vvEPA
United States
Environmental Protection
Agency
Office of Research and
Development
Washington DC 20460
EPA/600/8-90/057Ba
December 1994
External Review Draft
Health Assessment
Document for
Diesel Emissions
Volume I of II
Review
Draft
(Do Not
Cite or
Quote)
Notice
This document is a preliminary draft. It has not been formally
released by EPA and should not at this stage be construed to
represent Agency policy. It is being circulated for comment on its
technical accuracy and policy implications.
-------
DRAFT-DO NOT QUOTE OR CITE
EPA/600/8-90/057Ba
December 1994
External Review Dra
Health Assessment Document
for Diesel Emissions
Volume I of II
NOTICE
This document is a preliminary draft. It has not been formally
released by EPA and should not at this stage be construed to
represent Agency policy. It is being circulated for comment on
Its technical accuracy and policy implications.
Environmental Criteria and Assessment Office
Office of Health and Environmental Assessment
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
n s Environmental Protection Agency
Region 5 Library (PL-12J) „ ® P^ted on Pecycled Paper
?fwest JacS Boulevard, 12th Floor
Chicago, IL 60604-3590
-------
DISCLAIMER
This document is an external draft for review purposes only and does not constitute
U.S. Environmental Protection Agency policy. Mention of trade names or commercial
products does not constitute endorsement or recommendation for use.
December 1994 I-ii DRAFT-DO NOT QUOTE OR CITE
-------
Health Assessment Document for Diesel Emissions
TABLE OF CONTENTS
Volume I
1. EXECUTIVE SUMMARY 1-1
2. DIESEL EMISSIONS 2-1
3. DIESEL-DERIVED POLLUTANTS: ATMOSPHERIC
CONCENTRATIONS, TRANSPORT, AND TRANSFORMATIONS . . 3-1
4. DOSIMETRIC FACTORS 4-1
5. NONCANCER HEALTH EFFECTS OF DIESEL EXHAUST 5-1
6. QUALITATIVE AND QUANTITATIVE ASSESSMENT OF
NONCANCER HEALTH-EFFECTS-DERIVATION OF THE
INHALATION REFERENCE CONCENTRATION 6-1
Volume II
7. CARCINOGENICITY OF DIESEL EMISSIONS IN LABORATORY
ANIMALS 7-1
8. EPIDEMIOLOGIC STUDIES OF THE CARCINOGENICITY OF
EXPOSURE TO DIESEL EMISSIONS 8-1
Addendum to Chapter 8 8A-1
9. MUTAGENICITY 9-1
10. METABOLISM AND MECHANISM OF ACTION IN DIESEL
EMISSION-INDUCED CARCINOGENESIS 10-1
11. QUALITATIVE AND QUANTITATIVE EVALUATION OF THE
CARCINOGENICITY OF DIESEL ENGINE EMISSIONS 11-1
12. HEALTH RISK CHARACTERIZATION FOR DIESEL ENGINE
EMISSIONS 12-1
APPENDIX A: EXPERIMENTAL PROTOCOL AND COMPOSITION OF
EXPOSURE ATMOSPHERES A-l
December 1994 I_iii DRAFT-DO NOT QUOTE OR CITE
-------
TABLE OF CONTENTS (cont'd)
APPENDIX B: CONTRACTOR REPORT: ASSESSMENT OF RISK
FROM EXPOSURE TO DIESEL ENGINE EMISSIONS . . B-l
APPENDIX C: ALTERNATIVE MODEL FOR DIESEL CANCER RISK
ASSESSMENT C-l
APPENDIX D: MODELS FOR CALCULATING LUNG BURDENS .... D-l
December 1994 I-iv DRAFT-DO NOT QUOTE OR CITE
-------
TABLE OF CONTENTS
VOLUME I
Page
LIST OF TABLES I-x
LIST OF FIGURES I-xiii
AUTHORS, REVIEWERS, AND CONTRIBUTORS I-xv
ACKNOWLEDGMENTS I-xviii
1. EXECUTIVE SUMMARY 1-1
1.1 INTRODUCTION 1-1
1.2 DIESEL EMISSIONS 1-3
1.3 DIESEL-DERIVED POLLUTANTS: ATMOSPHERIC
CONCENTRATIONS, TRANSPORT, AND
TRANSFORMATIONS 1-7
1.4 DOSIMETRIC FACTORS IN ASSESSING CARCINOGENIC
RISK OF EXPOSURE TO DIESEL ENGINE EMISSIONS .... 1-9
1.5 NONCANCER HEALTH EFFECTS OF DIESEL EXHAUST .. 1-11
1.6 QUANTITATIVE AND QUALITATIVE ASSESSMENT OF
NONCANCER HEALTH EFFECTS—DERIVATION OF THE
INHALATION REFERENCE CONCENTRATION 1-15
1.7 CARCINOGENICITY OF DIESEL ENGINE EMISSIONS IN
LABORATORY ANIMALS 1-16
1.8 EPIDEMIOLOGIC STUDIES OF DIESEL EXHAUST
CARCINOGENICITY 1-18
1.9 MUTAGENICITY OF DIESEL ENGINE EMISSIONS 1-23
1.10 METABOLISM AND MECHANISM OF ACTION OF DIESEL
EMISSIONS-INDUCED CARCINOGENICITY 1-23
1.11 WEIGHT OF EVIDENCE CLASSIFICATION FOR
CARCINOGENICITY AND QUANTITATIVE ESTIMATE OF
UNIT RISK 1-25
1.12 RISK CHARACTERIZATION 1-27
REFERENCES 1_28
2. DIESEL EMISSIONS 2-1
2.1 INTRODUCTION ' .' 2-1
2.2 OVERVIEW OF DIESEL POLLUTANTS AND POLLUTANT
FORMATION 2-4
2.2.1 Gas-Phase Pollutant Emissions 2-4
2.2.1.1 Oxides of Nitrogen Formation 2-4
2.2.1.2 Hydrocarbon and Carbon Monoxide
Formation 2-5
2.2.2 Particle Formation and Emission 2-6
2.2.3 Gas-to-Particle Conversion 2-7
2.2.3.1 Condensation of Organic Matter 2-7
December 1994 j.v DRAFT-DO NOT QUOTE OR CITE
-------
TABLE OF CONTENTS (cont'd)
2.2.3.2 Oxidation of Sulfur Oxides 2-7
2.2.4 Nitroarene Formation 2-8
2.3 EMISSIONS FACTORS AND EMISSIONS INVENTORIES ... 2-9
2.3.1 Gaseous Pollutant Emission Factors 2-9
2.3.2 Paniculate Matter Emissions Factors 2-15
2.4 DIESEL ENGINE CONTROL TECHNOLOGY 2-17
2.4.1 Engine Modifications 2-17
2.4.2 Engine Control Systems 2-17
2.4.3 Turbocharging and Intercooling 2-18
2.4.4 Intake Manifold Tuning 2-20
2.4.5 Lubricating Oil Control 2-20
2.4.6 Aftertreatment Systems 2-21
2.4.7 Fuel Modifications 2-22
2.4.7.1 Fuel Additives 2-23
REFERENCES 2-24
3. DIESEL-DERIVED POLLUTANTS: ATMOSPHERIC
CONCENTRATIONS, TRANSPORT, AND TRANSFORMATIONS . . 3-1
3.1 INTRODUCTION 3-1
3.2 PRIMARY DIESEL EMISSIONS 3-1
3.2.1 Gaseous Emissions 3-2
3.2.2 Paniculate Emissions 3-6
3.2.2.1 Diesel Paniculate Matter 3-6
3.2.2.2 Paniculate Phase Inorganics 3-8
3.2.2.3 Paniculate Phase Organic Compounds 3-8
3.2.3 Gaseous/Paniculate Phase Emission Partitioning
of Polycyclic Aromatic Hydrocarbons 3-16
3.3 ATMOSPHERIC TRANSFORMATIONS OF PRIMARY DIESEL
EMISSIONS 3-18
3.3.1 Long-Range Transport and Fate of Primary Diesel
Emissions 3-18
3.3.2 Chemical Transformations 3-21
3.3.2.1 Gas-Phase Reactions 3-21
3.3.2.2 Particulate-Phase Reactions 3-36
3.3.3 Physical Removal Processes 3-41
3.3.3.1 Dry Deposition 3-41
3.3.3.2 Wet Deposition 3-43
3.4 ATMOSPHERIC CONCENTRATIONS OF PRIMARY
DIESEL EMISSIONS AND THEIR TRANSFORMATION
PRODUCTS 3-44
3.4.1 Volatile Organic Compounds Attributable
to Traffic 3-46
3.4.2 Polycyclic Aromatic Hydrocarbons 3-48
3.4.3 Nitroarene Concentrations in Ambient Air 3-51
December 1994 I-vi DRAFT-DO NOT QUOTE OR CITE
-------
TABLE OF CONTENTS (cont'd)
3.4.4 Need for Atmospheric Tracers of Diesel
Emissions 3-54
3.5 BIOASSAY-DIRECTED CHEMICAL ANALYSIS 3-60
3.6 SUMMARY 3-65
REFERENCES 3-68
4. DOSIMETRIC FACTORS 4-1
4.1 INTRODUCTION 4-1
4.2 REGIONAL DEPOSITION OF INHALED PARTICLES 4-2
4.2.1 Physical Processes, Physiological/Anatomical
Considerations, and Particle Characteristics 4-3
4.2.2 Species Variability in Regional Dose 4-4
4.3 RESPIRATORY TRACT CLEARANCE RATES 4-6
4.3.1 Tracheobronchial Clearance 4-6
4.3.2 Clearance from the Alveolar Region 4-11
4.3.2.1 Alveolar Clearance in Humans 4-11
4.3.2.2 Alveolar Clearance in Animals 4-11
4.3.2.3 Lung Burden and Pulmonary Overload
Resulting in Impaired Clearance 4-15
4.3.3 Role of Alveolar Macrophages in the Clearance
of Paniculate Matter 4-17
4.3.3.1 Alveolar Macrophage-Mediated Clearance
of Paniculate Matter 4-17
4.3.3.2 Translocations of Particles to Extraalveolar
Macrophage Compartment Sites 4-18
4.3.3.3 Potential Mechanisms for an Alveolar
Macrophage Sequestration Compartment for
Particles During Particle Overload 4-21
4.4 BIOAVAILABILITY OF ORGANIC CONSTITUENTS
PRESENT ON DIESEL EXHAUST PARTICLES 4-22
4.4.1 Whole-Animal Studies 4-22
4.4.2 Extraction of Diesel Particle Associated Organics in
Biological Fluids 4-25
4.4.3 Extraction of Diesel Particle Associated Organics
by Alveolar Lung Cells and Other Cell Types 4-27
4.4.4 Bioavailability of Adsorbed Compounds as a Function
of Particle Clearance Rates and Extraction Rates of
Adsorbed Compounds 4-28
4.5 CONSIDERATIONS FOR DOSIMETRY MODELING 4-31
4.6 SUMMARY 4.33
REFERENCES 4.35
December 1994 I.vii DRAFT-DO NOT QUOTE OR CITE
-------
TABLE OF CONTENTS (cont'd)
5. NONCANCER HEALTH EFFECTS OF DIESEL EXHAUST 5-1
5.1 HEALTH EFFECTS OF WHOLE EXHAUST 5-1
5.1.1 Human Data 5-1
5.1.1.1 Short-Term Exposures 5-1
5.1.1.2 Diesel Exhaust Odor 5-2
5.1.1.3 Long-Term Exposure 5-6
5.1.2 Animal Studies 5-11
5.1.2.1 Acute Exposures 5-11
5.1.2.2 Short-Term Exposures 5-16
5.1.2.3 Chronic Exposures 5-20
5.2 COMPARISON OF HEALTH EFFECTS OF FILTERED AND
UNFILTERED DIESEL EXHAUST 5-65
5.3 INTERACTIVE EFFECTS OF DIESEL EXHAUST 5-69
5.4 COMPARISON OF THE EFFECTS OF DIESEL EXHAUST
AND GASOLINE EXHAUST 5-72
5.5 DOSE-RATE AND PARTICULATE CAUSATIVE ISSUES . . . 5-78
5.6 SUMMARY AND DISCUSSION 5-81
5.6.1 Effects of Diesel Exhaust on Humans 5-81
5.6.2 Effects of Diesel Exhaust on Animals 5-83
5.6.2.1 Effects on Survival and Growth 5-84
5.6.2.2 Effects on Pulmonary Function 5-84
5.6.2.3 Histopathological and Histochemical
Effects 5-85
5.6.2.4 Effects on Defense Mechanisms 5-85
5.6.2.5 Neurological and Behavioral Effects 5-86
5.6.2.6 Other Noncancerous Effects 5-87
5.6.3 Comparison of Filtered and Unfiltered Diesel
Exhaust 5-87
5.6.4 Interactive Effects of Diesel Exhaust 5-87
5.6.5 Comparisons with Gasoline Exhausts 5-88
5.6.6 Summary 5-89
REFERENCES 5-90
6. QUALITATIVE AND QUANTITATIVE ASSESSMENT OF
NONCANCER HEALTH EFFECTS-DERIVATION OF THE
INHALATION REFERENCE CONCENTRATION 6-1
6.1 INTRODUCTION 6-1
6.2 QUALITATIVE EVALUATION OF DIESEL EXHAUST
EMISSIONS 6-2
6.3 APPROACH FOR DERIVATION OF THE INHALATION
REFERENCE CONCENTRATION 6-3
6.4 THE PRINCIPAL STUDIES FOR INHALATION REFERENCE
CONCENTRATION DERIVATION 6-7
6.5 SUPPORTING STUDIES FOR INHALATION REFERENCE
CONCENTRATION DERIVATION 6-11
December 1994 I_viii DRAFT-DO NOT QUOTE OR CITE
-------
TABLE OF CONTENTS (cont'd)
6.6 DERIVATION OF THE INHALATION REFERENCE
CONCENTRATION 6-15
6.7 SUMMARY 6-17
REFERENCES 6-19
December 1994 j.ix DRAFT-DO NOT QUOTE OR CITE
-------
LIST OF TABLES
Number page
2-1 Some Nitroarenes Identified in Vehicle Exhaust 2-10
2-2 Mobile 4 and Chassis Dynamometer Emissions, Urban Average
Vehicle Speed, 1980 Calendar Year 2-14
3-1 Emission Rates of Volatile Organic Compounds from Diesel
and Gasoline Engines 3-4
3-2 Paniculate Matter Emission Rates and Their Distribution
Between Total Carbon and Organic Carbon for Heavy- and
Light-Duty Diesel and Gasoline Engines 3-7
3-3 Summary of Composition and Emission Rates of Airborne
Paniculate Matter from On-Road Vehicles, Tuscarora
Mountain Tunnel 1977 Experiment 3-9
3-4 Classes of Organic Compounds Identified in Particulate-Phase
Combustion Emissions 3-10
3-5 Polycyclic Aromatic Hydrocarbons Identified and
Quantified in Extracts of Diesel Particles 3-12
3-6 Emission Rates of Particle-Bound Polycyclic Aromatic
Hydrocarbons from Heavy- and Light-Duty Diesel and
Gasoline Engines 3-14
3-7 Concentrations of Nitro-Polycyclic Aromatic
Hydrocarbons Identified hi a Light-Duty Diesel
Paniculate Extract 3-15
3-8 Vapor Pressures at 25 °C for a Series of Polycyclic
Aromatic Hydrocarbons 3-17
3-9 Calculated Atmospheric Lifetimes for Gas-Phase
Reactions of Selected Compounds Present in Automotive
Emissions with Atmospherically Important Reactive
Species 3-23
3-10 Summary of the Nitroarenes Produced from the Gas-Phase
Hydroxyl Radical-Initiated and Dinitrogen Pentoxide Reactions
and Electrophilic Nitration of Polycyclic Aromatic
Hydrocarbons 3-29
December 1994 I-x DRAFT-DO NOT QUOTE OR CITE
-------
LIST OF TABLES (cont'd)
3-11 Average Atmospheric Lifetimes of Particles Due to Dry
Deposition 3-42
3-12 Examples of Dry Deposition Velocities 3-42
3-13 Mean Particle, Gas, and Overall Scavenging Ratios for
Neutral Organic Compounds 3-45
3-14 Concentrations of Individual Hydrocarbons and
Aldehydes Measured in the Raleigh, NC, Roadside Study 3-47
3-15 Particle- and Vapor-Phase Polycyclic Aromatic
Hydrocarbon Concentrations for Baltimore Harbor
Tunnel Samples 3-49
3-16 Average Ambient Concentrations of Polycyclic Aromatic
Hydrocarbons Measured in Glendora, CA 3-50
3-17 The Maximum Concentrations of Nitrofluoranthene and
Nitropyrene Isomers Observed at Three South Coast Air
Basin Sampling Sites 3-52
3-18 Contribution of Nitrofluoranthene Isomers to the
Direct Mutagenicity of Ambient Paniculate Extracts 3-66
4-1 Predicted Doses of Inhaled Diesel Exhaust Particles per
Minute Based on Total Lung Volume, Total Airway
Surface Area, or Surface Area in Alveolar Region 4-7
5-1 Human Studies of Diesel Exhaust Exposure 5-12
5-2 Short-Term Effects of Diesel Exhaust on Laboratory
Animals 5-17
5-3 Effects of Chronic Exposures to Diesel Exhaust on Survival
and Growth of Laboratory Animals 5-21
5-4 Effects of Chronic Exposures to Diesel Exhaust on
Organ Weights and Organ-to-Body-Weight Ratios 5-22
5-5 Effects of Diesel Exhaust on Pulmonary Function of
Laboratory Animals 5.24
5-6 Histopathological Effects of Diesel Exhaust in the
Lungs of Laboratory Animals 5_29
December 1994 !_xi DRAFT-DO NOT QUOTE OR CITE
-------
LIST OF TABLES (cont'd)
5-7 Effects of Exposure to Diesel Exhaust on the Pulmonary
Defense Mechanisms of Laboratory Animals 5-39
5-8 Effects of Exposures to Diesel Exhaust on the Immune System
of Laboratory Animals 5-47
5-9 Effects of Exposure to Diesel Exhaust on the Liver of Laboratory
Animals 5-50
5-10 Effects of Exposure to Diesel Exhaust on the Hematological
and Cardiovascular Systems of Laboratory Animals 5-51
5-11 Effects of Chronic Exposures to Diesel Exhaust on Serum
Chemistry of Laboratory Animals 5-55
5-12 Effects of Chronic Exposures to Diesel Exhaust on
Microsomal Enzymes of Laboratory Animals 5-57
5-13 Effects of Chronic Exposures to Diesel Exhaust on
Behavior and Neurophysiology 5-61
5-14 Effects of Chronic Exposures to Diesel Exhaust on
Reproduction and Development in Laboratory Animals 5-63
5-15 Composition of Exposure Atmospheres in Studies Comparing
Unfiltered and Filtered Diesel Exhaust 5-66
5-16 Emission Rates for Diesel and Gasoline Engines 5-73
5-17 Exposure Atmosphere for Chronic Toxicity Study in
Beagle Dogs 5-76
6-1 Human Equivalent Continuous Concentrations from the
Principal Studies 6-11
December 1994 I-xii DRAFT-DO NOT QUOTE OR CITE
-------
LIST OF FIGURES
Number Page
2-1 Full load diesel engine pressure trace 2-3
2-2 Mobile 4 projections: Carbon monoxide emissions—heavy-duty
diesel 2-13
2-3 Mobile 4 projections: Hydrocarbon emissions—heavy-duty
diesel 2-13
2-4 Mobile 4 projections: Nitrogen oxide emissions—heavy-duty
diesel 2-14
3-1 Typical size distribution of diesel exhaust particles 3-7
3-2 Vapor/particle phase polycyclic aromatic hydrocarbon
distribution in samples collected in Baltimore Harbor
Tunnel 3-19
3-3 Diesel-derived pollutants: Emission-to-deposition
atmospheric cycle 3-20
3-4 Mass chromatograms of the molecular ion of the
nitrofluoranthenes and nitropyrenes formed from the gas-phase
reaction of fluoranthene and pyrene with the hydroxyl
radicals and present in the ambient particulate sample collected
at Torrance, CA 3.53
3-5 Mass chromatograms of the molecular ion of the
nitrofluoranthenes and nitropyrenes present in ambient particulate
samples collected in Torrance, CA, and Claremont, CA 3-55
3-6 Protocol for bioassay-directed chemical analysis 3-62
3-7 Distribution of direct-acting mutagenicity between
moderately polar and polar fractions of extracts of
particulate matter in ambient air, wood smoke, and
exhaust from heavy-duty and light-duty motor vehicles 3-65
4-1 Deposition distribution patterns of inhaled diesel exhaust
particles in the airways of different species 4-6
December 1994 I.xiii DRAFT-DO NOT QUOTE OR CITE
-------
LIST OF FIGURES (cont'd)
4-2 Clearance of insoluble particles deposited in tracheobronchial
and alveolar regions 4-8
4-3 Short-term thoracic clearance of inhaled particles as determined
by model prediction and experimental measurement 4-9
December 1994 I-xiv DRAFT-DO NOT QUOTE OR CITE
-------
AUTHORS, REVIEWERS, AND CONTRIBUTORS
Authors
Dr. Ronald Bradow
Atmospheric Research and Exposure
Assessment Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC
Dr. Chao Chen
Human Health Assessment Group
U.S. Environmental Protection Agency
Washington, DC
Dr. Kenney Crump
Clement International Corporation
Ruston, LA
Dr. Daniel Guth
Environmental Criteria and Assessment
Office
U.S. Environmental Protection Agency
Research Triangle Park, NC
Dr. John Johnson
Houghton, MI
Dr. Aparna Koppikar
Human Health Assessment Group
U.S. Environmental Protection Agency
Washington, DC
Ms. Tammie Lambert
Clement International Corporation
Ruston, LA
Dr. Bruce Lehnert
Pulmonary Biology-Toxicology Section
Los Alamos National Laboratory
Los Alamos, NM
Dr. Samuel Lestz
State College, PA
Dr. Trent Lewis
Cincinnati, OH
Dr. Kumar Menon
Pikesville, MD
Dr. Gimter Oberdorster
Department of Biophysics
University of Rochester Medical Center
Rochester, NY
Dr. Dennis Opresko
Biomedical and Environmental Information
Analysis
Health and Safety Research Division
Oak Ridge National Laboratory
Oak Ridge, TN
Dr. William Pepelko
Human Health Assessment Group
U.S. Environmental Protection Agency
Washington, DC
Mr. Chris Rambin
Clement International Corporation
Ruston, LA
Dr. Larry Valcovic
U.S. Environmental Protection Agency
Washington, DC
Mr. Michael Walsh
Arlington, VA
Dr. Ronald K. Wolff
Lilly Research Laboratories
Greenfield, IN
Dr. Robert A. Young
Biomedical and Environmental Information
Analysis
Health Sciences Research Division
Oak Ridge National Laboratory
Oak Ridge, TN
December 1994
I-xv DRAFT-DO NOT QUOTE OR CITE
-------
AUTHORS, REVIEWERS, AND CONTRIBUTORS (cont'd)
Authors (cont'd)
Dr. Chia-Ping Yu
State University of New York at Buffalo
Department of Mechanical and Aerospace
Engineering
Buffalo, NY
Dr. Barbara Zielinska
Desert Research Institute
Energy and Environmental Engineering
Center
Reno, NV
Project Managers
Mr. William Ewald
Environmental Criteria and Assessment
Office
U.S. Environmental Protection Agency
Research Triangle Park, NC
Dr. William Pepelko
Office of Health and Environmental
Assessment
U.S. Environmental Protection Agency
Washington, DC
Reviewers and Contributors
The following individuals reviewed the current and/or an earlier draft of this document and
participated in a peer review workshop on July 18 and 19, 1990.
Dr. Roy Albert
University of Cincinnati
Cincinnati, OH
Dr. James Bond
Chemical Industries Institute of
Toxicology
Research Triangle Park, NC
Dr. Glen Cass
California Institute of Technology
Pasadena, CA
Dr. Eric Garshik
Harvard Medical School
Charming Laboratory
Boston, MA
Dr. Judith Graham
Environmental Criteria and Assessment
Office
U.S. Environmental Protection Agency
Research Triangle Park, NC
Dr. Uwe Heinrich
Department of Environmental Hygiene
Fraunhofer Institute
Hanover, Germany
Dr. Joellen Lewtas
Health Effects Research Laboratory
Research Triangle Park, NC
Dr. Joe Mauderly
Lovelace Inhalation Research Institute
Albuquerque, NM
December 1994
I-xvi
DRAFT-DO NOT QUOTE OR CITE
-------
AUTHORS, REVIEWERS, AND CONTRIBUTORS (cont'd)
Reviewers and Contributors (cont'cD
Dr. Roger McClellan
Chemical Industries Institute of
Toxicology
Research Triangle Park, NC
Dr. Fred Miller
Duke University Medical Center
Durham, NC
Dr. Otto Raabe
University of California
Davis, CA
Mr. Charles Ris
Human Health Assessment Group
U.S. Environmental Protection Agency
Washington, DC
Dr. Herbert Rosenkranz
Department of Environmental Sciences
School of Medicine
Case Western University
Cleveland, OH
Dr. Irving Salmeen
Ford Motor Company
Scientific Research Lab
Dearborn, MI
Dr. Andrew Sivak
Cambridge, MA
Dr. Jeanette Wiltse
Office of Health and Environmental
Assessment
U.S. Environmental Protection Agency
Washington, DC
Dr. Thomas Smith
University of Massachusetts Medical
Center
Worchester, MA
Dr. Frank Speizer
Charming Laboratory
Boston, MA
Dr. Leslie Stayner
National Institute for Occupational Safety
and Health - Taft Labs
Cincinnati, OH
Dr. Werner Stober
Chemical Industries Institute of
Toxicology
Research Triangle Park, NC
Dr. Jaroslav Vostal
General Motors Research Labs
Warren, MI
December 1994
I-xvii DRAFT-DO NOT QUOTE OR CITE
-------
ACKNOWLEDGMENTS
Word Processing Support
Ms. Glenda Johnson
Biomedical and Environmental
Information Analysis
Health Sciences Research Division
Oak Ridge National Laboratory
Oak Ridge, TN
Document Production
Ms. Marianne Barrier
Mr. John Barton
Ms. Sheila Lassiter
Ms. Wendy Lloyd
Ms. Edie Smith
ManTech Environmental Technology, Inc.
Research Triangle Park, NC
References
Mr. Douglas Fennell
Environmental Criteria and Assessment
Office
U.S. Environmental Protection Agency
Research Triangle Park, NC
Ms. Catherine Carter
Ms. Blythe Hatcher
Information Organizers, Inc.
Research Triangle Park, NC
Reprographics
Mr. Richard Wilson
Environmental Criteria and Assessment
Office
U.S. Environmental Protection Agency
Research Triangle Park, NC
December 1994
I-xviii DRAFT-DO NOT QUOTE OR CITE
-------
i 1. EXECUTIVE SUMMARY
2
3
4 1.1 INTRODUCTION
5 Diesel engines are used to power ships, heavy machinery, locomotives, and heavy duty
6 trucks, as well as a small number of light-duty passenger cars and trucks. Since passage of
7 the Clean Air Act of 1963 and the Motor Vehicle Air Pollution Control Act of 1965, efforts
8 have been made by the federal government to limit pollutant emissions from mobile sources
9 to levels that will be protective of human health and the environment. Regulation of
10 emissions from gasoline- and diesel-powered vehicles falls under the authority of the U.S.
11 Environmental Protection Agency's (EPA's) Office of Ah- and Radiation, Office of Mobile
12 Sources. The EPA particle emission standards for light-duty diesel (LDD) vehicles and
13 trucks were promulgated in 1980 and became effective in 1982. In 1987, the particle
14 emission standard for LDD vehicles became 0.20 g/mile, and the standard for LDD trucks
15 became 0.26 g/mile. For heavy-duty diesel (HDD) vehicles, a particle emission standard of
16 0.6 g/brake horsepower-hour (g/bhp-h) took effect in 1988. In 1991, the standards became
17 0.25 g/bhp-h for both urban buses as well as all other HDD vehicles. In 1993, the standard
18 for urban buses was lowered to 0.10 g/bhp-h. During 1994, standards were decreased to
19 0.07 g/bhp-h for urban buses and 0.10 g/bhp-h for other HDD vehicles. In 1995, the
20 standard for urban buses will decrease to 0.05 g/bhp-h.
21 The continued use of diesel vehicles has raised concerns regarding the potential health
22 and environmental effects associated with diesel exhaust. Polycyclic aromatic hydrocarbons
23 (PAHs) are adsorbed onto the diesel exhaust particles, and several of these, such as
24 benzo[a]pyrene (B[a]P) and 1-nitropyrene (1-NP), have been shown to be carcinogenic in
25 animal experiments. Moreover, recent evidence has indicated that the particles themselves
26 may have intrinsic toxic and carcinogenic properties. The particles average about 0.2 jzm in
27 diameter.
28 Suggestive evidence for a carcinogenic effect of inhaled diesel exhaust in animals was
29 first reported in 1983 in studies demonstrating an increased incidence of pulmonary
30 adenomas in Sencar mice treated with the tumor promotor urethane and subsequently exposed
31 to diesel exhaust. Recent studies with rats have since demonstrated a clear association
December 1994 !_! DRAFT-DO NOT QUOTE OR CITE
-------
1 between chronic inhalation of diesel exhaust at high concentrations (2 to 7 mg/m3) and
2 increased incidences of lung tumors.
3 Evidence for the potential carcinogenicity of diesel exhaust in humans is limited; based
4 on recent studies which have indicated a small but significant increased risk of lung cancer in
5 occupationally exposed workers.
6 In addition to the potential carcinogenicity of diesel exhaust, there has also been some
7 concern that diesel particulate matter (PM) may contribute to other health problems,
8 especially those associated with the respiratory tract. Respirable particles such as those in
9 diesel exhaust have been implicated as etiological factors in various types of chronic lung
10 disease. They may also increase the lung's susceptibility to bacterial and viral infections,
11 aggravate preexisting diseases such as bronchitis or emphysema, or aggravate specific
12 respiratory conditions such as bronchial asthma. There is also some evidence for adverse
13 behavioral and neurological effects. Other components of diesel exhaust, such as sulfur
14 dioxide (SO2), nitrogen dioxide (NO2), formaldehyde, acrolein, and sulfuric acid may
15 contribute to some of these potential health effects.
16 Several EPA analyses have been conducted to assess the economic, environmental, and
17 health impacts of the current and proposed changes in the diesel emissions standards,
18 especially as they pertain to PM. These analyses indicate that emissions of diesel PM in the
19 United States in 1986 totaled 274,000 metric tons, or about 3.9% of the total suspended
20 particle emissions from all sources. As a result of new and proposed mobile source emission
21 standards, diesel particle emissions were projected to drop to between 125,000 and
22 154,000 metric tons/year in 1995. The Office of Mobile Sources estimated the annual mean
23 exposure of the U.S. population to diesel PM in 1986 to be 2.6 /zg/m3. The exposure level
24 for 1995 was projected to decrease to between 1.2 and 1.6 /xg/m3 as a result of the new and
25 proposed emission standards.
26 In an EPA-sponsored study, a quantitative evaluation of the carcinogenicity of diesel
27 exhaust was carried out using the comparative potency method. In this 1983 study, the upper
28 confidence limit on unit risk was found to range from 0.2 x 10"4 to 0.35 X lO^/^g/m3 for
29 LDD engines. This was considered to be the upper limit of risk from continuous exposure to
30 1 /ig/m3 of exhaust PM for a lifetime. Using these data, EPA's Office of Mobile Sources
31 estimated that the annual lung cancer incidence in the U.S. population in 1986 caused by
December 1994 1-2 DRAFT-DO NOT QUOTE OR CITE
-------
1 diesel exhaust PM ranged from 178 to 860. It was also estimated that the incidence rate in
2 1995 would fall between 92 and 443. It should be noted, however, that these are 95% upper
3 bound estimates. The actual number of incidences are not likely to be higher and be lower.
4 The latter figure took into account the more stringent emissions standards and also allowed
5 for a range of small to large increases in the use of diesel engines in light-duty vehicles.
6 Since EPA's initial evaluation of the carcinogenic risks associated with diesel exhaust,
7 additional animal and human carcinogenicity data have become available. The purpose of
8 this report is to reevaluate the carcinogenic potency of diesel PM in light of the new data
9 from these animal studies, as well as to reexamine the evidence available from human
10 epidemiological studies. Additional issues that are addressed include the interrelationship
11 between carcinogenic effects and rates of deposition and clearance of the exhaust particles
12 from the lungs, and the significance of potentially carcinogenic organic compounds adsorbed
13 to the exhaust particles and their subsequent desorption and bioavailability. Also included is
14 an overview of the potential noncarcinogenic health effects associated with exposure to diesel
15 exhaust.
16
17
18 1.2 DIESEL EMISSIONS
19 Diesel engines emit both gas phase pollutants (hydrocarbons [HCs], oxides of nitrogen
20 [NOJ, and carbon monoxide [CO]) and carbonaceous PM in quantities sufficient to impact
21 air quality. A description of the diesel engine, its combustion system, pollutant formation
22 mechanisms and emission factors are presented along with current and projected emission
23 control methods in Chapter 2.
24 The diesel engine compresses air to a high pressure and temperature. Fuel when
25 injected into this compressed air autoignites, thereby releasing its chemical energy; the
26 resulting combustion gases expand, doing work on the piston before being exhausted to the
27 atmosphere. Power output is controlled by the amount of injected fuel, not by throttling the
28 air intake.
29 Diesel engines may be either two- or four-stroke cycle, direct or indirect injected, and
30 naturally aspirated or supercharged. Further, they are frequently classified according to
31 service requirements such as light- or heavy-duty automotive, small or large industrial, and
December 1994 j_3 DRAFT-DO NOT QUOTE OR CITE
-------
1 rail or marine. Because they are currently regulated, emissions from motor vehicle diesel
2 engines will be of major concern in this report. Approximately one-half of the yearly
3 production of U.S. diesel engines goes into highway vehicles; the majority of the remainder
4 are divided among agricultural, construction, mining, marine, and a variety of stationary
5 applications.
6 Both the fuel and the lubricants that are used for diesel engine service are highly
7 finished petroleum-based products combined with appropriate additive packages. Diesel fuel
8 oil is a mixture of many different hydrocarbon molecules from about C7 to about C35, with
9 a boiling range from roughly 350 to 650 °F. Many of the fuel oil properties such as its
10 specific energy content, ignition quality, and specific gravity are related to its HC
11 composition. Therefore, one can easily surmise that fuel composition affects many aspects of
12 engine performance, including economy and exhaust emissions. For example, a decrease of
13 fuel aromatic content, sulphur, and volatility usually leads to a reduction of regulated
14 emissions. Current specifications and formulations for both fuels and lubricants will have to
15 be altered as part of any effort aimed at meeting the stringent diesel engine emission
16 standards. In particular, with respect to the fuel, sulphur and aromatic content are both very
17 important. With regard to the lubricating oil, ash content is of paramount importance, thus
18 careful attention will need to be paid to the contents of its additive package.
19 The goal of achieving high engine fuel efficiency while at the same time keeping the
20 exhaust emissions at an acceptable level is the immediate task confronting engine builders.
21 Although the combustion process itself is the source of both the emitted pollutants and good
22 efficiency, it should be obvious that the solution of this task does not lie in engine
23 modification alone. The use of modified fuel system components and injection parameters,
24 reformulated fuels and lubricants, and after-treating devices will also contribute to the
25 ultimate solution.
26 Because the diesel combustion process is a very complex one involving burning of fuel
27 droplets, it has proven difficult to predict quantitatively pollutant concentrations or emission
28 rates. Retarding injection or recirculating exhaust gases reduces NOX formation and emission
29 at the expense of increasing soot formation and HCs, all other factors being equivalent.
30 A variable-temperature model accounting for the average gas temperature at the time of
31 droplet burning accounted for the observed NOX considerably better than a
December 1994 1-4 DRAFT-DO NOT QUOTE OR CITE
-------
1 constant-temperature (peak cycle temperature) model. Process parameters that control NOX
2 formation included fuel jet momentum flux, in-cylinder air density at the start of injection,
3 swirl cross-flow momentum flux, and in-cylinder temperature. Air system design
4 characteristics that change air density and temperature can result in constant work but
5 decreasing NOX. Decreasing cylinder temperatures also result in increased HC emissions.
6 The net result experienced by the engine designer is a set of design trade-offs between NOX
7 and particulate material at high temperatures and between NOX and HCs at low operating
8 temperatures.
9 Small quantities of gaseous unburned HCs and of CO are emitted from diesel engines,
10 rather less than is the case with comparable spark ignition engines, however. In diesel
11 exhaust, CO is formed in concentrations of 2,000 ppm or slightly more. Typical gasoline
12 engines might have exhaust CO concentrations of 10,000 to 20,000 ppm.
13 Diesel combustion processes are also responsible for a small release of
14 low-molecular-weight HCs, principally methane, ethylene and acetylene in diesel exhaust.
15 Both fuel (primarily) and lubricant can supply organic matter to diesel engine exhaust HCs,
16 both in the gaseous and particulate states. Heavy HCs are characterized by substantial
17 quantities of aliphatic HCs and lesser amounts of long-chain substituted monoaromatics.
18 The chemical mechanism that accounts for carbon formation has not been completely
19 ascertained; the major weight of scientific opinion seems to support some role for
20 intermediate formation of PAHs in the process. Carbon is a stable combustion product
21 normally only of rich flames; carbon formation normally takes place over a rather narrow
22 temperature range. Formation of carbon particles is thought to involve polymerization of
23 gaseous intermediates at the surface of small particles. Growth and agglomeration of carbon
24 particles are probably gas-to-particle processes.
25 Studies of diesel particle composition have produced some information about the fate of
26 fuel sulfur. Sulfate has been found to be a significant component of diesel particles.
27 Generally, the sulfate found in particles accounts for only about 2% of the fuel sulfur
28 charged, the balance being emitted as SO2. Attempts to detect sulfite in diesel particle
29 samples have not demonstrated the presence of sulfur (IV). Currently, no means of reducing
30 sulfate formation is available other than reducing the sulfur concentration of diesel fuel.
December 1994 j_5 DRAFT-DO NOT QUOTE OR CITE
-------
1 The emissions factors for HDD CO and HC are somewhat lower but comparable with
2 those of the all-vehicle estimates (which is dominated by gasoline-fueled passenger cars).
3 Therefore, the control of these emissions has a small but discernable influence on overall CO
4 and HC emissions. In the case of NOX, however, heavy duty emissions are currently many
5 times that of the average traffic. Further, even with stringent controls, these emissions are
6 likely to be quite significant and an increasing fraction of the overall mobile source NOX
7 inventory.
8 Diesel passenger car and light-truck emissions of CO and HC are considerably lower
9 than those of gasoline vehicles whereas NOX values are similar. Because these classes
10 collectively only account for less than 1% of traffic, they do not materially influence the
11 emissions inventories for any of these gases.
12 There are no official EPA emissions data for PM emissions comparable to those for the
13 regulated gases. An experimental computer program has been developed for use in particle
14 assessments by the EPA Office of Mobile Sources. However, the program has not been
15 widely distributed nor has it been upgraded to take into account the most recent lead phase-
16 out rules for gasoline supplies or the future trends in diesel PM emissions. The AP-42
17 emissions factors have also not been upgraded in some years. Therefore, there is a need to
18 determine particle emissions, particularly for controlled HDD engines.
19 Diesel engine emissions are determined by the characteristics of the combustion process
20 within each cylinder. Primary engine parameters affecting diesel emissions are the fuel
21 injection system, the engine control system, the air intake port and combustion chamber
22 design, and the air charging system. Actions to reduce lubricating oil consumption can also
23 impact HC and PM emissions. Further, beyond the engine itself, exhaust aftertreatment
24 systems such as trap oxidizers and catalytic converters can play a significant role. Finally,
25 modifications to conventional fuels as well as alternative fuels can substantially lower or raise
26 emissions.
27 Modifications to diesel fuel composition have drawn considerable attention as a quick
28 and cost-effective means of reducing emissions from existing vehicles. The two
29 modifications that show the most promise are to reduce (1) sulphur content and (2) the
30 fraction of aromatic HCs in the fuel. Recently, EPA decided to reduce sulphur content in
31 diesel fuel to a maximum of 0.05% by weight.
December 1994 1-6 DRAFT-DO NOT QUOTE OR CITE
-------
1 The possibility of substituting cleaner-burning alternative fuels for diesel fuel has drawn
2 increasing attention during the last decade. Motivations advanced for this substitution include
3 conservation of oil products and energy security, as well as the reduction or elimination of
4 PM emissions and visible smoke. Care is necessary in evaluating the air quality claims for
5 alternative fuels, however. Although many alternative fuel engines do display greatly
6 reduced paniculate and SO2 emissions, emissions of other gaseous pollutants such as
7 unburned HCs, CO, and in some cases NOX and aldehydes may be much higher than from
8 diesels. The principal alternative fuels currently under consideration are natural gas, and
9 methanol made from natural gas, and in limited applications, liquified petroleum gas (LPG).
10 Diesel PM control technology has now advanced to where manufacturers have had no
11 difficulty complying with the 0.25 g/bhp-h standard that went into effect for 1991 model year
12 trucks. Without exception, manufacturers were able to comply solely with the use of engine
13 modifications.
14 Compliance with the 1991 urban bus standard of 0.1 g/bhp-h (which applies to all
15 heavy duty vehicles by 1994) was more problematic but appeared feasible if the Congress
16 didn't modify the law. Several manufacturers have indicated their intention to offer systems
17 that are expected to achieve these levels. Regarding 1994 compliance, most manufacturers
18 have indicated their intention to comply with the use of oxidation catalysts because
19 "low"-sulfur fuel will be available.
20
21
22 1.3 DIESEL-DERIVED POLLUTANTS: ATMOSPHERIC
23 CONCENTRATIONS, TRANSPORT, AND TRANSFORMATIONS
24
25 Major research programs were carried out in the late 1970s and early 1980s to
26 understand the physical and chemical characteristics of emissions from diesel engines and the
27 biological effects of these emissions. New control technologies are being introduced into
28 currently manufactured diesel vehicles, and the effect of these changes on diesel emissions is
29 likely to be visible in the future. Currently, diesel vehicles manufactured in the late 1970s
30 and early 1980s are still on the road, and in this sense, data collected from that time period
31 are still valid.
December 1994 ^7 DRAFT-DO NOT QUOTE OR CITE
-------
1 However, many of these data were collected using laboratory dynamometers with
2 selected new vehicles or at least those "well-tuned" to manufacturers specifications. The
3 well-controlled conditions of the dynamometer tests have many benefits but do not
4 necessarily represent vehicle emissions under real on-road conditions. The small number of
5 vehicles tested in the laboratory is not truly representative of the distribution within the
6 on-road vehicle fleet. Although several roadway and tunnel emission measurements were
7 performed in the past, the data base on mobile sources emission rates necessary to assess the
8 role of vehicle emissions in air pollution problems is still not sufficient. More measurements
9 carried out under realistic on-road conditions are necessary, in particular, for gaseous- and
10 particulate-phase organic compounds present in vehicle emissions.
11 Once released into the atmosphere, diesel emissions are subject to dispersion and
12 transport and, at the same time, to chemical and physical transformations into secondary
13 pollutants, which may be more harmful than their precursors. Thus, a knowledge of diesel
14 emissions at or near their sources is no longer sufficient to assess fully the impact of these
15 emissions on human health and welfare. The understanding of physical and chemical
16 changes that primary diesel emissions undergo during their transport through the atmosphere
17 is equally important. As a result of the last two decades of laboratory and ambient
18 experiments and computer modeling, a comprehensive set of data now exist concerning the
19 atmospheric loss processes and transformation of automotive emissions. However, our
20 knowledge concerning the products of these chemical transformations is still very limited.
21 Study is required to determine the products from the OH radical-initiated reactions of the
22 aromatic and aliphatic HCs, the major components of automobile emissions. The
23 atmospheric transformation products of PAHs and their oxygen-, sulfur-, and nitrogen-
24 containing analogs require study in the gaseous and adsorbed phases. In particular, the
25 reactions occurring in adsorbed phases on atmospherically relevant surfaces are poorly
26 understood and require further study. In addition, gas-to-particle conversion processes and
27 the chemical processes that lead to aerosol formation should be further investigated.
28 The quantitation of the contribution of diesel emissions to total ambient aerosol mass
29 concentration is not possible without developing a specific profile for diesel emissions, a
30 "fingerprint" that may be used in receptor source apportionment models. The existing data
31 indicate that it may be possible to use PAHs or alkylated PAHs, alkanes, and possibly,
December 1994 1-8 DRAFT-DO NOT QUOTE OR CITE
-------
1 certain unique compounds to assist in distinguishing between diesel and other pollutant
2 sources. However, the available data are not adequate for use in receptor modeling, and
3 study is required to determine the profile of diesel emissions by using sampling and
4 analytical methods appropriate to receptor modeling.
5
6
7 1.4 DOSIMETRIC FACTORS IN ASSESSING CARCINOGENIC RISK
8 OF EXPOSURE TO DIESEL ENGINE EMISSIONS
9 Because there is only very limited evidence for diesel exhaust-induced tumors at
10 nonpulmonary sites or for tumor induction by the gaseous fraction alone, dosimetry
11 considerations were limited to either whole exhaust or PM deposited in the lungs.
12 Dosimetric considerations were further limited to the gas exchange regions of the lung for
13 several reasons. First of all, most deposited PM is transported from the conducting airways
14 in less than 1 day. As a result, there is little evidence for the focal accumulation of particles
15 with accompanying lesions, such as seen in the alveolar regions. The rapid clearance also
16 allows little time for extraction of organics from the particle surface. Because few of these
17 particles are phagocytized but remain free in the lung fluid, which is ineffective in extracting
18 organics, the likelihood of elution is even less. Finally, most of the pathologic and
19 carcinogenic effects occur at or distal to the terminal bronchioles.
20 Clearance of the diesel particles from the alveolar regions varied from about 2 mo in
21 rats to an estimate of nearly 1 year in humans. Under high-exposure regimes, lung overload
22 occurred in experimental animals, resulting in slower or near cessation of clearance, thereby
23 increasing lung burdens even further. In addition, with large lung burdens, uptake of
24 particles by Type I alveolar cells, passage into the interstitium and transport to lung
25 associated lymph was increased. Factors considered to be involved in clearance inhibition
26 included loss of macrophage mobility with large particle loads and a tendency for
27 macrophages to aggregate.
28 Most biological fluids tested, including lung lavage fluid and serum, were relatively
29 ineffective in extracting organic agents adsorbed to the diesel particle surface. Particles
30 deposited in the alveolar region, however, are rapidly phagocytized by alveolar macrophages,
31 which are more effective in this regard. A fraction of the organics are eluted in a matter of
December 1994 !_9 DRAFT-DO NOT QUOTE OR CITE
-------
1 hours, with the more tightly bound fraction removed with a half-time of about a month.
2 Because elution rates are generally more rapid than particle clearance rates, most of the
3 organic fraction is assumed to become bioavailable, even with no inhibition of clearance.
4 In the development of a dosimetry model to allow both low-dose extrapolation as well
5 as extrapolation of diesel exhaust bioassay data from laboratory animals to humans, several
6 parameters must be accounted for. These include, at a minimum, deposition efficiency,
7 particle clearance rates, desorption rates of organics from the particle surface, lung surface
8 area, metabolic rate, and respiratory exchange rates. Adjustment for particle clearance rate
9 is necessary for two reasons. First of all, many of the animal experiments were conducted
10 under exposure regimes that resulted in an inhibition of clearance with an accompanying lung
11 burden overload. If lung burden of PM is considered to be the proper dosimetric variable,
12 then the disproportionately large lung burdens at high levels of exposure must be adjusted
13 for. Second, even under low-exposure regimes, clearance is slower in humans than in rats.
14 If the correct dosimetric variable, on the other hand, is particle-free organic matter, then a
15 smaller adjustment for variations in particle clearance rates is required because, even with
16 normal clearance rates, most of the organics are likely to be eluted from the particles
17 deposited in the alveolar region. Nevertheless, some adjustment is still necessary, because
18 the organics are seldom all eluted.
19 Another highly variable parameter in both experimental animals and humans is the
20 respiratory exchange rate. Animal estimates are often based on published values collected
21 under resting conditions. Respiration, however, may be inhibited by the irritant gases
22 present in diesel exhaust, especially at low dilution ratios. On the other hand, respiration
23 may be either greater or less than estimated resting values, depending on whether exposures
24 were carried out at night when the animal are likely to be awake and active, or during the
25 day when they are more likely to be asleep. Human respiratory exchange rates are also quite
26 variable, with the physically active segment of the population at potentially greater risk
27 because of greater exposure resulting from higher respiration rates.
28
29
30
December 1994 1-10 DRAFT-DO NOT QUOTE OR CITE
-------
1 1.5 NONCANCER HEALTH EFFECTS OF DIESEL EXHAUST
2 Symptoms of acute exposure of humans to high levels (i.e., above environmental
3 ambient concentrations) of diesel exhaust include mucous membrane, eye, and respiratory
4 tract irritation (including chest tightness and wheezing) and neuropsychological effects of
5 headache, lightheadedness, nausea, heartburn, vomiting, weakness, and numbness and
6 tingling in the extremities. The mere odor of diesel exhaust can cause nausea, headache, and
7 loss of appetite.
8 The effects of short-term exposures of humans to diesel exhaust have been investigated
9 primarily in occupationally exposed workers. In studies of underground miners, bus garage
10 workers, dock workers, and locomotive repairmen, changes in respiratory symptoms and
11 pulmonary function over the course of a workshift were generally found to be minimal and
12 not statistically significant. In a study of diesel bus garage workers, however, there was an
13 increased reporting of burning and watering of the eyes, cough, labored breathing, chest
14 tightness, and wheezing, but no reductions in pulmonary function were associated with
15 exposure to diesel exhaust. Pulmonary function was adversely affected in stevedores over a
16 workshift exposure to diesel exhaust but normalized after a few days without exposure to
17 diesel exhaust.
18 Chronic effects of diesel exhaust exposure in humans have been evaluated in
19 epidemiologic studies of occupationally exposed workers, including metal and nonmetal
20 miners, railroad yard workers, stevedores, and bus garage mechanics. Most of the
21 epidemiologic data indicate the absence of an excess of chronic respiratory disease associated
22 with exposure to diesel exhaust. In a few of these studies, a higher prevalence of respiratory
23 symptoms, primarily cough, phlegm, or chronic bronchitis, were observed among the
24 exposed. Reductions in forced vital capacity (FVC) and 1-s forced expiratory volume
25 (FEV!), and to a lesser extent forced expiratory flow at 50 and 75% of vital capacity (FEF50
26 and FEF75), have also been reported. Two studies, each with methodological problems,
27 detected statistically significant decrements in pulmonary function when compared with
28 matched controls. These two studies coupled with other reported nonsignificant trends in
29 respiratory flow-volume measurements suggest that exposure may impair pulmonary function
30 among occupational populations. Whereas a preliminary study of the association of
31 cardiovascular mortality and exposure to diesel exhaust found a fourfold higher risk ratio, a
December 1994 !_!! DRAFT-DO NOT QUOTE OR CITE
-------
1 more comprehensive study by the same investigators found no significant difference between
2 the observed and expected number of deaths caused by cardiovascular disease.
3 Caution is warranted in the interpretation of the results of epidemiologic studies that
4 have addressed noncarcinogenic health effects resulting from exposure to diesel exhaust.
5 These investigations suffer from a myriad of methodological problems, including
6 (1) incomplete information on the extent of exposure to diesel exhaust, necessitating in some
7 studies estimations of exposures from job titles and the resultant misclassification; (2) the
8 presence of confounding variables such as smoking or occupational exposures to other toxic
9 substance (e.g., mine dusts); and (3) the short duration and low intensity of exposure. These
10 limitations restrict definitive conclusions about diesel exhaust being the cause of any
11 noncarcinogenic health effects, observed or reported.
12 Animal studies of the toxic effects of diesel exhaust have employed acute, subchronic
13 and chronic exposure regimens. In acute exposure studies, toxic responses have been
14 associated primarily with high concentrations of CO, NO2, and aliphatic aldehydes.
15 In short-term and chronic exposure studies, toxic effects have been related to high
16 concentrations of PM. The data from short-term exposures indicate minimal effects on
17 pulmonary function, even though histological and cytological changes were observed in the
18 lungs. Exposures for several months or longer to levels markedly above environmental
19 ambient concentrations resulted in accumulation of particles in the lungs, increases in lung
20 weight, increases in alveolar macrophages and leukocytes, macrophage aggregation,
21 hyperplasia of alveolar epithelium, and thickening of the alveolar septa. Similar histological
22 changes, as well as reductions in growth rates and alterations in indices of pulmonary
23 function, have been observed in chronic exposure studies. Chronic studies have been carried
24 out using rats, mice, guinea pigs, hamsters, cats, and monkeys. Other studies have
25 demonstrated reductions in pulmonary resistance to respiratory tract infections in mice
26 exposed to diesel exhaust. In addition, there are limited animal data associating behavioral,
27 hematological, and clinical enzymatic changes and cytological alterations in the liver with
28 exposure to diesel exhaust.
29 In animals chronically exposed to diesel exhaust, reductions in growth rates have been
30 observed most often in studies in which the exhaust was diluted to produce concentrations of
31 PM of at least 2 mg/m3 and in which the exposure period lasted for 16 h or more per day.
December 1994 1-12 DRAFT-DO NOT QUOTE OR CITE
-------
1 In studies in which the daily exposures were only 6 to 8 h/day, no effects on growth or
2 survival were seen at levels of 6 to 8 mg/m3 of PM.
3 Alterations in pulmonary function indices have been observed in a number of different
4 species chronically exposed to diesel exhaust and include decreases in vital capacity, residual
5 lung volume, diffusing capacity, dynamic lung compliance, and expiratory flow rates, as well
6 as increases in airway resistance. The lowest exposure levels that resulted in impaired
7 pulmonary function varied among the species tested. In rats, mice, hamsters, and cats, the
8 testing results were consistent with restrictive lung disease such as caused by pulmonary
9 fibrosis. In monkeys, evidence of obstructive airway disease such as would occur with
10 chronic bronchitis was observed; this difference being attributed to the anatomical differences
11 of the lung and the lower dose of particles per gram of lung tissue retained in the monkey
12 lung.
13 Histological studies have demonstrated that chronic exposure to high concentrations of
14 diesel exhaust can adversely affect respiratory tract tissue. Typical findings include alveolar
15 histiocytosis, macrophage aggregation, tissue inflammation, increases in polymorphonuclear
16 leukocytes, hyperplasia of bronchiolar and alveolar Type II cells, thickened alveolar septa,
17 edema, fibrosis, and emphysema. Lesions in the trachea and bronchi were observed in some
18 studies. Associated with these histopathological findings were various biochemical changes
19 in the lung, including increases in lung DNA, total protein, alkaline and acid phosphatase,
20 and glucose-6-phosphate dehydrogenase; increased synthesis of collagen, and release of
21 inflammatory mediators such as leukotriene LTB and prostaglandin PGF2a. Although the
22 overall laboratory evidence is that prolonged exposure to diesel exhaust results in
23 histopathological and histochemical changes in the lungs of exposed animals, some studies
24 have also demonstrated that there may be a threshold of exposure to diesel exhaust below
25 which pathologic changes do not occur. These no-effect levels were reported to be 2 mg/m3
26 for cynomolgus monkeys, 0.11 to 0.35 mg/m3 for rats and 0.25 mg/m3 PM for guinea pigs
27 exposed for 7 to 20 h/day, 5 to 5.5 days/week for 104 to 130 weeks.
28 The pathological effects of diesel exhaust PM appear to be strongly dependent on the
29 relative rates of pulmonary deposition and clearance. At particle concentrations of about
30 1 mg/m3 or above, under varying durations of exposure, pulmonary clearance becomes
31 reduced with concomitant focal aggregations of particle-laden alveolar macrophages,
December 1994 1_13 DRAFT-DO NOT QUOTE OR CITE
-------
1 particularly in the peribronchiolar and alveolar regions, as well as in the hilar and
2 mediastinal lymph nodes. The principal mechanism of reduced particle clearance appears to
3 be the result of impaired pulmonary alveolar macrophage function. This impairment seems
4 to be nonspecific and applies to insoluble particles deposited in the alveolar region. Other
5 data suggest that the inability of particle-laden alveolar macrophages to translocate to the
6 mucociliary escalator is correlated to the average composite particle volume per alveolar
7 macrophage in the lung. Data from rats indicate that when this particle volume exceeds a
8 critical level, impairment appears to be initiated. Such data for other laboratory species and
9 humans, unfortunately, are very limited.
10 Rodent studies have revealed that exposure to diesel exhaust can reduce resistance to
11 respiratory infections. This effect, which can occur even after short exposures, does not
12 appear to be caused by direct impairment of the lymphoid or splenic immune systems;
13 however, in one study of influenza infection, interferon levels and hemagglutinin antibody
14 levels were adversely affected in the exposed mice.
15 The comparison of the toxic responses in laboratory animals exposed to whole diesel
16 exhaust or filtered exhaust containing no particles demonstrates across laboratories that diesel
17 particles are the principal etiologic agent of noncancerous health effects in laboratory animals
18 exposed to diesel exhaust.
19 Specific studies to test interactive effects of diesel exhaust with atmospheric
20 contaminants, other than coal dust, have not been conducted. Coal dust and diesel particles
21 had an additive effect only in rats and monkeys. Studies to assess the susceptibility of the
22 developing and mature lungs of rats and emphysematous versus normal rat lungs failed to
23 demonstrate an increased risk of the immature or impaired lung following exposure to diesel
24 exhaust.
25 Only one laboratory has made an evaluation of the comparative toxicities of diesel and
26 gasoline exhausts. Methodological and reporting limitations, however, resulted in data of
27 little relevance for a relative risk assessment. Data generated in the 1960s on gasoline
28 engine exhaust are difficult to compare with data generated in the 1980s on diesel exhaust
29 because different animal species were used. Likewise, gasoline engine exhaust from a 1990
30 catalyst equipped car is remarkably different than that from gasoline engines in the 1960s
31 with limited emission control devices. From a generic standpoint the major air pollutants
December 1994 1-14 DRAFT-DO NOT QUOTE OR CITE
-------
1 attributable to automotive exhaust in the 1980s were qualitatively the same as those in the
2 1960s. Low concentrations of ozone and NO2, primary pollutants affecting the lungs of
3 exposed subjects, produce their principal pathology in the centriacinar region airways (the
4 junction between the conducting and the gas exchange regions). Low doses of retained
5 particles, when fine and insoluble, tend to have their principal effect on the alveolar region
6 because of their long residence time and tendency to aggregate over time.
7 There is a considerable body of evidence that the major noncancerous health hazards
8 posed by exposure to diesel exhaust are to the lung. These data also denote that the
9 exposures that cause pulmonary injury are lower than those inducing detectable increases in
10 lung tumors. These same data further indicate that the inflammatory and proliferative
11 changes in the lung play a key role in the etiology of pulmonary tumors in exposed rats.
12 Subpopulations of the general public, the young, the old, and those with existing pulmonary
13 disease, may have a greater susceptibility to the toxic actions of diesel exhaust. Because
14 noncancerous pulmonary effects occur at lower doses than those inducing detectable increases
15 in lung tumors and these effects appear to be cofactors in the etiology of diesel
16 exhaust-induced tumors, noncancerous pulmonary effects must be considered in the risk
17 assessment of public exposure to diesel exhaust, notably the particulate component.
18
19
20 1.6 QUANTITATIVE AND QUALITATIVE ASSESSMENT OF
21 NONCANCER HEALTH EFFECTS—DERIVATION OF THE
22 INHALATION REFERENCE CONCENTRATION
23
24 A large number of chronic inhalation studies of diesel exhaust inhalation in
25 experimental animals characterize the respiratory effects and the concentration-response
26 relationship of those effects in detail. Many epidemiological studies are also available of
27 occupationally exposed humans. The epidemiological studies provide qualitative evidence
28 that supports the identification of a hazard to the respiratory system from animal studies.
29 The human studies are of limited value quantitatively due to inadequate exposure
30 characterization and confounding by concurrent exposure to other pollutants. The animal
31 studies were used for derivation of an inhalation reference concentration (RfC). The RfC is
32 defined as an estimate (uncertainty spanning perhaps an order of magnitude) of a daily
December 1994 j_15 DRAFT-DO NOT QUOTE OR CITE
-------
1 exposure (mg/m3) to the general public (including sensitive subgroups) that is likely to be
2 without an appreciable risk of deleterious effects during a lifetime exposure. The chronic
3 studies from the Inhalation Toxicology Research Institute (ITRI) and the Health Effects
4 Research Program (HERP) with rats were selected as the principal studies for RfC
5 development. Using the deposition and retention model discussed in Chapter 4 and Appendix
6 B to calculate human equivalent concentrations, a no observed adverse effect level (NOAEL)
7 of 0.155 mg/m3 was identified from the HERP studies and was used to calculate the RfC.
8 An uncertainty factor of 30 was applied for interspecies extrapolation and to account for
9 sensitive members of the population, resulting in an RfC of 5 pig/m3. The RfC is considered
10 to have high confidence, due to high confidence in the study and data base.
11
12
13 1.7 CARCINOGENICITY OF DIESEL ENGINE EMISSIONS IN
14 LABORATORY ANIMALS
15
16 As early as 1955, there was evidence for tumorigenicity and carcinogenicity of acetone
17 extracts of diesel exhaust following dermal application, and that variability in this response
18 was dependent on engine operating mode. By the 1970s, it was also reported that diesel
19 exhaust extracts were mutagenic. Until the early 1980s, however, no chronic studies
20 assessing the effects of inhalation of diesel exhaust, the relevant mode for human exposure,
21 had been reported. Since then, inhalation studies have been emphasized.
22 Studies employing rats exposed for two years or more to high PM concentrations
23 (up to 8 mg/m3), resulting in large particle loads in the lungs, were generally positive in
24 demonstrating diesel exhaust-induced increases in lung tumors. Inhalation of diesel exhaust
25 induced significant increases in lung tumors in two strains of female mice and an apparent
26 increase in a third strain. However, attempts to induce significant increases in lung tumor
27 incidence in Syrian golden hamsters, cats, or monkeys were unsuccessful. The negative
28 results in cats and monkeys may be explained by an inadequate exposure duration (2 years)
29 in these longer-lived species, whereas hamsters are generally less sensitive to lung tumor
30 induction by inhalation than are rats or mice.
31 Intratracheal instillation of diesel PM induced lung tumors in Fischer 344 rats.
32 Additional studies have shown not only that an extract of diesel particles was carcinogenic
December 1994 1-16 DRAFT-DO NOT QUOTE OR CITE
-------
1 when instilled in the lungs of rats, but also that most of the carcinogenicity resided in the
2 portion containing PAHs with 4 to 7 rings.
3 Alternate exposure routes including dermal exposure, skin painting, and subcutaneous
4 injection in mice provided additional evidence for tumorigenic effects of diesel exhaust.
5 However, negative results have also been reported for skin-painting studies using mice and
6 acetone extracts of diesel exhaust particle suspensions. Results of intraperitoneal injection of
7 diesel exhaust PM or particle extracts in Strain A mice were generally negative, suggesting
8 that this may be an inappropriate model for testing of diesel emissions.
9 Experiments using tumor initiators did not provide conclusive results regarding the
10 tumor-promoting potential of either filtered or whole diesel exhaust. One report, however,
11 did indicate that filtered exhaust may promote the tumor-initiating effects of
12 diethylnitrosamine (DEN) in hamsters.
13 It appears reasonably certain that with adequate exposures, inhalation of diesel exhaust
14 will induce lung cancer in rats and in at least some strains of mice. The relationship between
15 exposure levels and response, however, is less clearcut. Although significant increases in
16 lung tumors were not reported at PM concentrations of less than about 2 mg/m3, the
17 response at higher concentrations varies considerably. A significant percentage of this
18 variation can probably be attributed to the exposure regime. Cumulative exposure data
19 (concentration x daily exposure duration x days of exposure) from studies using rats
20 suggest a trend of increasing tumor incidence at exposures exceeding
21 1 x 104 mg4i/m3. The experimental designs, however, were not sufficiently sensitive to
22 rule out the possibility of responses at lower concentrations. A similar comparison could not
23 be adequately made for mice, because experimental designs were not comparable.
24 Several of the previously discussed studies indicated that only whole (unfiltered) diesel
25 exhaust is tumorigenic or carcinogenic and that these properties are eliminated or greatly
26 minimized in filtered diesel exhaust exposure. In one study, however, a significant increase
27 in lung tumors was seen in mice exposed to filtered exhaust, and some evidence suggests that
28 the gaseous fraction may promote the tumorigenic effects of DEN in hamsters.
29 Nevertheless, because of the lack of definitive positive data in rats and the limited positive
30 data in mice, the tumorigenicity of the gaseous fraction must be considered to be unresolved.
December 1994 1_17 DRAFT-DO NOT QUOTE OR CITE
-------
1 Evidence for the importance of the carbon core has been demonstrated by the induction
2 of lung tumors following exposure to carbon black (similar to the carbon core of the diesel
3 exhaust particle) that contains no more than traces of organics via intratracheal instillation in
4 one study and via inhalation in two others. Although the contribution to tumor induction by
5 the gaseous phase or the fractions of the PM phase may not be totally resolved, for
6 qualitative purposes, the EPA's evaluation is based on whole exhaust.
7 Based on positive data from rat and mouse inhalation studies, intratracheal instillation
8 studies using rats, skin painting studies in mice, and the supporting positive mutagenicity
9 studies, the evidence for carcinogenicity of diesel exhaust is considered to be adequate. The
10 contribution of the various fractions of diesel exhaust to the carcinogenic response is less
11 certain. The effects of the gaseous phase are equivocal. The presence of known carcinogens
12 adsorbed to diesel particles and the demonstrated tumorigenicity of particle extracts in a
13 variety of injection, instillation, and skin painting studies provides evidence that the organic
14 fraction, at high concentrations, can induce tumors. The relatively low concentration of
15 carcinogens in both the gaseous phase and the particle-absorbed organics, coupled with the
16 demonstrated ability of pure carbon to induce tumors at about the same concentration as
17 diesel exhaust renders it likely that the insoluble core of the diesel exhaust particle is
18 primarily responsible for the tumorigenic effects observed in the inhalation studies.
19
20
21 1.8 EPIDEMIOLOGIC STUDIES OF DIESEL EXHAUST
22 CARCINOGENICITY
23 It is difficult to study the health effects of diesel exhaust in the general population
24 because diesel emissions are diluted in the ambient air; hence, exposure is very low. Thus,
25 populations occupationally exposed to diesel exhaust are studied to determine the potential
26 health effects in humans. The occupations involving potential exposure to diesel exhaust are
27 miners, truck drivers, transportation workers, railroad workers, and heavy-equipment
28 operators.
29 All the occupational studies considered in this document have a similar problem—an
30 inability to measure accurately the actual exposure to diesel exhaust. Most studies compared
31 persons in job categories that would presumably have some exposure to diesel exhaust with
December 1994 1-18 DRAFT-DO NOT QUOTE OR CITE
-------
1 either standard populations (that have presumably no exposure to diesel exhaust) or with men
2 working in other job categories in industries with little or no potential for diesel exhaust
3 exposure. The study of the U.S. railroad workers was the only one in which the job
4 categories were verified based on an industrial hygiene survey. A few studies included
5 measurements of diesel fumes, but there was no standard method for the measurement.
6 Neither was any attempt made to correlate these exposures with the cancers observed in any
7 of these studies, nor was it clear exactly which extract should be measured to assess the
8 occupational exposure to diesel exhaust.
9 With the exception of one study, no known studies have assessed the relationship
10 between diesel exhaust exposure and lung cancer in miners. Virtually all metal mines use
11 diesel equipment, which was introduced in the early 1960s. Approximately 20,000 miners
12 are employed in metal mines but not all of them are currently working. Estimates of how
13 many miners are exposed to diesel fumes and for what length of time are not available. Coal
14 mines also use diesel equipment, which was introduced in these mines later than the 1960s.
15 The cohort studies reviewed in this document mainly demonstrated an increased risk of
16 lung cancer. Studies of bus company workers failed to demonstrate any statistically
17 significant excess risk of lung cancer, but these studies have certain methodological
18 problems, such as small sample size, short follow-up periods, lack of information on
19 confounding variables, and lack of analysis by duration of exposure or latency, that preclude
20 their use in determining the carcinogenicity of diesel exhaust.
21 A mortality study of heavy-equipment operators demonstrated a significantly increased
22 risk of liver cancer in the total cohort and in various subcohorts. This study also found a
23 nonsignificant positive trend for lung cancer with increasing length of membership and
24 latency. Analysis of a subcohort of deceased employees showed a significant excess of lung
25 cancer. Similarly, individuals without any work histories, who were probably in the same
26 job prior to and after 1967, also showed significant excess risks of lung cancer and stomach
27 cancer. However, because of several flaws in the methodology, this study cannot be used to
28 support or refute a causal association between exposure to diesel exhaust and occurrence of
29 cancers.
30 After controlling for age and smoking, a 2-year mortality analysis demonstrated an
31 excess risk of lung cancer in certain occupations with potential exposure to diesel exhaust.
December 1994 149 DRAFT-DO NOT QUOTE OR CITE
-------
1 These excesses were statistically significant among miners and heavy-equipment operators.
2 The elevated risks were nonsignificant in railroad workers and truck drivers, although a dose
3 response was observed for truck drivers. Despite the methodologic limitations, such as the
4 lack of representation of the study population (volunteers only) and the questionable
5 reliability of the exposure data (based on self-administered questionnaires that had not been
6 validated), this study is suggestive of a causal association between exposure to diesel exhaust
7 and excess risk of lung cancer.
8 Two mortality studies were conducted among railroad workers, one in Canada and one
9 in the United States. The Canadian study found relative risks of 1.2 (p < 0.10) and
10 1.35 (p < 0.001) among "possibly"- and "probably"-exposed groups, respectively. The
11 trend test in the Canadian study showed a highly significant dose-response relationship
12 between exposure to diesel exhaust and the risk of lung cancer. However, because of the
13 limitations, such as lack of detailed work descriptions and confounding factors of
14 simultaneous coal dust exposure, asbestos exposure, and smoking habits, the results of this
15 study are only suggestive of pulmonary carcinogenesis. It is also worthwhile to note that
16 coal dust has not, to date, been shown to be a pulmonary carcinogen.
17 The U.S. study provided evidence for probable causal association between diesel
18 exhaust exposure and occurrence of lung cancer. Relative risks of 1.57 and 1.34 were found
19 for workers in age groups 40 to 44 and 45 to 49, respectively, after the exclusion of workers
20 exposed to asbestos. This study also found that risk of lung cancer increased with increasing
21 duration of employment. This was a large cohort study, with lengthy follow-up and adequate
22 analysis, including dose response (based on duration of employment as a surrogate) as well
23 as adjustment for other confounding factors such as asbestos; thus, the observed association
24 between increased lung cancer and exposure to diesel exhaust is more meaningful.
25 Among the eight lung cancer case-control studies reviewed in this document, a study
26 adjusting for age and smoking did not find an increased risk of lung cancer resulting from
27 diesel exhaust exposure. A major limitation in this study was the lack of adequate exposure
28 data derived from the job titles obtained from occupational histories. On the other hand,
29 statistically nonsignificant excess risks were observed for diesel exhaust exposure in three
30 different studies. The first study found lung cancer excesses in railroad workers and truck
31 drivers. The second study found an excess risk for workers exposed to diesel exhaust versus
December 1994 1-20 DRAFT-DO NOT QUOTE OR CITE
-------
1 those who were not, and the third study found an excess risk for professional drivers. These
2 rates were adjusted for age and smoking. However, the first two studies had
3 nonparticipation rates of 47 and 36%, respectively. In the second study the use of self-
4 reported exposures was not validated, and the study also had low power to detect excess risk
5 of lung cancer for specific occupations. Therefore, these studies probably underestimate the
6 risk of lung cancer.
7 After adjusting for smoking, significantly increased risks of lung cancer were found
8 among French motor vehicle drivers and transport equipment operators. The main limitation
9 of this study was the inability to separate the exposures to diesel exhaust from that of
10 gasoline exhaust because both motor vehicle drivers and transport equipment operators
11 probably were exposed to both kinds of exhausts.
12 A group of investigators combined data from three studies (conducted in three different
13 states) to increase the ability to detect an association between lung cancer and different
14 occupations that have a high potential for exposure to diesel exhaust. They found that truck
15 drivers employed for more than 10 years had a significantly increased risk of lung cancer.
16 This study also found a significant trend of increasing risk of lung cancer with increasing
17 duration of employment among truck drivers. The relative odds were computed by adjusting
18 for birth cohort, smoking, and state of residence. The main limitation of this study is again
19 the mixed exposures to diesel and gasoline exhausts because information on type of engine
20 was lacking. Furthermore, the methods used to classify occupational categories in these
21 studies were different and, therefore, probably led to incompatibility of occupational
22 categories.
23 The most convincing evidence for diesel emission-induced lung cancer in humans comes
24 from the case-control studies among U.S. railroad workers and among truck drivers in the
25 Teamsters Union. In the study of U.S. railroad workers, after adjustment for asbestos and
26 smoking, the relative odds for continuous exposure were 1.39 (95% CI = 1.05, 1.83).
27 Among the younger workers with longer diesel exhaust exposure, the risk of lung cancer
28 increased with the duration of exposure, after adjusting for asbestos and smoking. After the
29 exclusion of recent diesel exhaust exposure (5 years before death), the relative odds increased
30 to 1.43. This study appears to be a well conducted and well-analyzed nested case-control
December 1994 l_2i DRAFT-DO NOT QUOTE OR CITE
-------
1 study with reasonably good power. Potential confounders were controlled adequately and
2 interactions between diesel exhaust and other lung cancer risk factors were tested.
3 In the study of truck drivers in the Teamsters Union who primarily drove diesel trucks
4 for 35 years, the relative odds were 1.89 (95% CI = 1.04, 3.42). This study also showed
5 increasing risks of lung cancer with increasing years of exposure when employment after
6 1959 was considered. The limitations of this study include possible misclassifications of
7 exposure and smoking, lack of levels of diesel exposure, smaller exposed population, and
8 insufficient latency period. Given these limitations, the findings of this study are probably
9 underestimated.
10 Of the seven bladder cancer case-control studies, three studies found an increased risk
11 in occupations with a high potential for diesel exhaust exposure. A significantly increased
12 risk of bladder cancer was found in Canadian railroad workers and in Argentinean truck and
13 railroad drivers. Significantly increased overall risks were observed with increasing duration
14 of employment of >20 years in truck drivers and railroad industry workers in the third
15 study. No significant increased risk was found for any diesel-related occupations in the
16 remaining four studies, but these studies had several limitations, such as inadequate
17 characterization of diesel exhaust exposure, lack of validation of surrogate measures of
18 exposure, and presence of other confounding factors (cigarette smoking, urinary retention,
19 concentrated smoke within the truck cab, etc.). Furthermore, most of the studies had small
20 sample sizes, and none presented any latency analysis.
21 In summary, an excess risk of lung cancer was observed in four out of seven cohort
22 studies and seven out of eight case-control studies. Of these studies, three cohort and three
23 case-control studies observed a dose-response relationship by using duration of employment
24 as a surrogate for dose. However, because of the lack of actual data on exposure to diesel
25 exhaust in these studies and other methodologic limitations, such as lack of latency analysis,
26 etc., the evidence of carcinogenicity in humans falls short of being sufficient, and hence, is
27 considered to be limited.
28 Five cohort studies (Gustavsson et al., 1990; Guberan et al., 1992; Emmelin et al.,
29 1993; Swanson et al., 1993; Hansen, 1993) and three case-control studies (Boffetta et al.,
30 1990; Cordier et al., 1993; Notani et al., 1993) have been published since the last update of
31 this chapter. The Addendum to Chapter 8 contains the reviews of these studies, which
December 1994 1-22 DRAFT-DO NOT QUOTE OR CITE
-------
1 indicate that the evidence of carcinogenicity in humans will not change. These studies will
2 be included in Chapter 8 in the final version of this document.
3
4
5 1.9 MUTAGENICITY OF DIESEL ENGINE EMISSIONS
6 Extensive studies with Salmonella have unequivocally demonstrated direct-acting
7 mutagenic activity in both paniculate and gaseous fractions of diesel exhaust. The induction
8 of gene mutations has been reported in several in vitro mammalian cell lines after exposure
9 to extracts of diesel particles. Dilutions of whole diesel exhaust did not induce sex-linked
10 recessive lethals in Drosophila or specific-locus mutations in male mouse germ cells.
11 Structural chromosome aberrations and sister chromatid exchanges (SCE) in mammalian
12 cells have been induced by particles. Whole exhaust induced micronuclei, but not SCE or
13 structural aberrations, in bone marrow of male Chinese hamsters exposed to whole diesel
14 emissions for 6 mo. In 7-week exposures, neither micronuclei nor structural aberrations
15 were increased in bone marrow of female Swiss mice. Likewise, whole diesel exhaust did
16 not induce dominant lethals or heritable translocations in male mice exposed for 7.5 and
17 4.5 weeks, respectively.
18
19
20 1.10 METABOLISM AND MECHANISM OF ACTION OF DIESEL
21 EMISSIONS-INDUCED CARCINOGENICITY
22
23 Currently, it is uncertain whether the carcinogenicity of diesel exhaust is the result of
24 genetic or epigenetic mechanisms or a combination thereof. The genetic mechanism is
25 supported by data showing the formation of DNA adducts in diesel-exposed animals and by
26 the known carcinogenic and mutagenic potential of many of the compounds in diesel exhaust.
27 Several studies affirm the bioavailability from inhaled diesel exhaust particles of
28 compounds such as B[a]P and 1-NP, which are known to be carcinogenic or mutagenic in
29 experimental animals. Furthermore, the biotransformation of xenobiotics to reactive
30 intermediates following their entry into the body via inhalation of diesel exhaust particles has
31 been demonstrated for B[a]P and various nitroarenes.
December 1994 1_23 DRAFT-DO NOT QUOTE OR CITE
-------
1 It is generally accepted that the formation of covalent adducts with DNA and the
2 subsequent alteration of cellular genetic information may represent one mechanism of diesel
3 exhaust-induced animal carcinogenicity. Several reports have provided data indicating the
4 formation of such adducts in animals following administration of known carcinogens and
5 after long-term exposure of animals to diesel exhaust. Molecular dosimetry studies have
6 suggested that DNA adduct formation is one step in diesel exhaust-induced carcinogenesis.
7 This premise is substantiated by several findings, including an increase in DNA adducts in
8 the same pulmonary regions where tumors occur, the fact that DNA adduct levels are greater
9 in species known to be susceptible to diesel exhaust-induced tumors, and that DNA adduct
10 levels are greater following exposure to diesel exhaust particles versus particles lacking
11 adsorbed organic chemicals.
12 The DNA adduct mechanism is based to a considerable extent on the presumed
13 availability of carcinogenic agents such as B[a]P to the target cells. Although such
14 compounds are present, they represent only a small percentage of all the organic agents
15 adsorbed to the exhaust particle. At this time, it is uncertain if their concentration at the
16 target cells, under the exposure conditions used, is adequate to account for the tumorigenic
17 effects reported. Studies with pyrolized pitch condensate have increased this uncertainty.
18 Pitch particles, containing about three orders of magnitude greater concentration of B[a]P
19 than diesel exhaust particles, were not significantly more effective in lung tumor induction
20 than was diesel exhaust.
21 Another proposed mechanism is based on carcinogenic activity of the particle itself.
22 Support for a particle effect is provided by reports that carbon black, which is essentially
23 devoid of organics but is otherwise similar in composition to the carbon core of the diesel
24 exhaust particle, is about as effective in the induction of lung cancer as is whole diesel
25 exhaust. It has been hypothesized that the particles induce releases of mediators from the
26 macrophages that include at least some of the following: reactive oxygen species, chemotactic
27 factors, lysosomal hydrolases, other proteinases, prostaglandins, plasminogen activators, and
28 growth factors. Many of these factors are considered to act via promotion, although the
29 mechanisms remain to be elucidated. Tumor initiation, however, is also likely to be induced
30 because carbon black exposure was also shown to induce increases in DNA adducts in the
31 lungs.
December 1994 1-24 DRAFT-DO NOT QUOTE OR CITE
-------
1 Lung damage with accompanying increases in tumor incidence is exacerbated by
2 exposure to high particle concentrations, which results in particle overload and concomitant
3 inhibition of clearance. The effectiveness of biochemically inert particles in inducing tumors
4 has also been shown to be related to particle surface area. Particle surface area and the
5 particle load in the macrophages are likely to act via release of one or more of the previously
6 listed factors. Although preliminary data provide support for this belief, further research is
7 needed.
8 In summary, although the mechanisms are still uncertain, current evidence indicates that
9 the diesel exhaust particles themselves may be involved in the tumor induction process.
10 Although organics adsorbed to the particle surface are likely to play a role in the tumorigenic
11 process, it is uncertain if concentrations of carcinogenic agents present in this moiety are
12 adequate to account for the observed responses. Evidence for the role of the gaseous fraction
13 of diesel exhaust in lung tumor induction is very limited. Risk estimates based on particle
14 concentration per unit surface area of lung were therefore considered to be reasonable,
15 although additional estimates based on bioavailable organics were also developed.
16
17
18 1.11 WEIGHT OF EVIDENCE CLASSIFICATION FOR
19 CARCINOGENICITY AND QUANTITATIVE ESTIMATE
20 OF UNIT RISK
21
22 On the basis of limited evidence for carcinogenicity of diesel engine emissions in
23 humans, supported by adequate evidence in animals and positive mutagenicity data, diesel
24 engine emissions are considered to best fit the weight-of-evidence Category Bl. Agents
25 classified into this category are considered to be probable human carcinogens.
26 Risk estimates can be derived from either human or animal experiments. Each source
27 of information carries with it its own set of strengths and limitations. Estimates based on
28 human studies reflect the direct observation of an association between exposure and human
29 cancer. These human estimates, however, are limited by the difficulty of reconstructing
30 reliable estimates of exposures many years in the past and distinguishing the influence of
31 confounding exposures to other carcinogens. Conversely, estimates based on animal studies
32 benefit from precisely measured exposures and the absence of many potentially confounding
December 1994 !_25 DRAFT-DO NOT QUOTE OR CITE
-------
1 factors. The use of animal estimates, however, involves uncertainty in the extrapolation of
2 dose and response rates to humans.
3 From human studies, published unit risk estimates range from 6 x 10"4 to
4 3 x 10~3//zg/m3. An upper bound risk estimate of 3 x 10~3//ig/m3 was reported for London
5 transport workers. (In view of the nonpositive findings of this study, a lower bound estimate
6 would encompass 0.) Upper bound risk estimates of 6 x 10"4 and 2 x 10"3//ug/m3 were
7 calculated for railroad workers, assuming mean occupational exposure concentrations of
8 500 or 125 pg/m3, respectively.
9 From animal experiments, upper bound unit risk estimates can be calculated using the
10 linearized multistage procedure, a default method used when the mechanism of action is
11 unknown, the information required by a mechanistic model is unavailable, or the suspected
12 mechanism or background conditions are consistent with linearity at low incremental
13 exposures. The linearized multistage procedure was applied to three rat experiments that
14 collectively span a 50-fold range of doses; this yielded unit risk estimates ranging from
15 1.6 to 7.1 x 10~5/ftg/m3. These estimates are based on two parallel assumptions to the
16 RfCs: (1) that carbon particles are primarily responsible for both toxic and carcinogenic
17 effects, and (2) that equivalent sensitivity occurs across species when dose is expressed as
18 mass per unit surface of the pulmonary region. Dosimetric adjustments from rats to humans
19 and from experimental regimes to continuous lifetime exposure also were similar to those
20 used for RfCs. The primary difference is the assumption of a threshold in the development
21 of RfCs.
22 In addition, an alternative low-dose extrapolation model was developed to account for
23 the possible tumor-initiating effects of the particles. Much of the data needed to estimate the
24 model's parameters, however, are lacking.
25 In view of the uncertainty inherent in these types of calculations, the human and animal
26 estimates should be viewed as complementary. For a bounding estimate intended to
27 determine whether an exposure level has a potential to pose a hazard to human health, the
28 published human estimates may be practical for exposure levels in the range of observations
29 in these studies. On the other hand, projection of the public health impact of an exposure
30 level may benefit from using estimates derived from animal experiments, because of the
31 closely controlled conditions and their precisely measured exposure levels, absence of many
December 1994 1-26 DRAFT-DO NOT QUOTE OR CITE
-------
1 confounding factors, and narrow confidence limits around the tumor incidence rates. A unit
2 risk estimate of 3.4 x 10~5//zg/m3 for continuous lifetime exposure, which is the geometric
3 mean of the upper bound estimates calculated from the three rat experiments, is therefore
4 recommended. The proper use and understanding of these risk estimates is discussed in
5 Chapter 12.
6
7
8 1.12 RISK CHARACTERIZATION
9 The chapter on risk characterization includes critical analyses of information, and is
10 presented such that summarizing its contents is not feasible. The chapter is, in essence, a
11 summary and interpretation of many aspects of this document. Therefore, an executive
12 summary is not provided and the reader is referred directly to Chapter 12.
December 1994 !_27 DRAFT-DO NOT QUOTE OR CITE
-------
1 REFERENCE
2 U.S. Congress. (1990) Clean Air Act amendments of 1990. Washington, DC: U.S. Government Printing Office.
December 1994 1-28 DRAFT-DO NOT QUOTE OR CITE
-------
i 2. DIESEL EMISSIONS
2
3
4 2.1 INTRODUCTION
5 The diesel engine was patented in 1892 by Rudolf Diesel, when it was conceived as a
6 prime mover that would provide much improved fuel efficiency compared with contemporary
7 engines. To the present day, the diesel engine's high efficiency remains its strongest selling
8 point. Currently, in the United States the diesel engine is used mainly in trucks, buses,
9 agricultural and other off-highway equipment, locomotives, ships, and in many stationary
10 applications.
11 The aim of this chapter is to present an accurate perspective on the diesel engine as a
12 contributor to the mobile source emissions inventory. Diesel engines emit both gas-phase
13 pollutants (hydrocarbons [HC], oxides of nitrogen [NOJ, and carbon monoxide [CO]) and
14 carbonaceous paniculate matter. In the following sections a brief description of the diesel
15 engine, its combustion system, pollutant formation mechanisms, and emission factors will be
16 presented.
17 The diesel engine compresses air to high pressure and temperature. Fuel when injected
18 into this compressed air autoignites, thereby releasing its chemical energy, and the resulting
19 combustion gases expand doing work on the piston before being exhausted to the atmosphere.
20 Power output is controlled by the amount of injected fuel rather than throttling the air intake.
21 Compared to its spark-ignited (SI) counterpart, the diesel engine's superior efficiency derives
22 from a higher compression ratio and no part-load throttling. Because of its poorer air
23 utilization, a larger piston displacement is required by a diesel engine for the same power
24 output as a comparable SI engine. To ensure structural integrity for prolonged reliable
25 operation at higher peak pressures brought about by a higher compression ratio and
26 autoignition, the structure of a diesel engine generally needs to be more massive than its
27 SI counterpart.
28 Diesel engines may be broadly identified as being either two- or four-stroke cycle,
29 injected directly or indirectly, and naturally aspirated or supercharged. Further, they are
30 frequently classified according to service requirements such as light-duty (LD) or heavy-duty
31 (HD) automotive, small or large industrial, and rail or marine engines. Here, because they
December 1994 2-1 DRAFT-DO NOT QUOTE OR CITE
-------
1 are currently regulated, emissions from automotive diesel engines will be the focus.
2 Approximately one half of the yearly production of American diesel engines go into highway
3 vehicles; the majority of the remainder is divided among agricultural, construction, mining,
4 marine, and a variety of stationary applications (Schulz, 1988).
5 All diesel engines use hydraulic fuel injection in one form or another. The fuel system
6 must meet four main objectives if a diesel engine is to function properly over its entire
7 operating range: (1) meter the correct quantity of fuel, (2) distribute the metered fuel to the
8 correct cylinder, (3) inject the metered fuel at the correct time, and (4) inject the fuel so that
9 it is atomized and mixes well with the in-cylinder air. The first two of these objectives are
10 the functions of a well-designed injection pump, and the last two are mostly a function of the
11 injection nozzle. As a part of the effort to obtain lower exhaust emissions without
12 diminishing fuel efficiency, current fuel injection systems are moving toward the use of
13 electronics for more flexible control than is available with purely mechanical systems.
14 Both the fuel and the lubricants that are used to service diesel engines are highly
15 finished petroleum-based products combined with chemical additives. Diesel fuel oil is a
16 mixture of many different hydrocarbon molecules from about C7 to about C35, with a
17 boiling range from roughly 350 to 650 °F. Many of the fuel oil properties such as its
18 specific energy content, ignition quality, and specific gravity are related to its hydrocarbon
19 composition. Therefore, one can easily surmise that fuel composition affects many aspects of
20 engine performance, including economy and exhaust emissions. For example, a decrease of
21 fuel aromatic content, sulphur, and volatility usually leads to a reduction of regulated
22 emissions (Ullman, 1989). Figure 2-1 (Obert, 1973) shows a typical pressure-tune trace for
23 a diesel engine that illustrates the four stages of the combustion process.
24
25 1. Ignition delay period—the elapsed time from the start of injection until the start of
26 combustion. This is the time required to atomize fuel, evaporate droplets, and mix
27 vapor with air and for the necessary preflame reactions to occur. The ignition
28 delay period is really two inseparable overlapping delay periods, physical delay
29 and chemical delay.
30
31 2. Uncontrolled burning period—during this stage the fuel that has passed entirely
32 through the first stage autoignites and burns in a premixed fashion and then
33 diffusion takes over control of the burning (Lyn, 1963; Kahn, 1970). During this
34 stage a high rate of pressure rise and noise associated with diesel knock occur.
35
December 1994 2-2 DRAFT-DO NOT QUOTE OR CITE
-------
1.200
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
120°
90
60° 30" TDC 30" 60°
Time (degrees of crankshaft rotation)
120°
Figure 2-1. Full load diesel engine pressure trace.
Source: Obert, 1973.
3. Controlled burning period—during this stage, the fuel bums as it is injected in
what is essentially a diffusion-controlled process. The burning rate and the rate at
which energy is released in this stage is lower than during the second stage.
4. Afterburning period—the elapsed time from the end of fuel injection until the end
of combustion. This stage is characterized by the diffusion mode giving way to the
premixed mode of combustion.
December 1994
2-3
DRAFT-DO NOT QUOTE OR CITE
-------
1 2.2 OVERVIEW OF DIESEL POLLUTANTS AND POLLUTANT
2 FORMATION
3 2.2.1 Gas-Phase Pollutant Emissions
4 2.2.1.1 Oxides of Nitrogen Formation
5 Because the diesel combustion process is very complex and involves burning of fuel
6 droplets, it has proven difficult to predict pollutant concentrations or emission rates
7 quantitatively. In SI gasoline engines, NO emission can be quantitatively explained by
8 adiabatic compression of the initially burned mixture (nearest the spark plug) by the
9 combustion pressure development in the later stages of combustion. Thus, the originally
10 burned gases are raised to a much higher temperature than that achieved in the flame by the
11 subsequent compression. Shahed (1985) has reviewed the work on this subject for diesel
12 engines. He reports that, qualitatively, NO formation in diesel engines cannot be explained
13 by this phenomenon. The time-temperature history of the burning droplets seems to
14 determine the extent of NO formation. Yu and Shahed (1981) have defined relevant engine
15 operating parameters that control NO emissions for DI HD engines. For instance, retarding
16 injection or recirculating exhaust gases reduces NO formation and emission at the expense of
17 increasing soot formation and hydrocarbons, all other factors being equivalent. Wu and
18 Peterson (1986) studied NO formation kinetics in an IDI passenger car diesel engine over a
19 wide range of operating conditions. These authors have found that a variable-temperature
20 model accounting for the average gas temperature at the time of droplet burning accounted
21 for the observed NO considerably better than a constant-temperature (peak-cycle-temperature)
22 model. Global NO formation rates found in their study suggested that the NO must be
23 formed in the vicinity of the droplet flame zone.
