at the composition of gasoline and of gasoline vapor as well as the
composition of exhaust emissions. Gasoline normally contains more than 200
hydrocarbon compounds, mainly in the C4 to CIQ range. Alkanes and aromatics
generally constitute the largest fraction, but olefins and naphthenes are also
present. Final gasoline composition is the result of the mixture of about six
blending stocks. The highest octane blending components in gasoline are
aromatics and branched-chain paraffins. Different gasoline blends are required
for different climatic regions of the United States. A summary of gasoline
composition data, for a composite blend that is representative of U.S. con-
139
sumption, is given in Table 6-22. The average composition of gasoline
vapor, determined from weighted averages of gasoline blending stocks and vapor
-1 OQ
pressures of respective compounds, is shown in Table 6-23.
6-61
-------
TABLE 6-22. SUMMARY DATA ON GASOLINE
COMPOSITION, REPORTED AS WEIGHT PERCENT
139
Hydrocarbon
Paraffins
Isobutane (iC.)
n-butane (nC»)
Isopentane ^lCs^
n-pentane ("^5)
Dimethyl butanes (Cg;
Methyl pentanes (Cg)
n-hexane
Dimethyl pentanes
Methyl hexanes
Tri methyl pentanes
n- heptane
Dimethyl hexanes
Methyl ethyl pentanes
Dimethyl hexanes
Tri methyl hexanes
n-octane
Naphthenes
Methyl cycl opentane
Cycl ohexane
Methyl cycl ohexane
Other saturates
Olefins
Methyl butene
Pentene
Methyl pentene
Other olefins
Aromatics
Benzene
Toluene
Ethyl benzene
Xylenes
Propyl benzene
Methyl ethyl benzenes
Trimethyl benzene
Other aromatics
Total
Saturates
Olefins
Aromatics
Morris and
Dishart3
0.7
4.8
8..5
3.4
2.0
4.6
2.0
2.4
5.9
11.1
1.2
1.3
1.4
1.3
1.4
1.3
2.0
0.9
1.2
7.5
2.5
0.8
0.8
7.5
0.9
6.5
1.3
8.8
1.4 •
2.8
7.1
5.2
62
11
27
Myers
et al.a
7.0
9.3
4.5
1.4
6.2
3.3
1.4
6.3
8.5
2.0
2.9
1.8
1.0
2.5
0.8
1.5
2.5
1.5
5.9
1.3
5.9
1.5
1.7
Maynard and
Sanders
4.3
10.2
5.8
2.0
6.0
1.5
1.9
3.2
9.4
2.0
2.0
0.4
.
0.4
0.6
0.2
0.3
1.3
1.6
1.1
3.5
0.8
12.2
1.7
7.3
0.3
1.6
2.3
2.3
62
9
29
aCited in Reference 139.
6-62
-------
TABLE 6-23. AVERAGE GASOLINE VAPOR COMPOSITION
139
Compound
Mean volume
percent
Alkanes
Propane
n-butane
Isobutane
Isopentane
n-Pentane
Cyclopentane
Dimethyl butane
Methylpentanes
n-Hexane
Methylcyclopentane
D1methylpentanes
Trimethylpentanes
Alkenes
Isobutylene
Methylbutenes
Aromatics
Benzene
Toluene
Xylene
0.8
38.
5.
22.9
7.0
;7
.7
0.
0.
3.6
1.5
1.3
1.1
0.5
1.1
2.8
0.7
1.8
0.5
6-63
-------
Evaporative emissions from automobiles, plus evaporative losses from
gasoline marketing and storage, result in the presence of gasoline vapor
components in all urban atmospheres. Evaporative emissions consist of the
lighter components of gasoline, primarily C. and C5 hydrocarbons. The exact
composition of the emissions depends on the composition of the gasoline, on
the temperature it reaches (in carburetor, fuel tank, storage tank; during
delivery), and the degree of "weathering," that is, prior evaporation.
Analyses from several studies indicate that the light paraffins and light
olefins constitute about 70 percent of carburetor emissions and about 90
percent of fuel-tank emissions. Isopentane and n-butane were by far the
predominant hydrocarbons, making up about 50 percent of the total. Despite
variations in gasoline composition, these two paraffins account for the major
portion of all gasoline evaporation losses.
Exhaust emissions from gasoline-fueled vehicles typically contain fuel
components and low-molecular-weight hydrocarbons that are not present in the
fuel. These low-molecular-weight hydrocarbons include methane, ethane, ethylene,
acetylene, propylene, the C^ olefins, and sometimes propadiene and methyl-
acetylene. The fuel components include hydrocarbons with carbon numbers
higher than butane (C,). The predominant hydrocarbons in auto exhaust, as
reported in three separate studies, appear in Table 6-24. As this table
shows, the low-molecular-weight hydrocarbons—methane, ethylene, acetylene,
and propylene—are high on each list. Toluene and isopentane seem to be the
major fuel components.
Extensive studies have been conducted on the effect of gasoline composition
on exhaust composition. Dishart and coworkers concluded that ethylene is
formed from saturates and olefins—propylene and butene primarily from saturates,
6-64
-------
TABLE 6-24. PREDOMINANT HYDROCARBONS
IN EXHAUST EMISSIONS FROM GASOLINE-FUELED AUTOS
10
Hydrocarbon
Methane .
Ethyl ene ,
Acetylene.
Propylene
n-Butane
Isopentane
Toluene.
Benzene
n-Pentane
m- + p_-Xvlene
1-Butene13
Ethane0
2-Methylpentane
n-Hexane
Isooctane
All others
Fraction
62-Car
survey
16.7
14.5
14.1
6.3
5.3
3.7
3.1
2.4
2.5
1.9
1.8
1.8
1.5
1.2
1.0
22.2
of total
15-Fuel
study140
18
17
12
7
4
4
5
—
~3C
•m _
—
30
HC, vol %
Engine- variable
study3'146
13.8
19.0
7.8
9.1
2.3
2.4
7.9
—
*'5c
6.0C
2.3
__
—
26.9
.Variables were air:fuel ratio and spark timing.
Combustion products.
Includes isobutylene.
6-65
-------
and diolefins primarily from olefins—and that additional amounts of toluene,
benzene, and xylenes are formed from higher aromatics and additional 2-methyl
-2-butene from higher saturates.140'142 Doelling et al.143 concluded that
gasoline composition had no effect on the total concentration of hydrocarbons
in exhaust, but that the percentages of paraffins, olefins, and aromatics in
exhaust were correlated with fuel composition. Neligan et al.144 concluded
that fuel olefinic exhaust emission is proportional to the olefin content of
the fuel, but that the nature and concentration of the light, cracked products
are independent of the olefin content of the fuel.
The use of catalytic converters has had a pronounced effect not only on
the amount of hydrocarbons emitted from cars but also on the kinds. In general,
oxidation catalysts have resulted in an increase in the percentage of paraffins,
especially methane; and a decrease in percentage of olefins and acetylene.
Typically, exhaust from a catalyst-equipped automobile contains about 62
percent paraffins, 17 percent aromatics, 18 percent olefins, and 3 percent
acetylene; as compared with 40, 24, 26, and 11 percent, respectively, for
automobiles without exhaust emission converters. The methane levels generally
range from about 10 to 30 percent.147'148 A study by Jackson produced the
emission data for non-catalyst-equipped and catalyst-equipped cars presented
149
in Table 6-25. These data corroborate the decrease in olefins and acetylene
cited above. The decrease in acetylene seen with the catalytic converter has
implications for the use of acetylene as an indicator of auto exhaust emissions
in source reconciliation studies (Section 6.3.1), since the mole percent of
acetylene attributable to auto exhaust is declining. Furthermore, auto exhaust
is not the only source of acetylene in urban atmospheres, as demonstrated by
6-66
-------
TABLE 6-25. SUMMARY OF EXHAUST EMISSION DATA
FOR NONCATALYST AND CATALYST-EQUIPPED
GASOLINE-FUELED CARS BY MODEL YEAR149
Noncatalyst-equipped
to. of Calif. Cars
to. of Fed. Cars
Mi 1 eage
(km)
tonreactive hydrocarbons
Acetyl ene
Methane
Benzene
Ethane
Propane
Nonreactive total
teactive hydrocarbons
Ethyl ene ,
Toluene
Xylenes
Propylene
Trimethylpentanes
nrButane
i-Pentane
Butenes
Methyl pentanes
n-Pentane
Ethyl benzene
i- Butane
Other hydrocarbons
Reactive total
lydrocarbon classes
Paraf f i ns
Olefins
Acetyl ene
Aromatics
1970
1
3,600
5,792
7.4
6.2
6.4
1.2
0.1
21.4
10.1
11.7
9.7
8.0
2.1
1.6
4.4
6.0
2.0
0.6
1.7
0.8
19.8
78.6
23.9
32.0
7.4
36.7
"lolar-based reactivity scales
NOp formation rate
Altshuller
Methane exclusion
Dimitriades
0.0542
0.0571
0.0167
0.1612
1972
1973
1974
Catalyst-equipped
1975
389
42,105 19,802 9,162 16,075
67,747 31,861 14,742 25,865
Carbon, % of total hydrocarbon
10.7
6.3
3.8
0.9
0.0
21.7
15.5
11.4
2.3
9.6
3.0
7.0
5.2
2.3
1.2
2.1
0.6
2.3
15.6
78.3
34.5
31.1
10.7
23.7
7.8
4.9
3.4
1.0
0.1
17.1
12.8
9.1
4.1
6.3
5.5
2.0
4.5
2.9
2.9
1.4
1.1
0.8
29.7
82.9
36.8
25.7
7.8
29.7
Hydrocarbon
0.0511
0.0556
0.0190
0.1773
0.0467
0.0499
0.0166
0.1548
7.7
4.7
4.9
1.0
0.1
18.4
12.4
7.5
7.7
6.1
4.0
4.7
3.6
4.0
2.0
1.3
.... 1.7
0.6
25.6
81.6
33.2
25.8
7.7
33.3
reactivity
0.0485
0.0518
0.0165
0.1560
2.5
11.3
3.7
2.6
0.3
20.4
7.4
6.5
5.9
3.1
5.4
5.3
5.4
2.5
3.0
2.1
1.4
1.0
29.6
79.6
52.9
16.4
2.5
28.2
1976
1977
4
2
3,218 9,859
5,178 15,863
3.6
9.8
4.8
3.0
0.3
21.6
7.2
7.2
7.3
3.9
5.0
3.9
4.5
4.2
3.2
2.1
1.9
0.8
27.2
78.4
46.4
17.0
3.6
33.0
0.3
28.0
1.6
4.5
0.3
34.7
7.8
7.7
2 0
f— • \J
1 5
-l- * ij
15.6
7 2
/ * if.
5.9
0.4
1.5
0.3
0.6
0.3
14.0
65.3
74.3
9.9
0 3
V/ * *J
15.6
per gram
0.0366
0.0162
0.0132
0.1210
0.0380
0.0892
0.0139
0.1232
0.0222
0.0234
0.0109
0.1078
6-67
-------
TABLE 6-25 (continued).
Noncatalyst-equipped
No. of Calif. Cars
No. of Fed. Cars
Mileage
(km)
1970
1
3,600
5,792
1972
1 •
42,105
67,747
1973
3
19,802
31,861
1974
1
8
9,162
14,742
Catalyst-equipped
1975 1976 1977
5-4
92-
16,075 3,218 9,859
25,865 5,178 15,863
Carbon, % of total hydrocarbon
Carbon-based reactivity scales
Methane exclusion 0.9380 0.9370 0.9510 0.9530 0.8870 0.9020 0.7200
Cal. Air Res. Bd. 0.4808 0.4469 0.4160 0.4215 0.3300 0.3443 0.2456
Relative reactivity per gram (nonconverter car average = 100.0)
Molar-based reactivity scales
N0« formation rate 111.3
Altshuller 109.8
Methane exclusion -100.0
Dimitriades 101.3
Carbon-based reactivity scales
Methane exclusion 98.7
Cal. Air Res. Bd. 112.8
104.9
106.9
113.8
112.5
98.6
104.8
95.9
96.0
99.4
98.2
100.1
97.6
99.5
99.6
98.3
99.0
100.3
98.9
75.2
69.6
79.0
76.8
93.4
77.4
78.0
75.4
83.2
78.2
94.3
30.7
Exhaust emissions, g/mi
.-a
45.6
45.0
65.
58.
,3
.4
75.8
57.6
Methane
Nonmethane HC
Total HC
0.21 0.06 0.20 0.09 0.08 0.09 0.08
3.22 1.30 3.75 1.87 0.63 0.86 0.19
3.43 1.36 3.95 1.96 0.71 0.95 0.27
To convert g/mi to g/km, divide g/mi by 1.609.
6-68
-------
Trijonis and Arledge, who found acetylene from major sources in Los Angeles in
the following amounts: 2 mole percent from petroleum refinery operations; 5
mole percent from stationary source fuel combustion; 8 mole percent from waste
burning and other fires; and 11 mole percent from gasoline-powered, light-duty
vehicles. To assume that auto exhaust is the only source of acetylene
could mean the introduction of significant error depending on the quantity and
profile of emissions being studied. For Los Angeles, the error introduced has
been calculated to be less than 3 percent, however.
The decrease in ethylene in exhaust emissions from catalyst-equipped
vehicles is worthy of particular mention, because ethylene is known to be a
phytotoxicant (Section 6.7). Data from the Jackson study149 show that ethylene
averaged 12.5 percent of the total hydrocarbons emitted from cars without
oxidation catalysts (model year 1970 through 1974) and only 7.4 percent of
total hydrocarbons emitted from catalyst-equipped cars (1975 through 1977
model years).
Jackson also evaluated the emissions in the same study by six scales used
to indicate potential photochemical reactivity of hydrocarbons. All six
showed that catalyst-equipped cars produced exhaust hydrocarbon mixtures that
were less reactive than those from non-catalyst cars. The reductions in
photochemical reactivity per gram ranged from about 10 to 35 percent. Thus,
catalytic converters have not only reduced total hydrocarbon emissions but
have also reduced the oxidant-forming potential of those emissions.149
Exhaust gases from gasoline-fueled autos contain organic compounds besides
hydrocarbons, such as aldehydes, ketones, alcohols, ethers, esters, acids, and
6-69
-------
phenols. The concentration in exhaust of these oxygenates is about one-tenth
that of the total hydrocarbon concentration. Aldehydes are generally believed
to be the most important class of oxygenates. Formaldehyde is by far the
predominant aldehyde, constituting about 60 to 70 percent of the total (by
volume); acetaldehyde is next, at about 10 percent; and propionaldehyde,
acrolein (acrylic aldehyde), benzaldehyde, and the tolualdehydes are all found
in appreciable amounts. As might be expected, the compositon of the gasoline
burned influences the compositionuof the aldehydes formed.
Gasoline composition and catalytic converters are but two of many factors
that influence the composition of exhaust gases from cars. Other factors
include fuel composition, driving patterns, configuration of emission controls,
ambient temperature and humidity, and individual automobile, characteristics
such as make, model year, age, engine size, air:fuel ratio, and spark timing.
In addition, fuel additives can alter the exhaust hydrocarbon emissions. Of
particular interest is the effect of lead additives. The use of tetraethyl
lead (TEL) as a gasoline antiknock agent tends to increase exhaust hydrocarbon
emissions directly and also indirectly, by promoting the buildup of engine
deposits. According to one study, TEL directly increases hydrocarbon emissions
by about 5 percent; the indirect effect, from accumulation of deposits, results
in an additional 7 percent.152 The composition of exhaust hydrocarbons is not
affected by"TEL.153
Just as the use of lead- additives affects exhaust hydrocarbon emissions,
so does their non-use; again, indirectly. To achieve the same octane and
antiknock properties found in leaded gasoline, refineries must process more of
the lower octane blending stocks such as' light, straight-run gasoline; hydro-
cracked gasoline; and thermally cracked gasoline. These processes primarily
6-70
-------
involve conversion of naphthenes to aromatics and normal paraffins to iso-
139
paraffins. In actual practice, gasolines of all grades generally have
lower octane ratings now than in previous years. While aromatics in all
gasolines, leaded as well as unleaded, have gradually increased over a period
of-years, the aromatic content of unleaded gasoline has not increased
commensurately with the removal of lead; that is, the aromatic content is not
high enough to attain the combined octane and antiknock properties of leaded
154
gasoline. Nevertheless, the increases in aromatic content that have occurred,
along with possible future increases, are expected to affect the hydrocarbon
composition of exhaust emissions. Benzene is considered the most toxic component
of the aromatic fraction of gasoline vapor (Section 6.6). Toluene, xylenes,
and other alkylated benzene derivatives are more toxic than benzene in acute
exposures but less toxic in chronic exposures; they are also much less volatile
139
than benzene. On the other hand, benzene has low reactivity in atmospheric
photochemical processes, whereas primary and secondary monoalkyl benzenes are
highly reactive and tertiary monoalkyl benzenes and tolualdehyde are moderately
reactive. Benzaldehyde, like benzene, is of low reactivity.7'73
An investigation by Nebel of the benzene content of exhaust emissions
from five catalyst-equipped cars (1975 Federal Test Procedure, model year 1975
cars) showed that benzene constituted an average of 2.86 percent of the total
HC exhaust emissions over odometer readings of 80 to 32,180 kilometers (50
to 20,000 miles). Total HC exhaust emission increased with accumulated mileage
from 0.25 g/km to about 0.56 g/km at 32,180 kilometers (0.4 g/mi to about 0.9
g/mi at 20,000 miles). Benzene exhaust emissions, as a percentage of total HC
exhaust emissions, increased irregularly with the accumulation of mileage.
138
Black et al. have recently examined both exhaust and evaporative emissions
6-71
-------
for total hydrocarbons and for individual species of hydrocarbons, including
benzene and toluene. Their results confirm other data in the literature which
show that oxidation catalysts not only reduce the mass of exhaust hydrocarbons
emitted but also remove a greater percentage of unsaturated olefinic, aromatic,
and acetylenic emissions than of saturated paraffinic emissions. While evapora-
138
tive control devices clearly reduced total hydrocarbon emissions, Black et al.
found that the effect of these devices on the composition of evaporative HC
emissions was well-defined. Their study showed that both the benzene and
total aromatic content of fuel influenced the benzene emission rate. Fuel
benzene had a greater impact on evaporative emissions of benzene—since all
evaporative benzene emissions come from the fuel—than on benzene in the
exhaust, at least some of which is derived from the combustion process. In
fact, dealkylation of higher aromatics appeared to be a significant source of
exhaust benzene emissions when the vehicle was operated on a fuel-rich mixture.
Data from the study showing differences in THC, benzene, and toluene emissions
with differences in fuel composition and vapor pressure and emission control
139
configuration are given in Table 6-26, derived from Black et al. Increased
gasoline aromaticity also affects exhaust emission rates for aromatic aldehydes
-I Cf-
from non-catalyst-controlled vehicles, as shown in Table 6-27. Increased
aromaticity likewise produces increased amounts of phenols in exhaust emissions
in uncontrolled cars and, with as much as 46 percent aromatic fuels, in
catalyst-equipped cars as well. The effects of increased aromaticity on
ambient air concentrations of aromatics appear to be negligible to date, as
discussed in Section 6.4.
6-72
-------
en
rH
CO CO
Z —1
O Of LU
I-H =3 =
CO O U_
CO LL.
£ = i
^SS
LU U.
I— a
«C LU Z
CO Z «£
LU LU
ex =3^-,
ir^i
LU
si"
LU
• »Q
LU LU O
I-H LU
t^S
^ Z r^
cc LU en
OCOrH
-rH
^ ^-^
. —ICO
to
3
m
S
•CJ
Sr^
ai
f— O
£*>
Ǥg
§•-
c o
a>
caac
O"E
JE^,
-o
a> 2:
m1"
31^
t— O
•2«
-0
ai 3:
S1"
NM-
C 0
ai
CQd«
•*'r-
O E
S^
-CJ
ss
0)
i— O
•2*
-o
1-
c o
,S^
si
1— OJ
«
VI
at
at
N
-p c
c ai
at ca
*»
8«
i— U>
ai u
F
Aromat
JQ r~*
O>>
30£
U_*— '
re
c in
Or-
•t- O
in s-
in +J
Si
O)
1—
U
•1—
J=
O)
>
rH rH f»rHCn ID to to CM CM CM CO
PO CO *i~ ^" PO O O"* GO PO CO f^ rH
rH rH rH rH rH i-H
cn CM CM o en in «? co cn oocn
r-l in ID O CO CM U3 CM rH «*• r-- O
o m rH ro to CM cn o
co i * i «* CM i co o r^ i cn
• i • i . i . . . i
en in to 1-4 i-t CM o CM
r» co co co CM OCMCOP* r— oo o
i-HrH CO 00 CM r».COOi-H rH CM rH
CM CM CM
coco oo CM ID inr*.coco CM rH in
O CO rH CO rH O in i-H O CM^O
co co co «*• ID en to CM
rH CM in O i-H CM O CO
i-i O O O OO O O
-Hi +1 i +1 +1 I +1 +1 +1 i +1
r— in rH o r» in rH cn
in co to CM * r^ co co
in CM to o oo o c\j
rH CM 00 CO O CO * r— rH Ol Cn CO
in in o oo ID rH o r-» * in co CM
CM rH rH rH rH i-t rH
00 O th cn CO O> ID «3- to rHOOO
«o r*. o CM to CM to co CM mr»cM
§ 3 a S 33 S rH
0 O O O OO O O
+ij +ii +i +i j +i +i +i ! +i
CO CO CO CO Ol r^ CO rH
o r-- co o oo CM «3- in
^ CM CM rH O i-5 O d
COCO IrtrHO COrHinO in rH O
rH in rH f^ CM O r~ rH CM rH l-» CM
en o> «* to h- «s- <* «s- r~ ^- to r—
in co co to in r» CM co in co to in
CM CM «* «3" CM CM CO «S- CM «4- «»• CM
COCO 00 CO CO "3- •* CO OO CO ro
CM CM cn en CM oo co cn CM en cn CM
rH rH rH rH i-H
l-H rH rH rH
O.O COOQCJ
H- 1 1— 1 H-4 I-H
I-H
-U I-H
ai i
i— O) >,j:
£C i- +> u re -o i
ai in t. c s- a •
J= 3 (DO O r-
o ^ ES: u-— i
CO r^ CO CT*
«o r» r» r»
s s s a
a)
• in
*— s a)
in f
- C 4->
ce o c
*-H *r- a>
*«. re
aTZ °-
do c
LU in -i-
ai
-•o c
> 01
^8 £
.'5 ..»
j. 01 ai «
ai -a c at
4-> ai s-
in < — NI 3
c c- ai in
xi ai
u c i-
O •DO.
^ (J O)
c -as-
ms- -o o
+-> o re a.
--"" ^5
0) rH +J
3 CM i-~O
H- 1 S-r-
to ai
S a. "^
•5.S " "
«-i2 i >
o co ce
+» CO <— »
ai ai •• =
(.in T3 • c
3>_- CO) O
JQ 01 C -i-
J- +J i—OI +->
re u> j2 NI ID
u >, c c
i— t- a> T-
o) re a> _a E
i— 4-> E (-
O) 18 ETD 0)
•i- in *o a)
tn TJ -o ca
c: -o re
II ro QJ a)
•o j= >
H- 1 ** re 4^ -r-
I-H OS 01 1- •!->
M3 -E* £
iT . ""= o.
in ce «|| re
£>3 §rH L5
2 - ei u
re > cu o
0. Q.T3
CO)
ae . . II 01 c
O S- r- •!-
LU o co .a ui
•!-> 3
» ai •" s- o
> s- a> a) 31
CJ 3 = E
Q. J3 0) E TJ
(- IM 3 Ol
•• re c in •—
s- u co re
O> J2T3 01
in c TJ T3 =
•i— re a) re ^^
c -a ai
re ^ -or— a>
u c re c-
re • 3
it +> JT t. *o
c +-> re a)
re i— -^ r- o
4-> O) 23 O
3 O) S.
r— *- «*'fl> o. en
-i :•• i 2
II i a co
~-5 «.. a, ^
I-H re ..-o c
• - c ai in E
cj at i— J3 o>
CL. 4-> J3 U>
al = no evaporative controls,
catalyst; IV = one canis
A = regular, unleaded summer
C = regular, unleaded winter
cComposite evaporative emissior
To convert g/mi to g/km dividt
6-73
-------
TABLE 6-27. INCREASE IN EXHAUST EMISSION RATES OF AROMATIC ALDEHYDES FOR
NON-CATALYST GASOLINE CARS WITH INCREASE IN FUEL AROMATICITY100
Fuel
Unleaded premium
Leaded premium
Leaded regular
Aromatics,
mole %
46.6
30.8
27.3
Total aldehydes,
ppm
65
72
69
Aromati c
al dehydes ,
ppm
13.6
6.1
5.8
Precipitated by the energy crisis, an additional development that will
affect the concentration and composition of hydrocarbon emissions from mobile
sources is the possible use of alternative liquid fuels such as gasoline-alcohol
or gasoline-ether mixtures. Comparison of vehicle emissions data from different
studies is difficult because of the variety of test conditions, measurement
techniques, fuels, engine and other modifications, and emission control system
configurations used in the respective studies. Nevertheless, certain qualitative
trends in emissions from alternate fuels are by now discernible from data in
the literature.
-I CO
Work reported in 1975 by Brinkman et al. indicated that adding methanol
to gasoline—without carburetor modifications—decreased CO emissions; decreased
volume-based fuel economy; decreased driveability and performance; and did or
did not affect the mass of HC and NO emissions, depending on the air-fuel
J\
ratio, car-fuel combinations, and other factors. They found also that the
research octane number (RON) increased almost linearly with the addition of
methanol but that the motor octane number (MON) did not. They further found
6-74
-------
that RON and MON of methanol-gasoline blends (hereafter referred to as "methagas")
are not predictive of the road octane number of the blends. A study by Bernhardt
159
and Lee showed that organic emissions from a 100 percent methanol-fueled
car contain only a small percentage of hydrocarbons, mainly methane, ethane,
and ethylene. Emissions of methanol and total aldehydes, however, using the
same engine with the same compression ratio, are significantly higher with
methanol than with gasoline. Total aldehydes in the exhaust are about twice
as high with methanol as with gasoline. The total aldehydes can be reduced
substantially, however, by increasing the compression ratio or by adding water
to the pure ethanol. The overall emission rates for hydrocarbons, as determined
by hot FID corrected by a factor obtained from GC analysis, were as shown in
Table 6-28. (Using the Federal emission test cycle, the vehicle tested
emitted 2.79 g/km (4.5 g/mi) from gasoline and 1.55 g/km (2.5 g/mi) from pure
methanol. The vehicle would not have met the 1975 Federal emission standard
for HC of 0.93 g/km (1.5 g/mi).159
TABLE 6-28. TOTAL HYDROCARBON EMISSION RATES (ECE TEST CYCLE) FOR ,,Q
CAR FUELED WITH GASOLINE, METHANOL, OR A 15% METHANOL-85% GASOLINE MIXTURE159
Fuel
Gasoline
Methanol
15% methanol-
85% gasoline
Test
Cold start
Hot start
Cold start
Hot start
Cold start
Hot start
HC emissions,
g/test
9.8
7.9
12.0
5.7
8.3
6.5
HC emission, % of HC
emissions from gasoline
100
100
122
72
85
82
6-75
-------
Asked to grant a waiver permitting the use of 10 percent ethanol in
gasoline, EPA in 1978 studied exhaust and evaporative emissions from 11
passenger cars tested by the Federal Test Procedure (FTP). Two gasoline and
three gasohol (ethanol-gasoline) fuels were used in four cars equipped with
three-way conversion (TWC) catalysts and in seven equipped with oxidation
catalysts. Evaporative emissions were determined by the SHED procedure
("Sealed Housing for Evaporative Determination") and were also measured by gas
chroraatography. The researchers concluded from their data that combustion of
a 10 percent ethanol-90 percent gasoline mixture: (1) increased evaporative
HC emissions by about 50 percent; (2) decreased exhaust HC emissions by about
9 percent; (3) increased aggregate HC emissions (exhaust plus evaporative);
(4) decreased fuel economy (carbon balance method) by about 2 percent; (5)
increased NO emissions by about 7 percent; and (6) decreased CO emissions by
yv
about 35 percent.
Three-way conversion catalyst systems (TWC), as employed in the EPA tests
cited above, provide for control of HC, CO, and NO and generally require the
use of a closed-loop feedback control system for the air/fuel metering system
in order to maintain a stoichiometric air/fuel ratio. When a fuel such as
ethanol is added to gasoline, the stoichiometric air/fuel ratio is changed.
Conventional air/fuel metering systems do not compensate for this change. A
further study of HC emissions (FTP) from a TWC-equipped car (Volvo) fueled by
ethanol-gasoline or methanol-gasoline blends resulted in the exhaust emissions
shown in Table 6-29.161
Ifi?
Lang and Black, in a recent EPA study, found similar results in tests
of two catalyst-equipped passenger cars fueled with five different fuel
blends. The FTP cycles and SHED procedures were used to determine exhaust
6-76
-------
TABLE 6-29. EXHAUST EMISSIONS FROM VEHICLE EQUIPPED WITH
THREE-WAY CONVERSION CATALYST AND FUELED WITH GASOLINE,
ETHANOL-GASOLINE, OR METHANOL-GASOLINEIbl
Fuel9
No catalyst, 100% gasoline
TWC
100% gasoline
10% EtOH- 905 gas
20% EtOH-80% gas
30% EtOH-70% gas
40% EtOH- 60% gas
50% EtOH- 50% gas
10% MeOH-90% gas
20% MeOH-80% gas
Exhaust
HC
0.93
0.14
0.31
0.24
0.11
0.14
0.22
0.16
0.15
emissions,
CO
13.68
3.19
3.90
2.77
2.61
0.78
0.90
3.20
3.20
/ -b
g/irn
NOX
.2.47
0.11
0.11
0.14
0.07
1.77
1.55
0.09
0.05
aEtOH = ethanol; gas = gasoline; MeOH = methanol.
To convert g/mi to g/km, divide g/mi by 1.609.
6-77
-------
(tailpipe) and evaporative HC emissions. In addition, tailpipe and
evaporative emissions were collected in Tedlar bags and analyzed for THC and
ethanol by gas chromatography. The two cars tested, a 1977 Ford Mustang II
and a 1979 Ford LTD II, differed somewhat in their emission control con-
figurations. Both were equipped with EGR, catalyst, and PCV. The LTD had an
AIR injection system and two carbon canisters for vapor control, one each for
the fuel tank and carburetor. The Mustang had one canister for fuel tank
vapor control. In these tests, the addition of ethanol to the base fuel
produced for both cars: (1) a decrease in exhaust THC emissions; (2) a
decrease in CO emissions; (3) an increase in NO emissions; and (4) a sub-
/\
stantial increase in evaporative emissions. Little change in fuel economy was
observed with either vehicle. Lang and Black noted that increases in fuel
vapor pressure when ethanol is added contributes to differences in evaporative
emissions, but that those changes do not account for all the increase observed.
One gasohol (fuel 5) was blended to match the vapor pressure of the base fuel
(fuel 3) and increases in evaporative emissions were still observed— They
concluded that the vapor control canisters may selectively adsorb and retain
ethanol because it is a more polar compound than the components of gasoline;
and that the purge period between tests was thus too short for effective
removal of ethanol vapor and regeneration of canister storage capacity. The
•1 CO
work of Furey and King had indicated the same selective adsorption of polar
compounds. The effects of fuel composition on the emissions of THC and ethanol
emission rates from the two test vehicles are shown in Figure 6-3.
specifications for the five test fuels are given in Table 6-30.
162
Basic
6-78
-------
TABLE 6-30. TEST FUEL SPECIFICATIONS
162
Properties
Vapor pressure (RVP)
IBP, °F
Ethanol (% vol)
API gravity
% Paraffin3
% Olefin3
% Aromatic3
1
9.15
91
1.4
59.8
69.7
0.4
28.5
2
9.10
101
6.2
58.7
67.5
0.6
25.7
Fuel
3
9.85
89
0.86
57.5
52.1
17.2
29.8
4
9.65
95
8.1
56.5
46.4
16.6
28.9
5
9.40
94
10.1
52.6
37.7
17.6
34.6
Determined by fluorescence indicator analysis (FIA).
