EPA600/8-91/049aF
August 1993
Air Quality Criteria for
Oxides of Nitrogen
Volume II of III
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
Printed on Recycled Paper
-------
DISCLAIMER
i
I I i
This document has been reviewed in accordance with U S Environmental Protection
i
Agency policy and approved for publication. Mention of trade names or commercial
products does not constitute endorsement or recommendation for use
I-ii
-------
PREFACE
The U S Environmental Protection Agency (EPA) promulgates the National Ambient
Air Quality Standards (NAAQS) on the basis of scientific information contained in criteria
documents In 1971, the first air quality criteria document for nitrogen oxides (NOX) was
issued by the National Air Pollution Control Admimsl ration, a predecessor of EPA On the
basis of scientific information contained in that document, NAAQS were promulgated for
•3
nitrogen dioxide (NO2) at levels of 0 053 ppm (100 /ig/m ), averaged over 1 year The last
full-scale NOX criteria document revision was completed by EPA in 1982, leading to an
Agency decision in 1985 to reaffirm the annual average NO2 NAAQS of 0 053 ppm The
present, revised criteria document, Air Quality Criteria for Oxides of Nitrogen, assesses the
current scientific basis for periodic reevaluation of the NO2 NAAQS in accordance with the
provisions identified in Sections 108 and 109 of the Clean Air Act
Key chapters in this document evaluate the latesl scientific data on (a) health effects of
NOX measured in laboratory animals and exposed human populations and (b) effects of NOX
on agricultural crops, forests, and ecosystems, as well as (c) NOX effects t>n visibility and
nonbiological materials Other chapters describe the nature, sources, distribution,
measurement, and concentrations of NOX in the environment These chapters were prepared
and peer reviewed by experts from various state and Federal government offices, academia,
and private industry for use by EPA to support decision making regarding potential risks to
public health and the environment Although the document is not intended to be an
exhaustive literature review, it is intended to cover all the pertinent literature through early
1993
The Environmental Criteria and Assessment Office of EPA's Office of Health and
Environmental Assessment acknowledges with appreciation the contributions provided by the
authors and reviewers and the diligence of its staff and contractors in the preparation of this
document at the request of EPA's Office of Air Quality Planning and Standards
I-iii
-------
-------
Air Quality Criteria for Oxides of Nitrogen
TABLE OF CONTENTS
Volume I Page
1 EXECUTIVE SUMMARY OF AIR QUALITY CRITERIA FOR
OXIDES OF NITROGEN . . . . 1-1
2 INTRODUCTION 2-1
3 GENERAL CHEMICAL AND PHYSICAL PROPERTIES OF
OXIDES OF NITROGEN AND OXIDES OF NITROGEN-DERIVED
POLLUTANTS 3-1