24 More recently, Lipkea et al. (1987) and Lipkea and DeJoode (1987) have constructed a
25 successful engine model that adequately explains NO formation. Process parameters that
26 control NO formation included fuel jet momentum flux, in-cylinder air density at the start of
27 injection, swirl cross-flow momentum flux, and in-cylinder temperature. Air system design
28 characteristics that change air density and temperature can result in constant work but
29 decreasing NOX. On the other hand, decreasing cylinder temperatures also result in
30 increased hydrocarbon emissions (Uyehara, 1987; Gill, 1988). The net result experienced by
December 1994 2-4 DRAFT-DO NOT QUOTE OR CITE
-------
1 the engine designer is a group of design trade-offs between NOX and paniculate material at
2 high temperatures and between NOX and hydrocarbon at low operating temperatures.
3
4 2.2.1.2 Hydrocarbon and Carbon Monoxide Formation
5 Small quantities of gaseous unburned HCs and of CO are emitted from diesel engines,
6 rather less than is the case with comparable SI engines, however. Myers and Uyehara
7 (1947) have explained the observed CO on the basis of locally rich combustion. During the
8 ignition delay period, especially small amounts of fuel vaporize from the initial droplets.
9 The gas-phase reactions of this material are responsible ultimately for its ignition and, thus,
10 the ignition of the droplets. However, this premixed patch is likely to be locally quite rich,
11 even though the overall fuel-air mixture has considerable excess air. Thus, CO is formed in
12 concentrations of 2,000 ppm or even slightly more in diesel exhaust. By comparison, typical
13 gasoline engines might have exhaust CO concentrations of 10,000 to 20,000 ppm.
14 These locally rich combustion processes are also responsible for a small release of low
15 molecular weight hydrocarbons, principally methane, ethylene, and acetylene, in diesel
16 exhaust. Black and High (1979) reported a gas chromatographic study of diesel exhaust
17 hydrocarbon species from a passenger car. Generally, the hydrocarbon emission1 rates were
18 well within the 0.41 g/mi allowed by the Federal emissions standards of recent years.
19 About 10% of these materials are C1-C4 combustion derived compounds. The bulk of the
20 emission in the gas phase was diesel fuel in the CIO to C25 molecular weight range; these
21 materials account for 70 to 80% of the HC emitted. The balance of the material, including
22 particle-bound HC, was in the same molecular weight range as lubricating oil. Hare et al.
23 (1976) and Hare and Bradow (1979) have reported similar findings with HD diesel engines,
24 both of the two-stroke cycle and four-stroke cycle types. Therefore, both fuel and lubricant
25 can supply organic matter to diesel engine exhaust HCs, both in the gaseous and paniculate
26 states.
27 Hampton et al. (1983) reported a GC-MS study of heavy HCs in a Pennsylvania
28 turnpike roadway tunnel having varying amounts of diesel and gasoline passenger car traffic.
29 There were characteristic differences in gaseous HC content of the tunnel gases between the
30 diesel and gasoline vehicles. Typically, diesel traffic was characterized by substantial
31 quantities of aliphatic hydrocarbons with lesser amounts of long-chain substituted
December 1994 2-5 DRAFT-DO NOT QUOTE OR CITE
-------
1 monoaromatics. Gasoline traffic was dominated by methyl- and ethyl-benzene emissions,
2 based on the porous polymer trapped samples used in this study. These results are in general
3 agreement with the Black and High (1979) conclusion that diesel HC emissions are primarily
4 fuel-derived.
5
6 2.2.2 Particle Formation and Emission
7 In 1980, a major symposium dealt with the theory and experimental evidence relating to
8 carbon formation in flames; the proceedings (Siegla and Smith, 1980) contain reviews of
9 most of the pertinent evidence on this subject. The chemical mechanism that accounts for
10 carbon formation is not completely established; the major weight of scientific opinion seems
11 to support some role for intermediate formation of polycyclic aromatic hydrocarbons in the
12 process. Diffusion flames, whether rich or lean, usually form some carbon. However,
13 carbon is consumed on the lean side by reaction with hydroxyl (OH) radicals (Fenimore and
14 Jones, 1967), and only when the OH radical population is reduced by other reactions (with
15 fuel hydrocarbons, for example) is carbon found to be a major combustion product. Thus,
16 carbon is normally a stable combustion product only of rich flames.
17 Uyehara and Beck (1988) have pointed out that there is a very good linear correlation
18 between CO emission rate and carbon emission rate from HD engines. Once formed, both1
19 substances are difficult to remove, requiring highly energetic OH radicals for reaction.
20 These authors argue that carbon formation normally takes place over a rather narrow
21 temperature range and that the maximum rate of carbon formation is found at temperatures
22 around 2,250 K. At temperatures of 2,400 K and above, carbon is burned out and at
23 temperatures of 1,900 K and below, it is never formed. However, if attempts are made to
24 raise temperature and thus burn out the carbon, high NO results. If flame zone temperatures
25 are lowered, by adding water, alcohol, or exhaust gas, HC emissions are increased (Ball,
26 1987; Kadota and Henein, 1981). Uyehara and Beck have described a qualitative model of
27 droplet combustion in which the burning rate of fuel droplets is controlled by boiling rate.
28 It is shown that an ideal condition would involve high-pressure, high-velocity injection with a
29 minimum ignition-delay period. Under these conditions, carbon can be minimized by
30 reducing rich gaseous combustion and NO can be controlled by delayed-injection timing.
31 Under these conditions, the fuel cetane number might become a critical parameter CGntroffing
December 1994 2-6 DRAFT-DO NOT QUOTE OR CITE
-------
1 the ignition delay and the region of uncontrolled combustion. In practice, Ullman (1989) has
2 shown that reducing aromatic content and, hence, decreasing the ignition delay of diesel fuel
3 is beneficial in the control of paniculate material and of NOX emissions in three HD engines
4 built to meet the 1988 California or 1991 Federal emissions standards. Tosaka et al. (1989)
5 have studied the effect of fuel aromatics in promoting of diesel carbon formation. These
6 authors have found an aliphatic radical-benzene condensation process that apparently accounts
7 for the additional amounts of carbon that result in the diesel combustion of aromatic fuels.
8
9 2.2.3 Gas-to-Particle Conversion
10 2.2.3.1 Condensation of Organic Matter
11 Generally, the formation of carbon particles is thought to involve growth of particles by
12 polymerization of gaseous intermediates at the surface of small particles (Kadota and Henien,
13 1981; Plee et al., 1981). Thus, the growth and agglomeration of carbon particles are
14 probably gas-to-particle processes. Ross et al. (1982) studied the properties of diesel
15 particles obtained from an engine operated on high-purity dodecane. In this case, the fuel
16 was too volatile to have an impact on particle composition. These authors found that the
17 carbon particles contained an HC film, which must have been condensed from the gas phase.
18 The carbon had rather low specific surface area, about 0.5 m2/g, which could be materially
19 increased by high temperature treatment.
20 Heats of sorption of a variety of HCs were determined by a GC technique. It was
21 found that the absorptivity of gaseous organic compounds on these particles was controlled
22 by Henry's Law absorption in the organic surface film. The presorbed organic layer was
23 several layers thick, and this material essentially acted as an organic droplet, dissolving
24 materials from the gas phase. Thus, the heat of sorption was adequately explained by the
25 heat of vaporization of the organics (Ross et al., 1982).
26
27 2.2.3.2 Oxidation of Sulfur Oxides
28 Studies of diesel particle composition have produced some information about the fate of
29 fuel sulfur. In the earliest studies, sulfate was found to be a significant component of diesel
30 particles (Hare et al., 1976). Generally, the sulfate found in particles accounted for only
31 about 2% of the fuel sulfur charged, the balance being emitted as sulfur dioxide (SO^).
December 1994 2-7 DRAFT-DO NOT QUOTE OR CITE
-------
1 Attempts to detect sulfite in diesel particle samples have involved extraction in
2 tetrachloromercurate or conversion to the formaldehyde busulfite addition product followed
3 by West-Gaeke colorimetry and direct ESCA detection. Neither of these methods has
4 demonstrated the presence of sulfur (IV).
5 Sulfate emissions rates have been measured using both engine and chassis dynamometer
6 test procedures. Hare et al. (1976) measured composite particle emissions rates on the older
7 federal compliance test for HD engines and found that with overall emissions rates from
8 0.3 to 1.0 g/kW-h (0.4 to 1.3 g/bhp-h) of particle mass, sulfate emissions were relatively
9 constant at about 0.02 g/kW-h (0.03 g/bhp-h). Dietzmann et al. (1980) measured sulfate
10 emission rates from a number of HD vehicles driven over simulated urban driving schedules
11 and found emission rates from about 0.03 to 0.05 g/km for overall particle emissions rates
12 from 0.5 to 1.6 g/km. Thus, with previous engines, sulfate was a significant but small (2 to
13 3%) component of diesel particle mass. With newer engine designs and particle emission
14 rates characteristically below 0.25 g/bhp-h, this emission rate amounts to 10 to 15% of the
15 emitted particle mass, a very important portion of the allowable limit. Currently, no means
16 of reducing this sulfate formation is available other than reducing the sulfur concentration of
17 diesel fuel.
18 Earlier in this document, the evidence regarding the oxidation of nitrogen to nitric
19 oxide was documented. Dietzmann et al. (1980) have reported analysis of HD diesel vehicle
20 exhaust for NO2 during simulated urban driving. These authors have reported NO2
21 concentrations from about 1 to 5 ppm, accounting for 2 to 5% of the NO emitted. Harris
22 et al. (1987) have measured NO2 and HNO3 in air-diluted diesel exhaust. In the exhaust
23 from HD engines, the NO2 concentration was from 1 to 30 ppm, whereas the HNO3 ranged
24 from 0.08 to 0.8 ppm. Generally, HNO3 accounted for a few percent of the NO2, which, in
25 turn, accounted for a few percent of the NO.
26
27 2.2.4 Nitroarene Formation
28 The soluble extract from diesel-generated paniculate material was shown to cause
29 mutations when subjected to a bacterial assay (Huisingh et al., 1978). It was not long before
30 evidence began to point toward the class of compounds consisting of nitrated PAHs that have
31 come to be called nitroarenes (Pederson and Siak, 1981; Newton et al., 1982). Bradow
December 1994 2-8 DRAFT-DO NOT QUOTE OR CITE
-------
1 et al. (1979) reported a summary of the EPA findings to that date on the formation of
2 semipolar mutagens in both HD and LD diesel engines. Dietzmann et al. (1980, 1981)
3 reported emissions of mutagenic material from a variety of HD trucks operated over transient
4 driving cycles, and Gibbs et al. (1980) reported the emission of bacterial mutagens from a
5 large number of in-use diesel passenger cars.
6 Very-high-resolution organic analytical procedures have been applied to diesel exhaust
7 samples (Liberti et al., 1984; Schuetzle and Frazer, 1986; Schuetzle and Perez, 1983).
8 Generally, a variety of nitrated polynuclear aromatic compounds has been found, which
9 accounts for a substantial portion of the mutagenicity found. However, not all the bacterial
10 mutagenicity has been identified in this way, and the identity of the remainder of the
11 mutagenic compounds remains unknown. The identity of the nitrated aromatics thus far
12 identified in diesel exhaust was the subject of review in the International Agency for
13 Research on Cancer (1989) monograph on diesel exhaust (Table 2-1).
14 However, it has been shown recently that nitroarenes are not an inevitable consequence
15 of combustion of hydrocarbon fuels (Salmeen et al., 1989). A laboratory pulse-flame
16 combustion burning toluene, cyclohexane, or diesel fuel, under conditions believed to
17 enhance the nitroarene formation (excess NOX), did not produce direct-acting mutagens and
18 therefore, presumably, no nitroarenes.
19
20
21 2.3 EMISSIONS FACTORS AND EMISSIONS INVENTORIES
22 2.3.1 Gaseous Pollutant Emission Factors
23 There are two systems of units by which emissions from diesel-powered vehicles may
24 be expressed. Generally, emissions are expressed as mass of the pollutant divided by some
25 measure of societal value associated with the use of the vehicle in question. For example,
26 the societal value of LD vehicles (cars and trucks) is primarily to convey people from place
27 to place. Somewhat less significantly, these vehicles may also be used to move some light
28 loads of goods or materials. The unit of societal value in either case has been taken to be the
29 distance traveled in this transport. Because the number of people moved or the weight of
30 goods is small in either case, it can be argued that it is the travel itself which serves as the
December 1994 2-9 DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 2-1. SOME NTTROARENES IDENTIFIED IN VEHICLE EXHAUST8
1,3-Dihydroxynitropyrenel
2,5-Dinitrofluorene
2,7-Dinitrofluorene
2,7-Dinitro-9-fluorenone
1,3-Dinitropyrene
1,6-Dinitropyrene
1,8-Dinitropyrene
9-Methylcarbazole
1 -Nitro-3-acetoxypyrene
9-Nitroanthracene
2-Nitroanthracene or -phenanthrene
x-Nitroanthracene or -phenanthrene (two isomers)b
6-Nitrobenzo[fl]pyrene
x-Nitrobenzoquinolineb
2-Nitrobiphenyl
3-Nitrobiphenyl
4-Nitrobiphenyl
1-Nitrochrysene
x-Nitrodibenzothiophene (two isomers)b
x-Nitro-y,z-dimethylanthracene or -phenanthrene (five isomers)b
1 -Nitrofluoranthene
3-Nitrofluoranthene
7-Nitrofluoranthene
8-Nitrofluoranthene
2-Nitrofluorene
3-Nitro-9-fluorenone
10-Nitro-l-methylanthracene or -phenanthrene
10-Nitro-9-methylanthracene or -phenanthrene
x-Nitro-y-methylanthracene or -phenanthrene6
1 -Nitro-2-methylnaphthalene
3-Nitro-1 -methylpyrene
6-Nitro-1 -methylpyrene
8-Nitro-1 -methylpyrene
1 -Nitronaphthalene
2-Nitronaphthalene
2-Nitrophenanthrene
1-Nitropyrene
5-Nitroquinoline
8-Nitroquinoline
x-Nitroterphenylb
x-Nitro-y,z,z'-trimethylanthracene or -phenanthrene (six isomers)b
x-Nitrotrimethylnaphthalene (three isomers)6
aFrom International Agency for Research on Cancer, 1989.
bx, y, z and z' imply position is unknown.
December 1994 2-10 DRAFT-DO NOT QUOTE OR CITE
-------
1 use index. Therefore, LD vehicle emissions are generally expressed as grams per mile or
2 grams per kilometer.
3 In the case of HD vehicles, the generally accepted index of value is the amount of work
4 done by the engine. By fixing the index as work done, large multicylinder engines can be
5 rated on more-or-less the same scale as less-capable smaller engines. By this means, some
6 equity between relatively small and very large truck engines can be maintained in the task of
7 meeting emissions standards. Therefore, emissions standards are normally expressed in
8 terms of grams per unit work (e.g., grams per brake horsepower-hour or g/Kw-h for all
9 HD engines). It should also be noted that characteristically, HD engines are manufactured
10 separately from trucks by different manufacturers; therefore, it is the engine whose emissions
11 are certified and guaranteed, not those of the overall truck or bus in which the engine is
12 used.
13 Although this procedure does produce equitable treatment for engines of a wide variety
14 of uses and displacements, it does create problems when one tries to estimate the influence of
15 vehicle class emissions on ambient air quality. Ordinarily, the only measure of vehicle use
16 available, either for urban areas or for particular roadways, is traffic count data plus some
17 estimate of average vehicle speed. From these two volume indices, estimates of vehicle-
18 distance-traveled can be made. Thus, to estimate traffic emissions, it is necessary to have
19 them expressed as grams/mile. Further, it is desirable to have some measure of in-use
20 emissions performance of HD vehicles such as that for LD vehicles. For these reasons,
21 chassis dynamometer performance tests that relate engine certification data to overall truck or
22 bus emissions rates have been developed and used to a limited extent (Dietzmann et al.,
23 1981).
24 For the most part, however, the primary data base used for estimated diesel vehicle
25 emission rates has been the certification engine performance test data converted from a work
26 to a distance-traveled basis. The emissions data for HD diesel vehicles is contained within
27 the emissions factor tables in the AP-42 handbook (U.S. Environmental Protection Agency,
28 1985) or generated by the Mobile 4 computer model (U.S. Environmental Protection
29 Agency, 1989). These tables, using assumptions about the degree to which certification
30 practice is reflected in in-use trucks, are also used to calculate both past and future emissions
31 factors.
December 1994 2-11 DRAFT-DO NOT QUOTE OR CITE
-------
1 For example, Figures 2-2, 2-3, and 2-4 present emissions factors for CO, HCs, and
2 NOX respectively, as predicted by the Mobile 4 computer model as a function of forward
3 vehicle speed. In these runs, default values for air temperature, cold-hot start mix, vehicle
4 model year mix, and I/M program details were used. It is assumed that the heavy-duty
5 mileage is accounted for by trucks and buses in the ratio of 3:1, the national average value.
6 In sea- or riverport cities, this value may be an underestimate of truck traffic, whereas, in
7 other cities the bus use may be underestimated. User-supplied values may be substituted for
8 virtually any of these defaults in the ordinary use of this model. Generally, the emissions of
9 diesel engines are somewhat more stable and less influenced by air temperature than are
10 other vehicle classes. Therefore, the emissions factors listed here are a fair measure of
11 overall heavy-duty vehicle emissions performance. By comparison, the emissions of all
12 vehicles on the road for these periods are given in Table 2-2.
13 It is clear that the emissions factors for heavy-duty vehicle CO and HC are somewhat
14 lower but comparable with those of the all-vehicle estimates (which is dominated by gasoline-
15 fueled passenger cars). Therefore, the control of these emissions has a small but discernable
16 influence on overall CO and HC emissions. In the case of NOX, however, heavy-duty
17 emissions are currently many times that for average traffic. Further, even with stringent
18 controls, these emissions are likely to be quite significant and to be an increasing fraction of
19 the overall mobile source NOX inventory.
20 In a few instances, chassis dynamometer measurements of in-use truck and bus
21 emissions have been made (Dietzmann et al., 1981; Warner-Selph and Dietzmann, 1984).
22 The number of vehicles tested in this manner has been relatively small (e.g., 6 in the first
23 two studies and 28 in the last). However, comparison of these test results to the Mobile 4
24 projects is instructive, and some of these data are shown in Table 2-2. The discrepancies are
25 generally fairly small, except in the CO emissions from in-use buses. Generally, these
26 values are substantially greater than the Mobile 4 predictions for overall HD engines. This
27 phenomenon has been investigated further by comparing engine-versus-chassis dynamometer
28 tests with the same engine (Warner-Selph and Dietzmann, 1984). For most categories of HD
29 engines, gaseous emissions from the two types of tests are generally comparable, although
30 differences of 25 to 30% are fairly common. However, for the bus engines, the chassis
31 dynamometer procedure generally produces more than twice the emissions of CO or of
December 1994 2-12 DRAFT-DO NOT QUOTE OR CITE
-------
Figure
2-2. Mobile 4 projections: Carbon
50 60
Vehicle Speed (mph)
m0noxideennssions-beavy-duty diesel.
Vehicle Speed (mph)
Figure 2-3
December 1994
. Mobile 4 projections: Hydrocarbon
2.13
DRAFT-DO NOT QUOTE OR CITE
-------
I
w
g
'
(0
E
LU
50
45
40-
35
30
251
20
15
10
10
20
30 40 50
Vehicle Speed (mph)
60
70
Figure 2-4. Mobile 4 projections: Nitrogen oxide emissions—heavy-duty diesel.
TABLE 2-2. MOBILE 4 AND CHASSIS DYNAMOMETER EMISSIONS8,
URBAN AVERAGE VEHICLE SPEED, 1980 CALENDAR YEAR
Study or Source
HC
CO
NO.
Particles
Mobile 4 HDD
Mobile 4 All Vehicles
Dietzmann et al. (1980)
(three 4-stroke engines)
Dietzmann et al. (1980)
(one bus engine)
Warner-Selph and Dietzmann (1984)
5 single-axle trailers
17 dual-axle trailers
6 city buses
4.92
9.33
1.98
2.20
3.18
2.78
2.73
15.55
58.63
6.72
60.50
6.00
11.50
43.84
28.91
4.40
24.50
30.55
14.90
27.20
19.84
0.92
2.88
1.71
2.35
1.54
aln grams per mile.
December 1994
2-14
DRAFT-DO NOT QUOTE OR CITE
-------
1 particulate material as are found in the engine dynamometer test. No more recent test
2 comparisons are available for controlled engine types. Therefore, it is not clear whether this
3 disparity still exists.
4 Diesel passenger car and light-truck emissions of CO and HC are considerably lower
5 than those of gasoline vehicles, whereas NOX values are similar. Because these classes
6 collectively only account for less than 1 % of traffic, they do not materially influence the
7 emissions inventories for any of these gases.
8
9 2.3.2 Particulate Matter Emissions Factors
10 Particle measurements on HD diesel engines were initially made on the 13-mode test
11 procedure used in the early 1970s (Hare et al., 1976). One two-stroke cycle engine
12 produced an emission rate of 0.9 g/kW-h, whereas a comparable four-stroke cycle engine
13 produced only 0.34 g/kW-h. Particle mass emissions rates generally increased with
14 increasing power output at steady speeds. The two-stroke cycle particulate material was
15 quite oily (e.g., 50% of the mass could extracted in nonpolar solvents, whereas only 9% of
16 the four-stroke cycle material could be so extracted). The molecular weight distribution of
17 the organic matter from the two-stroke cycle engine matched that of the lubricating oil;
18 furthermore, the particulate material was rich in calcium and phosphorous, which are
19 lubricating oil constituents. Apparently, lubricant consumption can materially influence both
20 the rate of particle emission and the particle composition.
21 Similar results were obtained by Hare and Bradow (1979) in a study of fuel effects on
22 diesel particulate material. Two engines used in this study produced about 0.9 and
23 1.9 g/kW-h with the four- and two-stroke cycle engines, respectively. In this study also,
24 lubricant composition and lubricant consumption appeared to have significant influence on
25 particle mass emissions. However, the main result was the influence of overall fuel
26 composition on particleemissions. Generally, low-sulfur, low-aromatics fuels produced the
27 smallest particle mass emissions. The range of emission rates ranged from 1.92 to
28 1.19 g/kW-h for the two-stroke cycle and from 0.98 to 0.52 g/kW-h as fuel properties
29 improved. These same effects have been seen more recently with current HD engines
30 (Ullman, 1989; Tosaka et al., 1989; Springer, 1990). Similarly, Springer (1989) has shown
31 that lubricant consumption and lubricant trace elements can contribute 30% or more;
December 1994 2-15 DRAFT-DO NOT QUOTE OR CITE
-------
1 therefore, there is a need to determine particle emissions, particularly for controlled
2 HD engines. McGeehan et al. (1988b) have shown that low-ash oils (1 % ash or less)
3 provide adequate protection, probably lower oil consumption, and lower paniculate emissions
4 than conventional oils. Yost et al. (1985) have suggested that overall combustion
5 performance including emissions performance is enhanced with low-aromatics, high-cetane
6 fuels for DI type engines; volatility is also a significant, if secondary, factor. However,
7 Wade and Jones (1984) suggest that cetane number alone is not an adequate descriptor of
8 combustion or emissions performance in passenger-car diesel engines.
9 Springer (1989, 1990) has produced good general discussions of the influence of fuel
10 and lubricant composition, respectively, on future particle mass emissions from HD diesel
11 engines. He points out that, as the particle emissions target moves from 0.20 g/bhp-h (the
12 1991 target) to 0.08 g/bhp-h, the manufacturers target for a 0.15-g/bhp-h EPA standard, oil
13 consumption, fuel sulfur oxidation, and engine wear products all become significant
14 contributors to emissions. The lubricant consumption issue may be influenced by the choice
15 of noble-metal catalyzed trap oxidizer control technology (see Section 2.4); however, this
16 choice will only put more pressure on reductions in fuel sulfur. To meet such stringent
17 standards, it is likely that reductions in fuel sulfur, together with changes in other fuel
18 properties, will be necessary (Springer, 1989; Balzotti et al., 1990).
19 Particle emissions from HD vehicles on a chassis dynamometer have been measured in
20 addition to the gaseous emissions mentioned earlier. Table 2-2 presents data from the same
21 three studies (Dietzmann et al., 1980, 1981; Warner-Selph and Dietzmann, 1984).
22 Generally, the emission rates of 1.5 g/km (2.4 g/mi) for trucks and 2.5 (4.0 g/mi for buses)
23 would dominate the emissions inventory for low-level sources in cities. For example,
24 Bradow (1980) showed that an all-traffic emissions level of 0.2 g/mi would produce an
25 estimated 10 /ig/m3 annual average fine-particle concentration in downtown St. Louis, MO.
26 On the same basis, the composite heavy-duty emission factor for the national average 4% of
27 the total traffic would account for about 5 jig/m3 annually.
28
29
30
December 1994 2-16 DRAFT-DO NOT QUOTE OR CITE
-------
1 2.4 DIESEL ENGINE CONTROL TECHNOLOGY
2 Primary engine parameters affecting diesel emissions are the fuel injection system,
3 engine control system, air intake port and combustion chamber design, and air charging
4 system. Actions to reduce lubricating oil consumption can also impact affect HC and
5 paniculate (PM) emissions. Further, beyond the engine itself, exhaust aftertreatment
6 systems, such as trap oxidizers and catalytic converters, can play a significant role. Finally,
7 modifications to conventional fuels as well as alternative fuels can substantially lower or raise
8 emissions.
9
10 2.4.1 Engine Modifications
11 The geometries of the combustion chamber and the air-intake port control the air
12 motion in the diesel combustion chamber and thus play an important role in air/fuel mixing
13 and emissions. A number of different combustion chamber designs, corresponding to
14 different basic combustion systems, are currently in use in HD diesel engines.
15 Changes in the engine combustion chamber and related areas have demonstrated a
16 major potential for emission control. Design changes to reduce the crevice volume in DI
17 diesel cylinders increase the amount of air available in the combustion chamber. Changes in
18 combustion chamber geometry—such as the use of a reentrant lip on the piston bowl—can
19 markedly reduce emissions by improving air/fuel mixing and minimizing wall impingement
20 by the fuel jet. Optimizing the intake port shape for best swirl characteristics has also
21 yielded significant benefits.
22
23 2.4.2 Engine Control Systems
24 Traditionally, diesel engine control systems have been closely integrated with the fuel
25 injection system, and the two systems are often discussed together. These earlier control
26 systems are entirely mechanical. The last few years have seen the introduction of an
27 increasing number of computerized electronic control systems for diesel engines. With the
28 introduction of these systems, the scope of the engine control system has been greatly
29 expanded. The basic functions of mechanical systems include basic fuel metering, engine
30 speed governing, maximum power limitation, torque curve "shaping", limiting smoke
31 emissions during transient acceleration, and (sometimes) limited control of fuel-injection
December 1994 2-17 DRAFT-DO NOT QUOTE OR CITE
-------
1 timing. Engine speed governing is accomplished through a spring and flyweight system that
2 progressively (and quickly) reduces the maximum fuel quantity as engine speed exceeds the
3 rated value. The maximum fuel quantity itself is generally set through a simple mechanical
4 stop on the rack that controls injection quantity. More sophisticated systems allow some
5 shaping of the torque curve to change the maximum fuel quantity as a function of engine
6 speed.
7 The advent of computerized electronic engine control systems has greatly increased the
8 potential flexibility and precision of fuel metering and injection timing controls. In addition,
9 it has made possible whole new classes of control functions, such as road speed governing,
10 alterations in control strategy during transients, synchronous idle speed control, and adaptive
11 learning—including strategies to identify and compensate for the effects of wear and
12 component-to-component variation in the fuel injection system.
13 By continuously adjusting the fuel-injection timing to match a stored "map" of optimal
14 timing versus speed and load, an electronic timing control system can significantly improve
15 on the NOx/particulate and NOx/fuel economy tradeoffs possible with static or mechanically
16 variable injection timing. Most electronic control systems also incorporate the functions of
17 the engine governor and the transient smoke limiter. This helps to reduce excess paniculate
18 emissions resulting from mechanical friction and lag time during engine transients while
19 simultaneously improving engine performance. Potential reductions in paniculate matter
20 emissions of up to 40% have been documented by using this approach.
21 Other electronic control features, such as cruise control, upshift indication, and
22 communication with an electronically controlled transmission, will also help to reduce fuel
23 consumption, and thus will probably reduce in-use emissions. Because the effect of these
24 technologies is to reduce the amount of engine work necessary per mile, rather than the
25 amount of pollution per unit of work, their effects will not be reflected in dynamometer
26 emissions test results, however.
27
28 2.4.3 Turbocharging and Interceding
29 A turbocharger consists of a centrifugal air compressor that feeds the intake manifold
30 which is mounted on the same shaft as an exhaust gas turbine in the exhaust stream.
31 By increasing the mass of air in the cylinder prior to compression, turbocharging
December 1994 2-18 DRAFT-DO NOT QUOTE OR CITE
-------
1 correspondingly increases the amount of fuel that can be burned without excessive smoke and
2 thus increases the potential maximum power output. The fuel efficiency of the engine is
3 improved as well. The process of compressing the air, however, increases its temperature,
4 increasing the thermal load on critical engine components. By cooling the compressed air in
5 an intercooler before it enters the cylinder, the adverse thermal effects can be reduced. This
6 also increases the density of the air, allowing an even greater mass of air to be confined
7 within the cylinder, thus further increasing the maximum power potential.
/
8 Increasing the air mass in the cylinder and reducing its temperature can reduce both
9 NOX and particulate emissions as well as increase fuel economy and power output from a
10 given engine displacement. Most HD diesel engines are currently equipped with
11 turbochargers, and most of these have intercoolers. Recent developments in air charging
12 systems for diesel engines have been primarily concerned with (1) increasing the
13 turbocharger efficiency, operating range, and transient response characteristics and
14 (2) employing unproved intercoolers to further reduce the temperature of the intake charge.
15 Tuned intake air manifolds (including some with variable tuning) have also been developed to
16 maximize air intake efficiency in a given speed range.
17 Turbochargers for HD diesel engines are already highly developed, but efforts to
18 improve their performance continue. The major areas of emphasis are improved matching of
19 turbocharger response characteristics to engine requirements, improved transient response,
20 and higher efficiencies. Engine/turbocharger matching is especially critical because of the
21 inherent conflict between the response characteristics of the two types of machines.
22 Currently, most intercoolers rely on the engine cooling water as a heat sink, because
23 this approach minimizes the number of components required. The relatively high
24 temperature of this water (about 90 °C) limits the benefits available, however. For this
25 reason, an increasing number of HD diesel engines are being equipped with low-temperature
26 charge air-cooling systems.
27 The most common type of low-temperature charge air-cooler rejects heat directly to the
28 atmosphere through an air-to-air heat exchanger mounted on the truck chassis in front of the
29 radiator. Although bulky and expensive, these charge air coolers are able to achieve the
30 lowest charge air temperatures—in many cases, only 10 or 15 °C above ambient.
31 An alternative approach is low-temperature air to water intercooling, which has been pursued
December 1994 2-19 DRAFT-DO NOT QUOTE OR CITE
-------
1 by Cummins Engine in the United States. Cummins has chosen to retain the basic water-air
2 intercooler but with drastically reduced radiator flow rates to reduce the water temperature
3 coming from the radiator. This water is then passed through the intercooler before it is used
4 for cooling the rest of the engine.
5
6 2.4.4 Intake Manifold Tuning
7 Tuned intake manifolds have been used for many years to enhance airflow rates on
8 high-performance gasoline engines and are being considered for some HD diesel engines.
9 A tuned manifold provides improved airflow and volumetric efficiency at speeds near its
10 resonant frequency, at the cost of reduced volumetric efficiency at other speeds. At least one
11 medium-HD manufacturer is considering a variable resonance manifold to improve airflow
12 characteristics at both low and high speeds.
13
14 2.4.5 Lubricating Oil Control
15 A significant fraction of diesel paniculate matter consists of oil-derived HCs and related
16 solid matter; estimates range from 10 to 50%. Reduced oil consumption has been a design
17 goal of HD diesel engine manufacturers for some time, and the current generation of diesel
18 engines already uses fairly little oil compared with their predecessors. Further reductions in
19 oil consumption are possible through careful attention to cylinder bore roundness and surface
20 finish, optimization of piston ring tension and shape, and attention to valve stem seals,
21 turbocharger oil seals, and other possible sources of oil loss. However, current technology
22 requires some oil consumption in the cylinder for lubricating and corrosion protective
23 functions.
24 Advances in piston/cylinder tribology could potentially eliminate or greatly reduce oil
25 consumption in the cylinder. Areas such as boundary lubrication and development of low-
26 friction ceramic coatings are currently the subjects of much research. The potential for
27 transforming this research into durable and reliable engines on the road remains to be
28 demonstrated, however.
29
30
December 1994 2-20 DRAFT-DO NOT QUOTE OR CITE
-------
1 2.4.6 Aftertreatment Systems
2 A trap oxidizer system consists of a durable particulate filter (the "trap") positioned in
3 the engine exhaust stream, along with some means for cleaning the filter by burning off
4 ("oxidizing") the collected particulate matter. The construction of a filter capable of
5 collecting diesel soot and other particulate matter from the exhaust stream is a
6 straightforward task, and several effective trapping media have been developed and
7 demonstrated. The most challenging problem of trap oxidizer system development has been
8 the process of "regenerating" the filter by burning off the accumulated particulate matter.
9 Diesel particulate matter consists primarily of a mixture of solid carbon coated with
10 HCs. The ignition temperature of this mixture is about 500 to 600 °C, which is above the
11 normal range of diesel engine exhaust temperatures. Thus, special means are needed to
12 ensure regeneration. Once ignited, however, this material burns to produce very high
13 temperatures, which can easily melt or crack the particulate filter. Initiating and controlling
14 the regeneration process to ensure reliable regeneration without damage to the trap is the
15 central engineering problem of trap oxidizer development.
16 Progress in-cylinder particle control has greatly reduced engine-out particle levels.
17 This progress has been most effective in reducing the solid soot fraction of the particulate
18 matter so that the SOF of the particulate matter now accounts for a much larger share than it
19 has previously. Depending on the engine and operating conditions, the SOF may account for
20 from 30% to more than 70% of the engine-out particulate matter.
21 Like a catalytic trap, a diesel catalytic converter oxidizes a large part of the HC
22 constituents of the SOF, as well as gaseous HC, CO, odor, and mutagenic constituents of
23 HC. Unlike a catalytic trap, however, a flow-through catalytic converter does not collect
24 any of the solid particulate matter, which simply passes through in the exhaust. This
25 eliminates the need for a regeneration system with its attendant technical difficulties and
26 costs. The particle-control efficiency of the catalytic converter is, of course, much less than
27 that of a trap. However, a particle-control efficiency of even 25 to 35% is enough to bring
28 many current development engines within the target range for the 1994 U.S. emissions
29 standard.
30 Diesel catalytic converters have a number of advantages. First, in addition to reducing
31 particulate emissions, the oxidation catalyst greatly reduces HC, CO, and odor emissions.
December 1994 2-21 DRAFT-DO NOT QUOTE OR CITE
-------
1 The catalyst is also very efficient in reducing emissions of gaseous and particle bound toxic
2 air contaminants. Although a precious-metal-catalyzed particle trap would have the same
3 advantages, the catalytic converter is much less complex, bulky, and expensive. Unlike the
4 trap, the catalytic converter has little impact on fuel economy or safety, and it will probably
5 not require replacement as often. Also, unlike the trap oxidizer, the catalytic converter is a
6 relatively mature technology—millions of catalytic converters are in use on gasoline vehicles
7 and diesel catalytic converters have been used in underground mining applications for more
8 than 20 years.
9 The disadvantage of the catalytic converter is the same as with the precious metal
10 catalyzed particle trap—sulfate emissions. The tendency of the precious metal catalyst to
11 convert SO2 to particulate sulfates requires the use of low-sulfur fuel; otherwise, the increase
12 in sulfate emissions would more than counterbalance the decrease in SOF.
13
14 2.4.7 Fuel Modifications
15 Modifications to diesel fuel composition have drawn considerable attention as a quick
16 and cost-effective means of reducing emissions from existing vehicles. The two
17 modifications that show the most promise are a reduction in sulfur content and in the fraction
18 of aromatic HCs in the fuel.
19 In addition to a direct reduction in emissions of SO2 and sulfate particles, reducing the
20 sulphur content of diesel fuel reduces the indirect formation of sulfate particles from SO2 in
21 the atmosphere. In Los Angeles, it is estimated that each pound of SQ2 emitted results in
22 roughly 1 Ib of fine particulate matter in the atmosphere. In this case, therefore, the indirect
23 PM emissions derived from SO2 from diesel vehicles are roughly as great as their direct
24 particle emissions. Sulfur dioxide conversion to particulate matter is highly dependent on
25 local meteorological conditions, however; thus, the effects could be greater or less.
26 A reduction in the aromatic HC content of diesel fuel may also help to reduce
27 emissions, especially where fuel aromatic levels are high. For existing diesel engines,
28 a reduction in aromatics from 35 to 20% by volume would be expected to reduce transient
29 particle emissions by 10 to 15% and NOX emissions by 5 to 10%. The HC emissions, and
30 possibly the mutagenic activity of the particulate SOF, would also be reduced. Modeling
31 studies of the refining industry have shown that aromatic reductions of this magnitude can
December 1994 2-22 DRAFT-DO NOT QUOTE OR CITE
-------
1 often be obtained through alterations in diesel fuel production and blending strategy, without
2 a necessitating major new investments in additional processing capacity.
3 Reduced aromatic content in diesel fuel would have other environmental and economic
4 benefits. The reduced aromatic content would improve the fuel's ignition quality, improving
5 cold starting and idling performance and reducing engine noise. The reduction in the use of
6 catalytically cracked blending stocks should also have a beneficial effect on deposit-forming
7 tendencies in the fuel injectors, reducing maintenance costs. On the negative side, however,
8 the reduced aromatics might result in some impairment of cold flow properties, because of
9 the increased paraffin content of the fuel.
10
11 2.4.7.1 Fuel Additives
12 A number of well-controlled studies have demonstrated the ability of detergent additives
13 in diesel fuel to prevent and remove injector tip deposits, thus reducing smoke levels. The
14 reduced smoke probably results in reduced PM emissions as well, but this has not been
15 demonstrated as clearly, because of the great expense of PM emissions tests on in-use
16 vehicles. Cetane-improving additives are also likely to result in some reduction in HC and
17 PM emissions in marginal fuels.
18
19
December 1994 2-23 DRAFT-DO NOT QUOTE OR CITE
-------
1 REFERENCES
2
3 Alkidas, A. C. (1988) Effects of injector-tip configuration on the performance and emissions of an uncooled
4 diesel. Presented at: International fuels and lubricants meeting and exposition; October; Portland, OR.
5 Warrendale, PA: Society of Automotive Engineers, Inc.; SAE technical paper no. 881613.
6
7 Ball, D. J. (1987) Paniculate carbon emissions and diesel vehicles. In: Vehicle emissions and their impact on
8 European air quality: proceedings of the Institution of Mechanical Engineers international conference;
9 November; London, United Kingdom. London, United Kingdom: Mechanical Engineering Publications
10 Limited; pp. 83-88.
11
12 Balzotti, A.; Coraetti, G. M.; Pidello, F.; Signer, M.; Scorsone, V. (1990) Italian city buses with particulates
13 traps. Presented at: International congress and exposition; February-March; Detroit, MI. Warrendale,
14 PA: Society of Automotive Engineers, Inc.; SAE technical paper no. 900114.
15
16 Black, F.; High, L. (1979) Methodology for determining paniculate and gaseous diesel hydrocarbon emissions.
17 Warrendale, PA: Society of Automotive Engineers, Inc.; SAE technical paper no. 790442.
18
19 Black, F.; Braddock, J.; Bradow, R. L.; Ingalls, M. (1985) Highway motor vehicles as sources of atmospheric
20 particles. Environ. Int. 11: 1-32.
21
22 Bradow, R. L. (1980) Diesel particle emissions. Bull. N. Y. Acad. Med. 56: 797-811.
23
24 Bradow, R. L. (1982) Diesel particle and organic emissions: engine simulation, sampling, and artifacts.
25 In: Lewtas, J., ed. lexicological effects of emissions from diesel engines: proceedings of the
26 Environmental Protection Agency 1981 diesel emissions symposium; October 1981; Raleigh, NC.
27 New York, NY: Elsevier Biomedical; pp. 33-47. (Developments in toxicology and environmental science:
28 v. 10).
29
30 Broukhiyan, E. M. H.; Lestz, S. S. (1981) Ethanol fumigation of a light duty diesel engine. In: SP-503 alternate
31 fuels for diesel engines. Warrendale, PA: Society of Automotive Engineers; pp. 15-25; SAE technical
32 paper no. 811209.
33
34 Clark, D. H.; Johnson, D. M.; Swedberg, S. E. (1984) Modem engine oils. Warrendale, PA: Society of
35 Automotive Engineers, Inc.; SAE technical paper no. 841207; SAE SP-603.
36
37 Dietzmann, H. E.; Parness, M. A.; Bradow, R. L. (1980) Emissions from trucks by chassis version of 1983
38 transient procedure. Presented at: Society of Automotive Engineers fuels and lubricants meeting; October;
39 Baltimore, MD. Warrendale, PA: Society of Automotive Engineers, Inc.; paper no. 801371.
40
41 Dietzmann, H. E.; Parness, M. A.; Bradow, R. L. (1981) Emissions from gasoline and diesel delivery trucks by
42 chassis transient cycle. Presented at: ASME energy-sources technology conference and exhibition;
43 January; Houston, TX. New York, NY: American Society of Mechanical Engineers; ASME paper no.
44 81-DGP-6.
45
46 Fenimore, C. P.; Jones, G. W. (1967) Oxidation of soot by hydroxyl radicals. J. Phys. Chem. 71: 593-597.
47
48 Gibbs, R. E.; Hyde, J. D.; Byer, S. M. (1980) Characterization of paniculate emissions from in-use diesel
49 vehicles. Presented at: Fuels and lubricants; October; Baltimore, MD. Warrendale, PA: Society of
50 Automotive Engineers, Inc.; SAE technical paper no. 801372.
51
52
December 1994 2-24 DRAFT-DO NOT QUOTE OR CITE
-------
1 Gibson, T. L.; Ricci, A. I.; Williams, R. L. (1981) Measurement of polynuclear aromatic hydrocarbons, their
2 derivatives, and their reactivity in diesel automobile exhaust. In: Cooke, M.; Dennis, A. J., eds.
3 Polynuclear aromatic hydrocarbons: chemical analysis and biological fate: proceedings of the
4 5th international symposium; Columbus, OH. Columbus, OH: Battelle Press; pp. 707-717.
5
6 Gill, A. P. (1988) Design choices for 1990's low emission diesel engines. Presented at: International congress
7 and exposition; February-March; Detroit, MI. Warrendale, PA: Society of Automotive Engineers, Inc.;
8 SAE technical paper no. 880350.
9
10 Gorse, R. A., Jr.; Salmeen, I. T.; Clark, C. R. (1982) Effects of filter loading and filter type on the
11 mutagenicity and composition of diesel exhaust particulate extracts. Atmos. Environ. 16: 1523-1528.
12
13 Hampton, C. V.; Pierson, W. R.; Schuetzle, D.; Harvey, T. M. (1983) Hydrocarbon gases emitted from
14 vehicles on the road. 2. Determination of emission rates from diesel and spark-ignition vehicles. Environ.
15 Sci. Technol. 17: 699-708.
16
17 Hardenberg, H. O.; Daudel, H. L.; Erdmannsdorfer, H. J. (1987) Experiences in the development of ceramic
18 fiber coil particulate traps. In: Diesel particulates: an update, SP-702. Warrendale, PA: Society of
19 Automotive Engineers; SAE technical paper no. 870015.
20
21 Hare, C. T.; Bradow, R. L. (1979) Characterization of heavy-duty gaseous and particulate emissions and effects
22 of fuel composition. Warrendale, PA: Society of Automotive Engineers, Inc.; SAE technical paper
23 no. 790490.
24
25 Hare, C. T.; Springer, K. J.; Bradow, R. L. (1976) Fuel and additive effects on diesel particulate development
26 and demonstration of methodology. Presented at: Automotive engineering congress and exposition;
27 February; Detroit, MI. Warrendale, PA: Society of Automotive Engineers, Inc.; SAE technical paper no.
28 760130.
29
30 Henein, N. A.; Bolt, J. A. (1967) Ignition lag in diesel engines. Warrendale, PA: Society of Automotive
31 Engineers, Inc.; SAE technical paper no. 670007.
32
33 Herr, J. D.; Yergey, J. A.; Dukovich, M.; Tejada, S.; Risby, T. H.; Lestz, S. S. (1983) The role of nitrogen in
34 the observed microbial mutagenic activity for diesel engine combustion. Warrendale, PA: Society for
35 Automotive Engineers; SAE technical paper no. 820467.
36
37 Huisingh, J.; Bradow, R.; Jungers, R.; Claxton, L.; Zweidinger, R.; Tejada, S.; Bumgarner, J.; Duffield, F.;
38 Waters, M.; Simmon, V. F.; Hare, C.; Rodriguez, C.; Snow, L. (1978) Application of bioassay to the
39 characterization of diesel particle emissions. In: Waters, M. D.; Nesnow, S.; Huisingh, J. L.; Sandhu,
40 S. S.; Claxton, L., eds. Application of short-term bioassays in the fractionation and analysis of complex
41 environmental mixtures: [proceedings of a symposium; February; Williamsburg, VA]. New York, NY:
42 Plenum Press; pp. 383-418. (Hollaender, A.; Probstein, F.; Welch, B. L., eds. Environmental science
43 research: v. 15).
44
45 International Agency for Research on Cancer. (1989) Diesel and gasoline engine exhausts and some nitroarenes.
46 Lyon, France: World Health Organization. (IARC monographs on the evaluation of carcinogenic risks to
47 humans: v. 46).
48
49 Kadota, T.; Henein, N. A. (1981) Time-resolved soot particulates in diesel spray combustion. In: Siegla, D. C.;
50 Smith, G. W., eds. Particulate carbon: formation during combustion [proceedings of an international
51 symposium; October 1981; Warren, MI]. New York, NY: Plenum Press; pp. 391-421.
J £
December 1994 2-25 DRAFT-DO NOT QUOTE OR CITE
-------
1 Kagami, M.; Akasaka, Y.; Date, K.; Maeda, T. (1984) The influence of fuel properties on the performance of
2 Japanese automotive diesels. Warrendale, PA: Society of Automotive Engineers, Inc.; SAE technical
3 paper no. 841082; SAE SP-581.
4
5 Kahn, I. M. (1970) Formation and combustion of carbon in a diesel engine. In: Diesel engine combustion
6 symposium; April. London, United Kingdom: Institute of Mechanical Engineers.
7
8 Lee, F. S.-C.; Pierson, W. R.; Ezike, J. (1980) The problem of PAH degradation during filter collection of
9 airborne particulates—an evaluation of several commonly used filter media. In: Bj0rseth, A.;
10 Dennis, A. J., eds. Polynuclear aromatic hydrocarbons: chemistry and biological effects: proceedings of
11 the 4th international symposium; Columbus, OH. Columbus, OH: Battelle Press; pp. 543-563.
12
13 Liberti, A.; Ciccioli, P.; Cecinato, A.; Brancaleoni, E.; Di Palo, C. (1984) Determination of
14 • nitrated-polyaromatic hydrocarbons (nitro-PAHs) in environmental samples by high resolution
15 chromatographic techniques. J. High Resolut. Chromatogr. Chromatogr. Commun. 7: 389-397.
16
17 Lipkea, W. H.; DeJoode, A. D. (1987) A model of a direct injection diesel combustion system for use in cycle
18 simulation and optimization studies. Presented at: International congress and exposition; February;
19 Detroit, MI. Warrendale, PA: Society of Automotive Engineers, Inc.; SAE technical paper no. 870573.
20
21 Lipkea, W. H.; DeJoode, A. D.; Christenson, S. R. (1987) The relationship between nitric oxide and work as
22 influenced by engine operating conditions and combustion system parameters for a direct injection diesel
23 engine. Presented at: International congress and exposition; February; Detroit, MI. Warrendale, PA:
24 Society of Automotive Engineers, Inc.; SAE technical paper no. 870269.
25
26 Lyn, W. T. (1960-61) Calculation of the effect of rate of heat release on the shape of cylinder pressure diagram
27 and cycle efficiency. Proc. Automob. Div. Inst. Mech. Eng. (1).
28
29 Lyn, W. T. (1963) Study of burning rate and nature of combustion in a diesel engine. In: ninth international
30 symposium on combustion. New York, NY: Academic Press, Inc.
31
32 Lyn, W. T.; Valdmanis, E. (1966-67) The effect of physical factors on ignition delay. Proc. Inst. Mech. Eng.
33 Part 2A (1).
34
35 McCann, J.; Ames, B. N. (1977) The Salmonella/Taicrosome mutagenicity test: predictive value for animal
36 carcinogenicity. In: Hiatt, H. H.; Watson, J. D.; Winsten, J. A., eds. Origins of human cancer: book C,
37 human risk assessment. New York, NY: Cold Spring Harbor Laboratory; pp. 1431-1450. (Cold Spring
38 Harbor conferences on cell proliferation: v. 4).
39
40 McGeehan, J. A.; Gilmore, J. T.; Thompson, R. M. (1988a) How sulfated ash in oils cause diesel exhaust valve
41 failure. Presented at: International fuels and lubricants meeting and exposition; October; Portland, OR.
42 Warrendale, PA: Society of Automotive Engineers, Inc.; SAE technical paper no. 881584.
43
44 McGeehan, J. A.; McNary, J. C.; Kahn, M. J. (1988b) Performance of 1.0% and 1.45% ash—SAE 15W-40 oils
45 in on-highway trucks with Cummins, Caterpillar, and Mack engines. Presented at: International congress
46 and exposition; February-March; Detroit, MI. Warrendale, PA: Society of Automotive Engineers, Inc.;
47 SAE technical paper no. 880260.
48
49 Myers, P. S.; Uyehara, O. A. (1947) Flame temperature measurements—electronic solution of temperature
50 equations. SAE Q. Trans. 1: 592-611.
51
December 1994 2-26 DRAFT-DO NOT QUOTE OR CITE
-------
1 Newton, D. L.; Erickson, M. D.; Tomer, K. B.; Pellazzari, E. D.; Gentry, P.; Zweidinger, R. B. (1982)
2 Identification of nitroaromatics in diesel exhaust paniculate using gas chromatography/negative ion
3 chemical ionization mass spectrometry and other techniques. Environ. Sci. Technol. 16: 206-213.
4
5 Nishioka, M. G.; Petersen, B.; Lewtas, J. (1983) Comparison of nitro-aromatic content and direct-acting
6 mutagenicity of passenger car engine emissions. In: Rondia, D.; Cooke, M.; Haroz, R. K., eds. Mobile
7 source emissions including polycyclic organic species. Dordrecht, The Netherlands: D. Reidel; pp.
8 197-210. (NATO advanced science institutes series C, mathematical and physical sciences: v. 112).
9
10 Obert, E. F. (1973) Internal combustion engines and air pollution. 3rd. ed. New York, NY: Harper and Row.
11 (Obert, E. F., ed. Intext series in mechanical engineering).
12
13 Pederson, T. C.; Siak, J.-S. (1981) The role of nitroaromatic compounds in the direct-acting mutagenicity of
14 diesel particle extracts. J. Appl. Toxicol. 1: 54-60.
15
16 Plee, S. L.; Ahmad, T. (1983) Relative roles of premixed and diffusion burning in diesel combustion. Presented
17 at: Fuels and lubricants meeting; October-November; San Francisco, CA. Warrendale, PA: Society of
18 Automotive Engineers, Inc.; SAE technical paper no. 831733.
19
20 Plee, S. L.; Ahmad, T.; Myers, J. P.; Siegla, D. C. (1981) Effects of flame temperature and air-fuel mixing on
21 emission of paniculate carbon from a divided-chamber diesel engine. In: Siegla, D. C.; Smith, G. W.,
22 eds. Paniculate carbon: formation in combustion. New York, NY: Plenum Press; pp. 423-487.
23
24 Risby, T. H.; Lestz, S. S. (1983) Is the direct mutagenic activity of diesel paniculate matter a sampling artifact?
25 Environ. Sci. Technol. 17: 621-624.
26
27 Ross, M. M.; Risby, T. H.; Steele, W. A.; Lestz, S. S.; Yashin, R. E. (1982) Physicochemical properties of
28 diesel paniculate matter. Colloids Surf. 5: 17-31.
29
30 Schuetzle, D.; Frazier, J. A. (1986) Factors influencing the emission of vapor and paniculate phase components
31 from diesel engines. In: Ishinishi, N.; Koizumi, A.; McClellan, R. O.; Stober, W., eds. Carcinogenic
32 and mutagenic effects of diesel engine exhaust: proceedings of the international satellite symposium on
33 lexicological effects of emissions from diesel engines; July; Tsukuba Science City, Japan. Amsterdam,
34 The Netherlands: Elsevier Science Publishers B. V.; pp. 41-63. (Developments in toxicology and
35 environmental science: v. 13).
36
37 Schuetzle, D.; Jensen, T. E. (1985) Analysis of nitrated polycyclic aromatic hydrocarbons by mass spectrometry.
38 In: White, C. M., ed. Nitrated polycyclic aromatic hydrocarbons. Heidelberg, Federal Republic of
39 Germany: A. Huthing Verlag; pp. 121-167.
40
41 Schuetzle, D.; Perez, J. M. (1983) Factors influencing the emissions of nitrated-polynuclear aromatic
42 hydrocarbons (nitro-PAH) from diesel engines. J. Air Pollut. Control Assoc. 33: 751-755.
43
44 Schuetzle, D.; Lee, F. S.-C.; Prater, T. J.; Tejada, S. B. (1981) The identification of polynuclear aromatic
45 hydrocarbon (PAH) derivatives in mutagenic fractions of diesel paniculate extracts. Int. J. Environ. Anal.
46 Chem. 9: 93-144.
47
48 Schulz, B. (1988) The United States engine markets in perspective. Diesel Prog. N. Am. 53(6): 8, 10, 12.
49
50 Shahed, S. M. (1985) The role of fuel-air mixing in diesel combustion and emissions. ASME international
51 conference on flows in internal combustion engines—III, v. 28, pp. 159-164.
52
53 Siegla, D. C.; Smith, G. W., eds. (1980) Paniculate carbon: formation during combustion [proceedings of an
54 international symposium; October 1980; Warren, MI]. New York, NY: Plenum Press.
December 1994 2-27 DRAFT-DO NOT QUOTE OR CITE
-------
1 Springer, K. J. (1988) Diesel lube oils—4th dimension of diesel paniculate control. In: Springer, K. J., ed.
2 Engine emission technology for the 1990's [proceedings of the internal combustion engine division
3 technical conference; October; San Antonio, TX]. New York, NY: American Society of Mechanical
4 Engineers; pp. 45-51.
5
6 Springer, K. J. (1989) Low emission diesel fuel for 1991—1994. In: Chrisman, B.; Serve, J. V., eds. Advances
7 in engine emissions control technology ICE, v. 5. American Society of Mechanical Engineers.
8
9 Tosaka, S.; Fujiwara, Y.; Murayama, T. (1989) The effect of fuel properties on diesel engine exhaust paniculate
10 formation. Presented at: International congress and exposition; February-March; Detroit, MI.
11 Warrendale, PA: Society of Automotive Engineers, Inc.; SAE technical paper no. 890421.
12
13 U.S. Environmental Protection Agency. (1985) Compilation of air pollutant emission factors, volume I: stationary
14 point and area sources; volume II: mobile sources. 4th ed. Research Triangle Park, NC: Office of Air
15 Quality Planning and Standards; EPA report no. AP-42-ED-4-VOL-1 and AP-42-ED-4-VOL-2. Available
16 from: NTIS, Springfield, VA; PB86-124906 and PB87-205266.
17
18 U.S. Environmental Protection Agency. (1986) Program to calculate size specific paniculate emissions for mobile
19 sources. Ann Arbor, MI: Emission Control Technology Division; EPA report no. EPA/SW/MT-86/011.
20 Available from: NTIS, Springfield, VA; PB86-179702/XAB.
21
22 U.S. Environmental Protection Agency. (1989) User's guide to MOBILE4 (Mobile Source Emissions Factor
23 Model). Ann Arbor, MI: Office of Mobile Sources; EPA report no. EPA-AA-TEB-89-01. Available
24 from: NTIS, Springfield, VA; PB89-164271.
25
26 Ullman, T. L. (1989) Investigation of the effects of fuel composition on heavy-duty diesel engine emissions.
27 Presented at: International fuels and lubricants meeting and exposition; September; Baltimore, MD.
28 Warrendale, PA: Society of Automotive Engineers, Inc.; SAE technical paper no. 892072.
29
30 Uyehara, O. A. (1987) Factors that affect BSFC and emissions for diesel engines: part 1—presentation of
31 concepts. Presented at: International congress and exposition; February; Detroit, MI. Warrendale, PA:
32 Society of Automotive Engineers, Inc.; SAE technical paper no. 870343.
33
34 Wade, W. R.; Jones, C. M. (1984) Current and future light duty diesel engines and their fuels. Presented at:
35 International congress and exposition; February-March; Detroit, MI. Warrendale, PA: Society of
36 Automotive Engineers, Inc.; SAE technical paper no. 840105.
37
38 Warner-Selph, M. A.; Dietzmann, H. E. (1984) Characterization of heavy-duty motor vehicle emissions under
39 transient driving conditions. Research Triangle Park, NC: U.S. Environmental Protection Agency;
40 contract no. 68-02-3722.
41
42 White, C. M. (1985) Analysis of nitrated polycyclic aromatic hydrocarbons by gas chromatography. In: White,
43 C. M., ed. Nitrated polycyclic aromatic hydrocarbons. Heidelberg, Federal Republic of Germany.
44 A. Huthing Verlag; pp. 1-86.
45
46 Wu, K.-J.; Peterson, R. C. (1986) Correlation of nitric oxide emission from a diesel engine with measured
47 temperature and burning rate. Presented at: International fuels and lubricants meeting and exposition;
48 October; Philadelphia, PA. Warrendale, PA: Society of Automotive Engineers, Inc.; SAE technical paper
49 no. 861566.
50
51 Yost, D. M.; Ryan, T. W., Ill; Owens, E. C. (1985) Fuel effects on combustion in a two-stroke diesel engine.
52 Presented at: International fuels and lubricants meeting and exposition; October; Tulsa, OK. Warrendale,
53 PA: Society of Automotive Engineers, Inc.; SAE technical paper no. 852104.
54
December 1994 2-28 DRAFT-DO NOT QUOTE OR CITE
-------
1 Yu, R. C.; Shahed, S. M. (1981) Effects of injection timing and exhaust gas recirculation on emissions from a
2 D.I. diesel engine. Presented at: Fuels and lubricants meeting; October; tulsa, OK. Warrendale, PA:
3 Society of Automotive Engineers, Inc.; SAE technical paper no. 811234.
December 1994 2-29 DRAFT-DO NOT QUOTE OR CITE
-------
i 3. DIESEL-DERIVED POLLUTANTS: ATMOSPHERIC
2 CONCENTRATIONS, TRANSPORT, AND
3 TRANSFORMATIONS
4
5
6 3.1 INTRODUCTION
7 Combustion of fuel in the diesel engine results in the formation of a complex mixture of
8 gaseous and paniculate exhaust. Because of concerns over possible health effects associated
9 with diesel paniculate emissions, measurements have been made to characterize chemically in
10 detail the exhausts from light-duty diesel (LDD) and, to a lesser extent, heavy-duty diesel
11 (HDD) engines. Most of these measurements are of primary pollutants, that is, gases and
12 paniculate matter emitted directly into the air from their sources.
13 Once emitted, however, the primary pollutants are subject to dispersion and transport
14 and, at the same time, to chemical and physical transformations into secondary pollutants,
15 which may be more harmful than their precursors. The time scales of these atmospheric
16 transformations and physical loss processes vary widely; atmospheric lifetimes range from
17 < 1 min for some highly reactive organic compounds to months or even years for other much
18 more inert constituents of direct emissions. Thus, to assess the impact of diesel emissions on
19 human health and welfare, it is necessary to determine the chemical and physical changes
20 that primary diesel emissions undergo during their transport through the atmosphere.
21
22
23 3.2 PRIMARY DIESEL EMISSIONS
24 Detailed chemical characterization of diesel engine emissions was performed mostly in
25 the late 1970s and early 1980s. Since that time substantial changes have occurred in engine
26 and emission control technologies, as well as improvements in chemical analysis
27 methodology. It is likely that the emissions from the currently manufactured diesel vehicles
28 may not be the same as those measured and reported earlier. When possible, the latest data
29 were used; however, the data presented in this chapter should not be considered to be fully
30 representative of either current emissions from the wide range of diesel engines currently
31 used or of those that may have occurred in the past.
December 1994 3_1 DRAFT-DO NOT QUOTE OR CITE
-------
1 3.2.1 Gaseous Emissions
2 Diesel passenger car and light-truck emissions of CO and total gaseous hydrocarbons
3 (THC) are considerably lower than those of gasoline vehicles, whereas NOX emission rates
4 are similar. For HDD vehicles, the CO and THC emission rates are somewhat lower than,
5 but comparable with, those of gasoline vehicles, but NOX emissions are currently many times
6 that of average traffic. In addition to these regulated pollutants, diesel exhausts also contain
7 some sulfur dioxide because of the presence of sulfur in the diesel fuel (0.2 wt % on the
8 average). Following combustion, approximately 98% of the sulfur is emitted as SO2 and
9 2% as paniculate sulfate (Pierson et al., 1978, 1979; Truex et al., 1980). Most of the
10 sulfate is in the form of sulfuric acid (H2SO4) (Truex et al., 1980).
11 The atmospheric concentration of nitric acid (HNO3) from LDD exhausts has been
12 reported to be negligible in comparison with that from other anthropogenic sources (Okamoto
13 et al., 1983; Harris et al., 1987). A range of concentrations from «100 ppbv
14 (= 250 fJLg/m3) to = 800 ppbv («2 mg/m3) (Harris et al., 1987) and an emission rate of
15 =1.3 mg/km were reported (Okamoto et al., 1983).
16 A small amount of ammonia was also detected in diesel engine exhausts (Pierson and
17 Brachaczek, 1983b). The highest value for NH3 (25 mg/km) was reported for HDDs;
18 =4 mg/km was reported for LDD, and «10 mg/km and =5 mg/km for gasoline-powered
19 vehicles, with and without catalyst, respectively (Pierson and Brachaczek, 1983b). The
20 emission rates of total aliphatic amines were reported to be below the detection limit of
21 0.04 to 0.3 mg/km for gasoline-powered vehicles and 0.08 to 0.7 mg/km for heavy-duty
22 trucks. It was concluded that motor vehicles are an insignificant source of atmospheric
23 NH3 and that amines emitted from motor vehicles cannot give rise to carcinogenic
24 nitrosoamines in the amount said to exist in ambient samples (Pierson and Brachaczek,
25 1983b).
26 In addition, low concentrations of phenols have been reported in HDD and LDD
27 emissions (Hare and Baines, 1979; Hare and Bradow, 1979). Aliphatic carboxylic acids
28 (mainly formic, acetic, propionic, and benzoic acids) were also reported in vehicle exhausts
29 (Kawamura et al., 1985); however, no data exist concerning diesel engine emissions of these
30 acids.
December 1994 3-2 DRAFT-DO NOT QUOTE OR CITE
-------
1 Table 3-1 shows the comparison of the emission rates of some representative alkanes,
2 alkenes, aromatic hydrocarbons, and aldehydes from HDD and LDD engines, and gasoline
3 engines with and without a catalytic converter. Data on catalyst-equipped gasoline vehicles
4 were obtained with a chassis dynamometer and are averaged from 46 in-use passenger cars,
5 1975 to 1982 models, selected to be representative of vehicles actually driven by the U.S.
6 public (Sigsby et al., 1987). Table 3-1 shows the data from the Federal Test Procedure
7 (FTP) only (which attempts to simulate a typical urban driving pattern with average speed of
8 =20 mph), although two other driving cycles (Crowded Urban Expressway and the New
9 York City cycle) are reported in the original publication.
10 The emission data on noncatalyst gasoline vehicles shown in Table 3-1 are averaged
11 from 25 in-use passenger cars, representing late 1980s vehicles in intensive use in the United
12 Kingdom (U.K.) (Bailey et al., 1990). The vehicles were driven "as received" and fueled by
13 leaded premium-grade gasoline, obtained locally from a single source. They were driven on
14 five routes chosen to cover the normal range of U.K. driving speeds and conditions, and the
15 exhaust samples were taken using a miniaturized constant-volume exhaust gas sampler.
16 To allow direct comparison, the urban roadway data (=13.5 mph average speed) are given
17 in Table 3-1. However, it has to be pointed out that hydrocarbon emission rates are highly
18 dependent on driving speeds; in general, THC emission rate expressed in grams per traveled
19 distance decreases as the driving speed increases, but the individual hydrocarbons display
20 various patterns, which relate to their origin. The gasoline components (hydrocarbons with
21 carbon number C > 4) are present in highest proportion at low speeds, whereas at higher
22 speeds these components are more efficiently used and the proportion of combustion-derived
23 products increases.
24 The data on LDD emissions are from the National Research Council Report (1982) and
25 from studies (Smith, 1989; Smith and Paskind, 1989) concerning the evaluation of particle
26 trap efficiencies for two diesel passenger cars, (i.e., a 1986 Mercedes Benz and a
27 Volkswagen prototype Jetta). However, little has been published recently; most of the
28 available data are from the late 1970s and early 1980s.
29 Quantitative data on emissions from heavy-duty vehicles are relatively sparse and are
30 generally expressed in terms of grams per unit work performed by HDD engines (see
31 Section 2.3.1). Most of the data on HDD vehicle emission rates expressed in terms of grams
December 1994 3.3 DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 3-1. EMISSION RATES OF VOLATILE ORGANIC
COMPOUNDS (VOC) FROM DIESEL AND GASOLINE ENGINES
VOC [g/mi(g/km)]
Diesel
VOC
THC
Methane
Ethylene
Acetylene
Propylene
fl-Pentane
wo-Pentane
n-Decane
n-Dodecane
Benzene
Toluene
Xylenes
Ethyl benzene
Naphthalene
Formaldehyde
Acetaldehyde
Acrolein
Benzaldehyde
Total aldehyde
HDD
3.65 (2.28)d
NA
NA
NA
NA
NA
NA
0.01 (0.007)e
0.027
(0.017)e
0.024
(0.015)d
0.01 (0.007)e
0.006
(0.004)d
0.005
(0.003)e
0.01 (0.007)e
NA
NA
0.053
(0.033)d
NA
NA
LDDa
0.23(0.14)f
0.01 (0.008)
0.04 (0.03)
NA
0.01 (0.008)
NA
NA
NA
NA
0.02 (0.015)8
0.006 (0.004)f
0.002 (0.001)f
0.001 (0.0006)f
0.003 (0.002)f
0.02 (0.01)
0.007 (0.004/
0.01 (0.006)
NA
0.03(0.02)f
Gasoline
Catalystc
5.4 (3.4)
0.27(0.17)
0.3 (0.2)
0.26(0.16)
0.15(0.09)
0.09 (0.06)
0.27(0.17)
0.003
0.003
0.31 (0.19)
0.7 (0.45)
0.96 (0.6)
0.21 (0.13)
NA
0.06 (0.04)8
NA
NA
NA
NA
Noncatalystb
1.8(1.2)
0.26(0.16)
0.14(0.09)
0.04 (0.02)
0.04 (0.03)
0.03 (0.02)
0.07 (0.04)
(0.0016)h
(0.002)'
0.06 (0.04)
0.1 (0.07)
0.08 (0.05)
0.02 (0.01)
NA
0.025 (0.015)
0.01 (0.007)
0.002 (0.001)
0.003 (0.002)
0.04 (0.03)
HDD = Heavy-duty diesel.
LDD = Light-duty diesel.
NA = Data not available.
aFrom National Research Council (1982), except as indicated.
bFrom Bailey et al. (1990), except as indicated.
cFrom Sigsby et al. (1987).
dFrom Westerholm et al. (1991).
eFrom Hampton et al. (1983), data from Allegheny Mountain Tunnel.
fFrom Smith (1989), and Smith and Paskind (1989); four-cycle FTP test, 1986 Mercedes Benz.
SFrom Schuetzle and Frazier (1986).
hFrom Hampton et al. (1983), no differentiation between vehicles with and without catalyst.
1 per traveled distance were obtained from tunnel field experiments, in particular from the
2 Allegheny and Tuscarora Mountain Tunnel experiments (Hampton et al., 1982, 1983).
3 Unfortunately, because of the sampling method selected, no quantitative data on
4 hydrocarbons with carbon number C < 8 could be obtained.
December 1994
3-4
DRAFT-DO NOT QUOTE OR CITE
-------
1 Recently, exhaust emissions from a HDD truck (Scania 143H, equipped with a turbo-
2 charged Scania DSC 1403 diesel engine) was characterized chemically and tested for
3 mutagenicity during transient driving conditions on a chassis dynamometer (Westerholm
4 et al., 1991). Table 3-1 gives selected data from this study.
5 As can be seen from Table 3-1, the emissions of gaseous organic compounds from
6 diesel engines and spark-ignition engines are qualitatively similar (e.g., similar chemical
7 components are present in both exhausts), although there are significant quantitative
8 differences. In theory, new spark-ignition vehicles equipped with catalytic converters emit
9 almost no reactive hydrocarbons in their exhausts. However, catalyst deterioration over the
10 lifetime of the vehicle and the evaporative and refueling emissions will result in an increase
11 in the amount of reactive material released.
12 Table 3-1 lists the emission rates for exhaust pipe emissions only. Currently, fuel
13 evaporation (e.g., from fuel lines and carburetors) accounts for 30 to 60% of the total hydro-
14 carbon emissions from passenger gasoline vehicles with and without catalytic converter
15 (International Agency for Research on Cancer, 1989). Under ambient conditions, the vapor
16 pressure of most diesel fuels in current use is so low that emissions resulting from
17 evaporation of diesel fuels are not significant. However, for both diesel and gasoline engine
18 emissions, methane, ethane, ethylene, acetylene, propane, and propylene originate strictly
19 from tailpipe emissions, as evidenced by the fact that the lowest molecular weight
20 components of gasoline are normally hydrocarbons with carbon number C4.
21 Differences between vehicles of the same category in the quantity of emitted material
22 are very significant, and such differences arise from many factors of engine design, fuel
23 control, engine conditions, and the general condition of the vehicle at the time of test. The
24 differences between test parameters (e.g., speed, cold or hot start, fuel composition,
25 dynamometer or road measurements, etc.) make the comparison of the data given in
26 Table 3-1 more uncertain. However, it is clear from these data, that the emission profile of
27 gaseous organic compounds is different for diesel and spark-ignition vehicles; the aromatic
28 hydrocarbons and low molecular weight alkanes (< C9) are more characteristic of spark-
29 ignition vehicle emissions, whereas the heavier alkanes (>C10) are more characteristic of
30 diesel emissions (Hampton et al., 1983; Carey and Cohen, 1980).
31
December 1994 3.5 DRAFT-DO NOT QUOTE OR CITE
-------
1 3.2.2 Participate Emissions
2 3.2.2.1 Diesel Participate Matter
3 Diesel exhaust particles are aggregates of spherical primary particles, and 75 to 95%
4 of the paniculate mass is in the accumulation mode centered about an aerodynamic diameter
5 of 0.2 /im. Figure 3-1 shows a typical size distribution of diesel exhaust particles. The size
6 distribution is important, because transport of the particles in the atmosphere and deposition
7 in the human respiratory tract depend essentially on aerodynamic diameter (see Chapter 4 for
8 discussion of deposition).
9 Diesel paniculate matter is generally defined as any material that is collected on a
10 filtering medium at a temperature of 52 °C or less after dilution of the raw exhaust.
11 In general, diesel engines produce more paniculate emissions than do gasoline engines with
12 or without catalytic converters. The main constituent of diesel particles is carbon, which
13 accounts for =80% of total particle mass. Approximately 70% of this so-called total carbon
14 (TC) occurs in the form of soot or elemental carbon (EC); the rest is in the form of organic
15 compounds and is called organic carbon (OC). Table 3-2 compares the emission rates of
16 paniculate matter and its distribution between TC and OC for HDD, LDD, and gasoline
17 engines. Data on HDD, LDD, and gasoline engines without catalytic converters are mostly
18 from the recently published survey of 13 HDD vehicles, 19 LDD vehicles, and 22 spark-
19 ignition vehicles currently in intensive use in Sydney, Australia (Williams et al., 1989a,b).
20 The vehicles were tested on a dynamometer using several test procedures, but Table 3-2 lists
21 the data from the ADR 37 cycle (for spark-ignition and LDD vehicles), which is essentially
22 identical to the 1979 U.S. Federal Test Procedure (FTP) cycle, and from the modified ADR
23 36 cycle (for HDD vehicles; multimode steady-state procedure). In addition, recent data
24 from a heavy-duty truck (Scania 143H), tested on a chassis dynamometer using transient
25 driving conditions (Westerholm et al., 1991) are also given in Table 3-2.
26 Because the dynamometer studies are not fully representative of the road conditions,
27 the data obtained from the field experiments in two highway tunnels, the Allegheny and
28 Tuscarora Mountain Tunnels of Pennsylvania Turnpike (Pierson and Brachaczek, 1983a;
29 Szkarlat and Japar, 1983), are also given in Table 3-2 for comparison.
30
December 1994 3-6 DRAFT-DO NOT QUOTE OR CITE
-------
O
O
1
o
o
a
i
n
s
333
S
•n -n >n
333
333
e
e
B> 3.
< to
£P
I: IB
« <_ (j
• o, »»
^ T3 c«
5«-|
n
s.
s*
W
W
on a
erho
et
et
00
•~- -~- K
H- H- p
VO
S
\o
00
8
H
00
o _
^g *•
^| /~,
00 -Q O
O\
o
<*
O
§
O
2
Ov
-J
3
>
-z
>
O
b
o
n
& C"1
W
VY-
D G
I
9
OO
to
a
•s
4?
•o
(a
N'
«
I
Mass Concentration Distribution
(dm/d In Dp [usual units mg m"3])
1 I I
I
-------
1 3.2.2.2 Participate Phase Inorganics
2 Both organic and elemental carbon, account for = 80% of the total participate matter
3 mass. The remaining 20% is composed of sulfate (mainly H2SO4) (Pierson and Brachaczek,
4 1983a) and some inorganic additives and adventitious components of fuel and motor oil.
5 Table 3-3 gives the average compositions of inorganic constituents of airborne paniculate
6 matter associated with vehicles on the road (from Pierson and Brachaczek, 1983a). All
7 airborne constituents of paniculate matter associated with vehicle traffic (other than
8 atmospheric transformation products of primary emissions) are included, whether emitted
9 from the exhaust or not (e.g., originated from tire wear debris and soil dust).
10
11 3.2.2.3 Particulate Phase Organic Compounds
12 Carbonaceous, diesel-emitted particles have high specific surface areas of 30 to
13 50 m2/g (Frey and Corn, 1967). Because of this high surface area, diesel particles are
14 capable of adsorbing relatively large quantities of organic material originating from unburned
15 fuel and lubricating oil and from pyrosynthesis occurring during combustion of fuel (see
16 Chapter 2). After removal of extractable organic material, the surface area of diesel particles
17 increases up to 90 m2/g (Pierson and Brachaczek, 1976).
18 The extractable fraction of diesel particles is typically in the range of 20 to 30%, but
19 it may be as high as 90% (Williams et al., 1989b), depending on vehicle type and operating
20 conditions. In general, if a diesel engine is running under low load, the incomplete
21 combustion results in a relatively low particle concentration and a higher proportion of
22 organic-associated particles (Dutcher et al., 1984). In addition, recent progress in in-cylinder
23 paniculate matter control has been most effective in reducing the elemental carbon fraction of
24 the particulate matter, so that the organic carbon fraction now accounts for much larger share
25 than it had previously (see Section 2.4, Chapter 2).
26 The extractable portion of total carbon, although commonly used as a measure of
27 organic compound content, is not totally equivalent to the OC fraction, as measured by the
28 thermal-optical carbon analysis technique (Japar et al., 1984). The average ratio of OC to
29 extractable mass was shown to be 0.70 ± 0.05, when toluene/propanol-1 mixture was used
30 as an extraction solvent, and this ratio was probably the result of the presence of both
31 oxygenated organic compounds and inorganic sulfates in the extracted mass.
December 1994 3-8 DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 3-3. SUMMARY OF COMPOSITION AND EMISSION RATES
(in milligrams per kilometer) OF AIRBORNE PARTICIPATE MATTER
FROM ON-ROAD VEHICLES, TUSCARORA MOUNTAIN TUNNEL 1977
EXPERIMENT (FROM PERSON AND BRACHACZEK, 1983a)
Constituent
Gasoline"
(mg/km [% of total mass])
Dieser
(mg/km [% of total mass])
H
B
C
N
Nac
Mgc
Alc
Sid
P6
S[S04-2]
Clf
Kd
Cac
Tid
Mn°-f
Fec
Cu
Zne
Brf
Baf
Pbf
5 ±4(10 ±6)
0.04 ± 0.6(0.07 ±0.11)
34 ± 21 (67 ± 42)
1.1 ± 0.8(2 ± 2)
0.09 ± 0.37 (0.2 ± 0.7)
0.7 ± 0.3 (1.3 ± 0.6)
0.2 ± 0.5 (0.3 ± 0.9)
0.5 ± 0.7(1.0 ± 1.3)
0.07 ± 0.06(0.13 ±0.11)
0.4[3.4 ± 0.9] (0.9[7 ± 3])
0.8 ± 0.4(1.6 ± 0.8)
0.17 ± 0.08(0.3 ± 0.2)
1.3 ± 0.3(2.5 ± 0.7)
0.006 ± 0.01 (0.01 ± 0.02)
0.08 ± 0.01 (0.16 ± 0.025)
0.32 ± 0.32 (0.6 ± 0.6)
0.04 ± 0.02 (0.07 ± 0.03)
0.04 ± 0.04 (0.08 ± 0.08)
5.75 ± 0.45(11.2 ±0.9)
0.03 ± 0.01 (0.07 ± 0.02)
12.4 ± 1.6 (24)
47 ± 11 (5 ± 1)
1.14 ± 0.16(0.13 ± 0.02)
725 ± 117(84 ± 14)
16 ± 2(1.9 ± 0.3)
6.6 ± 1.0(0.8 ± 0.1)
8 ± 1 (0.9 ± 0.15)
8.5 ± 1 (1.0 ± 0.2)
14 ±2(1.6 ±0.2)
1.3 ±0.2(0.15 ±0.02)
23[42 ± 5] (2.7[4.9 ± 0.9])
0(0)
1.5 ± 0.2(0.17 ± 0.03)
5.8 ± 1.4(0.7 ± 0.2)
0.12 ±0.03(0.014 ± 0.004)
0.34 ± 0.04 (0.04 ± 0.004)
5.0 ±0.9(0.6 ± 0.1)
0.22 ± 0.09 (0.025 ± 0.01)
1.4 ± 0.1 (0.16 ± 0.1)
0(0)
0.66 ± 0.03 (0.08 ± 0.033)
11.5 ± 3 (1.3 ±0.3)
"Mostly passenger cars, no distinction between catalytic and noncatalytic vehicles.
bMostly heavy-duty diesel trucks, average weight = 30 ton.
cPartially attributable to soil dust.
dWholly attributable to soil dust.
eAttributable to motor oil.
fAttributable to fuel additives.
1 Extraction and Fractionation Techniques
2 A variety of solvents and extraction techniques have been used in the past for the
3 separation of organic compounds from diesel particles (Levsen, 1988). Although the reports
4 on the extraction efficiencies are in part contradictory, it appears that Soxhlet extraction and
5 the binary solvent system composed of aromatic solvent and alcohol gave the best recovery
6 of PAHs, as determined by 14C-B[a]P (benzo[0]pyrene) spiking experiments (Schuetzle and
7 Perez, 1981). Direct chemical analysis of the entire extractable fraction of diesel particulate
December 1994
3-9
DRAFT-DO NOT QUOTE OR CITE
-------
1 matter is not generally possible because a large number of compounds of different polarity
2 are present. The separation of diesel paniculate organic matter (POM) into various fractions
3 according to chemical functionalities is a necessary preliminary step to chemical identification
4 of individual compounds. Open-column liquid chromatography (LC) and liquid-liquid
5 separation procedures have been the most widely used fractionation methods (Lee and
6 Schuetzle, 1983). Open-column LC is very often followed by normal-phase high-
7 performance liquid chromatography (HPLC), if the identification of less abundant
8 components is required.
9
10 Chemical Composition
11 Table 3-4 lists the general classes of extractable organic compounds identified in POM
12 from combustion emissions, including diesel emissions.
13
14
TABLE 3-4. CLASSES OF ORGANIC COMPOUNDS IDENTIFIED IN
PARTICULATE-PHASE COMBUSTION EMISSIONS
Hydrocarbons
Derivatives3 of hydrocarbons
Polycyclic aromatic hydrocarbons (PAH)
Derivatives2 of PAH
Multifunctional derivatives3 of PAH
Heterocyclic compounds
Derivatives3 of heterocyclics
Multifunctional derivatives3 of heterocyclic compounds
"Derivatives include acids, alcohols, aldehydes, esters, ketones, nitrates, and sulfonates.
Source: Adapted from Schuetzle (1988).
1 Liquid chromatography methods usually divide the complex environmental mixtures of
2 organic compounds into nonpolar, moderately polar, and polar fractions. This separation is
3 achieved by using specific solvents (or solvent mixtures) for the elution of compounds from
4 chromatographic columns. Schuetzle and coworkers (1985) proposed that standard chemical
5 compounds be selected for establishing reference points for the fractionation of diesel POM
December 1994 3-10 DRAFT-DO NOT QUOTE OR CITE
-------
1 into nonpolar, moderately polar, and polar fractions by normal phase HPLC. They proposed
2 that the elution of 1-nitronaphthalene would define the end of the elution of nonpolar
3 compounds and the beginning of the moderately polar fractions. In a similar way,
4 1,6-pyrene quinone would define the end of elution of moderately polar compounds and the
5 beginning of the polar region.
6 For diesel engine emissions, =57% of the extracted organic mass is contained in the
7 nonpolar fraction (Schuetzle, 1983). About 90% of this fraction consists of aliphatic
8 hydrocarbons from approximately C14 to about C40, with a carbon number maximum at C22
9 to C26 (Black and High, 1979; Pierson et el., 1983). Polycyclic aromatic hydrocarbons and
10 alkyl-substituted PAH account for the remainder of the nonpolar mass.
11 The moderately polar fraction ( = 9% w/w of extract) consists mainly of oxygenated
12 PAH species and nitrated PAH. The polar fraction ( = 32% w/w of extract) is composed
13 mainly of carboxylic and dicarboxylic acids of PAH, hydroxy-PAH, hydroxynitro-PAH,
14 nitrated N-containing heterocyclic compounds, etc. (Schuetzle, 1983; Schuetzle et al., 1985).
15 Limited recovery studies have shown that there is little degradation or loss of diesel
16 POM on the HPLC column. Greater than 90% of the mass and 70 to 100% of the Ames
17 S. typhimurium-active material injected onto the column have been recovered (Schuetzle
18 etal., 1985).
19
20 Polycyclic Aromatic Hydrocarbon
21 Particle-bound PAH and their derivatives (mainly nitrated PAH) attracted considerable
22 attention relatively early because of their mutagenic and, in some cases, carcinogenic
23 properties (see, for example, National Research Council, 1982). The most widely used
24 methods of PAH analysis included thin layer chromatography (TLC), capillary gas
25 chromatography (GC), gas chromatography/mass spectrometry (GC/MS), and HPLC with
26 UV or fluorescence detection (Levsen, 1988). Table 3-5 lists the PAH and thioarenes
27 identified and quantified by GC/MS in three LDD paniculate matter extracts (Tong and
28 Karasek, 1984). Data listed in this table reveal the presence of a large number of alkyl
29 derivatives of PAH, which are sometimes more abundant than the parent PAH.
30 Table 3-6 compares the emission rates of several representative PAHs from HDD,
31 LDD and gasoline (with and without catalytic converter) engines.
December 1994 3_H DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 3-5. POLYCYCLIC AROMATIC HYDROCARBONS IDENTIFIED
AND QUANTIFIED IN EXTRACTS OF DIESEL PARTICLES
(FROM TONG AND KARASEK, 1984)
Compound"-5
Acenaphthylene
Trimethylnaphthalene
Fluorene
Dimethylbiphenyl
C4-Naphthalene
Trimethylbiphenyl
Dibenzothiophene
Phenanthrene
Anthracene
Methyldibenzothiophene
Methylphenanthrene
Methylanthracene
Ethylphenanthrene
4H-Cyclopenta[
-------
TABLE 3-5 (cont'd). POLYCYCLIC AROMATIC HYDROCARBONS IDENTIFIED
AND QUANTIFIED IN EXTRACTS OF DIESEL PARTICLES
(FROM TONG AND KARASEK, 1984)
Compound3'6
Benzo[£]fluoranthene
Benzo[jt]fluoranthene
Benzo[e]pyrene
Benzo[a]pyrene
Benzo[a/z] anthracene
Indenofl ,2,3-[of]pyrene
Benzo[g/z/]perylene
Dibenzopyrene
Molecular Weight
252
252
252
252
278
276
276 j
302
Concentration
(ng/mg extract)Cid
421-1,090
91-289
487-946
208-558
50-96
30-93
443-1,050
136-254
"Compounds are arranged according to increasing GC retention times.
blsomeric alkyl derivatives are not listed separately.
cConcentration range as found in the paniculate extracts of three VW passenger cars.
dSoluble organic fractions accounted for 11.1, 12.1, and 14.7% of total paniculate matter (w/w) for these three
diesel samples.
1 Nitrated Polycyclic Aromatic Hydrocarbon
2 Nitro-PAH (nitroarenes) have been shown to be present in diesel paniculate extracts,
3 though in much lower concentration than the parent PAH (Schuetzle et al., 1981, 1982;
4 Paputa-Peck et al., 1983). Because many nitroarenes are potent direct-acting (e.g., without
5 metabolic activation) mutagens in the Ames assay using S. typhimurium strains (Rosenkranz
6 and Mermelstein, 1983), the analysis of nitro-PAH in diesel POM attracted considerable
7 attention in the early 1980s.
8 Numerous nitro-PAHs were identified in LDD paniculate extracts using capillary
9 GC with thermionic nitrogen-phosphorus detector (NPD) (Paputa-Peck et al., 1983).
10 Positive isomer identification for 16 nitro-PAH has been made utilizing the GC retention
11 times of authentic standards and low- and high-resolution mass spectra as identification
12 criteria. These include 1-nitropyrene; 2-methyl-l-nitronaphthalene; 4-nitrobiphenyl; 2-nitro-
13 fluorene; 9-nitroanthracene; 9-methyl-10-nitroanthracene; 2-nitroanthracene;
14 2-nitrophenanthrene; l-methyl-9-nitroanthracene; l-methyl-3-nitropyrene; 1-methyl-
15 6-nitropyrene; l-methyl-8-nitropyrene; 1,3-, 1,6-and 1,8-dinitropyrene; and 6-nitrobenzo
December 1994 3_13 DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 3-6. EMISSION RATES OF PARTICLE-BOUND POLYCYCLIC AROMATIC
HYDROCARBONS (PAHs) (in milligrams per kilometer) FROM HEAVY- (HDD)
AND LIGHT-DUTY (LDD) DIESEL AND GASOLINE ENGINES
Gasoline Cars
PAH
Pyrene
Fluoranthene
Benzo[a]pyrene
Benzo[y]pyrene
HDD
17.6 (ll)a
27.2 (17)a
<0.1 (0.06)a
2.3 (1.4)e
0.24 (.15)a
LDD
66 (42)b
50 (31)b
1 (0.6)d
NDb
3(1.9)d
Noncatalyst
45 (28)c
32 (20)c
3.2 (2)c
4.8 (3)c
Catalyst
7 (4.4)d
5 (3.1)d
0.4 (0.25)d
0.4 (0.25)d
aFrom Westerholm et al. (1991).
bFrom Smith (1989), four-cycle FTP test, 1986 Mercedes Benz.
cFrom Alsberg et al. (1985).
dFrom Schuetzle and Frazier (1986).
cFrom Dietzman et al. (1980), averaged value for four different engines.
ND = None detected.
1 [a]pyrene. In addition, two nitrated heterocyclic compounds were identified, namely 5- and
2 8-nitroquinoline. Forty-five additional nitro-PAHs were tentatively identified in this diesel
3 paniculate extract (Paputa-Peck et al., 1983).
4 The concentration of nitro-PAH adsorbed on diesel particles varies substantially from
5 sample to sample. Usually 1-nitropyrene is the predominant component, and concentrations
6 ranging from =7 to =165 /ng/g of particles are reported (Levsen, 1988).
7 Table 3-7 gives the approximate concentrations of several more abundant nitro-PAH in
8 LDD particulate extracts.
9
10 Oxygenated Poly cyclic Aromatic Hydrocarbons
11 The moderately polar fraction of diesel paniculate extract contains a variety of
12 oxy-PAHs (in particular aldehydes, ketones, quinones, and acid anhydrides) in much higher
13 amounts than nitro-PAH, which elute in the same fraction. Oxy-PAHs are nonmutagenic or
14 very weakly mutagenic, which explains the relatively low interest in this group of
15 compounds. The most detailed study of oxy-PAHs was published by Schuetzle et al. (1981),
16 who identified more than 100 compounds. A large number of oxy-PAH in the molecular
17 weight range of 182 to 272 were also identified by Tong et al. (1984). The main
December 1994 3-14 DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 3-7. CONCENTRATIONS OF NITRO-POLYCYCLIC AROMATIC
HYDROCARBONS (PAHs) IDENTIFIED IN A LIGHT-DUTY DIESEL (LDD)
PARTICIPATE EXTRACT
Concentration
Nitro-PAH3 Q*g/g of particles)
4-Nitrobiphenyl 2.2
2-Nitrofluorene 1.8
2-Nitroanthracene 4.4
9-Nitroanthracene 1.2
9-Nitrophenanthrene 1.0
3-Nitrophenanthrene 4.1
2-Methyl-l-nitroanthracene 8.3
1 -Nitrofluoranthene 1.8
7-Nitrofluoranthene 0.7
3-Nitrofluoranthene 4.4
8-Nitrofluoranthene 0.8
1-Nitropyrene 18.9; 75b
6-Nitrobenzo[a]pyrene 2.5
1,3-Dinitropyreneb 0.30
l,6-Dinitropyreneb 0.40
l,8-Dinitropyreneb 0.53
2,7-Dinitrofluorenec 4.2; 6.0
2,7-Dinitro-9-fluorenonec 8.6; 3.0
aFrom Campbell and Lee (1984) unless noted otherwise. Concentrations recalculated from /tg/g of extract to
Uglg of particles using a value of 44% for extractable material (w/w).
bFrom Paputa-Peck et al. (1983).
cFrom Schuetzle (1983).
1 components identified by Tong and co-workers in paniculate matter extracts of three LDD
2 cars (VWs) were 9-fluorenone, anthraquinone, 4H-cyclopenta[fife/]phenanthrene-4-one,
3 9-phenanthrene aldehyde, benzo[de]anthracene-7-one, and benzo[o/lpyrene-6-one. These
4 components are present at concentrations of 30 to 300 //g/g particles. Some of these
5 oxy-PAHs are formed during sampling (Levsen, 1988).