Although methanol and ethanol have been suggested as technologically
suitable fuel extenders, methyl tertiary butyl ether (MTBE) has also been
shown to be a suitable blending component for gasoline. The use of MTBE as an
extender provides a means for using methanol—since MTBE is synthesized from
.methanol plus isobutylene—without incurring some of the disadvangates
associated with adding methanol.directly to gasoline.
Hydrocarbon emissions from one prototype and one production car fueled
with gasoline, gasohol, gasohol purged with nitrogen to reduce volatility
("adjusted gasohol"), or gasoline containing MTBE were determined by Furey and
T63
King. Their test results showed that evaporative emissions were increased
significantly by the addition of 10 percent ethanol to gasoline but'were
increased less by the additon of 15 percent MTBE to gasoline. Total evaporative
emissions from the production car, not equipped with a closed-loop carburetor
system, were higher with gasohol, adjusted gasohol, and MTBE than with gasoline.
6-79
-------
ETHANOL
HYDROCARBON
TAILPIPE
EVAPORATIVE
AGGREGATE
Figure 6-3. Effects of fuel composition on THC and ethanol
162
emission rates.
6-80
-------
Total evaporative emissions from the prototype car, which was equipped with a
closed-loop system, were higher with the two gasohols than with gasoline, but
were essentially unchanged with the 15 percent MTBE-85 percent gasoline blend.
Exhaust emissions of HC, CO, and NO from the production car were lower with
s\
gasohol and MTBE than with gasoline, but differences in these exhaust emissions
from respective fuels in the prototype car were small. Furey and King also
determined, by comparing "inlet vapor" going to the canister with "break-through"
vapor from the canister, that ethanol and MTBE, both of which are polar compounds,
are selectively adsorbed in the canister. It should be noted that ethanol has
been reported to be moderately reactive photochemically but that MTBE is
expected to have low reactivity. The potential for photochemical reactivity
of evaporative emissions from gasohol-fueled or MTBE-gasoline fueled vehicles
will depend in part, then, on whether such emissions are from canister inlet
lines from the fuel tank and carburetor or whether they are vapors that have
already passed through the canister(s).
Naman and All sup determined exhaust and evaporative HC emissions and fuel
economy on eight test vehicles fueled by gasoline (Indolene), by gasohol (10
percent ethanol-90 percent gasoline), "methagas" (10 percent methanol-90 percent
gasoline), 7 percent MTBE in gasoline, and 7 percent tert-butyl alcohol (TBA)
in gasoline. Overall, decreases in fuel economy were seen with all the blends
compared to gasoline. The decreases were greatest with methagas and least
with MTBE. Emissions tests from the blends as compared with gasoline showed
that (1) exhaust HC emissions were reduced 5 to 32 percent; (2) exhaust NO
J\
emissions ranged from unchanged to slightly increased, depending on the fuel
and emission controls; (3) exhaust CO emissions were reduced,8 to 46 percent;
and (4) evaporative emissions increased, ranging from 5 percent with TBA to 93
6-81
-------
-I CC
percent with methagas. Tests by Harrington et al. showed clear directional
changes in exhaust emissions from vehicles fueled by 20 percent MTBE-80 percent
gasoline. With the MTBE blend, HC emissions decreased 20 to 30 percent, CO
emissions decreased 40 to 50 percent, and NO emissions increased 12 to 22
J\
percent. Fuel economy was not compromised by the addition of MTBE, apparently
because the reduction in heating value of the fuel was offset by improved
efficiency at the leaner air/fuel ratio produced by the introduction of MTBE.
Several studies have been conducted on the presence and levels of "unregu-
lated pollutants," i.e., non-criteria pollutants in exhaust from vehicles
fueled with gasohol, methagas, or MTBE blends. In the study cited earlier,
Naman and All sup measured aldehydes in exhaust from vehicles equipped with
oxidation catalysts or TWC catalysts and fueled with 10 percent ethanol, 10
percent methanol, 7 percent MTBE, or 7 percent TBA in gasoline. Exhaust
emissions of aldehydes in cars with oxidation catalysts were unchanged with
gasohol, were 50 percent higher with methagas, 22 percent lower with MTBE, and
6 percent lower with TBA relative to gasoline. Aldehydes in exhaust from cars
equipped with TWC catalysts were two times higher with gasohol, unchanged with
methagas, 1.8 times higher with MTBE, and 2.3 times higher with TBA relative
to gasoline. Aldehydes were measured in this study by FID, concentrations
being corrected according to the FID response to standards of these oxygenates.
Harrenstien used a combined 2,4-dinitrophenylhydrazone gas chromato-
graphic method specific for C, through C5 aldehydes to measure them in exhaust
from cars fueled with gasohol or methagas blends. Results of the study showed
the following:
6-82
-------
1. Total aldehydes in exhaust increased, relative to gasoline, 25
percent in 20 percent ethanol, 10 percent in 20 percent methanol,
and 30 percent in 30 percent methanol.
2. Aldehyde concentrations in exhaust are more a function of fuel blend
than equivalence ratio (air/fuel ratio relative to the stoichiometric
ratio).
3. Total aldehyde concentrations for the alcohol blends tested ranged
from 80 to 110 ppm (compound). Composition of the total was 70 to
90 percent formaldehyde, 5 to 22 percent acetaldehyde, 1 to 4 percent
acetone (a secondary aldehyde, or ketone), 1 percent propionaldehyde,
and 3 to 6 percent acrolein (acrylic aldehyde).
Data from a study by Matsuno et a!.167 indicate that in exhaust from
methagas-fueled vehicles, hydrogen cyanide, NOX, NH3, THC, and photochemical
reactivity decrease as the proportion of alcohol increases. Their data showed
that formaldehyde and unburned methanol can be eliminated by the use of catalysts
Using ethanol-gasoline blends, they showed that ethylene in exhaust increases
with an increase in ethanol, whereas hydrocarbons >C5 decrease as ethanol
increases. Again, catalysts reduce the amount of ethylene produced. When
neat ethanol was used, HC were barely detectable in the exhaust; 86 percent of
the carbon detected was attributable to unburned ethanol. The total aldehyde
content of exhaust was shown to increase with increases in the alcohol content,
methanol or ethanol, except for acrolein, which decreased as the alcohol
content increased. Aromatics and their aldehydes (e.g., benzaldehyde and
tolualdehyde) decreased with increasing alcohol content. Formic acid in the
exhaust increased rapidly with increases in methanol, but acetic acid decreased
slightly. .With ethanol, the reverse was true.167
6-83
-------
2 -
IU
o
I*
m
'§
ui
EC
iN 6.88
PARTICLE BOUND
OLDSMOBILE DIESEL .
TOTAL HYDROCARBON - 0.456 g/mi
PARTICLE BOUND - 0.181 g/mi
HYDROCARBON
|*|5.90
NISSAN DIESEL
TOTAL HYDROCARBON - 0.351 g/mi
PARTICLE BOUND - 0.058 g/mi
HYDROCARBON
* 5.78
TURBO-CHARGED RABBIT DIESEL
TOTAL HYDROCARBON • 0-295 g/mi
PARTICLE BOUND - 0.050 g/mi
HYDROCARBON,
246 8101214161820222426283032343638404244
CARBON NUMBER
Figure 6-4. Distribution of diesel hydrocarbon exhaust emissions, by
carbon number, between gas-phase and paniculate forms.
6-84
-------
-
UJ
o -
z
Q •
CO
RELATIVE A
f~ 1 1 IT— i I ^~
p
— •-
m*
PI Q NO. 20 DIESEL FUEL
^,
Ljf
gj .LUBRICANT
:
1
1
1
i
I
1
1
1
1
I
I
I
I
i
i
\
*~
••-
—
„
I
10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44
CARBON NUMBER
Figure 6-5. The distribution of hydrocarbons in diesel fuel and .-•
lubricant according to carbon number.170
6-85
-------
TABLE 6-31. EMISSION RATES FOR HYDROCARBONS
IN EXHAUST FROM TWO DIESEL VEHICLES171
(g/mi)a
Compound
Methane
Ethane
Propane
Acetylene
Benzene
Ethyl ene
Propylene
Total nonreactive HC
Nonreactive HC,
% of total HC
Methane, % of
total HC
Total HC (hot
FID analysis)
1975
Peugeot
0.0121
0.0021
0.0010
0.0120
0.0092
0.0413
0.0137
0.0364
12.0
4.0
0.304
1973
Nissan
0.0135
0.0009
~0
, 0.0129
0.0073
0.0280
0.0094
0.0346
14.0
5.5
0.247
JTo convert g/mi to g/km, divide g/mi by 1.609.
6-86
-------
Diesel emissions are almost exclusively exhaust emissions. The gas-phase
hydrocarbon fraction of diesel exhaust is complex and shows a bimodal distri-
bution by carbon number. The fraction consists of light, cracked hydrocarbons
and of heavy fuel-like components ranging up to as high as C. 134'168"170
Except for methane, the light, cracked hydrocarbons are mainly olefins.
Ethylene, acetylene, and propylene are the predominant light hydrocarbons,
with smaller amounts of C. olefins and even smaller amounts of C^ and Cc
• O 'O
olefins.134 Researchers at EPA170 have found that the C1()-C40 organics are
dominated by unburned fuel and lubricant components. Some of the organics in
this range are particle-bound and some are gas-phase. The distribution of
diesel exhaust hydrocarbons according to carbon number is shown in Figure 6-4.
The exhaust hydrocarbons in the (^ to CIQ range result from the combustion
process, cracking from higher-molecular-weight organics. Those emitted in the
C10 to C40 ran9e result mainly from unburned fuel in the CIQ to C25 range and
from lubricant in the C^ to C^g range. Partial combustion and rearrangement
compounds occur in this range also. For comparison, the distribution of
hydrocarbons, by carbon number, in diesel fuel and lubricant is shown in
Figure 6-5. The No.20 diesel fuel- used was composed of 66.2 percent
paraffins, 1.3 percent olefins, and 32.5 percent aromatics, as determined by
FIA. Emission rates for low-molecular-weight hydrocarbons in exhaust of
two diesel vehicles, obtained with the 1975 Federal Test Procedure (FTP), are
shown in Table 6-31.171
It should be pointed out here that the "hot FID" procedure used to
collect these and similar data uses a hot filter in the sampling probe
in order to preserve in the vapor phase those hydrocarbons emitted.
At ambient air temperatures, higher-molecular-weight hydrocarbons
6-87
-------
may react chemically, condense and be adsorbed or migrate to already formed
7 17?
participate matter, or possibly even evaporate or desorb from particles. ''
Though some ambient air data have been reported in which hydrocarbons >C12
were present in the vapor phase, complications with adsorption and desorption
of these compounds during collection, coupled with the fact that gas chromato-
graphic columns often produce breakdown products in that carbon number range,
detract from the acceptability of such data. Some of the compounds >C-,2
should, according to their individual vapor pressures, occur in the vapor
phase. However, hydrocarbons emitted from automobiles are not emitted as pure
compounds in isolation from other compounds and from particles. Consequently,
the vapor pressures of the individual compounds cannot be used to determine
the phase in which compounds >C12 are likely to be found at ambient air tempera
tures. Furthermore, once hydrocarbons are absorbed by or adsorbed onto
particles, the Kelvin effect influences the vapor pressures of the materials
in the particulate matter. '
Particulate emissions from mobile sources—including their formation,
their role in photochemical processes, and their health effects—are treated
in two recent EPA publications. '
Low-molecular-weight aldehydes have also been found in diesel exhaust by
many workers. Formaldehyde has been reported most often, followed by
acrolein. Vogh measured the following aldehydes in one diesel exhaust
sample: formaldehyde (18.3 ppm); acetaldehyde (3.2 ppm); acrolein, acetone,
propionaldehyde, and isobutyraldehyde (2.9 ppm); n-butyraldehyde (0.3 ppm);
crotonaldehyde and valeraldehyde (0.4 ppm); hexanaldehyde (0.2 ppm); and
benzaldehyde (0.2 ppm). Data obtained by Braddock and Gabele showed the
dependence of diesel aldehyde emissions on fuel. These data are presented in
Table 6-32.171
6-88
-------
TABLE 6-32. ALDEHYDE EMISSIONS FROM DIESEL VEHICLES
OPERATING ON FIVE DIFFERENT FUELS, 1975 FTP171
(mg/mi)a
Test fuel and number (n) of tests
Compound
Formal dehyde
Acetaldehyde
Propional dehyde
+ acrolein
Crotonal dehyde
Hexanal dehyde
Benzal dehyde
Total aldehydes
Total HC (hot FID)
% Aldehydes
Jet A Local no. 1
(n=7) (n=7)
21.9
7.5
5.4
0.5
0.3
0.9
36.3
344
10.6%
16.9
7.4
6.2
~o
0.5
7.8
38.9
304
12.8%
Nat'l avg. no.
2 (n=4)
48.2
11.9
8.7
0.6
0.4
0.8
60.1
471
12.8%
No. 2-D
n=7
40.8
8.2
5.5
~0
0.1
1.4
55.9
489
11.4%
Nissan no. 2
n=5
25.4
10.8
4.7
6.4
~o
0.3
47.6
334
14.2%
To convert mg/mi to mg/km, divide mg/mi by 1.609.
177
In a study of catalyst-controlled diesel engines, Marshall et al. found
that all four of the noble metal catalytic converters tested effected 10 to 50
percent reductions in peak emission rates of hydrocarbons. Detailed examina-
tion by gas-liquid chromatography indicated that the catalysts tended to
oxidize preferentially the unsaturated hydrocarbons. The remaining
high-molecular-weight hydrocarbons were, as a result, predominately the
saturated compounds. Because peak emission rates of aldehydes occurred at
light load (low exhaust temperature), the catalysts had little influence on
6-89
-------
peak emission rates. Average emission rates, however, were reduced sub-
stantially.
Since diesels emit hydrocarbons almost exclusively from the tailpipe, and
gasoline-powered cars emit from both tailpipe and evaporative sources, diesels
generally emit fewer net gas-phase hydrocarbons. Diesel exhaust from uncontrol-
led vehicles can, however, contain higher concentrations of gas-phase hydro-
178
carbons than the exhaust of catalyst-equipped gasoline-fueled vehicles.
While the net production of exhaust HC emissions is important from the health
standpoint, the contribution of the kind of hydrocarbons emitted from diesels
to the formation of photochemical smog products is also important. Spicer and
17Q
Levy critically reviewed the nature and photochemical smog reactivity of
diesel engine exhaust organic components. Although an enumeration of the
assumptions made and a description of the reactivity scales employed are
beyond the scope of this paper, a summary of their findings on the reactivity
of the two exhaust organic fractions is presented in table 6-33.
179
Though a
direct comparison of gasoline and diesel exhaust emissions is difficult because
180
different test cycles are used, these data and those of Landen and Perez
indicate that the calculated reactivity of diesel exhaust, on a ppm emission
basis, is fairly low relative to gasoline exhaust.
Because transportation sources in general, and highway vehicles in
particular, are major contributors to hydrocarbons in ambient air, emissions
from these sources strongly influence the composition of hydrocarbon mixtures
in urban atmospheres. Table 6-34 summarizes emission characteristics of autos
fueled by gasoline, alcohol-gasoline blends, ether-gasoline blends, and diesel
fuels, since changes in patterns of fuel use or in the percentage of diesel-
versus gasoline-fueled vehicles in operation will result in qualitative as
well as quantitative changes in hydrocarbons emitted from mobile sources.
6-90
-------
TABLE 6-33. SUMMARY OF REACTIVITY OF DIESEL EXHAUST
HYDROCARBON EMISSIONS AND PERCENTAGE CONTRIBUTIONS
OF LOWER-MOLECULAR-WEIGHT HYDROCARBONS AND ALDEHYDES
179
Gas-phase
HC fraction
C1-C5
C1-C22
Al dehydes
Total HC
r — P v
ul U5» *
of total
Al dehydes ,
% of total
Concentration
ppm
38.7
70.0
32.2
102.2
55
46
Reactivity index units
% NO/ 03a
38 90
69 180
31 193
100 373
Percentage
50
108
.1 223.4
.3 483.0
.2 460.4
.5 943.4
contribution
46
95
E.I.a
38.1
98.0
119.3
217.3
39
122
Aerosol3
36.6
314.0
32.2
346.2
12
10
^Op photooxidation reactivity; ozone-forming potential; potential for forming
components causing eye irritation; and aerosol-forming potential.
6-91
-------
TABLE 6-34. SUMMARY OF EMISSION CHARACTERISTICS FOR AUTOS FUELED BY GASOLINE
DIESEL, AND ALCOHOL-GASOLINE OR ETHER-GASOLINE BLENDS
Auto, control device,
or fuel
Emission characteristics
Gasoline-fueled,
uncontrolled
Gasoline-fueled,
catalytic con-
verters
Lead additives in
gasoline
Ethanol-gasoli ne
blends (relative
to gasoline)
MTBE-gasoline blends
(relative to
gasoline)
1. Exhaust emissions about 40% paraffins; 24%
aromatics; 26% olefins; 11% acetylene.
2. Main components of exhaust emissions: methane,
ethane, acetylene, ethylene, propylene, C, olefins,
toluene, benzene, ji-butane, rrpentane, isopentane,
xylene; aldehydes and some organic acids, ketones,
phenols.
3. Evaporative emissions: 70% of carburetor emissions
are light paraffins and olefins; 90% of fuel-tank
emissions are light paraffins and olefins.
1. Exhaust emissions: about 62% paraffins, 17% aromatics,
18% olefins; 3% acetylene. Methane is about 10-30%
of exhaust emissions.
2. Catalysts preferentially oxidize unsaturated HC.
3. Lower reactivity per gram HC emissions than from un-
controlled gasoline-fueled.
4. Lower net HC emissions than from uncontrolled gas-
oline-fueled.
1. Presence of TEL increased HC emissions.
2. Absence of TEL (or TML) necessitates higher
aromaticity of gasoline to achieve higher
octane ratings.
1. Decreased exhaust THC emissions.
2. Increased evaporative THC emissions.
3. Increased aggregate (exhaust plus evaporative) THC
emissions.
4. Aldehydes in exhaust increased from non-catalyst cars;
unchanged from cars with oxidation catalysts; increased
from cars with TWC catalysts.
5. Increased in ethylene from non-catalyst cars; no increase
from cars with oxidation catalysts.
6. Increases in acetic acid with increasing alcohol content.
1. Decreased exhaust THC emissions.
2. Increased evaporative THC emissions, but less than with
alcohol-gasoline blends.
3. Aldehydes increased from cars with TWC catalysts;
decreased from cars with oxidation catalysts.
6-92
-------
TABLE 6-34 (continued).
Auto, control device,
or fuel
Emission characteristics
Methanol-gasoline blends 1.
(relative to gasoline)
2.
3.
4.
5.
1.
2.
100% Methanol
Diesels
3.
4.
1.
2.
4.
5.
Emission rate of exhaust THC not appreciably changed;
composition changed.
Higher methanol in exhaust.
Aldehydes in exhaust increased from non-catalyst cars;
no significant change in aldehydes with TWC catalysts;
higher aldehydes with oxidation catalysts.
Decreases in exhaust NH3, HCN, photochemical reactivity
with increases in % alcohol.
Increases in exhaust formic acid with increases in % alcohol,
Higher cold-start HC emissions than from gasoline-
fueled.
Significantly lower hot-start HC emissions than
from gasoline-fueled.
Exhaust emissions: mainly methane, ethane, ethylene.
Significantly higher methanol and aldehyde emission
than gasoline-fueled (aldehydes reduced by increasing
compression ratio or adding water to methanol).
Almost exclusively exhaust emissions.
Emissions: light, cracked HC, mainly methane,
ethylene, acetylene, propylene; also aldehydes
(C-.-Cg->, including acrolein), and acetone.
Lower reactivity per gram HC emissions than from
gasoline-fueled.
Lower net HC emissions than from gasoline-fueled.
Higher carbonyl emissions (aldehydes, ketones).
6-93
-------
6.4 AMBIENT AIR CONCENTRATIONS
Ambient air monitoring for hydrocarbons is not required under EPA regulations
for purposes of enforcing the HC NAAQS. The monitoring of hydrocarbons that
is conducted by the states is for the purpose of controlling oxidant levels.
Thus, states have measured ambient air levels of HC primarily to obtain data
that can be used as input to the respective models used to determine the
percentage reductions in HC emissions that are needed to attain and maintain
the NAAQS for ozone. Hydrocarbons in ambient air are also measured by the
states and by EPA for other purposes, including: (1) to verify models; (2) to
determine the effectiveness of control strategies; (3) to study oxidant transport
phenomena; and (4) to obtain empirical data to verify reaction mechanisms and
kinetics.
Since routine monitoring has not been required, most of the data available
have been obtained from field studies conducted by EPA, alone or in conjunction
with state and local control agencies or contractors. Most of the data presented
in this section were obtained by the Environmental Sciences Research Laboratory
(ESRL), ORD, EPA, Research Triangle Park; or by its contractors and grantees.
In many instances these data have not yet been published. In other instances,
data from the open literature are available and have been presented. Some of
the data given here are 24-hour averages of total NMHC, calculated from 2-hour
or 3-hour averages obtained during diurnal studies. Some are "instantaneous"
concentrations measured in grab samples that are taken over a period of 5 to
10 minutes. Some of the data describe concentrations of individual hydrocarbons,
some classes of hydrocarbons, and some total nonmethane hydrocarbons. These
data are presented in these various forms partly because they are available
only in these forms and partly in order (1) to compare them with data that
6-94
-------
appeared in the 1970 criteria document; (2) to compare ranges in ambient air
levels with levels used in clinical and toxicological ^studies and with levels
found in occupational exposures; (3) to compare profiles—qualitative and
quantitative—of hydrocarbon mixtures in ambient air from one city to another;
and (4) to compare profiles of ambient air hydrocarbon mixtures with auto
exhaust and with gasoline vapor.
One quite important point must be noted at this juncture since it relates
to comparisons between ambient air HC concentrations and exposure levels
reported in occupational, clinical, or toxicologic studies. Most of the
ambient air data reported in this section have been calculated in terms of ppm
or ppb carbon (C), largely because the calibration gas used for most FID
measurements is methane, CH*, which contains one carbon atom per molecule; and
because mixtures cannot be expressed in terms of compound concentration. Data
reported by physicians and toxicologists are generally expressed as ppm or ppb
compound. Atmospheric concentrations are expressed as ppm C or ppb C by
multiplying the concentration of the compound by the number of carbon atoms in
the compound. An example at this point may serve to clarify the difference.
Pentane contains five carbons. A concentration of 20 ppb compound of pentane
is equivalent to a concentration of 100 ppb C, since the compound.concentration
is multiplied by the number of carbons in the compound.
oc
In a recently published compendium, Graedel has documented the detection
in ambient air of more than 1000 organic compounds. Five sources account for
the bulk (in terms of number of entries) of the detected, identified substances:
gasoline-powered vehicles, diesel vehicles, tobacco smoke, turbines, and
vegetation. Ambient air concentrations for gas-phase hydrocarbons as reported
in Graedel's monograph are summarized in Table 6-35. These data do not include
6-95
-------
TABLE 6-35*
AMBIENT AIR CONCENTRATIONS OF HYDROCARBONS
REPORTED BY GRAEDEL3
Compound
Concentration range, ppb
(unless specified)
ALKANES
Methane
Ethane
Propane
Butane
Isobutane
2,3-Dimethyl butane
Pentane
Isopentane
2-Methylpentane
3-Methylpentane
2,2-Dimethylpentane
2,3-Dimethylpentane
2,4-Dimethylpentane
2,2,4-Tri methylpentane
Hexane
2-Methy1hexane
3-Methylhexane
Heptane
2-Methylheptane
Octane
Nonane
Decane
Undecane
Dodecane
Hexadecane
ALKENES AND ALKYNES
Acetylene
Ethene
Propene
Propadiene
Propyne
1-Butene
2-Methyl-1-butene
Isobutene
1,3-Butadiene
Isoprene
cis-2-Butene
trans-2-Butene
1.3-4.0 ppm
0.05-95
12-94
0.01-182
0.06-35
3.8
0.023-64
0.1-101
5-12
3-11
1-2
2
0.3-13
17
4-27
10-28
10
0.2-34
3.4
0.04-3.4
0.1-9.0
1.0-11.2
0.95-8.8
1.3-5.1
0.16-1.0
0.2-227
0.7-700
1-52
2-4
1-6
1-6
1-19
1-6
1-9
0.2-2.9
1-11
1-3
6-96
-------
TABLE 6-35 (continued)
Compound
Concentration range, ppb
(unless specified)
2-Methy1-2-butene 2-18
1-Pentene 1-12
4-Methyl-l-pentene 1-3
ci^-2-Pentene 2-6
trans-2-Pentene 2-4
2,4-Dimethyl-2-pentene 3-10
1-Hexene 3
cis-2-Hexene 4-8
TERPENESb
a-Pinene 0.93-1.20
B-Pinene 0.14-0.40
Myrcene , 0.74-1.0
Limonene 0.06-5.7
ALICYCLICS
Cyclopentane 2-14
Cyclopentene 2-6
Cyclohexane 3-6
Methylcyclohexane 3-7
cis-l,2-Dimethylcyclohexane 3
AROMATICS (BENZENE AND DERIVATIVES)
Benzene 0.025-57
Toluene 0.005-129
o-Xylene • 0.5-33
m-Xylene 1-61
p_-Xylene 1-25
1,2,3-Trimethylbenzene 1-2
1,2,4-Trimethyl benzene 3-15.3
1,3,5-Trimethyl benzene 1.3-11
1,2,3,5-Tetramethylbenzene 1.3-5.3
1,2,4,5-Tetramethylbenzene 1.5-3.9
Ethylbenzene 0.1-22
o-Ethyl toluene 0.7-2.6
m-Ethyltoluene 1.1-13
p_-Ethyl toluene 1.1-13
Ethyl dimethyl benzene 0.74-2
Styrene 1.5-5
n-Propylbenzene 1-6
Cumene 1-12
p_-Cymene 0.12-2
sec-Butyl benzene 4-15
tert-Butylbenzene 2-6
Compiled from Graedel.
Isoprene is emitted from manmade sources as well as from natural sources
(vegetation); terpenes are emitted from natural sources only.
6-97
-------
the unpublished EPA data presented later in this section, but are the result
of a comprehensive survey of the open literature.
Many of the published detailed studies of organic compounds in the urban
atmosphere were conducted in the 1960s. Results of those studies were summarized
2
and discussed in the 1970 criteria document for hydrocarbons. Published data
on total levels of organic compounds in urban atmospheres from 1970 through
the present are relatively abundant, but most of these data suffer from consider-
able measurement error because they were obtained by FID instrumentation
(Sections 5.2 and 6.2). Nevertheless, FID measurements constitute the most
comprehensive published data base on NMHC levels in ambient air. Some statistics
derived from FID-measured NMHC are shown for 1967 through 1972 for six cities
in Table 6-36.181
Because the Los Angeles, California, area has population and traffic
density patterns, meteorological characteristics, and emission sources that
are uniquely conducive to photochemical oxidant formation, NMHC concentrations
have been measured there for many years and these measurements make up the
best long-term hydrocarbon data base available. For comparison with earlier
data from Los Angeles that were reported in the 1970 criteria document, more
recent data from the Los Angles area of the California South Coast Air Basin
are reported here. Trends in concentrations of total NMHC for Los Angeles for
1963 through 1972 are shown graphically in Figure 6-6, based on data reported
-I QO
by Trijonis et al. In a separate study of trends in Los Angeles, Mayrsohn
and Crabtree looked at long-term changes in concentrations of three "tracer"
hydrocarbons—acetylene, propane, and isopentane. These three hydrocarbons
are uniquely associated with known sources and thus can serve as tracers that
indicate the contributions of those sources to total NMHC concentrations.
6-98
-------
> -a >
• +> cu
CM
r-
en
• -o >
I +J f-
o d d d d
rH rHOO CO CO i-H
VD ID oo oo co in i-HtHinoo
r-i t-H d d
inini-Hi-HCMPO OCOCM10
r-i d O 0
r-< oo «D r- r- «- 10
CO ID in ID VD 4- <4-
do"
^mor-in
ddddd
r-l O P«. W «*
r-i i-H d do
en i- -
o
(/>
t 2
fSt
3 m
z u
cn o r*» ID I-H cn
d r-i do r-i d
«• vo PO co r~- in
o o o o d d
lOCOCMOOrOCMtv,
ddddddd
^oooooo
oooooo
d d d d d d
en r-i en r-i eo CM
CM UD r-l PO ^ 00
rH CM CM tH CM
.o .0 _a _n .0
r^« co cn o rH CM
ID ID ID P*. P** P^
CMrHOOOO ^J-OOOCO
oooooo dddd
> O O CM rH O CM
oooooo oddd
OOOOOO rH o O O OO
oooodd dddd
oooooo oooo oooo
oooooo oodd odd
ooooo
CM CM O ^J- CO
ddddd
rHCM O PO PO
o o o d d
O O O CM CM
ddddd
oo in ^r CM CM
ddddd
ID "O- 00 I-H CM
odd d o'
dddo'd
CM CMr-l 00
ddddd
O r-4 O O O
ddddd
OOOrHi-H OOOOO
ooodo ooodo'
ooooo ooooo
ooodo' ddddd
SSSSStS rf!§s gsggsgs 11*11 ggggg
Q.£}.a.Q ^ .a
tDOOCnOfHCM
ID ID 10 r~. r^ r-
.
0 CM
O I
t_> (
0
£= '—'
a>
o o
• rH
O <£
ii
.T en
O
rH
•
-------
T 1 1 1 1 1 "1
'O MAX. HOURLY AVERAGES - ALL DAYS OF YEAR
D MAX. HOURLY AVERAGES - JULY. AUG.. SEPT.
A 6-TO-9 A.M. AVERAGES - JULY. AUG.. SEPT.
UJ,
i . 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972
YEAR
Figure 6-6. Nonmethane hydrocarbon trends in Los Angeles, 1963
through 1972.182
6-100
-------
Acetylene has been a good indicator of automobile exhaust because, until the
widespread use of the catalytic converter, automobiles have been the most
significant source of this HC. Isopentane is a major component of both
automobile exhaust and of gasoline. Propane, present only in trace amounts in
gasoline and auto exhaust, is a major component of natural gas. All three
of these compounds have relatively long residence times in the atmosphere and
therefore serve as good indicators of the actual quantities in the ambient air
of auto exhaust, gasoline vapor, and natural gas, respectively.174 Table 6-37
shows the 1967 through 1977 concentrations of acetylene, propane, and isopentane
at two sites in the Los Angeles area. Mayrsohn et a!.183 measured individual
NMHC species and total NMHC at four sites in the Los Angeles area during the
smog season (June through September) in 1975. Monitoring was not done con-
currently at these Long Beach, El Monte, Azusa, and Upland sites, so that
direct comparison of data are not possible. However, the data are useful for
indicating the NMHC composition at the respective sites as well as the total
NMHC concentrations. These data are presented for C2 through C-,1 NMHC and for
TOO
total NMHC in Table 6-38.^ Although descriptions are not given in this
section of the specific methodologies employed in obtaining GC measurements of
ambient air samples presented, special note is made on Table 6-38 of the
methods used since they differ from the methods used by ESRL, EPA. Mayrsohn
183
et al. measured HC up through Clg, whereas ESRL researchers do not commonly
attempt to separate or quantify compounds greater than C-,Q because chromatograms
often show peaks resulting from column breakdown products that occur at the
retention times and temperatures at which compounds >C-,Q are resolved.184 The
miniscule amounts of >C12 compounds also complicates their separation, identifi-
cation, and quantification. The individual concentrations shown in Table 6-38
are for C2 through C-^ only, but the total NMHC concentrations include the C2
through C-,g compounds.
6-101
-------
TABLE 6-37. TRENDS IN 6-TO-9 a.m. CONCENTRATIONS OF INDICATOR HYDROCARBONS AT
TWO SITES IN THE CALIFORNIA SOUTH COAST AIR BASIN, 1967 THROUGH
(ppb C)
Acetylene
Year
1967
1968
1969
1870
1971
1972
1973
1974
1975
1976
1977
Azusa
90
-
75
64
-
-
37
51
46
40
33
DOLAa
186
153
124
119
-
-
84
95
-
99
86
Propane
Azusa
87
-
113
91
-
-
56
71
63
53
56
DO LA
121
102
151
147
-
-
92
102
-
92
104
Isopentane
Azusa
136
-
126
.105
-
•
91
107
113
88
79
DOLA
302
280
214
245
-
- .
169
175
-
185
154
DOLA = downtown Los Angeles.