4 AMBIENT AND INDOOR SOURCES AND EMISSIONS OF
NITROGEN OXIDES 4-1
5 TRANSPORT AND TRANSFORMATION OF NITROGEN
OXIDES 5-1
6 SAMPLING AND ANALYSIS FOR NITROGEN OXIDES
AND RELATED SPECIES . 6-1
7 AMBIENT AND INDOOR CONCENTRATIONS OF NITROGEN
OXIDES . 7-1
8 ASSESSING TOTAL HUMAN EXPOSURE TO NITROGEN
DIOXIDE 8-1
Volume n
9 EFFECTS OF NITROGEN OXIDES ON VEGETATION 9-1
10 THE EFFECTS OF NITROGEN OXIDES ON NATURAL
ECOSYSTEMS AND THEIR COMPONENTS 10-1
11 EFFECTS OF NITROGEN OXIDES ON VISIBILITY 11-1
»
12 EFFECTS OF NITROGEN OXIDES ON MATERIALS 12-1
I-v
-------
Air Quality Criteria for Oxides of Nitrogen
TABLE OF CONTENTS (cont'd)
Volume HI
13. STUDIES OF THE EFFECTS OF NITROGEN COMPOUNDS
ON ANIMALS . . . 13-1
14. EPIDEMIOLOGY STUDIES OF OXIDES OF NITROGEN 14-1
15. CONTROLLED HUMAN EXPOSURE STUDIES OF NITROGEN
OXIDES . 15-1
16. HEALTH EFFECTS ASSOCIATED WITH EXPOSURE TO
NITROGEN DIOXIDE . 16-1
APPENDKA- GLOSSARY OF TERMS AND SYMBOLS A-l
I-vi
-------
TABLE OF CONTENTS
Page
LIST OF TABLES I-xiv
LIST OF FIGURES I-xix
AUTHORS I-xxv
CONTRIBUTORS AND REVIEWERS I-xxvii
CLEAN AIR SCIENTIFIC ADVISORY COMMITTEE . . I-xxrx
PROJECT TEAM FOR DEVELOPMENT OF AIR QUALITY CRITERIA
FOR OXIDES OF NITROGEN I-xxxi
1 EXECUTIVE SUMMARY OF AIR QUALITY CRITERIA FOR
OXIDES OF NITROGEN . . 1-1
1 1 PURPOSE OF THE DOCUMENT 1-1
1 2 INTRODUCTION 1-1
1 3 CHEMICAL AND PHYSICAL PROPERTIES OF
NITROGEN OXIDES AND NITROGEN OXIDE-DERIVED
POLLUTANTS . 1-1
1 4 EMISSIONS OF NITROGEN OXIDES FROM AMBIENT
AND INDOOR SOURCES 1-2
1 5 TRANSPORT AND TRANSFORMATION OF NITROGEN
OXIDES . . 1-3
1 6 SAMPLING AND ANALYSIS FOR OXIDES OF
NITROGEN AND RELATED SPECIES 1-5
1 7 AMBIENT AND INDOOR CONCENTRATIONS OF
OXIDES OF NITROGEN 1-6
171 Ambient Nitrogen Dioxide Levels 1-6
172 Indoor Nitrogen Dioxide Levels 1-7
1 8 ASSESSING TOTAL HUMAN EXPOSURE TO NITROGEN
DIOXIDE . . 1-8
1 9 EFFECTS OF NITROGEN OXIDES ON VEGETATION 1-9
1 10 EFFECTS OF NITROGEN OXIDES ON ECOSYSTEMS 1-9
111 EFFECTS OF NITROGEN OXIDES ON VISIBILITY 1-14
1 12 EFFECTS OF NITROGEN OXIDES ON MATERIALS 1-15
1 13 STUDIES OF THE EFFECTS OF NITROGEN COMPOUNDS
ON ANIMALS 1-16
1 14 EPIDEMIOLOGY STUDIES OF NITROGEN DIOXIDE 1-17
1 15 CONTROLLED HUMAN EXPOSURE STUDIES OF
OXIDES OF NITROGEN . . . . 1-19
1 16 NITROGEN DIOXIDE HEALTH EFFECTS
CONCENTRATION-RESPONSE RELATIONSHIPS AND
SUBPOPULATIONS POTENTIALLY AT RISK . 1-20
1 16 1 Concentration-Response Relationships 1-20
1 16 2 Subpopulations Potentially at Risk 1-23
2 INTRODUCTION . 2-1
2 1 REGULATORY AND SCIENTIFIC BACKGROUND 2-2
X-vii
-------
TABLE OF CONTENTS (cont'd)
Page
2.2 CRITICAL ISSUES . . 2-4
2.3 ORGANIZATION OF THE DOCUMENT 2-5
REFERENCES ... 2-8
3. GENERAL CHEMICAL AND PHYSICAL PROPERTIES OF
OXIDES OF NITROGEN AND OXIDES OF NITROGEN-
DERIVED POLLUTANTS .... 3-1
3.1 INTRODUCTION AND OVERVIEW 3-1
3.2 NITROGEN OXIDES 3-4
3.2.1 Nitnc Oxide . . 3-4
3.2.2 Nitrogen Dioxide . 3-10