6
7 Polar Polycylic Aromatic Hydrocarbon Derivatives
8 According to Schuetzle et al. (1985), although 65 to 75% of the directly acting
9 mutagenicity (as tested by Ames S. typhimurium assay) for LDD paniculate extracts is
December 1994 3_15 DRAFT-DO NOT QUOTE OR CITE
-------
1 associated with the fraction of moderate polarity, more than 65% of the mutagenic activity
2 for HDD paniculate extract is concentrated in the most polar fraction. However, because of
3 the serious analytical difficulties, only preliminary data exist on the identification of
4 compounds that are responsible for the mutagenic activity of this fraction (so-called "polar
5 mutagens"). Schuetzle and co-workers (1985) employed the concept of "bioassay directed
6 chemical analysis" (see Section 3.5) for the isolation and identification of polar PAH
7 derivatives from the extracts of HDD paniculate matter (NIST standard reference material,
8 SRM 1650). Several hydroxynitro-PAHs, hydroxy-PAHs, and nitrated heterocyclic
9 compounds were tentatively identified in the polar fraction. It has to be noted, however, that
10 NIST SRM 1650 was not intended to be representative of HDD engines, but was a material
11 made available to investigators for the purpose of methods development.
12 In another study (Bayona et al., 1988), the polar HPLC fractions of the same NIST
13 SRM 1650 were analyzed by fused silica capillary GC with low- and high-resolution mass
14 MS, using electron impact (El) and negative ion chemical ionization (NICI) techniques.
15 In addition, direct-probe El and NICI-MS analyses were performed. Over 80 polycyclic
16 aromatic compounds (PAC) belonging to several different chemical classes (anhydrides,
17 carboxaldehydes, diazaarenes, cyclic imides, hydroxynitro-PAH, nitroaza-PAC,
18 nitrolactones, and quinones) were tentatively identified. Ten of them were positively
19 identified by comparison of retention times with authentic standards. Among them,
20 phenazine and phthalic anhydride were positively identified for the first time in diesel exhaust
21 particles. In addition, cyclic imides and their alkylated derivatives were tentatively
22 identified.
23
24 3.2.3 Gaseous/Particulate Phase Emission Partitioning of Polycyclic
25 Aromatic Hydrocarbons
26 The distribution of the emissions between the gaseous and paniculate phases is
27 determined by the vapor pressure of the individual species, by the amount and type of the
28 paniculate matter present (adsorption surface available), and by the temperature (Ligocki and
29 Pankow, 1989). Table 3-8 gives the vapor pressures at 25 °C of some representative PAH
30 ranging from naphthalene to benzo[a]pyrene.
December 1994 3-16 DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 3-8. VAPOR PRESSURES AT 25 °C FOR A SERIES
OF POLYCYCLIC AROMATIC HYDROCARBONS (PAHs)a
PAH
Naphthalene
Acenaphthylene
Acenaphthene
Fluorene
Phenanthrene
Anthracene
Fluoranthrene
Pyrene
Benzo[a]anthracene
Benzo[a]pyrene
Chryseneb
Vapor Pressure
at 298 K (torr)
8.0 x ID'2
6.7 x 10'3
2.2 x 10'3
6.0 X 10^
1.2 x lO"4
6.0 x 10-6
9.2 x lO"6
4.5 x ID"6
2.1 x 10'7
5.6 x 10-9
6.4 x 10'9
aSonnefeld et al. (1983), except as indicated.
bYamasaki et al. (1984).
1 The factor of «107 hi the range of vapor pressures is reflected in the fact that,
2 at equilibrium at ambient temperature, naphthalene exists almost entirely in the gas phase,
3 whereas B[a]P, other five-ring PAHs, and higher-ring PAHs are predominantly adsorbed on
4 particles. The intermediate three- and four-ring PAHs are distributed between the two
5 phases.
6 However, the vapor pressures of these intermediate PAHs can be significantly reduced
7 by their adsorption on various types of surfaces. Because of this phenomenon, the amount
8 and type of paniculate matter present plays an important role, together with temperature, in
9 the vapor-particle partitioning of semivolatile organic compounds (SOC).
10 The measurements of gas/particulate phase distribution are often accomplished by
11 using a high-volume filter followed by an adsorbent such as polyurethane foam (PUF),
12 Tenax, or XAD-2 (Cautreels and Van Cauwenberghe, 1978; Thrane and Mikalsen, 1981;
13 Yamasaki et al., 1982). However, the pressure drop behind a high-volume filter or cascade
14 impactor contributes to volatilization of the three- to five-ring PAHs, to a degree reflecting
15 their vapor pressures. The magnitude of this "blow-off" artifact depends on a number of
16 factors, including sampling temperature and the volume of air sampled (Van Vaeck et al.,
December 1994 3_17 DRAFT-DO NOT QUOTE OR CITE
-------
1 1984; Coutant et al., 1988). Despite these problems from volatilization, measurements with
2 the high-volume filters followed by a solid adsorbent have provided most estimates of vapor-
3 particle partitioning of SOC in ambient ah", as well as insights into the factors influencing
4 SOC adsorption onto aerosols.
5 Average distributions of PAH between high-volume filter and PUF plugs (positioned
6 downstream of the filter) in samples collected in a heavily traveled roadway tunnel
7 (Baltimore Harbor Tunnel) are shown in Figure 3-2. As discussed in the preceding text, the
8 "blow-off" from the filter precludes detailed quantitative interpretation. However, it can be
9 seen from this figure that significant fractions of phenanthrene, anthracene, and their
10 alkylated derivatives, along with fluoranthene and pyrene, exist in the gas phase. No PAHs
11 less volatile than pyrene were observed hi any of the PUF samples. Comparison of the
12 observed vapor-to-particle PAH ratios and those calculated based on the relationship derived
13 by Yamasaki and co-workers (1982) generally agreed within a factor of two (Benner et al.,
14 1989).
15
16
17 3.3 ATMOSPHERIC TRANSFORMATIONS OF PRIMARY
18 DIESEL EMISSIONS
19 3.3.1 Long-Range Transport and Fate of Primary Diesel Emissions
20 Once released into the atmosphere, primary diesel emissions (or any other direct
21 emissions) are subject to dispersion and transport and, at the same tune, to various physical
22 and chemical processes that determine their ultimate environmental fate. The role of the
23 atmosphere may be compared in some way with that of a giant chemical reactor hi which
24 materials of varying reactivity are mixed together, subjected to chemical and/or physical
25 processes, and finally removed (Schroeder and Lane, 1988). The main features of the
26 atmospheric cycle for primary diesel emissions, beginning with emission and ending with
27 deposition to the earth's surface, are shown on Figure 3-3.
28 Initial mixing describes the physical processes that act on pollutants immediately after
29 their release from an emission source. The dilution of diesel exhaust under roadway
30 conditions is an important factor to consider; whereas a dilution factor of ten is typical of
31 many dilution tunnels used in dynamometer studies of automobile exhaust, a dilution factor
December 1994 3-18 DRAFT-DO NOT QUOTE OR CITE
-------
Particle
Vapor Phase
Phenanthrene
Anthracene
3-Methylphenanthrene
2-Methylanthracene
1 -Methylphenanthrene
C2-phenanthrene(1)
C2-phenanthrene(2)
C2-phenanthrene(3)
C2-phenanthrene(4)
C2-phenanthrene(5)
C2-phenanthrene(6)
Fluoranthrene
Pyrene
0%
25% 50% 75%
Percent in Paniculate Phase
100%
Figure 3-2. Vapor/particle phase polycyclic aromatic hydrocarbon distribution in
samples collected in Baltimore Harbor Tunnel.
Source: Benner et al. (1989).
1 of «103 is more realistic under roadway conditions. This discrepancy leads to slightly
2 different particle size distributions under real driving conditions than those predicted from
3 laboratory data (Kittelson and Dolan, 1980); for example, because of slower coagulation
4 processes, more particles in the Aitken nuclei range (^0.08-^im diameter) may be expected
5 under typical roadway conditions.
6 Diffusion and transport processes occur simultaneously in the atmosphere and account
7 for the dispersion of emissions. The actual distance travelled by gaseous- and particulate-
8 phase pollutants depends on the amount of time a specific pollutant resides in the atmosphere
9 and is available for dispersion (Schroeder and Lane, 1988). As was discussed in Section 3.2,
10 primary diesel emissions are a very complex mixture containing thousands of organic and
11 inorganic constituents in the gas and paniculate phases. These compounds have different
December 1994
3-19
DRAFT-DO NOT QUOTE OR CITE
-------
Chemical
Transformation
Initial
Mixing
Transport and
Diffusion
Figure 3-3. Diesel-derived pollutants: Emission-to-deposition atmospheric cycle.
Source: Schroeder and Lane (1988).
1 chemical reactivities and are removed by dry and wet deposition processes with different
2 rates, as discussed more thoroughly in Sections 3.3.2 and 3.3.3. As a result, more reactive
3 compounds with short lifetimes will be removed from the atmosphere relatively quickly,
4 whereas more stable pollutants can be transported over greater distances. Clearly, a
5 knowledge of the atmospheric loss processes and lifetimes for automotive emissions is
6 important, because these lifetimes determine the geographic extent of the influence of these
7 emissions.
8 Anthropogenic pollutants can travel through the atmosphere over long distances.
9 In particular, the long-range transport of SO2 and its transformation to SO4= have been
10 studied extensively (Galloway and Whelpdale, 1980; Lowenthal and Rahn, 1985). The
11 organic pollutants, particularly those adsorbed on carbonaceous particles, are also subjected
December 1994
3-20
DRAFT-DO NOT QUOTE OR CITE
-------
1 to long-range transport. As will be discussed in Section 3.3.2.2, organic compounds, such
2 as PAHs, adsorbed on diesel paniculate matter are generally more resistant to atmospheric
3 reactions than those in the gas phase. In addition, particles of smaller diameter (< 1 /mi),
4 such as diesel paniculate matter (Figure 3-1), are removed less efficiently by wet and dry
5 deposition (see Section 3.3.3), and thus have longer atmospheric residence times.
6 It has been reported (Laflamme and Hites, 1978; Kites et al., 1980) that PAH and their
7 alkyl homologs are distributed in sediments throughout the world and that the PAH patterns
8 are similar to each other and to air particulate matter for most of the locations studied.
9 Furthermore, the quantities of PAH increase with proximity to urban areas. This suggests
10 anthropogenic combustion sources and long-range atmospheric transport of PAH.
11 Conclusive evidence for long-range transport of PAH was also reported from the recent
12 measurements of PAH in Siskiwit Lake, located on a wilderness island in northern Lake
13 Superior (McVeety and Hites, 1988). Because of its remote location, any PAH found in this
14 lake must have originated exclusively from atmospheric transport.
15 Earlier studies by Bj0rseth and co-workers (Bj0rseth and Olufsen, 1983) showed that
16 PAHs are transported from Great Britain and the European continent to remote locations in
17 Norway and Sweden. The specific sources of PAH emissions could not be identified,
18 however. The authors speculated that combustion engines were not the major sources of
19 PAHs, because the amounts of B[a]P found in the samples collected in Norway and Sweden
20 were higher than could be accounted for from gasoline and diesel fuel consumption in Great
21 Britain and because the coronene/B[a]P ratios in the samples were lower than those usually
22 found in gasoline and diesel exhaust. However, because of the lack of PAH profiles
23 specifically for combustion engines (or, as a matter of fact, any other specific tracer), the
24 problem of relative contribution of gasoline and diesel vehicle exhausts to long-range
25 transport of organics cannot be solved currently. The need for such "organic profiles" is
26 addressed in Section 3.4.4.
27
28 3.3.2 Chemical Transformations
29 3.3.2.1 Gas-Phase Reactions
30 The following chemical processes contribute to the removal of gas-phase compounds
31 from the atmosphere (Atkinson, 1988):
December 1994 3-21 DRAFT-DO NOT QUOTE OR CITE
-------
1 • Photolysis during daylight hours,
2
3 • Reaction with hydroxyl (OH) radicals during daylight hours,
4
5 • Reaction with ozone (O3) during daytime and nighttime,
6
7 • Reaction with hydroperoxyl (HO2) radicals typically during late daytime and early
8 nighttime hours,
9
10 • Reaction with gaseous nitrate (NO3) radicals during nighttime hours,
11
12 • Reaction with dinitrogen pentoxide (N2O5) during nighttime hours,
13
14 • Reaction with NO2 during daytime and nighttime hours, and
15
16 • Reaction with gaseous nitric acid (HNO3) and other species such as nitrous (HNO2)
17 acid and sulfuric acid (H2SO4).
18
19 It has been shown (Atkinson et al., 1990) that the N2O5 reactions with PAHs proceed
20 by initial NO3 addition to form an NO3-PAH adduct, which either dissociates back to
21 reactants or reacts exclusively with NO2 to form nitroarenes and other products. Because
22 under atmospheric conditions, where N2O5, NO3 radicals, and NO2 are in equilibrium, these
23 reactions are kinetically equivalent to a reaction with N2O5 with an effective N2O5 reaction
24 rate constant, we will further refer to these reactions as N2O5 reactions.
25 The reactive gaseous species, such as OH radicals, NO3 radicals, HO2 radicals, and
26 ozone, are present in the atmosphere either during the daytime (OH radicals) or nighttime
27 (N2O5 and NO3 radicals) hours or both time periods (ozone, NO^. For the routes of
28 formations of these species and their concentrations in the troposphere, see Finlayson-Pitts
29 and Pitts, 1986.
30 Table 3-9 gives the calculated atmospheric lifetimes for some selected compounds
31 present in automotive gas-phase emissions as the result of known tropospheric chemical
32 removal reactions (Atkinson, 1988). These lifetimes (i.e., the tune for the compound to
33 decay to 1/e or 37% percent of its original concentration) are calculated from the
34 corresponding measured reaction rate constants and the average ambient concentration of the
35 tropospheric species involved.
36 Although the individual rate constants are known to a reasonable degree of accuracy
37 (in general, to within a factor of two), the tropospheric concentrations of these key reactive
December 1994 3-22 DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 3-9. CALCULATED ATMOSPHERIC LIFETIMES FOR GAS-PHASE
REACTIONS OF SELECTED COMPOUNDS PRESENT IN AUTOMOTIVE
EMISSIONS WITH ATMOSPHERICALLY IMPORTANT REACTIVE SPECIES
(FROM ATKINSON, 1988, UNLESS NOTED OTHERWISE)
Atmospheric Lifetime Resulting from Reaction with
Compound
NO2
NO
HN03
SO2
NH3
Propane
7i-Butane
n-Octane
Ethylene
Propylene
Acetylene
Formaldehyde
Acetaldehyde
Benzaldehyde
Acrolein
Formic acid
Benzene
Toluene
m-Xylene
Phenol
Naphthalene
2-Methylnaphthalene
2 , 3 -Dimethy Inaphthalene
Acenaphthene
Acenaphthylene
Phenanthrene
Anthracene
Fluoranthene8
Pyrene6
OHa
2 days
4 days
180 days
26 days
140 days
19 days
9 days
3 days
3 days
11 h
30 days
3 days
1 day
2 days
1 day
50 days
18 days
4 days
11 h
10 h
1 day
5 h
4h
2h
2h
9h
2h
6h
6h
03"
12 h
1 min
—
> 200 year
—
> 7, 000 year
> 4,500 year
—
9 days
1.5 days
6 year
>2 x 104year
>7 year
—
60 days
—
600 year
300 year
75 year
—
> 80 days
>40 days
> 40 days
> 30 days
=50 min
—
—
—
—
NO3C HO2d
Ih 2h
3 min 20 min
— —
>4 x 104year >600year
— —
— —
9 year —
3 year —
3 year —
15 days —
> 14 year —
210 days 23 days
50 days —
60 days —
— —
— —
> 16 year —
9 year —
2 year —
20 min —
_f _
_f _
_f _
= 3h —
13 min —
— —
—
_f _
_f _
hve
2 min
—
—
—
—
—
—
—
—
—
4h
60h
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
aFor 12-h average concentration of OH radical of 1 x 106 molecule/cm3.
bFor 24-h average 03 concentration of 7 x 1011 molecule/cm3.
°For 12-h average NO3 concentration of 2 x 108 molecule/cm3.
dFor 12-h average HO2 concentration of 108 molecule/cm3.
eFor solar zenith angle of 0°.
lifetimes due to gas-phase reactions with a 12-h average concentration of N205 of 2 x 1010 molecule/cm3 are:
naphthalene =80 days, 2-methyInaphthalene, =35 days; 2,3-dimethylnaphthalene, =20 days; fluoranthene,
= 64 days; and pyrene, =20 days.
lifetimes calculated from kinetic data given in Atkinson et al. (1990).
December 1994
3-23
DRAFT-DO NOT QUOTE OR CITE
-------
1 species are much more uncertain. For example, the ambient concentrations of OH radicals at
2 any given time and/or location are uncertain to a factor of at least 5, and more likely
3 10 (Atkinson, 1988). The tropospheric diurnally and annually averaged OH radical
4 concentrations are more certain, to possibly a factor of 2. For this reason, the calculated
5 lifetimes listed in Table 3-9 are approximate only and are valid for those reactive species
6 concentrations listed in the table footnotes. However, these data permit estimation of the
7 contribution of each of these atmospheric reaction to the overall rates of removal of most
8 pollutants from the atmosphere.
9 As can be seen from Table 3-9, the major atmospheric loss process for most of the
10 automotive emission constituents listed is by daytime reaction with OH radicals. For some
11 pollutants, photolysis, reactions with ozone, and reactions with NO3 radicals during nighttime
12 hours are also important removal routes.
13 The atmospheric lifetimes do not take into consideration the potential chemical or
14 biological importance of the products of these various reactions. For example, the reaction
15 of gas-phase PAHs with N2O5 appears to be of minor significance as a PAH loss process,
16 but, as will be discussed in subsequent sections, to be more important as a route of formation
17 of mutagenic nitro-PAH.
18
19 Reactions of Nitrogen Oxides
20 Only the major atmospheric reactions of NOX are considered here; for detailed
21 discussion of the chemistry of these important species, see Finlayson-Pitts and Pitts (1986).
22 Oxides of nitrogen emitted by diesel engines include mainly NO, with lesser amounts
23 of NO2. Nitric oxide is easily oxidized to NO2 in the reactions with HO2 radicals and
24 alkylperoxy radicals (the reaction of NO with O2 is too slow at typical ambient
25 concentrations of NO). In addition, NO reacts rapidly (Table 3-1) with ozone via
26 Reaction 1:
27
28 NO + O3 -* NO2 + 02 (1)
29
30 Nitrogen dioxide is photolysed rapidly at wavelengths of <430 nm:
31
December 1994 3-24 DRAFT-DO NOT QUOTE OR CITE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
33
34
35
NO2 + hv -» NO + 0(3P)
The oxygen atom produced in this reaction reacts with O2 forming ozone:
02+ 0(3P)-03
(2)
(3)
The photolysis of NO2 is the only known significant anthropogenic source of ozone in
the ambient air, and is produced via Reactions 2 and 3. With this series of reactions, NO,
NO2, and O3 are in a photostationary state:
IN
/
\J 1
V
1- W3 ^l>Wj T *_/2
f
k
1 — O(3P)-
hv
with
[OJ =
k2[N02]
k,[NO]
where kt and k2 are the rate constants for the Reactions 1 and 2, respectively, and brackets
signify concentrations. This photostationary state is strongly affected by NO to NO2
conversions caused by reactions involving organic compounds.
The important atmospheric reactions of NO2 also include formation of NO3 radicals
and N2O5:
N02 + O3 -» NO3 + O2
NO2 + NO3 y N2O5
with N2O5 being in equilibrium with NO2 and NO3 radicals.
The other important atmospheric reactions of NO and NO2 include nitrous and nitric
acid formation, respectively, by reaction with OH radicals:
December 1994
3-25
DRAFT-DO NOT QUOTE OR CITE
-------
1 NO + OH -* HNO2
2 NO2 + OH -» HNO3
3
4 Reactions of Sulfur Dioxide
5 Reaction with OH radical is the dominant SO2 atmospheric gas-phase reaction process
6 (Stockwell and Calvert, 1983):
7
8 SO2 + OH -» HSO3
9
10 followed by the formation of HO2 radicals and H2SO4:
11
12 HSO3 + O2 - HO2 + SO3
15 * H2°
n H2S°4
20 Because SO2 is soluble in water, it undergoes scavenging by fog, cloud water, and
21 raindrops. In aqueous systems, SO2 is readily oxidized to sulfate (Calvert and Stockwell,
22 1983).
23
24 Reactions of Alkanes
25 Only a brief overview of the most important atmospheric reactions of alkanes is
26 presented here; for detailed discussion consult Finlayson-Pitts and Pitts (1986) and Atkinson
27 (1988, 1990).
28 Under atmospheric conditions, alkanes react with OH radicals during the daytime and
29 with NO3 radicals during the nighttime:
30
31 RH + OH -» R' + H2O
32 RH + NO3 -» R* + HNO3
33
34 Alkyl radical R" reacts with O2 forming an alkylperoxy radical:
35
36
December 1994 3-26 DRAFT-DO NOT QUOTE OR CITE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
R'
which, under polluted urban atmospheric conditions, reacts predominantly with NO by two
pathways: (a) oxidation of NO to NO2 and formation of alkoxy radical (RO*), the only
significant path for the smaller (
-------
1 For the simplest alkoxy radicals, reaction with O2 (pathway a) is predominant. For
2 larger radicals, however, isomerization (c) and decomposition (b) may become significant,
3 the relative importance of these two pathways, depending on the structure of the radical.
4 In all three cases, free radicals are produced, which then will carry on the chain reactions.
5 The first-generation products include aldehydes, ketones, and alkyl nitrates, which can react
6 further under atmospheric conditions.
7
8 Reactions ofAlkene
9 Lower molecular weight alkenes, such as ethylene, propylene, and isomeric butenes
10 are present in exhaust from gasoline and (although in lower amount) diesel engines (see
11 Table 2-2). Gas-phase alkenes are removed from the troposphere by reaction with
12 OH radicals, NO3 radicals, and O3 (Finlayson-Pitts and Pitts, 1986; Atkinson, 1988;
13 Atkinson and Carter, 1984). Reactions with OH radicals are rapid (see Table 3-10) and
14 proceed by OH radical addition to the double bond. For example, for propylene:
15
OH
CH3CH=CH2 + OH - CH3CHCH2OH + CH^HCHJ
(a) ~65% (b) «35%
25
26 Formation of the more stable radical (a) is dominant.
27 Because the OH-olefin adduct is essentially an alkyl radical, it reacts further in a
28 manner similar to alkyl radicals formed from the reaction of alkanes with OH radicals (e.g.,
29 by the addition of O2 followed by reaction with NO). For example, for radical (a):
30
31 NO O
32
CH3CHCH2OH + O2
33
NO2
34
CH3CHCH2OH
35 6-hydroxyalkoxy radical is the major product. This radical, as discussed above for alkanes,
36 can (a) react with O2 or (b) decompose (isomerization is not important in the case of smaller
37 alkenes):
December 1994 3-28 DRAFT-DO NOT QUOTE OR CITE
-------
o
n>
o
n>
TABLE 3-10. SUMMARY OF THE NITROARENES PRODUCED FROM THE GAS-PHASE HYDROXYL (OH)
RADICAL-INITIATED AND DINITROGEN PENTOXIDE (N2O5) REACTIONS AND ELECTROPHDLIC
NITRATION OF POLYCYCLIC AROMATIC HYDROCARBONS (PAHs)
(V
^3
U)
K>
O
Si
6
o
o
H
O
1
W
O
n
H
w
PAH
Naphthalene
1 -Methylnaphthalene
2-Methylnaphthalene
Acenaphthylene
Acenaphthene
Biphenyl
Phenanthrene
Anthracene
Fluoranthene
Pyrene
Acephenanthrylene
"Concurrent NO3 radical
Position of Nitration
OH
1- (0.3%); 2- (0.3%)
5->4->6->3->7-=2-»8-
Total yield (=0.4%)
5->6- = 7-=4- = 8-»3->l-
Total yield (=0.2%)
4- (2%)
5->3->4-
Total yield (=0.2%)
3- (5%)
Two isomers (not 9-nitrophenanthrene)
Total yield (<0.1%)
l-;2-
Total yield (=0.2%)
2- (3%); 7- (-0.15%); 8- (=0.15%)
2- (=0.25%); 4- (=0.045%)
Two isomers (not 4- or 5-nitro-)
Total yield (=0.1%)
reaction will dominate over N2O5 reaction
(Yield) in Reaction with
N2O5 Position of Electrophilic Nitration
1- (17%); 2- (7%) l-»2-
3->5->4->8-=6->7->2- 4->2->5->8->7- = 3->6-
Total yield (=30%)
4->l- = 5->8- = 3- = 7- = 6- l-»8->4->6->5->3->7-
Total yield ( = 30%)
None observed 1-
4-(40%);3-(=2%);5-(=2%)a 3-; 5-
No reaction observed 2-; 4-
Four isomers (including 9->3-; 2-; 1-
9-nitrophenanthrene)
Total yield (<1%)
1-; 2- 9-
Totalyield(<2%)
2-( = 25-30%) 3->8->7->l-
4-; 2- 1-
Total yield (<1%)
None observed 4-; 5-
in ambient air.
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
2
25
22
26
27
28
29
30
31
32
33
34
35
O-
CHjCHCHjOH
a
o
II
CHjCCHjOH -i- HO2
CH3CHO + -CH2OH
The available data suggest that the decomposition (b) dominates over reaction with O2.
Thus, OH radical reactions with alkenes lead to the formation of aldehyde and/or ketones.
The NO3 radical reaction becomes important under atmospheric conditions only for
alkenes more reactive than propylene (Table 3-1). Similarly to the OH radical reaction, this
reaction proceeds through NO3 radical addition to the double bond, followed by reaction with
O2 (Finlayson-Pitts and Pitts, 1986; Atkinson, 1988). Carbonyl compounds are formed as
major products, but minor products (possibly dinitrates) are not well defined (Atkinson,
1988).
The reactions of alkenes with ozone compete with the daytime OH radical reaction
(Table 3-1). These reactions proceed by addition to the double bond, followed by rapid
decomposition of so-called "ozom'de" or "molozonide" into a carbonyl compound and an
energy-rich biradical:
R'
\
R
c = c
R,R2COO + R,R4C=O
R,R2C=O + R3R4COO
December 1994
3-30 DRAFT-DO NOT QUOTE OR CITE
-------
1 The importance of pathways (a) and (b) are assumed to be equal (Atkinson and Carter,
2 1984).
3 The major uncertainty concerns the atmospheric fate of the energy-rich intermediate,
4 the so-called "Criegee biradical". It is believed that some of the Criegee intermediates
5 contain sufficient excess energy to spontaneously decompose. The remaining "thermalized"
6 species are expected to react in the atmosphere with NO, NO2, SO2, H2O vapor, CO, and
7 carbonyl compounds, although the rate constants and mechanisms of these reactions are
8 currently uncertain (Finlayson-Pitts and Pitts, 1986).
9
10 Reactions of Oxygen-Containing Organics
11 Diesel and gasoline vehicle exhausts contain aldehydes and carboxylic acids. The
12 major loss processes for aldehydes involve photolysis and reaction with OH radicals (see
13 Table 3-9). Photolysis is an important loss process for formaldehyde, forming H atom and
14 HCO radical in the first step:
—> H + HCO
HCHO + hv
*—•> H2 + CO
15
16
17 The rapid reaction of H atom and HCO radicals with O2 produces HO2 radicals:
18
19
20 H + O2 -» HO2
21
22 HCO + O2 -» HO2 + CO
23
24
25 The higher aldehydes also photodissociate, ultimately yielding H02 radicals.
26 The OH radical reaction with formaldehyde yields CO and H02:
HCHO -I- OH -» H2O + HCO*
* 02
HO2 + CO
December 1994 3.3 j DRAFT-DO NOT QUOTE OR CITE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
For higher aldehydes, the RCO radicals initially formed add O2 to yield acylperoxy
radical, which can react further with NO (pathway a) and NO2 (pathway b):
RCHO -i- OH -> RCO + H2O
02
RC(O)OO-
NO
RC(O)O- + NO2
!?N02
I
CO2
RC(O)OONO2
(pcroxyacyl nitrates)
Acetaldehyde forms peroxyacetyl nitrate (PAN), which has been shown to be a direct-acting
mutagen toward Ames S. typhimurium strain TA100 (Kleindienst et al., 1985) and is
phytotoxic.
Benzaldehyde, the simplest aromatic aldehyde, forms peroxybenzoyl nitrate by the
series of reactions analogous to pathway (b), above, and nitrophenols as a result of the
reaction with NO, analogous to pathway (a), above:
NO, "
C6H3C(O)OO- _
C6H5C(0)OONO2
NO *
C6H5- + C02
NO2
NO
C6H5OO-
NO2
NO,
HO,
NO2C6H4OH
(o- and p-)
C6H5OH + O2
December 1994
3-32
DRAFT-DO NOT QUOTE OR CITE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
Carboxylic acids react with OH radicals under atmospheric conditions (Table 3-1).
For formic acid, hydrogen atoms are produced (Atkinson, 1988):
HC(0)OH + OH -» H20 + C02 + H
For the higher carboxylic acids, reaction products are currently unknown.
Reactions of Monocyclic Aromatic Compounds
The monocyclic aromatic compounds are removed from the atmosphere solely by
reactions with OH radicals (see Table 3-1). These reactions proceed by two pathways
(Finlayson-Pitts and Pitts, 1986; Atkinson, 1988):
(1) a major pathway by OH radical addition to the aromatic ring, and
(2) a minor pathway by H atom abstraction, either from the aromatic ring (in the case of
benzene) or from an alkyl group (in the case of alkyl-substituted aromatics). This latter
pathway leads to the formation of aromatic aldehydes and is analogous to OH radical
/reaction of alkanes.
For example, for toluene:
H
+OH -
-90%
(+ lesser amount
of p-isomer)
N0
CH0
CHO
-i-HO,
NO2
The products arising from the OH radical addition pathway (a) are not well known.
Reaction with O2, again occurring by two pathways, is expected to predominate:
December 1994
3-33
DRAFT-DO NOT QUOTE OR CITE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
H02
Pathway (a) yields phenolic compounds and, for toluene, accounts for =20% of the overall
reaction yield (Atkinson, 1988). The major reactions involve ring cleavage (opening),
leading to a variety of bifunctional products (Finlayson-Pitts and Pitts, 1986).
For phenolic compounds, in addition to OH radical reaction (proceeding mainly by
initial OH radical addition to the ring), the NO3 radical reaction that yields nitrophenols
appears to be important (Atkinson, 1988):
NO3
HNO, +
NO2
+ o-isomer
Reactions of Poly cyclic Aromatic Compounds
As discussed in Section 3.2.4, two- to four-ring PAHs emitted from diesel and spark-
ignition engines are distributed between gas and particle phases. For those PAHs present in
the gas phase, the reaction with OH radical is predominant, leading to atmospheric lifetimes
of a few hours or less (see Table 3-9). The nighttime gas-phase reaction with N2O5 is of
minor significance as a PAH loss process but (as will be discussed in the following text) may
be important as a formation route of mutagenic nitro-PAH. In addition, for the
PAH-containing cyclopenta-fused ring, such as acenaphthene, acenaphthylene, and
December 1994
3-34
DRAFT-DO NOT QUOTE OR CITE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
acephenanthrylene, the NO3 radical reaction can be an important gas-phase loss process
during nighttime hours.
Relatively few product data are available concerning these gas-phase reactions. It has
recently been shown that, in the presence of NOX, the OH radical reactions with naphthalene,
1- and 2-methylnaphthalene, acenaphthylene, biphenyl, fluoranthene, pyrene, and
acephenanthrylene lead to the formation of nitroarenes (Arey at al., 1986, 1989; Atkinson
et al., 1987, 1990; Zielinska et al., 1988, 1989a). The postulated reaction pathway involves
initial OH radical addition to the most reactive ring position; for example, C-3 position for
fluoranthene (Pitts et al., 1985a):
8 9
OH +
H OH
followed by NO2 addition in the C-2 position. Subsequent elimination of water results in
2-nitrofluoranthene formation:
NO2
H OH
-H,O
HOH
The analogous reaction sequence for pyrene produces 2-nitropyrene (2-NP) (Pitts
et al., 1985a). In contrast, the electrophilic nitration reaction of fluoranthene, or pyrene,
involving the NO2+ ion, produces mainly 3-nitrofluoranthene (3-NF) from fluoranthene and
1-NP from pyrene.
December 1994
3-35
DRAFT-DO NOT QUOTE OR CITE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
1
2
The gas-phase reactions of N2O5 with naphthalene, 1- and 2-methylnaphthalene,
acenaphthene, phenanthrene, anthracene, fluoranthene, and pyrene yield, in general, the same
nitro-PAH isomers as the OH radical reaction, but with different yields (Arey et al., 1989;
Sweetman et al., 1986; Atkinson et al., 1987, 1990; Zielinska et al., 1986, 1989a). For
example, the same 2-NF is produced from both OH radical and N2O5 gas-phase fluoranthene
reactions, but the reaction with N2O5 produces a much higher yield. The postulated reaction
pathway involves initial NO3 radical addition to the most reactive position, followed by NO2
addition to the neighboring position and elimination of HNO3 (Zielinska et al., 1986):
NO,
Table 3-10 summarizes the nitroarene product data for OH radical-initiated and N2O5
gas-phase reactions with several PAHs studied to date in environmental chambers (Arey
et al., 1989; Zielinska et al., 1990). Section 3.4.3 will discuss the fact that, generally, the
same nitro-PAH isomers that are formed from OH radical and N2O5 reactions are observed
in ambient air samples.
3.3.2.2. Particulate-Phase Reactions
Organic compounds present in diesel exhaust are partitioned into the particulate phase
under atmospheric conditions. The following chemical processes are likely to contribute to
the degradation of these compounds in the troposphere (Atkinson, 1988):
• Photolysis,
• Reaction with O3,
December 1994
3-36
DRAFT-DO NOT QUOTE OR CITE
-------
1 • Reaction with N2O5 during nighttime hours,
2 • Reaction with NO2 during nighttime and daytime hours,
3 • Reaction with H2O2, and
4 • Reaction with HNO3, HNO2, and H2SO4.
5 However, the atmospheric lifetimes of particle-bound organic compounds are not well
6 known. This is partially because (1) these chemical processes are dependent upon the nature
7 of the substrate (e.g., Behymer and Kites, 1985, 1988) and (2) many of the laboratory
8 studies have been done using atmospherically unrealistic adsorbents, such as glass fiber and
9 Teflon-coated glass-fiber filters, silica gel, and alumina. The extrapolation of sometimes
10 contradictory results reported by different laboratories to atmospherically realistic conditions
11 presents major problems.
12 The atmospheric fate of particle-bound PAHs has received much attention since their
13 potential toxicity was first observed. In their recent publication, Behymer and Hites (1988)
14 define two opposite schools of thought on this subject. One says that particle-bound PAHs
15 degrade quickly in the atmosphere with lifetimes as short as a few hours (e.g., Kamens
16 et al., 1988; Nielsen, 1988; Behymer and Hites, 1988). The other says that PAHs degrade
17 slowly, if at all, in the atmosphere and eventually deposit on soil or water. The latter
18 conclusion is supported by the studies of marine and lacustrine sediments (the ultimate
19 environmental sinks of PAHs) which have shown that the relative abundances of PAH, even
20 at the most remote locations, are similar to those in combustion sources and in air paniculate
21 matter (Laflamme and Hites, 1978; Hites et al., 1980; McVeety and Hites, 1988).
22
23 Photooxidation of Particulate Polycyclic Aromatic Hydrocarbons
24 Laboratory studies of photolysis of PAHs adsorbed on 18 different fly ashes, carbon
25 black, silica gel, and alumina (Behymer and Hites, 1985, 1988) and several coal stack ashes
26 (Yokley et al., 1986; Dunstan et al., 1989) showed that the extent of PAH photodegradation
27 depended very much on the nature of the substrate to which they are adsorbed. The
28 dominant factor in the stabilization of PAH adsorbed on fly ash was the color of the fly ash,
29 which is related to the amount of black carbon present. It appeared that PAHs were
30 stabilized if the black carbon content of the fly ash was greater than « 5 %. On black
December 1994 3.37 DRAFT-DO NOT QUOTE OR CITE
-------
1 substrates, half-lives of PAHs studied were on the order of several days (Behymer and
2 Kites, 1988).
3 Similar conclusions were reached from studies of photolysis of PAHs adsorbed on coal
4 stack ashes (Yokley et al., 1986; Dunstan et al., 1989). The relative quantity of carbon in
5 coal ash was the main factor determining the extent of photochemical degradation of pyrene
6 and benzo[0]pyrene adsorbed on the surface. In addition, in coal ashes that contained a
7 relatively large quantity of iron, the magnetic particles played a minor role in stabilizing
8 adsorbed pyrene toward photodegradation (Dunstan et al., 1989).
9 On the other hand, the environmental chamber studies of Kamens and co-workers
10 (1988) on the daytime decay of PAH present on residential wood smoke particles and on
11 gasoline internal combustion emission particles showed PAH half-lives on the order of 1 h at
12 moderate humidities and temperatures. At very low-angle sunlight, very low water-vapor
13 concentration, or very low temperatures, PAH daytime half-lives increased to a period
14 of days.
15 In addition, the atmospheric studies by Nielsen (1988), carried out in rural areas
16 during the winter and early spring when ambient temperatures and concentrations of NC^ and
17 O3 were low, showed evidence for atmospheric decay of more reactive PAHs, such as
18 benzo[0]pyrene and cyclopenteno[oflpyrene. Although no estimation of these PAH lifetimes
19 was given, the author concluded that the decay appeared to be relatively fast.
20 Because of the limited understanding of the mechanisms of these complex
21 heterogenous reactions, it is currently impossible to draw any firm conclusion concerning the
22 photostability of particle-bound PAHs in the atmosphere. Because diesel paniculate matter
23 contains a relatively high quantity of elemental carbon (see Section 3.2.2.1), it is reasonable
24 to assume that PAH adsorbed onto these particles should be relatively stable under standard
25 atmospheric conditions. Clearly, additional comprehensive and systematic investigation of
26 adsorbed-phase reactions of PAH is needed.
27
28 Nitration of Particulate Polycyclic Aromatic Hydrocarbons Under Simulated Atmospheres
29 Since 1978, when Pitts and co-workers (Pitts et al., 1978) first demonstrated that
30 benzo [a] pyrene (B[a]P) deposited on glass-fiber filters and exposed to air containing
31 0.25 ppm of NO2 with traces of HNO3 formed nitro-B[0]P, numerous studies of the
December 1994 3-38 DRAFT-DO NOT QUOTE OR CITE
-------
1 heterogenous nitration reactions of PAHs adsorbed on a variety of substrates in different
2 simulated atmospheres have been carried out (e.g., Finlayson-Pitts and Pitts, 1986). It has
3 been shown that the rate of PAH nitration in a NO2/HNO3 system depends on the substrate
4 to which the PAHs are adsorbed (Jager and Hanus, 1980) and on the reactivity of particular
5 PAH (Nielsen, 1984).
6 It has also been shown that certain PAHs deposited on glass-fiber and Teflon-
7 impregnated glass-fiber filters react with gaseous N2O5 yielding their nitro derivatives (Pitts
8 et al., 1985b,c). The nitro-PAH isomers formed from the parent PAH are the same as those
9 formed from electrophilic nitration reactions involving NO2+ ions. Thus, the most abundant
10 isomers formed were 1-NP from pyrene, 6-nitro-B(a)P from B[a]P, and 3-nitroperylene from
11 perylene. For fluoranthene, 3-, 8-, 7-, and 1-NF isomers were formed in approximately
12 equal amounts in N2O5 reactions, whereas the nitrofluoranthene isomer distribution from
13 electrophilic nitration reaction is 3-> 8-» 7-> 1-NF. However, no 2-NF (the sole
14 isomer formed from gas-phase N2O5 reaction) was produced from this adsorbed-phase
15 reaction (Pitts et al., 1985c). It was speculated that N2O5 becomes ionized on the filter
16 surface prior to the reaction with fluoranthene, but the resulting NO2+ ion is not "free"
17 nitronium ion, that is, not completely dissociated (Zielinska et al., 1986).
18 Based on these laboratory studies, it has been proposed that some nitro-PAHs detected
19 in ambient particles may be formed from the reaction of the parent PAH with gaseous
20 copollutants in the atmosphere, during the collection of paniculate matter, or both (Pitts
21 et al., 1978, 1985a; Jager and Hanus, 1980; Brorstrom et al., 1983). However, the
22 extrapolation of the data obtained under laboratory conditions to the ambient atmosphere
23 requires several major assumptions. These include, for example, the assumptions that
24 substrate effect, PAH concentration, the presence of copollutants, relative humidity, etc.,
25 have no major impact on PAH nitration reactivities. If these assumptions are valid, the
26 available data indicate that the nitration of particle-bound PAH with NCyHNC^ and N2O5 is
27 probably not significant under atmospheric conditions (Pitts et al., 1985c). However, this
28 may not always be the case in air sheds that have high NO2 and nighttime N2O5
29 concentrations.
30 The formation of nitro-PAHs during sampling may be an important problem for diesel
31 paniculate matter collection, because NO2 and HNO3 are present in relatively high
December 1994 3.39 DRAFT-DO NOT QUOTE OR CITE
-------
1 concentrations. However, Schuetzle (1983) concluded that the artifact formation of 1-NP
2 during dilution tube sampling accounts for less than 10 to 20% of the total 1-NP present in
3 diesel particles if the sampling time is less than 23 min (one FTP cycle) and if the sampling
4 temperature is not higher than 43 °C.
5 The formation of nitroarenes during ambient high-volume sampling conditions has
6 been reported to be minimal, at least for the most abundant nitropyrene and nitrofluoranthene
7 isomers (Arey et al., 1988a).
8
9 Ownolysis of Particulate Polycyclic Aromatic Hydrocarbons
10 Numerous laboratory studies have shown that PAHs deposited on combustion-
11 generated fine particles and on model substrates undergo reaction with O3 (e.g., Katz et al.,
12 1979; Pitts et al., 1980, 1986; Van Vaeck and Van Cauwenberghe, 1984; Finlayson-Pitts and
13 Pitts, 1986). The dark reaction of several PAHs deposited on model substrates toward O3
14 has been shown to be relatively fast under simulated atmospheric conditions (Katz et al.,
15 1979; Pitts et al., 1980, 1986). Half-lives of the order of one to several hours were reported
16 for the more reactive PAHs, such as B[
-------
1 showed that significant degradation of more reactive PAHs adsorbed on ambient paniculate
2 matter, such as B[a]P, pyrene, and benz[a]anthracene, may occur in O3-polluted
3 atmospheres.
4
5 3.3.3 Physical Removal Processes
6 3.3.3.1 Dry Deposition
7 Dry deposition is the removal of particles and gases from the atmosphere through the
8 delivery of mass to the surface by nonprecipitation atmospheric processes and the subsequent
9 physical attachment to, or chemical reaction with, surfaces such as vegetation, soil, water, or
10 the built environment (Dolske and Gatz, 1985). It should be noted that the surface itself may
11 be wet or dry; the term "dry deposition" refers to the mechanism or transport to the surface,
12 not to the nature of the surface itself. Dry deposition plays an important role as a removal
13 mechanism of pollutant in the absence of precipitation. Even in remote locations such as
14 Siskiwit Lake, located on a wilderness island in northern Lake Superior, the dry deposition
15 of aerosol was found to exceed the wet removal mechanism by an average ratio of
16 9:1 (McVeety and Hites, 1988).
17 Dry deposition is usually characterized by a deposition velocity Vd, which is defined
18 as the flux (F), or deposition rate, of the species S to the surface divided by the concentra-
19 tion [S] at some reference height (generally 1 m):
20
V F
dlsj
21
22 The amount of species deposited per unit area per second in a given geographical
23 location, that is, the flux, can be either calculated, if the deposition velocity and the pollutant
24 concentration are known, or measured experimentally. The deposition velocity depends on
25 the specific gaseous or particle species, the properties of the surface to which the species is
26 being deposited, and the reference height. It also depends on a micrometeorological process
27 that transports the species to the surface (see Finlayson-Pitts and Pitts [1986] for a more
28 detailed discussion).
December 1994 3.41 DRAFT-DO NOT QUOTE OR CITE
-------
1 For particles, the deposition velocities depend on the particle size, exhibiting a
2 minimum for particles of mean diameter of «0.1 to =1 jim. Table 3-11 gives the average
3 calculated lifetimes of atmospheric particles as a function of their diameter. Table 3-12 gives
4 some selected examples of dry deposition velocity for several inorganic and organic species.
5
6
TABLE 3-11. AVERAGE ATMOSPHERIC LIFETIMES OF PARTICLES DUE TO
DRY DEPOSITION3
Diameter (/im) Lifetime (days)
0.002 0.01
0.02 1
0.2 10
2 10
20 1
200 0.01
aSource: Atkinson (1988).
TABLE 3-12. EXAMPLES OF DRY DEPOSITION VELOCITIES3
Depositing Species Mean Deposition Velocity (cm/s)
Ozone OA9
Paniculate sulfur 0.17
Particles:
0.18 ftm median diameter 0.16
0.25 nm median diameter 0.35
SO2 2.1
HNO3 2.5
Benzeno[g/»']peryleneb 0.99
Indeno[l,2,3-q/l pyreneb 0.99
"From Dolske and Gatz (1985), with grass as the surface, except as noted.
bFrom McVeety and Hites (1988), with water as the surface, and PAH on particles.
1 However, because of the differences in meteorology, nature of surfaces, and
2 measurement uncertainties, the reported values for deposition velocities of a given species
3 can differ by more than an order of magnitude. It is important to note that for certain
December 1994 3-42 DRAFT-DO NOT QUOTE OR CITE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
chemicals that have relatively slow gas-phase chemical removal rates, such as SO2 and
HNO3, dry deposition can be the major loss process under typical atmospheric conditions.
3.3.3.2 Wet Deposition
Wet deposition encompasses all processes by which airborne pollutants are transported
to the Earth's surface in aqueous form (i.e., in rain, snow, or fog). The mechanisms of wet
removal from the atmosphere may be very different for particle-associated compounds and
for gas-phase compounds. However, because many organic compounds are partitioned
between the aerosol and vapor phase (as discussed in Section 3.2.4), both processes of gas
and particle scavenging may be important for a given compound (Ligocki et al., 1985a,b;
Bidleman, 1988). When there is no exchange of material between the particulate and
dissolved phases in the rain, the total scavenging of a given compound can be expressed as
(Pankowetal., 1984):
W = W
g
where W is the overall scavenging ratio:
w = [rain, total]
[air, total]
W is the gas scavenging ratio:
_ [rain, dissolved]
8 [air, gas]
is the particle scavenging ratio:
_ [rain, particulate]
[air, particulate]
and 0 is the fraction of the atmospheric concentration that is associated with particles.
December 1994
3.43 DRAFT-DO NOT QUOTE OR CITE
-------
1 Particle scavenging is a complex process that depends on the meteorological conditions
2 in the cloud as well as the size and chemical composition of the aerosol particles. The
3 simplest model for in-cloud particle scavenging involves nucleation scavenging followed by
4 coalescence of the cloud droplet into raindrops. As many as «106 cloud droplets of
5 «10 /im diameter must combine to form one 1-mm raindrop. Hence, scavenging ratios
6 under these conditions are expected to be of the order 106. However, this process alone
7 seldom produces precipitation.
8 In cold clouds, ice crystals grow by vapor accretion and by collection of supercooled
9 droplets (riming). Scavenging ratios may be considerably lower than 106 under these
10 conditions. In the case of below-cloud scavenging, Wp values have been estimated to be
11 103 to 105 for 0.01 to 1.0 pirn particles (Slinn et al., 1978). From these data, one may
12 expect to observe overall particle scavenging ratios in the range of 103 to 106.
13 Ligocki and co-workers (Ligocki et al., 1985a,b) measured gas- and particle-
14 scavenging ratios for a number of organic compounds, including PAHs and their derivatives.
15 Table 3-13 gives mean gas, particle, and overall scavenging ratios for measured neutral
16 organic compounds. It can be seen from this table that particle scavenging ratios range from
17 102 to 105, whereas gas scavenging ratios range from 22 to 105. Gas scavenging dominates
18 over particle scavenging for compounds of lower molecular weights (mw < 252 for PAHs).
19 Particle scavenging dominates for the alkanes, which are essentially insoluble in water.
20 The complexity of liquid-phase inorganic acid formation from gaseous precursors and
21 the problems of acid rain and acid fog are beyond the scope of this chapter and are not
22 discussed here (see Finlayson-Pitts and Pitts [1986] for more information).
23
24
25 3.4 ATMOSPHERIC CONCENTRATIONS OF PRIMARY DIESEL
26 EMISSIONS AND THEIR TRANSFORMATION PRODUCTS
27 Most of the data collected on vehicle emissions are from laboratory studies that used
28 dynamometer/dilution tube measurements. The relevance of these measurements to the
29 atmosphere is always a question, because emissions from vehicles on the road have much
30 higher dilution ratios ( = 103 versus 10), are collected at lower temperatures, are composed of
31 a large number of individual vehicle exhausts, have usually experienced longer residence
December 1994 3-44 DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 3-13. MEAN PARTICLE, GAS, AND OVERALL SCAVENGING RATIOS
FOR NEUTRAL ORGANIC COMPOUNDS8
Compound
Toluene
1 ,2,4-Trimethylbenzene
Ethylbenzene
m+p-Xylene
o-Xylene
Naphthalene
2-Methylnaphthalene
1 -Methylnaphthalene
Diethylphthalate
Dibenzofuran
Fluorene
Phenanthrene + anthracene
9-Fluorenone
Methylphenanthrenes
Fluoranthene
Pyrene
Eicosane
9, 10-Anthracenedione
Dioctylphthalate
Docosane
Chrysene
Benz[a] anthracene
Benzo[e]pyrene
Benzo[a]pyrene
Benzo [b +j +k] fluoranthene
Perylene
Tricosane
Tetracosane
Benzo[g/z/]perylene
Coronene
Mean0b
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.008
0.009
0.011
0.021
0.027
0.053
0.071
0.14
0.21
0.56
0.61
0.71
0.75
0.97
0.98
0.98
1.0
1.0
1.0
1.0
1.0
Mean Wp
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
NA
11,000
15,000
17,000
15,000
13,000
11,000
9,300
40,000
2,400
36,000
27,000
2,600
1,300
2,000
1,700
2,200
1,800
22,000
16,000
3,100
5,900
Mean Wg
22 ±5
21 ±9
27 ± 1
33 ± 17
35 ± 15
250 ± 73
250 ± 78
330 ± 100
20,000
930
1,500
3,300
11,000
2,500
6,300
5,900
NA
27,000
20,000
NA
18,000
12,000
5,800
NA
7,400
NA
NA
NA
NA
NA
Mean W*
22 ±5
27 ±9
127 ± 11
33 ± 17
35 ± 15
250 ± 73
250 ± 78
330 ± 100
20,000
1,000
1,600
3,500
11,000
2,800
6,600
6,100
5,600
22,000
30,000
17,000
7,000
4,000
2,100
1,700
2,300
1,800
22,000
16,000
3,100
5,900
Dominant
Scav. Mech.d
g
g
g
g
g
g
g
g
g
g
g
g
g
g
g
g
P
g
P
P
g
g
P
P
P
P
P
P
P
P
"From Ligocki et al. (1985a,b).
b0 = (aerosol)/(vapor + aerosol).
CW = Wp0 + Wg(l-0).
dg = Gas;
p = Particle.
NA = Not available.
December 1994
3-45 DRAFT-DO NOT QUOTE OR CITE
-------
1 times (seconds to days, versus =5 s) before collection or measurement, and, as discussed in
2 Section 3.3, have the opportunity to interact with ambient air pollutants (including exhausts
3 of other vehicles and vehicle types).
4 Pierson and co-workers (Pierson et al., 1983; Salmeen et al., 1985) conducted field
5 experiments in the Allegheny Mountain Tunnel of the Pennsylvania Turnpike, to address this
6 problem. They found that the diesel-produced paniculate matter at tunnels was, in general,
7 very similar to that encountered in dilution-tube studies with respect to total paniculate
8 matter emission rates, percentage extractables, hydrocarbon molecular weight distribution,
9 HPLC profiles, particle size distribution, elemental compositions, and extract mutagenicities.
10 However, these findings did not preclude the possibility of substantial differences in detailed
11 chemical compositions. Indeed, the concentration of 1-NP in the extract of paniculate
12 samples collected in the Allegheny Mountain Tunnel was reported to be lower than would be
13 predicted on the basis of laboratory dilution tube measurements either for diesel or spark-
14 ignition vehicles (Gorse et al., 1983).
15 Some recent data on organic compound concentrations in air sheds heavily impacted
16 by motor vehicle emissions (tunnels, roadsides, etc.) are reviewed in the following text. The
17 possibility of using these data to distinguish emissions from different sources is discussed in
18 Section 3.4.4.
19
20 3.4.1 Volatile Organic Compounds Attributable to Traffic
21 Individual volatile hydrocarbons and aldehydes were measured along a section of U.S.
22 Highway 70 near Raleigh, NC (Zweidinger et al., 1988). Traffic volume during sampling
23 was determined by visual counting (=1,050 ± 10% vehicles per hour in each direction) and
24 was classified into four groups: (1) light-duty, including gasoline and diesel vehicles through
25 Class 2 trucks; (2) heavy-duty gasoline; (3) HDD; and (4) motorcycles. Typical distributions
26 were 91.5, 3.2, 5.1, and 0.2%, respectively.
27 Table 3-14 lists the mean concentrations from four roadsides for selected hydrocarbons
28 and aldehydes, expressed in ppb C and as a percentage contribution of individual
29 hydrocarbons and aldehydes to total nonmethane hydrocarbons (TNMHC) and total
30 aldehydes, respectively.
December 1994 3-46 DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 3-14. CONCENTRATIONS OF INDIVIDUAL HYDROCARBONS AND
ALDEHYDES MEASURED IN THE RALEIGH, NC, ROADSIDE STUDY"
Compound
Hydrocarbons
Ethane
Ethylene
Acetylene
Propane
Propylene
n-Butane
1-Butene
n-Pentane
wo-Pentane
Methylcyclopentane
Methylcyclohexane
n-Decane
Benzene
Toluene
m- and p-Xylenes
o-Xylene
Ethylbenzene
TNMHCd
Total paraffins
Total olefins
Total aromatics
Total unidentified NMHC
Aldehydes
Formaldehyde
Acetaldehyde
Acrolein
Benzaldehyde
Total aldehydes
Concentration (ppbC)
16.30
64.30
50.90
7.90
22.60
15.80
5.70
25.40
53.00
10.40
4.70
3.00
29.00
59.30
53.10
12.70
12.00
900.00
369.20
164.50
252.00
63.60
6.74
3.00
1.20
2.31
16.38
Percent Contribution1"'0
1.81
7.15
5.65
0.88
2.51
1.75
0.64
2.82
5.89
1.15
0.53
0.33
3.23
6.60
5.90
1.41
1.33
100.00
41.00
18.20
28.00
7.10
1.05
18.40
7.30
3.88
100.00
aFrom Zweidinger et al. (1988).
bPercent based on ppbC.
cPercent contribution of individual hydrocarbons to TNMHC and of individual aldehydes to total aldehydes.
dTNMHC = Total nonmethane hydrocarbons.
1 The roadside VOC distribution was compared with dynamometer/dilution tube results
2 on in-use vehicles, which were weighted hi an attempt to reflect the same model year
3 distribution as observed on the roadway (Sigsby et al., 1987; see also Table 3-1). The two
4 sets of data were similar in that the different driving cycles, like the different sampling sites,
December 1994 3.47 DRAFT-DO NOT QUOTE OR CITE
-------
1 generally show no significant differences in the distribution of hydrocarbons or aldehydes on
2 a percentage of total basis. There were, however, differences observed between the sets of
3 data, particularly for the contribution of combustion products (i.e., hydrocarbons below
4 C4 and aldehydes). For the roadside study, ethylene, formaldehyde, and acetaldehyde were
5 lower, whereas acetylene was higher than in the dynamometer study. However, noncatalyst
6 vehicles, which constituted 15% of all light-duty vehicles in the roadside study, were not
7 included in the dynamometer study, nor were LDD and HDD vehicles and trucks.
8
9 3.4.2 Polycyclic Aromatic Hydrocarbons
10 Paniculate and vapor phase samples were collected from the traffic passing through
11 the Baltimore Harbor Tunnel and analyzed for PAHs and related compounds (Benner et al.,
12 1989). High-volume air samplers equipped with Teflon filters backed by PUF plugs were
13 used for sample collection. There was no breakdown of traffic into numbers of diesel- and
14 gasoline-fueled vehicles.
15 The range of particle-phase PAH concentrations and the mean particle- and vapor-
16 phase PAH concentrations for 48 samples collected in the Baltimore Tunnel are tabulated in
17 Table 3-15. The ratios of mean particle-phase PAH concentrations to that of B[e]P, which is
18 considered to be a nonreactive PAH, are also given in this table.
19 As can be seen from Table 3-15, alkyl-substituted phenanthrenes in the tunnel samples
20 had relatively high concentrations compared with those of the parent compound. This
21 suggests a significant contribution from diesel vehicle emissions (particularly diesel-fueled
22 trucks) because extracts of diesel particulate matter are known to have significant
23 concentrations of methyl and dimethylphenanthrenes (see Table 3-5 and Yu and Kites, 1981).
24 Factor analysis was applied to the tunnel data in an attempt to identify factors
25 associated with different types of vehicles; two factors were obtained. The alkylated
26 phenanthrenes loaded significantly on factor 1, suggesting the diesel vehicles as the source of
27 these compounds. Several of the higher-molecular-weight PAHs loaded onto factor 2, which
28 may be associated with the contribution of gasoline-fueled emissions in the tunnel.
29 Ambient air sampling for PAH was also conducted during a summertime
30 photochemical air pollution episode in Glendora, CA, at a site situated less than 1 km from
31 the heavily traveled 1-210 freeway and generally downwind of Los Angeles; therefore, the
December 1994 3-48 DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 3-15. PARTICLE- AND VAPOR-PHASE POLYCYCLIC AROMATIC
HYDROCARBON CONCENTRATIONS FOR BALTIMORE HARBOR
TUNNEL SAMPLES8
Concentration (ng/m3)
Compound
Phenanthrene
Anthracene
3-Methylphenanthrene
2-Methylphenanthrene
2-Methylanthracene
9-and 4-Methyl-phenanthrene and
4H-cyclopenta[*fe/|-phenanthrene
1 -Methylphenanthrene
2 , 6-Dimethylphenanthrene
2,7-Dimethylphenanthrene
1,3-, 2,10-, 3,9-and 3,10-Dimethyl-
and phenanthrene
1,6- and 2,9-Di-methylphenanthrene
1 ,7-Dimethylphenanthrene
2 , 3-Dimethy Iphenanthrene
Fluoranthene
Pyrene
Benzo[£/»']fluoranthene
Cyclopenta[ofjpyrene
Benz[a]anthracene
Chrysene/triphenylene
Benzofluoranthenesf&j, +k]
Benzo[e]pyrene
Benzo[a]pyrene
Indeno[7, 2, 3-oflpyrene
Benzo[g/»']perylene
Coronene
Range,
particles
4.3-56
0.6-12
3.9-58
5.3-74
0.6-12
4.7-50
2.6-43
4.7-62
3.4-38
9.5-119
4.5-63
3.9-41
3.5-41
6.4-69
9.7-76
3.2-26
7.6-65
1.9-29
2.9-47
2.2-44
1.5-19
1.3-26
0.3-15
1.8-18
1.0-10
Mean,
particles
18.0
2.9
13.9
19.0
3.0
12.9
9.8
14.0
9.2
26.0
14.0
10.2
9.3
20.0
27.0
9.6
20.0
7.6
12.0
10.6
5.0
5.8
4.6
8.0
4.7
Mean,
vaporb
132
18
70
—
5.3
71
43
30
16
61
27
20
16
16
26
NDd
ND
ND
ND
ND
ND
ND
ND
ND
ND
Ratio0
to B[e]P
4.3
0.6
3.3
4.6
0.7
3.0
2.3
3.4
2.2
6.3
3.3
2.4
2.2
4.5
6.3
2.1
4.6
1.5
2.4
2.1
1.0
1.1
0.9
1.6
0.9
aFrom Benner et al. (1989).
bMean concentrations of PAH collected on PUF plug (calculated from data given in Table HI of Benner
et al., 1989).
cMean ratios to paniculate phase B[e]P.
dNone detected.
1
2
3
site was affected by motor vehicle emissions (Atkinson et al., 1988). Samples were collected
by means of high-volume samplers equipped with Teflon-impregnated glass fiber filters
backed by PUF plugs. Table 3-16 shows the average (from three daytime and three
December 1994
3-49 DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 3-16. AVERAGE AMBIENT CONCENTRATIONS OF POLYCYCLIC
AROMATIC HYDROCARBONS MEASURED IN GLENDORA, CAa
PAH
Phenanthrenec
Anthracene0
Fluoranthene
Pyrene
Eemo[ghi\ -fluoranthene
Cyclopentafof] -py rene
Benz[fl]anthracene
Chrysene/Tri-phenylene
Benzofluoranthenes[6,/' +k]
Benzo[tf]pyrene
Benzo[a]pyrene
Indeno[7, 2, 3-ofl-pyrene
Benzo[g/ii]perylene
Coronene
Total Concentration (ng/m3)
20.0
1.0
5.6 (0.26)d
4.1 (0.35)d
0.26
0.09
0.2
1.0
1.6
0.94
0.33
1.6
3.8
2.8
Ratio to B[e]Pb
—
—
0.27
0.37
0.28
0.1
0.22
1.1
1.7
1.0
0.35
1.7
4.0
3.0
"From Atkinson et al. (1988).
bRatios of particle-phase PAH to particle-phase B[e]P.
cPhenanthrene and anthracene were not present on filters, only on PUP plugs.
dFluoranthene and pyrene are distributed between gas and paniculate phases; numbers in parentheses represent
particle concentrations.
1 nighttime samples) concentrations of PAH measured and the ratios of their concentrations to
2 that of B[e]P. Unfortunately, no alkylated phenanthrenes were measured.
3 As can be seen from the comparison of Tables 3-15 and 3-16, the concentrations of all
4 PAHs measured in Glendora were much lower than those measured in the tunnel, as would
5 be expected. However, the ratios of the concentrations of particle-bound PAH to that of
6 B[e]P were also different for the two sites, usually much lower for the Glendora site (except
7 for higher molecular weight PAHs, indeno[l,2,3-crf]pyrene, benzo[gfti]perylene, and
8 coronene). This may indicate either contributions from sources other than motor vehicles in
9 the Glendora study or PAH photochemical transformations occurring on particles prior to or
10 during high-volume sample collections (or both). The generally higher PAH concentrations
11 for nighttime versus daytime sampling periods (Atkinson et al., 1988) seem to support the
12 latter possibility. However, the influence of meteorology cannot be excluded. This
13 conclusion is also consistent with high levels of photochemical pollutants observed in
14 Glendora; for example the daily maxima of O3 concentrations (which occurred always
December 1994 3-50 DRAFT-DO NOT QUOTE OR CITE
-------
1 between 1400 and 1700 hours, PST) ranged from 160 to 240 ppb throughout the entire
2 9 days of the study.
3
4 3.4.3 Nitroarene Concentrations in Ambient Air
5 Diesel paniculate matter contains a variety of nitroarenes, with 1-NP being the most
6 abundant among identified nitro-PAH. The concentration of 1-NP was measured in the
7 extract of paniculate samples collected at the Allegheny Mountain Tunnel on the
8 Pennsylvania Turnpike (Gorse et al., 1983). This concentration was 2.1 ppm and <5 ppm
9 (by mass) of the extractable material from diesel and spark-ignition vehicle paniculate
10 matter, respectively. These values are much lower than would be predicted on the basis of
11 laboratory dilution tunnel measurements either for diesel or for spark-ignition engines.
12 Unfortunately, there are no published tunnel or roadside data on concentrations of nitroatenes
13 other than 1-NP.
14 Several nitroarene measurements were conducted in air sheds heavily impacted by
15 motor vehicle emissions (Arey et al., 1987; Atkinson et al., 1988; Zielinska et al., 1989a,b;
16 Ciccioli et al., 1989). For example, ambient paniculate matter samples were collected at
17 three sites (Claremont, Torrance, and Glendora) situated in the Los Angeles Basin; the
18 Claremont and Glendora sites are =30 km and =20 km northeast, respectively, and the
19 Torrance site is =20 km southwest of downtown Los Angeles (Arey et al., 1987; Atkinson
20 et al., 1988; Zielinska et al., 1989a,b). The sampling was conducted during two
21 summertime periods (Claremont, September 1985, and Glendora, August 1986) and one
22 wintertime period (Torrance, January and February 1986). Table 3-17 lists the maximum
23 concentrations of nitropyrene and nitrofluoranthene isomers observed at these three sites
24 during the daytime and nighttime sampling periods.
25 As can be seen from Table 3-17, 1-NP, the most abundant nitroarene emitted from
26 diesel engines, is not the most abundant nitroarene observed in ambient paniculate matter
27 collected at three sites heavily impacted by motor vehicle emissions. Of the two nitropyrene
28 isomers present, 2-NP, the main nitropyrene isomer formed from the gas-phase OH radical
29 initiated reaction with pyrene (see Section 3.3.2.1), is sometimes more abundant. The 2-NF
30 was always the most abundant nitroarene observed in ambient particulate matter collected at
31 these three sites (see Table 3-7 and Ciccioli et al., 1989) and this nitrofluoranthene isomer is
December 1994 3.5! DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 3-17. THE MAXIMUM CONCENTRATIONS OF NITROFLUORANTHENE
(NF) AND NITROPYRENE (NP) ISOMERS OBSERVED AT THREE
SOUTH COAST AIR BASIN SAMPLING SITES
Concentration (pg/m3) at
Nitroarene, Collection Period
2-NF, day
2-NF, night
3-NF, day
3-NF, night
8-NF, day
8-NF, night
1-NP, day
1-NP, night
2-NP, day
2-NP, night
Claremonta-b
40
1,700
3
=3
2
2
3
10
1
8
Glendorac-d
350
2,000
NDf
ND
3
4
15
15
14
32
Torrance3'6
410
750
=3
70
8
50
60
50
50
60
"From Zielinska et al. (1989b).
bDaytime sample collected from 1200 to 1800 hours and nighttime sample from 1800 to 2400 hours on
September 13, 1985.
cFrom Atkinson et al. (1988).
dDaytime sample collected from 0800 to 2000 hours on August 20, 1986, and nighttime sample from 2000 to
0800 hours on August 20 and 21, 1986.
eDaytime sample collected from 0500 to 1700 hours on January 28, 1986, and nighttime sample from 1700 to
0500 hours on January 27 and 28, 1986.
fND = None detected.
1 not present in diesel and gasoline vehicle emissions. The 2-NF is the only nitroarene
2 produced from the gas-phase OH radical-initiated and N2O5 reactions with fluoranthene (see
3 Sections 3.3.2.1 and 3.3.2.2), whereas mainly 3-NF, and lesser amounts of 1-, 7-, and
4 8-nitroisomers are present in diesel paniculate matter and are produced from the electrophilic
5 nitration reactions of fluoranthene.
6 Figure 3-4 compares the nitroarenes formed from the OH radical-initiated reaction of
7 fluoranthene and pyrene in an environmental chamber (upper trace) with the ambient samples
8 collected at Torrance (lower trace). It is very unlikely that N2O5 could have been present
9 during the nighttime winter collections in Torrance, given the high level of NO present at
10 sunset. More likely a relatively high level of OH radicals was present because of the
11 measured high concentration of HNO2, which photolyzes to yield OH radicals. This suggests
December 1994 3-52 DRAFT-DO NOT QUOTE OR CITE
-------
100-
2-r
m/z 247
7-NF
A J
^F Products of the
OH Radical Reaction
with Fluoranthene
and Pyrene
2-NP
8-NF A
1 A A
23
24
25
26
27
i 100-1
m/z 247
2-NF
Ambient Sample
2-NP
7-NF
1-NP
8-NF
21
22
23
24
25
Figure 3-4. Mass chromatograms of the molecular ion of the nitrofluoranthenes (NF)
and nitropyrenes (NP) formed from the gas-phase reaction of fluoranthene
and pyrene with the OH radicals (top) and present in the ambient
particulate sample collected at Torrance, CA (bottom).
Source: Arey et al. (1989).
December 1994
3-53
DRAFT-DO NOT QUOTE OR CITE
-------
1 that all isomers observed in Figure 3-4 (lower trace), with the exception of 1-NP, are the
2 product of the OH-radical-initiated reactions of the parent PAH. Direct emissions may
3 account for the 1-NP (and 3-NF) observed at relatively low levels in these ambient samples.
4 (See Zielinska et al. [1989b] for full discussion of all the MW 247 nitroarenes observed in
5 ambient particles.)
6 Although the reaction with OH radicals is the major atmospheric loss process for gas-
7 phase fluoranthene and pyrene (Table 3-9) evidence for atmospheric formation of 2-NF from
8 N2O5 reaction with fluoranthene has also been reported (Zielinska et al., 1989b). Because
9 the 2-NF/2-NP yield ratio for N2O5 reactions, observed from environmental chamber
10 experiments, is > 100, compared to «10 for the OH radical reaction (Table 3-2) the high
11 2-NF/2-NP concentration ratio in ambient samples suggests a contribution from the N2O5
12 reaction with fluoranthene. Figure 3-5 shows a comparison of a wintertime samplecollected
13 in Torrance (upper trace) with a summertime sample collected hi Claremont (lower trace).
14 The 2-NF/2-NP ratio reached «200 for the summer night sample. The N2O5 concentration
15 was calculated to be =5 ppb for this night, which supports the suggested formation route of
16 2-NF via reaction with N2O5 (Zielinska et al., 1989b).
17 The evidence presented in the preceding text, as well as the observation that 2-NF has
18 been the most abundant MW 247 nitroarene in ambient samples collected worldwide
19 (Ramdahl et al., 1986), strongly suggests that the atmospheric formation from the parent
20 PAH, not the direct automotive emissions, is the major source of these nitroarenes in
21 ambient air. However, under certain sampling conditions, when ambient paniculate matter is
22 collected very close to emission sources, the MW 247 nitroarene profile may be different.
23 For example, in urban samples collected during wintertime rush hours at a central square in
24 Rome, Italy, at a height of 1.5 m above street level, 2-NF and 2-NP were not observed
25 (Cicciolietal., 1989).
26
27 3.4.4 Need for Atmospheric Tracers of Diesel Emissions
28 Receptor source apportionment models assist hi the identification of the principal
29 sources of airborne pollutants and in the determination of source contributions to ambient
30 aerosol mass concentrations (or gas- or particulate-phase species concentrations, light
31 extinction, etc.). The Chemical Mass Balance (CMB) is one of the most widely used of
December 1994 3-54 DRAFT-DO NOT QUOTE OR CITE
-------
100 r
m/z 247
2-NF Winter Night
7-NF
3-NF 2-NP
A 8'NF J
ft IM
'1 . 0
100 r
m/z 247
2-NF Summer Night
1-NP 2_NP
A >L
Relative Retention Time
Figure 3-5. Mass chromatograms of the molecular ion of the nitrofluoranthenes (NF)
and nitropyrenes (NP) present in ambient particulate samples collected in
Torrance, CA (top), and Claremont, CA (bottom).
December 1994
3-55
DRAFT-DO NOT QUOTE OR CITE
-------
1 several existing receptor models (see Watson et al. [1990] for more detailed discussion).
2 The CMB technique requires the measurement at receptor sites of selected chemical
3 constituents ("tracers") that can be attributed to specific emission sources. To use the model,
4 the concentration of constituents at each source type (source profile) must be known. The
5 chemical tracers used to construct the source profiles must be present in different proportions
6 in different source types, must remain relatively constant for each source type, and, in
7 addition, the changes in their concentrations between sources and receptors must be
8 negligible or able to be approximated.
9 Trace elements, OC, and EC are most widely used for the construction of the source
10 profile. However, emissions from certain sources are difficult to distinguish by using only
11 these tracers. For example, potassium, which is widely used as a wood-smoke tracer, is also
12 abundant in resuspended soil and cigarette smoke. Because of increasing use of unleaded
13 gasoline, the ambient concentrations of traditional motor vehicle tracers, lead and bromine,
14 are diminishing and there is a need to identify alternative tracers for mobile sources.
15 In addition, there is a need for tracers that can distinguish between gasoline-vehicle and
16 diesel-vehicle emissions. Finally, some sources of toxic air pollutants do not emit trace
17 metals.
18 Unique tracers, which could be used to distinguish diesel emissions from those of
19 spark-ignition engines, have not yet been identified in diesel exhausts. However, as has been
20 demonstrated by an experiment conducted in Vienna, Austria (Horvath et al., 1988), such
21 tracers can be added deliberately. In the Vienna case, the rare-earth element dysprosium was
22 added in the form of an organometallic compound to the entire diesel fuel supply in Vienna
23 and the vicinity. From the amount of this tracer in the atmospheric samples, the contribution
24 of diesel vehicle emissions to the particulate pollutants in Vienna was estimated («12 to
25 33%). This approach, however, is costly and not always practical.
26 A second, much less precise approach for estimating diesel exhaust contributions to
27 ambient aerosol is based on the fact that diesel- and gasoline-fueled vehicles coexist on the
28 highways (Cass, 1990). A "highway aerosol signature" can be constructed as an emissions-
29 weighted average of the chemical composition of the aerosol from gasoline-fueled
30 automobiles and trucks, diesel-fueled automobiles and trucks, tire wear, and brake wear.
31 This has been done successfully in receptor modeling studies of the Los Angeles ambient
December 1994 3-56 DRAFT-DO NOT QUOTE OR CITE
-------
1 aerosol (Cass and MacRae, 1983). The lead content of the leaded gasoline burned by the
2 gasoline-powered vehicles is used to determine the amount of "highway" aerosol in the
3 ambient air. The diesel exhaust contribution can be then estimated in proportion to the
4 relative contribution of diesel exhaust in the highway emission profile (which is based on
5 dynamometer emission tests and estimates of vehicle miles travelled by each vehicle type).
6 However, as mentioned above, the lead tracer approach will not be possible in the near
7 future, as lead is being removed from gasoline. In addition, the contribution of diesel
8 vehicles to total highway traffic is not always known, or is not known with required
9 accuracy. Furthermore, this method does not address the emissions from diesel engines used
10 in railroad locomotives, ships, off-highway construction equipment, etc.
11 Finally, it has been suggested that, because diesel particulate emissions are enriched in
12 EC, the EC content of an ambient particle, when scaled in proportion to the EC content of
13 diesel exhaust (roughly 70% EC by mass, see Table 3-2) places an upper limit on the amount
14 of diesel exhaust aerosol that can be present in an ambient sample (Cass, 1990). It has been
15 calculated from emission inventory data that diesel engines contribute approximately
16 67% (49% diesel highway vehicles and 18% diesel ships, rails, off-highway equipment, etc.)
17 of the fine EC particulate mass emitted to the Los Angeles atmosphere (Gray, 1986).
18 The contribution of diesel engine exhausts to ambient aerosol concentrations can be
19 quantified by atmospheric transport modeling. A model to predict the long-term average
20 concentration of EC in the Los Angeles area by simulating the transport of emissions by
21 atmospheric processes such as advection, diffusion, and deposition has been constructed
22 (Gray, 1986). A multiple-source modeling study of primary carbon particle emissions to the
23 Los Angeles atmosphere has been already conducted, and the diesel exhaust aerosol
24 concentrations were computed along with the contributions of other major primary carbon
25 particle sources to OC and EC particle concentrations (Gray, 1986). However, detailed
26 information regarding the sources of emissions, meteorology, atmospheric dispersion
27 parameters, deposition rates, and aerosol carbon background concentrations is required for
28 successful application of this air quality model. Particularly, because many combustion
29 sources contribute to ambient aerosol carbon concentration, the construction of a detailed,
30 spatially resolved inventory of fine carbon particle emissions is necessary for this method.
December 1994 3.57 DRAFT-DO NOT QUOTE OR CITE
-------
1 Because organic compounds are emitted from all combustion sources, the potential
2 exists for their use in receptor modeling. Reviews of existing literature data (see Daisey
3 et al., 1986) revealed the potential usefulness of certain classes of organic compounds in
4 developing "fingerprints" for specific emission sources. At the same time, it is clear that the
5 existing source emission data are not sufficient for receptor modeling purposes; thus,
6 appreciably more experimental data are required to develop consistent emission profiles for
7 specific emission sources.
8 The class of organic compounds most suitable for serving as a source tracer should be:
9
10 • Emitted in relatively high concentration to allow small sample sizes and
11 short sampling times,
12
13 • Relatively easy to separate from other classes of organic compounds,
14
15 • Relatively easy to identify and quantify on the basis of the
16 chromatographic and spectral properties of its members,
17
18 • Chemically stable to enssure the same composition at source and
19 receptor sites (i.e., it should not undergo atmospheric transformations),
20 and
21
22 • Emitted in reasonably stable proportion to fuel burned or to other emissions
23 (e.g., CO2, THC, and paniculate mass).
24
25
26 In addition, the composition pattern (profile) for this class should differ among different
27 emission sources to assist in distinguishing among them.
28 Polycyclic aromatic hydrocarbons have been advocated as potential tracers of various
29 types of combustion emissions (Daisey et al., 1986). Polycyclic aromatic hydrocarbons are
30 present in all combustion sources, and their relative proportions in emissions from a given
31 source type frequently vary over several orders of magnitude. In addition, good sampling
32 and analytical methods already exist for this class of compounds. However, although the
33 PAH concentrations in motor vehicle emissions were frequently measured in the past, the
34 purpose of these measurements has been to determine emission rates under different
35 operating conditions rather than to establish PAH profiles for source receptor modeling.
36 Therefore, there is an apparent lack of compatibility among available motor vehicle PAH
37 emission profiles.
December 1994 3-58 DRAFT-DO NOT QUOTE OR CITE
-------
1 Currently, it is not possible to make any firm judgment regarding the possibilities for
2 distinguishing diesel emissions from those of spark-ignition engines. However, some PAH
3 compounds appear to be promising in this regard. As mentioned in Section 3.2.2.3, diesel
4 exhaust is enriched in alkyl-substituted PAHs, particularly alkyl-phenanthrenes, relative to
5 parent compounds. It is known that low-temperature formation (petrogenesis) of PAHs
6 produces a mixture enriched in alkyl-substituted PAH and other kinetically favored
7 compounds, whereas high-temperature processes (combustion and pyrolysis) favor the
8 generation of unsubstituted compounds. In agreement with this, it has been reported that the
9 total concentration of alkyl-PAH in diesel emissions increases as the cylinder exhaust
10 temperature decreases (Jensen and Kites, 1983).
11 Benner and co-workers (1989) reported that the high concentrations of methyl- and
12 dimethylphenanthrenes measured in the Baltimore Harbor Tunnel suggest the contribution of
13 diesel emission sources in the tunnel (see also Section 3.4.2). However, alkyl-substituted
14 PAHs are also abundant in coal and coal-derived material (White, 1983) and shale oil
15 (Garrigues et al., 1987). More data on alkylated PAH concentrations in different combustion
16 sources are clearly needed.
17 The most important limitation of the use of PAH as emission markers is their
18 relatively high chemical reactivity (see Section 3.3). Thus, the PAH profile determined at the
19 emission source may differ considerably from the source PAH profile as it exists in the
20 ambient atmosphere. However, it has been suggested (Miguel and Pereira, 1989) that some
21 presumably more stable particle-bound PAHs, such as benzo[fc]fluoranthene,
22 benzo[g/H~]perylene, or ideno[l,2,3-o/]pyrene can be used as tracers of automotive emissions
23 at receptor sites that have no other major sources of PAHs.
24 Taking chemical reactivity and ambient abundance into account, alkanes seem to be
25 more suitable for tracing motor vehicle emissions than are PAHs. The lower homologs that
26 exist entirely in the gas phase react slowly with the OH radicals and their atmospheric
27 lifetimes are on the order of several days (Table 3-9). The higher, mostly particle-associated
28 homologs (C > 20) are relatively unreactive. The n-alkanes originating from natural sources
29 (e.g., plant waxes) could be distinguished from those originating from anthropogenic sources
30 on the basis of the odd-to-even carbon number preference. Also, the ratio of normal to
31 branched isomers is lower in emissions from fossil fuel combustion than for biogenic alkane
December 1994 3.59 DRAFT-DO NOT QUOTE OR CITE
-------
1 aerosol (Simoneit, 1984). In addition, the literature data indicate that a homologous series of
2 alkylcyclohexanes (CnH2n with n ranging from 16 to 29) are characteristic of spark-ignition
3 vehicle emission but are only trace components of diesel exhaust (Simoneit, 1984). It has
4 also been suggested that it may be possible to distinguish emissions from diesel and spark-
5 ignition engines based on the ratios of methylated isomers to total branched isomers (Boone
6 and Macias, 1987).
7 However, the very limited data sets available for source emissions indicate that the
8 relative concentrations of the particulate C-24 to C-36 alkanes vary only one order of
9 magnitude among different source types. It is possible that the inclusion of the semivolatile
10 alkanes would extend the range of relative concentrations. Clearly, much more measurement
11 of alkanes in source emissions is needed to allow comparison among sources.
12 In summary, the existing data indicate that it may be possible to use organic
13 compound profiles, perhaps in combination with inorganic species, to assist in distinguishing
14 among diesel-fueled vehicles, gasoline-fueled vehicles, and other particulate pollutant
15 sources. However, the determination of organic and inorganic compositions of emissions
16 from a number of sources, by using sampling and analytical methods appropriate for the
17 purpose of source receptor apportionment modeling, is clearly necessary.
18
19
20 3.5 BIOASSAY-DIRECTED CHEMICAL ANALYSIS
21 In 1976, Tokiwa and co-workers reported that organic extracts of ambient particles
22 collected in Japan were active in the Ames S. typhimurium assay when tested in the presence
23 of homogenized rat liver tissue, colloquially known as S9 mix (Tokiwa et al., 1976, 1977).
24 Soon thereafter, direct mutagenic activity (i.e., in the absence of S9 mix) of extracts of
25 ambient particles collected in major cities throughout the world was reported (Pitts et al.,
26 1977; Talcott and Wei, 1977; Tokiwa et al., 1980; Lofroth, 1981; Finlayson-Pitts and Pitts,
27 1986). It has been shown that this direct activity was primarily associated with organic
28 species present in inhalable particles of <2 /urn diameter.
29 Reports of the presence of direct-acting mutagenic species in extracts of ambient POM
30 and vehicle emissions resulted in investigations that identified a number of direct-acting PAH
31 derivatives, mostly in diesel particulate extracts (Schuetzle, 1983). The use of short-term
December 1994 3-60 DRAFT-DO NOT QUOTE OR CITE
-------
1 bacterial bioassays in conjunction with analytical procedures, so-called "bioassay directed
2 chemical analysis," proved to be extremely useful in identifying the in complex
3 environmental mixtures (Schuetzle and Lewtas, 1986).
4 The contribution of mutagenic species present in diesel paniculate extracts to the
5 mutagenic activities of ambient particulate extracts has been the subject of several studies
6 (Gibson, 1983; Tokiwa et al., 1983; Siak et al., 1985). The contribution of atmospherically
7 formed nitroarenes to the mutagenic activities of particulate samples collected in Southern
8 California has been also recently assessed (Arey et al., 1988b).
9 Most environmental samples are complex mixtures and comprise thousands of
10 chemical compounds. Identification of the biologically active compounds, often present in
11 minute quantities, by traditional analytical methods would present an enormous if not
12 impossible task. It became apparent in the late 1970s that short-term bioassays could be used
13 in combination with chemical fractionation to simplify the process of identifying significant
14 mutagens in complex environmental samples, such as diesel or ambient particulate extracts
15 (Schuetzle and Lewtas, 1986).
16 The Ames Salmonella bacterial strains, used with and without S9 mix, provide
17 information about the general classes of chemicals causing mutagenic response (e.g.,
18 frameshift versus base pair substitutions, promutagen, or direct-acting mutagen). More
19 recently, tester strains that are sensitive to certain classes of compounds have been
20 developed. For example, strain TA98NR, developed by Rosenkranz and co-workers (McCoy
21 et al., 1981), is deficient in nitroreductase enzymes and therefore gives a reduced response to
22 nitro-PAHs. Strain TA98 in conjunction with strain TA98NR is most frequently used for
23 assays of environmental samples. See Rosenkranz and Mermelstein (1983) for more detailed
24 discussion.
25 Figure 3-6 illustrates the principles of bioassay-directed chemical analysis. Proper
26 sampling, storage, and extraction of environmental samples are crucial parts of the analysis.
27 Sequential extraction with increasingly polar solvents or binary solvents is most frequently
28 used to separate organic material from particles (see Section 3.2.2.3). The subsequent step,
29 preparative fractionation, achieves crude separation of extract into several less complicated
30 fractions. The two most widely used prefractionation techniques are (1) chromatography on
31 an open silica column to separate groups of compounds on the basis of polarity and
December 1994 3_61 DRAFT-DO NOT QUOTE OR CITE
-------
Sampling |-> Extraction
1 fr Preparative
| ^ Fractionation
/ t
^
/ Mass/ \ No
( Mutagenicity \ ^
\ Recovery? J
^ Yes
Modify
3rocedure
/ " \ No
/ Mutagenicity \ fr
\ High? /
^Yes
Level 1 Fractionation
t
/ Mass/ \
( Mutagenicity
\ Recovery? J
^ Yes
/ \
/ Mutagenicity ^
\ High? A
^ Yes
Level 2 Fractionation
^
Id
J'
No
>->• •
No
r-*-
Modify
'rocedure
/ " \ No
/ Mutagenicity \ ^
tYes
Chemical Analysis
i
/ ' \ No
/ Mutagenicity \ ^
\ High? i *~
tYes
Synthesize Selected
Isomers
i
/ \ No
/ Mutagenicity \ ^
Vy High? i ^
Percent Contribution Percent Contribi
To Total •+ To
Sample Mutagenicity Fraction Mutage
^ Yes
Jtion Compound
licitv Quantitation
Figure 3-6. Protocol for bioassay-directed chemical analysis.
Source: Schuetzle and Lewtas (1986).
December 1994
3-62
DRAFT-DO NOT QUOTE OR CITE
-------
1 (2) separation of compounds into acidic, basic, and neutral fractions (Schuetzle and Lewtas,
2 1986). These fractions contain hundreds of compounds and are still too complicated to be
3 characterized chemically. However, the bioassay analysis is used to decide which fractions
4 should be analyzed further.
5 Normal-phase HPLC is usually employed as a Level 1 fractionation step. The
6 reference compounds 1-nitronaphthalene and 1,6-pyrenequinone were proposed as chemical
7 markers to designate the separation of samples into nonpolar, moderately polar, and polar
8 fractions. It was found that the nonpolar fractions accounted for less than 3% of the total
9 extract mutagenicity and that the distribution of mutagenicity between moderately polar and
10 polar fractions was dependent on the sample origin (Schuetzle et al., 1985).
11 Although a multitude of compounds were identified or tentatively identified from the
12 chemical analysis of fractions from Level 1 fractionation (see Section 3.2.2.3), it soon
13 became obvious that these fractions were still too complex to allow identification of many
14 less abundant chemical mutagens. Further separation of each fraction into subfractions using
15 Level 2 fractionation was necessary. This fractionation is usually achieved by employing
16 reversed-phase HPLC or normal-phase HPLC with a different solvent system and/or
17 chromatographic column than that used in Level 1 fractionation.
18 The analytical techniques most frequently used for the characterization of HPLC
19 fractions include high resolution capillary column GC with selective detectors and/or coupled
20 with mass spectrometry (GC/MS). Comparison of the results from different mass
21 spectrometric ionization techniques (i.e., electron impact versus chemical ionization) may be
22 helpful in the identification of individual compounds. However, very polar or labile
23 compounds cannot be analyzed with GC/MS techniques because of losses resulting from
24 adsorption, thermal decomposition, and chemical interactions occurring on GC columns.
25 A direct-insertion probe coupled with high-resolution MS or MS/MS techniques has been
26 used to screen for polar compounds. One of the most promising analytical techniques for the
27 analysis of polar PAHs appears to be super critical fluid chromatography and HPLC coupled
28 with MS. See Schuetzle et al. (1985) for more detailed discussion of new analytical
29 techniques for the identification of polar mutagens. However, the identification of less
30 abundant chemical mutagens present in complex environmental mixtures is still far from
31 being complete and remains one of the most difficult analytical task.
December 1994 3.53 DRAFT-DO NOT QUOTE OR CITE
-------
1 The contribution of atmospherically formed nitroarenes to the mutagenic activities of
2 paniculate samples collected in Southern California has been also assessed (Arey et al.,
3 1988b). The distribution of direct-acting mutagenicity between the moderately polar and
4 polar fractions for paniculate samples of different origins is shown in Figure 3-7.
5 As mentioned in the preceding text, the nonpolar fractions account for less than 3 % of the
6 total mutagenicity of extracts; thus, they are not shown in this figure. According to
7 Figure 3-7, for LDD paniculate extracts, most of the direct-acting mutagenicity (65 to 75%)
8 is associated with the moderately polar fraction. In contrast, the polar fractions of extracts
9 of ambient particles, HDD particles, and wood smoke particles are said to contain more than
10 65% of the total extract mutagenicity. The compounds responsible for the mutagenicity of
11 these polar fractions have not yet been identified. It is not clear why a difference between
12 LDD and HDD should exist; no explanation has been given.
13 Up to 40% of the direct-acting mutagenicity of total extracts of LDD particles can be
14 accounted for by six nitroarenes (1-NP, 3- and 8-NF, and 1,3-, 1,6-, and 1,8-dinitropyrene),
15 eluting in the moderately polar fraction (Salmeen et al., 1984). In contrast, these nitroarenes
16 accounted for not more than 3% of the total mutagenic activity of ambient paniculate
17 samples collected at urban and suburban sites (Siak et al., 1985).
18 The contribution of atmospherically formed nitroarenes, 2-NF, and 2-NP to the
19 mutagenicity of ambient paniculate samples collected in Southern California has been
20 recently assessed (Arey et al., 1988b). This contribution could be compared to that of
21 1-nitropyrene and 3- and 8-nitrofluoranthene, regarded as direct emissions from various
22 combustion sources.
23 Table 3-18 gives the mutagen density (revertants/m3) of eight ambient samples along
24 with the calculated percentage contributions of the measured nitroarenes to this mutagenicity.
25 The samples were collected in Torrance, CA, during wintertime and in Claremont, CA,
26 during summertime. (See Section 3.4.3 for more detailed description of sampling sites and
27 maximum nitroarene concentrations.)
28 Although 2-NF was always the most abundant nitroarene measured in these ambient
29 samples, the high much less abundant, highly mutagenic 8-NF, in one instance, contributed a
30 greater fraction of the ambient activity than did 2-NF. In all remaining cases, the
31 contribution of 2-NF to the ambient mutagenicity was higher than that of the other measured
December 1994 3-64 DRAFT-DO NOT QUOTE OR CITE
-------
100
50
*P
P
MP
MP
P
P
•I^^H
MP
Ambient HD LD
Air Diesel Diesel
P
Wood
Smoke
Figure 3-7. Distribution of direct-acting mutagenicity (TA98, -S9) between moderately
polar (PM) and polar (P) fractions of extracts of particulate matter in
ambient air, wood smoke, and exhaust from heavy-duty (HD) and light-
duty (LD) motor vehicles. Adapted from Schuetzle et al. (1985).
1
2
3
4
5
6
7
8
9
10
11
nitroarenes and ranged from «1 to «5%. In contrast, 1-NP never contributed more than
0.2%.
Thus, although nitroarenes directly emitted from combustion sources may, in some
cases, contribute significantly to ambient mutagenicity (Gibson, 1983), the contribution of
atmospherically formed nitroarenes should be also recognized.