6-102
-------
TABLE 6-38. MAXIMUM AND AVERAGE 6-TO-9 a.m.
NONMETHANE HYDROCARBON CONCENTRATIONS3 AT FOUR
SITES IN CALIFORNIA SOUTH COAST AIR,BASIN,
JUNE THROUGH SEPTEMBER 197515"5
(ppb C)
Compound
Ethane
Ethyl ene
Acetylene
Propane
Propylene
Allene
Isobutane
Butane
Butene-1
Butene-2
Isopentane
Pentane
Pentenes
C6
C7
C8
C9
C10 .
Cll
I NMHC
y r -r
2. u2 un
Long Beach
Avg.
56
26
22
72
10
1
36
83
3
7
86
49
2
2
94
80
66
84
48
921
889
Max.
170
71
63
279
48
4
111
340
10
19
306
179
21
21
394
257
174
342
224
3216
3084
El Monte
Avg.
114
60
72
101
30
4
52
111
8
24
171
87
25
167
284
271
200
256
198
2396
2232
Max.
239
118
159
212
69
7
176
250
16
57
397
196
93
426
785
762
463
828
896
6137
5601
Azusa
Avg.
77
40
46
63
20
2
31
71
5
15
113
55
5
129
202
231
184
294
229
2079
1811
Max.
122
64
78
135
33
4
65
143
10
29
233
,110
30
282
344
444
1375
2149
2933
19425
7418
Upland
Avg.
51
40
43
48
14
2
30
88
4
8
91
47
2
128
169
157
111
202
145
1493
1378
Max.
100
108
88
97
20
4
70
262
9
16
188
94
11
502
441
503
547 ,
1597
967
5666
5156
Ethane through pentane separated on Durapak phenylisocyanate porasil-c
column. Pentane through xylene separated and quantified by means of SE-30
column. Xylene (CQ) through C18 adsorbed on graphitized carbon black,
desorbed, and separated and quantified by means of an OV-101 column.
GC calibrated with known concentrations of individual components.
6-103
-------
Total NMHC data are available for St. Louis from special field studies
conducted in 1972 and 1973 by the Environmental Sciences Research Laboratory
(ESRL), EPA, at Research Triangle Park, North Carolina.185 St. Louis, next to
Los Angeles, probably is the most-studied city in the United States with
respect to air pollution in general and oxidant and its precursors in particular.
St. Louis is representative of an urban area having better ventilation characteristj
than Los Angeles. Consequently, the ambient air levels of NMHC in St. Louis
are considerably lower than those observed in the Los Angeles area. Like Los
Angeles, however, the hydrocarbon composition in St. Louis is dominated by
186
automotive-related sources. Total NMHC data are presented in Tables 6-39
and 6-40 for one urban site in 1972 and for one urban site in 1973. These are
concentrations in 2-hour samples collected over 24-hours for 6 and 4 days,
respectively.
During the 1972 ESRL field study, individual hydrocarbon measurements by
gas chromatography (GC) were made at the St. Louis University urban site.
Representative data are presented in Table 6-41. These are 24-hour average
hydrocarbon concentrations calculated from twelve 2-hour sample measurements
on September 13, 1972. Inspection of Table 6^41 easily verifies that methane
concentrations far exceed the concentrations of all other hydrocarbons com-
bined. As the table footnote indicates, the value for ethylene rs erroneous
as the result of contamination from a nearby ozone chemiluminescence monitor.
Consequently, the total NMHC concentration is undoubtedly erroneously high.
Conspicuous by its absence from this table, benzene is not quantitatively
determined by the GC method employed in these and many of the other individual
hydrocarbon measurements obtained by ESRL that are reported in this section.
(The GC method used by ESRL involves separation of lower-molecular weight
aliphatics (C« - Cc) on one column, intermediate aliphatics and alicyclics
on a second column, and higher-molecular-weight aliphatics, alicyclics,
6-104
-------
TABLE 6-39. TOTAL NONMETHANE HYDROCARBON CONCENTRATIONS IN 2-hour
SAMPLES MEASURED BY GAS CHROMATOGRAPHY AT URBAN SITE IN ST. LOUIS
FOR 6 DAYS, SEPTEMBER 13-25, 1972a'b
(ppm C)
Sampling
period
11.00 p.m. -1.00 a.m.
1.00 a.m. -3. 00 a.m.
3.00 a.m. -5. 00 a.m.
5.00 a.m. -7. 00 a.m.
7.00 a.m. -9. 00 a.m.
9.00 a.m. -11. 00 a.m.
11.00 a.m.- 1.00 p.m.
1.00 p.m.- 3.00 p.m.
3.00 p.m.- 5.00 p.m.
5.00 p.m.- 7.00 p.m.
7.00 p.m.- 9.00 p.m.
9.00 p.m. -11. 00 p.m.
Range of
values
0.416 to 3.847
0.478 to 4.056
0.402 to 3.937
0.486 to 5.491
0.431 to 6.585
0.433 to 4.750
0.405 to 3.184
0.263 to 3.450
0.352 to 2.194
0.384 to 1.854
0.247 to 1.728
0.242 to 1.912
Average
(6 days)
1.663
1.466
1. 244
2.031
2.125
1.681
1.311
1.172
1.080
0.903
0.938
1.043
St. Louis University site.
Summed from measurements of individual hydrocarbons; ethylene values are
excessively high because of contamination, while benzene measurements are
absent. Total NMHC, because of erroneously high ethylene data, are
accordingly high.
6-105
-------
TABLE 6-40. TOTAL NONMETHANE HYDROCARBON CONCENTRATIONS IN 2-hour
SAMPLES MEASURED BY GAS CHROMATOGRAPHY AT URBAN SITE IN ST. LOUIS
FOR 4 DAYS, JUNE 21-JULY 8,1973a'b
(ppm C)
Sampl i ng
period
Midnight-2:OQ a.m.
2:00-4:00 a.m.
4:00-6:00 a.m.
6:00-8:00 a.m.
8:00-10:00 a.m.
10:00-Noon
Noon-2:00 p.m.
2:00-4:00 p.m.
4:00-6:00 p.m.
6:00-8:00 p.m.
8:00-10:00 p.m.
10:00 p.m. -Midnight
Range of
values
0.467 .to 0.858
0.270 to 0.842
0.199 to 1.112
0.341 to 0.860
0.465 to 1.567
0.350 to 1.013
0.297 to 1.014
0.512 to 0.972
0.391 to 1.090
0.235 to 0.645
0.384 to 0.644
0.514 to 0.639
Average
(6 days)
0.643
0.636
0.635
0.559
0.974
0.727
0.636
0.685
0.641
0.393
0.454
0.572
a
CAMP site.
Summed from measurements of individual hydrocarbons. Ethylene values
within normal range; benzene measurements are absent.
6-106
-------
TABLE 6-41. 24-hour AVERAGE HYDROCARBON
CONCENTRATIONS MEASURED BY GAS CHROMATOGRAPHY
AT URBAN SITE IN ST. LOUIS, SEPTEMBER 13, 1972a'b'
(ppb C)
Compound
24-hr avg. concn.
Methane
Ethane
Ethyl ene
Propane
Acetylene
Iso-butane
n-Butane
Propylene
Propadiene
Neopentane
Iso-pentane
1-Butene + Iso-Butylene
trans-Butene-2 + Methyl acetylene
cis-Butene-2
1,3- Butadiene
jr-Pentane
Pentene-1
2-Methylbutene-l
t-Pentene-2
cis-Pentene-2
2-Methylbutene-2
Cyclopentane +
2-Methylpentane
3-Methylpentane
4-Methyl pentene-2
Hexane
Hexene-1
2,2-Dimethylpentane
2,4-Dimethylpentane
2-Methylpentane
cis-2-Hexene
3 , 3-Dimethyl pentane
Cyclohexane
3-Methyl hexane
2 , 3-Di methyl pentane
3-Methyl hexane
1-ci s-3-Dimethy 1 cycl opentane
2, 2, 4-Tri methyl pentane
ji-Heptane
Methyl cycl ohexane
Toluene
5473.8
27.3
2212. 5C
24.0
24.8
14.0
57.1
6.4 ,
N.D.d
N.D.
70.8
8.2
4.8
1.4
1.1
26.4 '
2.1
3.2
5.3
1.6
5.8
36.0
11.2
0.0
13.8
N.D.
0.7
6.6
9.1
0.0
N.D.
N.D.
7.9
8.8
10.1
1.6
11.6
5.4
1.5
61.1
6-107
-------
TABLE 6-41 (continued)
Compound
24-hr avg. concn.
n-Nonane 5.2
Ethylbenzene 9.0
p_-Xylene 7.8
m-Xylene 24.3
o-Xylene 11.1
Isopropylbenzene N.D.
+ styrene
nrDecane 13.0
ri-Propyl benzene 2.9
m + p_-Ethyl benzene 6.9
1,3,5-Trimethylbenzene 3.0
tert-Butylbenzene 1.6
+ o-Ethyltoluene
sec-Butyl benzene 15.7
+ 1,2,4-Trimethyl benzene
Unknown 3.6
1,2,3-Trimethylbenzene 3.6
n-Butyl benzene + p_-Di ethyl benzene 3.6
Z NMHC
I NMHC - ethylene
2 Paraffins6
Z Olefins1
Z Aromatics9 154.2
2793.5
581
362.1
64.7
aSt. Louis University site.
Gas chromatograph calibrated by standards of individual compounds.
cEthylene concentration not valid; contamination by nearby ozone monitojjgg
system is indicated when the ethylene/acetylene ratio exceeds about 3.
Not detected.
eAlicyclics counted with paraffins.
Acetylene counted with olefins.
^C unknown counted with aromatics.
6-108
-------
and aromatics on a third column. The retention time for benzene is too long"
on the second column, so that column breakdown products and the diffuse nature
of the peak prevent the quantitative measurement of benzene. The separation
characteristics and a too rapid elution time for benzene also preclude
quantitative measurement on the third column. Measurements of benzene con-
centrations obtained by other GC methods will be presented separately.)
In a 1974 field study, ESRL obtained total NMHC and individual hydrocarbon
-i OC
data at a rural site in Ohio. The sampling sites for this study were selected
as rural ground-level locations downwind of major metropolitan complexes.
Although geographically a rural area (Wilmington; McConnelsville; and Wooster,
Ohio), this area was found to be affected, as suspected, by transport of
polluted urban air masses. Hydrocarbon composition was also found to be
influenced by local HC point sources in surrounding communities. Nevertheless,
data collected in the Ohio study are probably representative of many such
1 ftfi
rural areas across the United States. The distribution of total NMHC
concentrations (in ppb C) by number of samples is shown in Figure 6-7 for
Wilmington and McConnelsville (the latter is located about 100 miles east-
187
northeast of Wilmington).
Individual hydrocarbon concentrations at the Wilmington site were measured
by gas chromatography. These data, consisting of 24-hour average concentrations
calculated from ten 2-hour and four 1-hour samples, are presented in Table
6-42. Note that the ethylene concentration is quite low. This is significant
for a semi-agricultural area, since ethylene is a known phytotoxicant. The
low level reflects the relatively light traffic density, since ethylene is
emitted mainly in auto exhaust. Though values are reported ,for acetaldehyde,
propionaldehyde, and acetone in Table 6-42, these concentrations are indicative
6-109
-------
30 | 1
w 25 1
in i
s!
1 20
£ 15
DC
m 10
S 6
0
—
—
-
1
^
H
' ' ' I — IMORNING SA'MPL'ES '
I I 700- 900 _
V77/7X EVENING SAMPLES
J^0i 1700-1900
n n r
'(///i.
///t
W/i
HZL ~
]
0 100 200 300
WILMINGTON, OHIO
TOTAL NONMETHANE HYDROCARBON CONCENTRATION, ppbC
0 100 200 300
MCCONNELSVILLE, OHIO
Figure 6-7. Sum of nonmethane hydrocarbon concentrations versus
number of samples for two around sites used in the 1974 Midwest
Oxidant Transport Study.187
6-110
-------
TABLE 6-42. 24-HOUR AVERAGE INDIVIDUAL HYDROCARBON MEASUREMENTS AT
WILMINGTON, OHIO, July 18, 1974, DETERMINED BY GAS CHROMATOGRAPHY
(ppb C)
Compound
Ethane
Ethyl ene
Propane
Acetylene
Isobutane
n-Butane
Propylene
Isobutylene
trans-Butene-2
cijS-Butene-2
Butadiene-1,3
Isopentane
n-Pentane
Pentene-1
2-Methylbutene-l
trans-Pentene-2
cis-Pentene-2
2-Methylbutene-2
Cyclopentane
2-Methylpentane
3-Methylpentane
4-Methylpentene-2
n-Hexane
Hexene-1
trans-Hexene-3 ,
2,4-Dimethylpentane
Methyl eye 1 opentane
3,3-Dimethylpentane
Cyclohexane
2-Methylhexane
2 , 3- Dimethyl pentane
3-Methyl hexane
1-ci s-3-Dimethyl cycl opentane
2,2,4-Trimethylp^entane
n-Heptane
Methyl cycl ohexane
Benzene
Toluene
Nonane
Ethyl benzene
p_-Xylene
Concn.
range
7.4 - 13.9
1.1- 5.1
3.6- 10.7
1.6 - 9.2
1.5 - 7.6
3.1-14.7
0.3 - 1.5
0.3 - 2.0
0.8 - 1.7
N.D.
N.D.- 0.7
2.5 - 21.4
1.2 - 8.1
N.D.
N.D.
N.D.
N.D.
•N.D.
0.4 - 5.7
0.9 - 4.1
0.7 - 2.8
N.D.- 0.3
0.8 - 5.1
N.D.- 3.2
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
0.8 - 4.5
1.0 - 6.8
1.9 - 5.8
0.3 - 2.3
0.3 - 4.8
24- hr
average concn.
9.9
2.9
6.6
5.4
3.9
8.0
0.7
0.8
1.2
0.1
8. 8
3.7
_
_
„,,,
_
1.6
2.2
1.4
0.0
2.1
0.2
_
_
_
_
_
_
_
_
-
_
_
2.3
3.5
3.2
0.6
0.9
6-111
-------
TABLE 6-42 (continued).
Compound
m-Xylene
o-Xyl ene
n-Decane
n-Propyl benzene
m,£-Ethyl toluene
1,3, 5-Trimethy 1 benzene
o-Ethyl toluene
1 , 2 , 4-Tr i methyl benzene
Concn.
range
N.D.- 1.0
0.2 - 6.1
0.7 - 2.0
0.2 - 1.2
N.D.- 0.8
N.D.- 1.0
N.D.
0.3 - 2.2
24- hr
average concn.
0.6
1.6
1.2
0.3
0.4
0.1
—
0.8
Unknowns
Methane,, ppm
Total NMHCD
Acetaldehyde
Propionaldehyde
Acetone
0.5 - 10.2
1.48 - 2.43
42.3 -223.5
1.3 - 7.5
N.D.- 11.4
3.0 - 20.7
5.4
1.6
101.6
4.4
2.6
8.5
N.D. - not detected.
Includes aldehydes, acetone, and unknowns.
6-112
-------
but not absolute, since oxygenates cannot be measured quantitatively on GC
unless calibrated by a specific standard or derivatized prior to measurement.
These same data are presented in Table 6-43 by time of day for total hydrocarbons,
hydrocarbon classes, and selected oxygenates.
TABLE 6-43. CONCENTRATIONS OF TOTAL NONMETHANE HYDROCARBONS,
METHANE, AND SELECTED OXYGENATES IN ELEVEN 2-hour AND TWO 1-hour
SAMPLES TAKEN OVER 24 hours, JULY 18, 1974, WILMINGTON, OHIO
Compound/class, ppb C
Time
Midnight-2:00 a.m.
2:00-4:00 a.m.
4:00-6:00 a.m.
6:00-8:00 a.m.
8:00-10:00 a.m.
10:00 a.m. -Noon
Noon-2:00 p.m.
2:00-3:00 p.m.
3:00-4:00 p.m.
4:00-5:00 p.m.
5:00-6:00 p.m.
6:00-8:00 p.m.
8:00-10:00 p.m.
10:00 p.m. -Midnight
Methane
1500
1520
1520
1590
1590
1520
1520
1480
1600
2430
1500
1530
2200
1730
NMHCa
71.7
42.3
55.4
65.6
66.0
52.7
128.1
149.0
149.7
223.5
150.5
104.3
84.5
78.5
IPb
31.8
24.5
31.8
36.5
40.9
34.1
78.5
91.0
90.7
161.6
93.0
59.3
61.5
59.1
I0b
3.3
2.9
4.1
5.5
7.0
3.4
5.5
6.3
5.3
10.3
6.3
4.5
7.1
, 7.4
IAb
20.9
5.9
7.1
8.6
9.2
6.8
10.0
12.6
14.7
14.5
19.2
14.8
6.0
5.2
Un-
knowns
9.0
2.2
4.0
2.8
1.9
3.2
10.2
9.7
7.8
7.3
9.6
6.9
0.5
0.5
Alde-
hydes
1.9
1.8
4.1
5.3
3.2
2.2
8.5
10.2
10.5
18.7
11.8
7.4
6.0
6.3
Ace-
tone
4.8
5.0
4.3
6.9
3.8
3.0
15.4
19.2
20.7
11.1
10.6
11.4
3.4
N.D.
.Includes aldehydes, acetone, and unknowns.
IP = total paraffins; 10 = total olefins; IA = total aromatics.
Note the absence in Table 6-42 of the natural hydrocarbons, isoprene (C,-)
3
and the monoterpenes (C10). These compounds can be resolved and identified
with the chromatographic columns used in this and subsequently cited studies
unless otherwise noted. Isoprene in this chromatographic system would be
eluted, if present, between the pentanes and the lower-molecular-weight alicyclics;
the monoterpenes would be eluted, if present, after toluene, the xylenes, and
6-113
-------
ethyl benzene.110 -The absence of natural hydrocarbons from ambient air at this
rural site is not atypical of results reported by other researchers for other
rural and remote, forested and nonforested sites. Though emission rates for
natural hydrocarbons from vegetation have been found to be reasonably high
(Section 6.3), ambient air concentrations, even within the canopy of forests,
have been found to be low.
Seila,104 sampling in January, found total NMHC concentrations of about
0.2 ppm C in Jones State Forest, which is primarily a loblolly pine forest,
with some oak and sweet gum present, that is located 38 miles north of Houston.
Only 2.2 percent of the NMHC concentration, measured in the forest, was natural
hydrocarbons, a concentration of about 0.004 ppm C (about 4 ppb C). Arnts and
Meeks188 reported similar concentrations in samples taken from various rural
and remote areas during warmer months. They found that total NMHC concen-
trations generally averaged 40 to 100 ppb C, of which natural hydrocarbons
were 10 percent or less (<4 to <10 ppb C). For rural sites in the East,
Ferman189 reported total NMHC concentrations of 0.1 ppm C or less and
terpene concentrations of 0.5 to 2 ppb C. He used a GC method, however,
that did not permit unequivocal identification of the terpenes. In his
studies, isoprene levels were generally around 2 ppb C. Westberg
reported values for natural hydrocarbons in a northern Idaho pine and fir
forest of 50 to 1500 parts per trillion C (ppt), with an average of 200 ppt C
(0.2 ppb C). Using a specialized mass spectrometric technique in conjunction
with GC, Westberg reported the concentrations for natural hydrocarbons in
rural ambient air shown in Table 6-44.110 The effect of distance from the
6-114
-------
forest is not surprising in view of similar data reported by other researchers
(cf. Zimmerman, in Section 6.3). Outside the forest, only benzene and
toluene, which are associated with manmade sources, are detectable.
TABLE 6-44. CONCENTRATIONS OF NATURAL HYDROCARBONS IN
AMBIENT AIR SAMPLES TAKEN WITHIN AND OUTSIDE A FOREST CANOPY
(ppt C)
110
Compound
Isoprene
Benzene
Toluene
ot-Pinene
p-Pinene
A-3-Carene
Forest samples,
ground level
580a
610
190
310
630
370
Outside canopy,
3 ft above ground
N.D.b
580
190
N.D.
N.D.
N.D.
For ppb, multiply by 0.001.
bN.D. = Not detectable.
Data representative of a major suburban area were obtained by ESRL in the
greater Boston area during the Northeast oxidant transport study in 1975.185 The
individual hydrocarbon data are presented in Table 6-45 as 24-hour average
concentrations, in ppb C, that were calculated from twelve 3-hour samples.
The data presented here were obtained at sites 10 to 15 miles north and south,
respectively, of central Boston. Middleton and Medfield are suburbs north of
town; Chickatawbut Hill is an Audubon Society reserve south of town. The
qualitative composition is expected to be similar to that in the central city;
6-115
-------
-I OC
the concentrations, however, are lower as a result of dilution. It is also
quite possible that the most highly reactive hydrocarbons may have been con-
sumed by the time concentrations were measured 10 to 15 miles out, given the
time required for dispersion and transport. The mobile sampling and analysis
lab at Chickatawbut Hill was located at the crest of the hill and was surrounded
on all sides by trees. Nevertheless, no natural hydrocarbons were detected
(a-, p-pinene; A-carene; and limonene in Table 6-45).
Data from one final urban area are presented. Houston, Texas, is a good
example of an urban area whose atmospheric hydrocarbon burdens are influenced
by both automotive and industrial emissions. Houston has a high density of
petroleum production and refining activities as well as other major chemical
and petrochemical industrial activities. In addition, the shipping channel
from Houston to Galveston (another petroleum/industrial city) is responsible
for emissions (ship and possibly fugitive emissions from petroleum and other
cargo) that influence the ambient air of Houston. The Houston area has a
greater number of hydrocarbon-polluting industries than St. Louis or Boston;
and certainly more than the rural area represented by Wilmington and
McConnelsville, Ohio, even thdugh local influences there included a
186
Consequently, the Houston area is expected to have a higher
refinery.
proportional contribution from industrial hydrocarbons than would be found in
186
St. Louis or Boston.
The Houston data presented here were obtained at numerous sites in and
around the city. A complete analysis of the available data has not yet been
made. However, the ranges in total NMHC from day to day and from site to site
demonstrate both the effects of local meteorology and the effects of unique
local point sources (versus the more homogeneous emissions of auto exhaust and
6-116
-------
TABLE 6-45. 24-hour AVERAGE INDIVIDUAL HYDROCARBON
CONCENTRATIONS MEASURED BY GAS
CHROMATOGRAPHY, BOSTON AREA, 1975
(ppb C)
Middleton, 8/11/75
Avg.
Ethane
Ethyl ene
Propane
Acetylene
Isobutane
ji-Butane
Propylene
Isobutylene
trans-Butene-2
cis-Butene-2 + Butadiene
Isopentane
rrPentane
Pentene-1
2-Methylbutene-l
trans- Pentene-2
cis-Pentene-2
2-Methylbutene-2
Acetaldehyde
Cycl opehtane
2-Methylpentane
3-Methylpentane
4-Methylpentane
nrHexane
Hexene-1
Unknown
trans-Hexene-3
2,4-Dimethylpentane
Cyclopentane
cis-Hexene-2
Unknown
Propionaldehyde
Acetone
3,3-Dimethylpentane
Cyclohexane
2-Methylhexane
2,3-Dimethylpentane
3-Methyl hexane
Unknown
l-c-3-Dimethyl cycl o-
pentane
2,2,4-Trimethylcyclo-
pentane
l-t-3-Di methyl cycl o-
pentane
n-Heptane
Methyl cycl ohexane
Unknown
9.
7
Contain.
11.
10.
9.
22.
3.
3.
1.
1.
25.
14.
0.
0.
N.
N.
0.
(6.
6.
10.
6.
'N.
6.
N.
- o.
0.
0.
4.
0.
N.
(N.
(6.
N.
0.
3.
1.
3.
1.
0.8
3.
0.
2.
1.
--
9
7
9
4
6
5
0
8
6
3
6
8
D.
D.
6
6)
0
3
3
D.
6
D.
5
3
9
6
1
D.
D.)
8)
D.
8
2
8
3
6
0
3
8
9
-
Range
7.3 -
12.1
Contain.
6.9 -
7.5 -
5.8 -
14.6 -
2.0 -
1.3 -
0.2 -
0.8 -
15.7 -
9.0 -
0.3 -
0.3 -
N.D.
N.D.
N.D.
(3.5 -
N.D.
N.D.
3.9 -
N.D.
3.9 -
N.D.
N.D.
0.2 -
0.2 -
2.8 -
N.D.
N.D.
(N.D.
(1.7
N.D.
0.2 -
1.3 -
0.4 -
1.5 -
N.D.
N.D.
1.8 -
N.D.
0.9 -
N.D.
17.4
17.6
13.1
32.4
6.8
6.8
2.2
2.7 .
39.3
22.5
1.3
1.5
- 1.5
8.9)
- 9.3
- 16.3
10.7
10.8
- 0.8
0.7
1.9
6.2
- 0.4
- 0.2
)
- 12.6)
1.3
5.4
3.4
6.3
- 12.1
-1.5
5.1
- 0.6
4.7
- 7.3
Medfield, 8/21/75 Chickatawbut,
Avg.
3.9
Range
2.0
- 5.2
Contain. Contam.
9.0
5.7
4.1
10.3
2.3
2.4
1.2
1.0
15.1
10.5
0.5
0.7
1.0
.0.4
0.9
(2.0)
5.2
6.1
3.9
0.1
4.2
0.2
0.4
0.4
0.9
3.1
0.4 .
0.1
(0.3)
(2.0)
N.D.
0.6
2.8
1.4
2.8
0.1
0.4
2.3
0.3
1.9
2.8
1.6
1.6
2.1
1.6
3.4
0.6
1.4
0.5
0.5
5.6
2.8
0.2
0.3
N.D
N.D
N.D
N.D
1.3
2.2
1.4
N.D
1.6
N.D
N.D
N.D
N.D
1.4
N.D
N.D
(N.
(N.
N.D
0.3
1.1
0.7
1.0
N.D
N.D
0.5
, N.D
0.9
1.6
N.D
- 21.6
- 10.1
- 6.7
- 15.8
- 5.4
- 4.0
- 1.7
- 1.9
- 24.1
- 28.7
- 0.9
-1.5
. - 3.1
. - 1.7
. - 2.0
. - 4.0
- 8.2
- 11.2
- 6.4
. - 0.3
- 6.9
. - 0.6
. - 0.9
. - 0.8
. - 1.8
- 5.2
. - 0.9
. - 0.4
D. - 1.3)
D. - 9.2)
.
- 1.0
- 4.8
- 2.2
- 4.7
. - 0.6
. - 1.4
- 4.8 ,
. - 1.0
- 3.2
- 4.5
. - 11.1
Avg.
2.9
5.2
2.7
3.9
2.8
6.2
1.8
1.8
0.7
0.7
8.2
4.3
0.2
0.2
N.D.
N.D.
0.1
(2.4)
5.4
3.8
2.6
0.0
3.5
0.0
0.1
0.1
0.2
2.4
N.D.
N.D.
N.D.
2.0
N.D.
0.6
2.1
1.0
2.2
0.5
2.0
0.5
2.0
2.1
—
8/21/75
Range
2.
1.
1.
0.
0.
1.
0.
0.
0.
N.
2.
1.
N.
N.
N.
N.
N.
1.
1.
1.
0.
N.
1.
N.
N.
N.
N.
1.
N.
N.
N.
1.
N.
N.
N.
N.
N.
N.
N.
N.
N.
N.
.--
0
2
2
9
8
8
4
6
6i_
Db
8
5
D.
D.
D.
D.
D.
2
1
9
7
D.
0
D.
D.
D.
D.
0
D.
D.
D.
1
D.
D.
D.
D.
D.
D.
D.
D.
D.
D.
-
-4.2
- 6.8
- 4.2
- 6.9
- 6.0
- 12.3
- 2.8
- 4.0
- 0.9
. - 1.3
- 13.8
- 6.4
- 0.4
- 0.6
- 0.8
- 3.8
- 11.5
- 5.7
- 4.2
- 0.2
- 9.1
- 0.3
- 0.4
- 0.3
- 0.6
- 5.1
- 3.8
- 1.8
- 4.0
- 1.8
- 4.2
- 2.1
- 3.0
- 1.3
- 4.4
- 4.9
6-117
-------
TABLE 6-45 (continued).
Middleton, MA
Toluene
Unknown
Unknown
Unknown
n-Nonane
Unknown
Unknown
Unknown
Unknown
Ethyl benzene
p_-Xylene
m-Xylene
Unknown
crPinene
£-Xylene
"n-Decane
Tsopropyl benzene
Unknown
p-Pinene
Unknown
nrPropyl benzene
jm,p, -Ethyl toluene
Unknown
Unknown
A-3-Carene
1,3, 5-TH methyl benzene
0- Ethyl toluene
1, 2, 4-Trimethyl benzene
Unknown (Undecane?)
D-Limonene
Avg.
26.5
1.5
N.D.
N.D.
2.8
0.9
0.2
0.1
0.2
7.2
6.8
21.5
0.0
N.D.
13.2
3.9
5.8
1.5
N.D.
N.D.
1.8
6.7
0.7
N.D.
N.D.
1.5
1.4
7.3
1.4
N.D.
Range
N.D.
N.D.
N.D.
N.D.
0.3
N.D.
N.D.
N.D.
N.D.
2.5
2.3
7.2
N.D.
N.D.
4.9
0.9
1.1
N.D.
N.D.
N.D.
N.D.
3.0
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
- 44.8
- 6.5
- 6.3
- 2.9
- 0.6
- 1.0
- 0.4
- 23.3
- 19.4
- 66.4
- 0.4
- 30.2
- 7.8
- 23.8
- 4.1
- 4.7
- 19.3
- 3.1
- 3.8
- 2.9
- 12.5
- 8.3
Medfield," MA Chickatawbut Hill, MA
Avg.
21.4
0.2
N.D.
N.D.
0.9
—
—
0.1
0.6
5.8
6.6
14.4
N.D.
N.D.
6.5
1.7
2.0
0.6
N.D.
N.D.
0.8
3.4
N.D.
N.D.
N.D.
0.9
0.8
5.7
Range
7.9
N.D.
N.D.
N.D.
0.2
—
—
N.D.
N.D.
3.2
3.2
N.D.
N.D.
N.D.
4.8
0.5
0.5
N.D.
N.D.
N.D.
N.D.
1.0
N.D.
N.D.
N.D.
N.D.
N.D.
2.0
- 43.6
- 0.8
- 2.2
- 0.5
- 5.0
- 8.7
- 17.0
- 24.2
- 9.6
- 4.. 4
- 3.5
- 2.1
- 1.6
- 6.4
- 1.6
- 1.4
- 8.6
Avg.
18.3
—
—
0.7
N.D.
N.D.
1.1
1.8
3.7
2.2
N.D.
N.D.
3.1
0.8.
0.8
0.2
0.6
2.0
—
0.3
0.3
4.1
3.7
Range
8.3
—
—
—
0.2
—
N.D.
N.D.
N.D.
0.6
0.4
N.D.
N.D.
N.D.
1.6
N.D.
N.D.
N.D.
—
—
N.D.
0.6
—
—
N.D.
N.D.
2.5
N.D.
- 41.3
-1.6
- 3.3
- 3.1
-8.5
- 6.3
- 4.4
- 1.1
- 1.7
- 1.0
- 1.1
- 2.6
- 0.6
- 0.7
- 5.4
- 9.2
HC sampler located adjacent to ozone chemiluminescence monitor that employed
ethylene; data contaminated.
5N.D. Not detected.
6-118
-------
fuel evaporation). Variations with time of day demonstrate the effects of
ventilation characteristics in respective urban areas. The data for both St.
Louis and Houston show that during the 6-to-9 a.m. period hydrocarbon concen-
trations can be as much as ten times greater than the concentrations during
the l-to-4 p.m. period, even though there is no variation with time of the
source emission rates.
Data presented in Table 6-46 demonstrate the diurnal variations in total
NMHC (measured by FID) at one site in Houston in 1976. The data show the
average concentration and the concentration range for samples measured hourly
over 24 hours for 23 days. The concentration during the 6-to-9 a.m. period is
about three times higher than that of the l-to-4 p.m. period.190
Additional hydrocarbon data for Houston have been acquired by ESRL in an
extensive field study conducted by EPA in conjunction with local and state air
pollution agencies. More recent data are still being reduced and prepared for
publication. Data for 1973 and 1974 portions of the Houston study are presented
in Tables 6-47 and 6-48. Concentrations of the three major classes of hydro-
carbons in air samples collected at 19 different sites, in three separate
1 Qi
phases of the study, are shown in Table 6-47. Paraffins include the alicyclics
for purposes of this tabulation. Detailed gas chromatographic analyses of
hydrocarbons in samples collected at three representative sites in Houston in
-|Q-|
1973 are presented in Table 6-48. yj~ Site 1 is located in the midwestern
sector of the city and is representative of the influence of auto exhaust.