3.2.3 Nitrous Oxide 3-10
3.2.4 Nitrogen Tnoxide . 3-12
3.2 5 Dimtrogen Tnoxide 3-12
3.2.6 Dinitrogen Tetroxide 3-13
3.2.7 Dinitrogen Pentoxide . . 3-13
3.3 NITRATES, NITRITES, AND NITROGEN ACIDS 3-14
3.3.1 Nitric Acid 3-14
3.3.2 Nitrous Acid - 3-14
3.3 3 Organic Nitrates 3-15
3.3 4 Aerosol Nitrates 3-16
3.4 AMMONIA . 3-16
3.5 AT-NITROSO COMPOUNDS 3-17
3.6 SUMMARY . 3-18
361 Nitrogen Oxides 3-19
3.6 2 Nitrates, Nitrites, and Nitrogen Acids 3-19
3.6.3 2V-Nitroso Compounds . 3-20
REFERENCES . 3-21
4. AMBIENT AND INDOOR SOURCES AND EMISSIONS OF
NITROGEN OXIDES 4-1
4.1 INTRODUCTION 4-1
4.2 AMBIENT SOURCES OF NITROGEN OXIDES 4-2
4.2.1 Anthropogenic Sources of Nitrogen Oxides 4-3
4.2.1.1 Transportation . . 4-4
4 2.1 2 Stationary Source Fuel Combustion • 4-8
4.2 13 Industrial Processes . 4-10
4214 Solid Waste Disposal . 4-10
4215 Miscellaneous Sources . 4-10
4.2 2 Natural Sources of Nitrogen Oxides 4-11
4.2 3 Global Estimates of Nitrogen Oxides Emissions 4-13
4.2 4 Analysis of United States Nitrogen Dioxide
Emission Sources, Levels, and Trends . . . 4-14
4 2.5 Comparison of Nitrogen Oxide Emissions Estimates 4-23
I-viii
-------
TABLE OF CONTENTS (cont'd)
4 3 INDOOR EMISSION SOURCES OF NITROGEN OXIDES 4-23
4.3.1 Introduction . . 4-23
432 Formation of Nitrogen Oxides in Combustion in
Gas-Fueled Household Appliances 4-24
433 Gas Stoves Used for Cooking 4-28
434 Unvented Space Heaters Fueled with Natural Gas
and Propane . . . 4-40
435 Kerosene Heaters , 4-45
436 Wood Stoves 4-47
437 Tobacco Products . 4-48
4 3.8 Comparison of Emissions from Sources Influencing
Indoor Air Quality 4-49
4 4 SUMMARY OF EMISSIONS OF NITROGEN OXIDES
FROM AMBIENT AND INDOOR SOURCES 4-51
REFERENCES . . 4-53
TRANSPORT AND TRANSFORMATION OF NITROGEN OXIDES 5-1
5 1 BACKGROUND 5-1
5 2 THE ROLE OF NITROGEN OXIDES IN OZONE
PRODUCTION , 5-3
521 Urban Plume Chemistry . . . 5-8
522 Ozone Production in Rural Environments 5-10
53 ODD NITROGEN SPECIES . 5-18
531 Nitric Acid 5-18
532 Nitrous Acid . . . 5-20
533 Peroxymtric Acid 5-20
534 Peroxyacylmtrates 5-21
535 Nitrate Radical 5-22
536 Dimtrogen Pentoxide . ... 5-24
537 Total Reactive Odd Nitrogen Species 5-25
538 Amines, Nitrosamines, and Nitramines 5-26
5 4 TRANSPORT 5-30
541 Transport of Reactive Nitrogen Species in
Urban Plumes 5-32
542 Transport and Chemistry in Combustion Plumes 5-34
543 Regional Transport 5-36
5 5 OXIDES OF NITROGEN AND THE GREENHOUSE
EFFECT . , 5-40
551 Ozone Greenhouse Effects Related to Nitrogen
Oxides 5-40
552 Nitrous Oxide Greenhouse Contributions 5-42
5 6 STRATOSPHERIC OZONE DEPLETION BY OXIDES
OF NITROGEN 5-44
5 7 DEPOSITION OF NITROGEN OXIDES; 5-48
I-ix
-------
TABLE OF CONTENTS (cont'd)
Page
5.7.1 Dry Deposition of Nitrogen Oxides 5-48
572 Methods for Determining Deposition Velocities 5-49
5721 Eddy Correlation . 5-50
5 7.2.2 Vertical Gradient Methods 5-50