3.6 SUMMARY
Major research programs were carried out in the late 1970s and early 1980s to
ascertain the physical and chemical characteristics of emissions from diesel engines and the
biological effects of these emissions. Although new control technologies are being
December 1994
3-65 DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 3-18. CONTRIBUTION OF NITROFLUORANTHENE (NF) ISOMERS TO
THE DIRECT MUTAGENICITY OF AMBIENT PARTICIPATE EXTRACTS3
Collection Date
and Location
January 27 and 28, 1986
Torrance, CA
January 28, 1986
Torrance, CA
February 24 and 25, 1986
Torrance, CA
February 25, 1986
Torrance, CA
September 14, 1985
Claremont, CA
September 14, 1985
Claremont, CA
September 14, 1985
Claremont, CA
September 15, 1985
Claremont, CA
Mutagen
Collection Density6
Time(PST) (rev./m3)
1700-0500
0500-1700
1800-0600
0600-1800
0600-1200
1200-1800
1800-2400
0000-0600
120
120
34
73
35
15
40
20
Percent Contribution to Mutagenicity
2-NF 3-NF
2.6 1.8
1.4 -c
3.9 -
1.6 -
1.0 -
0.8 -
5.2 0.4
2.1 -
8-NF 1-NP
3.1 0.1
0.5 0.1
- 0.2
- 0.1
- 0.2
- 0.2
0.4 0.2
0.3 0.1
2-NP
0.8
0.7
1.4
0.9
0.1
0.1
0.2
0.2
Total
8.4
2.7
5.5
2.6
1.3
1.1
6.4
2.7
aFrom Arey et al. (1988b).
bTested on strain TA98(-S9).
cNot quantified or only an upper limit determined.
1 introduced into currently manufactured diesel vehicles, the effect of these changes on diesel
2 emissions is likely to be visible in the future. Currently, diesel vehicles manufactured in the
3 late 1970s and early 1980s are still on the road and, in this sense, data collected from that
4 period are still valid.
5 However, many of these data were collected using laboratory dynamometers with
6 selected new vehicles, or vehicles well-tuned to manufacturers specifications. The well-
7 controlled conditions of the dynamometer tests have many benefits but do not necessarily
8 represent vehicle emissions under real on-road conditions, and the small number of vehicles
9 tested in the laboratory is not truly representative of the distribution within the on-road
10 vehicle fleet. Although several roadway and tunnel emission measurements were performed
11 in the past, the data base on mobile sources emission rates necessary to assess the role of
12 vehicle emissions in air pollution problems is still not sufficient. More measurements carried
December 1994
3-66
DRAFT-DO NOT QUOTE OR CITE
-------
1 out under realistic on-road conditions are necessary, in particular for gaseous- and
2 particulate-phase organic compounds present in vehicle emissions.
3 Once released into the atmosphere, diesel emissions are subject to dispersion and
4 transport and, at the same time, to chemical and physical transformation into secondary
5 pollutants, which may be more harmful than their precursors. Thus, a knowledge of diesel
6 emissions at or near their sources is no longer sufficient to assess fully the impact of these
7 emissions on human health and welfare. The understanding of physical and chemical
8 changes that primary diesel emissions undergo during their transport through the atmosphere
9 is equally important. As a result of the last two decades of laboratory and ambient
10 experiments and computer modeling, a comprehensive set of data now exist concerning the
11 atmospheric loss processes and transformation of automotive emissions. However, our
12 knowledge concerning the products of these chemical transformations is still very limited.
13 Study is required to determine the products from the OH radical-initiated reactions of the
14 aromatic and aliphatic hydrocarbons, the major components of automobile emissions. The
15 atmospheric transformation products of PAHs and their oxygen-, sulfur-, and nitrogen-
16 containing analogs require study in the gaseous and adsorbed phases. In particular, the
17 reactions occurring in adsorbed phases on atmospherically relevant surfaces are poorly
18 understood and require further study. In addition, gas-to-particle conversion processes and
19 the chemical processes that lead to aerosol formation should be further investigated.
20 The quantitation of the contribution of diesel emissions to total ambient aerosol mass
21 concentration is not possible without developing a specific profile for diesel emissions, a
22 "fingerprint" that may be used in receptor source apportionment models. The existing data
23 indicate that it may be possible to use PAHs and/or alkylated PAHs, alkanes, and possibly
24 certain unique compounds to assist in distinguishing between diesel and other pollutant
25 sources. However, the available data are not adequate for use in receptor modeling, and
26 study is required to determine the profile of diesel emissions by using sampling and
27 analytical methods appropriate to the purpose of receptor modeling.
December 1994 3.57 DRAFT-DO NOT QUOTE OR CITE
-------
1 REFERENCES
2 Alsberg, T.; Stenberg, U.; Westerholm, R.; Strandell, M.; Rannug, U.; Sundvall, A.; Romert, L.;
3 Bernson, V.; Pettersson, B.; Toftgard, R.; Franzen, B.; Jansson, M.; Gustafsson, J. A.; Egeback,
4 K. E.; Tejle, G. (1985) Chemical and biological characterization of organic material from gasoline
5 exhaust particles. Environ. Sci. Technol. 19: 43-50.
6
7 Arey, J.; Zielinska, B.; Atkinson, R.; Winer, A. M.; Ramdahl, T.; Pitts, J. N., Jr. (1986) The formation of
8 nitro-PAH from the gas-phase reactions of fluoranthene and pyrene with the OH radical in the presence
9 of NOX. Atmos. Environ. 20: 2339-2345.
10
11 Arey, J.; Zielinska, B.; Atkinson, R.; Winer, A. M. (1987) Polycyclic aromatic hydrocarbon and nitroarene
12 concentrations in ambient air during a wintertime high-NO^ episode in the Los Angeles basin. Atmos.
13 Environ. 21: 1437-1444.
14
15 Arey, J.; Zielinska, B.; Atkinson, R.; Winer, A. M. (1988a) Formation of nitroarenes during ambient
16 high-volume sampling. Environ. Sci. Technol. 22: 457-462.
17
18 Arey, J.; Zielinska, B.; Harger, W. P.; Atkinson, R.; Winer, A. M. (1988b) The contribution of
19 nitrofluoranthenes and nitropyrenes to the mutagenic activity of ambient paniculate organic matter
20 collected in southern California. Mutat. Res. 207: 45-51.
21
22 Arey, J.; Zielinska, B.; Atkinson, R.; Aschmann, S. M. (1989) Nitroarene products from the gas-phase
23 reactions of volatile polycyclic aromatic hydrocarbons with OH radical and N2O5. Int. J. Chem. Kinet.
24 21: 775-799.
25
26 Atkinson, R. (1988) Atmospheric transformations of automotive emissions. In: Watson, A. Y.; Bates, R. R.;
27 Kennedy, D., eds. Air pollution, the automobile, and public health. Washington, DC: National
28 Academy Press; pp. 99-132.
29
30 Atkinson, R. (1990) Gas-phase tropospheric chemistry of organic compounds: a review. Atmos. Environ.
31 Part A 24: 1-41.
32
33 Atkinson, R.; Carter, W. P. L. (1984) Kinetics and mechanisms of the gas-phase reactions of ozone with
34 organic compounds under atmospheric conditions. Chem. Rev. 84: 437-470.
35
36 Atkinson, R.; Arey, J.; Zielinska, B.; Aschmann, S. M. (1987) Kinetics and products of the gas-phase reactions
37 of OH radicals and N2O5 with naphthalene and biphenyl. Environ. Sci. Technol. 21: 1014-1022.
38
39 Atkinson, R.; Arey, J.; Winer, A. M.; Zielinska, B. (1988) A survey of ambient concentrations of selected
40 polycyclic aromatic hydrocarbons (PAH) at various locations in California [final report]. Sacramento,
41 CA: California Air Resources Board; contract no. A5-185-32.
42
43 Atkinson, R.; Arey, J.; Zielinska, B.; Aschmann, S. M. (1990) Kinetics and nitro-products of the gas-phase
44 OH and N03 radical-initiated reactions of naphthalene-dg, fluoranthene-di0 and pyrene. Int. J. Chem.
45 Kinet. 22: 999-1014.
46
47 Bailey, J. C.; Schmidl, B.; Williams, M. L. (1990) Speciated hydrocarbon emissions from vehicles operated
48 over the normal speed range on the road. Atmos. Environ. Part A 24: 43-52.
49
50 Bayona, J. M.; Markides, K. E.; Lee, M. L. (1988) Characterization of polar polycyclic aromatic compounds
51 in a heavy-duty diesel exhaust paniculate by capillary column gas chromatography and high-resolution
52 mass spectrometry. Environ. Sci. Technol. 22: 1440-1447.
53
December 1994 3-68 DRAFT-DO NOT QUOTE OR CITE
-------
1 Behymer, T. D.; Kites, R. A. (1985) Photolysis of polycyclic aromatic hydrocarbons adsorbed on simulated
2 atmospheric particulates. Environ. Sci. Technol. 19: 1004-1006.
3
4 Behymer, T. D.; Hites, R. A. (1988) Photolysis of polycyclic aromatic hydrocarbons adsorbed on fly ash.
5 Environ. Sci. Technol. 22: 1311-1319.
6
7 Benner, B. A., Jr.; Gordon, G. E.; Wise, S. A. (1989) Mobile sources of atmospheric polycyclic aromatic
8 hydrocarbons: a roadway tunnel study. Environ. Sci. Technol. 23: 1269-1278.
9
10 Bidleman, T. F. (1988) Atmospheric processes. Environ. Sci. Technol. 22: 361-367.
11
12 Bjorseth, A.; Olufsen, B. S. (1983) Long-range transport of polycyclic aromatic hydrocarbons.
13 In: Bj0rseth, A., ed. Handbook of polycyclic aromatic hydrocarbons. New York, NY: Marcel Dekker,
14 Inc.; pp. 507-524.
15
16 Black, F.; High, L. (1979) Methodology for determining paniculate and gaseous diesel hydrocarbon emissions.
17 Warrendale, PA: Society of Automotive Engineers, Inc.; SAE technical paper no. 790442.
18
19 Boone, P. M.; Macias, E. S. (1987) Methyl alkanes in atmosperic aerosols. Environ. Sci. Technol.
20 21: 903-909.
21
22 Brorstrom, E.; Grennfelt, P.; Lindskog, A. (1983) The effect of nitrogen dioxide and ozone on the
23 decomposition of particle-associated polycyclic aromatic hydrocarbons during sampling from the
24 atmosphere. Atmos. Environ. 17: 601-605.
25
26 Calvert, J. G.; Stockwell, W. R. (1983) Acid generation in the troposphere by gas-phase chemistry. Environ.
27 Sci. Technol. 17: 428A-443A.
28
29 Campbell, R. M.; Lee, M. L. (1984) Capillary column gas chromatographic determination of nitro polycyclic
30 aromatic compounds in paniculate extracts. Anal. Chem. 56: 1026-1030.
31
32 Carey, P.; Cohen, J. (1980) Comparison of gas phase hydrocarbon emissions from light-duty vehicles equipped
33 with diesel engines. U.S. Environmental Protection Agency technical report no. CTAB/PA/80-5.
34
35 Cass, G. R. (1990) Risk assessment of diesel exhaust: 1990 and beyond. Presented at: a workshop; March;
36 Los Angeles, CA. Sacramento, CA: California Air Resources Board.
37
38 Cass, G. R.; McRae, G. J. (1983) Source-receptor reconciliation of routine air monitoring data for trace
39 metals: an emission inventory assisted approach. Environ. Sci. Technol. 17: 129-139.
40
41 Cautreels, W.; Van Cauwenberghe, K. (1978) Experiments on the distribution of organic pollutants between
42 airborne paniculate matter and the corresponding gas phase. Atmos. Environ. 12: 1133-1141.
43
44 Ciccioli, A. P.; Cecinato, A.; Brancaleoni, E.; Draisci, R.; Liberti, A. (1989) Evaluation of nitrated polycyclic
45 aromatic hydrocarbons in anthropogenic emission and air samples: a possible means of detecting
46 reactions of carbonaceous particles in the atmosphere. Aerosol Sci. Technol. 10: 296-310.
47
48 Coutant, R. W.; Brown, L.; Chuang, J. C.; Riggin, R. M.; Lewis, R. G. (1988) Phase distribution and artifact
49 formation in ambient air sampling for polynuclear aromatic hydrocarbons. Atmos. Environ.
50 22: 403-409.
51
52 Daisey, J. M.; Cheney, J. L.; Lioy, P. J. (1986) Profiles of organic paniculate emissions from air pollution
53 sources: status and needs for receptor source apportionment modeling. J. Air Pollut. Control Assoc.
54 36: 17-33.
December 1994 3.59 DRAFT-DO NOT QUOTE OR CITE
-------
1 Dolske, D. A.; Gatz, D. F. (1985) A field intercomparison of methods for the measurement of particle and gas
2 dry deposition. J. Geophys. Res. 90: 2076-2084.
3
4 Dunstan, T. D. J.; Mauldin, R. F.; Jinxian, Z.; Hipps, A. D.; Wehry, E. L.; Mamantov, G. (1989)
5 Adsorption and photodegradation of pyrene on magnetic, carbonaceous, and mineral subfractions of coal
6 stack ash. Environ. Sci. Technol. 23: 303-308.
7
8 Dutcher, J. S.; Sun, J. D.; Lopez, J. A.; Wolf, I.; Wolff, R. K.; McClellan, R. O. (1984) Generation and
9 characterization of radiolabeled diesel exhaust. Am. Ind. Hyg. Assoc. J. 45: 491-498.
10
11 Finlayson-Pitts, B. J.; Pitts, J. N., Jr. (1986) Atmospheric chemistry: fundamentals and experimental
12 techniques. New York, NY: John Wiley & Sons.
13
14 Frey, J. W.; Corn, M. (1967) Diesel exhaust particulates. Nature (London) 216: 615-616.
15
16 Galloway, J. N.; Whelpdale, D. M. (1980) An atmospheric sulfur budget for eastern North America. Atmos.
17 Environ. 14: 409-417.
18
19 Garrigues, P.; Parlanti, E.; Radke, M.; Bellocq, J.; Willsch, H.; Ewald, M. (1987) Identification of
20 alkylphenanthrenes in shale oil and coal by liquid and capillary gas chromatography and high-resolution
21 spectrofluorimetry (Shpol'skii effect). J. Chromatogr. 395: 217-228.
22
23 Gibson, T. L. (1983) Sources of direct-acting nitroarene mutagens in airborne particulate matter. Mutat. Res.
24 122: 115-121.
25
26 Gorse, R. A., Jr.; Riley, T. L.; Ferris, F. C.; Pero, A. M.; Skewes, L. M. (1983) 1-Nitropyrene concentration
27 and bacterial mutagenicity in on-road vehicle particulate emissions. Environ. Sci. Technol. 17: 198-202.
28
29 Gray, H. A. (1986) Control of atmospheric fine primary carbon particle concentration [PhD thesis]. Pasadena,
30 CA: California Institute of Technology; EQL report no. 23.
31
32 Hampton, C. V.; Pierson, W. R.; Harvey, T. M.; Updegrove, W. S.; Marano, R. S. (1982) Hydrocarbon
33 gases emitted from vehicles on the road. 1. A qualitative gas chromatography/mass spectrometry
34 survey. Environ. Sci. Technol. 16: 287-298.
35
36 Hampton, C. V.; Pierson, W. R.; Schuetzle, D.; Harvey, T. M. (1983) Hydrocarbon gases emitted from
37 vehicles on the road. 2. Determination of emission rates from diesel and spark-ignition vehicles.
38 Environ. Sci. Technol. 17: 699-708.
39
40 Hare, C. T.; Baines, T. M. (1979) Characterization of particulate and gaseous emissions from two diesel
41 automobiles as functions of fuel and driving cycle. Presented at: automotive engineering congress and
42 exposition; February-March; Detroit, MI. Warrendale, PA: Society of Automotive Engineers; SAE
43 technical paper no. 790424.
44
45 Hare, C. T.; Bradow, R. L. (1979) Characterization of heavy-duty gaseous and particulate emissions and effects
46 of fuel composition. Warrendale, PA: Society of Automotive Engineers, Inc.; SAE technical paper no.
47 790490.
48
49 Harris, G. W.; Mackay, G. L; Iguchi, T.; Schiff, H. I.; Schuetzle, D. (1987) Measurement of NC^ and HNO3
50 in diesel exhaust gas by tunable diode laser absorption spectrometry. Environ. Sci. Technol.
51 21:299-304.
December 1994 3-70 DRAFT-DO NOT QUOTE OR CITE
-------
1 Hites, R. A.; Laflamme, R. E.; Windsor, J. G., Jr. (1980) Polycyclic aromatic hydrocarbons in marine/aquatic
2 sediments: their ubiquity. In: Petrakis, L.; Weiss, F. T., eds. Petroleum in the marine environment
3 [proceedings of a symposium at the 176th meeting of the American Chemical Society; September 1978;
4 Miami Beach, FL]. Washington, DC: American Chemical Society; pp. 289-311. (Comstock, M. J., ed.
5 Advances in chemistry series: v. 185).
6
7 Horvath, H.; Kreiner, I.; Norek, C.; Preining, O.; Georgi, B. (1988) Diesel emissions in Vienna. Atmos.
8 Environ. 22: 1255-1269.
9
10 International Agency for Research on Cancer. (1989) Diesel and gasoline engine exhausts and some nitroarenes.
11 Lyon, France: World Health Organization. (IARC monographs on the evaluation of carcinogenic risks
12 to humans: v. 46).
13
14 Jager, J.; Hanus", V. (1980) Reaction of solid carrier-adsorbed polycyclic aromatic hydrocarbons with gaseous
15 low-concentrated nitrogen dioxide. J. Hyg. Epidemiol. Microbiol. Immunol. 24: 1-15.
16
17 Japar, S. M.; Szkarlat, A. C.; Gorse, R. A., Jr.; Heyerdahl, E. K.; Johnson, R. L.; Rau, J. A.; Huntzicker,
18 J. J. (1984) Comparison of solvent extraction and thermal-optical carbon analysis methods: application
19 to diesel vehicle exhaust aerosol. Environ. Sci. Technol. 18: 231-234.
20
21 Jensen, T. E.; Hites, R. A. (1983) Aromatic diesel emissions as a function of engine conditions. Anal. Chem.
22 55: 594-599.
23
24 Kamens, R. M.; Guo, Z.; Fulcher, J. N.; Bell, D. A. (1988) Influence of humidity, sunlight, and temperature
25 on the daytime decay of polyaromatic hydrocarbons on atmosperic soot particles. Environ. Sci. Technol.
26 22: 103-108.
27
28 Katz, M., Chan, C.; Tosine, H.; Sakuma, T. (1979) Relative rates of photochemical and biological oxidation
29 (in vitro) of polycyclic aromatic hydrocarbons. In: Jones, P. W.; Leber, P., eds. Polynuclear aromatic
30 hydrocarbons: third international symposium on chemistry and biology—carcinogenesis and mutagenesis;
31 October 1978; Columbus, OH. Ann Arbor, MI: Ann Arbor Science Publishers, Inc.; pp. 171-189.
32
33 Kawamura, K.; Ng, L.-L.; Kaplan, I. R. (1985) Determination of organic acids (Q-Cjo) in the atmosphere,
34 motor exhausts and engine oil. Environ. Sci. Technol. 19: 1092-1096.
35
36 Kittelson, D. B.; Dolan, D. F. (1980) Diesel exhaust aerosols. In: Willeke, K., ed. Generation of aerosols and
37 facilities for exposure experiments: [technical papers presented at the symposium on aerosol generation
38 and exposure facilities]; April 1979; Honolulu, HI. Ann Arbor, MI: Ann Arbor Science; pp. 337-359.
39
40 Kleindienst, T. E.; Shepson, P. B.; Edney, E. O.; Claxton, L. D. (1985) Peroxyacetyl nitrate: measurement of
41 its mutagenic activity using the Salmonella/mammalian microsome reversion assay. Mutat. Res.
42 157: 123-128.
43
44 Laflamme, R. E.; Hites, R. A. (1978) The global distribution of polycyclic aromatic hydrocarbons in recent
45 sediments. Geochim. Cosmochim. Acta 42: 289-310.
46
47 Lee, F. S.-C.; Schuetzle, D. (1983) Sampling, extraction, and analysis of polycyclic aromatic hydrocarbons
48 from internal combustion engines. In: Bjerseth, A., ed. Handbook of polycyclic aromatic hydrocarbons.
49 New York, NY: Marcel Dekker, Inc.; pp. 27-94.
50
51 Levsen, K. (1988) The analysis of diesel paniculate. Fresenius' Z. Anal. Chem. 331: 467-478.
52
53
December 1994 3.7! DRAFT-DO NOT QUOTE OR CITE
-------
1 Ligocki, M. P.; Pankow, J. F. (1989) Measurements of the gas/particle distributions of atmospheric organic
2 compounds. Environ. Sci. Technol. 23: 75-83.
3
4 Ligocki, M. P.; Leuenberger, C.; Pankow, J. F. (1985a) Trace organic compounds in rain—II. gas scavenging
5 of neutral organic compounds. Atmos. Environ. 19: 1609-1617.
6
7 Ligocki, M. P.; Leuenberger, C.; Pankow, J. F. (1985b) Trace organic compounds in rain—HI. particle
8 scavenging of neutral organic compounds. Atmos. Environ. 19: 1619-1629.
9
10 Lofroth, G. (1981) Comparison of the mutagenic activity in carbon paniculate matter and in diesel and gasoline
11 engine exhaust. In: Waters, M. D.; Sandhu, S. S.; Huisingh, J. L.; Claxton, L.; Nesnow, S., eds.
12 Short-term bioassays in the analysis of complex environmental mixtures II: proceedings of the second
13 symposium on the application of short-term bioassays in the fractionation and analysis of complex
14 environmental mixtures; March 1980; Williamsburg, VA. New York, NY: Plenum Press; pp. 319-336.
15 (Hollaender, A.; Welch, B. L.; Probstein, R. F., eds. Environmental science research series: v. 22).
16
17 Lowenthal, D. H.; Rahn, K. A. (1985) Regional sources of pollution aerosol at Barrow, Alaska during winter
18 1979-80 as deduced from elemental tracers. Atmos. Environ. 19: 2011-2024.
19
20 McCoy, E. C.; Rosenkranz, H. S.; Mermelstein, R. (1981) Evidence for the existence of a family of bacterial
21 nitroreductases capable of activating nitrated polycyclics to mutagens. Environ. Mutagen. 3: 421-427.
22
23 McVeety, B. D.; Kites, R. A. (1988) Atmospheric deposition of polycyclic aromatic hydrocarbons to water
24 surfaces: a mass balance approach. Atmos. Environ. 22: 511-536.
25
26 Miguel, A. H.; Pereira, P. A. P. (1989) Benzo(k)fluoranthene, benzo(ghi)perylene, and
27 indeno(l,2,3-cd)pyrene: new tracers of automotive emissions in receptor modeling. Aerosol Sci.
28 Technol. 10: 292-295.
29
30 National Research Council. (1982) Diesel cars: benefits, risks and public policy, final report of the Diesel
31 Impacts Study Committee. Washington, DC: National Academy Press.
32
33 Nielsen, T. (1984) Reactivity of polycyclic aromatic hydrocarbons toward nitrating species. Environ. Sci.
34 Technol. 18: 157-163.
35
36 Nielsen, T. (1988) The decay of benzo[a]pyrene and cyclopenteno(cd)pyrene in the atmosphere. Atmos.
37 Environ. 22: 2249-2254.
38
39 Okamoto, W. K.; Gorse, R. A., Jr.; Pierson, W. R. (1983) Nitric acid in diesel exhaust. J. Air Pollut. Control
40 Assoc. 33: 1098-1100.
41
42 Pankow, J. F.; Isabelle, L. M.; Asher, W. E. (1984) Trace organic compounds in rain. 1. Sampler design and
43 analysis by adsorption/thermal desorption (ATD). Environ. Sci. Technol. 18: 310-318.
44
45 Paputa-Peck, M. C.; Marano, R. S.; Schuetzle, D.; Riley, T. L.; Hampton, C. V.; Prater, T. J.; Skewes,
46 L. M.; Jensen, T. E.; Ruehle, P. H.; Bosch, L. C.; Duncan, W. P. (1983) Determination of nitrated
47 polynuclear aromatic hydrocarbons in paniculate extracts by capillary column gas chromatography with
48 nitrogen selective detection. Anal. Chem. 55: 1946-1954.
49
50 Pierson, W. R.; Brachaczek, W. W. (1976) Paniculate matter associated with vehicles on the road. Warrendale,
51 PA: Society of Automotive Engineers; SAE technical paper no. 760039. SAE Trans. 85: 209-227.
52
53 Pierson, W. R.; Brachaczek, W. W. (1983a) Paniculate matter associated with vehicles on the road. II. Aerosol
54 Sci. Technol. 2: 1-40.
December 1994 3-72 DRAFT-DO NOT QUOTE OR CITE
-------
1 Pierson, W. R.; Brachaczek, W. W. (1983b) Emissions of ammonia and amines from vehicles on the road.
2 Environ. Sci. Technol. 17: 757-760.
3
4 Pierson, W. R.; Brachaczek, W. W.; Hammerle, R. H.; McKee, D. E.; Butler, J. W. (1978) Sulfate emissions
5 from vehicles on the road. J. Air Pollut. Control Assoc. 28: 123-132.
6
7 Pierson, W. R.; Brachaczek, W. W.; McKee, D. E. (1979) Sulfate emissions from catalyst-equipped
8 automobiles on the highway. J. Air Pollut. Control Assoc. 29: 255-257.
9
10 Pierson, W. R.; Gorse, R. A., Jr.; Szkarlat, A. C.; Brachaczek, W. W.; Japar, S. M.; Lee, F. S.-C.;
11 Zweidinger, R. B.; Claxton, L. D. (1983) Mutagenicity and chemical characteristics of carbonaceous
12 particulate matter from vehicles on the road. Environ. Sci. Technol. 17: 31-44.
13
14 Pitts, J. N., Jr.; Grosjean, D.; Mischke, T. M.; Simmon, V. F.; Poole, D. (1977) Mutagenic activity of
15 airborne particulate organic pollutants. Toxicol. Lett. 1: 65-70.
16
17 Pitts, J. N., Jr.; Van Cauwenberghe, K. A.; Grosjean, D.; Schmid, J. P.; Fitz, D. R.; Belser, W. L., Jr.;
18 Knudson, G. B.; Hynds, P. M. (1978) Atmospheric reactions of polycyclic aromatic hydrocarbons:
19 facile formation of mutagenic nitro derivatives. Science (Washington, DC) 202: 515-519.
20
21 Pitts, J. N., Jr.; Lokensgard, D. M.; Ripley, P. S.; Van Cauwenberghe, K. A.; Van Vaeck, L.; Shaffer,
22 S. D.; Thill, A. J.; Belser, W. L., Jr. (1980) "Atmospheric" epoxidation of benzo[a]pyrene by ozone:
23 formation of the metabolite benzo[a]pyrene-4,5-oxide. Science (Washington, DC) 210: 1347-1349.
24
25 Pitts, J. N., Jr.; Sweetman, J. A.; Zielinska, B.; Winer, A. M.; Atkinson, R. (1985a) Determination of
26 2-nitrofluoranthene and 2-nitropyrene in ambient particulate organic matter: evidence for atmospheric
27 reactions. Atmos. Environ. 19: 1601-1608.
28
29 Pitts, J. N., Jr.; Zielinska, B.; Sweetman, J. A.; Atkinson, R.; Winer, A. M. (1985b) Reaction of adsorbed
30 pyrene and perylene with gaseous N2O5 under simulated atmospheric conditions. Atmos. Environ.
31 19:911-915.
32
33 Pitts, J. N., Jr.; Sweetman, J. A.; Zielinska, B.; Atkinson, R.; Winer, A. M.; Harger, W. P. (1985c)
34 Formation of nitroarenes from the reaction of polycyclic aromatic hydrocarbons with dinitrogen
35 pentoxide. Environ. Sci. Technol. 19: 1115-1121.
36
37 Pitts, J. N., Jr.; Paur, H.-R.; Zielinska, B.; Arey, J.; Winer, A. M.; Ramdahl, T.; Mejia, V. (1986) Factors
38 influencing the reactivity of polycyclic aromatic hydrocarbons adsorbed on filters and ambient POM
39 with ozone. Chemosphere 15: 675-685.
40
41 Ramdahl, T.; Zielinska, B.; Arey, J.; Atkinson, R.; Winer, A.; Pitts, J. N., Jr. (1986) Ubiquitous occurrence
42 of 2-nitrofluoranthene and 2-nitropyrene in air. Nature (London) 231: 425-427.
43
44 Rosenkranz, H. S.; Mermelstein, R. (1983) Mutagenicity and genotoxicity of nitroarenes: all nitro-containing
45 chemicals were not created equal. Mutat. Res. 114: 217-267.
46
47 Salmeen, I. T.; Pero, A. M.; Zator, R.; Schuetzle, D.; Riley, T. L. (1984) Ames assay chromatograms and the
48 identification of mutagens in diesel particle extracts. Environ. Sci. Technol. 18: 375-382.
49
50 Salmeen, I. T.; Gorse, R. A., Jr.; Pierson, W. R. (1985) Ames assay chromatograms of extracts of diesel
51 exhaust particles from heavy-duty trucks on the road and from passenger cars on a dynamometer.
52 Environ. Sci. Technol. 19: 270-273.
53
December 1994 3.73 DRAFT-DO NOT QUOTE OR CITE
-------
1 Salmeen, I. T.; Young, W. C.; Zacmanidis, P.; Ball, J. C. (1989) Evidence from Ames assays of laboratory
2 pulse-flame combustor samples that direct-acting mutagens are not an inevitable consequence of
3 hydrocarbon fuel combustion. Mutat. Res. 227: 7-11.
4
5 Schroeder, W. H.; Lane, D. A. (1988) The fate of toxic airborne pollutants: the atmosphere plays a major role
6 in their transformation and transport. Environ. Sci. Technol. 22: 240-246.
7
8 Schuetzle, D. (1983) Sampling of vehicle emissions for chemical analysis and biological testing. Environ.
9 Health Perspect. 47: 65-80.
10
11 Schuetzle, D. (1988) Organic tracers for source apportionment of combustion emissions to ambient air
12 particulates. In: proceedings of the sixth symposium on environmental analytical chemistry and source
13 apportionment in ambient and indoor atmospheres; June; Provo, UT.
14
15 Schuetzle, D.; Frazier, J. A. (1986) Factors influencing the emission of vapor and paniculate phase components
16 from diesel engines. In: Ishinishi, N.; Koizumi, A.; McClellan, R. O.; Stober, W., eds. Carcinogenic
17 and mutagenic effects of diesel engine exhaust: proceedings of the international satellite symposium on
18 lexicological effects of emissions from diesel engines; July; Tsukuba Science City, Japan. Amsterdam,
19 The Netherlands: Elsevier Science Publishers B. V.; pp. 41-63. (Developments in toxicology and
20 environmental science: v. 13).
21
22 Schuetzle, D.; Lewtas, J. (1986) Bioassay-directed chemical analysis in environmental research. Anal. Chem.
23 58: 1060A-1076A.
24
25 Schuetzle, D.; Perez, J. M. (1981) A CRC cooperative comparison of extraction and HPLC techniques for
26 diesel paniculate emissions. Presented at: 74th annual meeting of the Air Pollution Control Association;
27 June; Philadelphia, PA. Pittsburgh, PA: Air Pollution Control Association; paper no. 81-56.4.
28
29 Schuetzle, D.; Perez, J. M. (1983) Factors influencing the emissions of nitrated-polynuclear aromatic
30 hydrocarbons (nitro-PAH) from diesel engines. J. Air Pollut. Control Assoc. 33: 751-755.
31
32 Schuetzle, D.; Lee, F. S.-C.; Prater, T. J.; Tejada, S. B. (1981) The identification of polynuclear aromatic
33 hydrocarbon (PAH) derivatives in mutagenic fractions of diesel paniculate extracts. Int. J. Environ.
34 Anal. Chem. 9: 93-144.
35
36 Schuetzle, D.; Riley, T. L.; Prater, T. J.; Harvey, T. M.; Hunt, D. F. (1982) Analysis of nitrated polycyclic
37 aromatic hydrocarbons in diesel particulates. Anal. Chem. 54: 265-271.
38
39 Schuetzle, D.; Jensen, T. E.; Ball, J. C. (1985) Polar PAH derivatives in extracts of particulates: biological
40 characterization and techniques for chemical analysis. Environ. Int. 11: 169-181.
41
42 Siak, J.; Chan, T. L.; Gibson, T. L.; Wolff, G. T. (1985) Contribution to bacterial mutagenicity from
43 nitro-PAH compounds in ambient aerosols. Atmos. Environ. 19: 369-376.
44
45 Sigsby, J. E., Jr.; Tejada, S.; Ray, W.; Lang, J. M.; Duncan, J. W. (1987) Volatile organic compound
46 emissions from 46 in-use passenger cars. Environ. Sci. Technol. 21: 466-475.
47
48 Simoneit, B. R. T. (1984) Organic matter of the troposphere—HI. characterization and sources of petroleum and
49 pyrogenic residues in aerosols over the western United States. Atmos. Environ. 18: 51-67.
50
51 Slinn, W. G. N.; Hasse, L.; Hicks, B. B.; Hogan, A. W.; Lai, D.; Liss, P. S.; Munnich, K. O.; Sehmel,
52 G. A.; Vittori, O. (1978) Some aspects of the transfer of atmospheric trace constituents past the air-sea
53 interface. Atmos. Environ. 12: 2055-2087.
54
December 1994 3-74 DRAFT-DO NOT QUOTE OR CITE
-------
1 Smith, L. R. (1989) Characterization of exhaust emissions from trap-equipped light-duty diesels [final report].
2 Sacramento, CA: California Air Resources Board; contract no. A5-159-32.
3
4 Smith, L. R.; Paskind, J. (1989) Characterization of exhaust emissions from trap-equipped light-duty diesels.
5 Presented at: Passenger car meeting and exposition; September; Dearborn, MI. Warrendale, PA:
6 Society of Automotive Engineers, Inc.; SAE technical paper no. 891972.
7
8 Sonnefeld, W. J.; Zoller, W. H.; May, W. E. (1983) Dynamic coupled-column liquid chromatographic
9 determination of ambient temperature vapor pressures of poly nuclear aromatic hydrocarbons. Anal.
10 Chem. 55: 275-280.
11
12 Stockwell, W. R.; Calvert, J. G. (1983) The mechanism of the HO-SO2 reaction. Atmos. Environ.
13 17: 2231-2235.
14
15 Sweetman, J. A.; Zielinska, B.; Atkinson, R.; Ramdahl, T.; Winer, A. M.; Pitts, J. N., Jr. (1986) A possible
16 formation pathway for the 2-nitrofluoranthene observed in ambient paniculate organic matter. Atmos.
17 Environ. 20: 235-238.
18
19 Szkarlat, A. C.; Japar, S. M. (1983) Optical and chemical properties of particle emissions from on-road
20 vehicles. J. Air Pollut. Control Assoc. 33: 592-597.
21
22 Talcott, R.; Wei, E. (1977) Airborne mutagens bioassayed in Salmonella typhimurium. J. Natl. Cancer Inst.
23 58:449-451.
24
25 Thrane, K. E.; Mikalsen, A. (1981) High-volume sampling of airborne polycyclic aromatic hydrocarbons using
26 glass fibre filters and polyurethane foam. Atmos. Environ. 15: 909-918.
27
28 Tokiwa, H.; Takeyoshi, H.; Morita, K.; Takahashi, K.; Saruta, N.; Ohnishi, Y. (1976) Detection of mutagenic
29 activity in urban air pollutants. Mutat. Res. 38: 351.
30
31 Tokiwa, H.; Morita, K.; Takeyoshi, H.; Takahashi, K.; Ohnishi, Y. (1977) Detection of mutagenic activity in
32 paniculate air pollutants. Mutat. Res. 48: 237-248.
33
34 Tokiwa, H.; Kitamori, S.; Takahashi, K.; Ohnishi, Y. (1980) Mutagenic and chemical assay of extracts of
35 airborne particulates. Mutat. Res. 77: 99-108.
36
37 Tokiwa, H.; Kitamori, S.; Nakagawa, R.; Horikawa, K.; Matamala, L. (1983) Demonstration of a powerful
38 mutagenic dinitropyrene in airborne paniculate matter. Mutat. Res. 121: 107-116.
39
40 Tong, H. Y.; Karasek, F. W. (1984) Quantitation of polycyclic aromatic hydrocarbons in diesel exhaust
41 paniculate matter by high-performance liquid chromatography fractionation and high-resolution gas
42 chromatography. Anal. Chem. 56: 2129-2134.
43
44 Tong, H. Y.; Sweetman, J. A.; Karasek, F. W.; Jellum, E.; Thorsrud, A. K. (1984) Quantitative analysis of
45 polycyclic aromatic compounds in diesel exhaust paniculate extracts by combined chromatographic
46 techniques. J. Chromatogr. 312: 183-202.
47
48 Truex, T. J.; Pierson, W. R.; McKee, D. E.; Shelef, M.; Baker, R. E. (1980) Effects of barium fuel additive
49 and fuel sulfur level on diesel paniculate emissions. Environ. Sci. Technol. 14: 1121-1124.
51 Van Vaeck, L.; Van Cauwenberghe, K. (1984) Conversion of polycyclic aromatic hydrocarbons on diesel
52 paniculate matter upon exposure to ppm levels of ozone. Atmos. Environ. 18: 323-328.
December 1994 3.75 DRAFT-DO NOT QUOTE OR CITE
-------
1 Van Vaeck, L.; Van Cauwenberghe, K.; Janssens, J. (1984) The gas-particle distribution of organic aerosol
2 constituents: measurements of the volatilisation artefact in hi-vol cascade impactor sampling. Atmos.
3 Environ. 18: 417-430.
4
5 Watson, J. G.; Robinson, N. F.; Chow, J. C.; Henry, R. C.; Kim, B. M.; Pace, T. G.; Meyer, E. L.;
6 Nguyen, Q. (1990) The USEPA/DRI chemical mass balance receptor model, CMB 7.0. Environ.
7 Software 5: 38-49.
8
9 Westerholm, R. N.; Almen, J.; Li, H.; Rannug, J. U.; Egeback, K.-E.; Gragg, K. (1991) Chemical and
10 biological characterization of paniculate-, semivolatile-, and gas-phase-associated compounds in diluted
11 heavy-duty diesel exhausts: a comparison of three different semivolatile-phase samplers. Environ. Sci.
12 Technol. 25: 332-338.
13
14 White, C. M. (1983) Determination of polycyclic aromatic hydrocarbons in coal-derived materials.
15 In: Bj0rseth, A., ed. Handbook of polycyclic aromatic hydrocarbons. New York, NY: Marcel Dekker,
16 Inc.; pp. 525-616.
17
18 Williams, D. J.; Milne, J. W.; Roberts, D. B.; Kimberlee, M. C. (1989a) Paniculate emissions from 'in-use'
19 motor vehicles—I. spark ignition vehicles. Atmos. Environ. 23: 2639-2645.
20
21 Williams, D. J.; Milne, J. W.; Quigley, S. M.; Roberts, D. B.; Kimberlee, M. C. (1989b) Paniculate
22 emissions from 'in-use' motor vehicles—II. diesel vehicles. Atmos. Environ. 23: 2647-2661.
23
24 Wolff, R. K.; Henderson, R. F.; Snipes, M. B.; Griffith, W. C.; Mauderly, J. L.; Cuddihy, R. G.; McClellan,
25 R. O. (1987) Alterations in panicle accumulation and clearance in lungs of rats chronically exposed to
26 diesel exhaust. Fundam. Appl. Toxicol. 9: 154-166.
27
28 Yamasaki, H.; Kuwata, K.; Miyamoto, H. (1982) Effects of ambient temperature on aspects of airborne
29 polycyclic aromatic hydrocarbons. Environ. Sci. Technol. 16: 189-194.
30
31 Yamasaki, H.; Kuwata, K.; Kuge, Y. (1984) Determination of vapor pressure of polycyclic aromatic
32 hydrocarbons in the supercooled liquid phase and their adsorption on airborne paniculate matter. Nippon
33 Kagaku Kaishi 8: 1324-1329.
34
35 Yokley, R. A.; Garrison, A. A.; Wehry, E. L.; Mamantov, G. (1986) Photochemical transformation of pyrene
36 and benzo[a]pyrene vapor-deposited on eight coal stack ashes. Environ. Sci. Technol. 20: 86-90.
37
38 Yu, M.-L.; Kites, R. A. (1981) Identification of organic compounds on diesel engine soot. Anal. Chem.
39 53: 951-954.
40
41 Zielinska, B.; Arey, J.; Atkinson, R.; Ramdahl, T.; Winer, A. M.; Pitts, J. N., Jr. (1986) Reaction of
42 dinitrogen pentoxide with fluoranthene. J. Am. Chem. Soc. 108: 4126-4132.
43
44 Zielinska, B.; Arey, J.; Atkinson, R.; McElroy, P. A. (1988) Nitration of acephenanthrylene under simulated
45 atmospheric conditions and in solution and the presence of nitroacephenanthrylene(s) in ambient
46 particles. Environ. Sci. Technol. 22: 1044-1048.
47
48 Zielinska, B.; Arey, J.; Atkinson, R.; McElroy, P. A. (1989a) Formation of methylnitronaphthalenes from the
49 gas-phase reactions of 1- and 2-methylnaphthalene with OH radicals and N2O5 and their occurrence in
50 ambient air. Environ. Sci. Technol. 23: 723-729.
51
52 Zielinska, B.; Arey, J.; Atkinson, R.; Winer, A. M. (1989b) The nitroarenes of molecular weight 247 in
53 ambient paniculate samples collected in southern California. Atmos. Environ. 23: 223-229.
December 1994 3.75 DRAFT-DO NOT QUOTE OR CITE
-------
1 Zielinska, R. B.; Sigsby, J. E., Jr.; Tejada, S. B.; Stump, (1990) The atmospheric formation of nitroarenes and
2 their occurence in ambient air. In: Proceedings of the fourth international conference on N-substituted
3 aryl compounds: occurrence, metabolism and biological impact of nitroarenes; July 1989; Cleveland,
4 OH.
5
6 Zweidinger, R. B.; Sigsby, J. E., Jr.; Tejada, S. B.; Stump, F. D.; Dropkin, D. L.; Ray, W. D.; Duncan,
7 J. W. (1988) Detailed hydrocarbon and aldehyde mobile source emissions from roadway studies
8 Environ. Sci. Technol. 22: 956-962.
December 1994 3.77 DRAFT-DO NOT QUOTE OR CITE
-------
i 4. DOSEMETRIC FACTORS
2
3
4 4.1 INTRODUCTION
5 Diesel engine emissions consist of a complex mixture of gases, vapors, and particles
6 made up of a carbon core with a great variety of organic agents adsorbed to the surface.
7 To adequately assess dose-response relationships, it is necessary to determine the exhaust
8 component(s) responsible for the effects of interest and the relationship between exposure
9 concentration and target organ dose for the active components. Assessment of dose-response
10 relationships will permit more advanced extrapolations from high experimental exposure
11 concentrations to ambient levels and from animal test species to humans. This chapter will
12 focus on these issues.
13 A review of animal carcinogenicity studies (Chapter 7) revealed that the gaseous phase
14 alone failed to induce increases in lung tumors in any of the long-term studies in rats,
15 although in one experiment positive results were reported in mice (Stober, 1986). Because
16 of the very limited positive data for this fraction and because the potential carcinogens likely
17 to be present in this fraction (formaldehyde and acetaldehyde) induce upper respiratory tract
18 tumors, which were not seen in the whole-exhaust studies, the gaseous phase is not
19 considered separately in determining carcinogenic risk of diesel exhaust. Noncancer
20 endpoints examined in these studies (Chapter 5) also were more affected by the whole
21 exhaust, compared to the gas phase of the exhaust. This chapter, therefore, focuses on the
22 dosimetry of particles.
23 With a single exception (Iwai et al., 1986), the tumors reported in the diesel exhaust
24 inhalation studies reviewed in Chapter 6 all occurred in the lungs. Although paniculate
25 matter deposited in the conducting airways of the respiratory tract is expectorated and
26 swallowed or expelled, and material deposited on the pelt of the animals ingested as a result
27 of preening, little information exists regarding possible uptake of carcinogens from the
28 gastrointestinal tract. Although organics adsorbed onto diesel exhaust particles may be
29 absorbed in the gastrointestinal tract (Bond et al., 1986), only lung tumors are seen in studies
30 in which animals are chronically exposed to diesel exhaust. Therefore, dosimetric
31 considerations will be confined to the respiratory tract.
December 1994 4_1 DRAFT-DO NOT QUOTE OR CITE
-------
1 The relative importance of the inorganic core of the particle vs the particle-adsorbed
2 organics in the causation of noncancer effects has not been investigated sufficiently to attain a
3 full understanding. The relative contributions of the carbon core versus that of the adsorbed
4 organics to carcinogenesis has been studied, and current interpretation of the contributions
5 indicates that the carbon core may play a key role. This subject is discussed in considerable
6 detail in Chapter 10. However, because both components may be involved in the
7 carcinogenic process, dosimetric variables relating to each of them are pertinent. The
8 concept of bioavailability relative to the particle-adsorbed organics in diesel exhaust is also
9 difficult to assess accurately because of the many uncertainties inherent in determining the
10 actual dose of these compounds following inhalation of diesel exhaust. This uncertainty may
11 be caused, in part, by the fact that the organic compounds may be unevenly distributed on
12 the soot particles. Furthermore, dose determination for inhaled compounds remains
13 problematic among toxicologists (Dahl et al., 1991).
14 The dosimetric aspects considered will include deposition in the conducting airways and
15 alveolar regions, normal particle clearance mechanisms and rates in both regions, clearance
16 rates during lung overload, elution of organics from the particles, particle transport to
17 extraalveolar sites, and the interrelationships of these factors in determining the target organ
18 dose.
19
20
21 4.2 REGIONAL DEPOSITION OF INHALED PARTICLES
22 The regional deposition of paniculate matter in the respiratory tract is dependent on the
23 interaction of a number of factors, including respiratory tract anatomy (airway dimensions
24 and branching configurations), ventilatory characteristics (breathing mode and rate,
25 ventilatory volumes and capacities), physical processes (diffusion, sedimentation, impaction,
26 and interception), and the physicochemical characteristics (particle size, shape, and density)
27 of the inhaled particles. Regional deposition of particulate material is usually expressed as
28 deposition fraction of the total particles or mass inhaled and may be represented by the ratio
29 of the particles or mass deposited in a specific region to the number or mass of particles
30 inspired. The factors affecting deposition in these various regions and their importance in
31 understanding the fate of inhaled diesel exhaust particulate matter is discussed in the
December 1994 4-2 DRAFT-DO NOT QUOTE OR CITE
-------
1 following sections. It is beyond the scope of this document to present a comprehensive
2 account of the complexities of respiratory mechanics, physiology, and toxicology. Where
3 appropriate, the reader is referred to publications that provide a more in-depth treatment of
4 these topics (Weibel, 1963; Brain and Mensah, 1983; Raabe et al., 1988).
5
6 4.2.1 Physical Processes, Physiological/Anatomical Considerations, and
7 Particle Characteristics
8 Deposition of particles may occur through several processes or combinations thereof,
9 including diffusion, sedimentation (gravitational settling), interception, electrostatic
10 precipitation, and impaction. It is important to appreciate that these processes are not
11 necessarily independent but may, in some instances, interact with one another such that total
12 deposition in the respiratory tract resulting from these processes may be less than the
13 calculated probabilities for deposition by the individual processes (Raabe, 1982). Depending
14 on the particle size and mass, varying degrees of deposition may occur in the nasopharyn-
15 geal, tracheobronchial, and alveolar regions of the respiratory tract.
16 Upon inhalation of particulate matter such as diesel exhaust, deposition will occur
17 throughout the respiratory tract. Because of high air-flow velocities and abrupt directional
18 changes in the nasopharyngeal and tracheobronchial regions, inertial impaction is a primary
19 deposition mechanism (especially for particles larger than 2.5 /*m mass median diameter
20 [MMD]). Although inertial impaction is a prominent process for deposition of larger
21 particles in the tracheobronchial region, it is of minimal significance as a determinant of
22 regional deposition patterns for diesel exhaust particles, with an MMD less than 1 pm and
23 small aspect ratio.
24 Because the MMD of diesel exhaust particles is generally less than 1 /mi, they are
25 subject to deposition in the alveolar region. Based on animal data regarding the site of origin
26 of diesel exhaust-induced tumors, particle deposition in the alveolar region may be of greatest
27 concern relative to the carcinogenic potential of diesel particulate matter and/or the adsorbed
28 organics. However, such data for humans is not available. For such small particles,
29 diffusion would be especially prevalent in this region, whereas sedimentation would become
30 less significant, especially for particles of MMD < 0.5 /im.
December 1994 4.3 DRAFT-DO NOT QUOTE OR CITE
-------
1 Respiratory tract anatomy and ventilatory characteristics are crucial in determining
2 regional deposition patterns and are also responsible for interspecies and interindividual
3 variability in both deposition of particulate matter and inhaled dose. The variability in size
4 and branching configurations of conducting airways is an important determinant of
5 interspecies variability in deposited dose. Because of the anatomical complexity and
6 variability in ventilatory patterns, a precise categorization of air-flow dynamics for a given
7 species or for a specific portion of the respiratory tract is difficult. For more extensive
8 discussions of deposition processes, refer to reviews by Morrow (1966), Raabe (1982), U.S.
9 Environmental Protection Agency (1982), Phalen and Oldham (1983), Lippmann and
10 Schlesinger (1984), and Raabe et al. (1988).
11 Exposure to whole diesel exhaust will also result in inhalation of gas-phase components
12 such as formaldehyde, acrolein, and sulfur dioxide, all of which have been demonstrated to
13 be sensory irritants. It is also known that these irritants affect respiratory rates (Kane and
14 Alarie, 1978, 1979). The sensory irritant-induced reduction of respiratory rate is mediated
15 through stimulation of free nerve endings of the afferent trigeminal nerve (Ulrich et al.,
16 1972). This physiologic reflex response has been shown to be a concentration-dependent
17 response (Alarie, 1966, 1973; Kane and Alarie, 1978). Several studies have also shown that
18 mice appear to be more responsive to sensory irritants relative to alteration of respiratory
19 patterns (Alarie, 1973; Kane and Alarie, 1978, 1979). However, only very low levels of
20 these irritants are present in diesel engine exhaust; consequently, their significance in
21 affecting delivered dose through changes in respiration may be small.
22
23 4.2.2 Species Variability in Regional Dose
24 The variability in the anatomy of conducting airways among species results in
25 interspecies variability of inhaled dose. Because different species breathing the same aerosol
26 will not receive the same dose to the respiratory tract, it is generally accepted that exposure
27 concentration is not an accurate description of respiratory tract dose (Brain and Mensah,
28 1983).
29 The deposition of inhaled diesel particles in the respiratory tract of humans and
30 mammalian species has been reviewed by Schlesinger (1985). He showed that physiological
31 differences in the breathing mode for humans (nasal or oronasal breathers) and experimental
December 1994 4-4 DRAFT-DO NOT QUOTE OR CITE
-------
1 animals (obligatory nose breathers), combined with different airway geometries, resulted in
2 significant differences in lower respiratory tract deposition for larger particles (> 1 jun).
3 In particular, a much lower fraction of inhaled larger particles is deposited in the alveolar
4 region of the rat compared with humans. However, relative deposition of the much smaller
5 diesel exhaust particles was not affected as much by the differences among species as was
6 demonstrated by Xu and Yu (1987). These investigators modeled the deposition efficiency of
7 inhaled diesel exhaust particles in rats, hamsters, and humans based on lung model
8 calculations of the models of Schum and Yeh (1980) and Weibel (1963). In Figure 4-1,
9 relative deposition patterns in the lower respiratory tract (trachea = generation 1; alveoli =
10 generation 23) are very similar among hamsters, rats, and humans. Variations in alveolar
11 deposition of diesel exhaust particles over one breathing cycle in these different species were
12 predicted to be within 30% of one another. Xu and Yu (1987) attributed this similarity to
13 the fact that deposition of the submicron diesel particles is dominated by diffusion rather than
14 sedimentation or impaction. Although these data assumed nose-breathing by humans, the
15 results would not be very different for mouth-breathing because of the low filtering capacity
16 of the nose for particles smaller than 0.1 /im.
17 However, for dosimetric calculations and modeling, it would be of much greater
18 importance to consider the actual dose deposited per unit surface area of the respiratory tract
19 rather than the relative deposition efficiencies per lung region. Table 4-1 compares the
20 predicted deposited doses of diesel exhaust particles inhaled in 1 min for the three species,
21 based on either the total lung volume, the surface area of all lung airways, or the surface
22 area of the epithelium of the alveolar region only. In Table 4-1, the absolute deposited dose
23 is lower in humans than in the two rodent species as a result of the greater respiratory
24 exchange rate in rodents and smaller size of the rodent lung. Such differences in the
25 absolute deposited dose in relevant target areas are highly important and have to be
26 considered when extrapolating the results from diesel exhaust exposure studies in animals to
27 humans. The differences are less on a surface area basis than on a lung volume basis
28 (Table 4-1). This is due to larger alveolar diameters in humans and concomitantly lower
29 surface area per unit of lung volume.
December 1994 4.5 DRAFT-DO NOT QUOTE OR CITE
-------
0
4 8 12 16 20
Generation Number
24
Figure 4-1. Deposition distribution patterns of inhaled diesel exhaust particles in the
airways of different species.
Source: Xu and Yu (1987).
1 4.3 RESPIRATORY TRACT CLEARANCE RATES
2 4.3.1 Tracheobronchial Clearance
3 The dynamic relationship between deposition and clearance is responsible for
4 determining lung burden at any point in time. Clearance of highly insoluble particles from
5 the tracheobronchial region is mediated primarily by mucociliary transport and is a more
6 rapid process than those operating in alveolar regions. Mucociliary transport (often referred
7 to as the mucociliary escalator) is accomplished by the rhythmic beating of cilia that line the
8 respiratory tract from the trachea through the terminal bronchioles. This movement propels
9 the mucous layer containing deposited particles (or particles within AMs) toward the larynx.
10 Clearance rate by this system is determined primarily by the flow velocity of the mucus,
11 which is greater in the proximal airways and decreases distally. These rates also exhibit
12 interspecies and individual variability. Considerable species-dependent variability in
December 1994
4-6
DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 4-1. PREDICTED DOSES OF INHALED DIESEL EXHAUST PARTICLES
PER MINUTE BASED ON TOTAL LUNG VOLUME (M), TOTAL AIRWAY
SURFACE AREA (M^, OR SURFACE AREA IN ALVEOLAR REGION (M2)
M Mj M2
Species (10~3 /xg/min-cm3) (10~6 ptg/min-cm2) (10~6 /ig/mincm2)
Hamsters 3.548 3.088 2.382
Fischer rat 3.434 3.463 2.608
Human 0.249 1.237 0.775
M and Mj = mass of particles deposited in total lung.
M2 = mass of particles deposited in the alveolar region only.
Based on the following conditions: (1) MMAD = 0.2 ^m; a = 1.9; 0 = 0.3; and p = 1.5 g/cm3; (2) particle
concentration = 1 mg/m3; and (3) nose-breathing.
Source: Xu and Yu (1987).
1 tracheobronchial clearance has been reported, with dogs generally having faster clearance
2 rates than guinea pigs, rats, or rabbits (Felicetti et al., 1981). The half-times (t1/2) values for
3 tracheobronchial clearance of relatively insoluble particles are usually on the order of hours:
4 those for alveolar clearance may be hundreds of days in humans and dogs. The clearance of
5 paniculate matter from the tracheobronchial region is generally recognized as being biphasic
6 or multiphasic (Raabe, 1982). Some studies have shown that particles are cleared from
7 large, intermediate, and small airways with t1/2 of 0.5, 2.5, and 5 h, respectively. However,
8 recent reports have indicated that clearance from conducting airways is biphasic and that the
9 long-term component for humans may take much longer for a significant fraction of particles
10 deposited in this region and may not be complete within 24 h, as generally believed
11 (Stahlhofenetal., 1990).
12 Although most of the particulate matter cleared from the tracheobronchial region will
13 ultimately be swallowed, the contribution of this fraction relative to carcinogenic potential is
14 unclear. With the exception of conditions of impaired bronchial clearance, the desorption
15 t1/2 for particle-associated organics is generally longer than the tracheobronchial clearance
16 times, thereby making uncertain the importance of this fraction relative to carcinogenesis in
17 the respiratory tract (Pepelko, 1987). Gerde et al. (1991) showed that for low-dose
18 exposures, particle-associated PAHs were rapidly released, thereby suggesting retained
December 1994 4.7 DRAFT-DO NOT QUOTE OR CITE
-------
1
2
3
4
5
6
1
2
3
4
5
6
particle-associated PAHs may be of lesser importance in tumorigenic responses than
originally believed. The relationship between the early clearance of insoluble particles (4
aerodynamic diamter) from the tracheobronchial regions and their longer term clearance from
the alveolar region is illustrated in Figure 4-2.
Tracheobronchial
Deposition
Alveolar Deposition
20 40 60
Hours after Inhalation
100
Figure 4-2. Clearance of insoluable particles depositied in tracheobronchial and alveolar
regions.
Source: Cuddihy and Yeh, 1986.
Cuddihy and Yeh (1986) reviewed respiratory tract clearance of particles inhaled by
humans. Depending on the type of particle (ferric oxide, teflon discs, or albumin
microspheres), the technique employed, and the anatomic region (midtrachea, trachea, or
main bronchi), particle velocity (moved by mucociliary transport) ranged from 2.4 to
21.5 mm/min. The highest velocities were recorded for midtracheal transport, and the
lowest were for main bronchi. In one study, an age difference was noted for tracheal
December 1994
4-8
DRAFT-DO NOT QUOTE OR CITE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
mucociliary transport velocity (5.8 mm/min for individuals less than 30 years of age and
10.1 mm/min for individuals over 55 years of age).
Cuddihy and Yeh (1986) described salient points to be considered when estimating
particle clearance velocities from tracheobronchial regions: respiratory tract airway
dimensions, calculated inhaled particle deposition fractions for individual airways, and
thoracic clearance measurements. Predicted clearance velocities for the trachea and main
bronchi were found to be similar to those experimentally determined for inhaled radiolabeled
particles but not for intratracheally instilled particles. The velocities observed for inhalation
studies were generally lower than those of instillation studies. Figure 4-3 illustrates a
comparison of the short-term clearance of inhaled particles by human subjects and the model
predictions for this clearance.
1.0-
"*•>
"w
a
CD
Q
_o
£
t_
0
H
'+3
jc
*5
c
_g
S.
-
0.8-
0.6-
-
0.4-
0.2-
0_
^>. / nange 01 i nree
\>v^ / Measurements
\r^—
_
7
/
/
/
^ Model
Same
~~
^"^^^•™^^^™
— • — _j
Projection
as Lower
Limit of Range
1 i i i I I I I I I
0 20 40 60 80
i i i
100 1,
Hours after Inhalation
Figure 4-3. Short-term thoracic clearance of inhaled particles as determined by model
prediction and experimental measurement.
Source: Cuddihy and Yeh, 1986 (from Stahlhofen et al., 1980).
December 1994
4-9
DRAFT-DO NOT QUOTE OR CITE
-------
1 Exposure of F344 rats to whole diesel exhaust at soot concentrations of 0.35, 3.5, or
2 7.0 mg/m3 for up to 24 mo did not significantly alter tracheal mucociliary clearance of
3 98mTc-macroaggregated albumin instilled into the trachea (Wolff et al., 1987). The
4 assessment of tracheal clearance was determined by measuring the amount of material
5 retained 1 h after instillation. The authors stated that measuring retention would yield
6 comparable estimates of clearance efficiency compared to measuring the velocity for
7 transport of the markers in the trachea. The results of this study were in agreement with
8 similar findings of unaltered tracheal mucociliary clearance in rats exposed to diesel exhaust
9 (0.21, 1.0 or 4.4 mg/m3) for up to 4 mo (Wolff and Gray, 1980). However, the 1980 study
10 by Wolff and Gray as well as an earlier study by Battigelli et al. (1966) showed that acute
11 exposure to high concentrations of diesel exhaust soot (1.0 and 4.4 mg/m3 in the study by
12 Wolff et al. and 8 to 17 mg/m3 in the study by Battigelli et al.) produced transient reductions
13 in tracheal mucociliary clearance. Battigelli et al. (1966) also noted that the compromised
14 tracheal clearance was not observed following cessation of exhaust exposure.
15 The fact that tracheal clearance does not appear to be significantly impaired or is
16 impaired only transiently following exposure to high concentrations of diesel soot is
17 consistent with the limited pathological effects observed in the tracheobronchial region of the
18 respiratory tract in experimental animals. However, the apparent retention of a fraction of
19 the deposited dose in the airways is cause for some concern regarding possible carcinogenic
20 effects in this region especially in light of the results from simulation studies by Gerde et al.
21 (1991) which suggested that release of poly cyclic aromatic hydrocarbons (PAHs) from
22 particles may occur within minutes and at the site of initial deposition. Moreover,
23 impairment of mucociliary clearance function as a result of exposure to either occupational or
24 environmental respiratory tract toxicants or to cigarette smoke will significantly enhance the
25 retention of particles in this region. For example, Vastag et al. (1986) demonstrated that not
26 only smokers with clinical symptoms of bronchitis but also symptom-free smokers have
27 significantly reduced mucocilairy clearance rates. Such impaired clearance function could
28 conceivably have a significant impact on effects of deposited diesel exhaust particles in the
29 conducting airways.
30
31
December 1994 4-10 DRAFT-DO NOT QUOTE OR CITE
-------
1 4.3.2 Clearance from the Alveolar Region
2 4.3.2.1 Alveolar Clearance in Humans
3 A number of investigators have reported on the alveolar clearance kinetics of human
4 subjects. Bohning et al. (1980) examined alveolar clearance in eight humans who had
5 inhaled <0.4 mg of 85Sr-labeled polystyrene particles (3.6 ± 1.6 ^im diameter). A double-
6 exponential model best described the clearance of the particles and provided ti/2 values of
7 29 ± 19 days and 298 ±114 days for short- and long-term phases, respectively. It was
8 noted that of the particles deposited in the alveolar region, 75 ± 13% were cleared via the
9 long-term phase. Alveolar retention t1/2 values of 330 and 420 days were reported for
10 humans who had inhaled aluminosilicate particles (Bailey et al., 1982).
11 Quantitative data on clearance rates in humans having large lung burdens of paniculate
12 matter is lacking. Bohning et al. (1982) and Cohen et al. (1979), however, did provide
13 evidence for slower clearance in smokers, and Freedman and Robinson (1988) reported
14 slower clearance rates in individuals that had mild pneumoconiosis. Although information on
15 particle burden and particle overload relationships in humans is much more limited than for
16 experimental animal models, inhibition of clearance does seem to occur. Stober et al. (1967)
17 estimated a clearance t1/2 of 4.9 years in coal miners with nil or slight silicosis, based upon
18 post mortem lung burdens. The lung burdens ranged from 2 to 50 mg/g of lung or more,
19 well above the value for which sequestration is observed in the rat. Furthermore, impaired
20 clearance resulting from smoking or exposure to other respiratory toxicants may increase the
21 possibility of an enhanced particle accumulation effect resulting from exposure to other
22 particle sources such as diesel exhaust.
23
24 4.3.2.2 Alveolar Clearance in Animals
25 Normal alveolar clearance rates in animals have been reported by a number of
26 investigators. Because the rat is the species for which experimentally induced lung cancer
27 data are available and for which most clearance data exist, it is the species most often used
28 for assessing human risk and reviews of alveolar clearance studies have been generally
29 limited to this species.
30 Chan et al. (1981) subjected 24 male F344 rats to nose-only inhalation of diesel exhaust
31 (6 mg/m3) labeled with 131Ba or 14C for 40 to 45 min and assessed total lung deposition,
December 1994 4_H DRAFT-DO NOT QUOTE OR CITE
-------
1 retention, and elimination. Based on radiolabel inventory, the deposition efficiency in the
2 respiratory tract was 15 to 17%. Measurement of 131Ba label in the feces during the first
3 4 days following exposure indicated that 40% of the deposited diesel exhaust particles were
4 eliminated via mucociliary clearance. Clearance of the particles from the lower respiratory
5 tract followed a two-phase elimination process consisting of a rapid (ty2 of 1 day) elimination
6 by mucociliary transport, and a slower (t\/2 of 62 days) macrophage-mediated alveolar clear-
7 ance. This study provided data for normal alveolar clearance rates of diesel exhaust particles
8 not affected by prolonged exposure or particle overloading.
9 Several studies have investigated the effects of exposure concentration on the alveolar
10 clearance of diesel exhaust particles by laboratory animals.
11 Wolff et al. (1986, 1987) provided clearance data (ti/2) and lung burden values for
12 F344 rats exposed to diesel exhaust for 7 h/day, 5 days/week for 24 mo. Exposure
13 concentrations of 0.35, 3.5, and 7.0 mg of soot/m3 were employed in this whole
14 body-inhalation exposure experiment. Intermediate (hours-days) clearance of 67Ga2O3
15 particles (30 min, nose-only inhalation) was assessed after 6, 12, 18, and 24 mo of exposure
16 at all of the diesel exhaust concentrations. A two-component function described the
17 clearance of the administered radiolabel:
18
F(t) = A exp(-0.693 t/r^ + B exp(-0.693 t/r2) ,
19
20 where F(t) was the percentage retained throughout the respiratory tract, A and B were the
21 magnitudes of the two components (component A representing the amount cleared from
22 nasal, lung, and gastrointestinal compartments and component B representing intermediate
23 clearance from the lung compartment), and TJ and T2 were the half-times for the A and
24 B compartments, respectively. The early retention half-times (TJ), representing clearance
25 from primary, ciliated conducting airways, were similar for rats in all exposure groups at all
26 time points except for those in the high exposure (7.0-mg/m3) group following 24 mo of
27 exposure where the clearance rate was faster than that of the controls. Significantly longer
28 B compartment retention half-times, representing the early clearance from nonciliated
29 passages such as alveolar ducts and alveoli, were noted after as few as 6 mo exposure to
30 diesel exhaust at 7.0 mg/m3 and 18 mo exposure to 3.5 mg/m3.
December 1994 4-12 DRAFT-DO NOT QUOTE OR CITE
-------
1 Nose-only exposures to 134Cs fused aluminosilicate particles (FAP) were used to assess
2 long-term (weeks-months) clearance. Following 24-mo exposure to diesel exhaust, long-term
3 clearance of 134Cs-FAP was significantly (p < 0.01) altered in the 3.5 (cumulative exposure
4 [C x T] of 11,760 mg-h/m3) and 7.0 mg/m3 (C X T = 23,520 mg-h/m3) exposure groups
5 (ty2 of 264 and 240 days, respectively) relative to the 0.35 mg/m3 and control groups (ty of
6 81 and 79 days, respectively). Long-term clearance represents the slow component of
7 particle removal from the alveoli. The decreased clearance correlated with the greater
8 particle burden in the lungs of the 3.5- and 7.0-mg/m3 exposure groups. Based on these
9 findings, the cumulative exposure of 11,760 mg-h/m3 represented a particle overload
10 condition resulting in compromised alveolar clearance mechanisms.
11 Heinrich et al. (1986) exposed rats 19 h/day, 5 days/week for 2.5 years to diesel
12 exhaust at a particle concentration of about 4 mg/m3. This is equal to a C x T of
13 53,200 mg-h/m3. The lung particle burden was sufficient to result in particle overload
14 conditions and impairment of clearance mechanisms. With respect to the organic matter
15 adsorbed onto the particles, the authors estimated that over the 2.5-year period, 6 to 15 mg
16 of particle-bound organic matter had been deposited and was potentially available for
17 biological effects. This estimation was based on the analysis of the diesel exhaust used in the
18 experiments, values for rat ventilatory functions, and estimates of deposition and clearance.
19 Accumulated burden of diesel soot particles in the lungs following an 18-mo, 7 h/day,
20 5 days/week exposure to diesel exhaust was reported by Griffis et al. (1983). Male and
21 female F344 rats exposed to 0.15, 0.94, or 4.1 mg soot/m3 were sacrificed at 1 day and 1,
22 5, 15, 33, and 52 weeks after exposure, and diesel soot was extracted from lung tissue
23 dissolved in tetramethylammonium hydroxide. Following centrifugation and washing of the
24 supernatant, diesel soot content of the tissue was quantitated using spectrophotometric
25 techniques. The analytical procedure was verified by comparing results to recovery studies
26 using known amounts of diesel soot with lungs of unexposed rats. Long-term retention for
27 the 0.15- and 0.94-mg/m3 groups had estimated half-times of 87 ± 28 and 99 ± 8 days,
28 respectively. The retention t1/2 for the 4.1-mg/m3 exposure group was 165 ± 8 days, which
29 was significantly (p < 0.0001) greater than those of the lower exposure groups. The 18-mo
30 exposures to 0.15 or 0.96 mg/m3 levels of diesel exhaust (C x T equivalent of 378 and
31 2,368 mg-h/m3, respectively) did not affect clearance rates, whereas the exposure to the
December 1994 4_13 DRAFT-DO NOT QUOTE OR CITE
-------
1 4.1-mg/m3 concentration (C x T = 10,332 mg-h/m3) resulted in significant lung soot
2 burdens and impaired clearance.
3 In a subsequent study (Lee et al. 1983), a three-phase model was used to describe the
4 clearance of diesel exhaust particles (7 mg/m3 for 45 min or 2 mg/m3 for 140 min) by F-344
5 rats (24 per group) exposed by nose-only inhalation with no apparent particle overload in the
6 lungs. The exposure protocols provided comparable total doses based on a 14C radiolabel.
7 The 14CO2 resulting from combustion of 14C-labeled diesel fuel was removed by a diffusion
8 scrubber to avoid erroneous assessment of 14C intake by the animals. Retention of the
9 radiolabeled particles was determined up to 335 days after exposure and resulted in the
10 derivation of a three-phase clearance of the particles. The resulting retention t1/2 values for
11 the three-phases were 1,6, and 80 days. The three clearance phases are taken to represent
12 removal of tracheobronchial deposits by the mucociliary escalator, removal of particles
13 deposited in the respiratory bronchioles, and alveolar clearance, respectively. Species
14 variability in clearance of diesel exhaust particles was also demonstrated by the fact that
15 Hartley guinea pigs exhibited negligible alveolar clearance from Day 10 to Day 432
16 following a 45-min exposure to a diesel particle concentration of 7 mg/m3. Initial deposition
17 efficiency (20 ±2%) and short-term clearance were, however, similar to that for rats.
18 Lung clearance in male F344 rats pre-exposed to diesel exhaust at 0.25 or 6 mg/m3
19 20 h/day, 7 days/week for periods lasting from 7 to 112 days was studied by Chan et al.
20 (1984). Following this pre-exposure protocol, rats were subjected to 45-min nose-only
21 exposure to 14C-diesel exhaust and alveolar clearance of radiolabel monitored for up to
22 1 year. First order clearance for the two-compartment model, R(t) = Ae'ult+Be~u2t, yielded
23 alveolar retention t1/2 values of 166 and 562 days for rats preexposed to 6.0 mg/m3 for 7 and
24 62 days, respectively. These values were significantly (p < 0.05) greater than the retention
25 t1/2 of 77 ± 17 days for control rats. The same retention values for rats of the 0.25-mg/m3
26 groups were 90 ± 14 and 92 ± 15 days, respectively. The two-compartment model
27 represents overall clearance of the tracer particles, even if some of the particles were
28 sequestered in particle-laden macrophages with substantially slower clearance rates. A lung
29 retention model for preexposed rats was developed that accounts for mucociliary clearance,
30 an active clearance phase by alveolar macrophages (AMs), and a residual fraction
31 representing macrophage aggregates with limited clearance capabilities. Clearance was
December 1994 4-14 DRAFT-DO NOT QUOTE OR CITE
-------
1 shown to be dependent on the initial burden of particles and, therefore, the clearnace t,/2
2 would increase in higher exposure scenarios. This study emphasizes the importance of
3 particle overloading of the lung, and the ramifications on clearance of particles. Based on
4 these data, a particle overload effect was demonstrated for both the high and low exposure
5 levels (equivalent to C x T dose of 840 and 7,440 mg-h/m3).
6 Long-term alveolar clearance rates of particles in various laboratory animals and
7 humans have been reviewed by Pepelko (1987). Although retention t1/2 vary both among and
8 within species and are also dependent on the physicochemical properties of the inhaled
9 particles, the retention ti/2 for humans is generally much longer (>8 mo) than the average
10 retention t1/2 of 60 days for rats.
11
12 4.3.2.3 Lung Burden and Pulmonary Overload Resulting in Impaired Clearance
13 The fact that particle overload impairs alveolar clearance is well documented for
14 animals. Furthermore, particle overload appears to be an important factor in the diesel
15 emissions-induced pulmonary carcinogenicity observed in animals. Some of the studies
16 described in more detail in Section 4.3.2.3 provide data affirming impaired alveolar
17 clearance resulting from an increased lung burden and particle overload. A study by Griffis
18 et al. (1983) demonstrated that exposure (7 h/day, 5 days/week) of rats to whole diesel
19 exhaust at concentrations of 0.15, 0.94, or 4.1 mg/m3 for 18 mo resulted in lung burdens of
20 35, 220, and 1,890 pig/g of lung tissue, respectively. The alveolar clearance of those rats
21 with the highest lung burden (1,890 /*g/g of lung) was impaired as determined by a
22 significantly greater (p < 0.0001) retention t1/2 for diesel exhaust particles. This is
23 reflected in the greater lung burden/exposure concentration ratio at the highest exposure
24 level. Similarly, in the study by Chan et al. (1984) rats exposed for 20 h/day, 7 days/week
25 to whole diesel exhaust (6 mg/m3) for 112 days had a total lung particle burden of 11.8 mg,
26 with no alveolar particle clearance being detected over 1 year.
27 Muhle et al. (1990) indicated that overloading of rat lungs occurred when lung particle
28 burdens reached 0.5 to 1.5 mg/g of lung tissue and that clearance mechanisms were totally
29 compromised at lung particle burdens greater than 10 mg/g for particles with a specific
30 density close to one.
December 1994 4_15 DRAFT-DO NOT QUOTE OR CITE
-------
1 Pritchard (1989), utilizing data from a number of diesel exhaust exposure studies,
2 examined alveolar clearance in rats as a function of cumulative exposure. The resulting
3 analysis noted a significant increase in retention t1/2 values at exposure rates above
4 10 mg-h/m3 and also showed that normal lung clearance mechanisms appeared to be
5 compromised as the lung soot burden approached 0.5 mg/g of lung.
6 Morrow (1988) has proposed that the condition of dust overloading in the lungs is
7 caused by a loss in the mobility of particle engorged AMs and that such an impediment is
8 related to the cumulative volumetric load of particles in the AM. Morrow (1988) has further
9 estimated that the clearance function of an AM may be completely impaired when the particle
10 burden in the AM is of a volumetric size equivalent to about 60% of the normal volume of
11 the AM. Oberdorster and co-workers (1991) assessed the alveolar clearance of smaller
12 (3.3 /*m diameter) and larger (10.3 ^m diameter) polystyrene particles, the latter of which
13 are volumetrically equivalent to about 60% of the average normal volume of a rat AM, after
14 intratracheal instillation into the lungs of rats. Whereas both sizes of particles were found to
15 be phagocytized by AM within a day after deposition and the smaller particles were cleared
16 at a normal rate, only minimal lung clearance of the larger particles was observed over an
17 approximately 200-day postinstillation period, thus supporting the volumetric overload
18 hypothesis.
19 Animal studies have revealed that impairment of alveolar clearance can occur following
20 chronic exposure to diesel exhaust particulate matter (Griffis et al., 1983; Wolff et al., 1987;
21 Vostal et al., 1982; Lee et al., 1983) or a variety of other diverse aerosols (Lee et al., 1986,
22 1988; Ferin and Feldstein, 1978; Muhle et al. 1990). Because high lung burdens of different
23 types of particles result in diminution of normal lung clearance kinetics or in what is now
24 called "particle overloading", this effect appears to be more related to the mass and/or
25 volume of particles in the lung than to the nature of the particles per se. Regardless, as
26 pointed out by Morrow (1988), particle overloading in the lung modifies the dosimetry for
27 particles in the lung and thereby can alter toxicologic responses.
28 Although quantitative data are limited regarding lung overload associated with impaired
29 alveolar clearance in humans, impairment of clearance mechanisms does appear to occur and
30 at a lung burden generally in the range reported to impair clearance in rats. Stober et al.
31 (1967) in their study of coal miners, reported lung particle burdens of 2 to 50 mg/g lung
December 1994 4_16 DRAFT-DO NOT QUOTE OR CITE
-------
1 tissue in individuals for which estimated clearance t1/2 values were very long (4.9 years).
2 Freedman and Robinson (1988) also reported slower alveolar clearance rates in coal miners,
3 some of whom had a mild degree of pneumoconiosis.
4
5 4.3.3 Role of Alveolar Macrophages in the Clearance of Participate Matter
6 4.3.3.1 Alveolar Macrophage-Mediated Clearance of Participate Matter
7 Aleveolar macrophages constitute an important, first-line cellular defense mechanism
8 against inhaled particles that deposit in the alveolar region of the lung. It is well established
9 that a host of diverse materials, including diesel paniculate matter, are phagocytized by the
10 AMs shortly after deposition (White and Garg, 1981; Lehnert and Morrow, 1985) and that
11 such cell-contained particles are generally rapidly sequestered from both the extracellular
12 fluid lining in the alveolar region and the potentially sensitive alveolar epithelial cells.
13 In addition to this role in compartmentalizing particles from other lung constituents, AMs are
14 prominently involved in mediating the clearance of relatively insoluble particles from the air
15 spaces (Lehnert and Morrow, 1985). Although the details of the actual process have not
16 been delineated, AMs with their particle burdens gain access and become coupled to the
17 mucociliary escalator and are subsequently transported from the lung via the conducting
18 airways. Although circumstantial in nature, numerous lines of evidence indicate that such
19 AM-mediated particle clearance is normally the predominant mechanism by which relatively
20 insoluble particles are removed from the lungs (Gibb and Morrow, 1962; Ferin, 1982;
21 Harmsen et al., 1985; Lehnert and Morrow, 1985; Powdrill et al., 1989).
22 The removal characteristics for particles deposited in the lung's alveolar region have
23 been descriptively represented by numerous investigators as a multicompartment or
24 multicomponent process in which each component follows simple first-order kinetics (Snipes
25 and Clem, 1981; Snipes et al., 1988; Lee et al., 1983). Although the various compartments
26 can be described mathematically, the actual physiologic mechanisms determining these
27 differing clearance rates have not been well characterized.
28 Lehnert et al. (1988), Lehnert et al. (1989) and Lehnert et al. (unpublished) performed
29 a study using laboratory rats to examine particle-AM relationships over the course of alveolar
30 clearance of low to high lung burdens of noncytotoxic microspheres (2.13 /tm diameter) to
31 obtain information on potential AM-related mechanisms that form the underlying bases for
December 1994 4_17 DRAFT-DO NOT QUOTE OR CITE
-------
1 kinetic patterns of alveolar clearance as a function of particle lung burdens. The
2 intratracheally instilled lung burdens studied varied from 1.6 x 107 particles (about 85
3 for the low lung burden to 2.0 X 108 particles (about 1.06 mg) for the mid-dose and
4 6.8 x 108 particles (about 3.6 mg) for the highest lung burden. The lungs were lavaged at
5 various times postexposure and the numbers of spheres in each macrophage counted.
6 The t1/2 values of both the early and later components of clearance were virtually
7 identical following deposition of the low and medium lung burdens. For the highest lung
8 burden, significant prolongations were found in both the early, more rapid as well as the
9 slower component of alveolar clearance. The percentages of the particle burden associated
10 with the earlier and later components, however, were similar to those of the lesser lung
11 burdens. Based on the data, the authors concluded that translocation of AMs from alveolar
12 spaces by way of the conducting airways is fundamentally influenced by the particle burden
13 of the cells so translocated. In the case of particle overload that occurred at the highest lung
14 burden, the translocation of AMs with the heaviest cellular burdens of particles (i.e., greater
15 than about 100 microspheres per AM) was definitely compromised.
16 On the other hand, analysis of the disappearance of AMs with various numbers of
17 particles indicates that they may not exclusively reflect the translocation of AM from the
18 lung. The observations are also consistent with a gradual redistribution of retained particles
19 among the lung's AMs concurrent with the removal of particle-containing AMs via the
20 conducting airways per se. Experimental support suggestive of potential processes for such
21 particle redistribution comes from a variety of investigations involving AM and other
22 endocyte cell types (Heppleston and Young, 1974; Evans et al., 1986; Aronson, 1963;
23 Sandusky et al., 1977; Heppleston, 1961; Riley and Dean, 1978).
24
25 4.3.3.2 Translations of Particles to Extraalveolar Macrophage Compartment Sites
26 Although the phagocytosis of particles by lung-free cells and the mucociliary clearance
27 of the cells with their paniculate matter burdens represent the most prominent mechanisms
28 that govern the fate of particles deposited in the alveolar region, other mechanisms exist that
29 can affect both the retention characteristics of relatively insoluble particles in the lung and
30 the lung clearance pathways for the particles. One mechanism is endocytosis of particles by
31 Type I cells (Sorokin and Brain, 1975; Adamson and Bowden, 1978, 1981) that normally
December 1994 4-18 DRAFT-DO NOT QUOTE OR CITE
-------
1 provide >90% of the cell surface of the alveoli in the lungs of a variety of mammalian
2 species (Crapo et al., 1983). This process may be related to the size of the particles that
3 deposit in the lungs and the numbers of particles that are deposited. Adamson and Bowden
4 (1981) found that with increasing loads of carbon particles (0.03 /xm diameter) instilled in the
5 lungs of mice, more free particles were observed in the alveoli within a few days thereafter.
6 The relative abundance of particles endocytosed by Type I cells also increased with
7 increasing lung burdens of the particles, but instillation of large particles (1.0 /un) rarely
8 resulted in their undergoing endocytosis. A 4-mg burden of 0.1 /mi-diameter latex particles
9 is equivalent to 8 x 1012 particles, whereas a 4-mg burden of 1.0-um particles is composed
10 of 8 x 109 particles. Regardless, diesel particles with volume median diameters between
11 0.05 and 0.3 /xm (Frey and Corn, 1967; Kittleson et al., 1978) would be expected to be
12 within the size range for engulfment by Type I cells, should suitable encounters occur.
13 Indeed, it has been demonstrated that diesel particles are endocytosed by Type I cells in vivo
14 (White and Garg, 1981).
15 Unfortunately, information on the kinetics of particle endocytosis by Type I cells
16 relative to that by AMs is scanty. Even when relatively low burdens of paniculate matter
17 are deposited in the lungs, some fraction of the particles usually appears in the regional
18 lymph nodes (Ferin and Fieldstein, 1978; Lehnert, 1989). As will be discussed, endocytosis
19 of particles by Type I cells is an initial, early step involved in the passage of particles to the
20 lymph nodes. Assuming particle phagocytosis is not sufficiently rapid or perfectly efficient,
21 increasing numbers of particles would be expected to gain entry into the Type I epithelial cell
22 compartment during chronic aerosol exposures. Additionally, if particles are released on a
23 continual basis by AMs that initially sequestered them after lung deposition, some fraction
24 the "free" particles so released could also undergo passage from the alveolar space into
25 Type I cells.
26 The endocytosis of particles by Type I cells represents only the initial stage of a process
27 that can lead to the accumulation of particles in the lung's interstitial compartment and the
28 subsequent translocation of particles to the regional lymph nodes. As shown by Adamson
29 and Bowden (1981), a vesicular transport mechanism in the Type I cell can transfer particles
30 from the air surface of the alveolar epithelium into the lung's interstitium, where particles
31 may be phagocytized by interstitial macrophages or they may remain in a "free" state for a
December 1994 4_19 DRAFT-DO NOT QUOTE OR CITE
-------
1 poorly defined period of time that may be dependent on the physicochemical characteristics
2 of the particle. The lung's interstitial compartment, accordingly, represents an anatomical
3 site for the retention of particles in the lung. Whether or not AMs and perhaps
4 polymorphonuclear lymphocytes (PMNs) that have gained access to the alveolar space
5 compartment and phagocytize particles there also contribute to the particle translocation
6 process into the lung's interstitium remains a controversial issue. However, it is widely
7 believed that once AMs, at least, assume occupancy in the alveoli, they do not reenter the
8 lung's interstitium (Roser, 1970; Brain et al., 1977; Adamson and Bowden, 1978). It should
9 be pointed out, however, that migration of AMs into the interstitium may be species
10 dependent. Evidence that such migration of AMs may contribute significantly to the passage
11 of particles to the interstitial compartment and also may be involved in the subsequent
12 translocation of particles to draining lymph nodes has been obtained with the dog model
13 (Harmsenetal., 1985).
14 The fate of particles once they enter the lung's interstitial spaces remains unclear.
15 Some particles, as previously indicated, are phagocytized by interstitial macrophages whereas
16 others apparently can remain in a free state in the interstitium for some time without being
17 engulfed by interstitial macrophages. It is currently unknown what fraction of the interstitial
18 macrophages may subsequently enter the alveoli with their engulfed burdens of particles and
19 thereby contribute to the size of the resident AM population over the course of lung
20 clearance. Moreover, no investigations have been conducted to date to assess the influence
21 that the burden of particles with an interstitial macrophage may have on its ability to migrate
22 into the alveolar space compartment.
23 It appears that at least some particles that gain entry into the interstitial compartment
24 can further translocate to the extrapulmonary regional lymph nodes. This process apparently
25 can involve the passage of free particles as well as particle-containing cells via lymphatic
26 channels in the lungs (Harmsen et al., 1985; Ferin and Fieldstein, 1978; Lee et al., 1985).
27 It is conceivable that the mobility of the interstitial macrophages could be particle burden
28 limited, and under conditions of high cellular burdens, a greater fraction of particles that
29 accumulate in the lymph may reach these sites as free particles. Whatever the process,
30 existing evidence indicates that when lung burdens of particles result in particle overload
December 1994 4-20 DRAFT-DO NOT QUOTE OR CITE
-------
1 condition, particles accumulate both more rapidly and abundantly in lymph nodes that receive
2 lymphatic drainage from the lung (Ferin and Feldstein, 1978; Lee et al., 1985).
3
4 4.3.3.3 Potential Mechanisms for an Alveolar Macrophage Sequestration
5 Compartment for Particles During Particle Overload
6
7 Several factors may be involved in the particle-load-dependent retardations in the rate
8 of particle removal from the lung and the corresponding functional appearance of an
9 abnormally slow-clearing, or particle sequestration compartment. As previously mentioned,
10 one potential site for particle sequestration is the containment of particles in the Type I cells.
11 Information on the retention kinetics for particles in the Type I cells is nonexistent, and no
12 information on how the vesicular transport of particles across the Type I cell may be
13 exhausted or otherwise modified during particle overload is currently available. Also, no
14 morphometric analyses have been performed to date to estimate what fraction of a retained
15 lung burden may be contained in the lung's Type I cell population during lung overloading.
16 Another anatomical region in the lung that may be a slow clearing site is the interstitial
17 compartment. Little is known about either the kinetics of removal of free particles or
18 particle-containing macrophages from the interstitial spaces or what fraction of a retained
19 burden of particles is contained in the lung's interstitium during particle overload. The
20 gradual accumulation of particles in the regional lymph nodes and the appearance of particles
21 and cells with associated particles in lymphatic channels and in the peribronchial and
22 perivascular lymphoid tissue (Lee et al., 1985; White and Garg, 1981) suggest that the
23 mobilization of particles from interstitial sites via local lymphatics is a continual process.
24 Indeed, it is clear from histologic observations of the lungs of animals chronically
25 exposed to diesel particles that Type I cells, the interstitium, the lymphatic channels, and
26 pulmonary lymphoid tissues are sites that could represent subcompartments of a more
27 generalized slow-clearing compartment.
28 Although these sites must be considered to be potential contributors to the increased
29 retention of particles during particle overload, a disturbance in particle-associated
30 AM-mediated clearance is undoubtedly the predominant cause inasmuch as the AMs are the
31 primary reservoirs of deposited particles. The factors responsible for a failure of AMs to
32 translocate from the alveolar space compartment in lungs with high particulate-matter
December 1994 4_2i DRAFT-DO NOT QUOTE OR CITE
-------
1 burdens remains uncertain, although a hypothesis concerning the process has recently been
2 offered involving volumetric AM burden (Morrow, 1988).
3 Other processes may also be involved in preventing particle-laden AMs from leaving
4 the alveolar compartment under conditions of particle overload in the lung. Clusters or
5 aggregates of particle-laden AMs in the alveoli are typically found in the lungs of
6 experimental animals that have received large lung burdens of a variety of types of particles
7 (Lee et al., 1985), including diesel exhaust particulate matter (White and Garg, 1981;
8 McClellan et al., 1982). The aggregation of AMs may explain, in part, the reduced
9 clearance of particle-laden AM during particle overload. The definitive mechanism(s)
10 responsible for this clustering of AMs has not been elucidated to date. Whatever the
11 underlying mechanism(s) for the AM agglutinating response, it is noteworthy that AMs
12 lavaged from the lungs of diesel exhaust-exposed animals continue to demonstrate a
13 propensity to aggregate (Strom, 1984). This observation suggests that the surface
14 characteristics of AMs are fundamentally altered in a manner that promotes their adherence
15 to one another in the alveolar region and that AM aggregation may not simply be directly
16 caused by their abundant accumulation as a result of immobilization by large particle loads.
17 Furthermore, even though overloaded macrophages may redistribute particle burden to other
18 AMs, clearance may remain inhibited (Lehnert, 1988). This may, in part, be due to
19 attractants from the overloaded AMs causing agglutination of those that are not carrying a
20 particle burden.
21
22
23 4.4 BIOAVAILABILITY OF ORGANIC CONSTITUENTS PRESENT
24 ON DIESEL EXHAUST PARTICLES
25 4.4.1 Whole-Animal Studies
26 Because it has been shown that diesel soot extract is not only mutagenic but also
27 contains known carcinogens, the organic fraction was originally considered to be the
28 primary source of carcinogenicity in animal studies. Evidence presented in more-recent
29 studies, however, indicates that the insoluble carbon core of the particle may fully explain
30 the pathogenic and carcinogenic processes observed in the inhalation studies. (See Chapter
31 10 for a discussion of this issue.) Nevertheless, the organic constituents may be involved.
December 1994 4-22 DRAFT-DO NOT QUOTE OR CITE
-------
1 An initial step and possibly the rate-limiting step in the bioavailability of carcinogenic
2 organics present on diesel particles is their dissociation from the particle surface (Vostal,
3 1983). This section will therefore focus on the dissociation of these organics from the
4 particles.
5 Sun et al. (1984) compared the disposition of diesel particle-adsorbed benzo[a]pyrene
6 (B[a]P) (0.1% by weight) and pure B[a]P following nose-only inhalation by F344 rats.
7 Long-term retention (percentage retained after 7 days) of particle-adsorbed 3H-B[a]P was
8 approximately 230-fold greater than that for pure 3H-B[a]P. Alveolar clearance of particle-
9 associated 3H was biphasic, with a long-term t1/2 of 18 days, the latter representing
10 clearance of 59% of the initially deposited radiolabel. Clearance of pure B[a]P aerosol was
11 >99% within 2 h and was apparently the result of alveolar and tracheobronchial epithelial
12 absorption into the blood, rather than the result of mucociliary clearance and subsequent
13 ingestion (Sun et al., 1982). The data therefore indicate that adsorption to the carbonaceous
14 diesel particle prolongs retention of the organic components.
15 A companion study (Bond et al., 1986) examined the biological fate of
16 14C-l-nitropyrene (14C-NP), both in pure form and adsorbed to diesel exhaust particles,
17 following 1-h nose-only inhalation by male F344 rats. Concentrations of I4C-NP ranged
18 from 0.05 to 1.1 mg/m3 of air, and diesel particle concentrations, where utilized, ranged
19 from 3.7 to 6.1 mg/m3 of air. The results indicated that long-term lung retention of 14C-NP
20 adsorbed onto diesel exhaust particles was 80-fold greater (t,/2 = 36 days) than that for pure
21 14C-NP, demonstrating again that adsorption onto the diesel particles prolongs the release of
22 the PAHs.