The site is the farthest of the three from the Houston shipping lanes and is
the closest of the three sites to a major traffic artery.191 The influence of
auto emissions is clearly discernible in the concentrations and proportions of
ethylene, acetylene, isopentane, 2- and 3-methylpentane, toluene, and the
6-119
-------
TABLE 6-46. TOTAL NONMETHANE HYDROCARBON
CONCENTRATIONS IN HOUSTON BY TIME OF DAY FOR 23 DAYS,
JULY 3-25, 1976, DETERMINED BY FID190
(ppm C)
Time of day
1:00 a.m.
2:00
3:00
4:00
5:00
6:00
7:00
8:00
9:00
10:00
11:00
Noon
1:00 p.m.
2:00
3:00
4:00
5:00
6:00
7:00
8:00
9:00
10:00
11:00
Midnight
Concn.
range
0.1 -
0.1 -
0.1 -
0.1 -
0.1 -
0.0 -
0.2 -
0.2 -
0.2 -
0.2 -
0.0 -
0.1 -
0.1 -
0.1 -
0.1 -
0.1 -
0.1 -
0.1 -
0.1 -
0.1 -
0.3 -
0.2 -
0.2 -
0.2 -
1.6
1.4
1.6
2.1
1.6
2.9
1.9
1.7
1.5
0.9
1.3
0.7
0.7
0.8
0.6
0.6
0.9
1.3
1.3
2.0
1.9
2.4
1.3
1.3
Avg. concn.
0.6
0.6
0.6
0.7
0.7
0.8
1.0
1.0
0.7
0.5
0.4
0.4
oTs
0.3
0.3
0.3
0.4
0.5
0.6
0.6
0.7
0.8
0.6
0.6
6-120
-------
TABLE 6-47. CONCENTRATIONS OF HYDROCARBONS
BY CLASS IN AIR SAMPLES COLLECTED AT 19 SITES9 IN HOUSTON, 1973
AND 1974, DETERMINED BY GAS CHROMATOGRAPHY191
(ppb C)
Date
September
1973
January
1974
April
1974
Sites
7 (HOI, H03,
H04, H06, H07, H08, H09)
7 (H12, H13, H14,
H15, H16, H17, H18)
5 (H21, H22,
H23, H24, H25)
Class
Paraffins5
Olefins
Aromatics
Total
Paraf f i ns
Olefins
Aromatics
Total
Paraffins
Olefins
Aromati cs
Total
Avg.
concn.
298.8
79.6
179.1
_ 55775
3315.8
533.1
904.4
4753.3
470.9
136.4
143.0
75O
% of
total
53.6
14.3
32.1
looTo
69.8
11.2
19.0
TOOTO
62.8
18.2
19.0
100.0
Sampling sites in traffic tunnels were excluded from this tabulation in order
to obtain averages more representative of Houston ambient air.
Paraffins include the alicyclics in this tabulation.
6-121
-------
TABLE 6-48. CONCENTRATIONS OF INDIVIDUAL HYDROCARBONS IN AIR
SAMPLES COLLECTED IN HOUSTON ON SEPTEMBER 11, 1973,
1Q1
. DETERMINED .BY GAS CHROMATOGRAPHY-13-1-
(ppb C)
Compound Site 1
Ethane
Ethyl ene
Propane
Acetyl ene
Isobutane
n-Butane
Propyl ene
Isobutylene
trans-2-Butene
cis-2-Butene
1,3-Butadiene
Isopentane
in-Pentane
1-Pentene
2-Methyl -1-butene
trans-2-Pentene
cis-2-Pentene
2-Methyl -2-butene
Acetaldehyde
Cycl opentane
Isoprene
2-Methyl pentane
3-Methy 1 pentane
4-Methyl -2-pentene
n-Hexane
1-Hexene
Unknown
trans-3-Hexene
2,4-Dimethylpentane
Methyl cycl opentane
cis-2-Hexene
Unknown
Propionaldehyde
Acetone
3,3-Dimethylpentane
Cyclohexane
2-Methyl hexane
2 , 3-Dimethyl pentane
3-Methyl hexane
35.1
174.7
28.0
41.9
41.5
111.3
22.5
20.8
4.1
2.5
4.1
117.8
58.5
4.8
7.1
9.4
11.4
7.9
— — —
___
___
52.3
21.8
0.0
30.4
0.0
0.0
7.2
12.7
— — —
0.0
— — —
8.0
___
15.3
22.5
22.9
l-cis-3-Di methyl cycl opentane —
2,2,4-Trimethylpentane 32.0
l-t-3-Dimethyl cycl opentane
ji-Heptane
16.2
Site 2a
24.8
30.8
22.7
8.8
21.4
38.0
7.3
8.4
4.0
1.0
2.8
42.1
20.8
1.7
1.8
2.9
7.8
3.1
_ — _
mmmmmf
— — —
21.7
7.3
0.0
14.3
0.0
0.0
— -
5.0
10.1
— — —
0.0
— "" —
0.0
0.0
0.0
0.0
0.0
0.0
0.0
— — —
0.0
Site 3a
21.7
24.5
19.0
20.7
11.4
29.2
9.6
11.6
4.0
0.0
4.0 .
33.5
20.0
1.3
1.1
2.2
1.3
1.8
— — —
""•*"""
— ~T"
19.8
8.5
0.0
8.5
0.0
0.0
---
2.1
4.7
___
0.0
---
_«_
0.0
0.0
0.0
0.0
0.0
0.0,
0.0
___
0.0
Avg.
concn.
27.2
76.7
23.2
23.8
24.8
59.5
13.2
13.6
4.0
1.2
3.6
64.5
33.1
2.6
3f3
4.8
6,8
4.3
__••>•
, ~* «•
W«M
31.3
12.5
0
17.7
0
0
— — —
4.8
9.2
_ mm mm
0
— — —
mm— mm
2.7
0
5.1
7.5
7.6
0.0
10.7
HIM —
5.4
6-122
-------
Table 6-48 (continued).
Compound
Methyl cycl ohexane
Toluene
n-Nonane
Ethyl benzene
p_-Xylene
m-Xylene
o-Xylene
Isopropyl benzene
ri-Decane
n-Propyl benzene
p_-Ethyl toluene
m- Ethyl toluene
1, 3 ,5-Tri methyl benzene
o- Ethyl toluene
1,2, 4-Tr i methyl benzene
Unknown
1,2, 3-Tr i methyl benzene
n-Butyl benzene +
p_-Di ethyl benzene
m-Di ethyl benzene
Unknown
Methane, ppm
Site la
16.8
132.0
19.9
- 19.8
20.0
86.1
75.1
13.5
20.4
3.8
15.5
—
15.5
5.7
182.9
36.6
0.0
0.0
0.0
0.0
2.01
Site 2a
0.0
35.6
6.4
6.7
5.0
13.0
8.8
7.3
10.1
4.5
10.9
0.0
9.6
31.8
2.3
0.0
0.0
0.0
1.77
Site 3a
0.0
44.2
7.8
5.3
5.3
21.0
8.4
8.7
1.3
5.8
2.6
1.5
6.8
10.1
0.0
0.0
0.0
0.0
1.75
Avg.
concn.
5.6
70.6
11.4
10.6
10.1
40.0
30.8
10.4
13.1
3.2
10.7
6.0
5.6
73.8
23.4
0.8
0
0
0
1.84 ppm
See text for site information.
6-123
-------
xylenes; all of these are considered to be among the predominating hydrocarbons
in auto exhaust. Sites 2 and 3 are in the northeast and southeast sectors
of Houston, respectively; they are much closer to the shipping lanes than site
1 and are not close to any major traffic arteries.
More recent data are available from a study conducted by EPA in Houston
in September through December 1978 to investigate visibility and photochemical
192
ozone problems in that area. One important aspect of the study was to
obtain data on the hydrocarbon composition of ambient air at various locations
in the Houston area for subsequent use in a sophisticated photochemical model
for ozone. The 1978 investigation of air quality in the Houston area consisted
of two individual but complementary studies, the Houston Aerosol Characterization
Study (HACS) and the Houston Ozone Modeling Study (HOMS), both of which were
the result of a Congressional mandate. The data reported here were collected
as part of HOMS.
Grab samples were collected and analyzed by gas chromatography. A total
of 109 peaks were seen on the chromatograms, of which unidentified peaks
represented 8 to 12 percent of the total NMHC concentration. These unknown
compounds may have resulted from sample contamination by outgassing of the
sample container, but that is uncertain. The problem is more likely to have
192
affected the sum of unknown aromatics than of paraffins or olefins. It is
possible, since a three-column GC procedure was used, that one or more unknown
compounds may have been resolved by more than one column, and thus may have
192
been counted twice in the total NMHC concentration. For these reasons, plus
the relatively low percentage contribution of the unknowns to the total concen^-
tration, the unknowns were excluded from the total NMHC concentration. Only
identified compounds are included in the data reported here. Table 6-49 shows
6-124
-------
the average percentage composition of paraffins, olefin, and aromatics observed
at nine HOMS sites. It also includes, for comparison, similar information
obtained from two sites in St. Louis in 1972 and 1977.192
The results from Houston agree reasonably well with those from St. Louis.
Likewise, comparison of these Houston data with those presented in Table 6-47,
which were obtained in 1973 and 1974, shows good agreement between the composite
averages of the earlier data and the 1978 data from individual sites. It
should be noted that Table 6-49 reports only the average composition observed.
The composition not only varies from site to site, as shown by these averages,
but also fluctuates day to day at each site. The standard deviation reported
in Table 6-49 gives an indication of the day-to-day fluctuations that occur.
The largest variations are seen in the olefin and aromatic percentages.
Variations in the more abundant paraffins are smaller.192 Overall, the largest
variations were seen at sites near industrial areas, where emission patterns
are expected to be more variable. The ratio of NMHC to NO at the respective
s\
sites is included in Table 6-49 and compared with limited data from St. Louis
(1977). The data from Houston agree with the limited data from St. Louis
except for Houston site 12, which is a site in a "boundary" area located close
to Galveston Bay.
The effects of meteorology on hydrocarbon concentrations are demonstrated
in Table 6-50, which shows the effects of wind direction on site 03, an urban
monitoring site. When the winds were from nonindustrial areas, north to
northeast, a higher percentage of the NMHC concentration could be attributed
192
to vehicular tailpipe sources. ' When the prevailing winds shifted and were
from the east and southeast, two effects were observed: (1), the percentage
tailpipe emissions decreased; and (2) the total NMHC concentrations increased
for both the morning and afternoon time periods.192
6-125
-------
TABLE 6-49. AVERAGE PERCENTAGE COMPOSITION OF HYDROCARBONS IN
AMBIENT AIR AT SELECTED SITES IN HOUSTON (1978) AND ST. LOUIS
1
(1972 AND 1977), 6 TO 9 A.M. PERIOD
Site
HOMS
02
03
05
06
10
12
15
17
20
ST.
1972
1977
Type of
site
-
Industrial
Urban
Suburban
Suburban
Industrial
Boundary
Boundary
Suburban
Boundary
LOUIS
Paraf f i ns
63.0 ± 10.9
60.0 ± 4.6
65.2 ± 5.7
59.2 ± 4.7
55.8 ± 11.0
66.8 ± 6.8
66.2 ± 3.5
70.2 ± 12.4
68.1 ± 7.7
62.0
64.0
Olefins
16.1 ± 5.0
13.5 ±1.3
11.1 ± 3.2
13.9 ± 2.7
21.4 ± 9.1
14.2 ± 3.8
10.3 ±1.4
12.4 ± 5.8
10.0 ± 6.5
11.3
10.7
Aromatics
17.8 ± 6.7
23.3 ±4.1
21.1 ± 3.2
23.3 ± 2.0
20.6 ± 8.7
16.7 ± 6.3
20.7 ±-2.8
16.5 ± 12.0
19.2 ±4.4
22.7
24.4
NMHC:
NOX
15.9 ± 16.6
— —
11.9 ± 6.1
— —
10.5 ± 1.6
39.3 + 13.0
14.9 ±9.0
15.1 ±8.4
—
17.0
aBetween suburban and rural, outside the northern perimeter of Houston.
6-126
-------
TABLE 6-50. EFFECT OF WIND DIRECTION OfcLHYDROCARBON
COMPOSITION AT SITE 03, HOUSTON1^
(ppb C)
Time,
CDT
0600-0900
1300-1600
0600-0900
1300-1600
Total
NMHC
concn.
September
1205.6
(49)a
694.7
(40)
September
1005.3
(60)
504.7
(48)
Paraffins
18-20, Prevail
708. 5
(40)
377.3
(35)
25-27, Prevail
557.9
(50)
293.9
(40)
Olefins
ing winds
194.5
(60)
117.3
(45)
ing winds
166.7
(67)
72.7
(64)
Aromatics
90°-104°
302.6
(64)
200.1
(45)
0°-60°
280.7
(68)
139.0
(57)
Percentage of emissions attributable to vehicular emissions
is given in parentheses.
6-127
-------
Concentrations of two specific hydrocarbons are of special interest
because of their adverse welfare and health effects, respectively. Ethylene
concentrations in various parts of the country are of particular interest
since the 1970 criteria document reviewed the phytotoxic effects of this
compound. Concentrations of ethylene are reported for several cities,
including Houston, in conjunction with discussion of its phytotoxic properties
in the section on welfare effects (Section 6.7)."
Concentrations of benzene in ambient air are of concern because of its
reported toxicity in humans. As noted earlier in this section, benzene
concentration data are sparse among the data obtained from field studies
directed toward precursor/oxidant relationships. Since it is considered
photochemically nonreactive, and since sampling and measuring it call for
specialized techniques, benzene was not measured routinely in the special
studies drawn upon in this section, (The data for Wilmington, Ohio, are an
exception and do include benzene because of the GC techniques used in that
particular study.) Some published data are available, however, that indicate
the range of concentrations found in urban air.
In the Wilmington, Ohio, study by ESRL, cited earlier, benzene concentra-
tions ranged from 0.8 ppb to 4.5 ppb over 24 hours (July 18, 1974), with an
average of 2.3 ppb. Toluene concentrations measured at the same time ranged
from 1.0 ppb to 6.8 ppb, with an average of 3.5 ppb. Westburg's rural air
data showed the benzene and toluene concentrations to be 0.58 and 0.19 ppb,
respectively.110 Benzene concentrations in Toronto, Canada, were reported in
193
1973 by Pilar and Graydon to average 13 ppb C, with a maximum of 98 ppb C.
Variations in benzene concentrations with the location of sampling sites were
6-128
-------
linked with traffic patterns by these authors, who also concluded that
automobiles accounted for virtually all benzene contamination and most of the
193
toluene contamination. They attributed the relatively high toluene/benzene
ratio in Toronto to the direct evaporation of gasoline, rather than to auto
-| nq
exhaust emissions alone. Toluene/benzene ratios in auto emissions were
found in 1966 to range from 1:1 to 2.8:1, with an overall average of 1.8:I.194
A study in Los Angeles in 1968 showed toluene/benzene ratios there of 1.4:1 to
4:1, with an average of 2.5:I.26 The ratios in Toronto ranged from 1.5:1 to
4:1, for an average of 2.4:1, which is close to results from Los Angeles. The
toluene/benzene ratio from Wilmington, Ohio (a rural area), is 1.5:1. The
ratio of toluene to benzene obtained from Westberg's measurements110 of ambient
air immediately outside a forest canopy is 0.33:1, which reflects the effects
of residence time, during transport to remote areas, on the disappearance of
toluene that results from its greater reactivity relative to benzene. Disparities
between auto exhaust ratios and ambient air ratios may indicate other souces
of toluene (or benzene) such as gasoline evaporation or solvent losses.26
Concentrations of benzene in ambient air were measured at four California
sites in summer 1975 by Mayrsohn et al.195 Average concentrations compiled
from their data are shown in Table 6-51.
West German researchers determined the effect of reducing the lead
content in gasoline, with necessary concomitant increases in aromaticity, on
ambient air levels of benzene and toluene. Lead in gasoline was reduced in
West Germany from 0.40 g/liter to 0.15 g/liter on January 1, 1976. Data from
sixty 15-minute air samples taken in 1975, before the reduction, showed a
concentration range for benzene of 7 to 171 ug/m3 (2.3 to 57, ppb C), with a
6-129
-------
QX
HH
^f
t—
•z.
LU
1— i
en
^£
z to
HH tn
LU en
Z rH
LU
M OC
Z LU
LU s:
CO "Z.
LU CO
o
co co
Z LU
CD 1—
I-H HH
r- CO
fV^ ^f^
1— rH
z z
LU o;
0 O
•Z. LU
O rH
0 -J
LU O
C3
ill CD
> LU
CM
|| CO rH
II II
C. C 1 C.
1
<^- CD CM
• • •
vj ^i"> • CO
CM rH CM
,— ^
en /"^s
CM Lf)
, it 1L ,
1 \~s ^~s |
1 1
en «=i-
• •
en Lfj
rH rH
^Q
^^N ^^^ ^^^
CO ȣ> ^> CO
rH CO f>s rH
t It It it
^^ v^^/ SM^ s^y
ro O CO CM
• • • •
cn CD oo co
CM Kf CM CO
r>
/*"N ^*~N /^**> /^*N
ljD r** rH LO
rH CO rH rH
II II II II
C C C C
s«— ^ "v-^ ^— ^ S»^
rH r*. CO rH
• • • •
«^ cn en co
CM CM rH CO
LO
LO ^s
r^ co
^> rH
LO ">s.
CM r«- LO
to ^x i r^.
r~~ en LO ~^^
^*«x I t^* rH
CO LO •>>. \
>^ r^. o co
en ^v. i— i i
i en ~~-s LO
LO CM 10 r^
"^^ r^ •» en
**J" *** *^^
^x* *» O ^O
CO CU tO
-P O) -
c OQ ~a
to o c.
05 S C35 ro
=J C •—
N r— O Q.
<: LU —i ID
LO
en
rH
(0
•P
CU
•
C 0}
.C C
0 0
0) •!-
S- ^
^> tO
fO >
2: i-
ai
e o>
C3 f^
S- 0
1-
M-
-a o
0)
r- S-
•r- CU
^*> -^2
£ £
b 3
0 Z
ro .a
6-130
-------
mean of 16 ppb C; and a range for toluene of 9 to 355 |jg/m3 (2 to 95 ppb C),
o
with a mean of 100 ug/m (27 ppb C). A series of three studies conducted in
1976 after the reduction in lead content produced the data shown in Table 6-52.
The conclusion reached was that increases in aromatics in gasoline did not, in
the 6 months immediately following the reduction in lead content, result in
higher toluene or benzene levels in ambient air.196
TABLE 6-52. MEAN VALUES AND RANGES IN AROMATIC
HYDROCARBON CONCENTRATIONS IN-WEST GERMANY, JULY 1976
(ug/nO
196
Compound
Benzene
Toluene
aTo convert
No.
samples
25
12
29
25
12
29
ug/m to ppm,
Sampling
time, min.
30
15
5
30
15
5
divide by 3175
Range
24 to 84a
28 to 54
17 to 144
57 to 171
60 to 111
40 to 328
for benzene and
Mean
46
43
55
101
93
113
3780 for
Ratio of
means
101/46 = 2.
93/43 = 2.
113/55 = 2.
toluene.
2
2
0
The NAS document on ozone and other photochemical oxidants9 depicted
graphically the similarities and differences between the hydrocarbon profiles
of urban air in the nation as a whole and in the nation with California excluded;
and the profiles in auto exhaust and in gasoline. That graph, somewhat modified,
is presented as Figure 6-8 to facilitate the same comparison.9
6.5 HYDROCARBON-OZONE RELATIONSHIP
The chemical relationship between precursor hydrocarbons and ozone and
other photochemical oxidants was summarized in Section 5.1 and updated in
Section 6.1. Early efforts to quantify the relationship of hydrocarbons and
6-131
-------
$
1
n-ALKANES
BRANCHED
CYCLIC
ACETYLENICS
ALKYL-
—
POLYNUCLEAR
^ALKENES
A2ALKENES
A3ALKENES
A4ALKENES
DIOLEFINS
OLEFINS
STYRENE,
INDENE.
TERPENES
NO. CARBON ATOMS
5 10 15 20
III II III I I I I I I
fe^M^m^^mM^J
:S:Ji:5i:i$;:Sf:i:S:i:i:i::::::::::::::::::::::::
<^%^M
HH22H22a
^SSSiSiSsSS:^
y///////s//////////////////A
!%™%^
Wfa
i
W/////////W/////A
'/////////////A
- •- n
'&&ffifS£&tfi!.
ty////////
V/////MW//A
W//////M
////X ^
K^
/////
fflffifri
•X'X*ivi'
VA
y//2
ii i i i i i | i i i i
5 10 15 2(
CARBON NUMBER
| GASOLINE
I EXHAUST
' (GASOLINE-
FUELED)
URBAN AIR
IURBAN AIR
3 EXCLUDING
CALIFORNIA
Figure 6-8. Occurrence of gas-phase hydrocarbons,
by carbon number, in gasoline, exhaust, and urban
ambient air.9
6-132
-------
oxidants were summarized in Section 5.5. That section included a brief review
of the aerometric data base and of the underlying assumptions that were used
to derive the Appendix J model for relating NMHC emissions to oxidant air
quality via NMHC ambient air concentrations. The relationship between hydrocarbons
and photochemical oxidants was perceived in 1971, when the existing hydrocarbon
standard and the Appendix J model were promulgated, to be such that the attainment
and maintenance of a given level of ambient air NMHC (an NAAQS) would ensure
the attainment and maintenance of the 1971 photochemical oxidant standard.
The empirical data that formed the basis of the "envelope curve," or "upper-limit
curve," used to derive the 1971 NAAQS for hydrocarbons, and also used to
derive the Appendix J model, did not depict an absolute relationship between
hydrocarbon ambient air concentrations and resulting oxidant concentrations.
\
Rather, the upper-limit curve was thought to demonstrate that maximum oxidant
concentrations in ambient air were primarily a function of 6:00 to 9:00 a.m.
concentrations of NMHC in ambient air. Thus, when the oxidant standard was
set at 0.08 ppm maximum 1-hour average, not to be exceeded more than once per
year, the NAAQS for hydrocarbons was set at 0.24 ppm C, the level shown by the
upper-limit curve to be the 6:00-to-9:00 a.m. level that would ensure the
production of no more than 0.08 ppm of ambient air oxidant (maximum 1-hour
average).
On its promulgation of a new standard for ozone in February 1979, EPA197
determined that the Appendix J model "no longer represents an acceptable
analytical relationship between hydrocarbons and ozone. Appendix J is, therefore,
being deleted. EPA will now allow states to use any of four analytical techniques
to determine the amount of hydrocarbon reduction necessary to demonstrate
attainment of the national ozone air quality standards:..." In taking this
6-133
-------
regulatory action, EPA has made it clear that the "upper-limit curve," which
formed the empirical basis for the Appendix J method, does not adequately
describe the relationship between emissions of NMHC and ambient air con-
centrations of NMHC, and between ambient air concentrations of NMHC and
ambient air concentrations of ozone and other photochemical oxidants.-
The replacement of the Appendix J method and the setting of the new
standard for ozone at 0.12 ppm maximum 1-hour average, not to be exceeded more
than once per year, both impact strongly on the existing NAAQS for hydro-
carbons. The-second fact stated at the beginning of this issue paper must be
discussed, at least in a general way, in order to provide am adequate basis
for a subsequent regulatory decision on the NAAQS for hydrocarbpns. That fact
is: According to present knowledge, no fixed level of nonmethane hydrocarbpns
can be used nationwide to ensure the attainment and maintenance of the ozone
NAAQS. In discussing that fact, this section (1) surveys the factors that
affect the hydrocarbon-ozone relationship; and (2) briefly discusses the
models promulgated by EPA for use in determining reductions needed in
precursor hydrocarbon emissions in order to attain and maintain the NAAQS for
ozone.
The fact that the original NAAQS was promulgated as a control strategy
for the attainment and maintenance of an oxidant standard rather than an ozone
standard does not affect this discussion and will not be addressee! here. The
rationale for promulgation in 1979 of an ozone standard rather than an oxidant
standard and the use of hydrocarbon emission reductions as appropriate and
effective strategy for the control of ozone and other photochemical oxidants
have been fully explained and documented in Air Quality Criteria for Ozone and
Other Photochemical Oxidants and in the notice of rulemaking on ozone.
6-134
-------
6.5.1 Factors Affecting the Hydrocarbon-Ozone Relationship
As summarized in Sections 5.1 and 6.1, the presence of ozone in ambient
air results largely from a series of multiple, complex photochemical reactions
involving nitrogen oxides and numerous organic compounds, including hydrocarbons.
In the absence of appreciable amounts of organic compounds, resulting levels
of ozone remain low as the result of establishment of chemical equilibrium
among ozone, nitric oxide, and nitrogen dioxide. In the presence of appreciable
amounts of organic compounds, this equilibrium is disturbed and larger amounts
10 198
of ozone are produced. '•L3° The basic chemistry involved was summarized in
simplified form in Sections 5.1 and 6.1. A basic understanding of the chemical
relationships among hydrocarbons, nitrogen oxides, and ozone is a prerequisite
for the development of techniques that relate precursor emissions to ozone
concentrations in ambient air. Three relatively recent documents have
presented thorough discussions of the chemical relationships between ozone and
7_Q
its precursors. All three also discussed the meteorologic and geographic
factors that affect the photochemical formation of ozone, and two of the three
reviewed the approaches used to describe quantitatively those relationships
for various predictive applications.7'9 The relative importance of the
respective factors—chemical, meteorologic, and geographic—that influence the
formation of ozone was evaluated in the cited documents and will not be
covered here.
The simple enumeration given in Table 6-53 of the main factors that affect
the NMHC-ozone relationship produces a recognition of the complexity of that
relationship that is heightened by the realization that these factors are all
interactive. Examination of the number and kind of factors that affect ozone
production reveals that the relationship between ambient air concentrations of
6-135
-------
TABLE 6-53. SOME FACTORS THAT AFFECT THE
PHOTOCHEMICAL FORMATION OF OZONE FROM PRECURSORS
Chemical factors
NMHC and NO emissions (starting reactants.)
Rates/source strength
Spatial variations (location, point or area source)
Temporal variations (diurnal, seasonal, year-to-year)
Species of NMHC emitted
Reaction rates
Reaction products
Stoichiometry (main and side reactions)
Ratio of NMHC/NO ; ratio of N02/N0
Presence of other reactants (inducing oxygenated HC)
Presence of background levels of NMHC and NQX
Presence of scavengers/sinks
Meteorological factors
Advection
Mixing heights.
Turbulent diffusion
Frequency/duration of inversions
Prevailing winds
Direction
Speed
Humidity
Temperature
Sunlight intensity
Diurnal
Seasonal
Wavelength distribution
Geographi c/topographi c
Longitude and latitude (both affect sunlight intensity, given above)
Altitude
Terrai n
Ventilation characteristics
Sinks
6-136
-------
hydrocarbons and those of ozone cannot, in a given locality much less
nationwide, be described by a simple proportionality; that is, hydrocarbon
ambient air concentrations are not constant vis-a-vis ozone concentrations.
Emissions of NMHC and NOX and background, or starting, ambient air concentrations,
as well as the ratio of NMHC to NOX, differ from place to place, from city to
city. Likewise, the spatial and temporal distribution of these sources of
differing strengths varies from city to city.
In addition to those basic chemical factors that affect the photochemical
formation of ozone, meteorological factors play an important role. The
dependence of ozone concentrations on NO and hydrocarbons (and other
J\
organics) can be dominated by meteorological conditions.198 Some of the
meteorological variables, such as sunlight intensity and its spectral
distribution, directly influence the rates of the ozone-forming photochemical
reactions, thereby influencing equilibria among reactants.198 Geographic
.location influences photochemical reactions inasmuch as sunlight intensity
\
varies with latitude. Temperature, the net product of geographic and meteoro-
logic factors, directly influences photochemical reactions; summer temperatures,
for example, are conducive to ozone formation throughout at least the contiguous
United States.7
A number of meteorologic and climatic factors other than temperature and
sunlight intensity exert major influences on the atmospheric formation of
ozone. For example, mixing height and its diurnal variations determine the
extent to which local emissions disperse and the extent to which transported
ozone and its precursors are likely to affect local ozone concentrations.198
Atmospheric mixing inhibits the accumulation of precursors and the photochemical
formation of ozone and other oxidants.7 The interaction of locally emitted HC
6-137
-------
and NO with transported pollutants is influenced also by prevailing wind
J\
speeds and wind trajectories. Geographic factors that affect the hydrocarbon-
A. .
ozone relationship are chiefly those that affect the meteorology of an area,
such as variations in insolation with latitude. Geography and topography are
the main factors in phenomena such as the seabreeze-landbreeze patterns observed
in the South Coast Air Basin of California.7 Topographic influences such as
mountains, canyons, and valleys are responsible for unique ventilation patterns
that can serve to reduce ozone levels in some areas and to increase them in
199
other areas.
The factors listed in Table 6-53 and surveyed here are not all-inclusive,
but they are representative of the complex influences that affect the hydrocarbon-
ozone relationship such that the reductions in hydrocarbon .emissions needed to
attain the NAAQS for ozone will differ from city to city, from season to
season, and from year to year.
One factor listed in Table 6-53, the differing reactivities of individual
hydrocarbons, is worthy of some detailed discussion inasmuch as it bears upon
the question of whether measurements of ambient air concentrations of total
NMHC have utility in the determination of needed reductions in hydrocarbon
emissions; and because it further emphasizes the added complexity of the
oxidant-precursor relationship that is introduced by the observed spatial and
temporal variations in the composition of hydrocarbons in the ambient air.
The fact that organic substances, including hydrocarbons, differ widely
in their respective reactivities is significant for understanding atmospheric
photochemical oxidation processes. This variation in reactivity is also
significant because it introduces selective organic compound control as a
possible refinement of the practice of controlling all organics to the same
6-138
-------
extent, irrespective of their reactivity. Variations in reactivity are
particularly important conceptually with respect to an ambient air standard
for hydrocarbons.
That individual hydrocarbon species participate to differing degrees in
the formation of ozone and other secondary products (aldehydes, ketones,
organic aerosols, etc.) was acknowledged in the 1970 criteria document for
hydrocarbons:
As a consequence of the differing reactivities of individual
hydrocarbons, it is impossible to predict accurately the rate of
consumption of hydrocarbons in photochemical air pollution unless
the detailed composition of the hydrocarbon component is known or
can be estimated. Knowledge of the total concentration of hydrocarbons
is insufficient, since two atmospheres having the same total hydrocarbon
measurement may contain individual hydrocarbons of very different
reactivity and thus exhibit very different rates of hydrocarbon
consumption and photochemical air pollution development.
Conceptually, then, a national ambient air quality standard for
hydrocarbons is of limited value in controlling photochemical oxidant levels
in ambient air. The measurement of total hydrocarbon mass in ambient air does
not describe the potential of that atmospheric loading of hydrocarbons for
producing ozone or other secondary smog components. Because of the differing
reactivities of individual hydrocarbons, ambient air measurements of total
NMHC may have limited utility for predicting the impact of hydrocarbon emissions
on photochemical oxidant levels, although day-to-day variations in the composition
of the hydrocarbon mix within a given city, at a given site, are likely to be
much less than the differences that exist between sites in the same city or
between different cities.
6-139
-------
Reactivity data on hydrocarbons have been obtained from smog chamber
studies in which a given hydrocarbon or hydrocarbon mixture is mixed with
clean air, NO, and N02 at prescribed concentrations and irradiated with
artificial sunlight.7'73 Experimental conditions for such measurements are,
to the extent feasible and practical, similar to the conditions typically
present in polluted ambient atmospheres. Thus, the reactant concentrations,
the intensity and spectral distribution of the radiation, the temperature, and
often the relative humidity are comparable to actual summer atmospheric
conditions. Because residence time and irradiation time in the atmosphere are
prolonged in situations in which the hydrocarbons are transported from the
source area, separate hydrocarbon reactivity data must be obtained for
simulations of transported air masses. Most available reactivity data are
applicable only to urban situations in which the hydrocarbons react for a few
hours and cause oxidant pollution in the vicinity of their sources.