23 Residence time is also prolonged when organics are adsorbed to other types of
24 particles. For example, Creasia et al. (1976) found that when crystalline B[a]P was instilled
25 into the lungs of mice, it was removed from the respiratory tract with a t,/2 of «1.5 h, but
26 when the B[a]P was adsorbed to 0.5 to 1.0 urn carbon particles, its t,/2 in the respiratory
27 tract increased to «36 h. Hence, the adsorption of B[a]P to the carbon particles increased
28 the lung retention of the B[a]P greater than 20-fold. Similar results have also been
29 obtained with B[a]P adsorbed to other particle types, including insoluble Ga2O3 (Sun et al.,
30 1982) and insoluble ferric oxide (Saffiotti et al., 1964). Consistent with a gradual elution of
31 B[a]P in AMs, Creasia and co-workers (1976) found that the removal of B[a]P when bound
December 1994 4.23 DRAFT-DO NOT QUOTE OR CITE
-------
1 to the carbon was faster than the lung clearance of carbon particles only, which had a
2 clearance t1/2 of »7 days.
3 Ball and King (1985) studied the disposition and metabolism of 14C-labeled
4 1-NP (>99.9% purity) coated onto diesel exhaust particles. A single dose of 14C-NP
5 (380 ng/g particle) was intratracheally administered (in 0.2-mL buffered saline) at a particle
6 dose of 5 mg per rat. Another group of rats (number not specified) received the labeled
7 14C-NP in 0.5 mL of buffered saline intragastrically. Additional groups of AGUS strain
8 rats raised conventionally or germ-free received ip injections of 14C-NP to determine the
9 role of gastrointestinal flora on the metabolism of 1-NP.
10 Regardless of the route of administration, >50% of the 14C was excreted within the
11 first 24 h; 20 to 30% of this appeared in the urine, and 40 to 60% was excreted in the
12 feces. The 14C excretion pattern for the intratracheally instilled compound was nearly
13 identical to that of the orally administered compound. For animals receiving intratracheally
14 instilled compound, 16 to 38% of the unexcreted dose was in the gastrointestinal tract and
15 5 to 8% remained in the lungs. Traces of radiolabel were detected in the trachea and
16 esophagus. Five to 12% of the radiolabel in the lung co-purified with the protein fraction,
17 indicating protein binding of the 1-NP-derived 14C. However, the corresponding DNA
18 fraction contained no 14C above background levels. The similar excretion kinetics and
19 metabolic profiles for these various routes of administration indicate that 1-NP becomes
20 bioavailable both in the lungs and the gastrointestinal tract.
21 Bevan and Ruggio (1991) assessed the bioavailability of B[a]P adsorbed to diesel
22 exhaust particles from a 5.7-L Oldsmobile engine. In this study, exhaust particles were
23 supplemented with exogenous 3H-B[a]P to provide 2.62 jag B[a]P/g of exhaust particle.
24 Distribution of the radiocativity was assessed at 1, 6, 24, or 72 h after intratracheal
25 instillation of these particles into Sprague-Dawley rats (1 mg of particles suspended in
26 0.3 mL of 0.15 M NaCl). At 24 h after administration, 68.5% of the radiolabel remained in
27 the lungs. This is approximately a 3.5-fold greater proportion than that reported by Sun
28 et al. (1984), the difference being attributed to slower pulmonary absorption, less
29 mucociliary transport in intratracheally instilled animals, and differences in administered
30 dose. At 3 days following administration, over 50% of the radioactivity remained in the
31 lungs, nearly 30% had been excreted into the feces, and the remainder was distributed
December 1994 4-24 DRAFT-DO NOT QUOTE OR CITE
-------
1 throughout the body. Experiments using rats with cannulated bile ducts showed that
2 approximately 10% of the administered radioactivity appeared in the bile over a 10-h period
3 and that less than 5% of the radioactivity entered the feces via mucociliary transport. The
4 in vitro elution of E[a]P into dimyristoylphosphatidylcholine (DMPC) vesicles was similar
5 for 3H-B[a]P-supplemented exhaust particles and for native exhaust particles (no additionnal
6 B[a]P added), thereby indicating that the estimation of in vivo bioavailability of B[a]P from
7 the 3H-B[a]P-supplemented exhaust particles was reasonably accurate.
8
9 4.4.2 Extraction of Diesel Particle Associated Organics in Biological
10 Fluids
11 For mutagenicity testing or biochemical analysis, diesel particles are usually extracted
12 with organic solvents such as dichloromethane. The efficiency of extraction may be much
13 different, however, than that of the fluids surrounding the particles in the in vivo state.
14 A number of studies have therefore been conducted in which attempts were made to extract
15 diesel exhaust particles with serum or lung lavage fluid. The efficiency of extraction was
16 usually estimated by performing mutagenicity tests on the extracts.
17 The utility of evaluating extraction by lung fluid or serum may be somewhat limited
18 because particles deposited in the alveoli are normally rapidly ingested by AMs. However,
19 as large lung burdens of diesel particles are attained, such as during chronic, high-
20 concentration exposures to diesel particles, AMs that become heavily laden may reach their
21 phagocytic capacity, thereby reducing their subsequent phagocytic ability. Under these
22 conditions, an increasing fraction of deposited particles could escape the phagocytic
23 mechanism and thereby be relatively more available over time in the extracellular lung fluid
24 prior to (1) their removal from the lung by extra-macrophagic clearance via the
25 tracheobronchial route, (2) their subsequent engulfment by newly recruited phagocytes,
26 and/or (3) their engulfment by Type I cells. Even under such conditions, a relatively large
27 mass of the particles will be within AM phagolysosomes, where low pH and enzyme
28 activity would be expected to act on the particles and adsorbed organics.
29 Particles from a 5.7-L engine operated at idle and from a 2.1-L engine operating on a
30 cycle of varying speed and load were incubated in lavage fluid, serum, saline, albumin,
31 dipalmitoyl lecithin, or dichloromethane (Brooks et al., 1981). The efficiency of extraction
December 1994 4.25 DRAFT-DO NOT QUOTE OR CITE
-------
1 by biological fluids was only 3 to 10% that of dichloromethane, based on mutagenicity
2 testing, and did not increase with incubation time up to 120 h. Similar findings were
3 reported by King et al. (1981). In this study, lung lavage fluid and lung cytosol fluid
4 extracts of diesel exhaust particles were not mutagenic. Serum extracts of diesel particles
5 did exhibit some mutagenic activity, but this was considerably less than that for organic
6 solvent extracts. Furthermore, the mutagenic activity of the solvent extract was
7 significantly reduced when combined with serum or lung cytosol fluid, suggesting protein
8 binding or biotransformation of the mutagenic components.
9 Siak et al. (1980) assessed the mutagenicity of material extracted from diesel particles
10 by bovine serum albumin in solution, simulated lung surfactant, fetal calf serum (PCS), and
11 physiologic saline. Only PCS was found to extract some mutagenic activity from the diesel
12 particles. These investigators concluded that the mutagens in diesel particles would not be
13 readily available in vivo. This conclusion lacks definitive proof because extracellular lung
14 fluid is a complex mixture of constituents that undoubtedly have a broad range of
15 hydrophobicity (George and Hook, 1984; Wright and Clements, 1987), and it fundamentally
16 differs from serum in terms of chemical composition (Gurley et al., 1988). Moreover,
17 assessments of the ability of lavage fluids, which actually represent substantially diluted
18 extracellular lung fluid, to extract mutagenic activity from diesel particles clearly do not
19 reflect the in vivo condition.
20 Creasia et al. (1976) reported that when B[a]P was adsorbed onto carbon particles
21 larger than would be expected to be easily phagocytized by AMs (15 to 30 ^m), the rates
22 of elimination of the B[o]P and the particles from the lung were virtually identical. The
23 data thus indicate little extraction from the particles not phagocytized by AM but only
24 surrounded by epithelial lining fluid.
25 In summary, because lung fluids appear to be relatively ineffective in the extraction of
26 particle-adsorbed organics and relatively few particles escape phagocytosis, free particles
27 are likely to contribute very little to the acute bioavailability of adsorbed organics.
28 However, during a particle-overload condition, as occurred in the chronic inhalation studies,
29 an increased fraction of the deposited diesel exhaust particles will not be phagocytized and
30 their adsorbed organics could be released over a long period with a clearance rate that may
31 be equivalent to that of the particles themselves. In this case, the organics are bioavailable
December 1994 4-26 DRAFT-DO NOT QUOTE OR CITE
-------
1 for even longer periods than those from phagocytized particles, from which organics in turn
2 are retained for longer periods than nonadsorbed organics. It is unknown if AMs in
3 particle-overload conditions retain the same capacity to dissolve organics or whether this
4 process is slowed as well, which would also increase the time over which carcinogenic
5 organics are released and available in the overloaded lung. Data of Creasia et al. (1976)
6 showing increased retention of organics adsorbed to large particles could be consistent with
7 this possibility because 15-um particles may still be phagocytized by AM (Snipes and
8 Clem, 1981; Oberdorster et al., 1991).
9
10 4.4.3 Extraction of Diesel Particle Associated Organics by Alveolar Lung
11 Cells and Other Cell Types
12 Another, more likely mechanism by which organic carcinogens (e.g., PAHs) may be
13 extracted from diesel particles in the lung is either particle dissolution or extraction of
14 organics from the particle surface within the phagolysosomes of AMs. This mechanism
15 presupposes that the particles are internalized by these phagocytes. Specific details about
16 the physicochemical conditions of the intraphagolysosomal environment, where particle
17 dissolution in AMs presumably occurs in vivo, have not been well characterized. However,
18 it is known that the phagolysosomes constitute an acidic (pH 4 to 5) compartment in
19 macrophages (Nilsen et al., 1988; Ohkuma and Poole, 1978). The relatively low pH in the
20 phagolysosomes has been associated with the dissolution of some types of inorganic
21 particles (some metals) by macrophages (Marafante et al., 1987; Lundborg et al., 1984), but
22 there are few studies that provide quantitative information concerning how organic
23 constituents of diesel particles (e.g., B[a]P) may be extracted in the phagolysosomes (Bond
24 et al., 1983). Whatever the mechanism, the end result is a prolonged exposure of the
25 respiratory epithelium to the gradual release of carcinogenic agents.
26 Quantitative data on how readily carcinogenic organics may be extracted from diesel
27 particles in the human lung is not available. As shown by Creasia et al. (1976), B[a]P
28 adsorbed onto diesel particles is removed from the mouse lung about five times faster than
29 carbon particles without B[a]P. For the rat, Sun and coinvestigators (1984) have reported
30 that the t1/2 for the lung clearance of B[a]P adsorbed onto diesel particles over a period of
31 time consistent with alveolar phase clearance was about 18 days. This latter value is
December 1994 4_27 DRAFT-DO NOT QUOTE OR CITE
-------
1 similar to the t1/2 for the removal of particles without B[a]P from the rat's lung during the
2 early, more rapid component of alveolar phase clearance (Snipes et al., 1988; Snipes and
3 Clem, 1981; Ferin, 1982; Lehnert et al., 1989; Lehnert, 1989). These findings may suggest
4 that the extraction of organic components from carrier particles by AMs may differ among
5 species, although differences in the lung burdens administered in the investigations
6 mentioned here may have influenced the outcomes of the studies.
7 It should be pointed out that studies designed to examine the extraction of organics by
8 AMs have generally focused on B[a]P as a representative procarcinogen associated with
9 diesel particles. Numerous other agents with carcinogenic activity are also associated with
10 diesel particles; these chemical constituents may be extracted from diesel particles with in
11 vivo kinetics that differ more or less from those of B[a]P. Thus, existing dosimetry models
12 that incorporate desorption of B[a]P from diesel particles as a representative organic
13 constituent (Yu and Yoon, 1988) may not accurately reflect the actual bioavailability of
14 other procarcinogenic agents on diesel particles. As discussed in the next section, however,
15 any error in this respect is likely to be minor.
16
17 4.4.4 Bioavailability of Adsorbed Compounds as a Function of Particle
18 Clearance Rates and Extraction Rates of Adsorbed Compounds
19 The bioavailabilty of toxic organic compounds adsorbed to particles can be influenced
20 by a variety of factors. Although the agent may be active while present on the particle,
21 most particles are taken up by AMs, a cell type not generally considered to be a target site.
22 To reach the target site, therefore, the agent must first elute from the particle surface.
23 Although elution can be considered to be a necessary step, it may not always be sufficient.
24 The agent must then diffuse out of the AM into the extracellular fluid and be absorbed by a
25 target cell (e.g., a Type I cell). In analyzing phagolysosomal dissolution of various ions
26 from particles in the lungs of Syrian golden hamsters, Godleski et al. (1988) demonstrated
27 that solubilization did not necessarily result in clearance of the ions and that binding of the
28 solubilized components to cellular and extracellular structures occurred. It is reasonable to
29 assume that phagocytized diesel soot particles may be subject to similar processes and that
30 these processes would be important in determining the rate of bioavailability of the particle-
31 bound constituents of diesel exhaust. Inability of these constituents to penetrate target cells
December 1994 4-28 DRAFT-DO NOT QUOTE OR CITE
-------
1 or to diffuse into the bloodstream is another possible factor limiting bioavailability to lung
2 target cells. Nevertheless, until further research demonstrates otherwise, it is assumed that
3 the rate limiting factor in the bioavailability of particle bound organics is the desorption rate
4 from the particle surface.
5 The long-term clearance t1/2 of diesel particles from the lungs of rats in the
6 nonoverloaded state was shown to range from about 2 to 3 mo (Chan et al., 1981, 1984;
7 Griffis et al., 1983; Lee et al., 1983). Clearance rate data for diesel particles in humans is
8 not available. For other types of insoluble particles such as polystyrene, however, t1/2
9 values are close to 1 year (Bohning et al., 1982). The clearance t,/2 values for B[a]P and
10 1-NP from the diesel particle surface, on the other hand, were reported to be only about
11 18 and 36 days, respectively (Sun et al., 1984; Bond et al., 1986). The lower t1/2 values for
12 clearance of the organics compared with particles themselves indicate that most of the
13 organics are being eluted prior to particle clearance, especially in humans. Quantitative
14 estimates of percentage eluted can be obtained using the following formula reported by
15 Pepelko (1987) which combines the two first-order rates.
16
Percentage elution = L [1 - e]"(kp + ks)t
kp +ks
18
19
20 where kp and ks are rate constants for kp + ks particle clearance and desorption,
21 respectively, and t is time post exposure.
22 For humans, assuming an elution t1/2 on the order of 2 to 4 weeks, well over 90% of
23 the organics should desorb from the particles. Even in rats with more rapid particle
24 clearance rates, most of the organics can be expected to be eluted. Whether particle
25 overload in the lung results in a change in elution rates of the organics is not known.
26 If lung burden of particulate matter is the proper dosimetric factor for induction of
27 pathology or carcinogenesis, on the other hand, target organ dose would be predicted to
28 increase more rapidly than exposure concentration under lung overload conditions.
29 Gerde et al. (1991a,b) described models simulating the effect of particle aggregation
30 and PAH content on the rate of PAH release in the lung. The investigators used three
31 models, one of which simulated a low-dose situation where only the adsorbed layer of PAH
December 1994 4_29 DRAFT-DO NOT QUOTE OR CITE
-------
1 is released from the carrier particle, and two of which simulated desorption of PAHs using
2 high PAH levels (with and without an inert carrier dust). Based on the theoretical results
3 obtained, particle retention would be of lesser importance for low-dose situations in which
4 particle-associated PAHs would be rapidly released at the site of particle deposition and not
5 necessarily at the site of particle retention. Frequent low-level exposure, therefore, may
6 result in sustained exposure of target cells and subsequently greater likelihood of tumor
7 formation at the site of initial deposition. For the high-dose situations, as represented by
8 instillation experiments in animals, critical doses to cells are likely to occur at the site of
9 retention because slow release from the particles may increase the dose of metabolites (and
10 increase the risk of tumor formation) before the parent PAH compound can be cleared from
11 the lungs. Generally, the models suggested that the local disposition of PAHs would be
12 more dependent upon the behavior of dissolved PAHs in the tissues following their release
13 from the carrier particles than by interactions between the PAHs and the carrier particles.
14 The model predictions were consistent with findings from animal studies that showed
15 longer PAH retention with higher exposures and longer retention for instillation
16 administration vs inhalation exposure.
17 Studies by Gerde et al. (1993a,b,c) using beagle dogs provided additional data
18 regarding the dosimetry of inhaled PAH supportive of the previously discussed models.
19 In the Gerde et al. (1993a) study, the dogs were exposed to an aerosolized bolus of PAH
20 crystals (phenanthrene or B[#]P) in a single breath. The PAH clearance was measured by
21 monitoring PAH levels in the systemic circulation. Clearance from the alveolar region was
22 dependent upon lipophilicity of the PAH; clearance of highly lipophilic PAHs (i.e., B[a]P)
23 was limited by diffusion of the chemical through the alveolar septa, while clearance of
24 moderately lipophilic PAHs (i.e., phenanthrene) was limited by rate of perfusion of the
25 blood. Therefore, bronchi, with their thicker epithelia, would be at greater risk than alveoli
26 for PAH-induced toxicity at the portal of entry. In the Gerde et al. (1993b) study, small
27 volumes of saline containing either dissolved B[a]P or phenanthrene, or a suspension of
28 particulate solvent green or macroaggregated albumin (MAA) were instilled into mucous
29 lining layer of a primary bronchus or distal tracheas. The highly lipophilic B[«]P was
30 cleared via the mucociliary escalator, some being cleared very rapidly (>90 mm/min). The
31 portion of B[a]P that penetrated the bronchial epithelium exhibited a half-time in the range
December 1994 4-30 DRAFT-DO NOT QUOTE OR CITE
-------
1 of 1.4 h indicating a diffusion-limited uptake of B[a]P by the airways. Although
2 mucociliary clearance is rapid for most of the lipophilic toxicants, the long retention time
3 for that portion that penetrates the epithelium is sufficient for substantial metabolism to
4 occur, resulting in a potential for local toxicity. Gerde et al. (1993c) used the data from the
5 previously described studies to validate models of alveolar clearance (Gerde et al., 1991b),
6 mucous lining penetration (Gerde and Scholander, 1987), and bronchial wall penetration
7 (Gerde et al., 1991b). The analysis provided a reasonable validation of the transport models
8 for these structural regions of the lung. Specifically, alveolar clearance and mucociliary
9 clearance are primarily via molecular diffusion, whereas clearance from bronchial walls
10 involves diffusion, metabolism of a portion of the PAH load, and endocytosis. Such
11 findings suggest that the bronchial epithelium may be especially vulnerable to toxicity
12 induced by diffusion-limited lipophilic substances. Additionally, the findings indicate that
13 all PAHs, and not just particle-retained fractions, are of importance in carcinogenic
14 responses to inhaled particle-associated PAHs.
15
16
17 4.5 CONSIDERATIONS FOR DOSIMETRY MODELING
18 Although more than one approach is possible in the development of dosimetry models
19 for inhaled diesel particulate matter, several dosimetry parameters are common to any
20 approach. These include ventilatory rates and volumes, tracheobronchial and alveolar
21 surface area, tracheobronchial and alveolar deposition efficiency, and tracheobronchial and
22 alveolar clearance rates. If both the adsorbed organics and the particles are considered to
23 be involved in the carcinogenic response of the lung, elution t,/2 values of these chemicals
24 should be included in the model. The dosimetry models must take into consideration not
25 only species differences but also the effects of extrapolating from high exposure
26 concentrations which results in an inhibition of particle clearance from the lungs caused by
27 overload. If adequate data can be obtained, the model could be expanded to include
28 transport to lung-associated lymph nodes, ingestion of particles by Type I cells, cell
29 proliferation rates, clearance of the interstitial compartment, and release of mediators for
30 chemotaxis and cell proliferation.
December 1994 4.3 j DRAFT-DO NOT QUOTE OR CITE
-------
1 An important consideration is ventilation rate. Numerous estimates of alveolar
2 ventilation rates are available for both humans and rats. Nevertheless, use of such estimates
3 is still a source of considerable potential error because they are usually resting values.
4 Actual respiration can vary widely with activity or with exposure conditions. Human
5 activity levels are also highly variable. Although it may be necessary to use a single mean
6 value in a model, it would be useful to include risk estimates for individuals having much
7 greater daily ventilation exchange rates as a result of participation in endurance sports or
8 performance of heavy labor.
9 Another variable not discussed previously is the adjustment of dose based on
10 metabolic rate. It has been U.S. Environmental Protection Agency (EPA) policy to consider
11 that effective dose varies with metabolic rate. Arguments for and against this presumption
12 are beyond the scope of this document. The primary consideration here is the degree of
13 adjustment. The EPA has traditionally adjusted for species differences in metabolic rate
14 based on the 2/3 power of body weight as a surrogate for body surface area. This factor is
15 currently being reappraised within EPA, and a preliminary proposal to alter this adjustment
16 to the 3/4 power has been made.
17 A major consideration in the development of dosimetry models is a judgement
18 concerning which fraction of exhaust is responsible for the induction of lung cancer. As is
19 discussed in Chapter 10, two approaches are used for quantitative assessment based on
20 animal data. In the first, cancer is assumed to be induced by the organic constituents
21 present on the particle surface, primarily PAHs and nitropyrenes. As discussed herein, in
22 an appropriate model, the effective dose will correlate closely with the deposited dose, with
23 only minor corrections for lung overloading. In a second approach, it is assumed that
24 retained particle burden in the lung fully accounts for the induced lung tumors. A more
25 specific dose parameter for this second approach may be the surface area of the retained
26 particles as discussed in Chapter 10. In this case, modelling must account for slowing of
27 clearance during lung overload, as well large differences in normal clearance rates between
28 rats and humans. For this approach, low-dose extrapolation, if exposures are at overload
29 levels, may result in a considerably different risk estimate than one based on target organ
30 dose of organics. A third approach could be proposed in which both exhaust components
December 1994 4-32 DRAFT-DO NOT QUOTE OR CITE
-------
1 may be operating simultaneously, with the PAHs initiating the carcinogenic process and the
2 particles promoting the process through induction of cell proliferation.
3
4
5 4.6 SUMMARY
6 Because only very limited evidence exists for diesel exhaust-induced tumors at
7 nonpulmonary sites or for tumor induction by the gaseous fraction alone, dosimetry
8 considerations were limited to either whole exhaust or particulate matter deposited in the
9 lungs. Dosimetric considerations were further limited to the alveolar region of the
10 respiratory tract for several reasons. First of all, most deposited particulate matter is
11 transported by mucociliary transport from the conducting airways in less than 1 day and
12 exposure to diesel exhaust does not appear to significantly inhibit this process. However,
13 as previously discussed, there may also be a long-phase retention, especially if
14 tracheobronchial clearance is impaired. In general, the rapid clearance in the conducting
15 airways reduces the time for extraction of organics from the particle surface (although there
16 is evidence for some relatively rapid removal of organics), and some AMs in the
17 mucociliary escalator may contain particles desorbed of organics. Finally, most of the
18 pathologic and carcinogenic effects occur at or distal to the terminal bronchioles in the rat.
19 Clearance of the diesel particles from the alveolar region varied from about 2 mo in
20 rats to an estimate of 1 year in humans. Under high-exposure regimes, lung overload
21 occurred in rats, leading to slower or near cessation of clearance, thereby increasing lung
22 burdens even further. In addition, with large lung burdens, uptake of particles by Type I
23 cells, passage into the interstitium, and transport to lung-associated lymph nodes was
24 increased. Factors considered to be involved in clearance inhibition included loss of AM
25 mobility with large particle loads and a tendency for AMs to aggregate and thereby become
26 immobilized.
27 Most biological fluids tested, including lung lavage fluid and serum, were relatively
28 ineffective in the extraction of organic agents adsorbed to the diesel particle surface.
29 Particles deposited in the alveolar region, however, are rapidly phagocytized by AMs which
30 are more effective in this regard. Although actual elution t]/2 values of organics from
31 phagocytosed particles were difficult to obtain, they were generally less than those for the
December 1994 4.33 DRAFT-DO NOT QUOTE OR CITE
-------
1 particles themselves, indicating that most organics are released even without inhibition of
2 particle clearance. The gradual elution also prolonged the residence time of organics in the
3 lungs compared with pure organic agents such as B[a]P, possibly avoiding overloading of
4 biological activation systems and thus increasing their effectiveness.
5 In the development of a dosimetry model to allow both low dose extrapolation and
6 extrapolation of diesel exhaust bioassay data from laboratory animals to humans, several
7 parameters must be accounted for. These include, at a minimum, deposition efficiency,
8 particle clearance rates, desorption rates of organics from the particle surface, and lung
9 surface area. The respiratory rates and volumes are highly variable in both experimental
10 animals and humans and are also determinants of deposition efficiency. Animal estimates
11 are often based on published values collected under resting conditions. Respiration,
12 however, may be inhibited by the irritant gases present in diesel exhaust, however, less so
13 at low dilution ratios. On the other hand, respiration may be either greater or less than
14 estimated resting values, depending on whether exposures were carried out at night when
15 the animals are likely to be awake and active or during the day when they are more likely
16 to be asleep. Human respiratory exchange rates are also quite variable, with the physically
17 active segment of the population at potentially greater risk because of higher doses resulting
18 from higher respiration rates.
19 Adjustment for particle clearance rate is necessary for two reasons. First of all, many
20 of the animal experiments were conducted under exposure regimes resulting in an inhibition
21 of clearance caused by an accompanying lung burden overload. If lung burden of
22 particulate matter is considered to be the proper dosimetric variable, then the
23 disproportionately large lung burdens at high levels of exposure must be adjusted for.
24 Second, even under low exposure regimes, clearance is slower in humans than in rats.
25 If the correct dosimetric variable, on the other hand, is particle-free organic matter, a
26 smaller adjustment for variations in particle clearance rates is required because most of the
27 organics are likely to be eluted from the particles deposited in the alveolar region, even at
28 normal clearance rates. Nevertheless, some adjustment is still necessary, because all of the
29 organics are seldom all eluted.
30
December 1994 4.34 DRAFT-DO NOT QUOTE OR CITE
-------
1 REFERENCES
2 Adamson, I. Y. R.; Bowden, D. H. (1978) Adaptive responses of the pulmonary macrophagic system to carbon:
3 II. morphologic studies. Lab. Invest. 38: 430-438.
4
5 Adamson, I. Y. R.; Bowden, D. H. (1981) Dose response of the pulmonary macrophagic system to various
6 particulates and its relationship to transepithelial passage of free particles. Exp. Lung Res. 2: 165-175.
7
8 Alarie, Y. (1966) Irritating properties of airborne materials to the upper respiratory tract. Arch. Environ. Health
9 13: 433-449.
10
11 Alarie, Y. (1973) Sensory irritation to the upper airways by airborne chemicals. Toxicol. Exp. Pharmacol.
12 24: 279-297.
13
14 Aronson, M. (1963) Bridge formation and cytoplasmic flow between phagocytic cells. J. Exp. Med.
15 118: 1083-1088.
16
17 Bailey, M. R.; Fry, F. A.; James, A. C. (1982) The long-term clearance kinetics of insoluble particles from the
18 human lung. Ann. Occup. Hyg. 26: 273-290.
19
20 Ball, L. M.; King, L. C. (1985) Metabolism, mutagenicity, and activation of 1-nitropyrene in vivo and in vitro.
21 Environ. Int. 11:355-361.
22
23 Battigelli, M. C.; Hengstenberg, F.; Mannela, R. J.; Thomas, A. P. (1966) Mucociliary activity. Arch. Environ.
24 Health 12: 460-466.
25
26 Bevan, D. R.; Ruggio, D. M. (1991) Bioavailability in vivo of benzo[a]pyrene adsorbed to diesel paniculate.
27 Toxicol. Ind. Health 7: 125-139.
28
29 Bohning, D. E.; Cohn, S. H.; Lee, H. D.; Atkins, H. L. (1980) Two-phase deep-lung clearance in man.
30 In: Sanders, C. L.; Cross, F. T.; Dagle, G. E.; Mahaffey, J. A., eds. Pulmonary toxicology of
31 respirable particles: proceedings of the nineteenth annual Hanford life sciences symposium; October
32 1979; Richland, WA. Washington, DC: U.S. Department of Energy; pp. 149-161. Available from:
33 NTIS, Springfield, VA; CONF-791002.
34
35 Bohning, D. E.; Atkins, H. L.; Cohn, S. H. (1982) Long-term particle clearance in man: normal and impaired.
36 In: Walton, W. H., ed. Inhaled particles V: proceedings of an international symposium; September 1980;
37 Cardiff, Wales. Ann. Occup. Hyg. 26: 259-271.
38
39 Bond, J. A.; Mitchell, C. E.; Li, A. P. (1983) Metabolism and macromolecular covalent binding of
40 benzo[a]pyrene in cultured Fischer-344 rat lung type II epithelial cells. Biochem. Pharmacol.
41 32: 3771-3776.
42
43 Bond, J. A.; Sun, J. D.; Medinsky, M. A.; Jones, R. K.; Yeh, H. C. (1986) Deposition, metabolism, and
44 excretion of l-[14C]nitropyrene and l-[14C]nitropyrene coated on diesel exhaust particles as influenced by
45 exposure concentration. Toxicol. Appl. Pharmacol. 85: 102-117.
46
47 Brain, J. D.; Mensah, G. A. (1983) Comparative toxicology of the respiratory tract. Am. Rev. Respir. Dis.
48 128: S87-S90.
49
50 Brain, J. D.; Sorokin, S. P.; Godleski, J. J. (1977) Quantification, origin, and fate of pulmonary macrophages.
51 In: Brain, J. D.; Proctor, D. F.; Reid, L. M., eds. Respiratory defense mechanisms (in two parts): part
52 II. New York, NY: Marcel Dekker, Inc.; pp. 849-891. (Lung biology in health and disease: v. 5)
53
December 1994 4.35 DRAFT-DO NOT QUOTE OR CITE
-------
1 Brooks, A. L.; Wolff, R. K.; Royer, R. E.; Clark, C. R.; Sanchez, A.; McClellan, R. O. (1981) Deposition and
2 biological availability of diesel particles and their associated mutagenic chemicals. Environ. Int.
3 5: 263-267.
4
5 Chan, T. L.; Lee, P. S.; Hering, W. E. (1981) Deposition and clearance of inhaled diesel exhaust particles in the
6 respiratory tract of Fischer rats. J. Appl. Toxicol. 1: 77-82.
7
8 Chan, T. L.; Lee, P. S.; Hering, W. E. (1984) Pulmonary retention of inhaled diesel particles after prolonged
9 exposures to diesel exhaust. Fundam. Appl. Toxicol. 4: 624-631.
10
11 Cohen, D.; Arai, S. F.; Brain, J. D. (1979) Smoking impairs long-term dust clearance from the lung. Science
12 (Washington, DC) 204: 514-517.
13
14 Crapo, J. D.; Young, S. L.; Fram, E. K.; Pinkerton, K. E.; Barry, B. E.; Crapo, R. O. (1983) Morphometric
15 characteristics of cells in the alveolar region of mammalian lungs. Am. Rev. Respir. Dis. 128: S42-S46.
16
17 Creasia, D. A.; Poggenburg, J. K., Jr.; Nettesheim, P. (1976) Elution of benzo[a]pyrene from carbon particles
18 in the respiratory tract of mice. J. Toxicol. Environ. Health 1: 967-975.
19
20 Cuddihy, R. G.; Yeh, H. C. (1986) Model analysis of respiratory tract clearance of particles inhaled by people.
21 In: Muggenburg, B. A.; Sun, J. D., eds. Annual report of the Inhalation Toxicology Research Institute.
22 Albuquerque, NM: Lovelace Biomedical and Environmental Research Institute; report no. LMF-115;
23 pp. 140-147.
24
25 Dahl, A. R.; Schlesinger, R. B.; Heck, H. D' A.; Medinsky, M. A.; Lucier, G. W. (1991) Comparative
26 dosimetry of inhaled materials: differences among animal species and extrapolation to man. Fundam.
27 Appl. Toxicol. 16: 1-13.
28
29 Evans, M. J.; Shami, S. G.; Martinez, L. A. (1986) Enhanced proliferation of pulmonary alveolar macrophages
30 after carbon instillation in mice depleted of blood monocytes by Strontium-89. Lab. Invest. 54: 154-159.
31
32 Felicetti, S. A.; Wolff, R. K.; Muggenburg, B. A. (1981) Comparison of trachea! mucous transport in rats,
33 guinea pigs, rabbits, and dogs. J. Appl. Physiol.: Respir. Environ. Exercise Physiol. 51: 1612-1617.
34
35 Ferin, J. (1982) Pulmonary alveolar pores and alveolar macrophage-mediated particle clearance. Anat. Rec.
36 203: 265-272.
37
38 Ferin, J.; Feldstein, M. L. (1978) Pulmonary clearance and hilar lymph node content in rats after particle
39 exposure. Environ. Res. 16: 342-352.
40
41 Freedman, A. P.; Robinson, S. E. (1988) Noninvasive magnetopneumographic studies of lung dust retention and
42 clearance in coal miners. In: Frantz, R. L.; Ramani, R. V., eds. Respirable dust in the mineral
43 industries: health effects, characterization and control. University Park, PA: Perm State University Press;
44 pp. 181-186.
45
46 Frey, J. W.; Corn, M. (1967) Physical and chemical characteristics of particulates in a diesel exhaust. Am. Ind.
47 Hyg. Assoc. J. 28: 468-478.
48
49 George, G.; Hook, G. E. R. (1984) The pulmonary extracellular lining. Environ. Health Perspect. 55: 227-237.
50
51 Gerde, P.; Scholander, P. (1987) A hypothesis concerning asbestos carcinogenicity: the migration of lipophilic
52 carcinogens in adsorbed lipid bilayers. Ann. Occup. Health 31: 395-400.
53
December 1994 4-36 DRAFT-DO NOT QUOTE OR CITE
-------
1 Gerde, P.; Medinsky, M. A.; Bond, J. A. (1991a) Particle-associated polycyclic aromatic hydrocarbons—a
2 reappraisal of their possible role in pulmonary carcinogenesis. Toxicol. Appl. Pharmacol. 108: 1-13.
3
4 Gerde, P.; Medinsky, M. A.; Bond, J. A. (1991b) The retention of polycyclic aromatic hydrocarbons in the
5 bronchial airways and in the alveolar region—a theoretical comparison. Toxicol. Appl. Pharmacol.
6 107: 239-252.
7
8 Gerde, P.; Muggenburg, B. A.; Hoover, M. D.; Henderson, R. F. (1993a) Disposition of polycyclic aromatic
9 hydrocarbons in the respiratory tract of the beagle dog: I. the alveolar region. Toxicol. Appl. Pharmacol.
10 121:313-318.
11
12 Gerde, P.; Muggenburg, B. A.; Sabourin, P. A.; Harkema, J. R.; Hotchkiss, J. A.; Hoover, M. D.; Henderson,
13 R. F. (1993b) Disposition of polycyclic aromatic hydrocarbons in the respiratory tract of the beagle dog:
14 II. the conducting airways. Toxicol. Appl. Pharmacol. 121: 319-327.
15
16 Gerde, P.; Muggenburg, B. A.; Henderson, R. F. (1993c) Disposition of polycyclic aromatic hydrocarbons in
17 the respiratory tract of the beagle dog: III. mechanisms of the dosimetry. Toxicol. Appl. Pharmacol.
18 121:328-334.
19
20 Gibb, F. R.; Morrow, P. E. (1962) Alveolar clearance in dogs after inhalation of an iron 59 oxide aerosol.
21 J. Appl. Physiol. 17: 429-432.
22
23 Godleski, J. J.; Stearns, R. C.; Katler, M. R.; Brain, J. D. (1988) Particle dissolution in alveolar macrophages
24 assessed by electron energy loss analysis using the Zeiss CEM902 electron microscope. J. Aerosol Med.
25 1: 198.
26
27 Griffis, L. C.; Wolff, R. K.; Henderson, R. F.; Griffith, W. C.; Mokler, B. V.; McClellan, R. O. (1983)
28 Clearance of diesel soot particles from rat lung after a subchronic diesel exhaust exposure. Fundam.
29 Appl. Toxicol. 3: 99-103.
30
31 Gurley, L. R.; Spall, W. D.; Valdez, J. G.; London, J. E.; Dethloff, L. A.; Lehnert, B. E. (1988) An HPLC
32 procedure for the analysis of proteins in lung lavage fluid. Anal. Biochem. 172: 465-478.
33
34 Harmsen, A. G.; Muggenburg, B. A.; Snipes, M. B.; Bice, D. E. (1985) The role of macrophages in particle
35 translocation from lungs to lymph nodes. Science (Washington, DC) 230: 1277-1280.
36
37 Heinrich, U.; Muhle, H.; Takenaka, S.; Ernst, H.; Fuhst, R.; Mohr, U.; Pott, F.; Stober, W. (1986) Chronic
38 effects on the respiratory tract of hamsters, mice, and rats after long-term inhalation of high
39 concentrations of filtered and unfiltered diesel engine emissions. J. Appl. Toxicol. 6: 383-395.
40
41 Heppleston, A. G. (1961) Observations on the disposal of inhaled dust by means of the double exposure
42 technique. In: Davies, C. N., ed. Inhaled particles and vapours: proceedings of an international
43 symposium; March-April 1960; Oxford, United Kingdom. New York, NY: Pergamon Press-
44 pp. 320-326.
45
46 Heppleston, A. G.; Young, A. E. (1973) Uptake of inert paniculate matter by alveolar cells: an ultrastructural
47 study. J. Pathol. Ill: 159-164.
48
49 Iwai, K.; Udagawa, T.; Yamagishi, M.; Yamada, H. (1986) Long-term inhalation studies of diesel exhaust on
50 F344 SPF rats. Incidence of lung cancer and lymphoma. In: Ishinishi, N.; Koizumi, A.; McClellan,
51 R. O.; Stober, W., eds. Carcinogenic and mutagenic effects of diesel engine exhaust: proceedings of the
52 international satellite symposium on lexicological effects of emissions from diesel engines; July; Tsukuba
53 Science City, Japan. Amsterdam, Holland: Elsevier Science Publishers B. V.; pp. 349-360.
54 (Developments in toxicology and environmental science: v. 13).
December 1994 4.37 DRAFT-DO NOT QUOTE OR CITE
-------
1 Kane, L. E.; Alarie, Y. (1978) Evaluation of sensory irritation from acrolein-formaldehyde mixtures. Am. Ind.
2 Hyg. Assoc. J. 39: 270-274.
3
4 Kane, L. E.; Alarie, Y. (1979) Interactions of sulfur dioxide and acrolein as sensory irritants. Toxicol. Appl.
5 Pharmacol. 48: 305-315.
6
7 King, L. C.; Kohan, M. J.; Austin, A. C.; Claxton, L. D.; Huisingh, J. L. (1981) Evaluation of the release of
8 mutagens from diesel particles in the presence of physiological fluids. Environ. Mutagen. 3: 109-121.
9
10 Kittelson, D. B.; Dolan, D. F.; Verrant, J. A. (1978) Investigation of a diesel exhaust aerosol. Warrendale, PA:
11 Society of Automotive Engineers, Inc.; report no. CONF-780208-22.
12
13 Lee, P. S.; Chan, T. L.; Hering, W. E. (1983) Long-term clearance of inhaled diesel exhaust particles in
14 rodents. J. Toxicol. Environ. Health 12: 801-813.
15
16 Lee, K. P.; Trochimowicz, H. J.; Remhardt, C. F. (1985) Transmigration of titanium dioxide (TiO2) particles in
17 rats after inhalation exposure. Exp. Mol. Pathol. 42: 331-343.
18
19 Lee, K. P.; Henry, N. W., Ill; Trochimowicz, H. J.; Reinhardt, C. F. (1986) Pulmonary response to impaired
20 lung clearance in rats following excessive TiO2 dust deposition. Environ. Res. 41: 144-167.
21
22 Lee, K. P.; Ulrich, C. E.; Geil, R. G.; Trochimowicz, H. J. (1988) Effects of inhaled chromium dioxide dust on
23 rats exposed for two years. Fundam. Appl. Toxicol. 10: 125-145.
24
25 Lehnert, B. E. (1988) Distribution of particles in alveolar macrophages during lung clearance. J. Aerosol Med.
26 1: 206.
27
28 Lehnert, B. E. (1989) Rates of disappearance of alveolar macrophages during lung clearance as a function of
29 phagocytized paniculate burden [abstract]. Am. Rev. Respir. Dis. 139(suppl.): A161.
30
31 Lehnert, B. E.; Morrow, P. E. (1985) Association of 59iron oxide with alveolar macrophages during alveolar
32 clearance. Exp. Lung Res. 9: 1-16.
33
34 Lehnert, B. E.; Valdez, Y. E.; Stewart, C. C. (1986) Translocation of particles to the tracheobronchial lymph
35 nodes after lung deposition: kinetics and particle-cell relationships. Exp. Lung Res. 10: 245-266.
36
37 Lehnert, B. E.; Cline, A.; London, J. E. (1989) Kinetics of appearance of polymorphonuclear leukocytes and
38 their particle burdens during the alveolar clearance of a high lung burden of particles. Toxicologist 9: 77.
39
40 Lippmann, M.; Schlesinger, R. B. (1984) Interspecies comparisons of particle deposition and mucociliary
41 clearance in tracheobronchial airways. J. Toxicol. Environ. Health 13: 441-469.
42
43 Lundborg, M.; Lind, B.; Camner, P. (1984) Ability of rabbit alveolar macrophages to dissolve metals. Exp.
44 Lung Res. 7: 11-22.
45
46 Marafante, E.; Lundborg, M.; Vahter, M.; Camner, P. (1987) Dissolution of two arsenic compounds by rabbit
47 alveolar macrophages in vitro. Fundam. Appl. Toxicol. 8: 382-388.
48
49 McClellan, R. 0.; Brooks, A. L.; Cuddihy, R. G.; Jones, R. K.; Mauderly, J. L.; Wolff, R. K. (1982)
50 Inhalation toxicology of diesel exhaust particles. In: Lewtas, J., ed. Toxicological effects of emissions
51 from diesel engines: proceedings of the Environmental Protection Agency diesel emissions symposium;
52 October 1981; Raleigh, NC. New York, NY: Elsevier Biomedical; pp. 99-120. (Developments in
53 toxicology and environmental science: v. 10).
54
December 1994 4-38 DRAFT-DO NOT QUOTE OR CITE
-------
1 Morrow, P. E. (1988) Possible mechanisms to explain dust overloading of the lungs. Fundam. Appl. Toxicol.
2 10: 369-384.
3
4 Muhle, H.; Bellman, B.; Creutzenberg, O.; Heinrich, U.; Keckar, M.; Mermelstein, R. (1990) Dust overloading
5 of lungs after exposure of rats to particles of low solubility: comparative studies. J. Aerosol Sci.
6 21: 374-377.
7
8 Nilsen, A.; Nyberg, K.; Camner, P. (1988) Intraphagosomal pH in alveolar macrophages after phagocytosis
9 in vivo and in vitro of fluorescein-labeled yeast particles. Exp. Lung Res. 14: 197-207.
10
11 Oberdorster, G.; Ferin, J.; Morrow, P. E. (1992) Volumetric loading of alveolar macrophages (AM): a possible
12 basis for diminished AM-mediated particle clearance. Exp. Lung Res. 18: 87-104.
13
14 Ohkuma, S.; Poole, B. (1978) Fluorescence probe measurement of the intralysosomal pH in living cells and the
15 perturbation of pH by various agents. Proc. Natl. Acad. Sci. U.S. A. 75: 3327-3331.
16
17 Pepelko, W. E. (1987) Feasibility of dose adjustment based on differences in long-term clearance rates of inhaled
18 particulate matter in humans and laboratory animals. Regul. Toxicol. Pharmacol. 7: 236-252.
19
20 Phalen, R. F.; Oldham, M. J. (1983) Tracheobronchial airway structure as revealed by casting techniques.
21 Am. Rev. Respir. Dis. 128: S1-S4.
22
23 Powdrill, J.; Buckley, C.; Valdez, Y. E.; Lehnert, B. E. (1989) Airway intra-luminal macrophages: origin and
24 role in lung clearance. Toxicologist 9: 77.
25
26 Pritchard, J. N. (1989) Dust overloading causes impairment of pulmonary clearance: evidence from rats and
27 humans. Exp. Pathol. 37: 39-42.
28
29 Raabe, 0. G. (1982) Deposition and clearance of inhaled aerosols. In: Witschi, H., ed. Mechanisms in
30 respiratory toxicology. Boca Raton, FL: CRC Press; pp. 27-76.
31
32 Raabe, O. G.; Al-Bayati, M. A.; league, S. V.; Rasolt, A. (1988) Regional deposition of inhaled monodisperse
33 coarse and fine aerosol particles in small laboratory animals. In: McCallum, R. I., ed. Inhaled particles
34 VI: proceedings of an international meeting. Cambridge, United Kingdom: British Occupational Hygiene
35 Society.
36
37 Riley, P. A.; Dean, R. T. (1978) Phagocytosis of latex particles in relation to the cell cycle in 3T3 cells. Exp.
38 Cell Biol. 46: 367-373.
39
40 Roser, B. (1970) The origins, kinetics, and fate of macrophage populations. J. Reticuloendothel. Soc.
41 8: 139-161.
42
43 Saffiotti, U.; Borg, S. A.; Grote, M. I.; Karp, D. B. (1964) Retention rates of particulate carcinogens in the
44 lungs. Chicago Med. Sch. Q. 24: 10-17.
45
46 Sandusky, C. B.; Cowden, M. W.; Schwartz, S. L. (1977) Effect of particle size on regurgitative exocytosis by
47 rabbit alveolar macrophages. In: Sanders, C. L.; Schneider, R. P.; Dagle, G. E.; Ragan, H. A., eds.
48 Pulmonary macrophage and epithelial cells: proceedings of the sixteenth annual Hanford biology
49 symposium; September 1976; Richland, WA. Washington, DC: Energy Research and Development
50 Administration; pp. 85-105. (ERDA symposium series: no. 43). Available from: NTIS, Springfield, VA;
51 CONF-760927.
52
53 Schlesinger, R. B. (1985) Comparative deposition of inhaled aerosols in experimental animals and humans:
54 a review. J. Toxicol. Environ. Health 15: 197-214.
December 1994 4.39 DRAFT-DO NOT QUOTE OR CITE
-------
1 Schum, M.; Yeh, H.-C. (1980) Theoretical evaluation of aerosol deposition in anatomical models of mammalian
2 lung airways. Bull. Math. Biol. 42: 1-15.
3
4 Siak, J. S.; Chan, T. L.; Lee, P. S. (1980) Diesel particulate extracts in bacterial test systems. In: Pepelko,
5 W. E.; Danner, R. M.; Clarke, N. A., eds. Health effects of diesel engine emissions: proceedings of an
6 international symposium, v. 1; December 1979; Cincinnati, OH. Cincinnati, OH: U.S. Environmental
7 Protection Agency, Health Effects Research Laboratory; pp. 245-262; EPA report no.
8 EPA-600/9-80-057b. Available from: NTIS, Springfield, VA; PB81-173809.
9
10 Snipes, M. B.; Clem, M. F. (1981) Retention of microspheres in the rat lung after intratracheal instillation.
11 Environ. Res. 24: 33-41.
12
13 Snipes, M. B.; McClellan, R. O. (1984) Disposition and retention of monodisperse aluminosilicate particles
14 inhaled by guinea pigs. In: Annual report of the Inhalation Toxicology Research Institute. Albuquerque,
15 NM: Lovelace Biomedical and Environmental Research Institute; report no. LMF-113; pp. 80-83.
16
17 Snipes, M. B.; Olson, T. R.; Yeh, H. C. (1988) Deposition and retention patterns for 3-, 9-, and 15-pm latex
18 microspheres inhaled by rats and guinea pigs. Exp. Lung Res. 14: 37-50.
19
20 Sorokin, S. P.; Brain, J. D. (1975) Pathways of clearance in mouse lungs exposed to iron oxide aerosols. Anat.
21 Rec. 181: 581-625.
22
23 Stahlhofen, W.; Gebhart, J.; Heyder, J. (1980) Experimental determination of the regional deposition of aerosol
24 particles in the human respiratory tract. Am. Ind. Hyg. Assoc. J. 41: 385-398a.
25
26 Stahlhofen, W.; Hoebrich, R.; Rudolf, G.; Scheuch, G. (1990) Short-term and long-term clearance of particles
27 from the upper human respiratory tract as a function of particle size. J. Aerosol Sci. 21(suppl. 1):
28 5407-5410.
29
30 Stober, W. (1986) Experimental induction of tumors in hamsters, mice and rats after long-term inhalation of
31 filtered and unfiltered diesel engine exhaust. In: Ishinishi, N.; Koizumi, A.; McClellan, R. O.;
32 Stober, W., eds. Carcinogenic and mutagenic effects of diesel engine exhaust: proceedings of the
33 international satellite symposium on lexicological effects of emissions from diesel engines; July; Tsukuba
34 Science City, Japan. Amsterdam, The Netherlands: Elsevier Science Publishers B. V.; pp. 421-439.
35 (Developments in toxicology and environmental science: v. 13).
36
37 Stober, W.; Einbrodt, H. J.; Klosterkotter, W. (1967) Quantitative studies of dust retention in animal and human
38 lungs after chronic inhalation. In: Davies, C. N., ed. Inhaled particles and vapours II: proceedings of an
39 international symposium; September-October 1965; Cambridge, United Kingdom. Oxford, United
40 Kingdom: Pergamon Press; pp. 409-418.
41
42 Strom, K. A. (1984) Response of pulmonary cellular defenses to the inhalation of high concentrations of diesel
43 exhaust. J. Toxicol. Environ. Health 13: 919-944.
44
45 Sun, J. D.; Wolff, R. K.; Kanapilly, G. M. (1982) Deposition, retention, and biological fate of inhaled
46 benzo(a)pyrene adsorbed onto ultrafine particles and as a pure aerosol. Toxicol. Appl. Pharmacol.
47 65: 231-244.
48
49 Sun, J. D.; Wolff, R. K.; Kanapilly, G. M.; McClellan, R. O. (1984) Lung retention and metabolic fate of
50 inhaled benzo(o)pyrene associated with diesel exhaust particles. Toxicol. Appl. Pharmacol. 73: 48-59.
51
52 Task Group on Lung Dynamics. (1966) Deposition and retention models for internal dosimetry of the human
53 respiratory tract. Health Phys. 12: 173-207.
54
December 1994 4-40 DRAFT-DO NOT QUOTE OR CITE
-------
1 Ulrich, C. E.; Haddock, M. P.; Alarie, Y. (1972) Airborne chemical irritants: role of the trigeminal nerve.
2 Arch. Environ. Health 24: 37-42.
3
4 Vastag, E.; Matthys, H.; Zsamboki, G.; Kohler, D.; Daikeler, G. (1986) Mucociliary clearance in smokers.
5 Eur. J. Respir. Dis. 68: 107-113.
6
7 Vostal, J. J. (1983) Unavailability and biotransformation of the mutagenic component of particulate emissions
8 present in motor exhaust samples. Environ. Health Perspect. 47: 269-281.
9
10 Vostal, J. J.; Schreck, R. M.; Lee, P. S.; Chan, T. L.; Soderholm, S. C. (1982) Deposition and clearance of
11 diesel particles from the lung. In: Lewtas, J., ed. lexicological effects of emissions from diesel engines:
12 proceedings of the 1981 EPA diesel emissions symposium; October 1981; Raleigh, NC. New York, NY:
13 Elsevier Biomedical; pp. 143-159. (Developments in toxicology and environmental science: v. 10).
14
15 Weibel, E. R. (1963) Morphometry of the human lung. New York, NY: Academic Press Inc.
16
17 White, H. J.; Garg, B. D. (1981) Early pulmonary response of the rat lung to inhalation of high concentration of
18 diesel particles. J. Appl. Toxicol. 1: 104-110.
19
20 Wolff, R. K.; Gray, R. L. (1980) Tracheal clearance of particles. In: Diel, J. H.; Bice, D. E.; Martinez B. S.;
21 eds. Inhalation Toxicology Research Institute annual report: 1979-1980. Albuquerque, NM: Lovelace
22 Biomedical and Environmental Research Institute; p. 252; report no. LMF-84.
23
24 Wolff, R. K.; Henderson, R. F.; Snipes, M. B.; Sun, J. D.; Bond, J. A.; Mitchell, C. E.; Mauderly, J. L.;
25 McClellan, R. O. (1986) Lung retention of diesel soot and associated organic compounds.
26 In: Ishinishi, N.; Koizumi, A.; McClellan, R.; Stober, W., eds. Carcinogenic and mutagenic effects of
27 diesel engine exhaust: proceedings of the international satellite symposium on lexicological effects of
28 emissions from diesel engines; July; Tsukuba Science City, Japan. Amsterdam, The Netherlands: Elsevier
29 Science Publishers B. V.; pp. 199-211. (Developments on toxicology and environmental science: v. 13).
30
31 Wolff, R. K.; Henderson, R. F.; Snipes, M. B.; Griffith, W. C.; Mauderly, J. L.; Cuddihy, R. G.; McClellan,
32 R. O. (1987) Alterations in particle accumulation and clearance in lungs of rats chronically exposed to
33 diesel exhaust. Fundam. Appl. Toxicol. 9: 154-166.
34
35 Wright, E. S. (1986) Effects of short-term exposure to diesel exhaust on lung cell proliferation and phospholipid
36 metabolism. Exp. Lung Res. 10: 39-55.
37
38 Wright, J. R.; Clements, J. A. (1987) Metabolism and turnover of lung surfactant. Am. Rev. Respir. Dis.
39 136: 426-444.
40
41 Xu, G. B.; Yu, C. P. (1987) Deposition of diesel exhaust particles in mammalian lungs: a comparison between
42 rodents and man. Aerosol Sci. Technol. 7: 117-123.
43
44 Yu, C. P.; Yoon, K. J. (1988) Determination of lung doses of diesel exhaust perticulates. Presented at:
45 5th annual conference of Health Effects Institute; Colorado Springs, CO.
December 1994 4.41 DRAFT-DO NOT QUOTE OR CITE
-------
-------
i 5. NONCANCER HEALTH EFFECTS
2 OF DIESEL EXHAUST
3
4
5 The objective of this chapter is to evaluate the noncarcinogenic health effects of diesel
6 exhaust. Data pertaining to exposures to whole diesel exhaust will be presented first,
7 followed by a comparison of the effects of filtered and unfiltered exhaust. Filtered exhaust
8 consists of the gaseous components of the exhaust without the associated paniculate matter.
9
10
11 5.1 HEALTH EFFECTS OF WHOLE EXHAUST
12 5.1.1 Human Data
13 5.1.1.1 Short-Term Exposures
14 Kahn et al. (1988) reported the occurrence of 13 cases of acute overexposure to diesel
15 exhaust among Utah and Colorado coal miners. Twelve miners had symptoms of mucous
16 membrane irritation, headache, and lightheadedness. Eight individuals reported nausea; four
17 reported a sensation of unreality; four reported heartburn; three reported weakness,
18 numbness, and tingling in their extremities; three reported vomiting; two reported chest
19 tightness; and two others reported wheezing. Each miner lost time from work because of
20 these symptoms, which resolved within 24 to 48 h. No air monitoring data were presented;
21 poor work practices were described as the predisposing conditions for overexposure.
22 El Batawi and Noweir (1966) reported that among 161 workers from two garages
23 where diesel-powered buses were serviced and repaired, 42% complained of eye irritation,
24 37% of headaches, 30% of dizziness, 19% of throat irritation, and 11% of cough and
25 phlegm. Ranges of mean concentrations of diesel exhaust components in the two diesel bus
26 garages were as follows: 0.4 to 1.4 ppm N02, 0.13 to 0.81 ppm SO2, 0.6 to 44.1 ppm
27 aldehydes, and 1.34 to 4.51 mg/m3 of particulate matter; the highest concentrations were
28 obtained close to the exhaust systems of the buses.
29 Eye irritation was reported by Battigelli (1965) in six subjects after 40 s of chamber
30 exposure to diluted diesel exhaust containing 4.2 ppm NO2, 1 ppm SO2, 55 ppm CO,
December 1994 54 DRAFT-DO NOT QUOTE OR CITE
-------
1 3.2 ppm total hydrocarbons, and 1 to 2 ppm total aldehydes; after 3 min and 20 s of
2 exposure to diluted diesel exhaust containing 2.8 ppm NO2, 0.5 ppm SO2, 30 ppm CO,
3 2.5 ppm total hydrocarbons, and < 1 to 2 ppm total aldehydes; and after 6 min of exposure
4 to diluted diesel exhaust containing 1.3 ppm NO2, 0.2 ppm SO2, <20 ppm CO, <2.0 ppm
5 total hydrocarbons, and < 1.0 ppm total aldehydes. The concentration of the exhaust
6 particles was not reported.
7 Katz et al. (1960) described the experience of 14 chemists and their assistants
8 monitoring the environment of a train tunnel utilized by diesel-powered locomotives.
9 Although workers complained on three occasions of minor eye and throat irritation, no
10 correlation was established with concentrations of any particular component of diesel exhaust.
11
12 5.1.1.2 Diesel Exhaust Odor
13 The odor of diesel exhaust is considered by most people to be objectionable; at high
14 intensities, it may produce sufficient physiological and psychological effects to warrant
15 concern about public health. Strong unpleasant and irritating odors may cause nausea,
16 headache, loss of appetite, psychological stress, and other health effects. The intensity of the
17 odor of diesel exhaust is an exponential function of its concentration such that a tenfold
18 change in the concentration will alter the intensity of the odor by one unit. Two human
19 panel rating scales have been used to measure diesel exhaust odor intensity. In the first
20 (Turk, 1967), combinations of odorous materials were selected to simulate diesel exhaust
21 odor; a set of 12 mixtures, each having twice the concentration of that of the previous
22 mixture, is the basis of the diesel odor intensity scale (D-scale). The second method is the
23 TIA (total intensity of aroma) scale based on seven steps, ranging from 0 to 3, with 0 being
24 undetectable, 1/2 very slight, and 1 slight and increasing in one-half units up to 3, strong
25 (Odor Panel of the CRC-APRAC Program Group on Composition of Diesel Exhaust, 1979;
26 Levins, 1981).
27 Surveys, utilizing volunteer panelists, have been taken to evaluate the general public's
28 response to the odor of diesel exhaust. Hare and Springer (1971) and Hare et al. (1974)
29 found that at a D rating of about 2 (TIA = 0.9, slight odor intensity), about 90% of the
30 participants perceived the odor, and almost 60% found it objectionable. At a D rating of
31 3.2 (TIA = 1.2, slight to moderate odor intensity), about 95% perceived the odor, and 75%
December 1994 5-2 DRAFT-DO NOT QUOTE OR CITE
-------
1 objected to it, and, at a D rating of 5 (TIA =1.8, almost moderate), about 95% objected to
2 it. Linnell and Scott (1962) evaluated the odor threshold for diesel exhaust in six subjects
3 and found that a dilution factor of 140 to 475 was required to reduce the odor to the
4 threshold level.
5
6 Pulmonary and Respiratory Effects
1 Battigelli (1965) exposed 13 volunteers to three dilutions of diesel exhaust obtained
8 from a one-cylinder, four-cycle, 7-hp diesel engine (fuel type unspecified) and found that
9 15-min to 1-h exposures had no significant effects on pulmonary resistance. Pulmonary
10 resistance was measured by plethysmography utilizing the simultaneous recording of
11 esophageal pressure and air flow determined by electrical differentiation of the volume signal
12 from a spirometer. The units of concentration of the constituents in the three diluted
13 exhausts were 1.3, 2.8, and 4.2 ppm NO2; 0.2, 0.5, and 1 ppm SO2, <20, 30, and 55 ppm
14 CO; and < 1.0, < 1 to 2, and 1 to 2 ppm total aldehydes, respectively. Particle
15 concentrations were not reported.
16 A number of studies have evaluated changes in pulmonary function occurring over a
17 workshift in workers occupationally exposed to diesel exhaust (specific time period not
18 always reported but assumed to be 8 h). In a study of coal miners, Reger et al. (1978) found
19 that both forced expiratory volume in 1 s (FEVj) and forced vital capacity (FVC) decreased
20 by 0.05 L in 60 diesel-exposed miners, an amount not substantially different from reductions
21 seen in non-diesel-exposed miners (0.02 and 0.04 L, respectively). Decrements in peak
22 expiratory flow rates were similar between diesel and non-diesel exhaust-exposed miners.
23 Miners with a history of smoking had an increased number of decrements over the shift than
24 did nonsmokers. Although the monitoring data were not reported, the authors stated that
25 there was no relationship between the low concentrations of measured respirable dust or NO2
26 (personal samplers) when compared with shift changes for any lung function parameter
27 measured for the diesel-exposed miners. This study is limited because results were
28 preliminary (abstract) and there was incomplete information on the control subjects.
29 Ames et al. (1982) compared the pulmonary function of 60 coal miners exposed to
30 diesel exhaust with that of a control group of 90 coal miners not exposed to diesel exhaust
31 for evidence of acute respiratory effects associated with exposure to diesel exhaust. Changes
December 1994 5.3 DRAFT-DO NOT QUOTE OR CITE
-------
1 over the workshift in FVC, FEVj, and forced expiratory flow rate at 50% FVC (FEF50)
2 were the indices for acute respiratory effects. The environmental concentrations of the
3 primary exhaust pollutants were 2.0 mg/m3 paniculate matter (< 10/mi, MMAD), 0.2 ppm
4 NO2, 12 ppm CO, and 0.3 ppm formaldehyde. The investigators reported a statistically
5 significant decline in FVC and FEVj over the workshift in both the diesel-exposed and
6 comparisons group. Current smokers had greater decrements in FVC, FEVlf and FEF50
7 than exsmokers and nonsmokers. There was a marked disparity between the ages and the
8 time spent underground for the two study groups. Diesel-exposed miners were about
9 15 years younger and had worked underground for 15 fewer years (4.8 versus 20.7 years)
10 than miners not exposed to diesel exhaust. The significance of these differences on the
11 results is difficult to ascertain because the responsiveness of airway smooth muscle is not
12 age-dependent, whereas predisposing pathology such as bronchitis or the formation of
13 macules or nodular masses proximal to small airways can affect the mechanical properties of
14 the airways.
15 Except for the expected differences related to age, 120 underground iron ore miners
16 exposed to diesel exhaust had no workshift changes in FVC and F£V! when compared with
17 120 matched surface miners (Jorgensen and Svensson, 1970). Both groups had equal
18 numbers (30) of smokers and nonsmokers. The frequency of bronchitis was higher among
19 underground workers, much higher among smokers than nonsmokers, and also higher among
20 older than younger workers. The authors reported that the underground miners had
21 exposures of 0.5 to 1.5 ppm NO2 and between 3 and 9 mg/m3 paniculate matter with 20 to
22 30% of the particles <5/mi, MMAD. The majority of the particles were iron ore; quartz
23 was 6 to 7% of the fraction <5 /un, MMAD.
24 Gamble et al. (1978) measured preshift FEVt and FVC in 187 salt miners and
25 obtained peak flow forced expiratory flow rates at 25, 50, and 75% of FVC (FEF25, FEF50,
26 or FEF75). Postshift pulmonary function values were determined from total lung capacity
27 and flows at preshift percentages of FVC. The miners were exposed to mean NO2 levels of
28 1.5 ppm and mean respirable particulate levels of 0.7 mg/m3. No statistically significant
29 changes were found between changes in pulmonary function and in NO2 and respirable
30 particles combined. Slopes of the regression of NO2 and changes in FEVj, FEF25, FEF50,
31 and FEF75 were significantly different from zero. The authors concluded that these small
December 1994 5.4 DRAFT-DO NOT QUOTE OR CITE
-------
1 reductions in pulmonary function were attributable to variations in NO2 within each of the
2 five salt mines that contributed to the cohort.
3 Gamble et al. (1987a) investigated the acute effects of diesel exhaust in 232 workers
4 in four diesel bus garages using an acute respiratory questionnaire and before and after
5 workshift spirometry. The prevalence of burning eyes, headaches, difficult or labored
6 breathing, nausea, and wheeze experienced at work was higher in the diesel bus garage
7 workers than in a comparison population of lead/acid battery workers who had not previously
8 shown a statistically significant association of acute symptoms with acid exposure.
9 Comparisons between the two groups were made without adjustment for age and smoking.
10 There was no detectable association of exposure to NO2 (0.23 ppm ± 0.24 S.D.) or
11 inhalable (less than 10 /mi MMAD) particles (0.24 mg/m3 ± 0.26 S.D.) and acute
12 reductions in FVC, FEVj, peak flows, FEF50, and FEF75. Workers who had respiratory
13 symptoms had slightly greater but statistically insignificant reductions in FEVj and FEF50.
14 Ulfvarson et al. (1987) evaluated workshift changes in the pulmonary function of
15 17 bus garage workers, 25 crew members of two types of car ferries, and 37 workers on
16 roll-on/roll-off ships. The latter group was exposed primarily to diesel exhaust; the first two
17 groups were exposed to both gasoline and diesel exhausts. The diesel-only exposures that
18 averaged 8 h, consisted of 0.13 to 1.0 mg/m3 paniculate matter, 0.02 to 0.8 mg/m3
19 (0.016 to 0.65 ppm) NO, 0.06 to 2.3 mg/m3 (0.03 to 1.2 ppm) NO2, 1.1 to 5.1 mg/m3
20 (0.96 to 4.45 ppm) CO, and up to 0.5 mg/m3 (0.4 ppm) formaldehyde. The largest
21 decrement in pulmonary function was observed during a workshift following no exposure to
22 diesel exhaust for 10 days. Forced vital capacity and FEVj were significantly reduced over
23 the workshift (0.44 L and 0.30 L, p < 0.01 and p < 0.001, respectively). There was no
24 difference between smokers and nonsmokers. Maximal midexpiratory flow, closing volume
25 expressed as the percentage of expiratory vital capacity, and alveolar plateau gradient
26 (phase 3) were not affected. Similar, but less pronounced, effects on FVC (-0.16 L) were
27 found in a second, subsequent study of stevedores (n = 24) only following 5 days of no
28 exposure to diesel truck exhaust. Pulmonary function returned to normal after 3 days
29 without occupational exposure to diesel exhaust. No exposure-related correlation was found
30 between the observed pulmonary effects and concentrations of NO, NO2, CO, or
31 formaldehyde; however, it was suggested that NO2 adsorbed onto the diesel exhaust particles
December 1994 5.5 DRAFT-DO NOT QUOTE OR CITE
-------
1 may have contributed to the overall dose of NO2 to the lungs. In a related study, six
2 workers (job category not defined) were placed in an exposure chamber and exposed to
3 diluted diesel exhaust containing 0.6 mg/m3 paniculate matter and 3.9 mg/m3 (2.1 ppm)
4 NO2. The exhaust was generated by a 6-cylinder, 2.38-L diesel engine, operated for 3 h and
5 40 min without interruption at constant speed, equivalent to 60 km/h, and at about one-half
6 full engine load. No effect on pulmonary function was observed.
7
8 5.1.1.3 Long-Term Exposure
9 Several epidemiologic studies have evaluated the effects of chronic exposure to diesel
10 exhaust on occupationally exposed workers. Battigelli et al. (1964) measured several indices
11 of pulmonary function, including vital capacity, FEVj, peak flow, nitrogen washout, and
12 diffusion capacity in 210 locomotive repairmen exposed to diesel exhaust in three engine
13 houses. The average exposure of these locomotive repairmen to diesel exhaust was
14 9.6 years. When compared with a control group matched for age, body size, "past
15 extrapulmonary medical history" (no explanation given), and job status (154 railroad yard
16 workers), no significant clinical differences were found in pulmonary function nor in the
17 prevalence of dyspnea, cough, or sputum between the diesel exhaust-exposed and nonexposed
18 groups. Exposure to the diesel exhaust showed marked seasonal variations because the doors
19 of the engine house were open in the summer and closed in the winter. For the exposed
20 group, the maximum daily workplace concentrations of air pollutants measured were 1.8 ppm
21 NO2, 1.7 ppm total aldehydes, 0.15 ppm acrolein, 4.0 ppm SO2, and 5.0 ppm total
22 hydrocarbons. The concentration of airborne particles was not reported.
23 Gamble et al. (1987b) examined 283 diesel bus garage workers from four garages in
24 two cities to determine if there was excess chronic respiratory morbidity associated with
25 exposure to diesel exhaust. Tenure was used as a surrogate of exposure; mean tenure of the
26 study population was 9 years ± 10 years S.D. Exposure-effect relationships within the study
27 population showed no detectable associations of symptoms with tenure. Reductions in FVC,
28 FEVi, peak flow, and FEF50 (but not FEF75) were associated with increasing tenure. When
29 compared with a control population (716 nonexposed blue collar workers) and after indirect
30 adjustment for age, race, and smoking, the exposed workers had a higher incidence of
31 cough, phlegm, and wheezing; however, there was no correlation between symptoms and
December 1994 5-6 DRAFT-DO NOT QUOTE OR CITE
-------
1 length of employment. Dyspnea showed an exposure-response trend but no apparent increase
2 in prevalence. Mean FEVl5 FVC, FEF50, and peak flow were not reduced in the total
3 cohort compared with the reference population but were reduced in workers with 10 years or
4 more tenure.
5 Purdham et al. (1987) evaluated respiratory symptoms and pulmonary function in
6 17 stevedores employed in car ferry operations that were exposed to both diesel and gasoline
7 exhausts and in a control group of 11 on-site office workers. Twenty-four percent of the
8 exposed group and 36% of the controls were smokers. If a particular symptom was
9 considered to be influenced by smoking, smoking status was used as a covariate in the
10 logistic regression analysis; pack-years smoked was a covariate for lung function indices.
11 The frequency of respiratory symptoms was not significantly different between the two
12 groups; however, baseline pulmonary function measurements were significantly different.
13 The latter comparisons were measured by multiple regression analysis using the actual (not
14 percentage predicted) results and correcting for age, height, and pack-years smoked. The
15 stevedores had significantly lower FEVj, FEVj/FVC, FEF50 and FEF75 (p < 0.021,
16 p < 0.023, p < 0.001, and p < 0.008, respectively) but not FVC. The results from the
17 stevedores were also compared with those obtained from a study of the respiratory health
18 status of Sydney, Nova Scotia, residents. These comparisons showed that the dock workers
19 had higher FVC, similar FEVj, but lower FEVj/FVC and flow rates than the residents of
20 Sydney. Based on these consistent findings, the authors concluded that the lower baseline
21 function measurements in the stevedores provided evidence of an obstructive ventilatory
22 defect but caution in interpretation was warranted because of the small sample size. There
23 were no significant changes in lung function over the work shift, nor was there a difference
24 between the two groups. The stevedores were exposed to significantly (p < 0.04) higher
25 concentrations of particulate matter (0.06 to 1.72 mg/m3, mean 0.50 mg/m3) than the
26 controls (0.13 to 0.58 mg/m3, mean not reported). Exposures of stevedores to SO2, NO2,
27 aldehydes, and PAHs were very low; occasional CO concentrations in the 20- to 100-ppm
28 range could be detected for periods up to Ih in areas where blockers were chaining gasoline-
29 powered vehicles.
30 Additional epidemiological studies on the health hazards posed by exposure to diesel
31 exhaust have been conducted for mining operations. Reger et al. (1982) evaluated the
December 1994 5.7 DRAFT-DO NOT QUOTE OR CITE
-------
1 respiratory health status of 823 male coal miners from six diesel-equipped mines compared
2 with 823 matched coal miners not exposed to diesel exhaust. The average tenure of
3 underground work for the underground miners and then- controls was only about 5 years; on
4 average, the underground workers in diesel mines spent only 3 of those 5 years underground
5 in diesel-use mines. Underground miners exposed to diesel exhaust reported a higher
6 incidence of symptoms of cough and phlegm but proportionally fewer symptoms of moderate
7 to severe dyspnea than their matched counterparts. These differences in prevalence of
8 symptoms were not statistically significant. The diesel-exposed underground miners, on the
9 average, had lower FVC, FEVl5 FEF50, FEF75, and FEF90 but higher peak flow and FEF25
10 than their matched controls. These differences, however, were not statistically significant.
11 Health indicators for surface workers and their matched controls were directionally the same
12 as for matched underground workers. There were no consistent relationships between the
13 findings of increased respiratory symptoms, decreased pulmonary function, smoking history,
14 years of exposure or monitored atmosphere pollutants (NOX, CO, particles and aldehydes).
15 Mean concentrations of NOX at the six mines ranged from 0 to 0.6 ppm for short-term area
16 samples, 0.13 to 0.28 ppm for full-shift personal samples, and 0.03 to 0.80 for full-shift area
17 samples. Inhalable particles (less than 10 /xm, MMAD) averaged 0.93 to 2.73 mg/m3 for
18 personal samples and 0 to 16.1 for full shift area samples. Ames et al. (1984), using a
19 portion of the miners studied by Reger, examined 280 diesel-exposed underground miners
20 initially in 1977 and again in 1982. Each miner in this group had at least 1 year of under-
21 ground mining work history in 1977. The control group was 838 miners with no exposure to
22 diesel exhaust. The miners were evaluated for the prevalence of respiratory symptoms,
23 chronic cough, phlegm, dyspnea and changes in FVC, FEVl5 and FEF50. No air monitoring
24 data were reported; exposure conditions to diesel exhaust gases and mine dust particles were
25 described as very low. These authors found no decrements in pulmonary function or
26 increased prevalence of respiratory symptoms attributable to exposure to diesel exhaust.
27 In fact, the 5-year incidences of cough, phlegm, and dyspnea were greater in miners without
28 exposure to diesel exhaust than those exposed to diesel exhaust.
29 Attfield (1978) studied 2,659 miners from 21 mines (8 metal, 6 potash, 5 salt, and
30 2 trona). Diesels were employed in only 18 of the mines, but those three mines not using
31 diesels were unidentified. The years of diesel usage, ranging from 8 in trona mines to 16 in
December 1994 5-8 DRAFT-DO NOT QUOTE OR CITE
-------
1 potash mines, were used as a surrogate for exposure to diesel exhaust. Based on a
2 questionnaire, an increased prevalence of persistent cough was associated with exposure to
3 aldehydes; this finding, however, was not supported by the pulmonary function data.
4 No adverse respiratory symptoms or pulmonary function impairments were related to CO2,
5 CO, NO2, inhalable dust, or inhalable quartz. The author failed to comment on whether the
6 prevalence of cough was related to the high incidence, 70%, of smokers in the cohort.
7 Questionnaire, chest radiographs, and spirometric data were collected by
8 Attfield et al. (1982) on 630 potash miners from six potash mines. These miners were
9 exposed for an average of 10 years (range, 5 to 14 years) to 0.1 to 3.3 ppm NO2; 0.1 to
10 4.0 ppm aldehyde; 5 to 9 ppm CO; and total dust concentrations of 9 to 23 mg/m3. The
11 ratio of total to inhalable (< 10 /zm, MMAD) dust ranged from 2 to 11. An increased
12 prevalence of respiratory symptoms was related solely to smoking. No association was found
13 between symptoms and tenure, dust exposure, NO2, CO, or aldehydes. A higher prevalence
14 of symptoms of cough and phlegm was found, but no differences in pulmonary function
15 (FVC and FEVj) were found in these diesel-exposed potash miners when compared with the
16 predicted values derived from a logistics model based on blue-collar workers working in
17 nondusty jobs.
18 Gamble et al. (1983) investigated respiratory morbidity in 259 miners from five salt
19 mines in terms of increased respiratory symptoms, radiographic findings, and reduced
20 pulmonary function associated with exposure to NO2, inhalable particles (< 10 /im, MMAD),
21 or years worked underground. Two of the mines used diesel extensively; no diesels were
22 used in one salt mine. Diesels were introduced into each mine in 1956, 1957, 1963, or
23 1963 through 1967. Several working populations were compared with the salt miner cohort.
24 After adjustment for age and smoking, the salt miners showed no increased prevalence of
25 cough, phlegm, dyspnea, or airway obstruction (FEVj/FVC) compared with aboveground
26 coal miners, potash miners, or blue-collar workers. The underground coal miners
27 consistently had an elevated level of symptoms. Forced expiratory volume at 1 s, FVC,
28 FEF50, and FEF75 were uniformly lower for salt miners in relation to all the comparison
29 populations. There was, however, no association between changes in pulmonary function
30 and years worked, estimated cumulative inhalable particles or estimated NO2 exposure. The
31 highest average exposure to paniculate matter was 1.4 mg/m3 (particle size not reported,
December 1994 5.9 DRAFT-DO NOT QUOTE OR CITE
-------
1 measurement includes NaCl). Mean NO2 exposure was 1.3 ppm, with a range of 0.17 ppm
2 to 2.5 ppm. In a continuation of these studies, Gamble and Jones (1983) grouped the salt
3 miners into low-, intermediate-, and high-exposure categories based on tenure in jobs with
4 diesel exhaust exposure. Average concentrations of inhalable particles and NO2 were 0.40,
5 0.60, and 0.82 mg/m3 and 0.64, 1.77, and 2.21 ppm for the three diesel exposure categories,
6 respectively. A statistically significant concentration-response association was found between
7 the prevalence of phlegm in the salt miners and exposure to diesel exhaust (p < 0.0001) and
8 a similar, but nonsignificant, trend for cough and dyspnea. Changes in pulmonary function
9 showed no association with diesel tenure. In a comparison with the control group of
10 nonexposed, blue-collar workers, adjusted for age and smoking, the overall prevalence of
11 cough and phlegm (but not dyspnea) was elevated in the diesel-exposed workers. Forced
12 expiratory volumes at 1 s and FVC were within 4% of expected, which was considered to be
13 within the normal range of variation for a nonexposed population.
14 In a preliminary study of three subcohorts from bus company personnel (clerks
15 [lowest exposure], bus drivers [intermediate exposure], and bus garage workers [highest
16 exposure]) representing different levels of exposure to diesel exhaust, Edling and Axelson
17 (1984) found a fourfold higher risk ratio for cardiovascular mortality in bus garage workers,
18 even after adjusting for smoking history and allowing for at least 10 years of exposure and
19 15 years or more of induction-latency. Carbon monoxide was hypothesized as the etiologic
20 agent for the increased cardiovascular disease but was not measured. However, in a more
21 comprehensive epidemiological study, Edling et al. (1987) evaluated mortality data covering
22 a 32-year period for a cohort of 694 bus garage employees and found no significant
23 differences between the observed and expected number of deaths from cardiovascular
24 disease. Information on exposure components and their concentrations was not reported.
25 The absence of reported noncancerous human health effects, other than infrequently
26 occurring effects related to respiratory symptoms and pulmonary function changes, is
27 notable. Unlike studies in laboratory animals to be described later in this chapter, studies of
28 the impact of diesel exhaust on the defense mechanisms of the human lung have not been
29 performed. No direct evidence is available in humans regarding doses of diesel exhaust, gas
December 1994 5-10 DRAFT-DO NOT QUOTE OR CITE
-------
1 phase, paniculate phase or total exhaust, that lead to impaired particle clearance or enhanced
2 susceptibility to infection.
3 A summary of epidemiology studies is presented in Table 5-1.
4
5 5.1.2 Animal Studies
6 Because of the large number of statistical comparisons made in the animal studies and
7 to permit uniform, objective evaluations within and among studies, data will be reported as
8 significantly different (i.e., p < 0.05) unless otherwise specified. The exposure regimens
9 used and the resultant exposure conditions employed in the animal inhalation studies are
10 summarized in Appendix A. Other than the pulmonary function studies performed by
11 Wiester et al. (1980) on guinea pigs during their exposure in inhalation chambers, the
12 pulmonary function studies performed by other investigators, although sometimes unreported,
13 were interpreted as being conducted on the following day or thereafter and not immediately
14 following exposure.
15
16 5.1.2.1 Acute Exposures
17 The acute toxicity of undiluted diesel exhaust to rabbits, guinea pigs, and mice was
18 assessed by Pattle et al. (1957). Four engine operating conditions were used, and four
19 rabbits, 10 guinea pigs, and 40 mice were tested under each exposure condition for
20 5 h (no controls were used). Mortality was assessed up to 7 days after exposure. With the
21 engine operating under light load, the exhaust was highly irritating but not lethal to the test
22 species, and only mild tracheal and lung damage was observed in the exposed animals. The
23 exhaust contained 74 mg/m3 paniculate matter (partical size not reported), 560 ppm CO,
24 23 ppm NO2, and 16 ppm aldehydes. Exhaust containing 5 mg/m3 paniculate matter,
25 380 ppm CO, 43 ppm NO2, and 6.4 ppm aldehydes resulted in low mortality rates
26 (mostly below 10%) and moderate lung damage. Exhaust containing 122 mg/m3 paniculate
27 matter, 418 ppm CO, 51 ppm NO2, and 6.0 ppm aldehydes produced high mortality rates
28 (mostly above 50%) and severe lung damage. Exhaust containing 1,070 mg/m3 paniculate
29 matter, 1,700 ppm CO, 12 ppm NO2, and 154 ppm aldehydes resulted in 100% mortality in
30 all three species. High CO levels, which resulted in a carboxyhemoglobin value of 60% in
31 mice and 50% in rabbits and guinea pigs, were considered to be the main cause of death in
December 1994 5_n DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 5-1. HUMAN STUDIES OF DIESEL EXHAUST EXPOSURE
Study
Description
Findings
Kahn et al. (1988)
El Batawi and
Noweir (1966)
Battigelli (1965)
Katz et al. (1960)
Hare and Springer
(1971)
Hare et al. (1974)
Linnell and Scott
(1962)
Battigelli (1965)
Reger et al. (1978)
Ames et al. (1982)
Jorgensen and
Svensson (1970)
13 Cases of acute exposure,
Utah and Colorado coal
miners.
161 Workers, two diesel bus
garages.
Six subjects, eye exposure
chamber, three dilutions.
14 Persons monitoring diesel
exhaust in a train tunnel.
Volunteer panelists who
evaluated general public's
response to odor of diesel
exhaust.
Odor panel under highly
controlled conditions
determined odor threshold for
diesel exhaust.
13 Volunteers exposed to three
dilutions of diesel exhaust for
15 min to 1 h.
Five or more VC maneuvers
by each of 60 coal miners
exposed to diesel exhaust at
the beginning and end of a
work shift.
Pulmonary function of
60 diesel-exposed compared
with 90 non-diesel-exposed
coal miners over work shift.
240 Iron ore miners matched
for diesel exposure,
smoking and age were given
bronchitis questionnaires
and spirometry pre- and
postwork shift.
Acute reversible sensory irritation, headache;
nervous system effects, broncho-constriction were
reported at unknown exposures.
Eye irritation (42%), headache (37%), dizziness
(30%), throat irritation (19%), and cough and
phlegm (11%) were reported in this order of
incidence by workers exposed in the service and
repair of diesel powered buses.
Time to onset was inversely related and severity
of eye irritation was associated with the level of
exposure to diesel exhaust.
Three occasions of minor eye and throat
irritation; no correlation established with
concentrations of diesel exhaust components.
Slight odor intensity, 90% perceived, 60%
objected; slight to moderate odor intensity, 95%
perceived, 75% objected; almost 75% objected;
almost 95% objected.
In six panelists, the volume of air required to
dilute raw diesel exhaust to an odor threshold
ranged from a factor of 140 to 475.
No significant effects on pulmonary resistance
were observed as measured by plethysmography.
FEVj, FVC, and PEFR were similar between
diesel and non-diesel-exposed miners. Smokers
had an increased increased number of
decrements over shift than nonsmokers.
Significant work shift decrements occurred in
miners in both groups who smoked; no significant
differences in ventilatory function changes
between miners exposed to diesel exhaust and
those not exposed.
Among underground (surrogate for diesel
exposure) miners, smokers and older age groups,
frequently of bronchitis was higher. Pulmonary
function was similar between groups and
subgroups except for differences accountable to
age.
December 1994
5-12 DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 5-1 (cont'd). HUMAN STUDIES OF DIESEL EXHAUST EXPOSURE
Study
Description
Findings
Gamble et al. 200 Salt miners performed
(1978) before and after workshift
spirometry. Personal
environmental NO2 and
inhalable particle samples were
collected.
Gamble et al. 232 Workers in four diesel bus
(1987a) garages administered acute
respiratory question naire and
before and after workshift
spirometry. Compared to
lead, acid battery workers
previously found to be
unaffected by their exposures.
Ulfvarson et al. Workshift changes in
(1987) pulmonary function were
evaluated in crews of roll-on/
roll-off ships and car ferries
and bus garage staff.
Pulmonary function was
evaluated in six volunteers
exposed to diluted diesel
exhaust, 2.1 ppm NO2, and
0.6 mg/m3 paniculate matter.
Battigelli et al. 210 Locomotive repairmen
(1964) exposed to diesel exhaust for
an average of 9.6 years in
railroad engine houses were
compared with 154 railroad
yard workers of comparable
job status but no exposure to
diesel exhaust.
Gamble et al. 283 Male diesel bus garage
(1987b) workers from four garages in
two cities were examined for
impaired pulmonary function
(FVC, FEV,, and flow rates).
Study population with a mean
tenure of 9 ± 10 years S.D.
was compared to a nonexposed
"blue collar" population.
Smokers had greater but not significant
reductions in spirometry than ex- or nonsmokers.
NO2, but not paniculate, levels
significantly decreased FEV1, FEF25,
FEF50, and FEF75 over the workshift.
Prevalence of burning eyes, headache,
difficult or labored breathing, nausea,
and wheeze were higher in diesel bus workers
than in comparison population.
Pulmonary function was affected
during a workshift exposure to
diesel exhaust, but it normalized after a
few days widi no exposure. Decrements
were greater with increasing intervals
between exposures. No effect on pulmonary
function was observed in the experimental
exposure study.
No significant differences in VC, FEV],
peak flow, nitrogen washout, or diffusion
capacity nor in the prevalence of dyspnea,
cough, or sputum were found between the
diesel exhaust-exposed and nonexposed
groups.
Analyses within the study populations population
showed no association of respiratory symptoms
with tenure. Reduced FEVj and FEF50 (but not
FEF75) were associated with increasing tenure.
The study population had a higher incidence of
cough, phlegm, and wheezing unrelated to tenure.
Pulmonary function was not affected in the total
cohort of diesel-exposed of diesel-exposed but
was reduced with 10 or more years of tenure.
December 1994
5-13
DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 5-1 (cont'd). HUMAN STUDIES OF DIESEL EXHAUST EXPOSURE
Study
Description
Findings
Purdham et al.
(1987)
Reger et al. (1982)
Ames et al. (1984)
Attfield (1978)
Attfield et al.
(1982)
Respiratory symptoms and
pulmonary function were
evaluated in 17 stevedores
exposed to both diesel and
gasoline exhausts in car ferry
operations; control group was
11 on-site office workers.
Differences in respiratory
symptoms and pulmonary
function were assessed in
823 coal miners from six
diesel equipped mines
compared to 823 matched coal
miners not exposed to diesel
exhaust.
Changes in respiratory
symptoms and function were
measured during a 5-year
period in 280 diesel-exposed
and 838 nonexposed U.S.
underground coal miners.
Respiratory symptoms and
function were assessed in
2,659 miners from
21 underground metal mines
(1,709 miners) and nonmetal
mines (950 miners). Years of
diesel usage in the mines were
surrogate for exposure to
diesel exhaust.
Respiratory symptoms and
function were assessed in
630 potash miners from six
potash mines using a
questionnaire, chest
radiographs and spirometry.
A thorough assessment of the
environment of each mine was
made concurrently.
No differences between the two groups for
respiratory symptoms. Stevedores had lower
baseline lung function consistent with an
obstructive ventilatory defect compared with
controls and those of Sydney, Nova Scotia,
residents. Caution in interpretation is warranted
due to small sample size. No significant size.
No significant changes in lung function over
workshift nor difference between two groups.
Underground miners in diesel-use mines reported
more symptoms of cough and phlegm and had
lower pulmonary function. Similar trends were
noted for surface workers at diesel-use mines.
Pattern was consistent with small airway disease
but factors other than exposure to diesel exhaust
thought to be responsible.
No decrements in pulmonary function or
increased prevalence of respiratory symptoms
were found attributable to diesel exhaust. In fact,
5-year incidences of cough, phlegm, and dyspnea
were greater in miners without exposure to diesel
exhaust than in miners exposed to diesel diesel
exhaust.
Questionnaire found an association between an
increased prevalence of cough and aldehyde
exposure; this finding was not substantiated by
spirometry data. No adverse symptoms or
pulmonary function decrements were related to
exposure to NO2, CO, CO2, dust, or quartz.
No obvious association indicative of diesel
exposure was found between health indices, dust
exposure, and pollutants. A higher prevalence of
cough and phlegm, but no differences in FVC
and FEVj, were found in these diesel-exposed
potash workers when compared to predicted
values from a logistic model based on blue- collar
staff working in nondusty jobs.
December 1994
5-14
DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 5-1 (cont'd). HUMAN STUDIES OF DIESEL EXHAUST EXPOSURE
Study
Description
Findings
Gamble et al.
(1983)
Gamble and Jones
(1983)
Edling and Axelson
(1984)
Edling et al. (1987)
Respiratory morbidity was
assessed in 259 miners in
5 salt mines by respiratory
symptoms, radiographic
findings and spirometry. Two
mines used diesels extensively,
2 had limited use, one used no
diesels in 1956, 1957, 1963,
or 1963 through 1967.
Several working populations
were compared to the salt
mine cohort.
Same as above. Salt miners
were grouped into low,
intermediate and high exposure
categories based on tenure in
jobs with diesel exposure.
Pilot study of 129 bus
company employees classified
into three dieselexhaust
exposure categories clerks (0),
bus drivers (1), and bus
garage workers.
Cohort of 694 male bus
garage employees followed
from 1951 through 1983
were evaluated for mortality
from cardiovascular disease.
Subcohorts categorized by
levels of exposure were clerks
(0), bus drivers (1), and bus
garage employees (2).
After adjustment for age and smoking, salt
miners showed no symptoms, increased
prevalence of cough, phlegm, dyspnea or air
obstruction (FEVj/FVC) compared to
aboveground coal miners, potash workers or blue
collar workers. FEVj, FVC, FEF50, and FEF75
were uniformly lower for salt miners in
comparison to all the comparison populations.
No changes in pulmonary function were
associated with years of exposure or cumulative
exposure to inhalable particles or NO2.
A statistically significant dose-related association
of phlegm and diesel exposure was noted.
Changes in pulmonary function showed no
association with diesel tenure. Age- and
smoking-adjusted rates of cough, phlegm, and
dyspnea were 145, 169, and 93% of an external
comparison population. Predicted pulmonary
function indices showed small but significant
reductions; there was no dose-response
relationship.
The most heavily exposed group (bus garage
workers) had a fourfold increase in risk of dying
from cardiovascular disease, even after correction
for smoking and allowing for 10 years of
exposure and 15 years or more of
inductionlatency time.
No increased mortality from cardiovascular
disease was found among the members of these
five bus companies when compared with the
general population or grouped as sub-cohorts with
different levels of exposure.
December 1994
5-15
DRAFT-DO NOT QUOTE OR CITE
-------
1 the latter case. High NO2 levels were considered to be the main cause of lung damage and
2 mortality seen in the other three tests. Aldehydes and NO2 were considered to be the main
3 irritants in the light load test.
4
5 5.1.2.2 Short-Term Exposures
6 A number of short-term inhalation studies have employed a regimen of 20 h/day,
7 7 days/week for varying exposure periods to differing concentrations of airborne paniculate
8 matter, vapor and gas concentrations of diluted diesel exhaust. Exposure regimens and
9 characterization of gas-phase components for these studies are summarized in Table 5-2.
10 Pepelko et al. (1980a) evaluated the pulmonary function of cats exposed under these
11 conditions for 28 days to 6.4 mg/m3 particulate matter. The only significant functional
12 change observed was a decrease in maximum expiratory flow rate at 10% vital capacity.
13 The excised lungs of the exposed cats appeared charcoal gray, with focal black spots visible
14 on the pleural surface. Pathologic changes included a predominantly peribronchial
15 localization of black-pigmented macrophages within the alveoli producing a focal pneumonitis
16 or alveolitis.
17 The effects of a short-term diesel exhaust exposure on arterial blood gases, pH, blood
18 buffering, body weight changes, lung volumes, and deflation pressure-volume (PV) curves of
19 young adult rats were evaluated by Pepelko (1982a). Exposures were 20 h/day, 7 days/week
20 for 28 days to a concentration of 6.4 mg/m3 particulate matter in the nonirradiated exhaust
21 (RE) and 6.75 mg/m3 in the irradiated exhaust (IE). In spite of the irradiation, levels of
22 gaseous compounds were not substantially different between the two groups (Table 5-2).
23 Body weight gains were significantly reduced in the reexposed rats and to an even greater
24 degree in rats exposed to IE. Arterial blood gases and standard bicarbonate were unaffected,
25 but arterial blood pH was significantly reduced in rats exposed to IE. Residual volume and
26 wet lung weight were not affected by either exposure, but vital capacity and total lung
27 capacity were increased significantly following exposure to RE. The shape of the deflation
28 PV curves were nearly identical for the control, RE, and IE groups.
29 In related studies, Wiester et al. (1980) evaluated pulmonary function in 4-day old
30 guinea pigs exposed for 20 h/day, 7 days/week for 28 days to IE having a concentration of
31 6.3 mg/m3 particulate matter. When housed in the exposure chamber, pulmonary flow
December 1994 5-16 DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 5-2. SHORT-TERM EFFECTS OF DIESEL EXHAUST ON
LABORATORY ANIMALS
1— »
1
1— >
u
£
H
6
o
1
0
o
H
M
§
n
3
Species/Sex
Rat, F-344, M;
Mouse, A/J;
Hamster, Syrian
Rat, F-344, M,
F; Mouse,
CD-l.M, F
Cat, Inbred, M
Rat, Sprague-
Dawley, M
Guinea Pig,
Hartley, M, F
Rat, F-344,
M
Guinea Pig,
Hartley
M, F
Exposure
Period
20 h/day
7 days/week
10-13 weeks
7 h/day
5 days/week
19 weeks
20 h/day
7 days/week
4 weeks
20 h/day
7 days/week
4 weeks
20 h/day
7 days/week
4 weeks
20 h/day
5.5
days/week
4 weeks
20 h/day
7 days/week
8 weeks
Particles
(mg/m3)
1.5
0.19 /tin, MMD
0.21
1.0
4.4
6.4
6.4
6.8a
6.8"
6.0
6.8 urn, MMD
6.3
C x T CO NO2 SO2
(mg-h/m3) (ppm) (ppm) (ppm) Effects
2, 100 to 2,730 6.9 0.49 — Increase in lung wt; increase in thickness
of alveolar walls; no species difference
140 — — — No effects on lung function; increase in
665 — — — PMNs and proteases and AM aggregation
2,926 — — — in both species
3,584 14.6 2.1 2.1 Few effects on lung function; focal
pneumonitis or alveolitis
3,584 16.9 2.49 2.10 Decreased body wt; arterial blood pH
3,808 16.1" 2.76" 1.86" reduced; vital total lung capacities
( < 0.0 1 ppm O3)" increased
3,808 16.7 2.9 1.9 Exposure started when animals were 4
days old; increase in pulmonary flow;
«0.01Ppm03)> bradycardia
2,640 — — — Macrophage aggregation; increase in
PMNs; Type II cell proliferation;
thickened alveolar walls
7,056 17.4 2.3 2.1 Increase in relative lung wt; AM
aggregation; hypertrophy of goblet cells;
(<0.01 ppm O3)" focal hyperplasia of alveolar epithelium
References
Kaplan et al.
(1982)
Mauderly et al.
(1981)
Pepelko et al.
(1980a)
Pepelko (1982a)
Wiester et al.
(1980)
White and Garg
(1981)
Weister et al.
(1980)
'Irradiated exhaust.
PMN = Poly morphonuclear leukocyte.
AM = Alveolar macrophage.
-------
1 resistance increased 35 %, and a small but significant sinus bradycardia occurred as compared
2 with controls housed and measured in control air chambers (p < 0.002). Respiratory rate,
3 tidal volume, minute volume, and dynamic compliance were unaffected, as were lead-1
4 electrocardiograms.
5 A separate group of adult guinea pigs was necropsied after 56 days of exposure to IE,
6 to diluted RE, or to clean air (Wiester et al., 1980). Exposure resulted in a significant
7 increase in the ratio of lung weight to body weight (0.68% for controls, 0.78% for IE, and
8 0.82% for RE). Heart/body weight ratios were not affected by exposure. Microscopically,
9 there was a marked accumulation of black pigment-laden alveolar macrophages (AM)
10 throughout the lung with a slight to moderate accumulation in bronchial and carinal lymph
11 nodes. Hypertrophy of goblet cells in the tracheobronchial tree was frequently observed, and
12 focal hyperplasia of alveolar lining cells was occasionally observed. No evidence of
13 squamous metaplasia of the tracheobronchial tree, emphysema, peribronchitis, or
14 peribronchiolitis was noted.
15 White and Garg (1981) studied pathologic alterations in the lungs of rats (16 exposed
16 and 8 controls) after exposure to diesel exhaust containing 6 mg/m3 paniculate matter. Two
17 rats from the exposed group and 1 rat from the control group (filtered room air) were
18 sacrificed after each exposure interval of 6 h, and 1, 3, 7, 14, 28, 42, and 63 days; daily
19 exposures were for 20 h and were 5.5 days/week. Evidence of AM recruitment and
20 phagocytosis of diesel particles was found at the 6-h sacrifice; after 24 h of exposure there
21 was a focal, scattered increase in the number of Type II cells. After 4 weeks of exposure,
22 there were morphologic changes in size, content, and shape of AM, septal thickening
23 adjacent to clusters of AMs, and an appearance of inflammatory cells, primarily within the
24 septa. At 9 weeks of exposure, focal aggregations of particle-laden macrophages developed
25 near the terminal bronchi, along with an influx of polymorphonuclear leukocytes, Type II
26 cell proliferation, and thickening of the alveolar walls. The affected alveoli occurred in
27 clusters that, for the most part, were located near the terminal bronchioles, but occasionally
28 were focally located in the lung parenchyma.
29 Mauderly et al. (1981) exposed rats and mice by inhalation to diluted diesel exhaust
30 for 545 h over a 19-week period on a regimen of 7 h/day, 5 days/week at concentrations of
31 0, 0.21, 1.02, or 4.38 mg/m3 paniculate matter. Indices of health effects were minimal
December 1994 5-18 DRAFT-DO NOT QUOTE OR CITE
-------
1 following 19 weeks of exposure. There were no significant exposure-related differences in
2 mortality or body weights of the rats or mice. There were also no significant differences in
3 respiratory function (breathing patterns, dynamic lung mechanics, lung volumes, quasistatic
4 PV relationships, forced expirograms, and CO-diffusing capacity), or tracheal mucociliary or
5 deep lung clearances among the experimental groups. Rats, but not mice, had elevated
6 immune responses in lung-associated lymph nodes at the two higher exposure levels.
7 Inflammation in the lungs of rats exposed to 4.38 mg/m3 paniculate matter was indicated by
8 increases in polymorphonuclear leukocytes and lung tissue proteases. Histopathologic
9 findings included AMs that contained diesel particles, an increase in Type II cells, and the
10 presence of particles in the interstitium and tracheobronchial lymph nodes.
11 Kaplan et al. (1982) evaluated the effects of subchronic exposure to diesel exhaust on
12 rats, hamsters, and mice. The exhaust was diluted to a concentration of 1.5 mg/m3
13 particulate matter; exposures were 20 h/day, 7 days/week. Hamsters were exposed for
14 86 days; rats and mice for 90 days. There were no significant differences in mortality or
15 growth rates between exposed and control animals. Lung weight relative to body weight of
16 15 rats exposed for 90 days was significantly higher than the mean for a group of control
17 group (15 rats). Histological examination of tissues of all three species indicated particle
18 accumulation in the lungs and mediastinal lymph nodes. Associated with the larger
19 accumulations, there was a minimal increase in the thickness of the alveolar walls, but the
20 vast majority of the particles elicited no response. After 6 mo of recovery, considerable
21 clearance of the diesel particles from the lungs occurred in all three species, as evaluated by
22 gross pathology and histopathology. However, no quantitative estimate of clearance was
23 provided.
24 Toxic effects in animals from short-term exposure to diesel exhaust appear to be
25 primarily attributable to the gaseous components (i.e., mortality from CO intoxication and
26 lung injury caused by cellular damage resulting from NO2 exposure). The results from
27 short-term exposures indicate that rats experience no to minimal lung function impairment
28 even at diesel exhaust levels sufficiently high to cause histological and cytological changes in
29 the lung. In subchronic studies, frank adverse health effects are not readily apparent and
30 when found are mild and result from exposure to concentrations of about 6 mg/m3 particulate
31 matter and durations of exposures of 20 h/day. There is ample evidence that subchronic
December 1994 5-19 DRAFT-DO NOT QUOTE OR CITE
-------
1 exposure to lower levels of diesel exhaust affects the lung, as indicated by accumulation of
2 particles, evidence of inflammatory response, AM aggregation and accumulation near the
3 terminal bronchioles, Type II cell proliferation, and thickening of alveolar walls adjacent to
4 AM aggregates. Little evidence exists, however, that subchronic exposure to diesel exhaust
5 impairs lung function.
6
7 5.1.2.3 Chronic Exposures
8 Effects on Growth and Longevity
9 Changes in growth, body weight, absolute or relative organ weights, and longevity can
10 be measurable indicators of chronic toxic effects. Such effects have been observed in some,
11 but not all, of the long-term studies conducted on laboratory animals exposed to diesel
12 exhaust. There was very limited conclusive evidence for an effect on survival in the
13 published chronic animal studies; deaths occurred intermittently early in one study in female
14 rats exposed to 3.7 mg/m3 particulate matter; however, the death rate began to decrease after
15 15 mo, and the survival rate after 30 mo was slightly higher than that of the control group
16 (Research Committee for HERP Studies, 1988). Studies on the effects of chronic exposure
17 to diesel exhaust on survival and body weight or growth are detailed in Table 5-3.
18 Increased lung weights and lung-to-body weight ratios have been reported in rats,
19 mice, and hamsters. These data are summarized in Table 5-4. In rats exposed for up to
20 36 weeks to 0.25 or 1.5 mg/m3 particulate matter, lung wet weights (normalized to body
21 weight) were significantly higher in the 1.5 mg/m3 exposure group than control values after
22 12 weeks of exposure (Misiorowski et al., 1980). Rats and Syrian hamsters were exposed
23 for 2 years (five 16-h periods per week) to diesel exhaust diluted to achieve concentrations of
24 0.7, 2.2, and 6.6 mg/m3 particulate matter (Brightwell et al., 1986). At necropsy, a
25 significant increase in lung weight was seen in both rats and hamsters exposed to diesel
26 exhaust compared with controls. This finding was more pronounced in the rats in which the
27 increase was progressive with both duration of exposure and particulate matter level. The
28 increase was greatest at 30 mo (after the end of a 6-mo observation period in the
29 high-concentration male group where the lung weight was 2.7 times the control and at 24 mo
30 in the high-concentration female group [3.9 times control]). Heinrich et al. (1986a,b; see
31 also Stober, 1986) found a significant increase in wet and dry weights of the lungs of rats
December 1994 5-20 DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 5-3. EFFECTS OF CHRONIC EXPOSURES TO DIESEL EXHAUST
ON SURVIVAL AND GROWTH OF LABORATORY ANIMALS
1
0>
1— >
s
1
to
t— '
o
£>
H
o
o
H
O
O
H
W
O
n
Species/Sex
Rat, F-344, M, F;
Monkey,
cynomolgus, M
Rat, F344, M;
Guinea Pig,
Hartley, M
Hamster, Chinese,
M
Rat, Wistar, M
Rat, F-344, M, F;
Mouse CD-I
Rat, Wistar, F;
Mouse, MMRI, F
Rat, F-344
M, F
Rat0
F-344/Jcl.
Exposure
Period
7h/day
5 days/week
104 weeks
20h/day
5 days/week
106 weeks
8h/day
7 days/week
26 weeks
6h/day
5 days/week
87 weeks
7h/day
5 days/week
130 weeks
19h/day
5 days/week
104 weeks
16h/day
5 days/week
104 weeks
16 h/day
6 days/week
130 weeks
Particles
(mg/m3)
2.0
0.23-0.36 pm, MMD
0.25
0.75
1.5
0.19pm, MMD
6.0
12.0
8.3
0.71 pm, MMD
0.35
3.5
7.0
0.25 /.m, MMD
4.24
0.35 urn, MMD
0.7
2.2
6.6
O.lld
0.41d
1.08d
2.31d
3. 72=
0.2-0.3 Mm, MMD
C x T
(mg-h/m3)
7,280
2,650
7,950
15,900
8,736
17,472
21,663
1,592
15,925
31,850
41,891
5,824
18,304
54,912
1,373
5,117
13,478
28,829
46,426
CO
(ppm)
11.5
2.7a
4.4"
7.1a
—
—
50.0
2.9
16.5
29.7
12.5
—
—
32.0
1.23
2.12
3.96
7.10
12.9
NO2
(ppm)
1.5
O.lb
0.27b
0.5"
—
—
4.0-6.0
0.05
0.34
0.68
1.5
—
—
—
0.08
0.26
0.70
1.41
3.00
SOz
(ppm) Effects
0.8 No effects on growth or survival
— Reduced body weight in rats at 1 .5 mg/m3
—
—
— No effect on growth
—
— No effect on growth or mortality rates
— No effect on growth or mortality rates
—
—
1 . 1 Reduced body wts; increased mortality in mice
— Growth reduced at 2.2 and 6.6 mg/m3
—
—
0.38 Concentration-dependent decrease in body
1 .06 weight; earlier deaths in females exposed to
2.42 3.72 mg/m3, stabilized by 15 mo
4.70
4.57
References
Lewis et al. (1989)
Schreck et al. (1981)
Vinegar et al.
(1981a,b)
Karagianes
et al. (1981)
Maude riy et al.
(1984, 1987b)
Heinrich et al.
(1986a)
Brightwell et al.
(1986)
Research Committee
for HERP Studies
(1988)
'Estimated from graphically depicted mass concentration data.
bEstimated from graphically presented
cData for tests with
dLight-duty engine.
'Heavy-duty engine
light-duty engine;
mass concentration data for NO2 (assuming 90%
NO and
10% NO2).
similar results with heavy-duty engine.
-------
O
n
o
8*
"
§
.£>.
1
N)
O
*>
3
T3
D
o
^^
o
H
O
G
H
W
O
TABLE 5-4. EFFECTS OF CHRONIC EXPOSURES TO DIESEL EXHAUST
ON ORGAN WEIGHTS AND ORGAN-TO-BODY- WEIGHT RATIOS
Species/Sex
Rat, F-344, M;
Mouse, A/J, M;
Hamster, Syrian,
M
Rat, F-344, M, F
Rat. F-344, M
Rat, F-344, F
Rat, F-344;
Guinea Pig,
Hartley
Hamster,
Chinese, M
Rat. Wistar, F;
Hamster, Syrian,
M, F;
Mouse NMRI, F
Rat, F-344;
Hamster, Syrian
Cat Inbred, M
Exposure
Period
20 h/day
7 days/week
12-13 weeks
7 h/day
5 days/week
52 weeks
20 h/day
5.5 days/week
36 weeks
7 h/day
5 days/week
104 weeks
20 h/day
5.5 days/week
78 weeks
8 h/day
7 days/week
26 weeks
19 h/day
5 days/week
120-140
weeks
16 h/day
5 days/week
104 weeks
8 h/day
7 days/week
124 weeks
Particles
(mg/m3)
1.5
0.19fim, MMD
2.0
0.23-0.36 itm, MMD
0.25
1.5
0.19 urn, MMD
2.0
0.23-0.36 itm, MMD
0.25
0.75
1.5
0.19pm, MMD
6.0
12.0
4.24
0.35 /am, MMD
0.7"
2.2b
6.6
6.0"
12.0b
C x T CO
(mg-h/m3) (ppm)
2,520-2,730 -
3,640 12.7
990 —
5,940 -
7,280 11.5
2,145 -
6,435 -
12,870 —
8,736 -
17,472 -
48,336-56,392 12.5
5,824 -
18,304 -
54,912 32.0
41,664 20.2
83,328 33.2
NO2 SO2
(ppm) (ppm) Effects
— — No effect on liver, kidney, spleen, or heart
weights
1 .6 0.83 No effects on weights of lungs, liver, hean,
spleen, kidneys, and testes
— — Increase in relative lung weight at 1.5 mg/m3
— — only initially seen at 12 weeks
1.5 0.81 No effects on heart weights
— — No effects on heart mass
— —
— _
— — Increase in lung weight and lung/body weight
— — ratio
1.5 1.1 Increase in rat, mouse, and hamster lung
weight and dry weights
— — Increase in lung weight concentration related
— — in rats; heart weight/body weight ratio
— — greater at 6.6 mg/m3
2.7 2.7 Decrease in lung and kidney weights
4.4 5.0
References
Kaplan et al. (1982)
Green et al. (1983)
Misiorowski
et al. (1980)
Vallyathan
et al. (1986)
Penney et al. (1981)
Vinegar et al.
(1981a,b)
Heinrich et al.
(1986a,b)
Stober (1986)
Brightwell et al.
(1986)
Pepelko et al.
(1980b, 1981)
Moorman et al.
(1985)
"1 to 61 weeks of exposure.
"A"? tf\ 1 *y A. u/*»ptc r»f Avrmcura
n
-------
1 and mice exposed at 4.24 mg/m3 paniculate matter for 1 year in comparison to controls.
2 After 2 years, the difference was a factor of 2 (mice) or 3 (rats). After the same exposure
3 periods, the hamsters showed increases of 50 to 75%, respectively. Exposure to equivalent
4 filtered diesel exhaust caused no significant effects in any of the species. Vinegar et al.
5 (1980, 1981a,b) exposed hamsters to two levels of diesel exhaust with resultant
6 concentrations of about 6 and 12 mg/m3 particulate matter for 8 h/day, 7 days/week for
7 6 mo. Both exposures significantly increased lung weight and lung weight to body weight
8 ratios. The difference between lung weights of exposed and control hamsters exposed to
9 12 mg/m3 particulate matter was approximately twice that of those exposed to 6 mg/m3.
10 No evidence was found in the published literature that chronic exposure to diesel
11 exhaust affected the weight of body organs other than the lung and heart (e.g., liver, kidney,
12 spleen, or testes). These data are contained in Table 5-4. Morphometric analysis of hearts
13 from rats and guinea pigs exposed to 0.25, 0.75, or 1.5 mg/m3 particulate matter 20 h/day,
14 5.5 days/week for 78 weeks revealed no significant alteration in mass at any exposure level
15 or duration of exposure (Penney et al., 1981). The analysis included relative wet weights of
16 the right ventricle, left ventricle, combined atria, and ratio of right to left ventricle.
17 Vallyathan et al. (1986) found no significant differences in heart weights and the ratio of
18 heart weight to body weight between rats exposed to 2 mg/m3 particulate matter for 7 h/day,
19 5 days/week for 24 mo and their respective clean air chamber controls. No significant
20 differences in the lungs, heart, liver, spleen, kidney, and testes of rats exposed for 52 weeks,
21 7 h/day, 5 days/week to diluted diesel exhaust containing 2 mg/m3 particulate matter
22 compared with their respective controls (Green et al., 1983).
23
24 Effects on Pulmonary Function
25 The effect of long-term exposure to diesel exhaust on pulmonary function has been
26 evaluated in laboratory studies of rats, hamsters, cats, and monkeys. These studies are
27 summarized in Table 5-5, along with more details on the exposure characteristics, in general
28 order of increasing dose (C x T) of the exhaust diesel particulate matter. The text will be
29 presented using the same approach.
30 Lewis et al. (1989) evaluated 10 control and 10 diesel-exposed rats (2 mg/m3
31 particulate matter, 7 h/day, 5 days/week for 52 or 104 weeks) for responses in functional
December 1994 5-23 DRAFT-DO NOT QUOTE OR CITE
-------
December
i— >
vO
•**•
i
Q
w
,>
H
"
D
o
^^
o
H
O
d
0
TABLE 5-5. EFFECTS OF DIESEL EXHAUST ON
PULMONARY FUNCTION OF LABORATORY ANIMALS
Species/ Sex
Rat, F-344
M, F
Monkey, M
Cynomolgus
Rat, F-344, M
Rat, Wistar, F
Hamster, Chinese,
M
Rat, F-344,
M, F
Hamster, Syrian
M, F
Rat, F-344;
Hamster Syrian
Rat, Wistar, F
Cat, inbred, M
Exposure
Period
7 h/day
5 days/week
104 weeks
7 h/day
5 days/week
104 weeks
20 h/day
5.5 days/week
87 weeks
7-8 h/day
5 days/week
104 weeks
8 h/day
7 days/week
26 weeks
7 h/day
5 days/week
130 weeks
19 h/day
5 days/week
120 weeks
16 h/day
5 days/week
104 weeks
19 h/day
5 days/week
140 weeks
8 h/day
7 days/week
124 weeks
Particles
(mg/m3)
2.0
0.23-0.36 Mm MMD
2.0
0.23-0.36 Mm, MMD
1.5
0.19 Mm, MMD
3.9
0. 1 MHI, MMD
6.0
12.0
0.35
3.5
7.0
0.23-0.26 urn, MMD
4.24
0.35 Mm, MMD
0.7
2.2
6.6
4.24
0.35 MHI, MMD
6.0"
12.0b
C x T CO
(mg-h/m3) (ppm)
7,280 11.5
7,280 11.5
14,355 7.0
14,196-16,224 18.5
8,736 -
17,472 -
1,593 2.9
15,925 16.5
31,850 29.7
48,336 12.5
5,824 -
18,304 -
54,912 -
56,392 12.5
41,664 20.2
83,328 33.3
NO2 SO2
(ppm) (ppm) Effects
1.5 0.8 No effect on pulmonary function
1 .5 0.8 Decreased expiratory flow; no effect on vital
or diffusing capabilities
0.5 — Increased functional residual capacity,
expiratory volume and flow
1.2 3.1 No effect on minute volume, compliance or
resistance
— — Decrease in vital capacity, residual volume,
— — and diffusing capacity; increase in static
deflation lung volume
0.05 — Diffusing capacity, lung compliance reduced
0.34 — at 3.5 and 7 mg/m3
0.68 -
1.5 1.1 Significant increase in airway resistance
— — Large number of pulmonary function changes
— — consistent with obstructive and restrictive
— — airway diseases at 6.6 mg/m3 (no specific data
provided)
1.5 1.1 Decrease in dynamic lung compliance;
increase in airway resistance
2.7 2.1 Decrease in vital capacity, total lung capacity,
4.4 5.0 and diffusing capacity after 2 years; no effect
on expiratory flow
References
Lewis et al. (1989)
Lewis et al. (1989)
Gross (1981b)
Heinrich et al. (1982)
Vinegar et al. (1980,
1981a,b)
Mauderly et al.
(1988)
McClellan et al.
(1986)
Heinrich et al.
(1986a)
Brightwell et al.
(1986)
Heinrich et al.
(1986a)
Pepelko et al.
(1980b, 1981)
Moorman et al.
(1985)
al to 61 weeks exposure.
b62 to 124 weeks of exposure.
O
3
-------
1 residual capacity and airway resistance and conductance. At the 104-week evaluation, the
2 rats were also examined for maximum flow volume impairments. No evidence of an
3 impairment of pulmonary function as a result of the exposure to diesel exhaust was found in
4 rats. Diesel exhaust-exposed monkeys were evaluated prior to exposure and at 6-mo
5 intervals up to 24 mo for pulmonary compliance and resistance, all static and dynamic lung
6 volumes, diffusing capacity, distribution of ventilation, and maximal ventilatory performance
7 (flow and volume). The monkeys exposed to diesel exhaust demonstrated small airway
8 obstructive disease. The obstructive impairment was most detectable using the forced
9 expiratory flow at 40% of the total lung capacity instead of the forced expiratory flow as a
10 percentage of the vital capacity. This significant finding is indicative of a shift in the flow
11 volume curve as a result of mild hyperinflation.
12 Gross (1981b) exposed rats for 20 h/day, 5.5 days/week for 87 weeks to diesel
13 exhaust containing 1.5 mg/m3 paniculate matter. When the data were normalized (e.g.,
14 indices expressed in units of air flow or volume for each animal by its own forced expiratory
15 volume), there were no apparent functionally significant changes occurring in the lungs at
16 38 weeks of exposure that might be attributable to the inhalation of diesel exhaust. After
17 87 weeks of exposure, functional residual capacity (FRC) and its component volumes
18 (expiratory reserve [ER] and residual volume [RV]), maximum expiratory flow (MEF) at
19 40% FVC, MEF at 20% FVC, and FEV0 { were significantly greater in the diesel-exposed
20 rats. An observed increase in airflow at the end of the forced expiratory maneuver when a
21 decreased airflow would be expected from the increased FRC, ER, and RV data (the typical
22 scenario of human pulmonary disease) showed these data to be inconsistent with known
23 clinically significant health effects . Furthermore, although the lung volume changes in the
24 diesel-exposed rats could have been indicative of emphysema or chronic obstructive lung
25 disease, this interpretation was contradicted by the air flow data, which suggest simultaneous
26 lowering of the resistance of the distal airways.
27 Heinrich et al. (1982) evaluated the pulmonary function of rats exposed to a
28 concentration of 3.9 mg/m3 paniculate matter for 7 to 8 h/day, 5 day/week for 2 years.
29 When compared with a control group, no significant changes in respiratory rate, minute
30 volume, compliance, or resistance occurred in the exposed group (number of rats per group
31 was not stated).
December 1994 5_25 DRAFT-DO NOT QUOTE OR CITE
-------
1 Hamsters (eight or nine per group) were exposed 8 h/day, 7 days/week, for 6 mo to
2 concentrations of either about 6 mg/m3 or about 12 mg/m3 paniculate matter (Vinegar et al.,
3 1980, 1981a,b). Vital capacity, vital capacity/lung weight ratio, residual lung volume by
4 water displacement, and CO2 diffusing capacity decreased significantly in hamsters exposed
5 to 6 mg/m3 particulate matter. Static deflation volume-pressure curves showed depressed
6 deflation volumes for diesel-exposed hamsters when volumes were corrected for body weight
7 and even greater depressed volumes when volumes were corrected for lung weight.
8 However, when volumes were expressed as percentage of vital capacity, the diesel-exposed
9 hamsters had higher lung volumes at 0 and 5 cm H2O. In the absence of confirmatory
10 histopathology, the authors tentatively concluded that these elevated lung volumes and the
11 significantly reduced diffusing capacity in the same hamsters were indicative of possible
12 emphysematous changes in the lung. Similar lung function changes were reported in
13 hamsters exposed at 12 mg/m3 particulate matter, but detailed information was not reported.
14 It was stated, however, that the decrease in vital capacity was 176% greater in the second
15 experiment than in the first.
16 Mauderly et al. (1988; see also McClellan, et al. 1986) examined the impairment of
17 respiratory function in rats exposed for 7 h/day, 5 days/week, for 24 mo to diluted diesel
18 exhaust with 0.35, 3.5, or 7.0 mg/m3 particulate matter. After 12 mo of exposure to the
19 highest concentration of diesel exhaust, the exposed rats (n = 22) had lower total lung
20 capacity (TLC), dynamic lung compliance (Cdyn), FVC, and CO diffusing capacity than
21 controls (n = 23). After 24 mo of exposure to 7 mg/m3 particulate matter, mean TLC,
22 Cd n, quasistatic chord compliance, and CO diffusing capacity were significantly lower than
23 control values. Nitrogen washout and percentage of FVC expired in 0.1 s were significantly
24 greater than control values. There was no evidence of airflow obstruction. The functional
25 alterations were attributed to focal fibrotic and emphysematous lesions and thickened alveolar
26 membranes observed by histological examination. Similar functional alterations and
27 histopathologic lesions were observed in the rats exposed to 3.5 mg/m3 particulate matter,
28 but such changes usually occurred later in the exposure period and were generally less
29 pronounced. There were no significant decrements in pulmonary function for the
30 0.35-mg/m3 group at any time during the study nor was there reported histopathologic
31 changes in this group.
December 1994 5-26 DRAFT-DO NOT QUOTE OR CITE
-------
1 Additional studies were conducted by Heinrich et al. (1986a,b; see also Stober, 1986)
2 on the effects of long-term exposure to diesel exhaust on the pulmonary function of hamsters
3 and rats. The exhaust was diluted to achieve a concentration of 4.24 mg/m3 paniculate
4 matter; exposures were for 19 h/day, 5 days/week for a maximum of 120 weeks (hamsters)
5 or 140 weeks (rats). After 1 year of exposure to the diesel exhaust, the hamsters exhibited a
6 significant increase in airway resistance and a nonsignificant reduction in lung compliance.
7 For the same time period, rats showed increased lung weights, a significant decrease in Cdyn,
8 and a significant increase in airway resistance. These indices did not change during the
9 second year of exposure.
10 Syrian hamsters and rats were exposed to 0.7, 2.2, or 6.6 mg/m3 paniculate matter
11 for five 16-h periods per week for 2 years (Brightwell et al., 1986). There were no
12 treatment-related changes in pulmonary function in the hamster. Rats exposed to the highest
13 concentration of diesel exhaust exhibited changes in pulmonary function (data not presented)
14 that were reported to be consistent with a concentration-related obstructive and restrictive
15 disease.
16 Pepelko et al. (1980b; 1981; see also Pepelko, 1982b) and Moorman et al. (1985)
17 measured the lung function of adult cats chronically exposed to diesel exhaust. The cats
18 were exposed for 8 h/day and 7 days/week for 124 weeks. Exposures were at 6 mg/m3 for
19 the first 61 weeks and 12 mg/m3 from weeks 62 to 124. No definitive pattern of pulmonary
20 function changes was observed following 61 weeks of exposure; however, a classic pattern of
21 restrictive lung disease was found at 124 weeks. The significantly reduced lung volumes
22 (TLC, FVC, FRC, and inspiratory capacity [1C]) and the significantly lower single-breath
23 diffusing capacity, coupled with normal values for dynamic ventilatory function (mechanics
24 of breathing), indicate the presence of a lesion that restricts inspiration but does not cause
25 airway obstruction or loss of elasticity. This pulmonary physiological syndrome is consistent
26 with an interstitial fibrotic response that was later verified by histopathology (Plopper et al.,
27 1983).
28 Pulmonary function impairment has been reported in rats, hamsters, cats and monkeys
29 chronically exposed to diesel exhaust. In all species but the monkey, the pulmonary function
30 testing results have been consistent with restrictive lung disease. The monkeys demonstrated
31 evidence of small airway obstructive responses. The disparity between the findings in
December 1994 5-27 DRAFT-DO NOT QUOTE OR CITE
-------
1 monkeys and those in rats, hamsters, and cats could be in part the result of increased particle
2 retention in the smaller species resulting from (1) exposure to diesel exhaust that has higher
3 airborne concentrations of gases, vapors, and particles and/or (2) longer duration of
4 exposure. The nature of the pulmonary impairment is also dependent on the site of
5 deposition and routes of clearance, which are determined by the anatomy and physiology of
6 the test laboratory species and the exposure regimen.
7
8 Lung Morphology, Biochemistry, and Lung Lavage Analysis
9 Several studies have examined the morphological, histological, and histochemical
10 changes occurring in the lungs of laboratory animals chronically exposed to diesel exhaust.
11 The histopathological effects of diesel exposure in the lungs of laboratory animals are
12 summarized in Table 5-6, ranked in order of C x T. The Table 5-6 also contains an
13 expanded description of exposures.
14 Kaplan et al. (1982) performed macroscopic and microscopic examinations of the
15 lungs of rats, mice, and hamsters exposed for 20 h/day, 7 days/week for 3 mo to diesel
16 exhaust containing 1.5 mg/m3 particulate matter. Gross examination revealed diffuse and
17 focal deposition of the diesel particles, that produced a grayish overall appearance of the
18 lungs with scattered, denser black areas. There was clearance of particles via the lymphatics
19 to regional lymph nodes. Microscopic examination revealed no anatomic changes in the
20 upper respiratory tract; the mucociliary border was normal in appearance. Most of the
21 particles were in macrophages, but some were free as small aggregates on alveolar and
22 bronchiolar surfaces. The particle-laden macrophages were often in masses near the
23 entrances of the lymphatic drainage and respiratory ducts. Associated with these masses was
24 a minimal increase in the thickness of the alveolar walls; however, the vast majority of the
25 particles elicited no response. After 6 mo of recovery, the lungs of all three species
26 contained considerably less pigment, as assessed by gross pathological and histopathological
27 examinations.
28 Lewis et al. (1989; see also Green et al., 1983) performed serial histological
29 examinations of rat lung tissue exposed to diesel exhaust containing 2 mg/m3 particulate
30 matter for 7 h/day, 7 days/week for 2 years. Accumulations of black-pigmented AMs were
31 seen in the alveolar ducts adjacent to terminal bronchioles as early as 3 mo of exposure, and
December 1994 5-28 DRAFT-DO NOT QUOTE OR CITE
-------
£
el
o
0 &
aw
II
ffl
Si
4) M-i
wj O
E 8
ry chan
rease in
weight
walls
ON
d
ft!
J3 rt ^H
8^2
ss
•a
u
J3
•a
og
ltlcf
X
vt
,1
!
:;!
I
«
o
xp
«
oo
^
«n
oo"
I I
SC*1
M
o d
. .
•" u SO
oo >n O
ON <- |
«
ts
d
S'
2
|( B,
- u
Ills 11
December 1994
5-29 DRAFT-DO NOT QUOTE OR CITE
-------
§a
-5
U O
§5
3^
H
S
•A
M
02
H
to
ug;
H^
||
U
£ .a
sill
«25Sc
8
§S^?8
o d 6 -^ rn
(N *O O
*—t O\ ^H
r*^ f*i r*-
CO t"» OO Os ^O
f- <-• t*- «s en
m —< ^- oo m
»-i >n m" oo" vcT
A at E4 (4
73
u
ja ^^
1 »
*4> O\
a c.
•3 o
2. is 8
Inflammatory changes; thickened alv
septa; bronchiolo-alveolarhyperplas:
emphysema (diagnostic methodolog)
described)
^
m
m
JS
NO
cn
rn
oo"
Tfr
Tfr
01
Tf
8) I/I
5- * "»
•a ™ S
— m «
PM
§ e"
>B >N
X vt
•s
u
i i
'« ON
EC C
M
al
1*
Inflammatory changes; bronchioloalt
hyperplasia; alveolar lipoproteinosis;
^
m
*
vO
C*">
oo*
(S
>•» # H)
-H m -*
2
z
8"
K
73
u
IN?
.S oo
E C
I »
" 3
Thickened alveolar septa; AM aggre
inflammatory changes; hyperplasia; 1
tumors
-
fS
s
C*l
NO"
in
•t
fS
M
*!^
>? f u
11*
£-3$
— m -•
U.
1
^
1
1
o °o 73
c2 s
Is"!
M C>
lljl
Minimal response at 0.25 and ultras)
changes at 0.75 mg/m3; thickened al
membranes; cell proliferation; flbros
6.0 mg/m3; increase in PMN at 0.75
and 1.5 mg/m3
1 1 1 1
1 1 1 1
1 1 1 1
\Q OO \Q S
oo in — «_ «
rl QO r^ oo
— \o
v>
-? ^
fs| I/-) •§ ^^
sf*
8 S1
ll
m
00
2
73 ^
S 73
b* *-•
A« ^ S
Is. -3
Inflammatory changes; AM aggregat
bronchiolar epithelial metaplasia; Ty
cell hyperplasia; peribronchiolar fibr
^°
(S ^'
O m
'•d' OO
vp cs
»-*" m"
•^- oo
o ^b
*O (S
M
J*
n*
*3*
oo r^ —
s
r*i
1
M
U
u S,
i g
1
B §
if
December 1994
5-30
DRAFT-DO NOT QUOTE OR CITE
-------
1 particles were seen within the interstitium of the alveolar ducts. These macular lesions
2 increased in size up to 12 mo of exposure. Collagen or reticulum fibers were seen only
3 rarely in association with deposited particles; the vast majority of lesions showed no evidence
4 of fibrosis. There was no evidence of focal emphysema with the macules. Multifocal
5 histiocytosis (24% of exposed rats) was observed only after 24 mo of exposure. These
6 lesions were most commonly observed subpleurally and were composed of collections of
7 degenerating macrophages and amorphous granular material within alveoli, together with
8 fibrosis and chronic inflammatory cells in the interstitium. Epithelial lining cells adjacent to
9 collections of pigmented macrophages showed a marked Type II cell hyperplasia;
10 degenerative changes were not observed in Type I cells. Histological examination of lung
11 tissue from monkeys exposed for 24 mo in the same regimen as used for rats revealed
12 aggregates of black particles, principally in the distal airways of the lung. Particles were
13 present within the cytoplasm of macrophages in the alveolar spaces as well as the
14 interstitium. Fibrosis, focal emphysema, or inflammation was not observed.
15 Histopathological effects of diesel exhaust on the lungs of rats have been investigated
16 by the Health Effects Research Program on Diesel Exhaust (HERP) in Japan. Both light-
17 duty (LD) and heavy-duty (HD) diesel engines were used. The exhaust was diluted to
18 achieve nominal concentrations of 0.1 (LD only), 0.4 (LD and HD), 1 (LD and HD),
19 2 (LD and HD), and 4 (HD only) mg/m3 paniculate matter. Rats were exposed for
20 16 h/day, 6 day/week for 30 mo. No histopathological changes were observed in the lungs
21 of rats exposed to 0.4 mg/m3 paniculate matter or less. At concentrations above 0.4 mg/m3
22 paniculate matter, severe morphological changes were observed. These changes consisted of
23 shortened and absent cilia in the tracheal and bronchial epithelium, marked hyperplasia of the
24 bronchiolar epithelium, swelling of the Type II cellular epithelium, and increased incidences
25 of lung adenomas and carcinomas at the 4 mg/m3 particle concentration. These lesions
26 appeared to increase in severity with increases in exhaust concentration and duration of
27 exposure. There was no difference in the degree of changes in pulmonary pathology at the
28 same level of concentrations between the LD and the HD series.
29 Histological examination of the respiratory tract of hamsters revealed significantly
30 higher numbers of hamsters exhibiting definite proliferative changes in the lungs in the group
31 exposed to diesel exhaust than were observed in the group exposed to particle-free diesel
December 1994 5.31 DRAFT-DO NOT QUOTE OR CITE
-------
1 exhaust or clean air (Heinrich et al., 1982). Sixty percent of these changes were described
2 as adenomatous proliferations. Exposures were for 7 to 8 h/day, 5 days/week for 104 weeks
3 to diesel exhaust diluted to achieve a concentration of 3.9 mg/m3 paniculate matter.
4 Iwai et al. (1986) performed serial histopathology on the lungs of rats at 1, 3, 6, 12,
5 and 24 mo of exposure to diesel exhaust. Exposures were for 8 h/day, 7 days/week for
6 24 mo; the exposure atmosphere contained 4.9 mg/m3 particulate matter. At 1 and 3 mo of
7 exposure, there were minimal histological changes in the lungs of the exposed rats. After
8 6 mo of exposure, there were particle-laden macrophages distributed irregularly throughout
9 the lung and a proliferation of Type II cells with adenomatous metaplasia in areas where the
10 macrophages had accumulated. After 1 year of exposure, foci of heterotrophic hyperplasia
11 of ciliated or nonciliated bronchiolar epithelium on the adjacent alveolar walls were more
12 common, the quantity of deposited particulate matter increased, and the number of
13 degenerative AMs and proliferative lesions of Type II or bronchiolar epithelial cells
14 increased. After 2 years of exposure, there was a fibrous thickening of the alveolar walls,
15 mast cell infiltration with epithelial hyperplasia in areas where the macrophages had
16 accumulated, and neoplasms.
17 Heinrich et al. (1986a; see also Stober, 1986) performed histopathologic examinations
18 of the respiratory tract of hamsters, mice, and rats exposed to diesel exhaust that had
19 4 mg/m3 particulate matter. Exposures were for 19 h/day, 5 days/week; the maximum
20 exposure period was 120 weeks for hamsters and mice and 140 weeks for rats. Histological
21 examination revealed different levels of response among the three species. In hamsters, the
22 exhaust produced thickened alveolar septa, bronchiolo-alveolar hyperplasia, and what were
23 termed emphysematous lesions (diagnostic methodology not described). In mice,
24 bronchioloalveolar hyperplasia occurred in 64% of the mice exposed to the exhaust and in
25 5% of the controls. Multifocal alveolar lipoproteinosis occurred in 71% and multifocal
26 interstitial fibrosis occurred in 43% of the mice exposed to exhaust but in only 4% of the
27 controls. In exposed rats, there were severe inflammatory changes in the lungs, as well as
28 thickened septa, foci of macrophages, and hyperplastic and metaplastic lesions.
29 The effects of diesel exhaust on the lungs of 18-week-old rats exposed to
30 8.3 ± 2.0 mg/m3 particulate matter were investigated by Karagianes et al. (1981).
31 Exposures were for 6 h/day, 5 days/week, for 4, 8, 16, or 20 mo. Histological examinations
December 1994 5-32 DRAFT-DO NOT QUOTE OR CITE
-------
1 of lung tissue noted focal aggregation of particle-laden AMs, alveolar histiocytosis,
2 interstitial fibrosis, and alveolar emphysema (diagnostic methodology not described). Lesion
3 severity was related to length of exposure. No significant differences in lesion severity
4 between diesel exhaust, diesel exhaust plus coal dust (5.8 ± 3.5 mg/m3), or the high
5 concentration (14.9 ±6.2 mg/m3) coal dust exposure groups following 20 mo of exposure.
6 Histological changes in the lungs of guinea pigs exposed to diluted diesel exhaust
7 containing either 0.25, 0.75, 1.5, or 6.0 mg/m3 paniculate matter were reported by Barnhart
8 et al. (1981, 1982). Exposures at 0.75 and 1.5 mg/m3 for 2 weeks to 6 mo resulted in an
9 uptake of exhaust particles by three alveolar cell types (AMs, Type I cells, and interstitial
10 macrophages) and also by granulocytic leukocytes (eosinophils). The alveolar-capillary
11 membrane increased in thickness as a result of an increase in the absolute tissue volume of
12 interstitium and Type II cells. In a continuation of these studies, guinea pigs were exposed
13 to diesel exhaust (up to 6.0 mg/m3 paniculate matter) for 2 years (Barnhart, et al., 1982).
14 A minimal tissue response occurred at the concentration of 0.25 mg/m3- After 9 mo of
15 exposure, there was a significant increase, about 30%, in Type I and II cells, endothelial
16 cells, and interstitial cells over concurrent age-matched controls; by 24 mo only macrophages
17 and Type II cells were significantly increased. As in the earlier study, ultrastructural
18 evaluation showed that Type I cells, AMs and eosinophils phagocytized the diesel particles.
19 Exposure to 0.75 mg/m3 for 6 mo resulted in fibrosis in regions of macrophage clusters and
20 in focal Type II cell proliferation. No additional information was provided regarding the
21 fibrotic changes with increasing concentration or duration of exposure. With increasing
22 concentration/duration of diesel exhaust exposure, Type II cell clusters occurred in some
23 alveoli. Intraalveolar debris was particularly prominent after exposures at 1.5 and
24 6.0 mg/m3 and consisted of secretory products from Type II cells.
25 In studies conducted on hamsters, Pepelko (1982b) found that the lungs of hamsters
26 exposed for 8 h/day, 7 days/week for 6 mo to 6 or 12 mg/m3 paniculate matter were
27 characterized by large numbers of black AM in the alveolar spaces, thickening of the
28 alveolar epithelium, hyperplasia of Type II cells, and edema.
29 Lungs from rats and mice exposed to 0.35, 3.5 or 7 mg/m3 (0.23 to 0.26 pm mass
30 median diameter [MMD]) for 7 h/day and 5 days/week showed pathologic lesions (Mauderly
31 et al., 1987b; Henderson et al., 1988). After 1 year of exposure at 7 mg/m3, the lungs of
December 1994 5.33 DRAFT-DO NOT QUOTE OR CITE
-------
1 the rats exhibited focal areas of fibrosis; fibrosis increased with increasing duration of
2 exposure and was observable in the 3.5-mg/m3 group of rats at 18 mo. The severity of
3 inflammatory responses and fibrosis was directly related to the exposure level. In the
4 0.35 mg/m3 group of rats, there was no inflammation or fibrosis. Although the mouse lungs
5 contained high lung burdens of diesel particles per gram of lung weight at each equivalent
6 exposure concentration, there was substantially less inflammatory reaction and fibrosis than
7 was the case hi rats. Fibrosis was observed only in the lungs of mice exposed at 7 mg/m3
8 and consisted of fine fibrillar thickening of an occasional alveolar septa.
9 Histological examinations were performed on the lungs of cats initially exposed to
10 6 mg/m3 particulate matter for 61 weeks and subsequently increased to 12 mg/m3 for
11 Weeks 62 to 124 of exposure. Plopper et al. (1983; see also Hyde et al., 1985) concluded
12 from the results of this study that exposure to diesel exhaust produced changes in both
13 epithelial and interstitial tissue compartments and that the focus of these lesions in the
14 peripheral lung was the centriacinar region where the alveolar ducts join the terminal
15 conducting airways. This conclusion was based on the following evidence. The epithelium
16 of the terminal and respiratory bronchioles in exposed cats consisted of three cell types
17 (ciliated, basal, and Clara cells), compared with only one type (Clara cells) in the controls.
18 The proximal acinar region showed evidence of peribronchial fibrosis and bronchiolar
19 epithelial metaplasia. Type II cell hyperplasia was present in the proximal interalveolar
20 septa. The more distal alveolar ducts and the majority of the rest of the parenchyma were
21 unchanged from controls. Peribronchial fibrosis was greater at the end of 6 mo in clean air
22 following exposure, whereas the bronchiolar epithelial metaplasia was most severe at the end
23 of exposure. Following an additional 6 months in clean air, the bronchiolar epithelium more
24 closely resembled the control epithelial cell population.
25 Wallace et al. (1987) used transmission electron microscopy (TEM) to determine the
26 effect of diesel exhaust on the intravascular and interstitial cellular populations of the lungs
27 of exposed rats and guinea pigs. Exposed animals and matched controls were exposed to
28 0.25, 0.75, 1.5, or 6.0 mg/m3 particulate matter for 2, 6, or 10 weeks or 18 mo. The
29 results inferred the following: (1) exposure to 6.0 mg/m3 for 2 weeks was insufficient to
30 elicit any cellular response, (2) both species demonstrated an adaptive multi-cellular response
31 to diesel exhaust, (3) increased numbers of fibroblasts were found in the interstitium from
December 1994 5.34 DRAFT-DO NOT QUOTE OR CITE
-------
1 Week 6 of exposure through Month 18, and (4) there was no significant difference in either
2 cell type or number in alveolar capillaries, but there was a significant increase at 18 mo in
3 the mononuclear population in the interstitium of both species.
4 Additional means for assessing the adverse effects of diesel exhaust on the lung are to
5 examine biochemical and cytological changes in bronchoalveolar lavage fluid (BALF) and in
6 lung tissue. Fedan et al. (1985) performed studies to determine whether chronic exposure of
7 rats affected the pharmacologic characteristics of the rat's airway smooth muscle.
8 Concentration-response relationships for tension changes induced with acetylcholine,
9 5-hydroxytryptamine, potassium chloride, and isoproterenol were assessed in vitro on isolated
10 preparations of airway smooth muscle (trachealis). Chronic exposure to diesel exhaust
11 significantly increased the maximal contractile responses to acetylcholine compared with
12 control values; exposure did not alter the sensitivity (EC50 values) of the muscles to the
13 agonists. Exposures were to diesel exhaust containing 2 mg/m3 paniculate matter for
14 7 h/day, 5 days/week for 2 years.
15 ' Biochemical studies of BALF obtained from hamsters and rats revealed that exposures
16 to diesel exhaust caused significant increases in lactic dehydrogenase, alkaline phosphatase,
17 glucose-6-phosphate dehydrogenase (G6P-DH), total protein, collagen, and protease (pH 5.1)
18 after approximately 1 year and 2 years of exposure (Heinrich et al., 1986a). These
19 responses were generally much greater in rats than in hamsters. Exposures were to diesel
20 exhaust containing 4.24 mg/m3 particulate matter for 19 h/day, 5 days/week for
21 120 (hamsters) to 140 (rats) weeks.
22 Protein, fi-glucuronidase activity, and acid phosphatase activity were significantly
23 elevated in BALF obtained from rats exposed to diesel exhaust containing 0.75 or 1.5 mg/m3
24 particulate matter for 12 mo (Strom, 1984). Exposure for 6 mo resulted in significant
25 increases in acid phosphatase activity at 0.75 mg/m3 and in protein, fi-glucuronidase, and
26 acid phosphatase activity at the 1.5 mg/m3 concentration. Exposure at 0.25 mg/m3
27 particulate matter did not affect the three indices measured at either time period. The
28 exposures were for 20 h/day, 5.5 days/week for 52 weeks.
29 Additional biochemical studies (Misiorowski et al., 1980) were conducted on
30 laboratory animals exposed under the same conditions and at the same site as reported on by
31 Strom (1984). In most cases, exposures at 0.25 mg/m3 did not cause any significant
December 1994 5.35 DRAFT-DO NOT QUOTE OR CITE
-------
1 changes. The DNA content in lung tissue and the rate of collagen synthesis were
2 significantly increased at 1.5 mg/m3 paniculate matter after 6 mo. Collagen deposition was
3 not affected. Total lung collagen content increased in proportion to the increase in lung
4 weight. The activity of prolyl hydroxylase was significantly increased at 12 weeks at
5 0.25 and 1.5 mg/m3; it then decreased with age. Lysal oxidase activity did not change.
6 After 9 mo of exposure, there were significant increases in lung phospholipids in rats and
7 guinea pigs exposed to 0.75 mg/m3 and in lung cholesterol in rats and guinea pigs exposed to
8 1.5 mg/m3. Pulmonary prostaglandin dehydrogenase activity was stimulated by an exposure
9 at 0.25 mg/m3 but was not affected by exposure at 1.5 mg/m3 (Chaudhari et al., 1980,
10 1981). Exposures for 12 or 24 weeks resulted in a concentration-dependent lowering of this
11 enzyme activity. Exposure of male rats and guinea pigs at 0.75 mg/m3 for 12 weeks did not
12 cause any changes in glutathione levels of the lung, heart, or liver. Rats exposed for
13 2 months at 6 mg/m3 showed a significant depletion of hepatic glutathione, whereas the lung
14 showed an increase of glutathione (Chaudhari and Dutta, 1982). Schneider and Felt (1981)
15 reported that similar exposures did not substantially change adenylate cyclase and guanylate
16 cyclase activities in lung or liver tissue of exposed rats and guinea pigs.
17 Bhatnagar et al. (1980; see also Pepelko, 1982a) evaluated changes in the biochemistry
18 of lung connective tissue of diesel-exposed rats and mice. The mice were exposed for
19 8 h/day and 7 days/week for up to 9 mo, to exhaust containing 6 mg/m3 particulate matter.
20 Total lung protein in rats exposed for 42 days increased about 40% over that of controls.
21 In vivo leucine incorporation was decreased, suggesting a decrease in overall protein
22 synthesis. In vivo proline incorporation, an estimate of collagen synthesis, was not affected
23 by exposure. Prolylhydroxylase activity was increased in rats exposed 42 days and in rats
24 exposed in utero, suggesting increased collagen synthesis. In mice exposed to diesel exhaust
25 for up to 9 mo, large increases in lung protein content and collagen were found, but overall
26 protein synthesis decreased. The increase in collagen synthesis suggested proliferation of
27 connective tissue and possible fibrosis (Pepelko, 1982a).
28 A number of reports (Henderson et al., 1985; McClellan et al., 1986; Mauderly et al.,
29 1987b; Henderson et al., 1988) have addressed biochemical and cytological changes in lung
30 tissue and BALF of rodents exposed for 7 h/day, 5 days/week for up to 30 mo at
31 concentrations of 0, 0.35, 3.5, or 7.1 mg/m3 particulate matter. At the lowest exposure
December 1994 5-36 DRAFT-DO NOT QUOTE OR CITE
-------
1 level (0.35 mg/m3), no biochemical or cytological changes occurred in the BALF or in lung
2 tissue in either Fischer 344 rats or CD-I mice. A chronic inflammatory response was seen
3 at the two higher exposure levels in both species, as evidenced by increases in inflammatory
4 cells (macrophages and neutrophils), cytoplasmic and lysosomal enzymes (lactate
5 dehydrogenase, glutathione reductase, and 6-glucuronidase), and protein (hydroxyproline) in
6 BALF. Analysis of lung tissue indicated similar changes in enzyme levels as well as an
7 increase in total lung collagen content. After 18 mo of exposure, lung tissue glutathione was
8 depleted in a concentration-dependent fashion in rats but was slightly increased in mice.
9 Lavage fluid levels of glutathione and glutathione reductase activity increased in a
10 concentration-dependent manner and were higher in mice than in rats. Rats exposed for
11 24 mo to diesel exhaust (3.5 mg/m3 particulate matter) had a fivefold increase in the
12 bronchoconstrictive prostaglandin PGF2a and a twofold increase in the inflammatory
13 leukotriene LTB4. In similarly exposed mice, there was a twofold increase in both
14 parameters. These investigators concluded that the release of larger amounts of such
15 mediators of inflammation from the alveolar phagocytic cells of rats accounted for the greater
16 fibrogenic response seen in that species.
17 Biochemical analysis of lung tissue from cats exposed for 124 weeks and held in clean
18 air for an additional 26 weeks indicated increases of lung collagen; this finding was
19 confirmed by an observed increase in total lung wet weight and in connective tissue fibers
20 estimated morphometrically in these cats (Hyde et al., 1985). Exposures were for 7 h/day,
21 5 days/week at 6 mg/m3 particulate matter for 61 weeks and at 12 mg/m3 for
22 Weeks 62 to 124.
23 Further effects of exposure to diesel exhaust on pulmonary cytology and lung
24 biochemistry may be found in Section 5.1.2.3.
25 The pathogenic sequence following the inhalation of diesel exhaust as determined.
26 histopathologically and biochemically begins with the phagocytosis of diesel particles by
27 AMs. These activated macrophages release chemotactic factors that attract neutrophils and
28 additional AMs. As the lung burden of diesel particles increases, there is an aggregation of
29 particle-laden AMs in alveoli adjacent to terminal bronchioles, increases in the number of
30 Type II cells lining particle-laden alveoli, and the presence of particles within alveolar and
31 peribronchial interstitial tissues and associated lymph nodes. The neutrophils and
December 1994 5.37 DRAFT-DO NOT QUOTE OR CITE
-------
1 macrophages release mediators of inflammation and oxygen radicals that deplete a
2 biochemical defense mechanism of the lung (i.e., glutathione). As will be described later in
3 more detail, other defense mechanisms are affected, particularly the decreased viability of
4 AMs which leads to decreased phagocytic activity and death of the macrophage. The latter
5 series of events may result in the presence of pulmonary inflammatory, fibrotic, or
6 emphysematous lesions. The data suggest that there may be a threshold of exposure to diesel
7 exhaust below which adverse structural and biochemical effects may not occur in the lung;
8 however, differences in the anatomy and pathological responses of laboratory animals
9 coupled with their lifespans compared with humans make a determination of human levels of
10 exposure to diesel exhaust without resultant pulmonary injury a difficult and challenging
11 endeavor.
12
13 Effects on Pulmonary Defense Mechanisms
14 The respiratory system has a number of defense mechanisms that negate or
15 compensate for the effects produced by the injurious substances that repeatedly insult the
16 upper respiratory tract, the tracheobronchial airways, and the alveoli. The effects of
17 exposure on the pulmonary defense mechanisms of laboratory animals as well as more details
18 on exposure atmosphere are summarized in Table 5-7 and ranked by cumulative exposure
19 (C x T).
20 Several studies have been conducted investigating the effect of inhaled diesel exhaust
21 on the deposition and fate of inert tracer particles or diesel particles themselves. Lung
22 clearance of deposited particles occurs in two distinct phases: a rapid phase (hours to days)
23 from the tracheobronchial region via the mucociliary escalator and a much slower phase
24 (weeks to months) from the nonciliated pulmonary region via primarily, but not solely, AMs.
25 Battigelli et al. (1966) reported that exposure to diesel exhaust 8 to 17 mg/m3 paniculate
26 matter impaired tracheal mucociliary clearance in rats. Lewis et al. (1989) found no
27 difference in the clearance of 59Fc3O4 particles (1.5 pirn MMAD, ag 1.8) 1 day after dosing
28 control and diesel exhaust-exposed rats (2 mg/m3, 7 h/day, 5 days/week for 8 weeks).
29 Wolff et al. (1987) and Wolff and Gray (1980) studied the effects of both subchronic
30 and chronic diesel exhaust exposure on the tracheal clearance of particles. Tracheal
31 clearance assessments were made by measuring the retention of radiolabeled technetium
December 1994 5_38 DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 5-7. EFFECTS OF EXPOSURE TO DIESEL EXHAUST ON THE PULMONARY
DEFENSE MECHANISMS OF LABORATORY ANIMALS
u>
o
o
o
G
O
H
W
O
Species
Exposure
Period
Particles
(mg/m3)
C x T
(mg-h/m3)
CO NO2 SQj
(ppm) (ppm) (ppm)
Effects
Reference
ALVEOLAR MACROPHAGE STATUS
Guinea Pig,
Hartley
Rat, F-344, M
Rat, F-344, M
Rat F-344/Crl,
M, F
Mouse, CD,
M,F
Rat
20 h/day
5.5 days/week
8 weeks
7 h/day
5 days/week
104 weeks
20 h/day
5.5 days/week
26, 48, or
52 weeks
7 h/day
5 days/week
104 weeks (rat),
78 weeks
(mouse)
7 h/day
5 day/week
12 weeks
0.25
1.5
0.19/tm, MMD
2.0
0.23-0.36 urn MMD
0.25a
0.751
1.5"
0.19 urn, MMD
0.35
3.5
7.0
0.25 /un, MMD
0.2
1.0
4.5
0.25f«m, MMD
220
1,320
7,280
715-8,580
1,274°
12.740F
25,480?
84
420
1,890
2.9 — —
j 5
11.5 1.5 0.81
2.9 - -
4.8 - -
7.5 — —
2.9 0.05 -
16.5 0.34 -
29.7 0.68 —
CLEARANCE
— — —
— — —
No significant changes in absolute numbers of alveolar
macrophages (AMs)
Little effect on viability, cell number, oxygen
consumption, membrane integrity, lyzomal enzyme
activity, or protein content of AMs; decreased cell
volume and ruffling of cell membrane and depressed
luminescence of AM
AM cell counts proportional to concentration of DP at
0.75 and 1.5 mg/m3; AM increased in lungs in
response to rate of DP mass entering lung rather man
total DP burden in lung; increased PMNs were
proportional to inhaled concentrations and/or duration
of exposure; PMNs affiliated with clusters of
aggregated AM rather than DP
Significant increases of AM in rats and mice exposed
to 7.0 mg/m3 DP for 24 and 18 mo, respectively, but
not at concentrations of 3.5 or 0.35 mg/m3 DP for the
same exposure durations; PMNs increased in a dose-
dependent fashion in both rats and mice exposed to
3.5 or 7.0 mg/m3 DP and were greater in mice man
rats
Evidence of apparent speeding of tracheal clearance at
the 4.5 mg/m3 level after 1 week of "Tc
macroaggregated-albuminand reduced clearance of
tracer aerosol in each of the three exposure levels at
12 weeks; indication of a lower percentage of ciliated
cells at the 1.0 and 4.5 mg/m3 levels
Chen et. al.
(1980)
Castranova et al.
(1985)
Strom (1984)
Vostal et al.
(1982)
Henderson et al.
(1988)
Wolff and Gray
(1980)
n
-------
O
O
z
s
€
O
TABLE 5-7 (cont'd). EFFECTS OF EXPOSURE TO DIESEL EXHAUST ON THE PULMONARY
DEFENSE MECHANISMS OF LABORATORY ANIMALS
1
n>
§
•^
l/i
k
Species
Rat, F-344
M, F
Rat, F-344, M
Rat, Sprague-
Dawley
Rat, F-344,
M, F
Exposure
Period
7 h/day
5 days/week
18 weeks
<0.5 A»m, MMD
7 h/day
5 days/week
26-104 weeks
4-6 h/day
7 days/week
0. 1 to 14.3 weeks
7 h/day
5 days/week
130 weeks
Particles
(mg/m3)
0.15
0.94
4.1
2.0
0.23-0.36 urn
MMD
0.9
8.0
17.0
0.35
3.5
7.0
0.25 nm, MMD
C x T
(mg-h/m3)
94.5
592
2,583
1,820-7,280
2.5-10,210
1,593
15,925
31,850
CO N02
(ppm) (ppm)
— —
— —
— —
11.5 1.5
- 5.0
- 2.7
— 8.0
2.9 0.1
16.5 0.3
29.7 0.7
S02
(ppm) Effects
— Lung burdens of DP were concentration-related;
— clearance half-time of DP almost double in
— 4.1 mg/m3 group compared to 0.15 mg/m3 group.
0.8 No difference in clearance of 59Fe3O4 particles
1 day after tracer aerosol administration; 120 days
after exposure tracer aerosol clearance was
enhanced; Lung burden of DP increased
significantly between 12 to 24 months of exposure
0.2 Impairment of tracheal mucociliary clearance in a
0.6 concentration-response manner
1.0
— No changes in tracheal mucociliary clearance after
— 6, 12, 18, 24, or 30 mo of exposure; increases in
— lung clearance half-times as early as 6 mo at
7.0 mg/m3 level and 18 mo at 3.5 mg/m3 level; no
changes seen at 0.35 mg/m3 level; after 24 mo of
diesel exposure, long-term clearance half-times
were increased in the 3.5 and 7.0 mg/m3 groups
Reference
Griffis et al. (1983)
Lewis et al. (1989)
Battigelli et al.
(1966)
Wolff et al. (1987)
MICROBIAL-INDUCED MORTALITY
Mice, CD-I, F
No change in mortality in mice exposed
intratracheally to 100 /xg of DP prior to exposure to
aerosolized Streptococcus sp.
Hatch et al. (1985)
O
90
n
>—i
3
-------
TABLE 5-7 (cont'd). EFFECTS OF EXPOSURE TO DIESEL EXHAUST ON THE PULMONARY
DEFENSE MECHANISMS OF LABORATORY ANIMALS
1
H- *
Exposure Particles C x T
Species Period (mg/m3) (mg-h/m3)
Mice CD-I, F 7 h/day 2.0 280-1,820
5 days/week 0.23-0.36 ^m MMD
4, 12, or
26 weeks
— --
Mice, CR/CD-1, F 8 h/day 5.3 to 7.9 11-20,350
7 days/week
2 hup to
46 weeks
CO NO2
(ppm) (ppm)
11.5 1.5
19 1.8
to to
22 3.6
SO,
(ppm)
0.8
0.9
to
2.8
Effects
Mortality similar at each exposure duration when
challenged with Ao/PR/8/34 influenza virus; in mice
exposed for 3 and 6 mo, but not 1 mo, there were
increases in the percentages of mice having lung
consolidation, higher virus growth, depressed
interferon levels and a four-fold reduction in
hemagglutin antibody levels
Enhanced susceptibility to lethal effects of
S. pyogenes infections at all exposure durations
(2 and 6 h; 8, 15, 16, 307, and 321 days);
inconclusive results with S. typhimunum because of
high mortality rates in controls; no enhanced
mortality when challenged with A/PR8-3 influenza
virus
Reference
Hahon et al.
(1985)
Campbell et al.
(1980, 1981)
'Chronic exposure lasted 52 weeks.
bChronic exposure lasted 48 weeks.
Calculated for 104-week exposure.
DP = Diesel exhaust particles.
AM = Alveolar macrophage.
PMN = Poly mo rphonuclear leukocyte.
-------
1 macroaggregated-albumin remaining 1 h after instillation in the distal trachea of rats. In the
2 subchronic studies, rats were exposed to 4.5, 1.0, or 0.2 mg/m3 particulate matter on a
3 7 h/day, 5 days/week schedule for up to 12 weeks. After 1 week there was an apparent
4 speeding of tracheal clearance at the 4.5 mg/m3 exposure level (p = 0.10), which returned
5 toward baseline after 6 weeks and was slightly below the baseline rate at 12 weeks. In the
6 1.0 mg/m3 group, there was a progressive significant reduction in the clearance rate at 6 and
7 12 weeks of exposure. There was a trend toward reduced clearance in the 0.2 mg/m3 group.
8 Scanning electron micrographs indicated minimal changes in ciliary morphology; however,
9 there was an indication of a lower percentage of ciliated cells at the 1.0 and
10 4.5 mg/m3 levels. In the chronic studies, rats were exposed to 0, 0.35, 3.5, or 7.0 mg/m3
11 for 7 h/day, 5 days/week for 30 mo. There were no significant differences in tracheal
12 clearance rates between the control group and any of the exposure groups after 6, 12, 18, 24,
13 or 30 mo of exposure. The preexposure measurements for all groups, however, were
14 significantly lower than those during the exposure period, suggesting a possible age effect.
15 The preexposure value for the 3.5-mg/m3 group was also significantly lower than the control
16 group.
17 There is a substantial body of evidence for an impairment of particulate clearance
18 from the bronchiole-alveolar region of rats following exposure to diesel exhaust.
19 Griffis et al. (1983) exposed rats 7 h/day, 5 days/week for 18 weeks to diesel exhaust at
20 0.15, 0.94, or 4.1 mg/m3 particulate matter. Lung burdens of the 0.15, 0.94, and
21 4.1 mg/m3 levels were 35, 220, and 1,890 pg/g, lung, respectively, 1 day after the 18-week
22 exposure. The clearance half-time of the diesel particles was significantly greater, almost
23 double, for the 4.1-mg/m3 exposure group than for those of the lower exposure groups,
24 165 ± 8 days versus 99 ± 8 days (0.94 mg/m3) and 87 ± 28 days (0.15 mg/m3),
25 respectively.
26 Chan et al. (1981) showed a dose-related slowing of 14C-diesel particle clearance in
27 rats preexposed to diesel exhaust at 0.25 or 6 mg/m3 particulate matter for 20 h/day,
28 7 days/week for 7 to 112 days. Clearance was inhibited in the 6-mg/m3 group when
29 compared by length of exposure or compared with the 0.25-mg/m3 or control rats at the
30 same time periods.
December 1994 5.42 DRAFT-DO NOT QUOTE OR CITE
-------
1 Heinrich et al. (1982) evaluated lung clearance in rats exposed for approximately
2 18 mo at 3.9 mg/m3 paniculate matter for 7 to 8 h/day, 5 days/week. Following exposure to
3 59Fe2O3-aerosol, the rats were returned to the diesel exhaust exposure and the radioactivity
4 was measured over the thoracic area at subsequent times. The biological half-life of the iron
5 oxide deposited in the rats' lungs was nearly twice that of controls.
6 Wolff et al. (1987) investigated alterations in particle clearance from the lungs of rats
7 chronically exposed to diesel exhaust at 0, 0.35, 3.5, or 7.0 mg/m3 paniculate matter for
8 7 h/day, 5 days/week for up to 24 mo. Progressive increases in lung burdens were observed
9 over time hi the 3.5- and 7.0-mg/m3 exposure groups. Levels of diesel particles in terms of
10 milligrams per lung were 0.60, 11.5, and 20.5 after 24 mo of exposure at the 0.35-, 3.5-, or
11 7.0-mg/m3 exposure levels, respectively. There were significant increases in 16-day
12 clearance half-times of inhaled radiolabeled particles of 67Ga2O3 (0.1 /zm, MMD) as early as
13 6 mo at the 7.0-mg/m3 level and 18 mo at the 3.5-mg/m3 level; no significant changes were
14 seen at the 0.35-mg/m3 level. Rats inhaled fused aluminosilicate particles (2 pm MMAD)
15 labeled with 134Cs after 24 mo of diesel exhaust exposure; long-term clearance half-times
16 were 79, 81, 264, and 240 days for the 0-, 0.35-, 3.5-, and 7.0-mg/m3 groups, respectively.
17 Differences were significant between the control and the 3.5 and 7.0 mg/m3 groups
18 (p < 0.01).
19 Lewis et al. (1989) conducted lung burden and 59Fe3O4 tracer studies in rats exposed
20 for 12 and 24 mo to 2 mg/m3 paniculate matter (7 h/day, 5 days/week). The slope of the
21 Fe3O4 clearance curve was significantly steeper than that of the controls, indicating a more
22 rapid alveolar clearance of the deposited 59Fe3O4. After 120 days from the inhalation of the
23 tracer particle, 19% and 8% of the initially deposited 59Fe3O4 was present in the lungs of
24 control and diesel exhaust-exposed rats, respectively. The lung burden of diesel particles,
25 however, increased significantly between 12 and 24 mo of exposure (0.52 to 0.97% lung dry
26 weight), indicating a later, dose-dependent inhibition of clearance.
27 Alveolar macrophages, because of their phagocytic and digestive capabilities, are one
28 of the prime defense mechanisms of the alveolar region of the lung against inhaled particles.
29 Thus, characterization of the effects of diesel exhaust on various properties of AMs provides
30 information on the integrity or compromise of a key pulmonary defense mechanism. The
31 physiological viability of AM from diesel-exposed rats was assessed after 2 years of exposure
December 1994 5.43 DRAFT-DO NOT QUOTE OR CITE
-------
1 by Castranova et al. (1985). The 7 h/day, 5 days/week exposure at 2 mg/m3 paniculate
2 matter had little effect on the following: viability, cell number, oxygen consumption,
3 membrane integrity, lysosomal enzyme activity, or protein content of the macrophages.
4 A slight decrease in cell volume, a decrease in chemiluminescence indicative of a decreased
5 secretion of reactive oxygen species and a decrease in ruffling of the cell membrane were
6 observed. These findings could be reflective of an overall reduction in phagocytic activity.
7 Exposure to diesel exhaust has been reported both to increase the number of
8 recoverable AMs from the lung (Strom, 1984; Vostal et al., 1982; Henderson et al., 1988)
9 or to produce no change in numbers (Chen et al., 1980; Castranova et al., 1985). Strom
10 (1984) found that in rats exposed to 0.25 mg/m3 paniculate matter for 20 h/day,
11 5.5 days/week for 6 mo or 1 year, as well as in the controls, BAL cells consisted entirely of
12 AMs, with no differences in the cell counts in the lavage fluid. At the higher concentrations,
13 0.75 or 1.5 mg/m3, the count of AM increased proportionally with the exposure
14 concentration; the results were identical for AMs at both 6 and 11 or 12 mo of exposure.
15 The increase in AM counts was much larger after exposure to 1.5 mg/m3 for 6 mo than after
16 exposure to 0.75 mg/m3 for 1 year, although the total mass (calculated as C X T) of
17 deposited paniculate burden was the same. These data suggested to the authors that the
18 number of lavaged macrophages was proportional to the mass influx of particles, rather than
19 to the actual diesel particulate burden in the lung. These results further implied that there
20 may be a threshold for the rate of mass influx of diesel particles into the lungs of rats above
21 which there was an increased recruitment of AMs. Henderson et al. (1988) reported similar
22 findings of significant increases of AMs in rats and mice exposed to 7.0 mg/m3 particulate
23 matter for 18 and 24 mo, respectively, for 7 h/day, 5 days/week, but not at concentrations of
24 3.5 or 0.35 mg/m3 for the same exposure durations. Chen et al. (1980), using an exposure
25 regimen of 0.25 and 1.5 mg/m3 particulate matter for 2 mo and 20 h/day and 5.5 days/week,
26 found no significant changes in absolute numbers of AMs from guinea pig BALF nor did
27 Castranova et al. (1985) in rat BALF following exposure to 2 mg/m3 particulate matter for
28 7 h/day, 5 days/week for 2 years.
29 A similar inflammatory response was noted by Henderson et al. (1988) and Strom
30 (1984), as evidenced by an increased number of PMNs present in BALF from rodents
31 exposed to diesel exhaust. Henderson et al. (1988) found these changes in rats and mice
December 1994 5.44 DRAFT-DO NOT QUOTE OR CITE
-------
1 exposed to 7.0 and 3.5 mg/m3 paniculate matter for 7 h/day, 5 days/week. Significant
2 increases in BALF PMNs were observed in mice at 6 mo of exposure and thereafter at the
3 7.0 and 3.5 mg/m3 exposure levels, but in rats only the 7.0 mg/m3 exposure level showed an
4 increase in BALF PMNs at 6 mo of exposure and thereafter. Significant increases in BALF
5 PMNs occurred in rats at 12, 18, and 24 mo of exposure to 3.5 mg/m3 paniculate matter.
6 Increases in PMNs were usually greater in mice. Strom (1984) reported that the increased
7 numbers of PMNs in BALF were proportional to the inhaled concentrations and/or duration
8 of exposure. The PMNs also appeared to be affiliated with clusters of aggregated AMs
9 rather than to the diesel particles per se. Proliferation of Type II cells likewise occurred in
10 response to the formed aggregates of AMs (White and Garg, 1981).
11 The integrity of pulmonary defense mechanisms can also be ascertained by assessing if
12 exposure to diesel exhaust affects the colonization and clearance of pathogens and alters the
13 challenged animals' response to respiratory tract infections. Campbell et al. (1980, 1981)
14 exposed mice to diesel exhaust followed by infectious challenge with Salmonella
15 typhimurium, Streptococcus pyogenes, or A/PR8-3 influenza virus and measured microbial-
16 induced mortality. Exposures to the diesel exhaust were to 6 mg/m3 paniculate matter for
17 8 h/day, 7 days/week for up to 321 days. Exposure to the diesel exhaust resulted in
18 enhanced susceptibility to the lethal effects of 5. pyogenes infection at all exposure durations
19 (2 h, 6 h; 8, 15, 16, 307, and 321 days). Tests with S. typhimurium were inconclusive
20 because of the high mortality rates in the controls. The mice exposed to diesel exhaust did
21 not exhibit an enhanced mortality when challenged with the influenza virus. Hatch et al.
22 (1985) found no changes in the susceptibility of mice to Group C Streptococcus sp. infection
23 following intratracheal injection of 100 /xg of diesel exhaust particles suspended in unbuffered
24 saline.
25 Hahon et al. (1985) assessed virus-induced mortality, virus multiplication with
26 concomitant interferon (IFN) levels (lungs and sera), antibody response, and lung
27 histopathology in mice exposed to diesel exhaust prior to infectious challenge with
28 Ao/PR/8/34 influenza virus. Weanling mice were exposed to the diesel exhaust containing
29 2 mg/m3 particulate matter for 7 h/day, 5 days/week. In mice exposed for 1, 3, and 6 mo,
30 mortality was similar between the exposed and control mice. In mice exposed for 3 and
31 6 mo, however, there were significant increases in the percentage of mice having lung
December 1994 5.45 DRAFT-DO NOT QUOTE OR CITE
-------
1 consolidation, higher virus growth, depressed interferon levels, and a fourfold reduction in
2 hemagglutinin antibody levels; these effects were not seen after the 1-mo exposure.
3 The effects of diesel exhaust on the pulmonary defense mechanisms are determined by
4 three critical factors related to exposure: the concentrations of the pollutants, the exposure
5 duration, and the exposure pattern. Higher doses of diesel exhaust as determined by an
6 increase in one or more of these three variables have been reported to increase the numbers
7 of AMs, PMNs, and Type II cells in the lung, whereas lower doses fail to produce such
8 changes. The single most significant contributor to the impairment of the pulmonary defense
9 mechanisms appears to be an excessive accumulation of diesel particles, particularly as
10 particle-laden aggregates of AMs. Such an accumulation would result from an increase in
11 deposition and/or a reduction in clearance. The deposition of particles does not appear to
12 change significantly following exposure to equivalent diesel exhaust doses over time.
13 Because of the significant nonlinearity in particle accumulation between low and high doses
14 of diesel exhaust exposure, coupled with no evidence of increased particle deposition, an
15 impairment in one or more of the mechanisms of pulmonary defense appears to be
16 responsible for the particle accumulation and subsequent pathological sequelae. The time of
17 onset of pulmonary clearance impairment was dependent both on the magnitude and on the
18 duration of exposures. For example, rats exposed for 7 h/day, 5 days/week for 104 weeks,
19 the concentration needed to induce pulmonary clearance impairment appears to lie between
20 0.35 and 2.0 mg/m3 paniculate matter.
21
22 Effects on the Immune System
23 The effects of diesel exhaust on the immune system of guinea pigs were investigated
24 by Dziedzic (1981). Exposures were to 1.5 mg/m3 particulate matter for 20 h/day,
25 5.5 days/week for up to 8 weeks. There was no effect of diesel exposure when compared
26 with matched controls for the number of B and T lymphocytes and null cells isolated from
27 the tracheobronchial lymph nodes, spleen, and blood. Cell viability as measured by trypan
28 blue exclusion was comparable between the exposed and control groups. The results of this
29 study and others on the effects of exposure to diesel exhaust on the immune system are
30 summarized in Table 5-8.
December 1994 5.46 DRAFT-DO NOT QUOTE OR CITE
-------
December ]
»
4^
Lf\
I
4±
~J
TABLE 5-8. EFFECTS OF EXPOSURES TO DIESEL EXHAUST ON
THE IMMUNE SYSTEM OF LABORATORY ANIMALS
Exposure
Species Period
Mouse, BDFI, F —
Guinea Pig; 20 h/day
Hartley, M 5.5 days/week_
4 or 8 weeks
Rat, F-344, M 7 h/day
5 days/week
52 or
104 weeks
Rat, F-344; 7 h/day
Mouse, CD-I 5 days/week
104 weeks
Particles
(mg/m3)
1.5
0.19pm MMD
2.0
0.23-0.36 Mm, MMD
0.35
3.5
7.0
0.25 urn, MMD
C x T CO NO2 SO2
(mg-h/m3) (ppm) (ppm) (ppm) Effects
— — — — Intranasally delivered doses of diesel particles as low as
1 ng exerted an adjuvant activity for IgE antibody
production
660 or 7,280 7.5 — — No alterations in numbers of B, T, and null lymphocytes
or cell viability among lymphocytes isolated from
tracheobronchial lymph nodes, spleen, or blood
3,640 or 7,280 11.5 1.5 0.8 Neither humoral immunity (assessed by enumerating
antibody-producing cells) nor cellular immunity (assessed
by the lymphocyte blast transformation assay) were
markedly affected
1,274 2.9 0.05 — Total number of anti-sheep red blood cell IgM AFC in the
12,740 16.5 0.34 — lung-associated lymph nodes was elevated in rats exposed
25,480 29.7 0.68 — to 3.5 or 7.0 mg/m3 DP (no such effects in mice); total
number of AFC per 106 lymphoid cells in lung-associated
lymph nodes and level of specific IgM, IgG, or IgA in rat
sera were not altered
References
Takafuji et al.
(1987)
Dziedzic (1981)
Mentnech et al.
(1984)
Bice et al.
(1985)
DP = Diesel particle.
AFC = Antibody forming cells.
-------
1 Mentnech et al. (1984) examined the effect of diesel exhaust on the immune system of
2 rats. Exposures were to 2 mg/m3 paniculate matter for 7 h/day, 5 days/week for up to
3 2 years. Rats exposed for 12 and 24 mo were tested for immunocompetency by determining
4 antibody-producing cells in the spleen 4 days after immunization with sheep erythrocytes.
5 The proliferative response of splenic T-lymphocytes to the mitogens concanavalm A and
6 phytohemagglutinin was assessed in rats exposed for 24 mo. There were no significant
7 differences between the exposed and control animals. Results obtained from these two assays
8 indicate that neither humoral immunity (assessed by enumerating antibody-producing cells)
9 nor cellular immunity (assessed by the lymphocyte blast transformation assay) were markedly
10 affected by the exposures.
11 Bice et al. (1985) evaluated whether or not exposure to diesel exhaust would alter
12 antibody immune responses induced after lung immunization of rats and mice. Exposures
13 were to 0.35, 3.5, or 7.0 mg/m3 for 7 h/day, 5 days/week for 24 mo. Chamber controls and
14 exposed animals were immunized by intratracheal instillation of sheep red blood cells
15 (SRBC) after 6, 12, 18, or 24 mo of exposure. No suppression in the immune response
16 occurred in either species. After 12, 18, and 24 mo of exposure, the total number of anti-
17 SRBC IgM antibody forming cells (AFCs) was elevated in rats, but not in mice, exposed to
18 3.5 or 7.0 mg/m3 particulate matter; after 6 mo of exposure, only the 7.0-mg/m3 level was
19 found to have caused this response hi rats. The number of AFC per 106 lymphoid cells in
20 lung-associated lymph nodes and the level of specific IgM, IgG, or IgA in rat sera were not
21 significantly altered. The investigators concluded that the increased cellularity and the
22 presence of diesel particles in the lung-associated lymph nodes had only a minimal effect on
23 the immune and antigen filtration function of these tissues.
24 Takafuji et al. (1987) evaluated the IgE antibody response of mice inoculated
25 intranasally at intervals of 3 weeks with varying doses of a suspension of diesel particles in
26 ovalbumin. Antiovalbumin IgE antibody liters, assayed by passive cutaneous anaphylaxis,
27 were enhanced by doses as low as 1 ptg of particles compared with immunization with
28 ovalbumin alone.
29 The inhalation of diesel exhaust appeared to have only minimal effects on the immune
30 status of rats and guinea pigs. Conversely, intranasally delivered doses as low as 1 /xg of
31 diesel particles exerted an adjuvant activity for IgE antibody production in mice. Further
December 1994 5-48 DRAFT-DO NOT QUOTE OR CITE
-------
1 studies of the effects of diesel exhaust on the immune system are needed to clarify the impact
2 of such variables as route of exposure, species, dose, and atopy.
3
4 Effects on the Liver
5 Meiss et al. (1981) examined alterations in the hepatic parenchyma of hamsters by
6 using thin-section and freeze-fracture histological techniques. Exposures to diesel exhaust
7 were for 7 to 8 h/day, 5 days/week, for 5 mo at about 4 or 11 mg/m3 paniculate matter.
8 The livers of the hamsters exposed to both concentrations of diesel exhaust exhibited
9 moderate dilatation of the sinusoids, with activation of the Kupffer cells and slight changes in
10 the cell nuclei. Fatty deposits were observed in the sinusoids, and small fat droplets were
11 occasionally observed in the peripheral hepatocytes. Mitochondria often had a loss of cristae
12 and exhibited a pleomorphic character. Giant microbodies were seen in the hepatocytes,
13 which were moderately enlarged, and gap junctions between hepatocytes exhibited a wide
14 range in structural diversity. The results of this study and others on the effect of exposure of
15 diesel exhaust on the liver of laboratory animals are summarized in Table 5-9.
16 Green et al. (1983) and Plopper et al. (1983) reported no changes in liver weights of
17 rats exposed to 2 mg/m3 paniculate matter for 7 h/day, 5 days/week for 52 weeks or of cats
18 exposed to 6 to 12 mg/m3, 8 h/day, 7 days/week for 124 weeks.
19 The use of light and electron microscopy revealed that long-term inhalation of varying
20 high concentrations of diesel exhaust caused numerous alterations to the hepatic parenchyma
21 of guinea pigs. A less sensitive index of liver toxicity, increased liver weight, failed to
22 denote an effect of diesel exhaust on the liver of the rat and cat following long-term exposure
23 to diesel exhaust. These results are too limited to understand potential impacts on the liver.
24
25 Blood and Cardiovascular Systems
26 Several studies have evaluated the effects of diesel exhaust exposure on hematological
27 and cardiovascular parameters of laboratory animals. These studies are summarized in
28 Table 5-10.
29 Standard hematological indices of toxicological effects on red and white blood cells
30 failed to denote dramatic and consistent responses. Erythrocyte (RBC) counts were reported
31 as being unaffected in cats (Pepelko and Peirano, 1983), rats and monkeys (Lewis et al.,
December 1994 5.49 DRAFT-DO NOT QUOTE OR CITE
-------
U
E
EXHAUST ON
LLS
NH M
Cd ^^
ll
0^
og
H§
W H
n S
Q PQ
S J
ta O
O ^
C/5 §
H S
U K
s J
a
Os'
Vi
U
s
^<
H
C/l
U
i
"o
Cli
a
u
*!
c5*^
z &
o 1
u §:
**k
~1
X J*>
u s
•§"g
1 »
to g
s
3^
||
w
«
I
co
2
•a
u
B
8
O
X
|
u
o
jes in absolute liver weigh
itio
•e £
o'C
Z S
00
d
VO
t-;
—
2
m
S
o jo Q
"?1
d
u
u w
I'll
n
U.
S
3
u.
1
^
oo
2
•3
ts
in
'53
S
1-al
J2 "S § -a
8 | 'gig
ijf!
sinusoids, with activated 1
inges of nuclei; fatty depo
istae and pleomorphic cha
llepatocytes had wide rang
t~
lllll
q q q
«•> O
d ~ —
o o o
r-i o\ •» J> ^
Is. "
* « *
op -a ^
t~ >/->
-------
TABLE 5-10. EFFECTS OF EXPOSURE TO DIESEL EXHAUST ON THE HEMATOLOGICAL AND
CARDIOVASCULAR SYSTEMS OF LABORATORY ANIMALS
I
§
Ji'
t/1
£»
O
£>
2
"7^
O
O
H
O
0
a
o
Q
H
Species/Sex
Monkey,
Cynomolgus, M
Rat, F-344, M,F
Guinea Pig,
Hartley, M, F
Hamster,
Syrian, M, F
Rat, F-344;
Guinea Pig,
Hartley
Rat, Wistar, M
Rat, F-3444/Jcl,
M, F
Exposure
Period
7 h/day
5 days/week
104 weeks
7 h/day
5 days/week
104 weeks
20 h/day
7 days/week
8 weeks
7-8 h/day
5 days/week
75 weeks
20 h/day
5.5 days/week
78 weeks
6 h/day
5 days/week
78 weeks
16 h/day
6 days/week
130 weeks
Particles
(mg/m3)
2
0.23-0.36 urn, MMD
2
0.23-0.36 urn, MMD
6.3"
6.8"
3.9
0.1 urn, MMD
0.25
0.75
1.5
0. 19 nm, MMD
8.3
0.71 ton, MMD
O.llc
0.41°
1.08C
2.31C
3.72d
0.1 /tm, MMD
C x T
(mg-h/m3)
7,280
7,280
7,056
7,616
10,238-11,700
2,145
6,435
12,870
19,422
1,373
5,117
13,478
28,829
46,426
CO
(ppm)
11.5
11.5
17.4
16.7
18.5
3.0
4.8
6.9
50.0
1.23
2.12
3.96
7.10
12.9
N02
(ppm)
1.5
1.5
2.3
2.9
1.2
0.11
0.27
0.49
4-6
0.08
0.26
0.70
1.41
3.00
SO2
(ppm) Effects
0.8 Increased mean corpuscular volume
(MCV)
0.8 Increase in banded neutrophils; no effect
on heart or pulmonary arteries
2.1 No effect on heart mass or ECG; small
1 .9 decrease in heart rate (IE only)
3.1 At 29 weeks, lower erythrocyte count;
increased MCV; reduced leukocyte
count
— No changes in heart mass or hematology
— at any exhaust level or duration of
— exposure in either species
— 3 % increase in COHb
0.38 At higher concentrations, RBC, Hb, Hct
1 .06 slightly elevated; MCV and mean
2 .42 corpuscular hemoglobin and
4.70 concentration were lowered
4.57
References
Lewis et al. (1989)
Lewis et al. (1989)
Vallyathan et al.
(1986)
Wiester, et al.
(1980)
Heinrich et al.
(1982)
Penney et al. (1981)
Karagianes et al.
(1981)
Research Committee
for HERP Studies
(1988)
-------
0 TABLE 5-10 (cont'd). EFFECTS OF EXPOSURE TO DIESEL EXHAUST ON THE HEMATOLOGICAL AND
8 CARDIOVASCULAR SYSTEMS OF LABORATORY ANIMALS
re Exposure
^ Species/Sex Period
\D Rat, F-344 16 h/day
5 days/week
104 weeks
Cat, Inbred, M 8 h/day
7 days/week
124 weeks
Particles
(mg/m3)
0.7
2.2
6.6
e.o6
12.0f
C XT
(mg-h/m3)
5,824
18,304
54,912
41,664
83,328
CO
(ppm)
—
—
32.0
20.2
33.3
NO2 SO2
(ppm) (ppm) Effects
— — Increases in RBC, Hb, Hct, and WBC, primarily banded
— — neutrophils; suggestion of an increase in prothrombin time;
— — increased heart/body weight and right ventricular/heart
ratios and decreased left ventricular contractility in
6.6-mg/m3 group
2.7 2.1 Increases in banded neutrophils; significant at 12 mo, but
4.4 5.0 not 24 mo
References
Brightwell et al.
(1986)
Pepelko and Peirano
(1983)
ANonirradiated diesel exhaust.
blrradiated diesel exhaust.
cLight-duty engine.
dHeavy-duty engine.
el to 61 weeks of exposure.
f62 to 124 weeks of exposure.
o
o
o
25
O
H
£>
I
o
90
O
i—i
H
W
-------
1 1989), guinea pigs and rats (Penney et al., 1981), and rats (Karagianes et al., 1981); lowered
2 in rats (Heinrich et al., 1982); and elevated in rats (Research Committee for HERP Studies,
3 1988; Brightwell et al., 1986). Mean corpuscular volume was significantly increased in
4 monkeys, 69 versus 64 (Lewis et al., 1989) and hamsters (Heinrich et al., 1982) and lowered
5 in rats (Research Committee for HERP Studies, 1988). The only other parameters of
6 erythrocyte status and related events were lowered mean corpuscular hemoglobin and mean
7 corpuscular hemoglobin concentration in rats (Research Committee for HERP Studies, 1988),
8 a 3 to 5% increase in carboxyhemoglobin saturation in rats (Karagianes et al., 1981), and a
9 suggestion of an increase in prothrombin time (Brightwell et al., 1986). The biological
10 significance of these findings regarding adverse health effects is deemed to be
11 inconsequential.
12 Three investigators (Pepelko and Peirano, 1983; Lewis et al., 1989; Brightwell et al.,
13 1986) reported an increase in the percentage of banded neutrophils in cats and rats. This
14 effect was not observed in monkeys (Lewis et al., 1989). The health implications of an
15 increase in abnormal maturation of circulating neutrophils are uncertain but do indicate a
16 toxic response of leukocytes following exposures to diesel exhaust. Leukocyte counts were
17 reported to be reduced in hamsters (Heinrich et al., 1982); increased in rats (Brightwell
18 et al., 1986); and unaffected in cats, rats, and monkeys (Pepelko and Peirano, 1983;
19 Research Committee for HERP Studies, 1988; Lewis et al., 1989). These inconsistent
20 findings indicate that the leukocyte counts are more indicative of the clinical status of the
21 laboratory animals than any direct effect of exposure to diesel exhaust.
22 An important consequence of particle retention in the lungs of exposed subjects can
23 be the development of pulmonary hypertension and cor pulmonale. Such pathology usually
24 arises from pulmonary fibrosis or emphysema obliterating the pulmonary vascular bed, or by
25 chronic anoxia. No significant changes in heart mass were found in guinea pigs or rats
26 exposed to diesel exhaust (Wiester et al., 1980; Penney et al, 1981; Lewis et al., 1989).
27 Rats exposed to diesel exhaust showed a greater increase in the medial wall thickness of
28 pulmonary arteries of differing diameters and right ventricular wall thickness; these
29 increases, however, did not achieve statistically significant levels (Vallyathan et al., 1986).
30 Brightwell et al. (1986) reported increased heart/body weight and right
December 1994 5.53 DRAFT-DO NOT QUOTE OR CITE
-------
1 ventricular/heart weight ratios and decreased left ventricular contractility in rats exposed to
2 6.6 mg/m3 paniculate matter for 16 h/day, 5 days/week for 104 weeks.