7,73
Reactivity data applicable to urban situations were compiled by Altshuller
and are presented in terms of a .reactivity classification in Table 6-54. ..
Examination of this table indicates that all hydrocarbons that are precursors
to peroxyacyl nitrates, formaldehyde, or aerosols are also precursors to
ozone. Consequently, irrespective of their reactivities with respect to the
various products listed, control of these ozone precursors should result not
only in the abatement of ozone but also in the abatement of these other products
as well. As this table indicates, however, not all ^hydrocarbons are equally
reactive in the formation of the respective smog components or effects of
smog. Note also that aldehydes, which are both primary and secondary
pollutants, are as reactive as many of the hydrocarbons in the production of
ozone.
6-140
-------
o
o
CM
to
O
«=c
C3
ce.
o
I i
*-'---
o
to
UJ
Q-
>-
f"~
1—
UJ
Q*
LU
U_
UL.
1—4
a
1 1
o
to
1 1 i
i— i
i—
t—t
>
I— 4
h-
O
• fHH
CO
CU
•a,
o
•p
-P
u
S- T-
O) -P
> u
o ro
cu
s-
-P CU
C O)
ro re
•— E
a. ro
"O
,
ro
CU -P C
•2TC.2
S~ -P
p— .
in
•p
u
•o
o
s-
n_
o
10
o
CU
i— CU
tO TJ
O CU
u_ -a
1 0)
x r— ~ re
0 >> S-
S-- U -P
cu re -r-
O_ C
S- .p
o c
ro
i CU Q.
O -P N
'i— i S S i
O
CO
co
o
*•
CM
CM
^j
Q
^"
"
S- CU
cu c
-P N
O C
cu
•a jo.
re ">>
^«— • ^ ^^
^_
cu re
c o
3 O
'o ^
I—
in
0
c
O)
u,
CO JD S
to in
CO CO
1 0 1
CO O O)
1 I-H +
*«>J
t^t i*^n
rH r-H <*•
1 1 1
r»» co CM
CD
*J3 CSJ r~l
f?J- CD LO
0 0
UD U? U>
1
^~
^)
^£
'tO
•f^
S-
•p
t
"O
3 re in
to in cu
CU C 1C
C -i- i— CU
CO <*- >, N
-^ cu .*: c
^^ ^~" ^^ CU
•*TT CD CO ^^
1 •!— • •!— '
rH O O
1
CO 1
O CU
CO CU
1 +-
o o
rH Q
U3 ^
-
to cu
1 +
CD
i—| >
3-1- CU
O 4^ TD
^3 CU r~~*
>— ro
r— O
ro cu -P
£•§ 5
OJ O Q.
-1-5 _P
O. D3 CU C
c x eu
CO CU *r-
ffl 1 •!-
JZ S- rH 4-
U CU 4-
1 -P J= 3
•O t- O3 in
cu ro 3 c
J= O *r-
u s- s-
c 3 j= re
re u -p -P
s- u ro
-Q O O> T3
"O ^>> CU *P
C re • r— 3
ro E CU 2>)-Q
r— CL
c to .a o -
•r- -p ro S- >>
ro U i— D-r-r
x: cu «r- .—
u 4- ro c ro
1 4- > O -P
•p cu ro c
.c in cu
CB s- re -P E
•r- O -P C •!-
ro ro cu s-
s- in -o E cu
•P "O CU Q.
W i— r— S- X
cu re 3 cu
s- ••- +* m
cu >> c ro -a
> cu cu cu
0 r- _E E -P
T7 re S- in c
CU E CU O)
Oi in Q.T3 +J
re x 3 u
S- >> CU i— CU
CU S- U 4-
> CU O C 4-
«=C > ^. HH LU
re ^ o ~a cu
6-141
-------
Results of systematic studies conducted by five groups of researchers*
were used by EPA in classifying organics into the three reactivity classes
shown in Table 6-55.7 While this table does not give numerical ratings for
the respective organics, it does illustrate the differing reactivities of
individual hydrocarbon and other organic species. Furthermore, it demonstrates
the differences in reactivity that exist, though not always systematically,
between compounds as a result of structural characteristics, including degree
and kind of substitution. For example, benzene shows low reactivity, but the
insertion of a methyl group results in toluene, which shows high reactivity.
Likewise, methanol shows low reactivity, and ethanol is moderately reactive;
but 2-ethoxy-ethanol is highly reactive. Partially halogenated paraffins show
low reactivity, but partially halogenated olefins demonstrate high reactivity.
These differences in reactivity can be useful in permitting selective or
preferential emission control as part of control strategies if the photochemical
behavior of a mixture of organic pollutants is consistent with—and can be
predicted from—the behavior of the individual components. Overall results
of smog chamber studies tend to support the view that the behavior of the
mixture can be predicted from the behavior of the components. The main effects
on ozone/oxidant production, however, of controlling highly reactive species
to a greater degree than moderately or slowly reactive species are: (1) a
delay in production of a daily oxidant maximum; and (2) a reduction in the
peak oxidant concentrations as a result of increased time for dispersion.
^Japanese Environment Agency; Bureau of Mines, U.S. Dept. of Interior; General
Motors Corp.; Battelle Memorial Institute Laboratories-Columbus; and Shell
Oil Co.
6-142
-------
TABLE 6-55. CLASSIFICATION OF ORGANICS WITH RESPECT TO OXIDANT-RELATED
REACTIVITY IN URBAN ATMOSPHERES7
Class I
(Low reactivity)
Class II
(Moderate reactivity)
Class III
(High reactivity)
C-^-Cg paraffins8
Acetylene3
Benzene
Benzaldehyde
Acetone
Methanol
Isopropanol
Tert-alkyl alcohols9
Methyl acetate9
Methyl benzoate
Ethyl amines
N, N-dimethyl
formamide
Perhalogenated
hydrocarbons
Partially halogenated
paraffins
Mono, dichlorobenzenes
Methyl ethyl ketone
Tert-monoalkyl benzenes
Cyclic ketones
Tolualdehydes
Tert-alkyl acetates3
2-Nitropropane
C4+ paraffins, cyclo-
paraffins
Ethanol
Prim, sec C2. alkyl
N, N-dimethyl acetamidec
n-alkyl C5+-ketones
Prim-, sec-, monoalkyl
benzenes
Dialkyl benzenes
Styrene
N-Methyl pyrrolidone
Partially halogenated
olefins
Aliphatic olefins
Tri-, tetra-alkyl
benzene
Methyl styrene
Branched alkyl ketones
Unsaturated ketones
Aliphatic aldehydes
Diacetone alcohol
Ethers3
2-Ethoxy-ethanol
Currently classified as not photochemically reactive under Los Angeles
County Rule 66 and similar regulations.
6-143
-------
The consequences of "carryover" of organic pollutants to the second day,
coupled with transport processes, are under intensive investigation. As yet,
few data are available on the reactivities of aged emissions during second-day
irradiations. The greater part of unreacted organic precursors will escape
the photochemical process during the first solar day and will be transported
downwind where, if sufficient NO is present on the second day, ozone will be
, , ^ - /^
produced from the NO and organic precursors. The few reactivity data available
J\
for "transport" situations indicate that under transport conditions the effective
range of reactivities is narrower than for the urban situation.
6.5.2 Models for Determining Hydrocarbon Emission Reductions
In the years following the 1971 promulgation of the Appendix J method,
experience has corroborated the theoretical deficiencies of that method. The
Agency has now officially recognized the deficiencies of the Appendix J method
by replacing it with other models for determining the HC emission reductions
needed to attain the ozone NAAQS. The revisions to 40 CFR, Part 51, that
accompanied the notice of promulgation of the new ozone standard in February
1979 stated that:
...Appendix J is being replaced by four analytical techniques. States
must use one of the four techniques to determine the amount of hydro-
carbon reductions necessary to demonstrate attainment of the national
ozone standard. The four techniques include: (1) Photochemical dis-
persion models, (2) Empirical Kinetics Modeling Approach (EKMA),ig£3)
Empirical and Statistical Models, and (4) Proportional Rollback.
EPA further stated in the same notice of promulgation that the use of photo-
chemical dispersion models was specified in certain instances:
EPA requires that States attempting to demonstrate attainment and
maintenance of the revised ozone standard by 1982 without adopting
reasonably available control technology (RACT) regulations for large
hydrocarbon sources must employ photochemical dispersion modeling.
6-144
-------
198
The use of other less rigorous analytical techniques are only
acceptable in areas where RACT measures are also scheduled for
implementation. Where States are unable to demonstrate attainment
by 1982, EPA believes any o-f the models are useful for indicating
the magnitude of the ozone problem and for identifying the need for
major cjQtrol programs to be implemented over the next several
years.
The techniques prescribed by EPA have as their chief objective the determina-
tion of reductions needed in hydrocarbon emissions to attain the ozone NAAQS.
The promulgation of these four models as required options to be used by
the states resulted from the efforts of an EPA working group that was formed
in 1975 to investigate the viability of alternatives to the Appendix J method.
Suggestions made as a result of working group sessions were reviewed periodically
by a group of experts outside of EPA. The conclusion reached as a result of
these efforts is that it is not possible to recommend a single approach for
198
all applications. The variety of applications, complexity of individual
situations, and differences in data availability and resources all preclude
-| no
use of a single procedure nationwide.
The four models prescribed in February 1979 as alternatives to the Appendix
J method are described briefly below.
1. Photochemical Dispersion Models. These are air quality simulation models
and are based on the most accurate available physical and chemical principles
TQ7
underlying the formation of ozone. Photochemical dispersion models have
the greatest potential for evaluating the effectiveness of oxidant control
strategies mainly because (1) the model permits spatial and temporal resolution
and (2) it can relate precursor emissions directly to ambient air ozone con-
centrations via atmospheric dispersion equations and chemical mechanisms.
Data requirements for these models may be extensive, however.
'198
6-145
-------
2. Empirical Kinetics Modeling Approach (EKMA). This model represents a
simpler alternative to the photochemical dispersion models. It includes the
use of a kinetic model that represents a detailed sequence, derived from smog
chamber data, of the chemical reactions that occur when a mixture of propylene,
n-butane, and NOV is irradiated.198 While retaining some of the vigorous
"""" *\
treatment of chemical and physical principles that is characteristic of photo-
chemical dispersion models, it does not require the extensive input data
197
needed for those models.
There are two variations of EKMA. The first involves the use of city-
specific ozone isopleths. The second involves the use of a standard set of
isopleths in which fixed assumptions have been made about sunlight intensity,
atmospheric dilution rate, reactivity of emissions, and diurnal emission
patterns. The city-specific approach is preferable because (1) it can be used
to evaluate a wider variety of control measures, including measures that are
initiated concurrently; and (2) it employs locally unique assumptions about
atmospheric dilution rate, sunlight intensity, and diurnal emission patterns.
The second advantage is not as crucial as it might appear, however, and using
the standard isopleths has the advantage that fewer input data are needed and
198
a computer is not needed.
EKMA makes use of a kinetics model to express maximum afternoon ozone
concentration as a function of morning ambient air levels of nonmethane hydro-
carbons (NMHC) and NO.,. EKMA is empirical in that it requires the use of
f\
observed second-highest hourly ozone concentrations and morning NMHC/NO
/\
ratios to estimate control requirements.
117
6-146
-------
3- Empirical and Statistical Models. These models are derived from observed
relationships between ozone and other variables.197 Two features of empirical-
statistical models are central. First, any one statistical relationship is
site-specific. Second, the use of these models is most appropriately limited
to applications involving moderate changes from the base control state, since
the imposition of drastic changes could appreciably alter the functional
relationship that was derived from data observed during the base set of control
1 Qft
conditions.
4- Proportional or Linear Rollback. Linear rollback assumes a linear relation-
ship between hydrocarbon emissions and ambient air concentrations of ozone.121
Other proportional rollback models embody the principle that reductions in
emissions are reflected by improvements in air quality, as shown by a curved
line or a complex surface that expresses some proportionality relationship.6
Rollback models work best if the geographic and temporal distribution of
emission sources is not changed. The linear rollback model has two chief
advantages: (1) it uses aggregated emissions data and (2) it is mathema-
tically simple. It has two chief disadvantages: (1) ideally, it should be
applied only to pollutants that do not undergo chemical transformations in the
atmosphere; and (2) it does not take into account the dependence of NMHC-ozone
relationships on the initial or base period NMHC/NOV ratio.198 Linear rollback
A
is less: data-intensive than EKMA and is a possible alternative for estimating
upper and lower bounds on control requirements for hydrocarbon and other
TOO
organic precursors.
A limitation of these and other available techniques is that most of
these models have received little validation. All of the above techniques,
including photochemical dispersion models, must employ simplifications to keep
6-147
-------
them from being intractable. The accuracy of methods incorporating such
simplifications can be determined or at least estimated by verification studies
that compare model results with observed data. There are few locations,
however, for which the data base is presently adequate for the verification of
198
analytical methods relating precursors and oxidants.
.-I
Each of these models has its own characteristic limitations and uncertain-
ties, many of which have not been included here but which are given in detail
in other documents.6>7'73'198 The uncertainties in each model are questions
more of degree than of direction. With respect to limitations, however, it
should be noted that:
...models that relate oxidant to ambient concentrations [of
precursors] rather than to emissions [of precursors] are
subject to the same validity limitation as the Appendix J
method, namely, that the model does not directly relate
air quality to precursors. Instead, the direct relationship
is between oxidant and the atmospheric dispersion factor. '
However, in spite of limitations inherent to each of the models briefly described
here, they appear to have fewer shortcomings and uncertainties than the Appendix J
method. All of the methods indicate that substantial reductions in organic
precursors are needed to reduce ozone levels appreciably. Therefore, the four
198
techniques prescribed by EPA can be regarded as approximations that will be
useful in estimating hydrocarbon emission control requirements.
The Federal Register issuance that promulgated these four techniques
took note of criticisms of the techniques, pointing out that reviewers and
commentators have cited various shortcomings. Among the shortcomings listed
is the fact that the respective models produce different results. EPA acknow-
ledged in that issuance "that the various techniques do produce different
results since different assumptions and different data bases are required for
6-148
-------
each specific model. Also, EPA agrees that control strategies should be based
on the most effective models."197
Present information indicates, then, that no one model can be used in all
urban areas in the United States for all applications. Thus, present knowledge
of available models, along with other scientific evidence, militates against
the concept that attainment of a uniform level of NMHC throughout the country
is either theoretically or pragmatically sound, from the scientific standpoint,
as a means of attaining the ozone NAAQS. This is demonstrated (1) in the
complexity and variability of the factors that influence the photochemical
formation of ozone; (2) in the fact that no one model can be successfully
applied to all situations in all urban areas, coupled with the fact that
different models give different results; and (3) the fact that models relating
NMHC concentrations, rather than NMHC emissions, to oxidant concentrations
really relate those oxidant concentrations to the atmospheric dispersion
factor.
6.6 HEALTH EFFECTS
Since promulgation of the hydrocarbon standards in 1971 the direct and
indirect effects of hydrocarbons on public health and welfare have been
extensively reviewed by EPA in a technical report (1972) entitled, Hydrocarbon
Pollutant System Study, and by the National Academy of Science in a monograph
(1976) entitled, Vapor-Phase Organic Pollutants.10 In 1977, NIOSH published a
criteria document addressing alkanes (C5-C8).202 Most recently (1978), the
Office of Air Quality Planning and Standards published a document entitled,
139
Assessment of Gasoline Toxicity. This document reviews the various hydrocarbons
present in gasoline either as individual compounds, as classes, or collectively
as a group of mixed hydrocarbons.
6-149
-------
A review of the literature since 1970 reveals once again that
hydrocarbons, p_er se, do not present a significant health hazard in the
atmosphere at the present detectable levels. They should be controlled or
restricted on the basis of their contribution to photochemical smog and the
resultant adverse effects of the smog products. To defend this position
properly it seems appropriate to present an up-to-date toxicological
evaluation of the hydrocarbon literature addressing each chemical family
individually. Components that represent major health hazards are identified
and their toxicities are discussed. Because of its objective, this toxicity
assessment reviews only the effects associated with environmental exposure to
hydrocarbons, either as mixtures (e.g., gasoline, solvents) or as individual
compounds, by inhalation.
6.6.1 Aliphatic Hydrocarbons
This class of straight chain or branched hydrocarbons includes the alkanes
(saturated), the alkenes (unsaturated, double bond) and the alkynes (unsaturated,
triple bond). A number of alkanes (methane, ethane, propane, butane), alkenes
(ethylene, propylene), and a single alkyne (acetylene) constitute a group of
inert gases and vapors collectively known as "simple asphyxiants." These
hydrocarbons, when present in high concentrations in air, act as simple
asphyxiants without other significant physiological effects. Any systemic
effects that may be observed are thought to be secondary due to the oxygen-
replacing capabilities of these gases at extremely high concentrations. No
threshold limit values (TLV) have been established for these "inert" substances
by the American Conference of Governmental Industrial Hygienists (1979 listing).
A TLV is not recommended for each simple asphyxiant because the limiting
factor is the available oxygen. The minimum oxygen content should be 18 percent
6-150
-------
by volume under normal atmospheric pressure (equivalent to a partial pressure,
p02, of 135 mm Hg). Atmospheres deficient in oxygen do not provide adequate
warning and most simple asphyxiants are odorless.
The recent use of isobutane, butane, and propane as substitutes for
fluorocarbons in aerosol products raises the question of the potential of
these gases to injure humans. Aviado et al.203 addressed this question in a
monograph entitled, Non-Fluorinated Propel!ants and Solvents for Aerosols.
In this monograph the literature was thoroughly reviewed in an attempt to
determine whether the mixture of three hydrocarbons (80.4 mol % isobutane, 25
mol % butane, 17.1 mol % propane) is more toxic than each of the individual
components. The threshold effective concentration (TEC, based on myocardial
and hemodynamic responses) for the hydrocarbon mixture (A-46) was 1.9 percent
by volume of air (19,000 ppm) compared to propane (TEC .3.3 percent), isobutane,
(TEC 2.0 percent), and n-butane (TEC 0.5 percent). The propellent mixture
A-46 produces the same pharmacologic effect as other hydrocarbons studied and
does not enhance the individual effects of isobutane, butane, and propane.
From a health point of view, isobutane is to be preferred over fluorocarbons
as a predominant component of aerosol propel!ant mixtures because it has a
higher threshold level for eliciting cardiac arrhythmia.
In general, the simple asphyxiants can be tolerated in high concentrations
in inspired air without producing significant physiological effects; however,
Reinhardt et al.204 reported that "aerosol sniffing" of these simple asphyxiants
in some cases has sensitized the heart, to the extent that small quantities of
epinephrine may lead to ventricular fibrillation and ultimately death through
cardiac arrest.
6-151
-------
Generally, alkanes from pentane (Cg) through octane (Cg) show increasingly
strong narcotic properties. With the exception of respiratory irritation and
central nervous system (CNS) depression manifested acutely upon exposure to
high vapor concentrations, the C5 through Cg aliphatic hydrocarbons are regarded
as relatively innocuous. Since exposure to only one alkane is infrequent,
o
NIOSH has recommended a time-weighted average concentration of 350 mg/m as
the occupational limit for pentane, hexane, heptane, octane, and total alkanes
(mixtures). On a volume/volume (v/v) basis, these concentrations are equal to
about 120 ppm pentane, 100 ppm hexane, 85 ppm heptane, and 75 ppm octane.
NIOSH also recommends a ceiling limit or maximum acceptable concentration
(MAC) of 1800 mg/m3 for either single hydrocarbons or mixtures, for 15 minutes
(about 610 ppm pentane, 500 ppm hexane, 440 ppm heptane and 385 ppm octane).
The recommended standard is based on the conclusion that acute intoxication by
these alkanes involves a transient central nervous system depression and that
202
chronic intoxication may involve a more persistent effect, polyneuropathy.
Polyneuropathy has usually been attributed to n-hexane, but occupational
exposures to rrhexane alone have not been described, and the recommended
standard is based on the belief that this neuropathy can be caused by other
alkanes and their isomers as well. Should sufficient evidence be developed
that this is not the case, the TLV limit of 350 mg/m of total alkanes
recommended by NIOSH might be considered for upward revision in the case of
those alkanes and/or isomers not causing chronic neurological disorders.
The effects of alkane (Cr-Cg) vapor exposure on humans and animals are
202
illustrated in Tables 6-29 and 6-30, respectively.
No studies were found to suggest that any of the volatile aliphatic
hydrocarbons are related to carcinogenic, mutagenic, or teratogenic effects in
6-152
-------
humans or experimental animals; nor is there any reason to suspect that they
will be found to produce such effects because these compounds are not chemically
related to substances known to have carcinogenic, mutagenic, or teratogenic
effects. On the basis of limited bioassays, both alkanes and alkenes have
been classified as noncarcinogens, although some of the long-chain (C1? or
greater) aliphatic hydrocarbons have been implicated as cocarcinogens or
tumor-promoters, based on mouse skin-painting experiments.10 The straight-
chain hydrocarbons above octane (Cg) are not sufficiently volatile, however,
to warrant serious consideration as vapor hazards at room temperature. In
1976 the American Conference of Governmental Industrial Hygienists (ACGIH) TLV
committee imposed a limit of 200 ppm on nonane based on analogy with other
t
members of this series and not on actual data. In 1978, Carpenter et al.205
established a no ill-effect level for nonane at 590 ppm based on extensive rat
inhalation studies. This work was in good agreement with work reported by Nau
206
et al. in 1966. Nonane concentrations commonly found in the ambient air
range between 0.0001 and 0.0009 ppm; acceptable workplace exposure levels are
200 ppm (TLV). In view of the low volatility of nonane and its analogy with
other aliphatic hydrocarbons, it would appear that this Cg aliphatic hydro-
carbon poses no danger to human health at these ambient levels.
6.6.2 Alicyclic Hydrocarbons
Toxicologically, the alicyclic compounds, as a class, resemble the pre-
viously discussed aliphatic hydrocarbons in that they act as general anesthetics
and central nervous system depressants with a relatively low order of acute
toxicity. The alicyclic compounds (naphthenes), which are present in modern
gasolines and gasoline vapors in relatively low concentrations compared to
other hydrocarbon components, have not been indicated as hematopoietic toxicants.
6-153
-------
Table 6-56. EFFECTS OF ALKANE VAPOR EXPOSURE ON HUMANS
.202
AT kane Subjects
Exposure
concentration
and duration
Effects
Pentane
Hexane
Hexane
Hexane
Hexane
Hexane
3-6 men and
women
3-6 men and
women
6 men and
women
93 men and
women
11 men and
women
4 men and
women
Up to 5000 ppm
10 min
5000 ppm
10 min
2500-1000 ppm
10-12 hr/d
2500-500 ppm
Hexane
Hexane
3-6 men and
women
3 women
2000 ppm
10 min
1300-650
8-10 hr/d
2-10 months
1000-500 ppm
3-6 months
not available
Heptane 3-6 men and 5000^
women 15 min
No symptoms.
Marked vertigo, giddiness.
Drowsiness in 0.5 hr, fatigue, IQSS pf
appetite in some,' paresthesia in distal
extremities.
Sensory impairment in distal portion of
extremities, muscle weakness in 13, cold
sensation of extremities in some, blurred
vision, headache, easy fatigability, anorexia,
weight loss at onset of pplyneuropathy,
muscular atrophy, demyelination and axonal
degeneration of peripheral nerves.
No symptoms.
Headache, burning sensation of face, ab-
dominal cramps, numbness, paresthesia,
weakness of distal extremities, bilateral
foot and wrist drop, absence of Achilles
tendon reflexes, fibrillation potentials,
decreased conduction time in motor and sensory
nerves, denervation-type injury of muscles,
numerous neurophathologic changes.
Fatigue, anorexia, paresthesia in distal
extremities, muscular atrophy.
Peripheral neurophathy, reduced motor and
sensory nerve conduction velocities, knee-
jerk and Achilles tendon reflexes absent,
muscular atrophy, diminished sensations of heat
and touch, pathologic abnormalities in muscles
and nerves.
Marked vertigo, incoordination, hilarity
for 30 min.
6-154
-------
Table 6-56 (continued).
Al kane Subjects
Exposure
concentration
and duration
Effects
Heptane
Heptane
Heptane
Heptane
Heptane
Octane
3-6 men and
women
3-6 men and
women
3-6 men and
women
3-6 men and
women
3-6 men and
5000 ppm
7 min
5000 ppm
4 min
3500 ppm
4 min
2000 ppm
4 min
1000 ppm
6 min
Marked vertigo, incoordination of space,
hilarity in some.
Marked vertigo, inability to walk straight,
hilarity.
Moderate vertigo.
Slight vertigo.
Slight vertigo.
(No data available.)
6-155
-------
Table 6-57. EFFECTS OF ALKANE VAPOR EXPOSURE ON ANIMALS
202
AT kane
Pentane
Pentane
Pentane
Pentane
Pentane
Pentane
Pentane
Hexane
Hexane
Hexane
Hexane
Hexane
Species No.
Mice 1
Mice 4
Mice 1
Mice
Mice 1
Mice 4
Mice 4
Mice 1
Mice 1
Mice 1
Mice
Mice
Exposure
concentration
and duration
129,200 ppjn
37 min
128,000 ppm
5 min
108,800 ppm
26 min
102,000-
68,000 ppm
2 hr.
91,800 ppm
66 min
64,000 ppm
5 min
32,000 ppm
5tn*i n
in i n
64,000 ppm
5 min
51,120 ppm
9 min
42,600 ppm
127 min
42,600-
34,080 ppm
2 hr
39,920 ppm
127 min
Effects
Decreased respiration rate, loss of reflexes,
death by 37 min of exposure.
Irritation, deep anesthesia, respiratory
arrest in 1 mouse by 4.75 min of exposure.
Lying down, weakened reflexes.
Lying down.
Temporary, lying down.
Irritation, anesthesia during recovery
period.
Anesthesia during recovery period.
Irregular respiratory pattern, respiratory
arrest by 2.5 - 4.5 min.
Death after spasms, no narcosis.
.Loss of reflexes, death.
Death.
Death.
6-156
-------
Table 6-57 (continued).
Al kane
Hexane
Hexane
Hexane
Hexane
Hexane
Hexane
Species No.
Mice 1
Mice
Mice
Mice 4
Mice 4
Mice 7
Exposure
concentration
and duration
34,080 ppm
123 min
32,000 ppm
5 min
28,400 ppm
2 hr
16,000 ppm
5 min
8000 ppm
5 min
1000-2000 ppm
Effects
Light narcosis.
Deep anesthesia.
Lying down.
No anesthesia.
No anesthesia.
Marked abnormal oosture and muscular atmnh\
Hexane Mice
Hexane Mice 19
Hexane Mice
Heptane Mice
Heptane Mice
Heptane Mice
-bWWW f~. \S\J\J |*'|>'II _ _
6 d/wk, 1 yr and degeneration; in electromyographic tests,
fibrillation at rest, complex NMU voltage and
high amplitude NMU voltage during movement, and
weakened interference waves during strong
contractions; increased electrical reaction
time; reversal of flexorextensor chronaxy
ratio.
500 ppm
6 d/wk, 1 yr
250-2000 ppm
6 d/wk, 1 yr
250 ppm
6 d/wk, 1 yr
64,000 ppm
5 min
32,000 ppm
5 min
18,300 ppm
2 hr.
Abnormal posture and muscular atrophy.
Higher strength-duration curve with increased
concentrations.
Slightly abnormal posture and muscular atrophy;
in electromyographic tests, some fibrillation
at rest.
Respiratory arrest in 3 mice by 3.75 min of
exposure.
Irregular respiratory pattern.
Death in 2 hr.
6-157
-------
Table 6-57 (continued).
Al kane
Heptane
Heptane
Heptane
Octane
Octane
Octane
Octane
Octane
Octane
Octane
Octane
Species No.
Mi ce 4
Mice
Cats
(decere-
brated)
Mice 4
Mice 4
Mice 1
Mice
Mice
Mice
Mice
Mice
Exposure
concentration
and duration
16,000 ppm
5 min
9760 ppm
2 hr
24,400-
6100 ppm
5 min
32,000 ppm
5 min
16,000 ppm
5 min
12,840 ppm
185
10,700 ppm
2 hr
8560 ppm
55 min
7490 ppm
2 hr
6634 ppm
1 yr
5350 ppm
48 min
Effects
No anesthesia.
Lying down.
Decreased blood pressure during exposure
return to normal during recovery period;
increased respiration, then decreased.
Respiratory arrest in 4 mice by 4 min of
*
Respiratory arrest in 1 during recovery
, rapid
initial
exposure
period.
Decreased respiration rate, death by following
day.
Loss of reflexes.
Narcosis.
Lying down.
Lying down.
No narcosis.
6-158
-------
These compounds are rapidly metabolized and eliminated in the urine in the
form of conjugated glucuronides and sulfates, thus eliminating the chance of
increased body burden upon repeated exposure to low atmospheric concentrations.
In general, the chronic exposure studies involving animals were very deficient
with respect to a dose-response relationship. No occupational or epidemiological
evidence was found to indicate that naphthenes have any systemic effects.
Unlike the aliphatic hydrocarbons, the degree of toxicity of the alicyclic
hydrocarbons does not correlate with structural characteristics (e.g., number
of carbon atoms, degree of unsaturation, and amount of branching.
Naphthene vapors at very high concentrations cause irritation of the
mucous membranes, with the saturated hydrocarbons generally causing less
irritation than the corresponding unsaturated ones. No studies were found,
regarding the carcinogenic, mutagenic, or teratogenic potential of any
alicyclic hydrocarbons.
6.6.3 Aromatic Hydrocarbons
The primary route of exposure to benzene is by vapor inhalation. High
concentrations of benzene, like most organic solvents, can cause depression of
the central nervous system. Symptoms are acute narcosis, accompanied by
drowsiness, vertigo, nausea, unconsciousness, and death. The effects of acute
benzene narcosis are usually completely reversible unless the initial severity
of exposure causes pathologic changes. Acute poisoning by high concentrations
of toluene or the xylenes is uncommon but, like benzene, both toluene and
xylene can produce effects on the central nervous system.
The health effects of chronic benzene exposure have been addressed in
three recent reviews.207'209 Benzene exposure by inhalation and other routes
is strongly implicated in three pathological conditions; namely, leukemia,
6-159
-------
pancytopenia, and chromosomal aberrations. The most severe long-term hazard
is to the hematopoietic system. Hematopoietic anomalies can affect a number
of blood parameters, including erythrocyte count, hemoglobin, mean corpuscular
volume of red blood cells, platelet counts, and leukocyte counts. Reportedly,
hematologic abnormalities have developed in humans as a result of repeated
210 211
exposure to benzene concentrations ranging down to 105 ppm and 60 ppm. '
rt-i rt
According to Pagnatto's data, suggestive but inconclusive hematologic
changes were noted from exposure to benzene concentrations as low as 20 to 25
ppm in rubber-coating plants.
It is now generally agreed that benzene can cause various forms of
leukemia which can arise with or without a previous history of aplastic anemia;
Knowledge of the carcinogenic potential of benzene has been gained primarily
through the experience of human occupational exposure. The exposure data in
these studies do not allow a scientific derivation of a dose-response curve.
Most studies in which exposure levels were determined involved doses in the
range of 100 to 500 ppm, though in some the benzene concentrations were lower.
Currently there is no convincing evidence that benzene causes neoplasias,
including leukemia, in animals. In animal studies, the most consistent physio-
OHO 91 A. 91 !•>
logical response to benzene has been leukopenia. ' '
The incidence of benzene-induced chromosomal aberrations in peripheral
blood lymphocytes and bone marrow has received considerable attention.
Available data from studies in which measurements ranged from 25 to 150 ppm
strongly suggest that chromosome breakage and rearrangement can result from
chronic exposure to benzene; in at least one study, significant effects were
91 C
noted at 2 to 3 ppm (time-weighted average). A dose-response relationship
has not been demonstrated for benzene-induced chromosome aberrations; however,
213
6-160
-------
the apparent lack of relationship may be the result of variations in
individual susceptibility.