3
4 Serum Chemistry
5 A number of investigators have studied the effects of exposure to diesel exhaust on
6 serum biochemistry. Such studies are summarized in Table 5-11.
7 The biological significance of changes in serum chemistry in female but not male rats
8 exposed at 2 mg/m3 particulate matter for 7 h/day, 5 days/week for 104 weeks (Lewis et
9 al., 1989) are difficult to interpret. Not only were the effects noted in one sex (females)
10 only, but the serum enzymes, LDH, SCOT, and SGPT were elevated in the control group, a
11 circumstance contrary to denoting organ damage in the exposed female rats. The elevations
12 of liver-related serum enzymes in the control versus the exposed female rats appear to be a
13 random event among these aged subjects. The incidence of age-related disease, such as
14 mononuclear cell leukemia, can markedly affect such enzyme levels, seriously compromising
15 the usefulness of a comparison to historical controls. The serum sodium values of
16 144 versus 148 mmol/L in control and exposed rats, respectively, although statistically
17 different would have no biological import.
18 The increased serum enzyme activities, alkaline phosphatase, SGOT, SGPT, gamma-
19 glutamyl transpeptidase, and decreased cholinesterase activity suggest an impaired liver;
20 however, such an impairment was not established histopathologically (Heinrich et al., 1982;
21 Research Committee for HERP Studies, 1988; Brightwell et al., 1986). The increased urea
22 nitrogen, electrolyte levels, and gamma globulin concentration and reduction in total blood
23 proteins are indicative of an impaired kidney function. Again there was no histopathological
24 confirmation of impaired kidneys in these studies.
25 Clinical chemistry studies suggest an impairment of both liver and kidney functions in
26 rats and hamsters chronically exposed to high concentrations of diesel exhaust. The absence
27 of histopathological confirmation, the appearance of such effects near the end of the lifespan
28 of the laboratory animal and the failure to find such biochemical changes in cats exposed to a
29 higher dose, however, tend to discredit the probability of hepatic and renal hazards to
30 humans exposed at atmospheric levels of diesel exhaust.
31
December 1994 5.54 DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 5-11. EFFECTS OF CHRONIC EXPOSURES TO DIESEL EXHAUST
ft>
1
l-t
VO
^
Species/Sex
Rat, F-344,
M, F
Hamster,
Syrian, M, F
Rat, F-344/JcL
M, F
Rat, F-344;
Exposure
Period
7 h/day
5 days/week
104 weeks
7-8 h/day
5 days/week
75 weeks
16 h/day
6 days/week
130 weeks
0.
16 h/day
Hamster, Syrian 5 days/week
i
C/l
0
i>
rfl
i
o
o
2;
o
H
O
0
H
ro
O
o
H
M
Cat inbred, M
104 weeks
8 h/day
7 days/week
124 weeks
\jn amtuivi ^nJMViio i n. i v^r
Particles C x T CO NO2
(mg/m3) (mg-h/rn3) (ppm) (ppm)
2.0 7,280 11.5 1.5
0.23
0.36 fisa, MMD
3.9 10,238-11,700 18.5 1.2
0.1 /im, MMD
0.1 la 1,373 1.23 0.08
0.41a 5,117 2.12 0.26
1.084 13,478 3.96 3.96
2.31" 28,829 7.10 7.10
3.72b 46,426 12.9 3.00
19-0.28 /tm, MMD
0.7 5,824 - -
2.2 18,304 — —
6.6 54,912 32.0 —
6.0° 41,664 20.2 2.7
12.0"1 83,328 33.3 4.4
L,t\B\J
S02
(ppm)
0.8
3.1
0.38
1.06
2.42
4.70
4.57
—
—
—
2.1
5.0
KJ\l\JK.l A11JUVJAL.CJ
Effects
Decreased phosphate, LDH, SGOT, and SGPT;
increased sodium in females but not males
After 29 weeks, increases in SGOT, LDH, alkaline
phosphatase, gamma-glutamyl transf erase, and BUN
Lower cholinesterase activity in males in both the light
and heavy-duty series and elevated gamma globulin and
electrolyte levels in males and females in both series
Rats, 6.6 mg/m3, reduction in blood glucose, blood
proteins, triglycerides and cholesterol; increase in BUN,
alkaline phosphate alamine and aspartate amino-
transferases (SGPT and SGOT); hamsters, 6.6 mg/m3,
decrease in potassium, LDH, aspartate amino-
transferase; increase in albumin and gamma-glutamyl
transferase
BUN unaltered; SGOT and SGPT unaffected; LHD
increase after 1 year of exposure
References
Lewis et al. (1989)
Heinrich et al.
(1982)
Research Committee
for HERP Studies
(1988)
Brightwell et al.
(1986)
Pepelko and Periano
(1983)
'Light-duty engine.
bHeavy-duty engine.
cl to 61 weeks of exposure.
""62 to 124 weeks of exposure.
Key: LDH =
SGOT =
BUN =
SGPT =
Lactate dehydrogenase.
Serum glutamic-oxaloacetic transaminase,
Blood urea nitrogen,
Serum glutamic-pyruvic transaminase.
-------
1 Effects on Microsomal Enzymes
2 Several studies have examined the effects of diesel exhaust exposure on microsomal
3 enzymes associated with the metabolism and possible activation of xenobiotics, especially
4 polynuclear aromatic hydrocarbons (PAH's). These studies are summarized hi Table 5-12.
5 Lee et al. (1980) measured the activities of aryl hydrocarbon hydroxylase (AHH) and
6 epoxide hydrase (EH) in liver, lung, testis and prostate gland of adult male rats exposed to
7 6.32 mg/m3 paniculate matter 20 h/day for 42 days. Maximal significant AHH activities
8 (pmol/min/mg microsomal protein) occurred at different times during the exposure period,
9 and differences between controls and exposed, respectively, were as follows: prostate
10 0.29 versus 1.31, lung 3.67 versus 5.11, and liver 113.9 versus 164.0. There was no
11 difference in AHH activity in the testis between exposed and control rats. Epoxide hydrase
12 activity was not significantly different from control values for any of the organs tested.
13 Pepelko and Peirano (1983) found no statistical differences in liver microsomal
14 cytochrome P448-450 levels and liver microsomal AHH between control and diesel-exposed
15 mice either at 6 and 8 mo of exposure. Small differences were noted in the lung microsomal
16 AHH activities, but these were believed to be artifactual differences, due to increases in
17 nonmicrosomal lung protein present in the microsomal preparations. Exposures were for
18 8 h/day, 7 days/week to 6 mg/m3 particulate matter.
19 Rabovsky et al. (1984) investigated the effect of chronic exposure to diesel exhaust on
20 microsomal cytochrome P450-associated benzo[a]pyrene (B[a]P) hydroxylase and
21 7-ethoxycoumarin deethylase activities in rat lung and liver. Male rats were exposed for
22 7 h/day, 5 days/week for 104 weeks to 2 mg/m3 particulate matter. The exposure had no
23 effect on B[a]P hydroxylase or 7-ethoxycoumarin deethylase activities in lung or liver.
24 In related studies, Rabovsky et al. (1986) examined the effects of diesel exhaust on vitally
25 induced enzyme activity and interferon production in female mice. The mice were exposed
26 for 7 h/day, 5 days/week for 1 mo to diesel exhaust diluted to achieve a concentration of
27 2 mg/m3 particulate matter. Following the exposure, the mice were inoculated intranasally
28 with influenza virus. Changes in serum levels of interferon and liver microsomal activities
29 of 7-ethoxycoumarin, ethylmorphine demethylase and NADPH-dependent cytochrome
30 c reductase were measured. In the absence of viral inoculation, exposure to diesel exhaust
31 had no significant effects on the activity levels of the two liver microsomal monooxygenases,
December 1994 5-56 DRAFT-DO NOT QUOTE OR CITE
-------
fa
WO
sT|
o|
of
I
.s
I
a
•S
o
(
December 1994
73
u
f !ii
*1? >5>o
S 1 a f 1
ii
-a
s
d
tt.
8
P
I:
rao
ey
If
s^
l-s
Tf
t-
d-f
o\
d
* .
o\
«^
d
9
cn
d
«lr>
8 vi
11
1
u,
p.
I
5-57
DRAFT-DO NOT QUOTE OR CITE
-------
H
tfi
DIESEL EXHAt
Y ANIMALS
si
c« 9P
cd Oj
2 "
s "fe
oo o y
^ "o •» in
0^* fcn 5
g. x^, o g
o ~ m S «
i, no induction of cyt. P-45
ndentcyt. c reductase; afte
r microsomal oxidation of
ear of exposure to either 0,
microsomal metabolism of
Illif
lllll
i i
i i
i i
o> in
n oo
S
5
m
ci.
1
^^
•a oo
rt ^,
^ i
S 2
S* 'S
a. cu
u
a
1
cQ
(A
1
.S
S
4>
i
M
ff
3 IA
11
a J
1 •
•5 S
O 'f
(N
CO
(S
^
r-'
^
.^
~
o
vd
11^3
oo r-- 01 to
S
^
^«
8
1
8
1
I
lg
8 S.
||
t «
u u
as ^
a S
December 1994
5-58 DRAFT-DO NOT QUOTE OR CITE
-------
1 and NADPH-dependent cytochrome c reductase. Exposure to diesel exhaust produced
2 smaller increases in ethylmorphine demethylase activity on Days 2 to 4 postvirus infection
3 and also abolished the Day 4 postinfection increase in NADPH-dependent cytochrome
4 c reductase when compared with nonexposed mice. These data suggested to the authors that
5 the relationship that exists between metabolic detoxification and resistance to infection in
6 unexposed mice was altered during a short-term exposure to diesel exhaust.
7 Chen and Vostal (1981) measured the activity of AHH and the content of cytochrome
8 P450 in the lungs and livers of rats exposed by inhalation or intraperitoneal (ip) injection of a
9 dichloromethane extract of diesel particles. In the inhalation exposures, the exhaust was
10 diluted to achieve concentrations of 0.75 or 1.5 mg/m3 paniculate matter, and the exposure
11 regimen was 20 h/day, 5.5 days/week for up to 9 mo. The concentration of total
12 hydrocarbons and particle-phase hydrocarbons was not reported. Parenteral administration
13 involved repeated ip injections at several dose levels for 4 days. Inhalation exposure had no
14 significant effect on liver microsomal AHH activity; however, lung AHH activity was
15 slightly reduced after 6 mo exposure to 1.5 mg/m3. An ip dose of diesel particulate extract,
16 estimated to be equivalent to the inhalation exposure, had no effect on AHH activity in liver
17 or lungs. No changes were observed in cytochrome P450 contents in lungs or liver
18 following inhalation exposure or ip treatment. Direct intratracheal administration of a
19 dichloromethane diesel particulate extract required doses greater than 6 mg/kg body weight
20 before the activity of induced AHH in the lung was barely doubled; liver AHH activity
21 remained unchanged (Chen, 1986).
22 In related studies, Navarro et al. (1981) evaluated the effect of exposure to diesel
23 exhaust on rat hepatic and pulmonary microsomal enzyme activities. The same exposure
24 regimen was employed (20 h/day, 5.5 days/week, for up to 1 year) and the exhaust was
25 diluted to achieve concentrations of 0.25 and 1.5 mg/m3 particulate matter (a few studies
26 were also conducted at 0.75 mg/m3). After 8 weeks of exposure, there was no evidence for
27 the induction of cytochrome P450, cytochrome P448, or NADPH-dependent cytochrome.
28 c reductase in rat liver microsomes. One year of exposure had little, if any, effect on the
29 hepatic metabolism of B[a]P. However, 1 year of exposure to 0.25 and 1.5 mg/m3
30 significantly impaired the ability of lung microsomes to metabolize B[a]P (0.15 and
December 1994 5.59 DRAFT-DO NOT QUOTE OR CITE
-------
1 0.02 nmole/30 min/mg protein, respectively versus 0.32 nmole/30min/mg protein for the
2 controls.)
3 There are conflicting results regarding the induction of microsomal AHH activities in
4 the lungs and liver of rodents exposed to diesel exhaust. One study reported induced AHH
5 activity in the lungs, liver, and prostate of rats exposed to diesel exhaust containing
6 6.32 mg/m3 paniculate matter for 20 h/day for 42 days; however, no induction of AHH was
7 observed in the lungs of rats and mice exposed to 6 mg/m3 particulate matter for 8 h/day,
8 7 days/week for up to 8 mo or to 0.25 to 2 mg/m3 for periods up to 2 years. Exposure to
9 diesel exhaust has not been shown to produce adverse effects on microsomal cytochrome
10 P450 in the lungs or liver of rats or mice. The weight of evidence suggests that the absence
11 of enzyme induction in the rodent lung exposed to diesel exhaust is caused either by the
12 unavailability of the adsorbed hydrocarbons or by their presence in insufficient quantities for
13 enzyme induction.
14
15 Effects on Behavior and Neurophysiology
16 Studies on the effects of exposure to diesel exhaust on the behavior and neuro-
17 physiology of laboratory animals are summarized in Table 5-13. Laurie et al. (1978) and
18 Laurie et al. (1980) examined behavioral alterations in adult and neonatal rats exposed to
19 diesel exhaust. Exposure for 20 h/day, 7 days/week, for 6 weeks to exhaust containing
20 6 mg/m3 particulate matter produced a significant reduction in adult spontaneous locomotor
21 activity (SLA) and in neonatal pivoting (Laurie et al., 1978). In a follow-up study, Laurie
22 et al. (1980) found that shorter exposure (8 h/day) to 6 mg/m3 particulate matter also
23 resulted in a reduction of SLA in adult rats. Laurie et al. (1980) conducted additional
24 behavioral tests on adult rats exposed during their neonatal period. For two of three
25 exposure situations (20 h/day for 17 days postparturition, or 8 h/day for the first 28 or
26 42 days postparturition), significantly lower SLA was observed in the majority of the tests
27 conducted on the adults after Week 5 of measurement. When compared with control rats,
28 adult 15-mo-old rats that had been exposed as neonates (20 h/day for 17 days) also exhibited
29 a significantly slower rate of acquisition of a bar pressing task to obtain food. The
30 investigators noted that the evidence was insufficient to determine whether the differences
December 1994 5-60 DRAFT-DO NOT QUOTE OR CITE
-------
o
8
fl>
1
Species/Sex
*O Rat, Sprague-
Dawley, M
Rat, Sprague
Dawley, F
Rat, Sprague-
Dawley, F
TABLE 5-13. EFFECTS OF CHRONIC EXPOSURES TO DIESEL EXHAUST
ON BEHAVIOR AND NEUROPHYSIOLOGY
Exposure
Period
8 h/day
7 days/week
1-4 weeks
20 h/day
7 days week
6 weeks
8 or 20 h/day
7 days/week
3, 4, 6, or
16 weeks
Particles C x T CO NO2 SO2
(mg/m3) (mg-h/m3) (ppm) (ppm) (ppm) Effects
6 336-1,344 19 2.5 1.8 Somatosensory and visual evoked potentials revealed longer
pulse latencies in pups exposed neonatally
6 5,040 19 2.5 1.8 Reduction in adult SLA and in neonatal pivoting
6 1,008-13,440 19 2.5 1.8 Reduction in SLA in adults; neonatal exposures for 20 or
8 h/day caused reductions in SLA. Neonatal exposures for
20 h/day for 17 days resulted in a slower rate of a bar-pressing
task to obtain food
References
Laurie and Boyes
(1980, 1981)
Laurie et al. (1978)
Laurie et al. (1980)
SLA = Spontaneous locomotor activity.
ON
-------
1 were the result of a learning deficit or to some other cause (e.g., motivational or arousal
2 differences).
3 These data are difficult to interpret in terms of health hazards to humans under
4 ambient environmental conditions because of the high concentration of diesel exhaust to
5 which the laboratory rats were exposed. Additionally, there are no further concentration-
6 response studies to assess at what exposure levels these observed results persist or abate.
7 A permanent alteration in both learning ability and activity resulting from exposures early in
8 life is a health hazard whose significance to humans should be pursued further.
9 Neurophysiological effects from exposure to diesel exhaust were investigated in rats by
10 Laurie and Boyes (1980, 1981). Rats were exposed to diluted diesel exhaust containing
11 6 mg/m3 paniculate matter for 8 h/day, 7 days/week from birth up until 28 days of age.
12 Somatosensory evoked potential, as elicited by a 1-mA electrical pulse to the tibial nerve in
13 the left hind limb, and visual evoked potential, as elicited by a flash of light, were the end
14 points tested. An increased pulse latency was reported for the rats exposed to diesel exhaust,
15 and this was thought to be caused by a reduction in the degree of nerve myelinization. There
16 was no neuropathological examination, however, to confirm this supposition.
17 Based on the data presented, it is not possible to specify the particular neurological
18 impairment(s) induced by the exposure to diesel exhaust. Again, these results occurred
19 following exposure to a high level of diesel exhaust and no additional concentration-response
20 studies were performed.
21
22 Effects on Reproduction and Development
23 Studies of the effects of exposure to diesel exhaust on reproduction and development
24 are summarized in Table 5-14. Twenty rats were exposed 8 h/day on Days 6 through 15 of
25 gestation to diluted diesel exhaust containing 6 mg/m3 paniculate matter (Werchowski et al.,
26 1980a,b; Pepelko and Peirano, 1983). There were no signs of maternal toxicity or decreased
27 fertility. No skeletal or visceral teratogenic effects were observed in 20-day-old fetuses
28 (Werchowski et al., 1980a). In a second study, 42 rabbits were exposed to 6 mg/m3
29 paniculate matter for 8 h/day, on Gestation Days 6 through 18. No adverse effects on body
30 weight gain or fertility were seen in the does exposed to diesel exhaust. No visceral or
December 1994 5-62 DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 5-14. EFFECTS OF CHRONIC EXPOSURES TO DIESEL EXHAUST
a
1
^0
L/l
W
O
^
*ft
H
0
O
O
H
0
G
0
O
O
H
W
Species/Sex
Mouse,
[C57B,]/
6XC3HJF,, M
Rat, Sprague-
Dawley, F
Rabbit, New
Zealand Albino,
F
Monkey,
Cynomolgus, M
Mouse,
A/Strong, M
Mouse, CD-I,
M, F
Exposure
Period
5 days
8 h/day
7 days/week
1 .7 weeks
8 h/day
7 days/week
1 .9 weeks
7 h/day
5 days/week
104 weeks
8 h/day
7 days/week
31 or
38 weeks
8 h/day
7 days/week
6 to 28 weeks
- -— — -«• — •*-• v — — .-«--. i * — -. , ^- .•_••.•_• T -M_«-a^* •v-' .M. ' • • • J • 1 M. -MJ. ^ JLJ^M-M^I^M^JTM. M. **JM*t M. 4HU ^ .11 Y-Ln_l_A_7
Particles C x T CO NO2 SO2
(mg/m3) (mg-h/rn3) (ppm) (ppm) (ppm) Effects
50, 100, or — — — — Dose related increase in sperm abnormalities; decrease
200 mg/kg in sperm number at highest dose; testicular wts
in corn oil; ip unaffected
injection
6 571 20 2.7 2.1 No signs of maternal toxicity or decreased fertility; no
skeletal or visceral teratogenic effects in 20-day-old
fetuses
6 638 20 2.7 2. 1 No adverse effects on maternal weight gain or
fertility; no skeletal or visceral teratogenic effects in
the fetuses
2 7,280 11.5 1.5 0.8 No effects on sperm motility, velocity, density,
morphology, or incidence of abnormalities
6 10,416-12,768 20 2.7 2.1 No effect on sperm morphology; high rate of
spontaneous sperm abnormalities may have masked
small effects
12 4,032-18,816 33 4.4 5.0 Overall fertility and survival rates were unaffected in
the three generation reproductive study; only
consistent change noted, an increase in lung weights,
was diagnosed as anthracosis
References
Quinto and
DeMarinis (1984)
- Werchowski et al.
(1980a)
Pepelko and Peirano
(1983)
Werchowski, et al.
(1980a)
Pepelko and Peirano
(1983)
Lewis et al. (1989)
Periera et al.
(1981b)
Pepelko and Peirano
(1983)
-------
1 skeletal developmental abnormalities were observed in the fetuses (Werchowski et al.,
2 1980b).
3 Pepelko and Peirano (1983) evaluated the potential for diesel exhaust to affect
4 reproductive performance hi mice exposed from 100 days prior to exposure throughout
5 maturity of the F2 generation. The mice were exposed for 8 h/day, 7 days/week to
6 12 mg/m3 paniculate matter. In general, treatment-related effects were minimal. Some
7 differences in organ and body weights were noted, but overall fertility and survival rates
8 were not altered by exposure to diesel exhaust. The only consistent change, an increase in
9 lung weights, was accompanied by a gross pathological diagnosis of anthracosis. These data
10 denoted that exposure to diesel exhaust at a concentration of 12 mg/m3 did not affect
11 reproduction. See Section 5.3, Mauderly et al. (1987a), which reports a lack of effects of
12 exposure to diesel exhaust on the developing rat lung.
13 Several studies have evaluated the effect of exposure to diesel exhaust on sperm.
14 Lewis et al. (1989) found no adverse sperm effects (sperm motility, velocity, densities,
15 morphology, or incidence of abnormal sperm) in monkeys exposed for 7 h/day, 5 days/week,
16 for 104 weeks to 2 mg/m3 paniculate matter. In another study in which A/Strong mice were
17 exposed to diesel exhaust containing 6 mg/m3 paniculate matter for 8 h/day for 31 or
18 38 weeks, no significant differences were observed in sperm morphology between exposed
19 and control mice (Pereira et al., 1981b). It was noted, however, that there was a high rate
20 of spontaneous sperm abnormalities in this strain of mice, and this may have masked any
21 small positive effect. Quinto and DeMarinis (1984) reported a statistically significant and
22 dose-related increase in sperm abnormalities in mice injected intraperitoneally for 5 days with
23 50, 100, or 200 mg/kg of diesel particulate matter suspended in corn oil. A significant
24 decrease in sperm number was seen at the highest dose, but testicular weight was unaffected
25 by the treatment.
26 No teratogenic, embryotoxic, fetotoxic, or female reproductive effects were observed
27 in mice, rats, or rabbits at exposure levels up to 12 mg/m3 particulate matter. Effects on
28 sperm morphology and number were reported in hamsters and mice exposed to high doses of
29 diesel particles; however, no adverse effects were observed in sperm obtained from monkeys
30 exposed at 2 mg/m3 for 7 h/day, 5 days/week for 104 weeks. Concentrations of 12 mg/m3
31 particulate matter did not affect male rat reproductive fertility in the F0 and Ft generation
December 1994 5-64 DRAFT-DO NOT QUOTE OR CITE
-------
1 breeders. Thus, exposure to diesel exhaust would not appear to be a reproductive or
2 developmental hazard.
3
4
5 5.2 COMPARISON OF HEALTH EFFECTS OF FILTERED AND
6 UNFILTERED DIESEL EXHAUST
7 In four chronic toxicity studies of diesel exhaust, the experimental protocol included
8 exposing test animals to exhaust containing no particles. Comparisons were then made
9 between the effects caused by whole, unfiltered exhaust, and those caused by the gaseous
10 components of the exhaust. Concentrations of components of the exposure atmospheres in
11 these four studies are given in Table 5-15.
12 Heinrich et al. (1982) compared the toxic effects of whole and filtered diesel exhaust
13 on hamsters and rats. Exposures were for 7 to 8 h/day and 5 days/week. Rats exposed for
14 24 mo to either whole or filtered exhaust exhibited no significant changes in respiratory
15 frequency, respiratory minute volume, compliance or resistance as measured by a whole-
16 body plethysmography, or in heart rate. In the hamsters, histological changes (adenomatous
17 proliferations) were seen in the lungs of animals exposed to either whole or filtered exhaust;
18 however, in all groups exposed to the whole exhaust, the number of hamsters exhibiting such
19 lesions was significantly higher than for the corresponding groups exposed to filtered exhaust
20 or clean air. Severity of the lesions was, however, not reported.
21 In a second study, Heinrich et al. (1986a, see also Stober, 1986) compared the toxic
22 effects of whole and filtered diesel exhaust on hamsters, rats, and mice. The test animals
23 (96 per test group) were exposed for 19 h/day, 5 days/week for 120 (hamsters and mice) or
24 140 (rats) weeks. Body weights of hamsters were unaffected by either exposure. Body
25 weights of rats and mice were reduced by the whole exhaust, but not by the filtered exhaust.
26 Exposure-related higher mortality rates occurred in mice after 2 years of exposure to whole
27 exhaust. After 1 year of exposure to the whole exhaust, hamsters exhibited increased lung
28 weights, a significant increase in airway resistance, and a nonsignificant reduction in lung
29 compliance. For the same time period, rats exhibited increased lung weights, a significant
30 decrease in dynamic lung compliance, and a significant increase in airway resistance. Test
31 animals exposed to filtered exhaust did not exhibit such effects. Histopathological
December 1994 5.55 DRAFT-DO NOT QUOTE OR CITE
-------
December 1994
TABLE 5-15. COMPOSITION OF EXPOSURE ATMOSPHERES IN STUDIES
COMPARING UNFILTERED AND FILTERED DIESEL EXHAUST2
Species/Sex
Rat Wistar, F;
Hamster, Syrian
Exposure6
Period
7 h/day UF
5 days/week F
104 weeks C
Particles
(mg/m3)
3.9
C x T
(mg-h/m3)
14,196
CO
(ppm)
18.5
18.0
NO2
(ppm)
1.2
1.0
SO2
(ppm)
3.1
2.8
Effects
No effect on pulmonary function or heart rate in
rats; increases in pulmonary adematous
proliferations in hamsters, UF significantly higher
than F or C
References
Heinrich et al. (1982)
•n
H
6
O
z
s
o
Rat, F-344, F
Rat, F-344, M, F;
Hamster, Syrian, M,
F
Rat, Wistar, F;
Hamster, Syrian, F;
Mouse NMRI, F
8 h/day
7 days/week
104 weeks
16 h/day
5 days/week
104 weeks
19 h/day
5 days/week
120 to
140 weeks
UF 4.9 28,538 7.0 1.8 13.1 Body weight decrease after 6 mo in UF, 18 mo in
F0 — — — — F; lung/body rate weight rate higher in both
^ — — — — groups at 24 mo; at 2 years, fibrosis and epithelial
hyperplasia in lungs of UF; nominal lung and
spleen histologic changes
UF 0.7 5,824 — — — UF; elevated red and white cell counts,
™ 2.2 18,304 — — — hematrocrit and hemoglobin; increased heart/body
C 6.6 54,912 32.0 — — weight and right ventricular/bean weight ratios;
— 32.0 — — lower left ventricular contractility; changes in
— 1.0 — — blood chemistry; obstructive and restrictive lung
disease; F: no effects
UF 4.24 48,336 12.5 1.5 3.1 UF: decreased body wt in rats and mice but not
™ — 56,392 11.1 1.2 1.02 hamsters; increased mortality, mice only;
— 0.16 — — decreased lung compliance and increased airway
resistance, rats and hamsters; species differences
in lung lavage enzymes and cell counts and lung
histopathology and collagen content, most
pronounced in rats; F: no effect on glucose-6-
phosphate dehydrogenase, total protein and lung
collagen
Iwai et al. (1986)
Brightwell et al. (1986)
Heinrich et al. (1986a)
i
o
HH
H
M
"Mean values.
"TJF = Unflltered whole exhaust,
F = Filtered exhaust,
C = Control.
cReported to have the same component concentrations as the unflltered, except particles that were present in undetectable amounts.
Concentrations reported for high concentration level only.
-------
1 examination indicated that different levels of response occurred in the three species.
2 In hamsters, filtered exhaust caused no significant histopathological effects in the lung; whole
3 exhaust caused thickened alveolar septa, bronchiolo-alveolar hyperplasia, and emphysematous
4 lesions. In mice, whole exhaust, but not filtered exhaust, caused multifocal bronchiolo-
5 alveolar hyperplasia, multifocal alveolar lipoproteinosis, and multifocal interstitial fibrosis.
6 In rats, there were no significant morphological changes in the lungs following exposure to
7 filtered exhaust. In rats exposed to whole exhaust, there were severe inflammatory changes
8 in the lungs, thickened alveolar septa, foci of macrophages, crystals of cholesterol, and
9 hyperplastic and metaplastic lesions. Biochemical studies of lung lavage fluids of hamsters
10 and mice indicated that exposure to filtered exhaust caused fewer changes than did exposure
11 to whole exhaust. The latter produced significant increases in lactate dehydrogenase, alkaline
12 phosphatase, glucose-6-phosphate dehydrogenase (G6P-DH), total protein, protease (pH 5.1),
13 and collagen. The filtered exhaust had a slight, but nonsignificant, effect on G6P-DH, total
14 protein, and collagen. Similarly, cytological studies showed that, while the filtered exhaust
15 had no effect on differential cell counts, the whole exhaust resulted in an increase in
16 leukocytes (161 ± 43.3//iL versus 55.7 ± 12.8/^iL in the controls), a decrease in
17 macrophages (30.0 ± 12.5 versus 51.3 ± 12.5//iL in the controls), and an increase in
18 granulocytes (125 ± 39.7 versus 1.23 ± 1.14//iL in the controls). All values presented for
19 this study are the mean with its standard deviation. The differences were significant for each
20 cell type. There was also a small increase in lymphocytes (5.81 ±4.72 versus
21 3.01 ± 1.23//
-------
1 exhaust, after 2 years there were only minimal histologic changes in the lungs, with slight
2 hyperplasia and stratification of bronchiolar epithelium and infiltration of atypical
3 lymphocytic cells in the spleen.
4 Brightwell et al. (1986) evaluated the toxic effects of whole and filtered diesel exhaust
5 on rats and hamsters. Three exhaust dilutions were tested, producing concentrations of 0.7,
6 2.2, and 6.6 mg/m3 paniculate matter. The test animals (144 rats and 312 hamsters per
7 exposure group) were exposed for five 16-h periods per week for 2 years. The four
8 exposure types were gasoline, gasoline catalyst, diesel, and filtered diesel. The results
9 presented were limited to statistically significant differences between exhaust-exposed and
10 control animals. The inference from the discussion section of the paper was that there was a
11 minimum of toxicity in the animals exposed to filtered diesel exhaust: "It is clear from the
12 results presented that statistically significant differences between exhaust-exposed and control
13 animals are almost exclusively limited to animals exposed to either gasoline or unfiltered
14 diesel exhaust." Additional results are described in Section 5.4.
15 A comparison of the toxic responses in laboratory animals exposed to whole exhaust
16 or filtered exhaust containing no particles demonstrates across studies that when the exhaust
17 is sufficiently diluted to limit the concentrations of gaseous irritants (NO2 and S02), irritant
18 vapors (aldehydes), CO, or other systemic toxicants, the diesel particles are the prime
19 etiologic agents of noncancer health effects, although additivity or synergism with the gases
20 cannot be ruled out. These toxic responses are both functional and pathological and
21 represent a cascading sequelae of lung pathology based on concentration and species. The
22 diesel particles plus gas exposures produced biochemical and cytological changes in the lung
23 that are much more prominent than those evoked by the gas phase alone. Such marked
24 differences between whole and filtered diesel exhaust are also evident from general
25 toxicological indices, such as decreases in body weight and increases in lung weights,
26 pulmonary function measurements, and pulmonary histopathology (e.g., proliferative changes
27 in Type II cells and respiratory bronchiolar epithelium, fibrosis). Hamsters, under equivalent
28 exposure regimens, have lower levels of retained particles in their lungs than do rats and
29 mice and, consequently, less pulmonary function impairment and pulmonary pathology.
30 These differences may result from lower paniculate inspiration and deposition during
December 1994 5-68 DRAFT-DO NOT QUOTE OR CITE
-------
1 exposure, greater paniculate clearance, or lung tissue less susceptible to the cytotoxicity of
2 deposited particles.
3
4
5 5.3 INTERACTIVE EFFECTS OF DIESEL EXHAUST
6 A multitude of factors may influence the susceptibility to exposure to diesel exhaust as
7 well as the resulting response. Some of these have already been discussed in detail (e.g., the
8 composition of diesel exhaust and concentration-response data); others will be addressed in
9 this section (e.g., the interaction of diesel exhaust with factors particular to the exposed
10 individual and the interaction of diesel exhaust components with other airborne
11 contaminants).
12 Mauderly et al. (1990a) compared the susceptibility of normal rats and rats with
13 preexisting laboratory-induced pulmonary emphysema exposed for 7 h/day, 5 days/week for
14 24 mo to diesel exhaust containing 3.5 mg/m3 particulate matter or to clean air (controls).
15 Emphysema was induced in one-half of the rats by intratracheal instillation of elastase
16 6 weeks before exhaust exposure. Measurements included lung burdens of diesel particles,
17 respiratory function, bronchoalveolar lavage, clearance of radiolabeled particles, pulmonary
18 immune responses, lung collagen, excised lung weight and volume, histopathology, and mean
19 linear intercept of terminal air spaces. None of the data for the 63 parameters measured
20 suggest that rats with emphysematous lungs were more susceptible than rats with normal
21 lungs to the effects of diesel exhaust exposure. In fact, each of the 14 emphysema-exhaust
22 interactions detected by statistical analysis of variance indicated that emphysema acted to
23 reduce the effects of diesel exhaust exposure. Diesel particles accumulated much less rapidly
24 in the lungs of emphysematous rats than in those of normal rats. The mean lung burdens of
25 diesel particles in the emphysematous rats were 39, 36, and 37% of the lung burdens of
26 normal rats at 12, 18, and 24 mo, respectively. No significant interactions were observed
27 among lung morphometric parameters. Emphysema prevented the exhaust-induced increase
28 for three respiratory indices of expiratory flow rate at low lung volumes, reduced the
29 exhaust-induced increase in nine lavage fluid indicators of lung damage, prevented the
30 expression of an exhaust-induced increase in lung collagen, and reduced the exhaust-induced
31 delay in particle clearance.
December 1994 5.69 DRAFT-DO NOT QUOTE OR CITE
-------
1 Mauderly et al. (1987a) evaluated the relative susceptibility of developing and adult
2 lungs to damage by exposure to diesel exhaust. Rats (48 per test group) were exposed to
3 diesel exhaust containing 3.5 mg/m3 particulate matter and about 0.8 ppm NO2. Exposures
4 were for 7 h/day, 5 days/week through gestation to the age of 6 mo, or from the age of 6 to
5 12 mo. Comparative studies were conducted on respiratory function, immune response, lung
6 clearance, airway fluid enzymes, protein and cytology, lung tissue collagen, and proteinases
7 in both age groups. After the 6-mo exposure, adult rats, compared with controls, exhibited:
8 (1) more focal aggregates of particle-containing macrophages in the alveolar ducts near the
9 terminal bronchioles, (2) a sixfold increase in the neutrophils (as a percentage of total
10 leukocytes) in the airway fluids, (3) a significantly higher number of total lymphoid cells in
11 the pulmonary lymph nodes, (4) delayed clearance of diesel particles and radiolabeled
12 particles (t1/2 = 90 days versus 47 days for controls), and (5) increased lung weights. These
13 effects were not seen in the neonatal rats. On a weight for weight (milligrams of particulate
14 matter per gram of lung) basis, diesel particle accumulation in the lungs was similar in young
15 and adult rats immediately after the exposure. During the 6-mo postexposure period, diesel
16 particulate clearance was much more rapid in the neonatal rats, approximately 2.5-fold.
17 During postexposure, diesel particle-laden macrophages became aggregated in the neonatal
18 rats, but these aggregations were located primarily in a subpleural position. The authors
19 concluded that exposure to diesel exhaust, using pulmonary function, structural (qualitative or
20 quantitive) biochemistry as the indices, did not affect the developing rat lung more severely
21 than the adult rat lung.
22 As a result of the increasing trend of using diesel powered equipment in coal mining
23 operations and the concern for adverse health effects hi coal miners exposed to both coal dust
24 or coal mine dust and diesel exhaust, Lewis et al. (1989) and Karagianes et al. (1981)
25 investigated the interaction of coal dust and diesel exhaust. Lewis et al. (1989) exposed rats,
26 mice, and cynomolgus monkeys to (1) filtered ambient air, (2) 2 mg/m3 diesel particulate
27 matter, (3) 2 mg/m3 inhalable coal dust, and (4) 1 mg/m3 of both items 2 and 3. Gaseous
28 and vapor concentrations were identical in both diesel exhaust exposures. Exposures were
29 for 7 h/day, 5 days/week for up to 24 mo. Synergistic effects between diesel exhaust and
30 coal dust were not demonstrated; additive toxic effects were the predominant effects noted.
December 1994 5-70 DRAFT-DO NOT QUOTE OR CITE
-------
1 Karagianes et al. (1981) exposed rats (24 per group) to diesel exhaust containing
2 8.3 mg/m3 of diesel exhaust alone or in combination with about 6 mg/m3 of coal dust.
3 No synergistic effects were found between diesel exhaust and coal dust; additive effects in
4 terms of visual dust burdens in necropsied lungs were related to dose (i.e., length of
5 exposure and airborne paniculate concentrations).
6 Additional studies of interactions between other airborne substances and components of
7 diesel exhaust have not been reported. However, general toxicological data from other
8 inhalation studies may be applied in assessing the potential health hazards posed by exposure
9 to diesel exhaust. Many of the individual components of diesel exhaust, notably irritants and
10 chemical asphyxiants, at sufficiently high concentrations can elicit acute adverse health
11 effects either individually, additively, or synergistically.
12 The health effects of airborne contaminants from sources other than diesel engines
13 may be altered in the presence of diesel particles by their adsorption onto the diesel particles.
14 When adsorbed onto diesel particles, the gases and vapors can be transported and deposited
15 deeper into the lungs, and because they are more concentrated on the particle surface, the
16 resultant cytotoxic effects or physiological responses may be enhanced. Nitrogen dioxide
17 adsorbed onto carbon particles caused pulmonary parenchymal lesions in mice, whereas NO2
18 alone produced edema and inflammation but no lesions (Boren, 1964). Collagen synthesis in
19 lung tissue was higher in animals exposed to NO2 and ammonium sulfate than in those
20 exposed to either agent alone (Last et al., 1983). Exposure to formaldehyde and acrolein
21 adsorbed onto carbon particles (1 to 4 /xm) resulted in the recruitment of polymorphonuclear
22 leukocytes to tracheal and intrapulmonary epithelial tissues but not when the aldehydes were
23 tested alone (Kilburn and McKenzie, 1978).
24 There is no direct evidence that diesel exhaust interacts with other substances in an
25 exposure environment or the physiological status of the exposed subject other than impaired
26 resistance to respiratory tract infections. Although there is experimental evidence that gases
27 and vapors can be adsorbed onto carbonaceous particles, enhancing the toxicity of these
28 particles when deposited in the lung, there is no evidence for an increased health risk from
29 such interactions with diesel particles under ambient urban atmospheric conditions.
30 Likewise, there is no experimental evidence in laboratory animals that the youth or
December 1994 5.71 DRAFT-DO NOT QUOTE OR CITE
-------
1 preexisting emphysema of an exposed individual enhances the risk of exposure to diesel
2 exhaust.
3
4
5 5.4 COMPARISON OF THE EFFECTS OF DIESEL EXHAUST AND
6 GASOLINE EXHAUST
7 Light-duty gasoline and diesel engines differ considerably in the composition of their
8 respective exhausts (see Chapter 2). Diesel engine exhaust generally has higher levels of
9 paniculate matter, whereas gasoline engine exhaust has higher levels of carbon monoxide
10 (Table 5-16). This difference in CO concentrations between gasoline and diesel engines
11 affects the way that animal inhalation studies are conducted. To prevent symptoms of acute
12 CO intoxication, gasoline engine exhaust is often more diluted in the high concentration
13 exposure group compared with studies of diesel engine exhaust. Consequentially, animals
14 exposed to gasoline exhaust are exposed to much lower concentrations of exhaust
15 components than are those exposed to diesel exhaust. A greater proportion of the exhaust
16 hydrocarbons in diesel exhaust consists of longer-chain hydrocarbons and of high molecular
17 weight organics and is associated with the particles (Carey and Cohen, 1980). Studies
18 undertaken to evaluate the relative health hazards posed by gasoline or diesel engine exhaust
19 are very limited.
20 Brightwell et al. (1986) compared the toxic effects of filtered and unfiltered diesel
21 exhaust, gasoline exhaust, and gasoline exhaust from an engine equipped with a catalytic
22 converter. The exhausts used in this study were generated by 1.6-L gasoline and 1.5-L
23 diesel engines. The type of fuel used was not stated. They were run according to the US-72
24 (FTP) driving cycle on computer-controlled test benches. The exhaust from each engine was
25 diluted by a constant 800 m3 of conditioned air per hour to give the highest exposure
26 concentration. Further dilutions of 1 in 3 or 1 in 9 were used for the medium- and low-
27 exposure concentrations.
28 Rats and hamsters were exposed for five 16-h periods per week for 2 years. Three
29 diesel paniculate concentrations were studied: 0.7, 2.2, and 6.6 mg/m3. From a toxicologic
30 perspective, the concentrations of CO and NOX in the gasoline exhaust atmosphere were
31 sufficiently high, 224 ppm and 49 ppm, respectively, to impart adverse health effects by
December 1994 5-72 DRAFT-DO NOT QUOTE OR CITE
-------
December 1994
TABLE 5-16.
Vehicle Type
Light-duty diesel
Heavy-duty diesel
Light-duty gasoline
Heavy-duty gasoline
EMISSION RATES
Particles
(g/km)
0.1-0.4
0.5-4.0
0.001-0.004
0.004-0.2
FOR DIESELa AND
CO
(g/km)
0.5-3.0
5-50
1-3
10-200
GASOLINE ENGINES
NO
(g/km)
0.5-2.0
3-20
0.2-1.0
1-11
Vapor Phase
Hydrocarbons
(g/km)
0.05-0.8
0.9-6.0
0.08-0.5
2-20
"Federal test procedures.
Source: Cuddihy et al. (1981), as adapted by Cuddihy et al. (1984).
H
6
o
z
3
O
d
9
W
-------
1 themselves. Body weights of rats were significantly lower in the highconcentration gasoline
2 group and both medium- and high-concentration diesel exhaust groups compared with
3 controls. The diesel-exposed rats exhibited alterations in respiratory physiology indicative of
4 concentration-related obstructive and restrictive airway disease (specific data not given).
5 These changes were not seen in rats exposed to gasoline exhaust nor in hamsters exposed to
6 filtered diesel or gasoline exhausts. The major biologically significant changes in blood
7 chemistry in the high concentration group after 18 or 24 mo were a decrease in glucose,
8 cholesterol and triglycerides and an increase in BUN for diesel exhaust compared with a
9 decrease in cholesterol and increase in lactate dehydrogenase (LDH) and a-hydroxybutyric
10 dehydrogenase (HBDH) for gasoline exhaust. The major significant changes in hamster
11 blood chemistry after 16 mo of exposure to the high level of exhaust were a decrease in
12 potassium, LDH, HBDH, and asparate aminotransferase and an increase in albumin and
13 gamma-glutamyl transpeptidase (GTP) for diesel exhaust compared with a decrease in
14 cholesterol and increase in glucose, albumin, sodium, GTP, and cholinesterase for gasoline-
15 catalyst exhaust.
16 The major significant differences in hematologic parameters between controls and
17 high-concentration groups were in gasoline- and diesel exhaust-exposed rats and in gasoline-,
18 gasoline catalyst-, and diesel exhaust-exposed hamsters. The major changes seen in both
19 gasoline- and diesel exhaust-exposed rats were increases in RBC count, hemoglobin, and
20 hematocrit. There was also an increase in white cell count, primarily attributable to
21 segmented neutrophils, in diesel exhaust-exposed rats and a suggestion of an increase in
22 prothrombin time with both exhaust types. The significant changes observed in hamsters
23 were in the gasoline emission-exposed groups with increases in RBC count, hemoglobin, and
24 hematocrit.
25 Cardiovascular function measurements on male rats showed significant differences
26 between controls and both gasoline- and diesel exhaust-exposed animals. Exposed rats
27 showed a significant increase in heart to body weight and right ventricular/heart weight ratio
28 (16.14 ± 0.58% in diesel versus 13.5 ± 1.37% in controls). In diesel exhaust-exposed rats
29 the left ventricular dP/dt maximum values (measured as an index of left ventricular
30 contractility) were significantly lower, 2,340 ± 595 mm Hg/s and 3,040 ± 388 mm Hg/s
31 (mean ± standard deviation) in diesel exhaust-exposed and control rats, respectively.
December 1994 5.74 DRAFT-DO NOT QUOTE OR CITE
-------
1 At necropsy a significant increase in lung weight was seen in both hamsters and rats. This
2 finding was more marked in rats where the increase was progressive with both duration of
3 exposure and concentration level.
4 The interpretation of these comparative differences in respiratory function, blood
5 chemistry and hematologic indices between diesel and gasoline exhaust exposure is
6 questionable. In most instances, only statistically significant changes were reported; the
7 absolute values of each parameter were not presented. The appropriateness of the t-test
8 comparing an exposed group with its relevant control for such a large number of parameters
9 is disputable. The responses observed were often limited to the high-concentration groups.
10 Certain toxicologic responses noted appear to be secondary effects (e.g., [1] decreased body
11 weight, glucose, cholesterol, and triglyceride and increased BUN during high-level diesel
12 exposure resulting from a reduced food intake; [2] heart and right ventricular mass increases
13 caused by hypoxia [gasoline]; or [3] increased pulmonary vascular resistance and/or airway
14 pathology caused by paniculate retention [diesel]). The other indices, although compatible
15 with impaired liver or kidney dysfunction, were not confirmed during histopathological
16 examinations.
17 The toxic effects of gasoline engine exhaust were studied in beagle dogs during an
18 exposure regimen of 16 h/day, 7 days/week for 68 mo (Stara et al., 1980). The dogs were
19 exposed to irradiated or nonirradiated exhaust with or without concurrent exposure to a
20 mixture of sulfur dioxide and sulfuric acid. The composition of the exposure atmospheres is
21 given in Table 5-17. During exposure, the dogs had significantly higher hemoglobin and
22 hematocrit values; RBCs were also elevated, but significance was not obtained at each 6-mo
23 measurement period (Orthoefer et al., 1980). There were no effects on white blood cells,
24 MCV, MCH, or MCHC, or clinical chemistry parameters (Orthoefer et al., 1980) or lung
25 collagen content (Bhatnagar, 1980). At 18 and 36 mo of exposure, pulmonary function
26 parameters were not significantly altered in the exposed dogs. After 61 mo, dogs exposed to
27 irradiated exhaust had higher total expiratory resistance and some evidence of right
28 ventricular hypertrophy when compared to the controls. Dogs that received the nonirradiated
29 gasoline exhaust containing added sulfur dioxide and sulfuric acid aerosol had a higher mean
30 residual volume to total lung capacity ratio, indicating evidence of pulmonary hyperinflation
31 (air-trapping). Pulmonary function studies conducted 2 years after the exposures were
December 1994 5.75 DRAFT-DO NOT QUOTE OR CITE
-------
n
-Q
J
O
i
a
M"^
H
£
y
H
W
fc
O
tf
ffi
u
1
OH
O
1
t/3
H
w
r»'
i-<
t
»/>
u
J
3
H
U?
W3>
*d
1 1
•S ,-v
JU g
8 ex
|a
o|
oo e£
/-^
CN^g
O ~gj.
p
xS
Zp*i [
ex
c?|
2 3<
0?
U &
^
1
.vj r^ to r*' ro
^ (S tS (S (S
i i § i i
-------
1 terminated demonstrated more functional abnormalities than those found at 61 mo of
2 exposure; all exhaust-exposed dogs had pulmonary function and structural differences from
3 the control groups (Stara et al., 1980). Auto-exhaust (gasoline engine) exposure injured both
4 airways and parenchyma, inducing abnormalities in ventilatory resistance, gas-exchange, and
5 lung volume values. Histological studies revealed substantial atypical bronchiolar epithelial
6 hyperplasia in both the nonirradiated and irradiated gasoline exhaust groups (Hyde et al.,
7 1978). The hyperplastic lesions were derived from nonciliated epithelial cells. Aggregations
8 of inflammatory cells were also observed in the distal regions of the terminal bronchioles.
9 Data on the degree of peribronchiolar fibrosis were not presented. The nonirradiated
10 gasoline exhaust containing added sulfur dioxide and sulfuric acid aerosol showed the greatest
11 severity of bronchiolar hyperplasia, the highest level of pulmonary resistance, and a
12 significant increase in squamous metaplasia in the trachea and bronchi.
13 The available data from limited long-term studies of exposure to diesel exhaust or
14 gasoline exhaust provide little to no relevant comparative data to equate relative health risks.
15 However, based on these data individually, coupled with studies on individual components of
16 emissions from internal combustion engines, one would expect certain pathologies to be
17 evident as a result of the atmospheric components present and their respective concentrations.
18 With lower concentrations of irritant, oxidant gases (e.g., nitrogen dioxide and ozone), the
19 principle pulmonary pathology lies in the distal portions of the bronchiolar airways;
20 increasing concentrations can produce important parenchymal pathology (U.S. Environmental
21 Protection Agency, 1993, 1994). Low doses of retained particles, depending on their size
22 and chemical composition, tend to affect lung parenchyma more so than airways because of
23 their much longer residence time and tendency to be aggregated over time (Stokinger, 1977).
24 With such evidence, gasoline exhaust would be more apt to produce terminal airway disease
25 and diesel exhaust would be more apt to produce parenchymal disease. The interactive
26 anatomical and physiological properties of lung tissue and the interactive cellular toxicity
27 between lung tissues and pollutants found in emissions from internal combustion engines,
28 however, often produce lesions in both the airways and parenchyma with varying degrees of
29 pathology, depending on the exhaust composition and concentration. The observation in dogs
30 that pulmonary function decrements continued following termination of exposure to gasoline
31 exhaust necessitates that such experimental regimens be examined more often.
December 1994 5.77 DRAFT-DO NOT QUOTE OR CITE
-------
1 5.5 DOSE-RATE AND PARTICULATE CAUSATIVE ISSUES
2 The purpose of animal toxicological experimentation is to identify the hazards and
3 dose-response effects posed by a chemical substance or complex mixture and to extrapolate
4 these effects to humans for subsequent health assessments. The cardinal principle in such a
5 process is that the intensity and character of the toxic action is a function of the dose of the
6 toxic agent(s) that reaches the critical site of action. The considerable body of evidence
7 reviewed clearly denotes that major noncancerous health hazards may be presented to the
8 lung following the inhalation of diesel exhaust. Based on pulmonary function and
9 histopathological and histochemical effects, a determination can be made concerning what
10 dose/exposure rates of diesel exhaust (expressed in terms of the diesel particulate
11 concentration) result in an injury to the lung and which appear to elicit no effect. The
12 inhalation of poorly soluble particles, such as those found in diesel exhaust, increases the
13 pulmonary particulate burden. When the dosing rate exceeds the ability of the pulmonary
14 defense mechanisms to achieve a steady-state lung burden of particles, there is a slowing of
15 clearance and the progressive retention of particles in the lung that can ultimately approach a
16 complete cessation of lung clearance (Morrow, 1988). This phenomenon has practical
17 significance both for the interpretation of experimental inhalation data and for the prevention
18 of disease in humans exposed to airborne particles.
19 Mauderly et al. (1989) reported that for those studies using exposures of 24 mo or
20 longer, the pulmonary tumor incidence appeared to be better correlated with exposure
21 intensity expressed as mg-h-m"3/week than with the total cumulative exposure. The lowest
22 exposure rate causing a statistically significant increase in tumor incidence was
23 122.5 mg-h-m"3/week (Mauderly et al., 1989). A 2-degree polynomial function fitted to
24 exposure rate-lung tumor incidence data from several studies revealed that the lung tumor
25 incidence of the exhaust-exposed rats exceeded the upper limit of the incidence among
26 control rats at an exposure rate of approximately 170 mg-h-m"3/week (Mauderly et al.,
27 1990b).
28 The lowest exposure rate (expressed in terms of the diesel particulate concentration)
29 was about 70 mg-h-m"3/week in monkeys (Lewis et al., 1989), 122 to 140 mg-h-m"3/week in
30 rats (Mauderly et al., 1988; Gross et al., 1981b), 336 to 403 mg-h-m'3 in hamsters (Vinegar
31 et al., 1981a,b; Heinrich et al., 1986a), and 504 mg-h-m"3/week in cats (Pepelko et al.,
December 1994 5-78 DRAFT-DO NOT QUOTE OR CITE
-------
1 1980, 1981; Moorman et al., 1986). The exposure rates at which some evidence of
2 pulmonary histopathological or histochemical effects were noted were about 70, 104, 176,
3 210, and 249 mg-h-m'3/week in rats (Lewis et al., 1989; Research Committee for HERP
4 Studies, 1988; Mauderly et al., 1987a,b; Heinrich et al., 1986a; Karagianes et al., 1981);
5 146, 403, and 504 mg-h-m"3/week in the hamster (Heinrich et al., 1982, 1986a; Pepelko,
6 1982b); 82 mg-h-rn3/week in the guinea pig (Barnhart et al., 1981, 1982) and 403
7 mg-h-m~3/week in the mouse (Heinrich et al., 1986a). From a histopathological or
8 histochemical perspective, no-observable-effect exposure rates were reported to be 27, 56,
9 and 70 mg-h-m~3/week for the guinea pig, rat and monkey, respectively (Barnhart et al.,
10 1981, 1982; Brightwell et al., 1986; Lewis et al., 1989).
11 The data for exposure intensities that cause pulmonary injury demonstrate that they are
12 less than the exposure intensities reported to be necessary to induce lung tumors. Using the
13 most widely studied laboratory animal species and the one reported to be the most sensitive
14 to tumor induction, the laboratory rat, the lowest no-adverse-effect exposure intensity for
15 lung injury was 56 mg-h-m"3/week. The lowest observed effect level for pulmonary injury in
16 rats was 70 mg-h-m~3/week, and, for pulmonary tumors, 122.5 mg-h-m"3/week. The results
17 clearly show that lower exposure intensities, and equivalent total doses (because these results
18 in rats were for 104 weeks or longer) produce noncancerous pulmonary disease in the
19 absence of pulmonary tumors. Such data are supportive of the position that inflammatory
20 and proliferative changes in the lung may play a key role in the etiology of pulmonary
21 tumors in exposed rats (Mauderly et al., 1990b). Adults who have a preexisting condition
22 that may predispose their lungs to increased particle retention (e.g., smoking or high
23 particulate burdens from nondiesel sources), inflammation (e.g., repeated respiratory
24 infections), epithelial proliferation (e.g., chronic bronchitis), and fibrosis (e.g., silica
25 exposure) and infants and children due to their developing pulmonary and immunologic
26 systems, may have a greater susceptibility to the toxic actions of diesel exhaust.
27 There is also the issue of whether the noncancerous health effects related to exposure
28 to diesel exhaust are caused by the carbonaceous core of the particle or substances adsorbed
29 onto the core, or both.
30 Current knowledge is that much of the toxicity resulting from the inhalation of diesel
31 exhaust relates to the carbonaceous core of the particles. Several studies on inhaled aerosols
December 1994 5.79 DRAFT-DO NOT QUOTE OR CITE
-------
1 demonstrate that lung reactions characterized by an appearance of particle-laden AMs and
2 their infiltration into the alveolar ducts, adjoining alveoli and tracheobronchial lymph nodes,
3 hyperplasia of Type II cells, and the impairment of pulmonary clearance mechanisms are not
4 limited to exposure to diesel particles. Such responses have also been observed following the
5 inhalation of coal dust (Lewis et al., 1989; Karagianes et al., 1986), titanium dioxide
6 (Lee et al., 1985), titanium tetrachloride hydrolysis products (Lee et al., 1986), quartz
7 (Klosterkotter and Buneman, 1961), volcanic ash (Wehner et al., 1986), amosite (Bolton
8 et al., 1983) and manmade mineral fibers (Lee et al., 1988). In more recent studies, animals
9 have been exposed to carbon black that is similar to the carbon core of the diesel exhaust
10 particle. Nikula et al. (1994) exposed rats for 24 mo to carbon black or diesel exhaust at
11 exposure rates of 200 or 520 mg-h-m"3. Both concentrations induced macrophage
12 hyperplasia, epithelial proliferation, inflammation, and fibrosis. Dungworth et al. (1994)
13 reported moderate to severe inflammation characterized by multifocal bronchoalveolar
14 hyperplasia, alveolar histiocytosis, and focal segmental fibrosis in rats exposed to carbon
15 black for up to 20 mo at exposure rates of 510 to 540 mg-h-m"3. The observed lung
16 pathology reflects notable dose-response relationships, and usually evolves in a similar
17 manner. With increasing dose, there is an increased accumulation and aggregation of
18 particle-laden AMs, Type II cell hyperplasia, a foamy (degenerative) macrophage response,
19 alveolar proteinosis, alveolar bronchiolization, cholesterol granulomas, and often squamous
20 cell carcinomas and bronchioalveolar adenomas derived from metaplastic squamous cells in
21 the areas of alveolar bronchiolization. Particle size, volume, surface area, and/or
22 composition may be the critical element(s) in the overload phenomenon following exposure to
23 diesel particles. The overloaded macrcphages secrete a variety of cytokines, oxidants, and
24 proteolytic enzymes that are responsible for inducing particle aggregation and damaging of
25 adjacent epithelial tissue (Oberdorster and Yu, 1991). For a more detailed discussion of
26 mechanism see Chapter 10.
27 The principal noncancerous health hazard to humans posed by exposure to diesel
28 exhaust is a structural or functional injury to the lung. Such effects are demonstrable at dose
29 rates or cumulative doses of diesel particles lower than those reported to be necessary to
30 induce lung tumors. Current knowledge indicates that the carbonaceous core of diesel
31 particles is the major causative factor in the injury to the lung but that other factors such as
December 1994 5-80 DRAFT-DO NOT QUOTE OR CITE
-------
1 the cytotoxicity of adsorbed substances on the particles may also play a role. The lung
2 injury appears to be mediated through effects on pulmonary AMs. Because noncancerous
3 pulmonary effects occur at lower doses than does tumor induction and because these effects
4 may be cofactors in the etiology of diesel exhaust-induced tumors, noncancerous pulmonary
5 effects must be considered in the total evaluation of diesel exhaust, notably the paniculate
6 component.
7
8
9 5.6 SUMMARY AND DISCUSSION
10 5.6.1 Effects of Diesel Exhaust on Humans
11 The most readily identified acute noncancer health effect of diesel exhaust on humans
12 is its ability to elicit subjective complaints of eye, throat, and bronchial irritation and
13 neuropsychological symptoms such as headache, lightheadedness, nausea, vomiting, and
14 numbness and tingling of the extremities. Studies of the perception and offensiveness of the
15 odor of diesel exhaust and a human volunteer study in an exposure chamber have
16 demonstrated that the time of onset of the human subjective symptoms is inversely related to
17 increasing concentrations of diesel exhaust and the severity is directly related to increasing
18 concentrations of diesel exhaust. In one study in which a diesel engine was operated under
19 varying load conditions, a dilution factor of 140 to 475 was needed to reduce the exhaust
20 level to an odor-detection threshold level.
21 A public health issue is whether short-term exposure to diesel exhaust might result in
22 an acute decrement in ventilatory function and whether the frequent repetition of such acute
23 respiratory effects could result in chronic lung function impairment. One convenient means
24 of studying acute decrements in ventilatory function is to monitor differences in pulmonary
25 function in occupationally exposed workers at the beginning and end of a workshift.
26 In studies of underground miners, bus garage workers, dock workers, and locomotive
27 repairmen, increases in respiratory symptoms (cough, phlegm, and dyspnea) and decreases in
28 lung function (FVC, FEV1} PEFR, and FEF25.75) over the course of a workshift were
29 generally found to be minimal and not statistically significant. In a study of acute respiratory
30 responses in diesel bus garage workers, there was an increased reporting of cough, labored
31 breathing, chest tightness, and wheezing, but no reductions in pulmonary function were
December 1994 5.81 DRAFT-DO NOT QUOTE OR CITE
-------
1 associated with exposure to diesel exhaust. Pulmonary function was affected in stevedores
2 over a workshift exposure to diesel exhaust but normalized after a few days without exposure
3 to diesel exhaust fumes. In a third study, there was a trend toward greater ventilatory
4 function changes during a workshift among coal miners, but the decrements were similar in
5 miners exposed and not exposed to diesel exhaust.
6 Smokers appeared to demonstrate larger workshift respiratory function decrements and
7 increased incidents of respiratory symptoms. Acute sensory and respiratory symptoms were
8 earlier and more sensitive indicators of potential health risks from diesel exposure than were
9 decrements in pulmonary function. Studies on the acute health effects of exposure to diesel
10 exhaust in humans, experimental and epidemiological, have failed to demonstrate a consistent
11 pattern of adverse effects on respiratory morbidity; the majority of studies offer, at best,
12 equivocal evidence for an exposure-response relationship. The environmental contaminants
13 have frequently been below permissible workplace exposure limits; in those few cases where
14 health effects have been reported, the authors have failed to identify conclusively the
15 individual or collective causative agents in the diesel exhaust.
16 Chronic effects of diesel exhaust exposure have been evaluated in epidemiological
17 studies of occupationally exposed workers (metal and nonmetal miners, railroad yard
18 workers, stevedores, and bus garage mechanics). Most of the epidemiological data indicate
19 an absence of an excess risk of chronic respiratory disease associated with exposure to diesel
20 exhaust. In a few studies, a higher prevalence of respiratory symptoms, primarily cough,
21 phlegm, or chronic bronchitis, were observed among the exposed. These increased
22 symptoms, however, were usually not accompanied by significant changes in pulmonary
23 function. Reductions in FEV} and FVC, and to a lesser extent FEF50 and FEF75, also have
24 been reported. Two studies detected statistically significant decrements in baseline
25 pulmonary function consistent with evidence of obstructive airway disease. One study was of
26 stevedores and had a very limited sample size of 17 exposed and 11 controls. The second
27 study was in coal miners and showed that both underground and surface workers at diesel-
28 use mines had somewhat lower pulmonary performance than their matched controls. The
29 proportion of workers in or at diesel-use mines, however, showed equivalent evidence of
30 obstructive airway disease and for this reason the authors of the second paper felt that factors
31 other than diesel exposure might have been responsible. A doubling of minor restrictive
December 1994 5-82 DRAFT-DO NOT QUOTE OR CITE
-------
1 airway disease was also observed in workers in or at diesel-use mines. These two studies
2 coupled with other reported nonsignificant trends in respiratory flow-volume measurements
3 suggest that exposure to diesel exhaust may impair pulmonary function among occupational
4 populations. Epidemiological studies of the effects of diesel exhaust on organ systems other
5 than the pulmonary system are scant. Whereas a preliminary study of the association of
6 cardiovascular mortality and exposure to diesel exhaust found a fourfold higher risk ratio, a
7 more comprehensive epidemiological study by the same investigators found no significant
8 difference between the observed and expected number of deaths caused by cardiovascular
9 disease.
10 Caution is warranted in the interpretation of the results of the epidemiological studies
11 that have addressed noncarcinogenic health effects from exposure to diesel exhaust. These
12 investigations suffer from a myriad of methodological problems including: (1) incomplete
13 information on the extent of exposure to diesel exhaust, necessitating in some studies
14 estimations of exposures from job titles and resultant misclassification; (2) the presence of
15 confounding variables such as smoking or occupational exposures to other toxic substances
16 (e.g., mine dusts); and (3) the short duration and low intensity of exposure. These
17 limitations restrict drawing definitive conclusions as to the cause of any noncarcinogenic
18 diesel exhaust effect, observed or reported.
19
20 5.6.2 Effects of Diesel Exhaust on Animals
21 Animal studies of the toxic effects of diesel exhaust have involved acute, subchronic,
22 and chronic exposure regimens. In acute exposure studies, toxic effects appear to have been
23 associated primarily with high concentrations of carbon monoxide, nitrogen dioxide, and
24 aliphatic aldehydes. In short- and long-term studies, toxic effects have been associated with
25 exposure to the complex exhaust mixture. Effects of diesel exhaust in various animal species
26 are summarized in Tables 5-2 to 5-14. In short-term studies, health effects are not readily
27 apparent and when found are mild and result from concentrations of about 6 mg/m3
28 particulate matter and durations of exposure approximating 20 h/day. There is ample
29 evidence, however, that short-term exposures at lower levels of diesel exhaust impact the
30 lung as indicated by an accumulation of particles, evidence of inflammatory response, AM
31 aggregation and accumulation near the terminal bronchioles, Type II cell proliferation, and
December 1994 5.33 DRAFT-DO NOT QUOTE OR CITE
-------
1 the thickening of alveolar walls adjacent to AM aggregation. Little evidence exists,
2 however, from short-term studies that exposure to diesel exhaust impairs lung function.
3 Chronic exposures cause lung pathology that results in reduced growth rates, altered
4 pulmonary function, and increased diesel paniculate retention in the lung. Exposures to
5 diesel exhaust have also been associated with increased susceptibility to respiratory tract
6 infection, neurological or behavioral changes, an increase in banded neutrophils, and
7 morphological alterations in the liver.
8
9 5.6.2.1 Effects on Survival and Growth
10 The data presented in Table 5-3 show limited effects on survival in mice and rats and
11 some evidence of reduced body weight in rats following chronic exposures to concentrations
12 of 1.5 mg/m3 paniculate matter or higher and exposure durations of 16 to 20 h/day,
13 5 days/week for 104 to 130 weeks. Increased lung weights and lung to body weight ratios in
14 rats, mice, and hamsters; an increased heart to body weight ratio in rats; and decreased lung
15 and kidney weights in cats have been reported following chronic exposure to diesel exhaust.
16 No evidence was found of an effect of diesel exhaust on other body organs (Table 5-4). The
17 lowest observed effect level in rats approximated 1 to 2 mg/m3 for 7 h /day, 5 days/week for
18 104 weeks.
19
20 5.6.2.2 Effects on Pulmonary Function
21 Pulmonary function impairment has been reported in rats, hamsters, cats, and
22 monkeys exposed to diesel exhaust and included lung mechanical properties (compliance and
23 resistance), diffusing capacity, lung volumes, and ventilatory performance (Table 5-5).
24 Pulmonary function studies were not conducted in all the chronic exposure investigations or
25 for all species utilized. The effects generally appeared only after prolonged exposures. The
26 lowest exposure levels (expressed in terms of diesel particulate concentrations) that resulted
27 in impairment of pulmonary function occurred at 2 mg/m3 in cynomolgus monkeys, 1.5 and
28 3.5 mg/m3 in rats, 4.24 and 6 mg/m3 in hamsters, and 11.7 mg/m3 in cats. Exposures in
29 monkeys, cats, and rats (3.5 mg/m3) were for 7 to 8 h/day, 5 days/week for 104 to
30 130 weeks. Exposures in hamsters and rats (1.5 mg/m3) varied in hours per day (8 to 20)
31 and weeks of exposure (26 to 130). In all species but the monkey, the testing results were
December 1994 5-84 DRAFT-DO NOT QUOTE OR CITE
-------
1 consistent with restrictive lung disease; alteration in expiratory flow rates indicated that
2 2 mg/m3 was a LOAEL. Monkeys demonstrated evidence of obstructive airway disease.
3 The nature of the pulmonary impairment is dependent on the dose of toxicants delivered to
4 and retained in the lung, the site of deposition and effective clearance or repair, and the
5 anatomy and physiology of the affected species; these variables appear to be factors in the
6 disparity of the airway disease in monkey versus the other species tested.
7
8 5.6.2.3 Histopathological and Histochemical Effects
9 Histological studies have demonstrated that chronic exposure to diesel exhaust can
10 result in effects on respiratory tract tissue (Table 5-6). Typical findings include alveolar
11 histiocytosis, macrophage aggregation, tissue inflammation, increase in polymorphonuclear
12 leukocytes, hyperplasia of bronchiolar and alveolar Type II cells, thickened alveolar septa,
13 edema, fibrosis, and emphysema. Lesions in the trachea and bronchi were observed in some
14 studies. Associated with these histopathological findings were various histochemical changes
15 in the lung, including increases in lung DNA, total protein, alkaline and acid phosphatase,
16 glucose-6-phosphate dehydrogenase; increased synthesis of collagen; and release of
17 inflammatory mediators such as leukotriene LTD and prostaglandin PGF2a. Although the
18 overall laboratory evidence is that prolonged exposure to diesel exhaust paniculate matter
19 results in histopathological and histochemical changes in the lungs of exposed animals, some
20 studies have also demonstrated that there may be a threshold of exposure to diesel exhaust
21 below which pathologic changes do not occur. These no-observed-adverse-effect levels were
22 reported to be 2 mg/m3 for cynomolgus monkeys, 0.11 to 0.35 mg/m3 for rats, and
23 0.25 mg/m3 paniculate matter for guinea pigs exposed for 7 to 20 h/day, 5 to 5.5 days/week
24 for 104 to 130 weeks.
25
26 5.6.2.4 Effects on Defense Mechanisms
27 The pathological effects of diesel exhaust paniculate matter appear to be strongly
28 dependent on the relative rates of pulmonary deposition and clearance (Table 5-7).
29 Clearance of particles from the alveolar region of the lungs is a multiphasic process involving
30 phagocytosis by AMs. Chronic exposure to diesel particle concentrations of about 1 mg/m3
31 or above, under varying exposure durations, causes pulmonary clearance to be reduced with
December 1994 5.85 DRAFT-DO NOT QUOTE OR CITE
-------
1 concomitant focal aggregations of particle-laden AMs, particularly in the peribronchiolar and
2 alveolar regions, as well as in the hilar and mediastinal lymph nodes. The exposure
3 concentration at which focal aggregates of particle-laden macrophages occur may vary from
4 species to species, depending on rate of uptake and pulmonary deposition, pulmonary
5 clearance rates, the relative size of the macrophage population per unit of lung tissue, the
6 rate of recruitment of macrophages and leukocytes, and the relative efficiencies for removal
7 of particles by the mucociliary and lymphatic transport system. The principal mechanism of
8 reduced particle clearance appears to be an effect on pulmonary AMs. This impairment of
9 particle clearance seems to be nonspecific and applies primarily to dusts that are persistently
10 retained in the lungs. Lung dust levels of approximately 1 to 2 mg/g lung tissue appear to
11 produce this effect in the Fischer 344 rat. Morrow (1988) suggested that the inability of
12 particle-laden AMs to translocate to the mucociliary escalator is correlated to an average
13 composite particle volume per AM in the lung. When this particle volume exceeds
14 approximately 60 /xm3 per AM in the Fischer 344 rat, impairment of clearance appears to be
15 initiated. When the particulate volume exceeds approximately 600 /mi3 per cell, evidence
16 suggests that AM-mediated particulate clearance virtually ceases and agglomerated particle-
17 laden macrophages remain in the alveolar region and increasingly nonphagocytized dust
18 particles translocate to the pulmonary interstitium. Data for other laboratory animal species
19 and humans are, unfortunately, very limited.
20 Several animal studies have indicated that exposure to diesel exhaust can reduce an
21 animal's resistance to respiratory infections. This effect, which can occur even after only
22 2 or 6 h of exposure to diesel exhaust containing 5 to 8 mg/m3 particulate matter, does not
23 appear to be caused by direct impairment of the lymphoid or splenic immune systems;
24 however, in one study of influenza virus infection, interferon levels and hemagglutinin
25 antibody levels were adversely affected in the exposed mice. Studies on the effects of
26 exposure to diesel exhaust or diesel particles on the immune system of laboratory animals
27 have produced equivocal results (Table 5-8).
28
29 5.6.2.5 Neurological and Behavioral Effects
30 Behavioral effects have been observed in rats exposed to diesel exhaust from birth to
31 28 days of age (Table 5-13). Exposure caused a decreased level of spontaneous locomotor
December 1994 5-86 DRAFT-DO NOT QUOTE OR CITE
-------
1 activity and a detrimental effect on learning in adulthood. In agreement with the behavioral
2 changes was physiological evidence for delayed neuronal maturation. Exposures were to
3 6 mg/m3 particulate matter for 8 h/day, 7 days/week from birth to about 7, 14, 21, or
4 28 days of age.
5
6 5.6.2.6 Other Noncancerous Effects
7 Essentially negative effects (based on the weight of evidence of a number of studies)
8 were noted for reproductive and teratogenic effects in mice, rats, rabbits, and monkeys;
9 clinical chemistry and hematology in the rat, cat, hamster, and monkeys; and enzyme
10 induction in the rat and mouse (Tables 5-10 through 5-12 and 5-14).
11
12 5.6.3 Comparison of Filtered and Unfiltered Diesel Exhaust
13 The comparison of the toxic responses in laboratory animals exposed to whole diesel
14 exhaust or filtered exhaust containing no particles demonstrates across laboratories that diesel
15 particles are the principal etiologic agent of noncancerous health effects in laboratory animals
16 exposed to diesel exhaust (Table 5-15). Whether the particles act additively or
17 synergistically with the gases cannot be determined from the designs of the studies. Under
18 equivalent exposure regimens, hamsters have lower levels of retained diesel particles in their
19 lungs than do rats and mice and consequently less pulmonary function impairment and
20 pulmonary pathology. These differences may result from a lower intake rate of diesel
21 particles, lower deposition rate and/or more rapid clearance rate, or lung tissue that is less
22 susceptible to the cytotoxicity of diesel particles. Observations of a decreased respiration in
23 hamsters when exposed by inhalation favor lower intake and deposition rates.
24
25 5.6.4 Interactive Effects of Diesel Exhaust
26 There is no direct evidence that diesel exhaust interacts with other substances in an
27 exposure environment, other than an impaired resistance to respiratory tract infections.
28 Young animals were not more susceptible. In several ways, animals with laboratory-induced
29 emphysema were more resistant. There is experimental evidence that both inorganic and
30 organic compounds can be adsorbed onto carbonaceous particles. When such substances
31 become affiliated with particles, these substances can be carried deeper into the lungs where
December 1994 5.37 DRAFT-DO NOT QUOTE OR CITE
-------
1 they might have a more direct and potent effect on epithelial cells or on AM ingesting the
2 particles. Specific studies to test interactive effects of diesel exhaust with atmospheric
3 contaminants, other than coal dust, have not been conducted. Coal dust and diesel particles
4 had an additive effect only.
5
6 5.6.5 Comparisons with Gasoline Exhausts
7 There has only been one study directly comparing toxic responses in laboratory
8 animals exposed to either gasoline or diesel exhausts. Design limitations, imparted by a
9 single equivalent dilution of each type of exhaust resulting in toxic concentrations of CO and
10 NOX, and the frequent reporting of statistically significant results without the absolute values
11 make scientific comparisons between the two types of exhaust impossible. The most
12 noteworthy contribution of this study was that, when a catalytic converter was used with the
13 gasoline exhaust and the particles were removed from the diesel exhaust, there was little
14 evidence of toxic responses in the exposed animals. In the absence of direct or relevant data
15 on the comparative toxicities of gasoline and diesel exhausts, one is limited to an
16 extrapolation of results performed independently on each type of exhaust. Low
17 concentrations of ozone and nitrogen dioxide produce a principal pathology in the distal or
18 terminal bronchiolar airways; increasing concentrations, however, can produce important
19 deep lung or alveolar injury. Low doses of retained particles, depending on their size and
20 chemical composition, tend to affect the deep lung or alveolar region more than the airways
21 because of the longer residence time and the tendency to be aggregated over time. With
22 such evidence, gasoline exhaust would be more apt to produce terminal airway disease and
23 diesel exhaust, parenchymal disease. Further evaluations reveal that emissions from internal
24 combustion engines often produce lesions in both the airways and parenchyma with varying
25 degrees of pathology, depending on composition and concentration of the toxic components.
26 The introduction of new control technologies (e.g., the catalytic converter to control
27 emissions from gasoline engines and prototype controls such as ceramic mufflers to control
28 diesel particle emissions) modify exhaust emissions both quantitatively and qualitatively.
29 Engineering controls to reduce particle emissions would likely result in a reduction of health
30 consequences as well.
31
December 1994 5-88 DRAFT-DO NOT QUOTE OR CITE
-------
1 5.6.6 Summary
2 The principal noncancerous health hazard to humans from exposure to diesel exhaust
3 is a structural, functional, and/or biochemical injury to the lung. Although most of the
4 reported effects were observed at relatively high exposure levels compared with existing or
5 projected urban levels, the noncancerous effects are demonstrable at dose rates or cumulative
6 doses lower than those reported to be necessary to induce lung tumors. Furthermore, prior
7 lung injury appears to be a major contributor to the neoplastic process. Current knowledge
8 indicates that the carbonaceous core of the diesel particle is the prune causative agent of lung
9 injury. The lung injury appears to be mediated by a progressive impairment of AMs.
10 Because noncancerous pulmonary effects occur at lower doses than those inducing tumors
11 and appear to be cofactors in the etiology of diesel exhaust-induced tumors, noncancerous
12 pulmonary effects are relevant factors in the development of risk assessments.
December 1994 5.89 DRAFT-DO NOT QUOTE OR CITE
-------
1 REFERENCES
2
3 Ames, R. G.; Attfield, M. D.; Hankinson, J. L.; Hearl, F. J.; Reger, R. B. (1982) Acute respiratory effects of
4 exposure to diesel emissions in coal miners. Am. Rev. Respir. Dis. 125: 39-42.
5
6 Ames, R. G.; Reger, R. B.; Hall, D. S. (1984) Chronic respiratory effects of exposure to diesel emissions in
7 coal mines. Arch. Environ. Health 39: 389-394.
8
9 Attfield, M. D. (1978) The effect of exposure to silica and diesel exhaust in underground metal and nonmetal
10 miners. In: Kelley, W. D., ed. Industrial hygiene for mining and tunneling: proceedings of an American
11 Conference of Governmental Industrial Hygienists topical symposium; November 1978; Denver, CO.
12 pp. 129-135.
13
14 Attfield, M. D.; Trabant, G. D.; Wheeler, R. W. (1982) Exposure to diesel fumes and dust at six potash
15 mines. Ann. Occup. Hyg. 26: 817-831.
16
17 Barnhart, M. I.; Chen, S.-t.; Salley, S. O.; Puro, H. (1981) Ultrastructure and morphometry of the alveolar
18 lung of guinea pigs chronically exposed to diesel engine exhaust: six month's experience. J. Appl.
19 Toxicol. 1: 88-103.
20
21 Barnhart, M. I.; Salley, S. O.; Chen, S.-t.; Puro, H. (1982) Morphometric ultrastructural analysis of alveolar
22 lungs of guinea pigs chronically exposed by inhalation to diesel exhaust (DE). In: Lewtas, J., ed.
23 Toxicological effects of emissions from diesel engines: proceedings of the Environmental Protection
24 Agency diesel emissions symposium; October, 1981; Raleigh, NC. New York, NY: Elsevier
25 Biomedical; pp. 183-200. (Developments in toxicology and environmental science: v. 10).
26
27 Battigelli, M. C. (1965) Effects of diesel exhaust. Arch. Environ. Health 10: 165-167.
28
29 Battigelli, M. C.; Mannella, R. J.; Hatch, T. F. (1964) Environmental and clinical investigation of workmen
30 exposed to diesel exhaust in railroad engine houses. Ind. Med. Surg. 33: 121-124.
31
32 Battigelli, M. C.; Hengstenberg, F.; Mannela, R. J.; Thomas, A. P. (1966) Mucociliary activity. Arch.
33 Environ. Health 12: 460-466.
34
35 Bhatnagar, R. S. (1980) Collagen and prolyl hydroxylase levels in lungs in dogs exposed to automobile exhaust
36 and other noxious gas mixtures. In: Stara, J. F.; Dungworth, D. L.; Orthoefer, J. G.; Tyler, W. S.,
37 eds. Long-term effects of air pollutants in canine species. Cincinnati, OH: U.S. Environmental
38 Protection Agency, Environmental Criteria and Assessment Office; report no. EPA-600/8-80-014;
39 pp. 71-87.
40
41 Bhatnagar, R. S.; Hussain, M. Z.; Sorensen, K. R.; Von Dohlen, F. M.; Danner, R. M.; McMillan, L.; Lee,
42 S. D. (1980) Biochemical alterations in lung connective tissue in rats and mice exposed to diesel
43 emissions. In: Pepelko, W. E.; Danner, R. M.; Clarke, N. A., eds. Health effects of diesel engine
44 emissions: proceedings of an international symposium, v. 1; December 1979; Cincinnati, OH.
45 Cincinnati, OH: U.S. Environmental Protection Agency, Health Effects Research Laboratory;
46 pp. 557-570; EPA report no. EPA-600/9-80-057a. Available from: NTIS, Springfield, VA;
47 PB81-173809.
48
49 Bice, D. E.; Mauderly, J. L.; Jones, R. K.; McClellan, R. O. (1985) Effects of inhaled diesel exhaust on
50 immune responses after lung immunization. Fundam. Appl. Toxicol. 5: 1075-1086.
51
52 Bolton, R. E.; Vincent, J. H.; Jones, A. D.; Addison, J.; Beckett, S. T. (1983) An overload hypothesis for
53 pulmonary clearance of UICC amosite fibres inhaled by rats. Br. J. Ind. Med. 40: 264-272.
54
December 1994 5-90 DRAFT-DO NOT QUOTE OR CITE
-------
1 Boren, H. G. (1964) Carbon as a carrier mechanism for irritant gases. Arch. Environ. Health 8: 119-124.
2
3 Brightwell, J.; Fouillet, X.; Cassano-Zoppi, A.-L.; Gatz, R.; Duchosal, F. (1986) Neoplastic and functional
4 changes in rodents after chronic inhalation of engine exhaust emissions. In: Ishinishi, N.; Koizumi, A.;
5 McClellan, R. 0.; Stbber, W., eds. Carcinogenic and mutagenic effects of diesel engine exhaust:
6 proceedings of the international satellite symposium on lexicological effects of emissions from diesel
7 engines; July; Tsukuba Science City, Japan. Amsterdam, Holland: Elsevier Science Publishers B. V.;
8 pp. 471-485. (Developments in toxicology and environmental science: v. 13).
9
10 Campbell, K. I.; George, E. L.; Washington, I. S., Jr. (1980) Enhanced susceptibility to infection in mice after
11 exposure to dilute exhaust from light duty diesel engines. In: Pepelko, W. E.; Danner, R. M.; Clarke,
12 N. A., eds. Health effects of diesel engine emissions: proceedings of an international symposium, v. 2;
13 December 1979; Cincinnati, OH. Cincinnati, OH: U.S. Environmental Protection Agency, Health
14 Effects Research Laboratory; pp. 772-785; EPA report no. EPA-600/9-80-057b. Available from: NTIS,
15 Springfield, VA; PB81-173817.
16
17 Campbell, K. I.; George, E. L.; Washington, I. S., Jr. (1981) Enhanced susceptibility to infection in mice after
18 exposure to dilute exhaust from light duty diesel engines. Environ. Int. 5: 377-382.
19
20 Carey, P.; Cohen, J. (1980) Comparison of gas phase hydrocarbon emissions from light-duty vehicles equipped
21 with diesel engines. U.S. Environmental Protection Agency technical report no. CTAB/PA/80-5.
22
23 Castleman, W. L.; Dungworth, D. L.; Schwartz, L. W.; Tyler, W. S. (1980) Acute respiratory bronchiolitis:
24 an ultrastructural and autoradiographic study of epithelial cell injury and renewal in rhesus monkeys
25 exposed to ozone. Am. J. Pathol. 98: 811-840.
26
27 Castranova, V.; Bowman, L.; Reasor, M. J.; Lewis, T.; Tucker, J.; Miles, P. R. (1985) The response of rat
28 alveolar macrophages to chronic inhalation of coal dust and/or diesel exhaust. Environ. Res.
29 36: 405-419.
30
31 Chan, T. L.; Lee, P. S.; Hering, W. E. (1981) Deposition and clearance of inhaled diesel exhaust particles in
32 the respiratory tract of Fischer rats. J. Appl. Toxicol. 1: 77-82.
33
34 Chaudhari, A.; Dutta, S. (1982) Alterations in tissue glutathione and angiotensin converting enzyme due to
35 inhalation of diesel engine exhaust. J. Toxicol. Environ. Health 9: 327-337
36
37 Chaudhari, A.; Farrer, R. G.; Dutta, S. (1980) Effect of exposure to diesel exhaust on pulmonary prostaglandin
38 dehydrogenase (PGDH) activity. In: Pepelko, W. E.; Danner, R. M.; Clarke, N. A., eds. Health
39 effects of diesel engine emissions: proceedings of an international symposium, v. 1; December 1979;
40 Cincinnati, OH. Cincinnati, OH: U.S. Environmental Protection Agency, Health Effects Research
41 Laboratory; pp. 532-537; EPA report no. EPA-600/9-80-057a. Available from: NTIS, Springfield VA-
42 PB81-173809.
43
44 Chaudhari, A.; Farrer, R. G.; Dutta, S. (1981) Effect of exposure to diesel exhaust on pulmonary prostaglandin
45 dehydrogenase (PGDH) activity. J. Appl. Toxicol. 1: 132-134.
46
47 Chen, K. C. (1986) Induction of aryl hydrocarbon hydroxylase in rat tissue following intratracheal instillation of
48 diesel paniculate extract and benzo[a]pyrene. J. Appl. Toxicol. 6: 259-262.
49
50 Chen, K.-C; Vostal, J. J. (1981) Aryl hydrocarbon hydroxylase activity induced by injected diesel paniculate
51 extract vs inhalation of diluted diesel exhaust. J. Appl. Toxicol. 1: 127-131.
•J £
53 Chen, S.; Weller, M. A.; Barnhart, M. I. (1980) Effects of diesel engine exhaust on pulmonary alveolar
54 macrophages. Scanning Electron Microsc. 3: 327-338.
December 1994 5.91 DRAFT-DO NOT QUOTE OR CITE
-------
1 Cuddihy, R. G.; Griffith, W. C.; Clark, C. R.; McClellan, R. O. (1981) Potential health and environmental
2 effects of light duty diesel vehicles II. Albuquerque, NM: Lovelace Biomedical and Environmental
3 Research Institute, Inhalation Toxicology Research Institute; report no. LMF-89.
4
5 Cuddihy, R. G.; Griffith, W. C.; McClellan, R. O. (1984) Health risks from light-duty diesel vehicles.
6 Environ. Sci. Technol. 18: 14A-21A.
7
8 Dungworth, D. L.; Mohr, U.; Heinrich, U.; Ernst, H.; Kittel, B. (1994) Pathologic effects of inhaled particles
9 in rat lungs: associations between inflammatory and neoplastic processes. In Mohr, U.; Dungworth,
10 D. L.; Mauderly, J. L.; Oberdorster, G., eds. Toxic and carcinogenic effects of solid particles in the
11 respiratory tract: [proceedings of the 4th international inhalation symposium]; March 1993; Hannover,
12 Germany. Washington, DC: International Life Sciences Institute Press; pp. 75-98.
13
14 Dziedzic, D. (1981) Differential counts of B and T lymphocytes in the lymph nodes, circulating blood and
15 spleen after inhalation of high concentrations of diesel exhaust. J. Appl. Toxicol. 1: 111-115.
16
17 Edling, C.; Axelson, O. (1984) Risk factors of coronary heart disease among personnel in a bus company. Int.
18 Arch. Occup. Environ. Health 54: 181-183.
19
20 Edling, C.; Anjou, C.-G.; Axelson, O.; Kling, H. (1987) Mortality among personnel exposed to diesel exhaust.
21 Int. Arch. Occup. Environ. Health 59: 559-565.
22
23 El Batawi, M. A.; Noweir, M. H. (1966) Health problems resulting from prolonged exposure to air pollution in
24 diesel bus garages. Ind. Health 4: 1-10.
25
26 Eskelson, C. D.; Strom, K. A.; Vostal, J. J.; Misiorowski, R. L. (1981) Lipids in the lung and lung gavage
27 fluid of animals exposed to diesel particulates. Toxicologist 1: 74-75.
28
29 Fedan, J. S.; Frazier, D. G.; Moorman, W. J.; Attfield, M. D.; Franczak, M. S.; Kosten, C. J.; Cahill, J. F.;
30 Lewis, T. R.; Green, F. H. Y. (1985) Effects of a two-year inhalation exposure of rats to coal dust
31 and/or diesel exhaust on tension responses of isolated airway smooth muscle. Am. Rev. Respir. Dis.
32 131: 651-655.
33
34 Forrest, L.; Lee, W. B.; Smalley, W. M. (1980) Assessment of environmental impacts of light-duty vehicle
35 dieselization [final report]. Washington, DC: U.S. Department of Transportation, National Traffic
36 Highway Safety Administration; report no. DOT-TSC-NHTSA-80-5. Available from: NTIS,
37 Springfield, VA; PB80-220759.
38
39 Gamble, J. F.; Jones, W. G. (1983) Respiratory effects of diesel exhaust in salt miners. Am. Rev. Respir. Dis.
40 128: 389-394.
41
42 Gamble, J.; Jones, W.; Hudak, J. (1983) An epidemiological study of salt miners in diesel and nondiesel mines.
43 Am. J. Ind. Med. 4: 435-458.
44
45 Gamble, J.; Jones, W.; Hudak, J.; Merchant, J. (1978) Acute changes hi pulmonary function in salt miners.
46 Presented at: ACGIH conference on mining and tunneling; November; Denver, CO. Cincinnati, OH:
47 American Conference of Governmental Industrial Hygienists; pp. 119-128.
48
49 Gamble, J.; Jones, W.; Minshall, S. (1987a) Epidemiological-environmental study of diesel bus garage workers:
50 acute effects of NC^ and respirable paniculate on the respiratory system. Environ. Res. 42: 201-214.
51
52 Gamble, J.; Jones, W.; Minshall, S. (1987b) Epidemiological-environmental study of diesel bus garage
53 workers: chronic effects of diesel exhaust on the respiratory system. Environ. Res. 44: 6-17.
54
December 1994 5-92 DRAFT-DO NOT QUOTE OR CITE
-------
1 Garg, B. D. (1983) Histologic quantitation of macrophage aggregates in the lungs of diesel exhaust exposed
2 rats. Acta Stereol. 2(suppl. 1): 235-238.
3
4 Gillespie, J. R. (1980) Review of the cardiovascular and pulmonary function studies on beagles exposed for
5 68 months to auto exhaust and other air pollutants. In: Stara, J. F.; Dungworth, D. L.; Orthoefer,
6 J. G.; Tyler, W. S., eds. Long-term effects of air pollutants: in canine species. Washington, DC: U.S.
7 Environmental Protection Agency, Environmental Criteria and Assessment Office; pp. 115-153; EPA
8 report no. EPA-600/8-80-014. Available from: NTIS, Springfield, VA; PB81-144875.
9
10 Green, F. H. Y; Boyd, R. L.; Danner-Rabovsky, J.; Fisher, M. J.; Moorman, W. J.; Ong, T.-M.; Tucker, J.;
11 Vallyathan, V.; Whong, W.-Z.; Zoldak, J.; Lewis, T. (1983) Inhalation studies of diesel exhaust and
12 coal dust in rats. Scand. J. Work Environ. Health 9: 181-188.
13
14 Griffis, L. C.; Wolff, R. K.; Henderson, R. F.; Griffith, W. C.; Mokler, B. V.; McClellan, R. O. (1983)
15 Clearance of diesel soot particles from rat lung after a subchronic diesel exhaust exposure. Fundam.
16 Appl. Toxicol. 3: 99-103.
17
18 Gross, K. B. (198la) Pulmonary function testing of animals chronically exposed to diluted diesel exhaust for
19 267 days. Environ. Int. 5: 331-337.
20
21 Gross, K. B. (1981b) Pulmonary function testing of animals chronically exposed to diluted diesel exhaust.
22 J. Appl. Toxicol. 1: 116-123.
23
24 Hahon, N.; Booth, J. A.; Green, F.; Lewis, T. R. (1985) Influenza virus infection in mice after exposure to
25 coal dust and diesel engine emissions. Environ. Res. 37: 44-60.