NIOSH, in recognition of accumulated clinical and epidemiological
evidence to the effect that benzene is leukemogenic and a probable cause of
other severe systemic toxicity, has recommended a much more stringent standard
for occupational exposure to benzene.217 Since it is not currently possible
to establish a safe level of exposure to a carcinogen, NIOSH has recommended
that exposure to benzene be kept as low as possible. In accordance with this
recommendation, OSHA established an emergency temporary standard for benzene
in air at 1 ppm, as an 8-hour, time-weighted, average concentration, with a
ceiling level of 5 ppm for any 15 minute period during the 8-hour day.218
This emergency temporary standard for benzene reduction in the workplace was
immediately challenged by industry. In February 1978, this temporary benzene
standard was designated permanent;219 however, this ordered reduction was set
aside by a New Orleans Federal Appeals Court the same year. The latter ruled
that OSHA cannot legally regulate benzene on the basis of health hazards alone
but must include a cost-benefit analysis to determine whether the benefits
expected from the revised permanent standard are in proportion to the costs
imposed. The Supreme Court on July 3, 1980, agreed with the Appeals Court
ruling (5/4 decision) that the benzene rules written by the Labor Department's
Occupational Safety and Health Administration were invalid, although the lower
court's opinion on cost versus benefit was adopted by only one of the five
Justices voting in the majority.
EPA has addressed the effects of benzene exposure in a health document
entitled Assessment of Health Effects of Benzene Germane to .Low-Level Exposure.
In combination with an exposure assessment document entitled Assessment of
209
6-161
-------
Human Exposures to-Atmospheric Benzene,220 the Carcinogen Assessment Group of
EPA, using a linear non-threshold model, estimated the leukemic risk at 90
cases per year upon exposure of the general population to ambient atmospheric
benzene levels (1 ppb). This amounts to between 0.23 and 1.62 percent of the
total leukemia deaths in the United States, based upon 1973 vital statistics.
Toluene (methyl benzene), xylenes (dimethylbenzenes), and trimethylbenzenes
are generally much less toxic and volatile than benzene. The primary and most
hazardous route of exposure to toluene in the workplace is inhalation. At
equilibrium the body retains approximately 50 percent of the inhaled toluene,
20 percent of which is excreted unchanged by the lungs while the remainder is
metabolized via the liver to benzoic acid. The majority of benzoic acid (80
percent) is conjugated with glycine and excreted as hippuric acid in the
urine. The remaining 20 percent of benzoic acid is conjugated with glucuronic
acid in the urine as the glucuronide.
Today there is general agreement in the literature that toluene, unlike
benzene, does not affect the hematopoietic system. The myelotoxic effects
previously attributed to toluene from early studies are presently judged by
updated investigations to be the result of concurrent exposure to benzene as a
contaminant. These updated investigations involve both experimental and
occupational inhalation exposure to pure toluene. EPA's Carcinogen Assessment
Group recently concluded, "there are no adequate studies that describe the
221
carcinogenic potential of toluene."
Effects from toluene have also been reported for other organs, and systems;
namely the nervous system, liver, kidney, heart, mucous membrane, immunologic
system, and the skin. Like many of the early investigations of the effects of
toluene on the hematopoietic system, the reported effects for some of these organs
6-162
-------
(liver, kidney, heart, peripheral nervous system) are inconsistent and may be
attributed to impure toluene and/or inaccurate exposure data. The most
sensitive indicators of toluene exposure are a number of symptoms in the
central nervous system and in the psychomotor functions. The effects of
toluene, however, on the central nervous system are not specific for this
aromatic hydrocarbon but typify the responses to other hydrocarbons.
The majority of reported effects have been observed at high exposure
levels. Very few observations have been made at exposure levels of 0 to 375
3
mg/m (0 to 100 ppm). There are no reports on the dose response relationship
over an 8-hour workday at these concentration levels. The present threshold
limit value for toluene has been established at 100 ppm. A detailed review of
the literature regarding the health effects of toluene has been reported by
222 993
NIOSH and, most recently, by Scandinavian scientists.
No data are available on actual occupational or environmental exposures
to xylenes that warrant their consideration as important environmental health
hazards. The current xylene standard of 100 ppm was designed primarily to
protect workers against the irritating and narcotizing properties of xylene.
6.6.4 Hydrocarbon Mixtures—Gasoline
Inhalation toxicity data based upon human exposure of gasoline vapors are
very limited. Inhalation of extremely high concentrations of gasoline vapor
can cause narcosis, coma, and sudden death. Death upon acute exposure to
gasoline fumes is generally attributed to severe central nervous system depression
terminating in respiratory paralysis. Gasoline vapors can sensitize the
myocardium to the extent that small quantities of epinephrine may precipitate
ventricular fibrillation. This may explain the type of sudden deaths observed
224—228
in cases of accidental gasoline exposure, since the edema observed in
6-163
-------
these individuals upon autopsy did not appear severe enough to be fatal. This
is a reasonable explanation in view of the fact that Chenoweth showed that
both gasoline and numerous constituents of gasoline are known to induce ventricular
229
fibrillation in the presence of epinephrine.
Acute inhalation exposures to milder concentrations,of gasoline vapor are
usually characterized by nonspecific anesthetic or narcotic effects. Symptoms
such as headache, vertigo, blurred vision, ataxia, tinnitus, nausea, anorexia,
230
and muscular weakness are not uncommon. Early studies by Drinker et al.
demonstrated only slight irritation to the eyes, nose, and throat after human
exposure to gasoline for 1 hour. Slight dizziness accompanied the eye irrita-
tion at 2600 ppm, while at 10,000 ppm marked intoxication was experienced
231
after 4 to 5 minutes. More recently, a study by Davis et al. revealed no
manifestations of intoxication in humans exposed to any of three different
unleaded gasolines for 30 minutes at concentrations of 200, 500, and 1000 ppm.
The only significant effect reported was eye irritation at the 1000 ppm level.
The effects of gasoline vapor on humans are summarized in Table 6-58. Little
definitive animal experimentation has been reported to date which assesses the
acute toxicity of commonly used gasolines. Machle43 furnished the only acute
toxicity data with respect to animals exposed to gasoline. He reported that
gasoline vapor exposures in excess of 10,000 ppm rapidly caused death in most
experimental animal species.
Little evidence was found on the health effects of exposure to low con-
centrations of gasoline vapor over long periods of time. Chronic gasoline
toxicity data appear to be limited to a few reports on occupational exposures
and some cases of gasoline abuse. Definitive, well-designed epidemiologic
studies are not available.
6-164
-------
In general, the symptoms of chronic exposure to gasoline vapor are
ill-defined. They may consist of fatigue, muscular weakness, nausea,
vomiting, abdominal pain, and weight loss. Exposure is also known to have
neurological effects which include confusion, atoxia, tremor, paresthesias,
neuritis, and paralysis of peripheral and cranial nerves.232
Cases of repeated self-induced gasoline intoxication, generally involving
higher concentrations of gasoline vapor, could possibly be considered chronic
inhalation exposures. Persons engaged in the regular habit of "gasoline
sniffing" experience loss of appetite and weight, muscular weakness, and
cramps. Other effects reported to result from this practice include abnormal
EEC's and organ damage. The greatest hazard associated with chronic gasoline
inhalation is exposure to the aromatic hydrocarbons, especially benzene. As
previously noted, chronic benzene intoxication ca.p result in severe
irreversible systemic effects such as encephalopathy, aplastic anemia, and
leukemia. .Nothing was found in the literature relating chronic gasoline
sniffing with fetal pathological conditions such as liver, kidney, or bone
poo
marrow lesions. McDermott and Killiany calculated a TLV for a gasoline
based on the hydrocarbon constituents present in gasoline and their respective
TLV values. A time-weighted average exposure of 300 ppm over an 8-hour period
was estimated to be reasonable along with a ceiling exposure of 1000 ppm for
15 minutes. The TLV was calculated as though benzene has additive effects in
the presence of other hydrocarbons. The assumption of additive effects in the
calculation of TLVs has been made by ACGIH for all mixtures containing
potentially harmful hydrocarbons. The reader is referred to the ACGIH
listings of TLVs for a discussion of the basis of TLV calculations for
O"3A
hydrocarbon mixtures.
6-165
-------
O)
u
0)
O)
re
re
•IT1
(A
a>
to
O)
re
re
re
0)
in
S
s- in
•i— IA
0)
UJ
•P
re
O) Ou
o
O)
OJ •« O
u) c -P
ct)
O)
U)
ro
N
O)
E
(A
O) *
(A Ol
Q) •£!
i 1
•o
S tl? vi
O)
•r- O)
in
in
(0
•i- S-
•i- S-
•r- s-
ao oo
co
CM
o
LO
CO
UD
«t
CM
LO
CO
CM
6-166
-------
235
Runion, prior the the Supreme Court decision to overturn the permanent
OSHA standard on benzene (1 ppm), re-examined the practical impact of such a
standard in relation to gasoline exposure. Gasoline TLVs were calculated
using the ACGIH method and incorporating the benzene TLV criterion of 10 ppm,
as well as the criterion of 1 ppm. The other constituents of gasoline used in
the calculations were estimated on actual ACGIH TLV values. The consequence
of'a lowered benzene limit on a gasoline TLV is apparent from Figure 6-9.235
Gasoline containing one percent benzene, for example, had a TLV of 300 ppm,
whereas the same gasoline, under the benzene standard which was recently
overtuned by the Supreme Court, would have had a TLV of less than 150 ppm.
Recently, Shell conducted a study of service station attendants' exposure
to benzene and gasoline vapors in response to the Emergency Temporary Standard
originally issued by OSHA. The overall exposure of service station attendants,
based on 84 time-weighted average results at seven stations were well below 1
ppm except one exposure at 2.08 ppm. Peak exposures sampled over a 15-minute
period did not exceed 1.21 ppm. Attendants' TWA exposures to total gasoline
vapor were 114 ppm or less, with measured 15 peak exposures not exceeding 100
poc
ppm during actual filling operations.
The permanent standard on benzene exposure, which was recently (July 3,
1980) ruled invalid by the Supreme Court in a five to four decision, did not
attempt to regulate exposures to gasoline in service stations.
6.6.5 Miscellaneous Hydrocarbons
In addition to gasoline, there are many other hydrocarbon mixtures known
collectively as "solvents" that are utilized in vast quantities in the United
States. The majority of these contain more than one constituent and have
boiling ranges varying from less than 10°F to more than 100°F. In general,
6-167
-------
0 2 4 6 8 10 12 14
PERCENT BENZENE IN GASOLINE - LIQUID PHASE
Figure 6-9. Calculated gasoline threshold limit Value, reflecting impact of
present vs. newly promulgated TLV standard as a function of the liquid
volume percent benzene in gasoline.235
6-168
-------
the hydrocarbon constituents that make up these solvents consist of aliphatics,
benzene, alkyl-benzenes, and mono- and dieye"loparaffins. The widespread usage
and accessibility of these solvents in industry under legitimate conditions
lead to potential exposure which may be further compounded by those who
intentionally inhale such mixtures for purposes of self-intoxication. Despite
the common use of hydrocarbon solvent mixtures, however, there is little
published information pertaining to their toxicity. Data up to 1940 have been
compiled by Von Oettingen,237 while information through the 1960s has been
35 238 93Q
summarized by Gerarde ' and by Browning. The majority of these reports
suffer from inadequate identification of physical properties and composition
of the solvent mixtures tested. Therefore, comparison of results from the
various studies is very difficult and application of a,study's findings to an
immediate solvent exposure situation is difficult.
Since 1965, investigators206'240'241 have published inhalation toxicity
data on animals that are more credible than previous work due to the advent of
more sophisticated analytical techniques, which ultimately lead to better
characterization of the mixtures tested. To date, the most definitive reports
on inhalation toxicology of hydrocarbon mixtures have been published as a
series of sixteen articles by Carpenter et al.242"257 (A standard protocol
was followed for 15 different solvent mixtures in evaluation of: (1) their
"no-ill-effect" level in rats and dogs; (2) LT5Q and LC5Q in rats; (3) central
nervous system effects in cats; (4) subacute toxicity in rats and dogs; (5)
respiratory irritation in mice, and (6) odor and irritation thresholds in
humans.) The suggested hygienic standards for inhalation of these various
solvents are based upon inhalation studies with, rats and dogs and the sensory
response of human subjects. The composition of the hydrocarbon solvent mixtures
6-169
-------
and the corresponding suggested hygienic standard are summarized in Table
6-59.205>242~257 some solvent mixtures (e.g., kerosene) were of such low
volatility (low vapor pressure) at 25°C that lethal vapor levels could not be
achieved. In these cases, no LC5Q could be obtained, but the suggested
hygienic standard was based upon the sensory response of human subjects at the
respective saturated vapor concentrations in air at 25°C.
These studies lend additional support to the concept that the. majority of
aliphatic, alicyclic, and aromatic hydrocarbons, with the exception of benzene,
appear to be relatively nontoxic, even when encountered as mixtures.
6.6.6 Health Effects Summary
Today as in 1970 there are no observed health effects associated with
volatile hydrocarbons as a class at ambient levels; however, there are health
effects reported at elevated levels for individual hydrocarbons.
Hydrocarbons, strictly speaking, are organic compounds composed of only
carbon and hydrogen. There are many other organic compounds which are loosely
referred to as hydrocarbons but, in addition to carbon and hydrogen, these
compounds contain such elements as the halogens (fluorine, chlorine, bromine,
iodine), sulfur, nitrogen, oxygen, etc. Examples of the latter so-called
hydrocarbons are perch!oroethylene, ethylene dichloride, and ethylene dibromide.
The only hydrocarbons considered in the preceding discussion were those that
were volatile and contained only carbon and hydrogen. This classification
limits the carbon number to approximately twelve and excludes the non-volatile
pblynuclear aromatic hydrocarbons and the volatile substituted hydrocarbons
(e.g., tetrachloroethylene). The volatile hydrocarbons considered can be
categorized as aliphatics, alicyclics, and aromatics.
6-170
-------
re
to
UJ
00
z
^3
t— :c
z
UJ U.
> 0
— 1
O UJ
oo oo
ll
CD 00
O£ UJ
c_j
o >-
C£ Of
a o
>- 00
UJ
00 00
— >
00
5 «
00
— J
os:
U_ HH
o Hr
z oo
UJ
l-l >•
>2 1;
:CO
>— 1
£2 >C
UJ O
00
UJZ
to ^
— 1
in K- <
1
ID Z
o
UJ
— 1 O
m uj
r-
»:
i— i
J2
|
^
-°
*e
in
OJ
s.
=1
•u
X
c
o
J2
re
u
Z
•u
(!„
o
in
c
o;
3
•J^
in
O
CJ
cu
a>
re
^~
f—
a.
" ro
\ c
U) •!-
ll
1-1 CU
01
0)
c
a)
CO
in
t Q)
i— C
>> 01
i— C
« Ol
in
IB
E
o
^£
in
cu
O
in[-i—
'C
TO
£
(0
a o
o c
r— O
Z= '
ol
in
>r.
J^
IB
s-
ra
Q.
*>" in
a> TO s.
<3> i- JZ
re
S- 0
> 0 E
< 1 Q.
a.
•c
Q) U -O
4J »r- i.
in c: TO
oj a -a E
OVi- C Q.
01 o> re o.
oo jz in
c
ai
o
00
CO
' I.I 1 1 . | 0 1
ro
r-l in
0 C5 ' 1 1 CM ro in
IO C3 O
rH rH in O o
0 0 rt ' ' ' 'o ' d
i^ t-i «3- r~ m oo o en c
Id 2 "^ SS!^00 orozi
•^ r-l en ^3* in p«- ro ^~
o
l^
1 1 1 O 1 1 I ii
o
1 1 1 *"' .
0 ' d ' '
^j. ^ CO i-H CM CM
cvl S ° " S °
1
ro ix) ID to in r- i— i
0 toco OCM-JT' 00^
CO CMin CM f— 1 CM t— ) O
•=*• r^ in oo' "in" ' co • 10
in r^,^ 'CO'ID'O ^^
a> «3-«*- CMr-ltD CniO
"* O0«3- tHcninro oro
•S- CM «S- r-l • r-| ^c
s- in ui
tv s- a; -
-^ OJ c 4J
S-jf **>,§! g-g 2 +,gg
w S.5 reS s- g x o 4 *5g •= £
•£_•£ 5> S> t3 W 0 M- Q. > JZ JZ
C ^ D. • T3 r— _Q n— . o> in «r- r— ^-> 4J
i£2°°-cie)xo o.— o
rorec n-Jin =m ^ 10 o «3-TOin o o
> OOCC Erp^rH COCn
rH CM CO «3-miDP^ OOCn
6-171
-------
•a
££
1
U
cn
in
ID
S
re
*o
S
tn
01
t-
5
X
•f—
"S
re
u
•o
!*-
O
tn
4J
c
m
3
4J
in
c
cS
OJ
c
1o
.e
CL
re
z
in
tn «i—
re re
HI 01
01
c
CU
N
C
1 01
>>§
JkT N
F— C
«S 01
in
u
re
§
«
in
c:
C
CU
o
M,
c a
ti-
re
S-
re
§g
'*0 E
Ol
in
c
C
*^
£
+J in
ei re t-
cn t. j=
S
cu in
>*»
to .C tn
in
*>
o>
o
CO
Cn CM I-H
1 >l • 1 '
CM CM I-H
H-t HH
CO O CM CM 00
. 1 . . .
o in cn o CM
ID
o o
i i i i
o o
,
0) O)
cn c c
i-H IDCOQlin rH CU CM
co co in r— cn o p— ^~
CM "3" O CO O CO
in
cn cn in r-i
.... t
i CM in cn r-H
* °° >p '
r-i
II
CM rH
i i tn
1 O 0 *»
V C
0)
c
o
f
o
to ID ID co cn u
i _:_:._•_,
p^ 01 O • cn n* T3
rH 0)
J£
5
c
(D O>
«3- CO CO CM tD rH *O
CM < co tn o o cn ; cz .
ro CM I-H in CM 3
CM Tl- «D CO Of-- *
in m co o cn *}• GO
in co co CM CM cn
00 CO * 00 «- * *
o o o o
SID ID 6
i cn cn CM
CO* CO*
.
u u u
^ tn o ID o o o
r-l CM CO CM CO CO Cn
^- co co in
U 0)
•" > s.
0 C t- 0
CU -r- (U O CL
•o s- ro ro 4-> u >
CU CU CU t-E -C -r- O
ME C *> 0 *> O.4J C -r- O)
>r-cu c: ojc £>c roc o»+J c
t-in -f- coi reoi c cu j= ro re
oo.cz at u > -><*^£ c
*OU 4J 3C ^Zf— Xlf— .CO O
oa> i— o o> o cn o a.i-z
cu ^ cz> o o •*— tn *r- in re re i
O rH, CM CO «• tn ID
rH rH i-l t-1 iH i-H rH
p>;
CM
1
CM
CM tn
* in c
in >> o
O r— "-
tn re t-
0) H->
U r— C
c re ai
cu s- u
i» 1 ^ C
01 U O
<*- cu o
CU CL
s. tn s- o
oo
£ tn o.tn
o tn re CM
s- re >
t- E •<->
•o re
TJ C CU
0) O 4-> S-
r- « •!-
•r-T3t.ro
tin -p cr
re re *^
O CO CO
ro J= u
6-172
-------
All three classes of hydrocarbons, namely the aliphatics, alicyclics, and
aromatics, are similar in that they give rise to irritation of the mucous
membranes and depression of the central nervous system. Mucous membrane
irritation is more pronounced from aromatic vapors compared to equivalent
concentrations of aliphatic and alicyclic hydrocarbons. Unlike the aliphatic
hydrocarbons, the degree of toxicity of the alicyclic hydrocarbons does not
correlate with structural characterisites (e.g., number of carbon atoms,
degree of unsaturation, and amount of branching). Aliphatic hydrocarbons
containing fewer than five carbon atoms are gases at room temperature. The
majority of these hydrocarbons, which include methane, ethane, ethylene,
propane, propylene, butane, and acetylene, are collectively known as "simple
asphyxiants."
Some of these, especially ethylene, propylene, and acetylene, were used
as general anesthetics in the past. Any systemic effects observed at extremely
high concentrations are considered to be secondary due to the oxygen-replacing
capabilities of these gases; however, there are some reports that "aerosol
sniffing" of these simple asphyxiants resulted ultimately in cardiac arrests.
To date, however,.these gases are still considered harmless, physiologically
speaking, due to the fact that they are still classified as "simple asphyxiants"
by the American Conference of Government Industrial Hygienists (ACGIH) with no
assigned threshold limit values.
This is not the case with certain saturated aliphatic hydrocarbons,
namely pentane, hexane, heptane, and octane (C-5 to C-8), all of which have
been implicated as causative agents in permanent impairment of the peripheral
nervous system—namely, toxic polyneuropathy. Based upon the belief that this
neuropathy is caused by alkanes other than n-hexane, NIOSH recommended a
6-173
-------
*3
threshold limit value of 350 mg/m for total airborne C-5 to C-8 alkanes,
which may in the future be revised upward in the case of those alkanes not
causing chronic neurological disorders. This recommended standard by NIOSH
lowers the TLV from 500 ppm for hexane, hepane, and octane for each compound
to about 100 ppm.
Unlike the aliphatic and alicyclic hydrocarbons, aromatic hydrocarbons
have been suspected of causing hematological disorders. There is general
agreement, however, from the recent literature that those hematological
disorders which have ultimately led to leukemia have been due to benzene
alone, and not to any other aromatic hydrocarbons. Knowledge of the
carcinogenic potential of benzene has been primarily gained through the
experience of human occupational exposure and not through experimental animal
studies. Based upon the leukemogenic nature of benzene and its role as the
probable causative agent of other severe systemic effects, NIOSH recommended
an occupational standard of 1 ppm in place of the previous standard of 10 ppm.
OSHA first adopted the 1 ppm standard as an emergency temporary standard and
subsequently as a permanent standard. This decision was immediately challenged
by industry and ruled invalid by the Court of Appeals of New Orleans. This
decision by the lower court was recently upheld by the Supreme Court in its
recent five to four ruling against OSHA; the previous 10 ppm standard remains
in place.
With the exception of cardiac sensitization, there is little firm evidence
that toluene exerts a specific toxic effect on any organ system in experimental
animals or in man. Early reports of toluene-induced myelotoxicity have been
questioned on the basis that toluene was probably contaminated by benzene in
these earlier studies. There is firm evidence, however, that toluene may
6-174
-------
markedly alter the metabolism of other solvents, especially if they are all
oxidized by a common enzyme system in the liver. The present threshold limit
value of 100 ppm was designed to protect workers from any CNS depressant
effects; however, more recent studies suggest that this situation may be
modified by exercise.
Like other hydrocarbons, xylene causes irritation of the mucous membranes
and depression of the central nervous system. Individual isomers are apparently
similar in narcotic potency/acute potency to mixed xylene vapors. Like that
of toluene, reported xylene-induced myelotoxicity was in fact due to contamina-
tion with benzene. The present TLV value of 100 ppm was designed to protect
workers against minimal irritation and CNS depression.
Many of the health effects associated with individual hydrocarbons are
likewise described for exposure to a combination of these components which are
commonly found in varying amounts in gasoline and commerical solvents. Health
effects observed upon exposure to individual hydrocarbons and to hydrocarbon
mixtures are irritation of the mucous membranes, central nervous system depression,
and neurological disorders. The majority of the earlier studies addressing
the toxicity associated with the inhalation of solvent mixtures lacked credit-
ability .because of inadequate characterization of the hydrocarbons present and
inaccurate exposure assessments. Since 1965, more reliable data have been
published due to the advent of more sophisticated analytical techniques for
characterizing mixtures. >To date, the most impressive inhalation toxicology
studies are those of Carpenter from the viewpoint that the properly designed
protocol was followed for all 17 studies, and hygienic standards were
suggested based upon both human and animal data.
6-175
-------
With the exception of benzene, it appears that aliphatics, alicyclics,
and aromatic hydrocarbons are relatively non-toxic even at levels well above
those found in the ambient air, whether present as individual compounds or as
mixtures.
6.7 WELFARE EFFECTS
Although ethylene as an air pollutant has been reviewed earlier by a
number of workers,258" most of the contemporary research and concern has
centered around other components of polluted air, such as S02, ozone (03),
peroxyacetylnitrate, CO, fluorides, and particulates. More attention has been
focused on the other gases because, unlike ethylene, they have a direct effect
on human health and comfort and cause visible vegetation damage in the form of
lesions and crop yield reductions.
Ethylene arises from both natural and anthropogenic sources. The former
sources include plants (endogenous), soil, natural gas, and burning vegetation,
while anthropogenic sources include industries, greenhouses, laboratories, and
automobiles. In 1971 Abeles et al.261 estimated that the total emissions from
cars and other manmade sources and activities in the United States was 15
million tons annually, with 93 percent due to the automobile. The contribution
from vegetation was estimated to be about 20,000 tons annually or about 0.1
percent of the total emissions. In spite of this constant production of large
amounts of ethylene, and occasional ambient measurements of ethylene in urban
areas of up to several hundred ppb, the levels in rural areas are still very
low (<5 ppb). These low levels existed prior to the introduction of emission
controls on mobile and stationary sources; consequently there appear to be
natural sinks for the removal of this unsaturated hydrocarbon from the ambient
air. These sink processes include oxidation by ozone, gas-phase photochemical
6-176
-------
reaction with nitrogen dioxide,262 and aerobic microbial reactions in the
soil.261
Maximum concentrations of ethylene have been observed in the ambient air
of urban areas that are consistent with the increasing traffic density.
Ethylene concentrations in urban areas have ranged from <5 ppb up to high
values of 700 ppb observed by Abeles and Heggesstad50 in Washington, DC. In
general, these values are peak values and are not truly representative of the
majority of observations. This was well demonstrated by Hanan263 in Denver,
where peak values as high as 180 ppb were reported but where the mean levels
for three different sampling stations were less than 50 ppb.
Since 1970, EPA has conducted several field studies for the monitoring of
total nonmethane hydrocarbon and/or individual hydrocarbons. These include
studies at an urban site at St. Louis (1972), rural sites in Ohio (1974), at
mixed sites in Houston (1973-1974), and at suburban sites in the Boston area
(1975). The data obtained from St. Louis and Boston showed extremely high
values of ethylene due to contamination from nearby ozone monitoring systems.
The Wilmington, Ohio, site, which is representative of rural areas, showed
ethylene levels averaging 1.44 ppb (2.88 ppb C) based on 12 samples taken
every 2 hours over a 24-hour period. This is low relative to urban levels,
which is important since ethylene has been well established as a phytotoxicant.
The data presented in Table 6-60191 are representative of the concentrations
of ethylene at the Houston study sampling sites. Figure 6-10191 gives the
geographical locations of the sampling sites. The majority of these sites are
in the ship channel area of Houston and represent the general location of the
industrial complex. The maximum concentration, measured at the various sites,
with the exception of the tunnels, was 166 ppb (332 ppb C) at site H14; the
6-177
-------
TABLE 6-60. HOUSTON STUDY ETHYLENE LEVELS
191
Sampling
site code
9-11-73
HOI
H02
H03
H04
H05
H06
H07
H08
H09
1-30-74
H12
H13
H14
H15
H16
H17
HIS
4-2-74
H21
H22
H23
H24
H25
Concentration
Location
Reagan Road at Euclid Road
Baytown Tunnel
Pasadena (TACB site)
Jacinto City (TACB site)
Washburn Tunnel
Industrial Drive
Channel View
Decker Road
Red Bluff
Minden and Dabney Streets
Jacinto City (TACB site)
Pasadena (TACB site)
Queens Road & Revere
Jacinto City (TACB site)
Pasadena site (TACB site)
Downtown Houston
Jacinto City
Industrial Drive
Jacinto Port Road
Deer Park
Pasadena
ppb
87.35
479.0
15.4
12.25
681.3
5.0
6.95
8.4
3.15
71.5 '
100.95
131.4
165.9
165.9
74.9
62.15
137.3
39.1
19.0
16.35
5.3
ppb C
174.7
958.0
30.8
24.5
1362.7
10.1
13.9
16.8
6.3
143.0
201.9
262.8
331.9
331.9
149.9
124.3
274.7
78.2
38.0
32.7
10.6
6-178
-------
CHANNELVIEW DRIVE
H08
• TUNNEL SAMPLES
Figure 6-10. Geographic location of the grab sample collection sites, Houston.
6-179
-------
lowest concentration measured was 5.0 ppb (10.6 ppb C). The highest average
levels of ethylene observed under different meteorological conditions were
seen in samples collected at sites H12 through HIS, on January 30, 1974. The
average ethylene concentration was 110.4 ppb (220.8 ppb C). Since a high
pressure system was centered over the Houston area on that date, the nonmethane
hydrocarbon (NMHC) levels measured then can be considered to approach the
191
maximum levels expected to occur in Houston. In view of the ethylene
contamination from adjacent ozone monitoring instruments, these figures are
probably ijiflat.ed and would average out at less than 100 ppb. In the absence
of a nationwide moni.tpring system, it is extremely difficult to pass judgment
on an appropriate mean urban level for ethylene; however, in urban regions
that have been monitored, concentrations of 25 ppb are common and 100 ppb
levels are not unusual.
Generally, ethylene concentrations required to affect plants in some way,
e.g., inhibition of stem or root elongation, should exceed the threshold
concentration of 10=ppb; Full or maximum effect;is reached at values, of 1,000
to 10,000 ppb.50 Half-maximum effects are observed at 100 to 500 ppb. The
effect, however, varies greatly with the duration of exposure to a specific
concentration and with plant species.
264
An effective dosage term (ppb-days) was used by Piersol and Hanan to
correlate ethylene concentrations and exposure duration. They reported plant
growth inhibition induced by ethylene dosages ranging from 75 to 10,000 ppb-days.
Although the highest dosages produced the greatest growth inhibition, 70 to 80
percent of the total maximum inhibition was induced by dosages of around 100
ppb-days.
The effects of urban ethylene production on vegetation have been primarily
associated with ornamental plants grown under optimum greenhouse conditions.
6-180
-------
The impact of ethylene on welfare has been a problem in certain localities of
the country. Ethylene is suspected of causing severe economic problems for
individuals involved in the production and retail sale of flowers and crops.
In 1962 California adopted an air quality standard for ethylene, namely, 500
ppb for 1 hour and 100 ppb for 8 hours. This standard was designed to protect
ornamental flowers that are commercially produced in abundance in California,
especially orchids, which are extremely sensitive to ethylene at low levels
CO
(100 ppb for 6 hours). Ethylene levels in certain areas of California and
Colorado are aggravated by frequent inversions combined with high-density
traffic. With the introduction of the catalytic converter on all 1975 model
cars and thereafter, the exhaust emissions of ethylene were decreased by a
factor of two, from ~ 12 percent to ~ 6 percent of total hydrocarbon emissions.
Due to natural sinks and the lack of good monitoring data nationwide, it is
not possible to accurately assess pre-1975 and post-1975 ethylene trends.
Since 90 percent of nationwide ethylene emissions are associated with urban
transportation, it is fairly safe to say that ethylene levels have decreased
since 1975 and will continue to do so as more new model cars replace pre-1975
models. This reduction in ethylene levels will hopefully be accelerated by
the application of more stringent hydrocarbon emission standards for all 1980
passenger cars, as well as by similar stringent standards for light- and
heavy-duty trucks for subsequent model years.
With the exception of plant damage by ethylene, hydrocarbons do not
directly affect welfare. Indirectly, however, they are among the pollutants
that contribute to the formation of acidic precipitation, which in turn has
direct effects on welfare. The direct effects of acid precipitation on public
welfare are documented in detail in Air Quality Criteria for Nitrogen Oxides.8
6-181
-------
As a result of the combustion of tremendous quantities of fossil fuels,
approximately 50 million tons of sulfur and nitrogen oxides are discharged
into the atmosphere annually in the United States. Through a series of complex
chemical reactions (see Section 6.1), these gases can be converted into acids
which may return to earth as components of either rain or snow. This acidic
precipitation, more commonly known as acid rain, may have severe effects on
the environment. For example, North American and Scandanavian lakes have
become so acidic that they no longer support fish life. More than 90 lakes in
the Adirondack mountains of New York State are fish!ess because acidic condi-
tions have inhibited reproduction. Recent data indicate that other areas of
the United States may be vulnerable to similar adverse impacts. Other kndwn
effects include: (1) the destruction of monuments and statues and other classi-
cal buildings throughout the world; (2) potential damage to crop yield; and
: iS
(3) possible decreases in the productivity of forests.