26
27 Hare, C. T.; Springer, K. J. (1971) Public response to diesel engine exhaust odors [final report]. San Antonio,
28 TX: Southwest Research Institute; report no. AR-804. Available from: NTIS, Springfield, VA;
29 PB-204012.
30
31 Hare, C. T.; Springer, K. J.; Somers, J. H.; Huls, T. A. (1974) Public opinion of diesel odor.
32
33 Hatch, G. E.; Boykin, E.; Graham, J. A.; Lewtas, J.; Pott, F.; Loud, K.; Mumford, J. L. (1985) Inhalable
34 particles and pulmonary host defense: in vivo and in vitro effects of ambient air and combustion
35 particles. Environ. Res. 36: 67-80.
36
37 Heinrich, U.; Peters, L.; Funcke, W.; Pott, F.; Mohr, U.; Stober, W. (1982) Investigation of toxic and
38 carcinogenic effects of diesel exhaust in long-term inhalation exposure of rodents. In: Lewtas, J., ed.
39 Toxicological effects of emissions from diesel engines: proceedings of the Environmental Protection
40 Agency diesel emissions symposium; October 1981; Raleigh, NC. New York, NY: Elsevier Biomedical;
41 pp. 225-242. (Developments in toxicology and environmental science: v. 10).
42
43 Heinrich, U.; Pott, F.; Mohr, U.; Stober, W. (1985) Experimental methods for the detection of the
44 carcinogenicity and/or cocarcinogenicity of inhaled polycyclic-aromatic-hydrocarbon-containing
45 emissions. Carcinogenesis (London) 8: 131-146.
46
47 Heinrich, U.; Muhle, H.; Takenaka, S.; Ernst, H.; Fuhst, R.; Mohr, U.; Pott, F.; Stober, W. (1986a) Chronic
48 effects on the respiratory tract of hamsters, mice, and rats after long-term inhalation of high
49 concentrations of filtered and unfiltered diesel engine emissions. J. Appl. Toxicol. 6: 383-395.
51
December 1994 5.93 DRAFT-DO NOT QUOTE OR CITE
-------
1 Heinrich, U.; Pott, F.; Rittinghausen, S. (1986b) Comparison of chronic inhalation effects in rodents after
2 long-term exposure to either coal oven flue gas mixed with pyrolized pitch or diesel engine exhaust.
3 In: Ishinishi, N.; Koizumi, A.; McClellan, R. O.; Stober, W., eds. Carcinogenic and mutagenic effects
4 of diesel engine exhaust: proceedings of the international satellite syposium on lexicological effects of
5 emissions from diesel engines; July; Tsukuba Science City, Japan. Amsterdam, Holland: Elsevier
6 Science Publishers B. V.; pp. 441-457. (Developments in toxicology and environmental science: v. 13).
7
8 Henderson, R. F.; Sun, J. D.; Jones, R. K.; Mauderly, J. L.; McClellan, R. O. (1983) Biochemical and
9 cytological response in airways of rodents exposed in life span studies to diluted diesel exhaust.
10 Am. Rev. Respir. Dis. 127: 164.
11
12 Henderson, R. F.; Pickrell, J. A.; Jones, R. K.; Sun, J. D.; Benson, J. M.; Mauderly, J. L.; McClellan, R. O.
13 (1988) Response of rodents to inhaled diluted diesel exhaust: biochemical and cytological changes in
14 bronchoalveolar lavage fluid and in lung tissue. Fundam. Appl. Toxicol. 11: 546-567.
15
16 Hinners, R. G.; Burkart, J. K.; Malanchuk, M. (1980) Facilities for diesel exhaust studies. In: Pepelko, W. E.;
17 Danner, R. M.; Clarke, N. A., eds. Health effects of diesel engine emissions: proceedings of an
18 international symposium; December 1979. Cincinnati, OH: U.S. Environmental Protection Agency,
19 Health Effects Research Laboratory; pp. 681-697; EPA report no. EPA-600/9-80-057b. Available from:
20 NTIS, Springfield, VA; PB81-173817.
21
22 Hyde, D.; Orthoefer, J.; Dungworth, D.; Tyler, W.; Carter, R.; Lum, H. (1978) Morphometric and
23 morphologic evaluation of pulmonary lesions in beagle dogs chronically exposed to high ambient levels
24 of air pollutants. Lab. Invest. 38: 455-469.
25
26 Hyde, D. M.; Plopper, C. G.; Weir, A. J.; Murnane, R. D.; Warren, D. L.; Last, J. A.; Pepelko, W. E.
27 (1985) Peribronchiolar fibrosis in lungs of cats chronically exposed to diesel exhaust. Lab. Invest.
28 52: 195-206.
29
30 Ingalls, M. N. (1985) Improved mobile source exposure estimation. Ann Arbor, MI: U.S. Environmental
31 Protection Agency, Emission Control Technology Division; EPA report no. EPA-460/3-85-002.
32 Available from: NTIS, Springfield, VA; PB85-248441.
33
34 Iwai, K.; Udagawa, T.; Yamagishi, M.; Yamada, H. (1986) Long-term inhalation studies of diesel exhaust on
35 F344 SPF rats. Incidence of lung cancer and lymphoma. In: Ishinishi, N.; Koizumi, A.; McClellan,
36 R. O.; Stober, W., eds. Carcinogenic and mutagenic effects of diesel engine exhaust: proceedings of the
37 international satellite symposium on lexicological effects of emissions from diesel engines; July;
38 Tsukuba Science Cily, Japan. Amsterdam, Holland: Elsevier Science Publishers B. V.; pp. 349-360.
39 (Developments in toxicology and environmenlal science: v. 13).
40
41 Jorgensen, H.; Svensson, A. (1970) Studies on pulmonary funclion and respiratory tract symptoms of workers
42 in an iron ore mine where diesel trucks are used underground. J. Occup. Med. 12: 348-354.
43
44 Kahn, G.; Orris, P.; Weeks, J. (1988) Acute overexposure to diesel exhaust: report of 13 cases. Am. J. Ind.
45 Med. 13: 405-406.
46
47 Kane, L. E.; Alarie, Y. (1978) Evaluation of sensory irritation from acrolein-formaldehyde mixtures. Am. Ind.
48 Hyg. Assoc. J. 39: 270-274.
49
50 Kane, L. E.; Alarie, Y. (1979) Interactions of sulfur dioxide and acrolein as sensory irritants. Toxicol. Appl.
51 Pharmacol. 48: 305-315.
52
53
December 1994 5.94 DRAFT-DO NOT QUOTE OR CITE
-------
1 Kaplan, H. L.; MacKenzie, W. F.; Springer, K. J.; Schreck, R. M.; Vostal, J. J. (1982) A subchronic study of
2 the effects of exposure of three species of rodents to diesel exhaust. In: Lewtas, J., ed. Toxicological
3 effects of emissions from diesel engines: proceedings of the Environmental Protection Agency diesel
4 emission symposium; October, 1981; Raleigh, NC. New York, NY: Elsevier Biomedical; pp. 161-182.
5 (Developments in toxicology and environmental science: v. 10).
6
7 Karagianes, M. T.; Palmer, R. F.; Busch, R. H. (1981) Effects of inhaled diesel emissions and coal dust in
8 rats. Am. Ind. Hyg. Assoc. J. 42: 382-391.
9
10 Katz, M.; Rennie, R. P.; Jegier, Z. (1960) Air pollution and associated health hazards from diesel locomotive
11 traffic in a railroad tunnel. Occup. Health Rev. 11: 2-15.
12
13 Kilburn, K. H.; McKenzie, W. N. (1978) Leukocyte recruitment to airways by aldehyde-carbon combinations
14 that mimic cigarette smoke. Lab. Invest. 38: 134-142.
15
16 Klosterkotter, W.; Bunemann, G. (1961) Animal experiments on the elimination of inhaled dust. In: Davies,
17 C. N., ed. Inhaled particles and vapours: proceedings of an international symposium; March-April
18 1960; Oxford, United Kingdom. New York, NY: Pergamon Press; pp. 327-341.
19
20 Last, J. A.; Gerriets, J. E.; Hyde, D. M. (1983) Synergistic effects on rat lungs of mixtures of oxidant air
21 pollutants (ozone or nitrogen dioxide) and respirable aerosols. Am. Rev. Respir. Dis. 128: 539-544.
22
23 Laurie, R. D.; Boyes, W. K. (1980) Neurophysiological alterations due to diesel exhaust exposure during the
24 neonatal life of the rat. In: Pepelko, W. E.; Danner, R. M.; Clarke, N. A., eds. Health effects of
25 diesel engine emissions: proceedings of an international symposium, v. 2; December 1979; Cincinnati,
26 OH. Cincinnati, OH: U.S. Environmental Protection Agency, Health Effects Research Laboratory;
27 pp. 713-727; EPA report no. EPA-600/9-80-057b. Available from: NTIS, Springfield, VA;
28 PB81-173817.
29
30 Laurie, R. D.; Boyes, W. K. (1981) Neurophysiological alterations due to diesel exhaust exposure during the
31 neonatal life of the rat. Environ. Int. 5: 363-368.
32
33 Laurie, R. D.; Lewkowski, J. P.; Cooper, G. P.; Hastings, L. (1978) Effects of diesel exhaust on behavior of
34 the rat. Presented at: 71st annual meeting of the Air Pollution Control Asscociation; June; Houston, TX.
35 Pittsburgh, PA: Air Pollution Control Association.
36
37 Laurie, R. D.; Boyes, W. K.; Wessendarp, T. (1980) Behavioral alterations due to diesel exhaust exposure.
38 In: Pepelko, W. E.; Danner, R. M.; Clarke, N. A., eds. Health effects of diesel engine emissions:
39 proceedings of an international symposium, v. 2; December 1979; Cincinnati, OH. Cincinnati, OH:
40 U.S. Environmental Protection Agency, Health Effects Research Laboratory; pp. 698-712; EPA report
41 no. EPA-600/9-80-057b. Available from: NTIS, Springfield, VA; PB81-173817.
42
43 Lee, S. D.; Campbell, K. I.; Laurie, D.; Hinners, R. G.; Malanchuk, M.; Moore, W.; Bhatnagar, R. J.;
44 Lee, I. (1978) Toxicological assessment of diesel emissions. Presented at: 71st annual meeting of the
45 Air Pollution Control Association; June; Houston, TX. Pittsburgh, PA: Air Pollution Control
46 Association; paper no. 78-33.4.
47
48 Lee, I. P.; Suzuki, K.; Lee, S. D.; Dixon, R. L. (1980) Aryl hydrocarbon hydroxylase induction in rat lung,
49 liver, and male reproductive organs following inhalation exposure to diesel emission. Toxicol. Appl.
50 Pharmacol. 52: 181-184.
51
52 Lee, K. P.; Trochimowicz, H. J.; Reinhardt, C. F. (1985) Pulmonary response of rats exposed to titanium
53 dioxide (TiO2) by inhalation for two years. Toxicol. Appl. Pharmacol. 79: 179-192.
54
December 1994 5.95 DRAFT-DO NOT QUOTE OR CITE
-------
1 Lee, K. P.; Kelly, D. P.; Schneider, P. W.; Trochimowicz, H. J. (1986) Inhalation toxicity study on rats
2 exposed to titanium tetrachloride atmospheric hydrolysis products for two years. Toxicol. Appl.
3 Pharmacol. 83: 30-45.
4
5 Lee, K. P.; Kelly, D. P.; O'Neal, F. O.; Stadler, J. C.; Kennedy, G. L., Jr. (1988) Lung response to ultrafme
6 Kevlar aramid synthetic fibrils following 2-year inhalation exposure in rats. Fundam. Appl. Toxicol.
7 11: 1-20.
8
9 Levins, P. L. (1981) Review of diesel odor and toxic vapor emissions. Washington, DC: U.S. Department of
10 Transportation, National Highway Traffic Safety Administration; report no. DOT-TSC-NHTSA-81-9.
11
12 Lewis, T. R.; Moorman, W. J. (1980) Pulmonary and cardiovascular physiology studies during exposure.
13 In: Stara, J. F.; Dungworth, D. L.; Orthoefer, J. G.; Tyler, W. S., eds. Long-term effects of air
14 pollutants: in canine species. Cincinnati, OH: U.S. Environmental Protection Agency, Office of Health
15 and Environmental Assessment, Environmental Criteria and Assessment Office; pp. 97-113; EPA report
16 no. EPA-600/8-80-014. Available from: NTIS, Springfield, VA; PB81-144875.
17
18 Lewis, T. R.; Campbell, K. I.; Vaughan, T. R., Jr. (1969) Effects on canine pulmonary function via induced
19 N02 impairment, paniculate interaction, and subsequent SOX. Arch. Environ. Health 18: 596-601.
20
21 Lewis, T. R.; Green, F. H. Y.; Moorman, W. J.; Burg, J. R.; Lynch, D. W. (1989) A chronic inhalation
22 toxicity study of diesel engine emissions and coal dust, alone and combined. J. Am. Coll. Toxicol.
23 8: 345-375.
24
25 Linnell, R. H.; Scott, W. E. (1962) Diesel exhaust composition and odor studies. J. Air Pollut. Control Assoc.
26 12: 510-515, 545.
27
28 Malanchuk, M. (1980) Exposure chamber atmospheres. Sampling and analysis. In: Stara, J. F.; Dungworth,
29 D. L.; Orthoefer, J. G.; Tyler, W. S., eds. Long-term effects of air pollutants: hi canine species.
30 Cincinnati, OH: U.S. Environmental Protection Agency, Office of Health and Environmental Criteria,
31 Environmental Criteria and Assessment Office; pp. 41-54; EPA report no. EPA-600/8-80-014.
32 Available from: NTIS, Springfield, VA; PB81-144875.
33
34 Mauderly, J. L.; Benson, J. M.; Bice, D. E.; Henderson, R. P.; Jones, R. K.; McClellan, R. O.; Mokler,
35 B. V.; Pickrell, J. A.; Redman, H. C.; Wolff, R. K. (1981) Observations on rodents exposed for
36 19 weeks to diluted diesel exhaust. In: Inhalation Toxicology Research Institute annual report
37 1980-1981. Albuquerque, NM: Lovelace Biomedical and Environmental Research Institute;
38 pp. 305-311; report no. LMF-91.
39
40 Mauderly, J. L.; Benson, J. M.; Bice, D. E.; Carpenter, R. L.; Evans, M. J.; Henderson, R. F.; Jones,
41 R. K.; McClellan, R. O.; Pickrell, J. A.; Redman, H. C.; Shami, S. G.; Wolff, R. K. (1983)
42 Life-span study of rodents inhaling diesel exhaust: results through 30 months. In: Inhalation Toxicology
43 Research Institute annual report 1982-1983. Albuquerque, NM: Lovelace Biomedical and Environmental
44 Research Institute; report nos. LMF-107.
45
46 Mauderly, J. L.; Benson, J. M.; Bice, D. E.; Carpenter, R. L.; Evans, M. J.; Henderson, R. F.; Jones,
47 R. K.; McClellan, R. O.; Pickrell, J. A.; Shami, S. G.; Wolff, R. K. (1984) Life-span study of rodents
48 inhaling diesel exhaust: effect on body weight and survival. Albuquerque, NM: Lovelace Biomedical
49 and Environmental Research Institute; report no. LMF-113; pp. 287-291.
50
51 Mauderly, J. L.; Bice, D. E.; Carpenter, R. L.; Gillett, N. A.; Henderson, R. F.; Pickrell, J. A.; Wolff,
52 R. K. (1987a) Effects of inhaled nitrogen dioxide and diesel exhaust on developing lung. Cambridge,
53 MA: Health Effects Institute; research report no. 8.
54
December 1994 5-96 DRAFT-DO NOT QUOTE OR CITE
-------
1 Mauderly, J. L.; Jones, R. K.; Griffith, W. C.; Henderson, R. F.; McClellan, R. O. (1987b) Diesel exhaust is
2 a pulmonary carcinogen in rats exposed chronically by inhalation. Fundam. Appl. Toxicol. 9: 208-221.
3
4 Mauderly, J. L.; Gillett, N. A.; Henderson, R. F.; Jones, R. K.; McClellan, R. O. (1988) Relationships of
5 lung structural and functional changes to accumulation of diesel exhaust particles. In: Dodgson, J.;
6 McCallum, R. I.; Bailey, M. R.; Fisher, D. R., eds. Inhaled particles VI: proceedings of an
7 international symposium and workshop on lung dosimetry; September 1985; Cambridge, United
8 Kingdom. Ann. Occup. Hyg. 32(suppl. 1): 659-669.
9
10 Mauderly, J. L.; Bice, D. E.; Cheng, Y. S.; Gillett, N. A.; Griffith, W. C.; Henderson, R. F.; Pickrell, J. A.;
11 Wolff, R. K. (1990a) Influence of preexisting pulmonary emphysema on susceptibility of rats to diesel
12 exhaust. Am. Rev. Respir. Dis. 141: 1333-1341.
13
14 Mauderly, J. L.; Cheng, Y. S.; Snipes, M. B. (1990b) Particle overload in toxicology studies: friend or foe and
15 how can we know? In: Proceedings of the particle lung interaction symposium; May; Rochester, NY.
16
17 McClellan, R. O.; Bice, D. E.; Cuddihy, R. G.; Gillett, N. A.; Henderson, R. F.; Jones, R. K.; Mauderly,
18 J. L.; Pickrell, J. A.; Shami, S. G.; Wolff, R. K. (1986) Health effects of diesel exhaust. In: Lee,
19 S. D.; Schneider, T.; Grant, L. D.; Verkerk, P. J., eds. Aerosols: research, risk assessment and
20 control strategies: proceedings of the second U.S.-Dutch international symposium; May 1985;
21 Williamsburg, VA. Chelsea, MI: Lewis Publishers, Inc.; pp. 597-615.
22
23 Meiss, R.; Robenek, H.; Schubert, M.; Themann, H.; Heinrich, U. (1981) Ultrastructural alterations in the
24 livers of golden hamsters following experimental chronic inhalation of diluted diesel exhaust emission.
25 Int. Arch. Occup. Environ. Health 48: 147-157.
26
27 Mentnech, M. S.; Lewis, D. M.; Olenchock, S. A.; Mull, J. C.; Roller, W. A.; Lewis, T. R. (1984) Effects
28 of coal dust and diesel exhaust on immune competence in rats. J. Toxicol. Environ. Health 13: 31-41.
29
30 Misiorowski, R. L.; Strom, K. A.; Vostal, J. J.; Chvapil, M. (1980) Lung biochemistry of rats chronically
31 exposed to diesel particulates. In: Pepelko, W. E.; Danner, R. M.; Clarke, N. A., eds. Health effects
32 of diesel engine emissions: proceedings of an international symposium; December 1979. Cincinnati,
33 OH: U.S. Environmental Protection Agency, Health Effects Research Laboratory; pp. 465-480; EPA
34 report no. EPA-600/9-80-057a. Available from: NTIS, Springfield, VA; PB81-173809.
35
36 Misiorowski, R. L.; Strom, K. S.; Vostal, J. J.; Tillema, L.; Chvapil, M. (1981) Collagen parameters in the
37 lung of rats chronically exposed to diesel particulates. Toxicologist 1: 75.
38
39 Moorman, W. J.; Clark, J. C.; Pepelko, W. E.; Mattox, J. (1985) Pulmonary fuction responses in cats
40 following long-term exposure to diesel exhaust. J. Appl. Toxicol. 5: 301-305.
41
42 Morrow, P. E. (1988) Possible mechanisms to explain dust overloading of the lungs. Fundam. Appl. Toxicol.
43 10: 369-384.
44
45 Navarro, C.; Charboneau, J.; McCauley, R. (1981) The effect of in vivo exposure to diesel exhaust on rat
46 hepatic and pulmonary microsomal activities. J. Appl. Toxicol. I: 124-126.
47
48 Nikula, K. J.; Snipes, M. B.; Barr, E. B.; Griffith, W. C.; Henderson, R. F.; Mauderly, J. L. (1994)
49 Influence of particle-associated organic compounds on the carcinogenicity of diesel exhaust.
50 In: Mohr, U.; Dungworth, D. L.; Mauderly, J. L.; Oberdorster, G., eds. Toxic and carcinogenic
51 effects of solid particles in the respiratory tract: [proceedings of the 4th international inhalation
52 symposium]; March 1993; Hannover, Germany. Washington, DC: International Life Sciences Institute
53 Press; pp. 565-568.
54
December 1994 5.97 DRAFT-DO NOT QUOTE OR CITE
-------
1 Oberdorster, G.; Yu, C. P. (1991) The carcinogenic potential of inhaled diesel exhaust: a particle effect?
2 J. Aerosol Sci. 21(suppl. 1): S397-S401.
3
4 Odor Panel of the CRC-APRAC Program Group on Composition of Diesel Exhaust. (1979) Development and
5 evaluation of a method for measurement of diesel exhaust odor using a diesel odor analysis system
6 (DOAS). New York, NY: Coordinating Research Council, Inc., Air Pollution Research Advisory
7 Committee; CRC-APRAC project no. CAPI-1-64.
8
9 Orthoefer, J. G.; Stara, J. F.; Yang, Y. Y.; Campbell, K. I. (1980) Clinical summary of beagle study. Physical
10 , examinations, ocular examinations, hematology examination (post-exposure) and blood chemistries.
11 In: Stara, J. F.; Dungworth, D. L.; Orthoefer, J. G.; Tyler, W. S., eds. Long-term effects of air
12 pollutants: in canine species. Cincinnati, OH: U.S. Environmental Protection Agency, Office of Health
13 and Environmental Criteria, Environmental Criteria and Assessment Office; pp. 55-70; EPA report no.
14 EPA-600/8-80-014. Available from: NTIS, Springfield, VA; PB81-144875.
15
16 Pattle, R. E.; Stretch, H.; Burgess, F.; Sinclair, K.; Edginton, J. A. G. (1957) The toxicity of fumes from a
17 diesel engine under four different running conditions. Br. J. Ind. Med. 14: 47-55.
18
19 Penney, D. G.; Baylerian, M. S.; Fanning, K. E.; Thill, J. E.; Yedavally, S.; Fanning, C. M. (1981) A study
20 of heart and blood of rodents inhaling diesel engine exhaust particulates. Environ. Res. 26: 453-462.
21
22 Pepelko, W. E. (1982a) Effects of 28 days exposure to diesel engine emissions in rats. Environ. Res.
23 27: 16-23.
24
25 Pepelko, W. E. (1982b) EPA studies on the lexicological effects of inhaled diesel engine emissions.
26 In: Lewtas, J., ed. Toxicological effects of emissions from diesel engines: proceedings of the
27 Environmental Protection Agency diesel emissions symposium; October 1981; Raleigh, NC. New York,
28 NY: Elsevier Biomedical; pp. 121-142. (Developments in toxicology and environmental science: v. 10).
29
30 Pepelko, W. E.; Peirano, W. B. (1983) Health effects of exposure to diesel engine emissions: a summary of
31 animal studies conducted by the U.S. Environmental Protection Agency's Health Effects Research
32 Laboratories at Cincinnati, Ohio. J. Am. Coll. Toxicol. 2: 253-306.
33
34 Pepelko, W. E.; Mattox, J. K.; Yang, Y. Y.; Moore, W., Jr. (1980a) Pulmonary function and pathology in
35 cats exposed 28 days to diesel exhaust. J. Environ. Pathol. Toxicol. 4: 449-458.
36
37 Pepelko, W. E.; Mattox, J.; Moorman, W. J.; Clark, J. C. (1980b) Pulmonary function evaluation of cats after
38 one year of exposure to diesel exhaust. In: Pepelko, W. E.; Danner, R. M.; Clarke, N. A., eds. Health
39 effects of diesel engine emissions: proceedings of an international symposium, v. 2; December 1979;
40 Cincinnati, OH. Cincinnati, OH: U.S. Environmental Protection Agency, Health Effects Research
41 Laboratory; pp. 757-765; EPA report no. EPA-600/9-80-057b. Available from: NTIS, Springfield, VA;
42 PB81-173817.
43
44 Pepelko, W. E.; Mattox, J.; Moorman, W. J.; Clark, J. C. (1981) Pulmonary function evaluation of cats after
45 one year of exposure to diesel exhaust. Environ. Int. 5: 373-376.
46
47 Pereira, M. A.; Sabharwal, P. S.; Kaur, P.; Ross, C. B.; Choi, A.; Dixon, T. (1981a) In vivo detection of
48 mutagenic effects of diesel exhaust by short-term mammalian bioassays. Environ. Int. 5: 439-443.
49
50 Pereira, M. A.; Sabharwal, P. S.; Gordon, L.; Wyrobek, A. J. (1981b) The effect of diesel exhaust on
51 sperm-shape abnormalities in mice. Environ. Int. 5: 459-460.
52
53 Plopper, C. G.; Hyde, D. M.; Weir, A. J. (1983) Centriacinar alterations in lungs of cats chronically exposed
54 to diesel exhaust. Lab. Invest. 49: 391-399.
December 1994 5-98 DRAFT-DO NOT QUOTE OR CITE
-------
1 Purdham, J. T.; Holness, D. L.; Pilger, C. W. (1987) Environmental and medical assessment of stevedores
2 employed in ferry operations. Appl. Ind. Hyg. 2: 133-139.
3
4 Quinto, I.; De Marinis, E. (1984) Sperm abnormalities in mice exposed to diesel paniculate. Mutat. Res.
5 130: 242.
6
7 Rabovsky, J.; Petersen, M. R.; Lewis, T. R.; Marion, K. J.; Groseclose, R. D. (1984) Chronic inhalation of
8 diesel exhaust and coal dust: effect of age and exposure on selected enzyme activities associated with
9 microsomal cytochrome P-450 in rat lung and liver. J. Toxicol. Environ. Health 14: 655-666.
10
11 Rabovsky, J.; Judy, D. J.; Rodak, D. J.; Petersen, M. (1986) Influenza virus-induced alterations of cytochrome
12 P-450 enzyme activities following exposure of mice to coal and diesel particulates. Environ. Res.
13 40: 136-144.
14
15 Reger, R. (1979) Ventilatory function changes over a work shift for coal miners exposed to diesel emissions.
16 In: Bridbord, K.; French, J., eds. lexicological and carcinogenic health hazards in the workplace:
17 proceedings of the first annual NIOSH scientific symposium; 1978; Cincinnati, OH. Park Forest South,
18 IL: Pathotox Publishers, Inc.,; pp. 346-347.
19
20 Reger, R.; Hancock, J.; Hankinson, J.; Hearl, F.; Merchant, J. (1982) Coal miners exposed to diesel exhaust
21 emissions. Ann. Occup. Hyg. 26: 799-815.
22
23 Research Committee for HERP Studies. (1988) Diesel exhaust and health risks: results of the HERP studies.
24 Tsukuba, Ibaraki, Japan: Japan Automobile Research Institute, Inc.
25
26 Schneider, D. R.; Felt, B. T. (1981) Effect of diesel paniculate exposure on adenylate and guanylate cyclase of
27 rat and guinea pig liver and lung. J. Appl. Toxicol. 1: 135-139.
28
29 Schreck, R. M.; Soderholm, S. C.; Chan, T. L.; Hering, W. E.; D'Arcy, J. B.; Smiler, K. L. (1980)
30 Introduction to and experimental conditions in the GMR chronic inhalation studies of diesel exhaust.
31 In: Pepelko, W. E.; Danner, R. M.; Clarke, N. A., eds. Health effects of diesel engine emissions:
32 proceedings of an international symposium; December 1979. Cincinnati, OH: U.S. Environmental
33 Protection Agency, Health Effects Research Laboratory; pp. 573-591; EPA report no.
34 EPA-600/9-80-057b. Available from: NTIS, Springfield, VA; PB81-173817.
35
36 Schreck, R. M.; Soderholm, S. C.; Chan, T. L.; Smiler, K. L.; D'Arcy, J. B. (1981) Experimental conditions
37 in GMR chronic inhalation studies of diesel exhaust. J. Appl. Toxicol. 1: 67-76
38
39 Stara, J. F.; Dungworth, D. L.; Orthoefer, J. G.; Tyler, W. S., eds. (1980) Long-term effects of air pollutants:
40 in canine species. Cincinnati, OH: U.S. Environmental Protection Agency, Office of Health and
41 Environmental Assessment, Environmental Criteria and Assessment Office; EPA report no.
42 EPA-600/8-80-014. Available from: NTIS, Springfield, VA; PB81-144875.
43
44 Stephens, R. J.; Freeman, G.; Evans, M. J. (1972) Early response of lungs to low levels of nitrogen dioxide:
45 light and electron microscopy. Arch. Environ. Health 24: 160-179
46
47 Stober, W. (1986) Experimental induction of tumors in hamsters, mice and rats after long-term inhalation of
48 filtered and unfiltered diesel engine exhaust. In: Ishinishi, N.; Koizumi, A.; McClellan, R. O.;
49 Stober, W., eds. Carcinogenic and mutagenic effects of diesel engine exhaust: proceedings of the
50 international satellite symposium on lexicological effects of emissions from diesel engines; July;
51 Tsukuba Science City, Japan. Amsterdam, The Netherlands: Elsevier Science Publishers B. V.;'
52 pp. 421-439. (Developments in toxicology and environmental science: v. 13).
53
December 1994 5.99 DRAFT-DO NOT QUOTE OR CITE
-------
1 Stokinger, H. E. (1977) Routes of entry and modes of action. In: Key, M. M.; Henschel, A. F.; Butler, J.;
2 Ligo, R. N.; Tabershaw, I. R.; Ede, L., eds. Occupational diseases: a guide to their recognition.
3 Washington, DC: U.S. Department of Health, Education, and Welfare, National Institute for
4 Occupational Safety and Health; pp. 11-42; DHEW (NIOSH) publication no. 77-181.
5
6 Strom, K. A. (1984) Response of pulmonary cellular defenses to the inhalation of high concentrations of diesel
7 exhaust. J. Toxicol. Environ. Health 13: 919-944.
8
9 Takafuji, S.; Suzuki, S.; Koizumi, K.;Tadokoro, K.; Miyamoto, T.; Ikemori, R.; Muranaka, M. (1987)
10 Diesel-exhaust particulates inoculated by the intranasal route have an adjuvant activity for IgE
11 production in mice. J. Allergy Clin. Immunol. 79: 639-645.
12
13 Turk, A. (1967) Selection and training of judges for sensory evaluation of the intensity and character of diesel
14 exhaust odors. Cincinnati, OH: U.S. Department of Health, Education, and Welfare, National Center
15 for Air Pollution Control; Public Health Service publication no. 999-AP-32. Available from: NTIS,
16 Springfield, VA; PB-174707.
17
18 U.S. Environmental Protection Agency. (1983) Diesel paniculate study. Ann Arbor, MI: Office of Mobile
19 Sources.
20
21 Ulfvarson, U.; Alexandersson, R.; Aringer, L.; Svensson, E.; Hedenstierna, G.; Hogstedt, C.; Holmberg, B.;
22 Rosen, G.; Sorsa, M. (1987) Effects of exposure to vehicle exhaust on health. Scand. J. Work Environ.
23 Health 13: 505-512.
24
25 Vallyathan, V.; Virmani, R.; Rochlani, S.; Green, F. H. Y.; Lewis, T. (1986) Effect of diesel emissions and
26 coal dust inhalation on heart and pulmonary arteries of rats. J. Toxicol. Environ. Health 19: 33-41.
27
28 Vinegar, A.; Carson, A. I.; Pepelko, W. E. (1980) Pulmonary function changes in Chinese hamsters exposed
29 six months to diesel exhaust. In: Pepelko, W. E.; Danner, R. M.; Clarke, N. A., eds. Health effects of
30 diesel engine emissions: proceedings of an international symposium, v. 2; December, 1979; Cincinnati,
31 OH. Cincinnati, OH: U.S. Environmental Protection Agency, Health Effects Research Laboratory;
32 pp. 749-756; EPA report no. EPA-600/9-80-057b. Available from: NTIS, Springfield, VA;
33 PB81-173817
34
35 Vinegar, A.; Carson, A.; Pepelko, W. E.; Orthoefer, J. O. (1981a) Effect of six months of exposure to two
36 levels of diesel exhaust on pulmonary function of Chinese hamsters. Fed. Proc. 40: 593.
37
38 Vinegar, A.; Carson, A.; Pepelko, W. E. (1981b) Pulmonary function changes in Chinese hamsters exposed six
39 months to diesel exhaust. Environ. Int. 5: 369-371.
40
41 Vostal, J. J.; Chan, T. L.; Garg, B. D.; Lee, P. S.; Strom, K. A. (1981) Lymphatic transport of inhaled diesel
42 particles in the lungs of rats and guinea pigs exposed to diluted diesel exhaust. Environ. Int. 5: 339-347.
43
44 Vostal, J. J.; White, H. J.; Strom, K. A.; Siak, J.-S.; Chen, K.-C.; Dziedzic, D. (1982) Response of the
45 pulmonary defense system to diesel paniculate exposure. In: Lewtas, J., ed. Toxicological effects of
46 emissions from diesel engines: proceedings of the Environmental Protection Agency diesel emissions
47 symposium; October 1981; Raleigh, NC. New York, NY: Elsevier Biomedical; pp. 201-221.
48 (Developments in toxicology and environmental science: v. 10).
49
50 Wallace, M. A.; Salley, S. O.; Barnhart, M. I. (1987) Analysis of the effects of inhaled diesel exhaust on the
51 alveolar intravascular and interstitial cellular components of rodent lungs. Scanning Microsc.
52 1: 1387-1395.
53
December 1994 5-100 DRAFT-DO NOT QUOTE OR CITE
-------
1 Waller, R. E.; Commins, B. T.; Lawther, P. J. (1961) Air pollution in road tunnels. Br. J. Ind. Med.
2 18: 250-259.
3
4 Wehner, A. P.; Dagle, G. E.; Clark, M. L.; Buschbom, R. L. (1986) Lung changes in rats following inhalation
5 exposure to volcanic ash for two years. Environ. Res. 40: 499-517.
6
7 Weller, M. A.; Chen, S.-T.; Barnhart, M. I. (1981) Acid phosphatase in alveolar macrophages exposed in vivo
8 to diesel engine exhaust. Micron 12: 89-90.
9
10 Werchowski, K. M.; Chaffee, V. W.; Briggs, G. B. (1980a) Teratologic effects of long-term exposure to diesel
11 exhaust emissions (rats). Cincinnati, OH: U.S. Environmental Protection Agency, Health Effects
12 Research Laboratory; EPA report no. EPA-600/1-80-010. Available from: NTIS, Springfield, VA;
13 PB80-159965.
14
15 Werchowski, K. M.; Henne, S. P.; Briggs, G. B. (1980b) Teratologic effects of long-term exposure to diesel
16 exhaust emissions (rabbits). Cincinnati, OH: U.S. Environmental Protection Agency, Health Effects
17 Research Laboratory; EPA report no. EPA-600/1-80-011. Available from: NTIS, Springfield, VA;
18 PB80-168529.
19
20 White, H. J.; Garg, B. D. (1981) Early pulmonary response of the rat lung to inhalation of high concentration
21 of diesel particles. J. Appl. Toxicol. 1: 104-110.
22
23 Wiester, M. J.; Iltis, R.; Moore, W. (1980) Altered function and histology in guinea pigs after inhalation of
24 diesel exhaust. Environ. Res. 22: 285-297.
25
26 Wolff, R. K.; Gray, R. L. (1980) Tracheal clearance of particles. In: Diel, J. H.; Bice, D. E.; Martinez,
27 B. S., eds. Inhalation Toxicology Research Institute annual report: 1979-1980. Albuquerque, NM:
28 Lovelace Biomedical and Environmental Research Institute; p. 252; report no. LMF-84.
29
30 Wolff, R. K.; Henderson, R. F.; Snipes, M. B.; Griffith, W. C.; Mauderly, J. L.; Cuddihy, R. G.; McClellan,
31 R. O. (1987) Alterations in particle accumulation and clearance in lungs of rats chronically exposed to
32 diesel exhaust. Fundam. Appl. Toxicol. 9: 154-166.
33
34 Wright, E. S. (1986) Effects of short-term exposure to diesel exhaust on lung cell proliferation and phospholipid
35 metabolism. Exp. Lung Res. 10: 39-55.
December 1994 5_10i DRAFT-DO NOT QUOTE OR CITE
-------
i 6. QUALITATIVE AND QUANTITATIVE
2 ASSESSMENT OF NONCANCER HEALTH
3 EFFECTS-DERIVATION OF THE INHALATION
4 REFERENCE CONCENTRATION
5
6
7 6.1 INTRODUCTION
8 Noncancer endpoints have been studied in detail in controlled animal studies of diesel
9 exhaust, and the progression of events from initial particle deposition through chronic
10 structural and functional alterations have been described. Some of these effects are seen
11 early in the course of a lifetime exposure and progress throughout the lifetime of the animal
12 in the absence of a tumor response. These findings raise the possibility of noncancer
13 respiratory disease as a human health hazard of long-term exposure to diesel exhaust. This
14 chapter presents a qualitative and quantitative assessment of the toxicological data on
15 noncancer endpoints for diesel emissions.
16 The quantitative assessment of noncancer health effects from exposure to diesel exhaust
17 emissions involves the development of an inhalation reference concentration (RfC). An RfC
18 is defined as an estimate (with uncertainty spanning perhaps an order of magnitude) of a
19 continuous inhalation exposure to the human population (including sensitive subgroups) that
20 is likely to be without appreciable risks of deleterious noncancer effects during a lifetime.
21 The RfC approach is based on the assumption that a threshold exists for the human
22 population below which no effect will occur. The RfC is an estimate of a likely subthreshold
23 concentration. To derive the RfC, the database on toxicological effects is reviewed and the
24 most relevant and sensitive endpoints for human risk assessment are identified. The lowest-
25 observed-adverse-effect level (LOAEL, the lowest concentration producing an adverse
26 effect), or the no-observed-adverse-effect level (NOAEL, the highest concentration which did
27 not produce any adverse effect), is used as the basis for derivation of the RfC. The NOAEL
28 for the data base (or LOAEL) is selected after calculation of the human equivalent
29 concentration for the exposure regimens used in the experimental studies. The NOAEL is
30 considered to be an operational estimate of a subthreshold exposure. The human equivalent
31 concentration of the NOAEL is then divided by the uncertainty factors to account for any
December 1994 6_1 DRAFT-DO NOT QUOTE OR CITE
-------
1 uncertainties or data gaps in order to extrapolate from the experimental conditions to a
2 no-adverse-effect level in a chronically and continuously exposed sensitive human.
3 The study or studies identifying the LOAEL and/or NOAEL selected as the basis for
4 derivation of the RfC are termed the principal study(ies). The principal studies are selected
5 as those that identify the threshold region of the concentration-response curve and are
6 representative of the entire data base in this regard. Other studies which are pertinent to
7 identifying the threshold for the effect are termed supporting studies. Supporting studies may
8 provide additional evidence identifying the concentration-response relationship, the relative
9 sensitivity of various effects or species, or the occurrence of other noncancer endpoints, such
10 as reproductive or developmental toxicity. Principal and supporting studies used in the
11 derivation of the RfC for diesel engine emissions are discussed in Sections 6.4 and 6.5,
12 respectively, and the derivation of the RfC is discussed in Section 6.6. The verified RfC for
13 diesel emissions is available on the U.S. Environmental Protection Agency's (EPA's)
14 Integrated Risk Information System (IRIS).
15
16
17 6.2 QUALITATIVE EVALUATION OF DIESEL EXHAUST
18 EMISSIONS
19 The noncarcinogenic effects of inhalation of diesel exhaust have been studied in many
20 chronic and subchronic experiments in several laboratory animal species (Chapter 5). The
21 pathogenic sequence following the inhalation of diesel exhaust as determined
22 histopathologically and biochemically begins with the phagocytosis of diesel particles by
23 alveolar macrophages. These activated macrophages release chemotactic factors that attract
24 neutrophils and additional alveolar macrophages. As the lung burden of diesel particles
25 increases, there are aggregations of particle-laden alveolar macrophages in alveoli adjacent to
26 terminal bronchioles, increases in the number of Type II cells lining particle-laden alveoli,
27 and the presence of particles within alveolar and peribronchial interstitial tissues and
28 associated lymph nodes. The neutrophils and macrophages release mediators of inflammation
29 and oxygen radicals, and particle-laden macrophages are functionally altered, resulting in
30 decreased viability and impaired phagocytosis and clearance of particles. The latter series of
31 events may result in the presence of pulmonary inflammatory, fibrotic, or emphysematous
December 1994 6-2 DRAFT-DO NOT QUOTE OR CITE
-------
1 lesions. Studies showing these effects were described in Chapter 5. Epidemiological studies
2 of people exposed in various occupations in which diesel engines are used provide suggestive
3 evidence for a respiratory effect. Although detailed information describing the pathogenesis
4 of respiratory effects in humans is lacking, the effects in human studies lend qualitative
5 support to the findings in controlled animal studies.
6 The weight-of-evidence from the available toxicological data on diesel exhaust indicates
7 with high confidence that inhalation of diesel exhaust can be a respiratory hazard, based on
8 findings in multiple controlled animal studies in several species with suggestive evidence
9 from human occupational studies. The endpoints of concern include biochemical,
10 histological, and functional changes in the pulmonary and tracheobronchial regions. There is
11 also some evidence for effects on respiratory system related immune function. Although
12 there is some suggestive evidence of liver and kidney changes in animals exposed to diesel
13 exhaust, these data are inadequate to indicate that a hazard exists for these endpoints. Study
14 of other endpoints, including reproductive and developmental toxicity, in controlled animal
15 exposures have shown no evidence of potential hazard.
16
17
18 6.3 APPROACH FOR DERIVATION OF THE INHALATION
19 REFERENCE CONCENTRATION
20 A total of 10 different long-term (> 1 year) animal inhalation studies of diesel engine
21 emissions have been conducted. The focus of these studies has been on the respiratory tract
22 effects in the pulmonary region. Effects in the upper respiratory tract and in other organs
23 were not found consistently in chronic animal exposures. The research programs on the
24 toxicology of diesel emissions at the Inhalation Toxicology Research Institute (ITRI) and the
25 Japanese Health Effects Research Program (HERP) consisted of large-scale chronic
26 exposures, with exposed animals being designated for the study of various endpoints and at
27 various time points (Ishinishi et al., 1986, 1988; Mauderly et al., 1987a,b, 1988; Henderson
28 et al., 1988; Wolff et al., 1987; Nikula et al., 1991). Each research program is represented
29 by multiple published accounts of results. These research programs were selected as the
30 principal basis for the derivation of the RfC because each contains studies which identify a
31 LOAEL and a NOAEL for respiratory effects after chronic exposure.
December 1994 6.3 DRAFT-DO NOT QUOTE OR CITE
-------
1 Four chronic laboratory animal studies have been performed to compare the effects of
2 exposure to whole exhaust with the effects of filtered exhaust containing no particles
3 (Heinrich et al., 1982, 1986; Iwai et al., 1986; Brightwell et al., 1986; see Section 5.2).
4 These studies demonstrate that when the exhaust is sufficiently diluted to limit the
5 concentrations of gaseous irritants (NO2 and SO2), irritant vapors (aldehydes), CO, or other
6 systemic toxicants, the diesel particles are the prime etiologic agents of noncancer health
7 effects. The whole diesel exposures produced changes in the lung that are much more
8 prominent than those evoked by the gas phase alone. Such marked differences between
9 whole and filtered diesel exhaust are evident from general toxicological indices, such as
10 decreases in body weight and increases in lung weights, pulmonary function measurements,
11 and pulmonary histopathology. Based on these results, the derivation of the RfC is based on
12 the dose of the particles to the lung surface.
13 Diesel paniculate matter is composed of an insoluble carbon core with a surface coating
14 of relatively soluble organic constituents. Since macrophage accumulation, epithelial
15 histopathology, and reduced clearance have been observed in rodents exposed to high
16 concentrations of chemically inert particles (Morrow, 1992), it appears possible that the
17 toxicity of diesel particles results from the carbon core rather than the associated organics.
18 However, the organic component of diesel particles, consisting of a large number of
19 poly cyclic aromatic hydrocarbons and heterocyclic compounds and their derivatives (Chapter
20 3), may also play a role in the pulmonary toxicity of diesel particles. It is not possible to
21 separate the carbon core from the adsorbed organics in order to compare the toxicity. As an
22 approach to this question, studies were performed at the Lovelace Inhalation Toxicology
23 Research Institute as well as the Fraunhofer Institute of Toxicology and Aerosol Research in
24 which rats were exposed to either diesel exhaust or to carbon black, an inert analog of the
25 carbon core of diesel particles. Rats were exposed for 16 h/day, 5 days/week, for up to
26 24 mo to particle concentrations of either 2.5 or 6.5 mg/m3 in the Lovelace study (Nikula
27 et al., 1991, 1994), 17 to 18 h/day, 5 days/week, for 10 or 20 mo at concentrations of
28 6 mg/m3 in the Fraunhofer study (Dungworth et al., 1994). This study has not been fully
29 reported; a preliminary report is provided in the ITRI Annual Report (Nikula et al., 1991).
30 Although the study is primarily concerned with the role of particle-associated organics in the
31 carcinogenicity of diesel exhaust, non-neoplastic effects are also mentioned. According to
December 1994 6-4 DRAFT-DO NOT QUOTE OR CITE
-------
1 these preliminary reports, both diesel exhaust and carbon black exposure resulted in
2 macrophage hyperplasia, epithelial hyperplasia, bronchiolar-alveolar metaplasia, and focal
3 fibrosis. The reports state that the number and intensity of the lesions seems to correspond
4 to the exposure duration and concentration and that the morphological characteristics of the
5 lesions were similar in the animals exposed to diesel and to carbon black. These preliminary
6 results suggest that the chronic noncancer effects of diesel exhaust exposure are caused by
7 the persistence of the insoluble carbon core of the particles, rather than by the extractable
8 organic layer. On this basis, the human equivalent concentration used for derivation of the
9 RfC is calculated as the exposure concentration at which humans achieve the same retained
10 mass of the carbon core per unit of pulmonary surface area as the laboratory species
11 exhibiting an adverse effect. A more complete analysis of this conclusion will await
12 publication of the complete results for the ITRI and Fraunhofer studies.
13 Using the data on deposition and retention of diesel particles in animals as well as
14 theoretical and empirical information on human deposition and retention of inhaled particles,
15 a mathematical model has been developed which accounts for these processes and can be
16 used to extrapolate between species (Yu and Yoon, 1990). This model is discussed in
17 Chapter 4 and Appendix C. The retention model takes into account the retardation of
18 particle clearance due to the particle overload effect. Assuming that the long-term retained
19 dose must be the same in rat and in humans to induce the same effect, a deposited dose for
20 the human lung can be calculated from the retained dose applying human-specific retention
21 half-tunes to arrive at the human equivalent concentration (HEC). The retention model used
22 by Yu and Yoon (1990) includes the three-compartment lung respiratory tract model, with
23 additional physiologically-based compartments that describe the blood, lymphatics, and
24 gastrointestinal tract. Transport due to dissolution of the organic phase is assumed to be
25 constant. All other transport processes are modeled using first order rate constants, with the
26 exception of the mechanical clearance of the carbon core from the alveolar compartment.
27 Several studies have shown that mechanical transport of diesel particles from the alveolar
28 compartment varies with particle lung burden after lung burden reaches a certain level. The
29 functional dependence of mechanical alveolar clearance rates on particle lung burden used in
30 the model was determined by fitting the experimental data in rats. The model
December 1994 6_5 DRAFT-DO NOT QUOTE OR CITE
-------
1 mathematically describes deposition and transport of the three particle components (carbon
2 core, weakly bound organics, and strongly bound organics) between compartments.
3 The use of a specific retention or physiologically based pharmacokinetic model is
4 considered the optimum method for RfC derivation and default approaches are described for
5 chemicals without applicable models. Consistent with this approach, the calculation of the
6 HEC for diesel particles is based on the model described above. The HEC calculation is
7 based on the assumption that the equivalent dose metric across species is the retained mass of
8 insoluble carbon core per surface area in the alveolar (pulmonary) region. Because the
9 dependence of mechanical alveolar clearance on particle lung burden in humans is not
10 known, it was assumed in development of the model for humans that the particle overload
11 phenomenon occurs in humans and in rats at equivalent lung burdens expressed as mass per
12 unit surface area. This assumption allows for the development of a diesel particle-specific
13 human retention model and therefore allows extrapolation from the rat studies to human
14 exposures. The model has not been extended to other species at this time because data
15 describing the dependence of the particle overload phenomenon on lung particle burden for
16 species other than the rat are not available.
17 The input data required to run the dosimetric model include the particle size
18 characterization expressed a mass median aerodynamic diameter (MMAD) and the geometric
19 standard deviation (sigma g). In the principal and supporting studies used for the RfC
20 derivation, these parameters are measured using different methods and are reported in
21 different levels of detail. Simulation data presented by Yu and Xu (1986) show that across a
22 range of MMAD and sigma g inclusive of the values reported in these studies, the pulmonary
23 deposition fraction differs by no more that 20%. The minimal effect of even a large
24 distribution of particle size on deposition probably results because the particles are still
25 mostly in the submicron range and deposition is influenced primarily by diffusion. However,
26 it has also been shown that the particle characteristics in a diesel exhaust exposure study
27 depend very much on the procedures used for generation of the chamber atmosphere.
28 Especially important are the volume and temperature of the dilution gas because of the rapid
29 coagulation of particles. The difference reported in particle sizes and distributions in various
30 studies likely reflected real differences in the exposure chambers as well as different
31 analytical methods. Since the particle diameter and size distribution were not reported in the
December 1994 6-6 DRAFT-DO NOT QUOTE OR CITE
-------
1 two lowest exposure concentrations in the HERP studies, it was decided to use a default
2 particle size of MMAD = 0.2 fjm and sigma g = 2.3 for modeling of lung burden. For
3 consistency, the lung burdens for the other studies were also calculated using the default
4 particle size assumption. The difference in the human equivalent concentration using the
5 default particle size compared with the actual reported particle size is no more than 4% in
6 the HERP study and 19% in the ITRI study.
7
8
9 6.4 THE PRINCIPAL STUDIES FOR INHALATION REFERENCE
10 CONCENTRATION DERIVATION
11 The experimental protocol and results for the principal studies are discussed in Chapter
12 5 and Appendix A and are briefly reviewed here. In studies conducted at ITRI, rats and
13 mice were exposed to target diesel particulate concentrations of 0, 0.35, 3.5, or 7 mg/m3 for
14 7 h/day, 5 days/week for up to 30 mo (rats) or 24 mo (mice) (Mauderly et al., 1988).
15 A total of 364 to 367 rats per exposure level were exposed and used for various studies
16 examining different endpoints including carcinogenicity, respiratory tract histopathology and
17 morphometric analysis, particle clearance, lung burden of diesel particulate matter,
18 pulmonary function testing, lung biochemistry, lung lavage biochemistry and cytology,
19 immune function, and lung cell labeling index. Subsets of animals were examined at 6, 12,
20 18, and 24 mo of exposure and surviving rats were examined at 30 mo. Diesel emissions
21 from a 5.7-L engine operated on a Federal Test Procedure urban driving cycle were diluted
22 and fed into the exposure chambers. Particle concentrations were measured daily using a
23 filter sample and weekly grab samples were taken for measurement of gaseous components
24 including carbon monoxide, carbon dioxide, nitrogen oxides, ammonia, and hydrocarbons.
25 The actual particle concentration for the low, medium, and high exposure levels were 0.353,
26 3.47, and 7.08 mg/m3, respectively. Mass median diameter (geometric standard deviations)
27 determined using a impactor/parallel flow diffusion battery were 0.262 (4.2), 0.249 (4.5),
28 and 0.234 (4.4) for the low, medium, and high exposure groups, respectively.
29 Lung wet weight to dry weight ratio was increased significantly in the two highest
30 exposure groups. Qualitative descriptions of the histological results in the respiratory tract
31 are found in Mauderly et al. (1987 and 1988), Henderson et al. (1988), and McClellan et al.
December 1994 5.7 DRAFT-DO NOT QUOTE OR CITE
-------
1 (1986). Aggregates of particle-laden macrophages were seen after 6 mo in rats exposed to
2 7-mg/m3 target concentrations, and after 1 year of exposure histological changes were seen,
3 including focal areas of epithelial metaplasia. Fibrosis and metaplasia increased with
4 increasing duration of exposure and were observable in the 3.5 and 7 mg/m3 groups of rats
5 at 24 mo. Changes in the epithelium included extension of bronchiolar cell types into the
6 alveoli. Focal thickening of the alveolar septa was also observed. Histological effects were
7 seen in areas near aggregations of particle laden macrophages. The severity of inflammatory
8 responses and fibrosis was directly related to the exposure level. In the 0.35-mg/m3 group
9 of rats, there was no inflammation or fibrosis. Although the mouse lungs contained higher
10 lung burdens of diesel particles per gram of lung weight at each equivalent exposure
11 concentration, there was substantially less inflammatory reaction and fibrosis than was the
12 case in rats. Fibrosis was observed only in the lungs of mice exposed at 7 mg/m3 and
13 consisted of fine fibrillar thickening of an occasional alveolar septa.
14 Groups of 16 rats and mice (8/sex) were subjected to bronchoalveolar lavage after 6,
15 12, 18, and 24 (rats only) mo of exposure (Henderson et al., 1988). Lung wet weights were
16 increased at 7 mg/m3 in mice and rats at all time points and in mice at 3.5 mg/m3 at all time
17 points after 6 mo. An increase in lavagable neutrophils, indicating an inflammatory response
18 in the lung, was seen at 3.5 and 7 mg/m3 in rats and mice at most time points. An increase
19 in protein content of the bronchoalveolar lavage fluid was observed in rats exposed to 3.5 or
20 7 mg/m3 at 12 and 18 mo but not at 24 mo. Increased protein content was also seen in mice
21 at the two higher concentrations at all time points. Increases in lavage fluid content of
22 lactate dehydrogenase, glutathione reductase, beta-glucuronidase, glutathione, and
23 hydroxyproline were observed in rats and mice exposed to 3.5 or 7 mg/m3 at various time
24 points. At the lowest exposure level, no biochemical or cytological changes occurred in the
25 lavage fluid or in lung tissue in either Fischer 344 rats or CD-I mice.
26 Mauderly et al. (1988; see also McClellan, et al. 1986) examined the impairment of
27 respiratory function in rats exposed according to the protocol described above. After 24 mo
28 of exposure to 7 mg/m3 paniculate matter, mean TLC, Cdyn, quasistatic chord compliance,
29 and CO diffusing capacity were significantly lower than control values and nitrogen washout
30 and percentage of FVC expired in 0.1 s were significantly greater than control values. There
31 was no evidence of airflow obstruction. Similar functional alterations were observed in the
December 1994 6-8 DRAFT-DO NOT QUOTE OR CITE
-------
1 rats exposed to 3.5 mg/m3 paniculate matter, but such changes usually occurred later in the
2 exposure period and were generally less pronounced. There were no significant decrements
3 in pulmonary function for the 0.35-mg/m3 group at any time during the study.
4 Wolff et al. (1987) investigated alterations in particle clearance from the lungs of rats
5 in the ITRI study. Progressive increases in lung burdens were observed over tune in the
6 3.5- and 7.0-mg/m3 exposure groups. There were significant increases in 16-day clearance
7 half-times of inhaled radiolabeled particles of gallium oxide (0.1 um, MMD) as early as
8 6 mo at the 7.0-mg/m3 level and 18 mo at the 3.5-mg/m3 level; no significant changes were
9 seen at the 0.35-mg/m3 level. Rats that inhaled fused aluminosilicate particles (2 pm
10 MMAD) radiolabeled with cesium after 24 mo of diesel exhaust exposure showed increased
11 clearance half-times in the 3.5- and 7.0-mg/m3 groups.
12 In the HERP studies, histopathological effects of diesel exhaust on the lungs of rats
13 were investigated (Ishinishi et al., 1986, 1988). In this study, both light-duty (LD, 1.8-L)
14 and heavy-duty (HD, 11-L) diesel engines were operated under constant velocity and load
15 conditions. The exhaust was diluted to achieve target concentrations of 0.1 (LD only),
16 0.4 (LD and HD), 1 (LD and HD), 2 (LD and HD), and 4 (HD only) mg/m3 paniculate
17 matter. Particle concentrations were determined by filter samples. Actual concentrations
18 were 0.11, 0.41, 1.18, and 2.32 mg/m3 for the light-duty engine and 0.46, 0.96, 1.84, and
19 3.72 mg/m3 for the heavy-duty engine. The number and frequency of sampling is not clear
20 from the published reports. Fischer 344 rats (120 male and 95 female per exposure level for
21 each engine type) were exposed for 16 h/day, 6 days/week for 30 mo. Particle size
22 distributions were determined using an Andersen cascade impactor and an electrical aerosol
23 analyzer. At the 24-mo sampling, the MMD and distribution (sigma g) were 0.22(2.93) and
24 0.19(2.71) for the light-duty engine groups at 2.32 and 1.18 mg/m3, respectively, and
25 0.27(3.18) and 0.22(2.93) for the heavy-duty engine groups at 3.72 and 1.84 mg/m3,
26 respectively (Ishinishi et al., 1988). The number and timing of the samples are not clear
27 from the published reports, nor is it clear which method was used for the results reported
28 above. Particle size data were not reported for the other exposure groups. Hematology,
29 clinical chemistry, urinalysis, and light and electron microscopic examinations were
30 performed. The body weight of females exposed to 4 mg/m3 was 15 to 20 % less than
31 controls throughout the study. No histopathological changes were observed in the lungs of
December 1994 6-9 DRAFT-DO NOT QUOTE OR CITE
-------
1 rats exposed to 0.4 mg/m3 paniculate matter or less. At concentrations above 0.4 mg/m3
2 paniculate matter, accumulation of particle-laden macrophages was observed. In areas of
3 macrophage accumulation, there was bronchiolization of the alveolar ducts, with bronchiolar
4 epithelium replacing alveolar epithelium. Proliferation of brochiolar epithelium and Type 2
5 cells was observed. In these areas, edematous thickening and fibrosis of the alveolar septum
6 were seen. Fibrosis of the alveolar septum developed into small fibrotic lesions. These
7 lesions are collectively referred to as hyperplastic lesions by the authors and their incidence
8 is reported. From a total of 123 to 125 animals examined (approximately equal numbers of
9 males and females), hyperplastic lesions were reported in 4, 4, 6, 12, and 87 animals in the
10 light-duty engine groups exposed to 0, 0.11, 0.41, 1.18, and 2.32 mg/m3, respectively, and
11 in 1, 3, 7, 14, and 25 animals in the heavy-duty engine groups exposed to 0, 0.46, 0.96,
12 1.84, and 3.72 mg/m3, respectively. Statistical analysis of these results was not reported,
13 but there was no difference in the severity ascribed to changes in pulmonary pathology at
14 similar exposure concentrations between the LD and the HD series.
15 The HERP study identifies LOAELs for rats exposed chronically at 1.18 and
16 0.96 mg/m3 (actual exposure) for LD and HD series, respectively, and NOAELs at 0.41 and
17 0.46 mg/m3 (actual) for LD and HD series. The ITRI studies identify a NOAEL for
18 biochemical, histological, and functional changes in the pulmonary region at 0.35 mg/m3
19 (LOAEL = 3.5 mg/m3). The HECs for the principal studies were obtained using the
20 deposition and retention model of Yu and Yoon (1990), as discussed previously. The HEC
21 calculation is based on the assumption that the estimate for the human exposure scenario
22 (a 70-year continuous exposure) should result in an equivalent dose metric, expressed as
23 mass of diesel particle carbon core per unit of pulmonary region surface area, as that
24 associated with no effect at the end of the 2-year rat study. To obtain the HEC, the lung
25 burden in the rat study is calculated using the exposure regimen (concentration, number of
26 hours per day and days per week) and values for rat tidal volume, functional residual
27 capacity, and breathing frequency. A continuous human exposure resulting in the same final
28 lung burden is calculated and is the HEC. The HEC values corresponding to the animals
29 exposure levels in the principal studies are shown in Table 6-1, along with a designation of
30 the concentrations as AEL (adverse-effects-level) or NOAEL; the LOAEL(HEC)s are 0.30,
31 0.36, and 0.36 mg/m3. These values, along with the LOAELS from other studies (discussed
December 1994 6-10 DRAFT-DO NOT QUOTE OR CITE
-------
TABLE 6-1. HUMAN EQUIVALENT CONTINUOUS CONCENTRATIONS FROM
THE PRINCIPAL STUDIES
1
2
3
4
5
6
7
8
9
10
11
12
Exposure
Concentration
Study (mg/m3)
HERP-Light Duty 0.11
0.41
1.18
2.32
HERP-Heavy Duty 0.46
0.96
1.84
3.72
ITRI 0.353
3.47
7.08
AEL/NOAEL
NOAEL
NOAEL
AEL
AEL
NOAEL
AEL
AEL
AEL
NOAEL
AEL
AEL
below), show strong support for an experimental threshold in rats in the
The highest NOAEL(HEC) which is below
all LOAEL(HEC)s is 0.155
HEC
(mg/m3)
0.038
0.139
0.359
0.571
0.155
0.303
0.493
0.911
0.042
0.360
0.582
range of 0.15 to 0.3.
mg/m3 from the
HERP heavy duty diesel study. This NOAEL(HEC) is selected as the basis for the RfC
calculation.
6.5 SUPPORTING STUDIES FOR INHALATION REFERENCE
CONCENTRATION DERIVATION
Chronic inhalation studies using male
carried out at the General Motors Research
Exposures to target concentrations of 0.25,
F-344 rats and male Hartley
Laboratories (Barnhart et al.
guinea pigs were
,, 1981, 1982).
0.75, and 1.5 mg/m3 were generated 20 h/day,
5.5 days/ week for up to 2 years. Exposures at 0.75 and 1.5 mg/m3 for
2 weeks to 6 mo
December 1994
6-11
DRAFT-DO NOT QUOTE OR CITE
-------
1 were reported by Barnhart et al. (1981, 1982). The focus of these studies is on electron
2 micrographic morphometry, and very little descriptive light microscopic histology is
3 reported. Quantitative morphometric analysis showed that the alveolar-capillary membrane
4 increased in thickness as a result of an increase in the absolute tissue volume of interstitium
5 and Type II cells. Exposure to 0.75 mg/m3 for 6 mo resulted in fibrosis in regions of
6 macrophage clusters and in focal Type II cell proliferation observable by light microscopy.
7 Increased cellular composition of the interstitium in animals exposed to 0.75 or 1.5 mg/m3
8 consisting of a variety of inflammatory cell types was observed. Hypertrophy and
9 proliferation of Type II cells was observed as early as 2 weeks at 0.75 mg/m3 or higher.
10 Mean thickness of the air-blood barrier remained elevated in the animals exposed to 0.75 and
11 1.5 mg/m3 exposures, although the peak thickness occurred at 6 mo to 1 year of exposure.
12 These data show that no appreciable changes in morphometric parameters occurred after a
13 2-year exposure to 0.25 mg/m3, while exposure to 0.75 or 1.5 mg/m3 resulted in increased
14 thickness of alveolar septa and increased number of various types of alveolar cells.
15 Increased numbers of polymorphonuclear leukocytes and monocytes were lavaged from rats
16 exposed to 0.75 or 1.5 mg/m3 and biochemical changes occurred in lung tissue at these
17 concentrations (Misiorowski et al., 1980; Eskelson et al., 1981; Strom, 1984). These studies
18 demonstrate a LOAEL of 0.796 and a NOAEL of 0.258 mg/m3 for male guinea pigs in a
19 chronic study for respiratory endpoints, including light and electron microscopy, lavage
20 cytology, and lung tissue biochemistry.
21 A 15-mo inhalation study was performed by Southwest Research Institute for General
22 Motors (Kaplan et al., 1983). Male F-344 rats, Syrian golden hamsters, and A/J mice were
23 exposed to diluted diesel exhaust at target concentrations of 0.25, 0.75, and 1.5 mg/m3 for
24 20 h/day and 7 days/week. Focal accumulation of particle-laden macrophages was associated
25 with minimal to mild fibrosis of the alveolar wall. Based on accumulation of particle-laden
26 macrophages, this study identifies a LOAEL at 0.735 mg/m3 and a NOAEL at 0.242 mg/m3.
27 In a study performed by NIOSH (Lewis et al., 1986, 1989; Green et al., 1983), male
28 and female F-344 rats and male cynomolgus monkeys were exposed to target levels of
29 2 mg/m3 diesel particles. Accumulations of black-pigmented alveolar macrophages were
30 seen in the alveolar ducts of rats adjacent to terminal bronchioles and epithelial lining cells
31 adjacent to collections of pigmented macrophages showed a marked Type II cell hyperplasia.
December 1994 6-12 DRAFT-DO NOT QUOTE OR CITE
-------
1 No evidence of an impairment of pulmonary function as a result of the exposure to diesel
2 exhaust was found in rats. Histological examination of lung tissue from monkeys exposed
3 for 24 mo in the same regimen as used for rats revealed aggregates of black particles,
4 principally in the distal airways of the lung. Fibrosis, focal emphysema, or inflammation
5 was not observed. The monkeys exposed to diesel exhaust demonstrated small airway
6 obstructive disease. The obstructive impairment was most detectable using the forced
7 expiratory flow at 40% of the total lung capacity instead of the forced expiratory flow as a
8 percentage of the vital capacity. This study demonstrates a LOAEL for rats and monkeys at
9 a diesel particle concentration of 2 mg/m3.
10 Heinrich et al. (1986; see also Stober, 1986) exposed male and female Syrian golden
11 hamsters, female NMRI mice, and female Wistar rats to diesel engine emissions with a
12 4.2 mg/m3 paniculate concentration. Lung weights were increased by a factor of 2 or 3 in
13 rats and mice after 2 years of exposure and in hamsters the lung weights were increased by
14 50 to 70%. Although histological examination revealed different levels of response among
15 the three species, histological effects were seen in all species and effects on pulmonary
16 function were observed in rats and hamsters. This study demonstrates a LOAEL in rats for
17 respiratory system effects of 4.2 mg/m3.
18 The effects of diesel exhaust on the lungs of 18-week-old male Wistar rats exposed to
19 8.3 ± 2.0 mg/m3 particulate matter were investigated by Karagianes et al. (1981).
20 Histological examinations of lung tissue noted focal aggregation of particle-laden alveolar
21 macrophages, alveolar histiocytosis, interstitial fibrosis, and alveolar emphysema. Lesion
22 severity was related to length of exposure. No exposure-related effects were seen in the
23 nose, larynx, or trachea. This study demonstrates a LOAEL of 8.3 mg/m3 for respiratory
24 effects after chronic exposure of rats to diesel emissions.
25 The lung function of adult cats chronically exposed to diesel exhaust concentrations of
26 6.34 mg/m3 for the first 61 weeks and 6.7 mg/m3 from weeks 62 to 124. No definitive
27 pattern of pulmonary function changes was observed following 61 weeks of exposure;
28 however, a classic pattern of restrictive lung disease was found at 124 weeks (Pepelko et al.,
29 1980).
30 Brightwell et al. (1986) evaluated the toxic effects of diesel exhaust on rats and
31 hamsters at concentrations of 0.7, 2.2, and 6.6 mg/m3 particulate matter. Respiratory
December 1994 6_13 DRAFT-DO NOT QUOTE OR CITE
-------
1 physiology measurements were not affected in high concentration hamsters, but were
2 significantly changed in high concentration rats. The changes in rats are not specified, but
3 are summarized as being consistent with obstructive and restrictive disease in the high
4 concentration group. No detailed information on histopathological results or pulmonary
5 function results are provided and no further published accounts of this study could be located
6 except a discussion of the tumorigenic response in rats and hamsters (Brightwell et al.,
7 1989). Because only results from the high concentration groups are reported, this study is
8 not useful for RfC development. The high concentration is shown to be an AEL in rats and
9 hamsters.
10 Werchowski et al. (1980a) reported a developmental study in rabbits exposed on Days 6
11 through 18 of gestation to a 1 in 10 dilution of diesel exhaust. The exposure protocol for this
12 and other EPA studies is reviewed by Pepelko and Peraino (1983) and the target exposure
13 level of 6 mg/m3 is indicated. Exposure to diesel emissions had no effect on maternal
14 toxicity or on the developing fetuses. In a companion study (Werchowski et al., 1980b),
15 20 SD rats were exposed for 8 h/day during Days 5 to 16 to a target concentration of
16 6 mg/m3 of diesel particles (protocol reviewed by Pepelko and Peraino, 1983). Fetuses were
17 examined for external, internal, and skeletal malformations and number of live and dead
18 fetuses, resorptions, implants, corpora lutea, fetal weight, litter weight, sex ratio, and
19 maternal toxicity were recorded. No conclusive evidence of developmental effects was
20 observed in this study.
21 In an EPA-sponsored reproductive study summarized by Pepelko and Peraino (1983),
22 CD-I mice were exposed to a target concentration of 12 mg/m3 for 8 h/day and
23 7 days/week. The FO and Fl animals were exposed for 100 days prior to breeding and
24 100 mating pairs were randomly assigned to four exposure groups of 25 each. Viability
25 counts and pup weights were recorded at 4, 7, and 14 days after birth and at weaning.
26 No treatment-related effects on body weight in FO mice, or in Fl animals through weaning
27 or in mating animals through gestation were found. No treatment-related effects on gestation
28 length, percent fertile, litter size, or pup survival was observed. The only organ weight
29 difference was an increase in lung weight in exposed FO and Fl mice (lung weight and lung
30 weight/body weight) and in F2 males (lung weight/body weight). Based on this study, a
31 NO AEL for reproductive effects in rats is identified at 12 mg/m3.
December 1994 6-14 DRAFT-DO NOT QUOTE OR CITE
-------
1 The reproductive and developmental studies described in the previous paragraphs serve
2 to show that the effects in the respiratory system are the most sensitive effects that result
3 from diesel exhaust exposures. These studies add to the confidence that a variety of
4 noncancer effects have been studied, and are required for a designation of high confidence in
5 the database and the RfC.
6 Several epidemiologic studies have evaluated the effects of chronic exposure to diesel
7 exhaust on occupationally exposed workers. The human studies, taken together, are
8 suggestive but inconclusive of an effect on pulmonary function, as described in Chapter 5.
9 The studies are not directly useful for derivation of the RfC because of inadequate ability to
10 directly relate the observed effects with known concentrations of diesel particles. The studies
11 are confounded by coexposures to other particles or by a lack of measurement of particle
12 exposure.
13
14
15 6.6 DERIVATION OF THE INHALATION REFERENCE
16 CONCENTRATION
17 Reports on chronic exposures to diesel emissions performed at ITRI and the HERP
18 studies were selected as the basis of the RfC. These studies were selected because they
19 identify both a NOAEL and a LOAEL for rats exposed chronically. The only other study
20 identifying both a NOAEL and a LOAEL was the G.M. study, which was not used because
21 limited information was available characterizing the pulmonary lesion in rats. The
22 availability of the dosimetric model for rats and not for other species, along with the
23 apparent comparability between the rat and other rodent species in response resulted in the
24 choice of the rat as the basis for development of the RfC. Although the data from the
25 monkey in the Lewis et al. (1989) study suggest that the pulmonary function effect in
26 primates more closely resembles that in humans, this study has only one exposed group,
27 making evaluation of dose-response impossible. Thus, this was not considered to be a strong
28 enough basis for eliminating consideration of the strong rodent database. The pulmonary
29 effects, including histological lesions, biochemical changes, pulmonary function impairment,
30 and impaired particle clearance were determined to be the critical noncancer effect.
31 Sufficient documentation from other studies showed that there is no effect in the extrathoracic
December 1994 6_15 DRAFT-DO NOT QUOTE OR CITE
-------
1 region of the respiratory system or in other organs at the lowest levels that produce
2 pulmonary effects in chronic exposures. In addition, adequate information is available from
3 the EPA studies showing no effect on development in two species or on reproduction in a
4 two-generation reproductive study.
5 Because the RfC is based on a NOAEL from a chronic animal study, with a thorough
6 database, uncertainty exists in the extrapolation from animals to humans and for extrapolation
7 to sensitive members of the population (inter- and intra-species extrapolation). A factor of
8 10 is normally applied for each area of uncertainty (i.e., a total uncertainty factor of 100)
9 when a chronic animal NOAEL is available. Since the dosimetry model which is specific for
10 diesel particles is available, the use of this model is considered to reduce the uncertainty in
11 extrapolating between animals and humans, compared to a case in which no chemical or
12 species-specific data on dosimetry are available. The usual uncertainty factor of 10 includes
13 aspects of pharmacokinetics and pharmacodynamics. An uncertainty factor of 3 rather than
14 10 was adopted for interspecies extrapolation when using default dosimetry adjustments, and
15 was also used for the diesel RfC as a result of the application of the dosimetry model. When
16 a reduction in an uncertainty factor is used, it is generally considered that the reduction can
17 be no more precise than a single significant figure and a reduction of the 10 to its geometric
18 halfway point, is the extent of the change considered appropriate. Therefore, 3 is adopted as
19 the square root of 10, to one significant figure. A total uncertainty factor of 30 results
20 (10 for intra-species and 3 for inter-species extrapolation).
21 There was some concern as to whether the extensive nature of the database and the
22 quality of the dosimetric model warranted a further reduction of the uncertainty factor.
23 A further reduction, to a total uncertainty factor of 10, could be argued on the grounds that
24 the NIOSH study suggests that monkeys (and perhaps humans as well) are less sensitive than
25 rats, suggesting that the interspecies factor could be further reduced. It could also be argued
26 that the uncertainty factor of 3 for interspecies extrapolation is used for cases in which
27 default dosimetry adjustments are applied, and a more refined dosimetry model requires a
28 further reduction in the uncertainty factor. A reduction in the overall uncertainty factor
29 could also be argued on the grounds that the dosimetric model includes information on the
30 developing and aging lung, so the intraspecies uncertainty factor could be reduced. The
31 latter argument might also be supported by studies of diesel exhaust exposure in animals with
December 1994 6-16 DRAFT-DO NOT QUOTE OR CITE
-------
1 experimentally induced lung disease. Mauderly et al. (1990) showed that a 2-year study of
2 rats with elastase-induced emphysema had less deposition of diesel exhaust and were less
3 susceptible to diesel exhaust toxicity. Also, Mauderly et al. (1987a) showed no differences
4 between developing lung and adult lung in susceptibility by comparing the effects of a 6-mo
5 exposure of rats to 3.5 mg/m3 diesel exhaust and 0.8 ppm NO2, starting either at gestation or
6 at 6 mo of age. These arguments were not considered to be sufficient to reduce the
7 uncertainty factor further, and a total uncertainty factor of 30 was adopted. Using the
8 NOAEL(HEC) of 0.155 mg/m3 from the HERP study, an RfC of 5 ^g/m3 was calculated.
9 The RfC is generally expressed to one significant figure due to the imprecision of the
10 uncertainty factors.
11 The RfC also includes confidence statements associated with the principal study, the
12 data base, and the resulting RfC. The studies used as the basis of the RfC were well-
13 conducted chronic studies with adequate numbers of animals, and the LOAELs and NOAELs
14 were consistent across studies thereby resulting in high confidence. The database contains
15 several chronic studies, including multiple species, which support the LOAEL observed in
16 the principal studies. There are also developmental and reproductive studies, resulting in a
17 high confidence database. Following from high confidence in the studies and database, the
18 RfC has high confidence.
19
20
21 6.7 SUMMARY
22 A large number of chronic inhalation studies of diesel exhaust inhalation in
23 experimental animals are available. These studies characterize the respiratory effects and the
24 concentration-response relationship of those effects in detail. Many epidemiological studies
25 are also available of occupationally exposed humans. The epidemiological studies provide
26 qualitative evidence that supports the identification of a hazard to the respiratory system from
27 animal studies. The human studies are of limited value quantitatively due to inadequate
28 exposure characterization and confounding by concurrent exposure to other pollutants. The
29 animal studies are used for derivation of an RfC. The chronic studies from ITRI and HERP
30 were selected as the principal studies for RfC development. Using the deposition and
31 retention model discussed in Chapter 4 and Appendix C to calculate human equivalent
December 1994 6-17 DRAFT-DO NOT QUOTE OR CITE
-------
1 concentrations, a NOAEL(HEC) of 0.155 mg/m3 was identified from the HERP studies and
2 a LOAEL(HEC) of 0.36 was identified from the ITRI studies. An uncertainty factor of
3 30 was applied for interspecies extrapolation and to account for sensitive members of the
4 population, resulting hi an RfC of 5 pig/m3. The RfC is considered to have high confidence,
5 due to high confidence in the study and database.
December 1994 6-18 DRAFT-DO NOT QUOTE OR CITE
-------
1 REFERENCES
2 Barnhart, M. I.; Chen, S.-t.; Salley, S. O.; Puro, H. (1981) Ultrastructure and morphometry of the alveolar lung
3 of guinea pigs chronically exposed to diesel engine exhaust: six month's experience. J. Appl. Toxicol.
4 1: 88-103.
5
6 Barnhart, M. I.; Salley, S. O.; Chen, S.-t.; Puro, H. (1982) Morphometric ultrastructural analysis of alveolar
7 lungs of guinea pigs chronically exposed by inhalation to diesel exhaust (DE). In: Lewtas, J., ed.
8 lexicological effects of emissions from diesel engines: proceedings of the Environmental Protection
9 Agency diesel emissions symposium; October, 1981; Raleigh, NC. New York, NY: Elsevier Biomedical;
10 pp. 183-200. (Developments in toxicology and environmental science: v. 10).
11
12 Brightwell, J.; Fouillet, X.; Cassano-Zoppi, A.-L.; Gatz, R.; Duchosal, F. (1986) Neoplastic and functional
13 changes in rodents after chronic inhalation of engine exhaust emissions. In: Ishinishi, N.; Koizumi, A.;
14 McClellan, R. O.; Stober, W., eds. Carcinogenic and mutagenic effects of diesel engine exhaust:
15 proceedings of the international satellite symposium on lexicological effects of emissions from diesel
16 engines; July; Tsukuba Science City, Japan. Amsterdam, Holland: Elsevier Science Publishers B. V.;
17 pp. 471-485. (Developments in toxicology and environmental science: v. 13).
18
19 Brightwell, J.; Fouillet, X.; Cassano-Zoppi, A.-L.; Bernstein, D.; Crawley, F.; Duchosal, R.; Gatz, R.;
20 Perczel, S.; Pfeifer, H. (1989) Tumors of the respiratory tract in rats and hamsters following chronic
21 inhalation of engine exhaust emissions. J. Appl. Toxicol. 9: 23-31.
22
23 Dungworth, D. L.; Mohr, U.; Heinrich, U.; Ernst, H.; Kittel, B. (1994) Pathologic effects of inhaled particles in
24 rat lungs: associations between inflammatory and neoplastic processes. In Mohr, U.; Dungworth, D. L.;
25 Mauderly, J. L.; Oberdorster, G., eds. Toxic and carcinogenic effects of solid particles in the respiratory
26 tract: [proceedings of the 4th international inhalation symposium]; March 1993; Hannover, Germany.
27 Washington, DC: International Life Sciences Institute Press; pp. 75-98.
28
29 Eskelson, C. D.; Strom, K. A.; Vostal, J. J.; Misiorowski, R. L. (1981) Lipids in the lung and lung gavage
30 fluid of animals exposed to diesel particulates. Toxicologist 1: 74-75.
31
32 Green, F. H. Y; Boyd, R. L.; Danner-Rabovsky, J.; Fisher, M. J.; Moorman, W. J.; Ong, T.-M.; Tucker, J.;
33 Vallyathan, V.; Whong, W.-Z.; Zoldak, J.; Lewis, T. (1983) Inhalation studies of diesel exhaust and
34 coal dust in rats. Scand. J. Work Environ. Health 9: 181-188.
35
36 Heinrich, U.; Peters, L.; Funcke, W.; Pott, F.; Mohr, U.; Stober, W. (1982) Investigation of toxic and
37 carcinogenic effects of diesel exhaust in long-term inhalation exposure of rodents. In: Lewtas, J., ed.
38 Toxicological effects of emissions from diesel engines: proceedings of the Environmental Protection
39 Agency diesel emissions symposium; October 1981; Raleigh, NC. New York, NY: Elsevier Biomedical;
40 pp. 225-242. (Developments in toxicology and environmental science: v. 10).
41
42 Heinrich, U.; Muhle, H.; Takenaka, S.; Ernst, H.; Fuhst, R.; Mohr, U.; Pott, F.; Stober, W. (1986) Chronic
43 effects on the respiratory tract of hamsters, mice, and rats after long-term inhalation of high
44 concentrations of filtered and unfiltered diesel engine emissions. J. Appl. Toxicol. 6: 383-395.
45
46 Henderson, R. F.; Pickrell, J. A.; Jones, R. K.; Sun, J. D.; Benson, J. M.; Mauderly, J. L.; McClellan, R. 0.
47 (1988) Response of rodents to inhaled diluted diesel exhaust: biochemical and cytological changes in
48 bronchoalveolar lavage fluid and in lung tissue. Fundam. Appl. Toxicol. 11: 546-567.
49
50
December 1994 6_19 DRAFT-DO NOT QUOTE OR CITE
-------
1 Ishinishi, N.; Kuwabara, N.; Nagase, S.; Suzuki, T.; Ishiwata, S.; Kohno, T. (1986) Long-term inhalation
2 studies on effects of exhaust from heavy and light duty diesel engines on F344 rats. In: Ishinishi, N.;
3 Koizumi, A.; McClellan, R. O.; Stober, W., eds. Carcinogenic and mutagenic effects of diesel engine
4 exhaust: proceedings of the international satellite symposium on toxicological effects of emissions from
5 diesel engines; July; Tsukuba Science City, Japan. Amsterdam, Holland: Elsevier Science Publishers
6 B. V.; pp. 329-348. (Developments in toxicology and environmental science: v. 13).
7
8 Ishinishi, N.; Kuwabara, N.; Takaki, Y.; Nagase, S.; Suzuki, T.; Nakajima, T.; Maejima, K.; Kato, A.;
9 Nakamura, M. (1988) Long-term inhalation experiments on diesel exhaust. In: Diesel exhaust and health
10 risks: results of the HERP studies. Tsukuba, Ibaraki, Japan: Japan Automobile Research Institute, Inc.,
11 Research Committee for HERP Studies; pp. 11-84.
12
13 Iwai, K.; Udagawa, T.; Yamagishi, M.; Yamada, H. (1986) Long-term inhalation studies of diesel exhaust on
14 F344 SPF rats. Incidence of lung cancer and lymphoma. In: Ishinishi, N.; Koizumi, A.;
15 McClellan, R. O.; Stober, W., eds. Carcinogenic and mutagenic effects of diesel engine exhaust:
16 proceedings of the international satellite symposium on toxicological effects of emissions from diesel
17 engines; July; Tsukuba Science City, Japan. Amsterdam, Holland: Elsevier Science Publishers B. V.;
18 pp. 349-360. (Developments in toxicology and environmental science: v. 13).
19
20 Kaplan, H. L.; Springer, K. J.; MacKenzie, W. F. (1983) Studies of potential health effects of long-term
21 exposure to diesel exhaust emissions. San Antonio, TX: Southwest Research Institute; SwRI project no.
22 01-0750-103.
23
24 Karagianes, M. T.; Palmer, R. F.; Busch, R. H. (1981) Effects of inhaled diesel emissions and coal dust in rats.
25 Am. Ind. Hyg. Assoc. J. 42: 382-391.
26
27 Lewis, T. R.; Green, F. H. Y.; Moorman, W. J.; Burg, J. A. R.; Lynch, D. W. (1986) A chronic inhalation
28 toxicity study of diesel engine emissions and coal dust, alone and combined. In: Ishinishi, N.;
29 Koizumi, A.; McClellan, R. O.; Stober, W., eds. Carcinogenic and mutagenic effects of diesel engine
30 exhaust: proceedings of the international satellite symposium on toxicological effects of emissions from
31 diesel engines; July; Tsukuba Science City, Japan. Amsterdam, The Netherlands: Elsevier Science
32 Publishers B. V.; pp. 361-380. (Developments in toxicology and environmental science: v. 13).
33
34 Lewis, T. R.; Green, F. H. Y.; Moorman, W. J.; Burg, J. R.; Lynch, D. W. (1989) A chronic inhalation
35 toxicity study of diesel engine emissions and coal dust, alone and combined. J. Am. Coll. Toxicol.
36 8: 345-375.
37
38 Mauderly, J. L.; Jones, R. K.; Griffith, W. C.; Henderson, R. F.; McClellan, R. O. (1987a) Diesel exhaust is a
39 pulmonary carcinogen in rats exposed chronically by inhalation. Fundam. Appl. Toxicol. 9: 208-221.
40
41 Mauderly, J. L.; Bice, D. E.; Carpenter, R. L.; Gillett, N. A.; Henderson, R. F.; Pickrell, J. A.; Wolff, R. K.
42 (1987b) Effects of inhaled nitrogen dioxide and diesel exhaust on developing lung. Cambridge, MA:
43 Health Effects Institute; research report no. 8.
44
45 Mauderly, J. L.; Gillett, N. A.; Henderson, R. F.; Jones, R. K.; McClellan, R. O. (1988) Relationships of lung
46 structural and functional changes to accumulation of diesel exhaust particles. In: Dodgson, J.; McCallum,
47 R. I.; Bailey, M. R.; Fisher, D. R., eds. Inhaled particles VI: proceedings of an international symposium
48 and workshop on lung dosimetry; September 1985; Cambridge, United Kingdom. Ann. Occup. Hyg.
49 32(suppl. 1): 659-669.
50
51 Mauderly, J. L.; Bice, D. E.; Cheng, Y. S.; Gillett, N. A.; Griffith, W. C.; Henderson, R. F.; Pickrell, J. A.;
52 Wolff, R. K. (1990) Influence of preexisting pulmonary emphysema on susceptibility of rats to diesel
53 exhaust. Am. Rev. Respir. Dis. 141: 1333-1341.
54
December 1994 6-20 DRAFT-DO NOT QUOTE OR CITE
-------
1 McClellan, R. O.; Bice, D. E.; Cuddihy, R. G.; Gillett, N. A.; Henderson, R. F.; Jones, R. K.; Mauderly,
2 J. L.; Pickrell, J. A.; Shami, S. G.; Wolff, R. K. (1986) Health effects of diesel exhaust. In: Lee,
3 S. D.; Schneider, T.; Grant, L. D.; Verkerk, P. J., eds. Aerosols: research, risk assessment and control
4 strategies: proceedings of the second U.S.-Dutch international symposium; May 1985; Williamsburg,
5 VA. Chelsea, MI: Lewis Publishers, Inc.; pp. 597-615.
6
7 Misiorowski, R. L.; Strom, K. A.; Vostal, J. J.; Chvapil, M. (1980) Lung biochemistry of rats chronically
8 exposed to diesel particulates. In: Pepelko, W. E.; Danner, R. M.; Clarke, N. A., eds. Health effects of
9 diesel engine emissions: proceedings of an international symposium; December 1979. Cincinnati, OH:
10 U.S. Environmental Protection Agency, Health Effects Research Laboratory; pp. 465-480; EPA report
11 no. EPA-600/9-80-057a. Available from: NTIS, Springfield, VA; PB81-173809.
12
13 Morrow, P. E. (1992) Dust overloading of the lungs: update and appraisal. Toxicol. Appl. Pharmacol.
14 113: 1-12.
15
16 Nikula, K. J.; Snipes, M. B.; Barr, E. B.; Mauderly, J. L. (1991) Histopathology and lung tumor responses in
17 rats exposed to diesel exhaust or carbon black. In: Annual report of the Inhalation Toxicology Research
18 Institute operated for the United States Department of Energy by the Lovelace Biomedical and
19 Environmental Research Institute: October 1, 1990 though September 30, 1991. Albuquerque, NM:
20 Inhalation Toxicology Research Institute, Lovelace biomedical and Environmental Research Institute;
21 pp. 87-129; report no. LMF-134.
22
23 Nikula, K. J.; Snipes, M. B.; Barr, E. B.; Griffith, W. C.; Henderson, R. F.; Mauderly, J. L. (1994) Influence
24 of particle-associated organic compounds on the carcinogenicity of diesel exhaust. In: Mohr, U.;
25 Dungworth, D. L.; Mauderly, J. L.; Oberdorster, G., eds. Toxic and carcinogenic effects of solid
26 particles in the respiratory tract: [proceedings of the 4th international inhalation symposium]; March
27 1993; Hannover, Germany. Washington, DC: International Life Sciences Institute Press; pp. 565-568.
28
29 Pepelko, W. E.; Peirano, W. B. (1983) Health effects of exposure to diesel engine emissions: a summary of
30 animal studies conducted by the U.S. Environmental Protection Agency's Health Effects Research
31 Laboratories at Cincinnati, Ohio. J. Am. Coll. Toxicol. 2: 253-306.
32
33 Pepelko, W. E.; Mattox, J.; Moorman, W. J.; Clark, J. C. (1980) Pulmonary function evaluation of cats after
34 one year of exposure to diesel exhaust. In: Pepelko, W. E.; Danner, R. M.; Clarke, N. A., eds. Health
35 effects of diesel engine emissions: proceedings of an international symposium, v. 2; December 1979;
36 Cincinnati, OH. Cincinnati, OH: U.S. Environmental Protection Agency, Health Effects Research
37 Laboratory; pp. 757-765; EPA report no. EPA-600/9-80-057b. Available from: NTIS, Springfield VA-
38 PB81-173817.
39
40 Stober, W. (1986) Experimental induction of tumors in hamsters, mice and rats after long-term inhalation of
41 filtered and unfiltered diesel engine exhaust. In: Ishinishi, N.; Koizumi, A.; McClellan, R. O.;
42 Stober, W., eds. Carcinogenic and mutagenic effects of diesel engine exhaust: proceedings of the"
43 international satellite symposium on lexicological effects of emissions from diesel engines; July; Tsukuba
44 Science City, Japan. Amsterdam, The Netherlands: Elsevier Science Publishers B. V.; pp. 421-439.
45 (Developments in toxicology and environmental science: v. 13).
46
47 Strom, K. A. (1984) Response of pulmonary cellular defenses to the inhalation of high concentrations of diesel
48 exhaust. J. Toxicol. Environ. Health 13: 919-944.
49
50 U.S. Environmental Protection Agency. (1990) Interim methods for development of inhalation reference
51 concentrations [review draft]. Research Triangle Park, NC: Office of HealtK and Environmental
52 Assessment, Environmental Criteria and Assessment Office; EPA report no. EPA/600/8-90/066 A.
53 Available from: NTIS, Springfield, VA; PB90-238890.
54
December 1994 6_2i DRAFT-DO NOT QUOTE OR CITE
-------
1 Werchowski, K. M.; Chaffee, V. W.; Briggs, G. B. (1980a) Teratolo/ic epcts of long-term exposure to diesel
2 exhaust emissions (rats). Cincinnati, OH: U.S. Environmenta^^ection Agency, Health Effects
3 Research Laboratory; EPA report no. EPA-600/1-80-010. Available from: NTIS, Springfield, VA;
4 PB80-159965.
5
6 Werchowski, K. M.; Henne, S. P.; Briggs, G. B. (1980b) Teratologic effects of long-term exposure to diesel
7 exhaust emissions (rabbits). Cincinnati, OH: U.S. Environmental Protection Agency, Health Effects
8 Research Laboratory; EPA report no. EPA-600/1-80-011. Available from: NTIS, Springfield, VA;
9 PB80-168529.
10
11 Wolff, R. K.; Henderson, R. F.; Snipes, M. B.; Griffith, W. C.; Mauderly, J. L.; Cuddihy, R. G.; McClellan,
12 R. O. (1987) Alterations in particle accumulation and clearance in lungs of rats chronically exposed to
13 diesel exhaust. Fundam. Appl. Toxicol. 9: 154-166.
14
15 Yu, C. P.; Xu, G. B. (1986) Predictive models for deposition of diesel exhaust particulates in human and rat
16 lungs. Aerosol Sci. Technol. 5: 337-347.
17
18 Yu, C. P.; Yoon, K. J. (1990) Retention modeling of diesel exhaust particles in rats and humans. Amherst, NY:
19 State University of New York at Buffalo (Health Effects Institute research report no. 40).
U S. Environmental Protection Agency
Region 5, Library (PL-12J)
77 West Jackson Boulevard, 12th noor
Chicago, IL 60604-3590
December 1994 6-22 DRAFT-DO NOT QUOTE OR CITE
------- |