6.7.1 Welfare Effects Summary
As early as 1871, hydrocarbons were suspected of causing injury to vegetation;
greenhouse plants were found to be highly susceptible. Early investigators
suspected other chemicals (e.g., CO, HCN, C2H4) as the primary cause of damage.
Later studies demonstrated, however, that the principal cause was the gaseous
hydrocarbon, ethylene, which was found in illuminating the gas commonly used
in greenhouses. Since then, research has demonstrated that ethylene is both an
endogenous plant-growth regulator and a serious phytotoxic air pollutant.
Dose-response curves have been established for ethylene damage; however, the
effect is highly dependent upon concentration, duration of exposure, and plant
species.
6-182
-------
Although there are natural emission sources of this phytotoxic air pollutant,
manmade sources, especially the internal combustion engine, contribute most of
the ethylene in ambient air. In general, rural areas have ethylene levels
less than 5 ppb. The higher levels observed in -urban centers are consistent
with traffic density and are highly dependent upon topography and meteorological
conditions. This is quite evident in certain areas of California (Los Angeles)
and Colorado (Denver) where traffic is dense and inversions are frequent. The
ethylene levels and the susceptibility of orchids and other ornamental flowers .
to ethylene were the prime reason why the State of California adopted an air
quality standard for ethylene in 1962. Although ethylene is ubiquitous in
nature, its impact on welfare as a phytotoxicant is more severe in certain
localities of the country depending on the sensitivity of the plant species
grown there. Ethylene is suspected of causing severe economic problems for
individuals involved in the production and retail sale of flowers and other
crops.
With the advent of the catalytic converter on all new model cars in 1975
and thereafter, ethylene levels should begin to decrease since the internal
combustion engine is responsible for greater than 90 percent of ethylene in
the ambient air.
Although the welfare effects of hydrocarbons, as a class, are limited to
the direct action of ethylene as a phytotoxicant, there are welfare effects
indirectly attributed to hydrocarbons. Photochemically reactive hydrocarbons
in the presence of sunlight and NOX react to produce intermediate radicals
(e.g., OH, H02). These radicals have been implicated in the production of
nitric and sulfuric acids from N0x and S0x, respectively, through oxidation.
Such acids are ultimately returned to earth as components of rain or snow.
6-183
-------
This type of acidic precipitation is commonly referred to as "acid rain." In
addition to severe impacts on the environment, other effects from acid rain
include damage to monuments and other classical structures, reduction in crop
yield, and potential decreases in forest productivity.
6-184
-------
7. REFERENCES.
1. The Clean Air Act as Amended August 1977. U.S. Congress. P.L 95-95
Washington, DC; U.S. Government Printing Office, 1977. '
2. Air Quality Criteria for Hydrocarbons. U.S. Department of Health, Education
and Welfare, Public Health Service, National Air Pollution Control Admin-
istration, Washington, DC. Publication No. AP-64. March 1970.
3. The Clean Air Act of 1970. U.S. Congress. 42 U.S.C. 1857 et seq
Washington, DC; U.S. Government Printing Office, 1971.
4. Federal Register. 36:8186, April 30, 1971.
5. Code of Federal Regulations. Title 40, Part 50, Chapter 1. Washington, DC-
U.S. Government Printing Office, 1975.
6. Federal Register. 44(28):8230-8237, February 8, 1979.
7. Air Quality Criteria for Ozone and Other Photochemical Oxidants U S
Environmental Protection Agency, Research Triangle Park, NC. Publication
No. EPA-600/8-78-004. April 1978.
8. Air Quality Criteria for Nitrogen Oxides. U.S. Environmental Protection.
Agency, Research Triangle Park, NC. Revised External Review Draft. June
1979.
9. Ozone and Other Photochemical Oxidants. Washington, DC. National Academy
of Sciences. 1977. •*
10. Vapor-Phase Organic Pollutants. Washington, DC; National Academy of
Sciences. 1976.
11. Andreatch, A. J. , and R. Fein!and. Continuous Trace Hydrocarbon Analysis
by Flame lonization. Anal. Chem. 32:1021-1024, July 1960 Cited in-
Reference 2. '
12. McElroy, Frank F. Environmental Monitoring and Support Laboratory
Memorandum, "Review of Issue Paper on Hydrocarbons," to Beverly E. Tilton
Environmental Criteria and Assessment Office. U.S. Environmental Protection
Agency, Research Triangle Park, NC. (Undated; received October 1979.)
13. Bruderreck, H., W. Schneider, and I. Halasz. Quantitative Gas Chrdmato-
graphic Analysis of Hydrocarbons with Capillary Columns and Flame lonization
Detector. Anal. Chem. 36:461-473, March 1964. Cited in: Reference 2.
14. Hauser, Thomas R. Environmental Monitoring and Support Laboratory.
Memorandum, "NMOC Measurement Methodology Pertinent to 0, Control for the
1982 SIP Revision," to Robert E. Neligan, Monitoring and Data Analysis
Division. U.S. Environmental Protection Agency, Research Trianqle Park NC
August 3, 1979. '
7-1
-------
15. Altshuller, A. P. Atmospheric analysis by gas chromatography. In:
Advances in Chromatography. Vol. 5. J. C. Giddings and R. A. Keller
(eds.). New York, Marcel Dekker, Inc. 1968. Cited in: Reference 2.
16. Pierce, L. B., and P. K. Mueller. Analysis of atmospheric hydrocarbons
by gas chromatography. Presented at 10th Conference on Methods in Air
Pollution and Industrial Hygiene Studies. San Francisco, CA. February
19-21, 1969. Cited in: Reference 2.
17. Migeotte, M. V. Spectroscopic evidence of methane in the earth's atmos-
phere. Phys. Rev. 73(5):519-520, 1948. Cited in: Reference 2.
18. Koyama, T. Gaseous metabolism in lake sediments and paddy soils and
production of atmospheric methane and hydrogen. J. Geophys. Res.
63(13):3971-3983, 1963. Cited in: Reference 2.
19. Rasmussen, R. A., and F. W. Went. Volatile organic material of plant
origin in the atmosphere. Proc. Nat. Acad. Sci. U.S. 53:215-220, 1965.
20. Mason, D. V. et al. Sources and air pollutant emission patterns in major
metropolitan areas. Presented at 62nd Annual Meeting, Air Poll. Contr.
Assoc., New York, NY. June 22-26, 1969. Cited in: Reference 2.
21. Compilation of Air Pollutant Emission Factors (Part B), 2nd Ed., 3rd
Printing. U.S. Environmental Protection Agency, Research Triangle Park,
N.C. Publication No. AP-42. February 1976. Cited in: Occurrence,
Chemistry, and Monitoring of Photochemical Oxidants and Their Precursors.
Environmental Criteria and Assessment Office. U.S. Environmental Protec-
tion Agency, Research Triangle Park, NC. In press.
22. 1962-1967 Summary of Monthly Means and Maximums of Pollutant Concentra-
tions, Continuous Air Monitoring Projects, National Air Surveillance
Networks. U.S. Department of Health, Education, and Welfare, Public
Health Service. National Air Pollution Control Administration,
Cincinnati, OH. Publication No. APTD-69-1. April 1969. Cited in:
Reference 2.
23. Stephens, E. R. and F. R. Burleson. Distribution of light hydrocarbons
in ambient air. Presented at 62nd Annual Meeting, Air. Poll. Contr.
Assoc., New York, NY, June 22-26, 1969. Cited in: Reference 2.
24. Neligan, R. E. Hydrocarbons in the Los Angeles atmosphere. Arch. Environ.
Health 5:581-591, 1962.
25. Altshuller, A. P. et al. A technique for measuring photochemical reactions
in atmospheric samples. Science, 1970. Cited in: Reference 2.
26. Unpublished data, 1966-1967. Los Angeles, Calif. Cited in: Reference 2.
27. Lonneman, W. A., T. A. Bellar, and A. P. Altshuller. Aromatic hydrocarbons
in the atmosphere of the Los Angeles Basin. Environ. Sci. Techno!.
2:1017-1020, 1968.
7-2
-------
28. Dickinson, J. E. Air Quality of Los Angeles County. Technical Progress
Report, Vol. II. Los Angeles County Air Pollution Control District
Los Angeles, CA, February 1961. Cited in: Reference 2.
29. Renzetti, N. A., and R. J. Bryan. Atmospheric sampling for aldehydes and
eye irritation in Los Angeles smog—1960. J. Air Poll. Contr. Assoc
11:421-424, 427, 1961.
30. Altshuller, A. P., and S. P. McPherson. Spectrophotometric analysis of
aldehydes in the Los Angeles atmosphere. J. Air Poll. Contr. Assoc.
13(3):109-111, 1963.
31. Scott Research Laboratories, Inc. Atmospheric Reaction Studies in the
Los Angeles Basin, Vol. I. Prepared for Coordinating Research Council
Inc., and National Air Pollution Control Administration under NAPCA
Contract No. CPA-22-69-19, 1969. Cited in: Reference 2.
32. Schuck E. A., A. P. Altshuller, D. S. Barth, and G. B. Morgan. Relation-
ship of hydrocarbons to oxidants in ambient atmospheres. J Air Poll
Contr. Assoc. 20:297-302, 1970.
33. Air Quality Criteria for Nitrogen Oxides. U.S. Environmental Protection
Agency, Washington, DC. Publication No. AP-84. January 1971.
34. Appendix J. Federal Register. 36(158):15502, August 14, 1971.
35. Gerarde, H. W. The Aliphatic Hydrocarbons. In: Industrial Hygiene and
Toxicology. Vol. II. F. A. Patty (ed.). New York, Interscience Publishers,
Inc. , 1963.
36. Graedel, T. E. Chemical Compounds in the Atmosphere. New York, Academic
Press, 1978.
37. Gerarde, H. W. The Alicyclic Hydrocarbons. In: Industrial Hygiene and
K^iSS: l^lllj-lZV*' *** "*'>' *" ^ Int-scie-e Publishers,
38. Greenburg, L., et al. The effects of exposure to toluene in industry
J. Amer. Med. Assoc. 118(8):573-578, 1942.
39. Gerarde, H. W. The Aromatic Hydrocarbons. In: Industrial Hygiene and
Tn?100™^' V°1'-,o.1; ™A- Patty (ed-}- New York> Interscience Publishers,
Inc., 1963. pp. 1219-1240.
40. Gerarde, H. W. Hydrocarbon Mixtures. In: Industrial Hygiene and Toxicology
vol. II. F. A. Patty (ed.). New York, Interscience Publishers, Inc
1963. p. 1199-1201.
41. Davis, A., L. J. Schafer, and Z. G. Bell. The effects on humans volunteers
or exposure to air containing gasoline vapor. Arch. Environ Health
1:548-564, 1960.
7-3
-------
42. Threshold Limit Values of Airborne Contaminants and Intended Changes
Adopted by the American Conference of Governmental Industrial Hygienists
for 1970. Cincinnati, OH; Amer. Conf. Gov't. Ind. Hyg., 1970.
43. W. Machle. Gasoline intoxication. J. Amer. Med. Assoc. 117:1965-1971,
1941.
44. Pratt, H. K., and J. D. Goeschl. Physiological roles of ethylene in
plants. Annu. Rev. Plant Physio!. 20:541, 1969.
45. Altshuller, A. P., and T. A. Bellar. Gas chromatographic analysis of
hydrocarbons in the Los Angeles atmosphere. J. Air Poll. Contr. Assoc.
13:81-87, 1963.
46. Clayton, G. C., and T. S. Platt. Evaluation of ethylene as an air pollutant
affecting plant life. Amer. Ind. Hyg. Assoc. J. 28:151-160, 1967.
47. Gordon, R. J., H. Mayrsohn, and R. M. Ingels. C2~C5 hydrocarbons in the
Los Angeles atmosphere. Environ. Sci. Techno!. 2:1117-1120, 1968.
48. Stephens, E. R., E. F. Parley, and F. R. Burleson. Sources and reactivity
of light hydrocarbons in ambient air. Proc. Amer. Petrol. Inst. (Refin.
Div). 47:466-483, 1967.
49. Scott, W. E., E. R. Stephens, P. C. Hanst, and R. C. Doerr. Further
developments in the chemistry of the atmosphere. Proc. Amer. Petrol.
Inst. 37:171, 1957.
50. Abeles, F. B., and H. E. Heggestad. Ethylene: An urban air pollutant.
J. Air Poll. Contr. Assoc. 23(6):517-521, 1973.
51. Abeles, F. B. Ethylene. In: Plant Biology. Academic Press, New York,
1973. p. 252.
52. Crocker, W. Physiological Effects of Ethylene and Other Unsaturated
Carbon-Containing Gases. In: Growth of Plants. New York, Reinhold,
1948.
53. California Standards for Ambient Air Quality and Motor Vehicle Exhaust.
Supplement No. 2 (Technical Report). California State Dept. of Public
Health, Berkeley, CA, 1962.
54. Katz, M. Quality standards for air and water. Occupational Health Rev.
17(1): 3, 1965.
55. Stern, A. C. (ed.). Air Pollution, Vol. Ill, 2nd Ed. New York, Academic
Press, 1968. p. 55, 97, 666.
56. Ethylene in community air quality guides. Am. Ind. Hyg. Assoc. J. 29:627,
1968.
7-4
-------
57.
58.
59.
60.
61.
62.
63.
64.
65.
66.
67.
Calvert, J. 6., and R. D. McQuigg. The computer simulation of the rates
and mechanisms of photochemical smog formation. In: Proc. Symposium,
Chemical Kinetics Data for the Upper and Lower Atmosphere. Warrenton,
VA, September 15-18, 1974. Int. J. Chem. Kinetics. New York, John Wiley
and Sons, Inc., 1975.
Demerjian, K. L., J. A. Kerr, and J. G. Calvert. The mechanisms of
photochemical smog formation. In: Advances in Environmental Sciences
and Technology. New York, John Wiley and Sons, Inc., 1974.
Stedman, D. H. , F. D. Norvis, E. E. Daley, H. Niki, and B. Weinstock.
The role of OH radicals in photochemical smog reactions. Paper presented
at 160th Nat'l. Meeting of American Chemical Society, Chicago, September
1970.
Heicklen, J., K. Westberg, and N. Cohen. The Conversion of NO to N0? in
Polluted Atmospheres. Center for Air Environmental Studies. Pennsylvania
State Univ., University Park, PA. Publication No. CAES 115-69. 1969.
Air Quality Criteria for Sulfur Oxides and Particulate Matter, External
Review Draft No. 1. U.S. Environmental Protection Agency, Research
Triangle Park, NC, April 1980.
Fried!ander, S. K.
pollution sources.
Chemical element balances and identification of air
Environ. Sci. Techno!. 1:235-240, 1972.
Grosjean, D., K. Van Cauwenberghe, J. P. Schmid, P. E. Kelley, and J. N.
Pitts, Jr. Identification of C~-Cin aliphatic dicarboxylic acids in
airborne particulate matter. Environ. Sci. Techno!. 12(3):313-317, 1978.
Miller, D. F., W. E. Schwartz, P. W. Jones, D. W. Joseph, C. W. Spicer,
C. J. Higgle, and A. Levy. Haze Formation and Its Origin. Prepared for
U.S. Environmental Protection Agency by Battelle Columbus Laboratories,
Columbus, Ohio. U.S. Environmental Protection Agency, Research Triangle
Park, NC. Publication No. EPA-650/3-74-002, 1974. Cited in: Reference
9. :
O'Brien, R. J., J. H. Cfabtree, J. R. Holmes, M. C. Hoggan, and A. H.
Bockian. Formation of photochemical aerosol from hydrocarbons.
Atmospheric analysis. Environ.. Sci. Techno!. 9:577-582, 1975.
Schuetzle, D., D. Cronn, A. L. Crittenden, and R. J. Charlson. Molecular
composition of secondary aerosol and its possible origin. Environ. Sci.
Techno!. 9:838-845, 1975.
Schuetzle, D., and R. A. Rasmussen. The molecular composition of secondary
aerosol particles formed from terpenes. J. Air Poll. Contr. Assoc.
28(3):236-240, 1978.
7-5
-------
68.
69.
70.
71.
72.
73.
74.
75.
76.
77.
78.
Pitts, J. N., Jr., D. Grosjean, G. J. Doyle, D. R. Fitz, P. L. Johnson,
S. L. Midland, T. M. Mischke, K. J. Pettus, M. P. Poe, J. P. Smith, and
J. K. Suder. Detailed Characterization of Gaseous and Size-Resolved
Particulate Pollutants at a South Coast Air Basin Smog Receptor Site:
Levels and Modes of Formation of Sulfate, Nitrate and Organic Particulates
and Their Implications for Control Strategies. Statewide Air Pollution
Research Center, University of California. Riverside, CA, December 1978.
436 p.
Eggleton, A. E. J., and R. A. Cox. Homogeneous oxidation of sulphur
compounds in the atmosphere. Atmos. Environ. 12: 227-230, 1978.
Urone, P., and W. H. Schroeder. S0£in the atmosphere, a wealth of
monitoring data, but few reaction rate studies. Environ. Sci. Techno!.
3:436-445, 1969.
Middleton, P., C. S. Kiang, and V. A. Mohnen. Theoretical estimates of
the relative importance of various urban sulfate aerosol production
mechanisms. Atmos. Environ. 14:463-472, 1980.
Penkett, S. A., B. M. R. Jones, K. A. Brice, and A. E. J. Eggleton. The
importance of atmospheric ozone and hydrogen peroxide in oxidizing sulphur
dioxide in cloud and rainwater. Atmos. Environ. 13:123-137, 1979.
Occurrence, Chemistry, and Monitoring of Photochemical Oxidants and Their
Precursors. Environmental Criteria and Assessment Office. U.S.
Environmental Protection Agency, Research Triangle Park, NC. In press.
Reckner, L. R. Survey of Users of EPA Reference Method for Measurement '
of NMHC in Ambient Air. U.S. Environmental Protection Agency, Research
Triangle Park, NC. Publication No. EPA-650/4-75-008. 1974. Cited in:
Reference 73.
McElroy, F. F., and V. L. Thompson. Hydrocarbon Measurement Discrepancies
Among Various Analyzers Using Flame lonization Detectors. U.S. Environmental
Protection Agency, Research Triangle Park, NC. n"un<'•-*-•'
1975. Cited in: Reference 73.
Publication No. EPA-600/4-75-010,
Harrison, J. W., M. L. Timmons, R. B. Denyszyn, and C. E. Decker.
Evaluation of the EPA Reference Method for Measurement of Nonmethane
Hydrocarbons. U.S. Environmental Protection Agency, Research Triangle
Park, NC. Publication No. EPA-600/4-77-033. June 1977.
Black, F. M., L. E. High, and J. E. Sigsby. Methodology for Assignment
of a Hydrocarbon Photochemical Reactivity Index for Emissions from Mobile
Sources. U.S. Environmental Protection Agency, Research Triangle Park,
NC. Publication No. EPA-650/2-75-025. March 1975.
McClenny, W. A. Environmental Sciences Research Laboratory. Memorandum
to Frank F. McElroy, Environmental Monitoring Support Laboratory. U.S.
Environmental Protection Agency, Research Triangle Park, NC. March 19,
1979.
7-6
-------
79. McClenny, W. A. Environmental Sciences Research Laboratory. Personal
communication to B. E. Til ton, Environmental Criteria and Assessment
Office. U.S. Environmental Protection Agency, Research Triangle Park, NC.
July and August 1980.
80. Neligan, R., and E. L. Martinez. Presentation at Oxidant Research Committee
Meeting. U.S. Environmental Protection Agency, Research Trianqle Park,
NC. 1979.
81. Brittain, Doyle T. Region IV, U.S. Environmental Protection Agency.
Memorandum to Edmond P. Lomasney, Office of Research and Development,
U.S. Environmental Protection Agency. June 28, 1979.
82. Deitz, W. A. Response factors for gas chromatographic analyses. J. Gas
Chromatog. 5:68-71, 1968. Cited in: Reference 73.
83. Seila, R. L., W. A. Lonneman, and S. A. Meeks. Evaluation of polyvinyl
fluoride as a container material for air pollution samples. J. Environ.
Sci. Health - Environ. Sci. Eng. 2:121-130, 1976. Cited in: Reference
73. ~
84. Lonneman, W. A., S. A. Meeks, and R. L. Kuntz. Personal communication to
R. Kamens, University of North Carolina, Chapel Hill, NC from W. A.
Lonneman, U.S. Environmental Protection Agency, Research Triangle Park
NC. September 1978. Cited in: Reference 73.
85. Meeks, S. A. Environmental Sciences Research Laboratory. Personal
communication to B. E. Tilton, Environmental Criteria and Assessment
Office. U.S. Environmental Protection Agency, Research Triangle Park NC
October 1979.
86. Lodge, J. P., Jr., et al. The use of hypodermic needles as critical
orifices in sampling. J. Air Poll. Contr. Assoc. 16:197-200. April
1966. Cited in: Reference 2.
87.
88.
89.
90.
91.
Black, F. Environmental Sciences Research Laboratory. Memorandum to B.
E. Tilton, Environmental Criteria and Assessment Office. U.S. Environmental
Protection Agency, Research Triangle Park, NC. May 29, 1979.
National Air Quality, Monitoring, and Emissions Trends Report, 1977.
U.S. Environmental Protection Agency, Research Triangle Park, NC.
Publication No. EPA-450/2-78-052. December 1978.
Emission Estimates, 1940 through 1976. U.S. Environmental Protection
Agency, Research Triangle Park, NC. Publication No. EPA-450/1-78-003,
1978.
Statistical Abstract of the United States. 1975. U.S. Department of
Commerce, Washington, DC; U.S. Government Printing Office. 1977.
1975 National Emissions Report, National Emissions Data System of the
Aerometric and Emissions Reporting System. U.S. Environmental Protection
Agency, Research Triangle Park, NC. Publication No. EPA-450/2-78-020.
May 1978.
7-7
-------
92. Heisler, S. L., S. K. Fried!ander, and R. B. Husar. The relationship of
smog aerosol size and chemical element distributions to source characteris-
tics. Atmos. Environ. 7:633-649, 1973.
93. Gartrell, G., Jr., and S. K. Friedlander. Relating particulate pollution
to sources: The 1972 California aerosol characterization study. Atmos.
Environ. 9:279-299, 1975.
94. Gordon, G. E. Receptor models. Environ. Sci. Technol. 14(7): 792-800,
1980.
95. Lioy, P. J., and T. J. Kneip. Aerosols: Anthropogenic and natural
sources and transport. Summary of a conference. J. Air Poll. Contr.
Assoc. 30(4):358-361, 1980.
96. Neligan, R. E. Hydrocarbons in the Los Angeles atmosphere. Arch. Environ.
Health 5:581, 1962.
97. Stephens, E. R., and F. R. Burleson. Distribution of light hydrocarbons
in ambient air. J. Air Poll. Contr. Assoc. 19:929, 1969.
98. Lonneman, W. A., S. L. Kopczynski, P. E. Darley, and F. D. Sutterfield.
Hydrocarbon composition of urban air pollution. Environ. Sci. Technol.
8(3):229-236, 1974.
99. Siddiqi, A. A., and F. L. Worley, Jr. Urban and industrial air pollution
in Houston, Texas—I. Hydrocarbons. Atmos. Environ. 11:131-143, 1977.
100. Kopczynski, S. L., W. A. Lonneman, T. Winfield, and R. Seila. Gaseous
pollutants in St. Louis and other cities. J. Air Poll. Contr. Assoc.
25(3):251-255, 1975.
101. Mayrsohn, H., and J. H. Crabtree. Source reconciliation of atmospheric
hydrocarbons. Atmos. Environ. 10:137-143, 1975.
102. Mayrsohn, H., J. H. Crabtree, M. Kuramoto, R. D. Sothern, and S. H. Mano.
Source reconciliation of atmospheric hydrocarbons, 1974. Atmos. Environ.
11:189-192, 1977.
103. Crabtree, J. H., and H. Mayrsohn. Urban and industrial air pollution in
Houston, Texas—I. Hydrocarbons (Discussion). Atmos. Environ. 12:1245,
1977.
104. Seila, R. L. Nonurban Hydrocarbon Concentrations in Ambient Air North of
Houston, Texas. U.S. Environmental Protection Agency, Research Triangle
Park, NC. Publication No. EPA-600/3-79-010. February 1979. ,38 p.
105. Robinson, E., and R. C. Robbins. Sources, abundance, and fate of gaseous
atmospheric pollutants. SRI Project PR-6755, Huntsville, Alabama. Menlo
Park, CA; Stanford Research Institute. 1968. 123 p.
106. Crutzen, P. J. Photochemical reactions initiated by and influencing
ozone in unpolluted tropospheric'air. Tell us. XXVI(1-2):47-57, 1974.
7-8
-------
107. Dimitriades, B. Issues regarding the role of natural organics in
photochemical air pollution. Presented at the 73rd Annual Meeting, Air
Poll. Contr. Assoc., Montreal, June 22-27, 1980. Preprint No. 80-69 1
Pittsburgh, PA. Air Poll. Contr. Assoc., 1980.
108. Hahn, J. H., G. Ketseridis, and G. Schebeske. The ocean as a source of
higher atmospheric n-alkanes. Presented at the 73rd Annual Meeting, Air
Poll. Contr. Assoc., Montreal, June 22-27, 1980. Preprint No. 80-69.4.
Pittsburgh, PA. Air Poll. Contr. Assoc., 1980.
109. Zimmerman, P. R. Testing of Hydrocarbon Emissions from Vegetation and
Development of a Methodology for Estimating Emission Rates from Foliage.
Final Report, Draft. Prepared by Washington State University for U.S.
Environmental Protection Agency under Contract No. DU-77-C063. U.S.
Environmental Protection Agency, Research Triangle Park, NC. May 1978.
110. Westberg, B. H. Biogenic hydrocarbon measurements. |n: Proceedings of
Symposium on Atmospheric Biogenic Hydrocarbons: Emission Rates, Concen-
trations, and Fates, January 8-9, 1980. J. J. Bufalini, ed. U.S.
Environmental Protection Agency, Research Triangle Park, NC. In press.
111. Tingey, D, T. The effect of environmental factors on the emissions of
biogenic hydrocarbons from live oak and slash pine. In: Proceedings of
Symposium on Atmospheric Biogenic Hydrocarbons: Emission Rates, Concen-
trations, and Fates, January 8-9, 1980. J. J. Bufalini, ed. U.S.
Environmental Protection Agency, Research Triangle Park, NC. In press.
112. Arnts, R. R., and B. W. Gay, Jr. Photochemistry of Some Naturally Emitted
Hydrocarbons. U.S. Environmental Protection Agency, Research Triangle
Park, NC. Publication No. EPA-600/ 3-79-081, 1979.
113. Arnts, R. R.,. and S. A. Meeks. Biogenic Hydrocarbon Contribution to the
Ambient Air of Selected Areas. U.S. Environmental Protection Agency,
Research Triangle Park, NC. Publication No. EPA-600/3-80-023. 1980.
114. Rasmussen, R. A. Isoprene: Identified as a forest-type emission to the
atmosphere. Environ. Sci. Techno!. 4:667-671, 1970.
115. Zimmerman, P. R. Testing of Hydrocarbon Emissions from Vegetation, Leaf
Litter, and Aquatic Surfaces, and Development of a Methodology for Compiling
Biogenic Emission Inventories. Final Report. Prepared by Washington
State University, Pullman, WA, for U.S. Environmental Protection Agency
under Contract No. DU-77-C063. .U.S. Environmental Protection, Research
Triangle Park, NC. Publication No. EPA-450/4-79-004. March 1979.
116. Flyckt, D. L., H. H. Westberg, and M. W. Holdren. Natural organic emissions
and their impact on air quality. Presented at the 73rd Annual Meeting.
Air Poll. Contr. Assoc., Montreal, June 22-27, 1980. Preprint No. 80-69.2.
Pittsburgh, PA. Air Poll. Contr. Assoc., .1980.
117. Quarles, T., B. Lamb, and E. Robinson. Measurements of isoprene fluxes
from a northeastern deciduous forest. Presented at 73rd Annual Meeting,
Air Poll. Contr. Assoc., Montreal, June 22-27, 1980. Preprint No. 80-69.3.
Pittsburgh, PA. Air Poll. Contr. Assoc., 1980.
7-9
-------
118. Bufallni, J. J., ed. Proceedings of Symposium on Atmospheric Biogenic
Hydrocarbons: Emission Rates, Concentrations, and Fates; January 8-9,
1980. U.S. Environmental Protection Agency, Research Triangle Park, NC.
In Press.
119. Zimmerman, P.. R. Natural, sources of ozone in Houston: Natural organics.
In: Proc. of Specialty Conf. on Ozone/Oxidants: Interactions with the
Total Environment, October 14-17, 1979. Pittsburgh, PA. Air. Poll.
Contr. Assoc., 1980.
120. Tingey, D. T. The effect of environmental factors on the emissions of
biogenic hydrocarbons from live oak. and slash pine. In: Proc. of Symposium
on Atmospheric Biogenic Hydrocarbons: Emission Rates, Concentrations,
and Fates, January 8-9, 1980. J. J. Bufalim", ed. U.S. Environmental
Protection Agency, Reasearch Triangle Park, NC. In Press.
121. Wells, C. G. Methods of estimationg foliar biomass. In: Proc. of
Symposium on Atmospheric Biogenic Hydrocarbons: Emission Rates, Concen-
trations, and Fates; January 8-9, 1980. J. J. Bufalini, ed. U.S.
Environmental Protection Agency, Research Triangle Park-, NC. In Press.
122. Compilation of Emission Factors. U.S. Environmental Protection Agency,
Research Triangle Park, NC. Publication No. AP-42.
123. Bucon, H. W., J. F. Macko, and H. J. Taback. Volatile Organic Compound
(VOC) Species Data Manual. Prepared by KVB Engineering, Inc., Tustin, .
CA, for U.S. Environmental Protection Agency under Contract No. 68-02-3029.
U.S. Environmental Protection Agency, Research Triangle Park, NC. Publication
No. EPA-450/3-78-119. December 1978.
124. Hammersma, W. Environmental Engineering Div., TRW, Incorporated, Redondo
Beach, CA. Private communication with B.E. Tilton, Environmental Criteria
and Assessment Office, U.S. Environmental Protection Agency, Research
Triangle Park, NC. August 16, 1979.
125. Darley, Ellis F. Hydrocarbon Characterization of Agricultural Waste
Burning. Prepared under Contract ARB A7-068-30 by Statewide Air Pollution
Research Center, University of California, Riverside, for California Air
Resources Board, Sacramento, CA. Report No. ARB-R-A7-068-30-79-99.
April 1979.
126. Mobile Source Emission Factors. U.S. Environmental Protection Agency,
Washington, DC. Publication No. EPA-400/9-78-005. March 1978.
127. Fisher, T. M. The evolution of automotive emission controls. Presented
at 73rd Annual Meeting, Air Poll. Contr. Assoc., Montreal, Canada, June
22-27, 1980. Preprint No. 80-1.1. Pittsburgh, PA. Air Poll. Contr.
Assoc., 1980.
128. Mobile Source Emission Factors. Appendix E. Derivation of Motor Vehicle
Exhaust Emission Rates. U.S. Environmental Protection Agency, Washington, DC.
Publication No. EPA-400/9-78-005. March 1978.
7-10
-------
129. Motor Vehicle Manufacturers Association of the United States, Inc
Highway motor vehicle emission factors. Ir±: Proc of Emission
Inventory/Factor Workshop, Vol. 2. U.S. Environmental Protection Aqency
Research Triangle Park, NC. Publication No. EPA-450/3-78-042b. 1977.
130. Williams, M. Comments on "Highway motor vehicle emission factors " In-
Proc. of Emission Inventory/Factor Workshop, Vol 2. U.S. Environmental"'
Protection Agency, Research Triangle Park, NC. Publication No
EPA-450/3-78-042b. 1977.
131. Burger, R. J. National Research Council, Washington, D.C. Letter to
Michael P. Walsh, Office of Mobile Source Air Pollution Control U S
Environmental Protection Agency, Washington, DC, June 30, 1980.'
132. National Research Council. Letter Report on the Feasibility of Light
Duty Vehicles to Comply with Existing 1980 Standards. Cited in-
Reference 131.
133. Bobalek E. Industrial Environmental Research Laboratory. Memorandum to
B. E. Tilton, Environmental Criteria and Assessment Office U S
Environmental Protection Agency, Research Triangle Park, NC. May 22,
134. Linnell, R. H., and W. E. Scott. Diesel exhaust composition and odor
studies. J. Air Poll. Contr. Assoc. 12:510,515,545; 1962. Cited in-
Reference 10.
135. McKee, H. C., J. M. Clark, and R. J. Wheeler. Measurement of diesel
engine emissions. J. Air Poll. Contr. Assoc. 12:516-521, 546; 1962.
136. Compilation of Emission Factors. Appendix D. Emission Factors U S
Environmental Protection Agency, Research Triangle Park, NC. Publication
No. AP-42. December 1975.
137. Technology Assessment of Changes in the Future Use and Characteristics of
the Automobile Transportation System. Summary and Findings. Washington, DC-
The Congress of the United States, Office of Technology Assessment
February 1979.
138. Black, F. M., L. E. High, and J. M. Lang. Composition of automobile
evaporative and tailpipe hydrocarbon emissions. J. Air Poll Contr
Assoc. In Press.
139. Assessment of Gasoline Toxicity. Prepared for U.S. Environmental Protec-
tion Agency by PEDCo Environmental, Inc., Cincinnati, Ohio, under Contract
No. 68-02-2515. U.S. Environmental Protection Agency, Research Triangle
Park, NC. July 1978.
140. Morris, W. E., and K. T. Dishart. influence of vehicle emission control
systems on the relationship between gasoline and vehicle exhaust hydrocarbon
composition. In: Effect of Automotive Emission Requirements on Gasoline
Characteristics. ASTM Special Publication 487. Philadelphia American
Society for Testing and Materials, 1971. Cited in: Reference 10
7-11
-------
141. Dishart, K. T. Exhaust hydrocarbon composition. Its relation to gasoline I
composition. Proc. Amer. Petrol. Inst. 50:514-540, 1970. Cited in:
Reference 10. I
142. Dishart, K. T., and W. C. Harris. The effect of gasoline hydrocarbon
composition on automotive exhaust emissions. Proc. Amer. Petrol. Inst.
49:612-642, 1968. Cited in: Reference 10.
143. Doelling, R. P., A. F. Gerber, and M. P. Walsh. Effect of gasoline
characteristics on auto exhaust emissions. In; Effect of Automotive
Emission Requirements on Gasoline Characteristics. ASTM Special Technical
Publication 487. Philadelphia, American Society for Testing and Materials,
1971. Cited in: Reference 10.
144. Neligan, R. E., P. F. Mader, and L. A. Chambers. Exhaust composition in
relation to fuel composition. J. Air Poll. Contr. Assoc. 11:178-186,
1961. Cited in: Reference 10.
145. Papa, L. J. Measuring exhaust hydrocarbons down to parts per billion.
In: Vehicle Emissions. Part III (Selected SAE Papers, 1967-1970). SAE
Prog."Tech. Ser. Vol 14. New York, Society of Automotive Engineers,
1971. Cited in: Reference 10.
146. Jackson, M. W. Effects of some engine variables and control systems on
composition and reactivity of exhaust emissions. Vehicle Emissions.
Part II. SAE Prog. Tech. Ser. 12:241-267, 1967. Cited in: Reference
10. ' -
147. Black, F. M. and R. L. Bradow. Patterns of hydrocarbon emissions from
1975 production cars. Presented at Society of Automotive Engineers'
National Fuels and Lubricants Meeting, Houston,_ TX.: June 1975. ; SAE,
Paper No. 750681. Cited in: Reference 73.
148. Black, F. M. Automotive hydrocarbon emission patterns and the measurement
of nonmethane hydrocarbon emission rates. Presented at Society of Automotive
Engineers' International Automotive Engineering Congress and Exposition,
Detroit, MI. February 1977.
149. Jackson, M. W. Effect of catalytic emission control on exhaust hydrocarbon
composition and reactivity. Presented at Society of Automotive Engineers
Passenger Car Meeting, Troy, MI. June 5-9, 1978. SAE Paper No. 780624.
150. Trijonis, J. C., and K. W. Arledge. Utility of Reactivity Criteria in
Organic Emission Control Strategies: Application to the Los Angeles
Atmosphere. Prepared by TRW Environmental Services, Redondo Beach, CA,
under Contract No. 68-02-1735 for U.S. Environmental Protection Agency.
U.S. Environmental Protection Agency, Research Triangle Park, NC.
Publication No. EPA-600/3-76-091. August 1976.
151. Jerskey, T. N., R. G. Lamb, G. Z. Whitten, L. E. Reid, and H. Hogo.
Sources of Ozone: An Examination and Assessment. Prepared by Systems
Applications, Inc., San Rafael, Calif., for the American Petroleum Institute,
Washington, DC. Report No. EF76-110R. August 1976.
7-12
-------
152. Leihkanen, H. E., and E. W. Beckman. The effect of leaded and unleaded
gasolines on exhaust emissions as influenced by combustion chamber deposits.
Presented at Society of Automotive Engineers' National Combined Fuels and
Lubricants, Powerplant, and Truck Meetings, St. Louis, Missouri, October
26-29, 1971. SAE Paper No. 71083. Cited in: Reference 10.
153. Fleming, R. D., W. F. Marshall, and J. R. Allsup. Fuel Additives and
Automobile Emissions. U.S. Bureau of Mines Tech. Progress Rept. No. 40.
U.S. Department of the Interior, Washington, DC, 1971. Cited in:
Reference 10.
154. Bradow, R. L.s Environmental Sciences Research Laboratory. Personal
communication with R. M. Bruce and B. E. Tilton, Environmental Criteria
and Assessment Office. U.S. Environmental Protection Agency, Research
Triangle Park, NC. August 13, 1979.
155. Nebel, G. J. Benzene in auto exhaust.
29(4):391-392, 1979.
J. Air Poll. Contr. Assoc.
156. Hinkanip, J. B. , M. E. Griffing, and D. W. Zutut. Aromatic aldehydes and
phenols in the exhaust from leaded and unleaded fuels. Presented at
Meeting of Petroleum Chemistry Division, American Chemical Society, Los
Angeles, California., March 28-April 2, 1971. Cited in: Santodonato,
J., D. Basu, and P. Howard. Health Effects Associated with Diesel Exhaust
Emissions. Prepared by Syracuse Research Corporation, Syracuse, N. Y.,
for U.S. EPA under Contract No. 68-02-2800. U.S. Environmental Protection
Agency, Research Triangle Park, NC. Publication No. EPA-600/1-78-063.
November 1978.
157. Gross, G. P. The effect of fuel and vehicle variables on polynuclear
aromatic hydrocarbons and phenol emissions. Presented at Society of
Automotive Engineers' Automotive Engineering Congress and Exposition,
Detroit, MI, 1972. SAE Paper No. 720210. Cited in: Reference 178.
158. Brinkman, N. D., N. E. Gallopoulos, and M. W. Jackson. Exhaust emissions,
fuel economy, and driveability of vehicles fueled with alcohol-gasoline
blends. Presented at Automotive Engineering Congress and Exposition,
February 24-28, 1975, Detroit, Michigan. SAE Paper No. 750120. Warrendale,
PA; Soc. Automotive Eng., Inc.
159. Bernhardt, W. E. and W. Lee. Engine performance and exhaust emission
characteristics of a methanol-fueled automobile. In: Future Automotive
Fuels. J. M. Colucci and N. E. Gallopoulos (eds). New York, Plenum
Press, 1977.
160. Lawrence, R. D. Emissions from gasohol-fueled vehicles. In: Proc. of
3rd Int'l. Symposium on Alcohol Fuels Technology, May 1979, Asilomar, CA.
U.S. Department of Energy, Washington, DC. Publication No. CONF-790520.
April 1980.
161. Mooney, J. J., J. G. Hansel, and K. R. Burns. Three-way conversion
catalysts on vehicles fueled with ethanol-gasoline mixtures. Presented
at Automotive Engineering Congress and Exposition, February 26-March 2,
1979, Detroit, Michigan. SAE Paper No. 790428. Warrendale, PA; Soc.
Automotive Eng., Inc.
7-13
-------
162. Lang, J. N., and F. M. Black. Impact of gasohol on automobile evaporative
and tailpipe emissions. In Press.
163. Furey, R. L., and J. B. King. Evaporative and exhaust emissions from
cars fueled with gasoline containing ethanol or methyl-tert-butyl ether.
Presented at Automotive Engineering Congress and Exposition, February
25-29, 1980, Detroit, Michigan. SAE Paper No. 800261, Warrendale, PA;
Soc. Automotive Eng., Inc.
164. Naman, T. M., and J. R. Allsup. Exhaust and evaporative emissions from
alcohol and ether fuel blends. U.S. Department of Energy. Bartlesville
Energy Technology Center, Bartlesville, OK. (Undated, 1979 or later).
165. Harrington, J. A., D. D. Brehob, and E. H. Schanerberger. Evaluation of
.methyl tertiary butyl ether as a gasoline blending component. In: Proc.
of 3rd Int'l. Symposium on Alcohol Fuels Technology, May 1979, Asilomar,
CA. U.S. Department of Energy, Washington, DC. Publication No. CONF-790520.
April 1980.
166. Harrenstien, M. S. Measurement of Aldehyde Concentrations in the Exhaust
of an Internal Combustion Engine Fueled by Alcohol/Gasoline Blends. M.S.
Thesis, University of Miami, Coral Gables, Florida. August 1978.
167. Matsuno, M., Y. Nakano, H. Kachi, K. Kimura, F. Tsuruga, N. Iwai, H. Suto,
H. Enokido, and T. Itoh. Alcohol engine emissions: Emphasis on unregulated
compounds. Japan Automobile Research Inst., Inc., Yatabe-cho, Tsukuba-gun,
Ibaraki-ken. Japan. (Undated; 1978 or later.)
168. Linnell, R. H., and W. E. Scott. Diesel exhaust analysis. Arch. Environ.
Health. 5:616-625, 1962. Cited in: Reference 10.
169. Merrion, D. F. Effect of design revisions on two stroke cycle diesel
engine exhaust. In: Vehicle Emissions, Part III. SAE Prog, in Tech.
Series. Vol. 14. New York, Society of Automotive Engineers, Inc., 1971.
Cited in: Reference 10.
170. Black, F. M., and L. E. High. Diesel hydrocarbon emissions, particulate
and gas phase. Presented at Society of Automotive Engineers' Symposium
on Diesel Particulate Emissions Measurement Characterization, Detroit,
Michigan, February 1979.
171. Braddock, J. N., and P. A. Gabele. Emission patterns of diesel-powered
passenger cars. Part II. Presented at Society of Automotive Engineers'
Automotive Engineering Congress and Exposition, Detroit, Michigan,
February 28 - March 4, 1977. SAE Paper No. 770168.
172. Heisler, S. L., and S. K. Friedlander. Gas-to-particle conversion in
photochemical smog: Aerosol growth laws and mechanisms for organics.
Atmos. Environ. 11:157-168, 1977.
173. Hidy, G. M., and C. S. Burton. Atmospheric aerosol formation by chemical
reactions. In: Proc. of Int'l. J. Chem. Kinetics Symposium No. 1. New
York, NY; John Wiley and. Sons, 1975.
7-14
-------
174. Mayrsohn, H. , and J. H. Crabtree. Atmospheric Hydrocarbon Trends, 1967-1977.
California Air Resources Board. Sacramento, CA. Paper No. RD-78-008D
May 1978.
175. Health Assessment Document for Polycyclic Organic Matter. Final Draft.
U.S. Environmental Protection Agency, Research Triangle Park, NC. June 1980.
176. Vogh, J. W. Mature of odor components in diesel exhaust. J. Air Poll
Contr. Assoc. 19:773-777, 1969.
177. Marshall, W. F., D. E. Seizinger, and R. W. Freedman. Effects of Catalytic
Reactors on Diesel Exhaust Composition. U.S. Bureau of Mines, Dept. of
the Interior. Washington, DC. Technical Prog. Rept. No. 105. April
1978.
178. Santodonato, J., D. Basu, and P. Howard. Health Effects Associated with
Diesel Exhaust Emissions. Prepared by Syracus Research Corporation,
Syracus, NY, for U.S. Environmental Protection Agency under Contract No.
68-02-2800. U.S. Environmental Protection Agency, Research Triangle
Park, NC. Publication No. EPA-600/1-78063. November 1978.
179. Spicer, C. W., and A. Levy. The Photochemical Smog Reactivity of Diesel
Exhaust Organics. Final Report. Prepared by Battelle Columbus Laboratories,
Columbus, Ohio, for Coordinating Research Council, Inc., New York, NY.
Report No. CRC-APRAC-CAPA-10-71-1. May 1975.
180. Landen, E. W., and J. M. Perez. Some diesel exhaust reactivity information
derived by gas chromatography. New York, Society of Automotive Engineers.
SAE Paper No. 740530. 1974.
181. Bufalini, J. J., B. Gay, Jr., and K. L. Brubaker. Hydrogen peroxide
formation from formaldehyde photoxidation and its presence in urban
atmospheres. Environ. Sci. Technol. 6:816, 1972. Cited in: Reference
73. ~
182. Trijonis, J., T. Peng, G. .McRae, and L. Lees. Oxidant and precursor
trends in the metropolitan Los Angeles region. In: Proc. Int'l. Conf.
on Photochemical Oxidant Pollution and its Control. U.S. Environmental
Protection Agency, Research Triangle Park, NC. Publication No.
EPA-600/3-77-001. January 1977.
183. Mayrsohn, H., M. Kuramoto, J. H. Crabtree, R. D. Sothern, and S. H. Mano.
Atmospheric Hydrocarbon Concentrations, June-September 1975. Division of
Technical Services, California Air Resources Board. Sacramento, CA.
January 1976.
184. Meeks, S. A. Environmental Sciences Research Laboratory. Personal
communication to B. E. Til ton, Environmental Criteria and Assessment
Office. U.S. Environmental Protection Agency, Research Triangle Park,
NC. February 1980.
185. Raw or reduced data provided by W. A. Lonneman and S. A. Meeks.
Environmental Sciences Research Laboratory, U.S. Environmental Protection
Agency, Research Triangle Park, NC. May 1979.
7-15
-------
186. Lonneman, W. A. Environmental Sciences Laboratory. Memorandum to B. E.
Tilton, Environmental Criteria and Assessment Office. U.S. Environmental
Protection Agency, Research Triangle Park, NC. May 9, 1979.
187. Lonneman, W. A. Ozone and hydrocarbon measurements in recent oxidant
transport studies. In: Proc. Int'l Conf. on Photochemical Oxidant
Pollution and its Control. U.S. Environmental Protection Agency, Research
Triangle Park, NC. Publication No. EPA-600/3-77-001. January 1977.
188. Arnts, R. R., and S. A. Meeks. Biogenic Hydrocarbon Contribution to the
Ambient Air of Selected Areas. U.S. Environmental Protection Agency,
Research Triangle Park, NC. Publication No. EPA-600/3-80-023, 1980.
189. Ferman, M. Rural nonmethane hydrocarbon concentration and composition.
In: Proc. of Symposium on Atmospheric Biogenic Hydrocarbons: Emission
Rates, Concentrations, and Fates, January 8-9, 1980. J. J. Bufalini, ed.
U.S. Environmental Protection Agency, Research Triangle Park, NC. In
Press.,
190. Westberg, H., K. Allwine, and E. Robinson. Measurement of Light Hydro-
carbons and Oxidant Transport: Houston Study 1976. Prepared by Washington
State University for the U.S. Environmental Protection Agency under
Contract No. 68-02-2298. U.S. Environmental Protection Agency, Research
Triangle Park, NC. Publication No. EPA-600/3-78-062. July 1978.
191. Lonneman, W. A., G. Namie, and J. J. Bufalini. Hydrocarbons in Houston
Air. U.S. Environmental Protection Agency, Research Triangle Park, NC.
Publication No. EPA-600/3-79-018. February 1979.
192. Lonneman, W. A. Ambient hydrocarbon measurements in Houston. Paper
presented at Air Pollution Control Association Specialty Conference on
Ozone and Oxidants, October 15-17, 1979, Houston, TX.
193. Pilar, S. and W. F. Graydon. Benzene and toluene distribution in Toronto
atmosphere. Environ. Sci. Technol. 7(7):628-631, 1973.
194. Bonamassa, F., and H. Wong-Woo. Paper presented at Div. of Water, Air,
and Waste Chemistry, New York, NY. September 1966. Cited in: Reference
193.
195. Mayrsohn, H., M. Kuramoto, J. H. Crabtree, R. D. Sothern, and S. H. Mano.
Atmospheric Hydrocarbon Concentrations, June-September 1975. California
Air Resources Board, Sacramento, CA. January 1976.
196. Lahmann, E., B. Seifert, and D. Ullrich. The pollution of ambient air
and rainwater by organic components of motor-vehicle exhaust-gases. I_n:
Proc. Fourth Int'l. Clean Air Congress, Tokyo, Japan, May 16-20, 1977.
Tokyo, Japanese Union Air Poll. Prevention Assoc., 1977.
197. Preparation, Adoption, and Submittal of Implementation Plans. Revisions
to 40 CFR, Part 51. Federal Register 44(28):8234, February 8, 1979.
7-16
-------
198. Uses, Limitations, and Technical Basis of Procedures for Quantifying
Relationships Between Photochemical Oxidants and Precursors. U.S.
Environmental Portection Agency, Research Triangle Park, NC. Publication
No. EPA-450/2-77-0219. November 1977.
199. Zeldin, M. D., and J. C. Cassmassi. Development of Improved Methods for
Predicting Air Quality Levels in the South Coast Air Basin. Final Report.
Prepared by Technology Service Corporation, Santa Monica, CA, for California
Air Resources Board, Sacramento, CA. March 1979.
200. Altshuller, A. P. An evaluation of techniques for the determination of
the photochemical reactivity of organic emissions. J. Air Poll. Contr.
Assoc. 16:257, 1966. Cited in: Reference 7.
201. Office of Air Programs, Hydrocarbon Pollutant System Study. U.S.
Environmental Protection Agency, Research Triangle Park, NC. Publication
No. APTD-1499. 1972.
202. National Institute for Occupational Safety and Health. Criteria for
Recommended Standard...Occupational Exposure to Alkanes, (C5-C8).
Department of Health Education and Welfare, Nat'l. Inst. for Occup.
Safety and Health. Publication No. 77-151. Cincinnati, OH. March 1977.
203. Aviado, D. M., S. Zakhari, and T. Watanabe. Non-fluorinated Propellents
and Solvents for Aerosols. Cleveland, OH; CRC Press. 1977.
204. Reinhardt, C. F., A. Azar, M. E. Maxfield, P. E. Smith, and L. S. Mullin.
Cardiac arrhythmias and aerosol "sniffing." Arch. Environ. Health 22:
1971. " —
205. Carpenter, C., E. Kinkead, D. Geary, Jr., L. Sullivan, and J. King, XVII.
Animal response to n-nonane vapor. Toxicol. Appl. Pharmacol. 44:53-61,
1978. ~ —
206. Nau, C. , J. .Neal, and M. Throton. Cg-C-,2 Fractions obtained from petroleum
distillates. An evaluation of their potential toxicity. Arch. Environ.
Health 12:382-93, 1966.
207. National Research Council Advisory Center on Toxicology. Health Effects
on Benzene; A Review. Report No. NAS/ACT/P-829. . June 1976. Prepared
for the Environmental Protection Agency, Washington, DC. June 1976.
208. Snyder, R. and J. J. Kocsis. Current concepts of chronic benzene toxicity,
CRC Critical Reviews. Toxicology 3:265-288, 1975.
209. Office of Research and Development Office of Health and Ecological Effects.
Assessment of Health Effects of Benzene Germane to Low-Level Exposure.
U.S. Environmental Protection Agency, Washington, DC. Publication No.
EPA-600/1-78-061. September 1978.
210. Hardy, H. L., and H. G. Elkins. Medical aspects of maximum allowable
concentrations-Benzene. J. Ind. Hyg. Toxicol. 30:196-200, 1948.
7-17
-------
211. Wilson, R. Benzene poisoning in industry. J. Lab Clin. Med. 27:1517-1521,
1942.
212. Pagnatto, L. D., H. B. ETkins, H. G. Brugsch, and E. J. Walkley. Industrial
benzene exposure from petroleum naptha - I. Rubber coating industry. J.
Am. Ind. Hyg. Assoc. 22:417-21.
213. Vigliani, E. C., and G. Saita. Benzene and leukemia. New Engl. J. Med.
217:872, 1964.
214. Cornish, H. H. Solvents and vapors. In: Toxicology. The Basic Science
of Poisons. L. J. Casarett and J. Doull, eds. MacMillan Publishing Co.,
Inc., New York, 1975. pp. 503-526.
215. Deichmann, W. B., W. E. MacDonald, and E. Bernal. The hemopoietic tissue
toxicity of benzene vapors. Toxicol. Appl. Pharmacol. 5:20124, 1963.
216. Blaney, L. Early detection of benzene toxicity. Ind. Med. 19:227-228,
1950.
217. National Institute for Occupational Safety and Health. Revised Recommenda-
tion for an Occupational Exposure Standard for Benzene. Cincinnati, OH.
August 1976.
218. Emergency temporary standard for occupational exposure to benzene; Notice
of hearing. Federal Register. 42(85):22516-22529, May 3, 1977.
219. Occupational exposure to benzene. Federal Register. 43(29):5918-5970,
February 10, 1978.
220. Carcinogen Assessment Group. Assessment of Human Exposures to Atmospheric
Benzene. U.S. Environmental Protection Agency, Washington, DC. 1978.
221. Carcinogen Assessment Group. Carcinogenic Assessment of Toluene. Draft.
U.S. Environmental Protection Agency, Research Triangle Park, NC. June
1980.
222. National Institute for Occupational Safety and Health. Criteria for a
Recommended Standard...Occupational Exposure to Toluene. Cincinnati, OH.
Report No. HSM 73-11023. 1973.
223. Conn, Kanl-Heinz. Toluene - A toxicologic review. Scand. J. Work Environ.
and Health. 5:71-90, 1979.
224. Ainsworth, R. W. Petrol-vapor poisoning. Brit. Med. -J. 1:1547-1548,
1960.
225. Wang, C. C., and G. V. Irons. Acute gasoline intoxication. Arch Environ.
Health. 2:114-16, 1961.
226. Aaidin, R. Petrol-vapor poisoning. Brit. Med. J. 2:369-370, 1958.
7-18
-------
227. Nelms, R. , Jr., R. Davis, and J. Bond. Verification of fatal gasoline
intoxication in confined spaces utilizing gas-liquid chromatography. Am.
J. Clin. Pathol. 53:641-6, 1970.
228. Poklis, A. Death resulting from gasoline "sniffing." A case report. J.
Forensic Sci. Soc. 16:43-46, 1976.
229. Chenoweth, M. Ventricular fibrillation induced by hydrocarbons and
epinephrine. J. Ind. Hyg. Toxicol. 28:151-158, 1946.
230. Drinker, P., C. Yaglou, and M. Warren. The threshold toxicity of gasoline
vapor. J. Ind. Hyg. Toxicol. 25:225-232, 1943.
231. Davis, A., L. Schafch, and Z. Bell. The effects on human volunteers of
exposure to air containing gasoline vapor. Arch. Environ. Health 1:548-554,
1960.
232. Swinyard, E. A. Noxious gases and vapors Ir\: The Pharmacological Basis
of Therapeutics, 4th. Ed. L. S. Goodman and A. Oilman, eds. MacMillan
Co., Toronto. 1970. pp. 930-943.
233. McDermott, H. J., and S. E. Killiany. Quest for a Gasoline TLV. Presented
at the 1976 American Industrial Hygiene Convention, Atlanta, Georgia, May
12-16,1976. .
234. American Conference of Governmental Industrial Hygienists. Threshold
Limit Values for 1976. ACGIH, Cincinnati, OH, 1976.
235. Runion, H. E. Benzene in Gasoline II. Amer. Ind. Hyg. Assoc. J. 38:391-393,
1977. —
236. McDermott, H. J., and G. A. Vos. Service station attendants' exposure to
benzene and gasoline vapors. J. Amer. Ind. Hygiene Assoc. 40:315-321,
1979. —
237. Von Oettingen, W. Toxicity and potential dangers of aliphatic and aromatic
hydrocarbons. Public Health Service, Washington, DC. U.S. Public Health
Bulletin No. 225. 1940.
238. Gerarde, H. Toxicology and biochemistry of aromatic hydrocarbons. New
York, Elsevier Publishing Co., 1960.
239. Browning, E. Toxicity of Industrial Organic Solvents. Medical Research
Council Report 80. London, Her Majesty's Stationery Office, 1953.
240. Rector, D., B. Steadman, R. Jones, and J. Siege!. Effects on experimental
animals of long-term inhalation exposure to mineral spirits. Toxicol.
Appl. Pharmacol. 9:257-268, 1966.
241. Hine, C., and H. Zuidema. The toxicological properties of hydrocarbon
solvent. Industr. Med. 39:215-220, 1970.
7-19
-------
242. Carpenter, C., E. Kinkead, D. Geary, Jr., L. Sullivan, and J. King.
Petroleum hydrocarbon toxicity studies. I. Methodology. Toxicol. Appl.
Pharmacol. 32:246-262, 1975.
243. Carpenter, C., E. Kinkead, D. Geary, Jr., L. Sullivan, and J. King. II.
Animal and human response to vapors of varnish makers and painters'
naphtha. Toxicol. Appl. Pharmacol. 32:263-281, 1975.
244. Carpenter, C., E. Kinkead, D. Geary, Jr., L. Sullivan, and J. King. III.
Animal and human response to vapors of Stoddard solvent. Toxicol. Appl.
Pharmacol. 32:282-297, 1975.
245. Carpenter, C., E. Kinkead, D. Geary, Jr., L. Sullivan, and J. King. IV.
Animal and human response to vapors of rubber solvent. Toxicol. Appl.
Pharmacol. 33:526-542, 1975.
246. Carpenter, C., E. Kinkead, D. Geary, Jr., L. Sullivan, and J. King. V.
Animal.and human response to vapors of mixed xylenes. Toxicol. Appl.
Pharmacol. 33:543-558, 1975.
247. Carpenter, C., E. Kinkead, D. Geary, Jr., L. Sullivan, and J. King. VI.
Animal and human responses to vapors of "60 solvent." Toxicol. Appl.
Pharmacol. 34:374-394, 1975.
248. Carpenter, C., E. Kinkead, D. Geary, Jr., L. Sullivan, and J. King. VII.
Animal and human response to vapors of "70 solvent." Toxicol. Appl.
Pharmacol. 34:395-412, 1975.
249. Carpenter, C., E. Kinkead, D. Geary, Jr., L. Sullivan, and J. King. VIII.
Animal and human response to vapors of "140° flash aliphatic solvent."
Toxicol. Appl. Pharmacol. 34:413-429, 1975.
250. Carpenter, C., E. Kinkead, D. Geary, Jr., L. Sullivan, and J. King. IX.
Animal and human response to vapors of "80 thinner." Toxicol. Appl.
Pharmacol. 36:409-425, 1976.
251. Carpenter, C., E. Kinkead, D. Geary, Jr., L. Sullivan, and J. King. X.
Animal and human response to vapors of "50 thinner." Toxicol. Appl.
Pharmacol. 36:427-442, 1976. .
252. Carpenter, C., E. Kinkead, D. Geary, Jr., L. Sullivan, and J. King. XI.
Animal and human response to vapors of deodorized kerosene. Toxicol.
Appl. Pharmacol. 36:443-456, 1976."
253. Carpenter, C., E. Kinkead, D. Geary, Jr., L. Sullivan, and J. King. XII.
Animal and human response to vapors of "40 thinner." Toxicol. Appl.
Pharmacol. 36:457-472, 1976.
254. Carpenter, C., E. Kinkead, D. Geary, Jr., L. Sullivan, and J. King. XIII.
Animal and human response to vapors of toluene concentrate. Toxicol.
Appl. Pharmacol. 36:473-490, 1976.
7-20
-------
255. Carpenter, C., E. Kinkead, D. Geary, Jr., L, Sullivan, and.J. King. XIV.
Animal and Human Response to vapors of high aromatic solvent. Toxicol
Appl. Pharmacol. 41:235-249.
256. Carpenter, C., E. Kinkead, D. Geary, Jr., L. Sullivan, and J. King. XV.
Animal response to vapors of high naphthenic solvent. Toxicol. Appl.
Pharmacol. 41:251-260, 1977.
257. Carpenter, C., E. Kinkead, D. Geary, Jr., L. Sullivan, and J. King. XVI.
Animal response to vapors of naphthenic aromatic solvent. Toxicol. Appl
Pharmacol. 41:261-270, 1977.
258. Clayton, G. D., G. D. Clayton and Assoc., Inc. Prepared for Auto Mfq.
Assoc., Detroit, MI, 1966.
259. Das Gupta, S. N. Proc. Indian SciV Congr., No. 2. 1957. p. 88. Cited in-
Reference 51. -—
260. Stahl, Q. R. Preliminary Air Pollution Survey of Ethylene: A Literature
Review. Nat'l. Air Pollution Control Admin., Public Health Service,
Department of Health, Education and Welfare. Cincinnati, OH. Publication
No. APTD 69-35. 1969.
261. Abeles, F. B., L. E. Craker, L. E. Forrence, and C. R. Leather. Fate of
air pollutants: Removal of ethylene, sulfur dioxide, and nitrogen dioxide
by soil. Science. 173:914, 1971.
262. Leighton, P. A. Photochemistry of Air Pollution. Academic Press, New
York, NY, 1961.
263. Hanan, J. J. Ethylene dosages in Denver and marketability of cut-flower
carnations. J. Air Poll. Contr. Assoc. 23(6):522-524, 1973.
264. Piersol, J. R., and J. J. Hanan. Effect of ethylene on carnation growth.
J. Amer. Soc. Horticult. Sci. 100(6):679-681, 1975.
7-21
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
REPORT NO.
EPA/600/8-80-045
3. RECIPIENT'S ACCESSION MO.
TITLE AND SUBTITLE
REVIEW OF CRITERIA FOR VAPOR-PHASE HYDROCARBONS-
5. REPORT DATE
August 1980
6. PERFORMING ORGANIZATION CODE
AUTHOR(S)
Beverly E. Til ton and Robert M. Bruce, Ph.D.
8. PERFORMING ORGANIZATION REPORT NO.
PERFORMING ORGANIZATION NAME AND ADDRESS
Environmental Criteria and Assessment Office
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
12. SPONSORING AGENCY NAME AND ADDRESS
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
Preprint '
14. SPONSORING AGENCY CODE
EPA/600/00
is.SUPPLEMENTARY NOTES This document formerly was announced in 45FR 15262, March 10,
1980, as "Facts and Issues Associated with the Need for a Hydrocarbon Criteria
Document."
16. ABSTRACT
Information on vapor-phase hydrocarbons presented in this document covers
basic atmospheric chemistry relative to secondary products, especially ozone; sources
and emissions; ambient air concentrations; relationship of precursor hydrocarbons to
resultant ozone levels in ambient air; health effects; and welfare effects. The
principal conclusions from this document are as follow. Hydrocarbons are a principal
contributor to the formation of ozone and other photochemical oxidants; however,
no fixed single quantitative relationship between precursor hydrocarbons and resulting
ozone concentrations can be defined. This relationship varies from site to site
depending on local precursor mixes, transport considerations, and meteorological
factors. Consequently no single quantitative relationship can be defined nationwide.
While specific hydrocarbon compounds can be of concern to public health and welfare,
as a class this group of materials cannot be considered a hazard to human health
or welfare at or even well above those concentrations observed in the ambient air.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
COSATI Field/Group
hydrocarbons/gases/alkanes/alkene hydrocarbons/ozone
aliphatic hydrocarbons/aromatic hydrocarbons/aerosols
ethene/isoprene/terpene hydrocarbons/plant growth
air pollution/sources/internal combustion engines
fuels/gasoline/stationary engines/diesel engines
human health effects/blood diseases/leukemia/asphyxia
nervous system dlsorders/benzene/toluene/xylenes
chemical reactions/atmospheric composition/
gas chromatography/measurement
photochemical
model s
AQSM
EKMA
acid rain
natural sources
manmade sources
fnol g
smog
04B
06A
06C
06E
06F
07C
13B
18. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19. SECURITY CLASS (This Report)
UNCLASSIFIED
21. NO. OF PAGES
20. SECURITY CLASS (Thispage)
UNCLASSIFIED
22. PRICE
EPA Form 2220-1 (Rev. 4-77)
PREVIOUS EDITION IS OBSOUETE
7-22
-------
-------
------- |