5.7 2 3 Chamber Methods 5-51
573 Deposition of Nitrogen Oxides 5-51
5.7.4 Nitric Acid Deposition ... . 5-52
5.7.5 Deposition of Peroxyacetylmtrate 5-52
5 7.6 Wet Deposition of Nitrogen Oxides 5-52
5.8 SUMMARY AND CONCLUSIONS 5-53
5 8.1 Ozone Production 5-53
582 Production of Odd Nitrogen Species 5-53
583 Transport 5-54
5.8 3 1 General Features 5-54
5 8.3 2 Transport of Reactive Nitrogen Species
in Urban Plumes 5-55
5833 Transport and Chemistry in Combustion
Plumes 5-55
5834 Regional Transport 5-55
584 Oxides of Nitrogen and the Greenhouse Effect 5-56
5841 Nitrous Oxide Greenhouse Contributions 5-56
5 8.4 2 Stratospheric Ozone Depletion by Oxides
of Nitrogen 5-57
5.85 Deposition of Nitrogen Oxides 5-57
REFERENCES . 5-59
6. SAMPLING AND ANALYSIS FOR NITROGEN OXIDES
AND RELATED SPECIES . 6-1
6.1 INTRODUCTION . 6-1
6 2 NITRIC OXIDE 6-2
6.2 1 Chemiluminescence 6-2
6.2 2 Laser-Induced Fluorescence 6-5
6.2 3 Absorption Spectroscopy 6-6
6.2.4 Passive Samplers 6-8
6.2.5 Calibration 6-10
6.2.6 Intercompansons 6-10
6.2.7 Sampling Considerations for Nitric Oxide
and Other Nitrogen-Containing Species 6-12
6.3 NITROGEN DIOXIDE 6-12
6.3.1 Chermluminescence, Nitric Oxide Plus Ozone 6-13
6 3.2 Chemiluminescence,, Luminol 6-17
6.3.3 Photofragmentation/Two-Photon
Laser-Induced Fluorescence 6-18
6.3.4 Absorption Spectroscopy 6-19
I-x
-------
TABLE OF CONTENTS (cont'd)
7age
635 Wet Chemical Methods . 6-21
6351 Gness-Saltzman Method 6-21
6352 Continuous Saltzman Method 6-22
6353 Alkaline Guaiacol Method . 6-22
6354 Jacobs-Hochheiser Method 6-23
6355 Sodium Arsemte Method (Manual
and Continuous) . 6-23
6356 Tnethanolamine-Guaiacol-Sulfite Method 6-24
6 3 5.7 Tnethanolamine Method 6-24
636 Other Active Methods 6-25
637 Passive Samplers 6-27
638 Calibration . 6-30
639 Intercompansons 6-31
6 3 10 Designated Methods 6-34
6 4 NITROGEN OXIDES 6-38
6 5 TOTAL REACTIVE ODD NITROGEN OXIDES 6-40
6 6 PEROXYACETYL NITRATE . 6-41
661 Gas Chromatography-Electron Capture Detection 6-42
662 Alkaline Hydrolysis . 6-43
663 Gas Chromatography—Alternate Detectors 6-43
664 Peroxyacetyl Nitrate Stability 6-44
665 Calibration .... . 6-45
666 Other Organic Nitrates 6-46
6 7 NITRIC ACID 6-47
671 Filtration 6-47
672 Denuders ... . . 6-49
673 Chemiluminescence 6-51
674 Absorption Spectroscopy 6-52
675 Calibration 6-52
676 Intercompansons ... . . 6-53
68 NITROUS ACID . 6-55
681 Denuders 6-55
6.8 2 Chemiluminescence 6-56
683 Photofragmentation/Laser-Induced Fluorescence . 6-57
684 Absorption Spectroscopy 6-57
685 Calibration 6-58
686 Intercompansons 6-58
6 9 DINITROGEN PENTOXTOE AND NITRATE RADICALS 6-59
6 10 PARTICULATE NITRATE 6-60
6 10 1 Filtration 6-60
6 10 2 Denuders/Filtration 6-64
6 10 3 Impactors .... . 6-64
6 10 4 Analysis . 6-65
6 11 NITROUS OXIDE - 6-69
I-xi
-------
TABLE OF COlSfTENTS (cont'd)
Page
6.12 SUMMARY . . 6-70
REFERENCES . . . 6-72
7. AMBIENT AND INDOOR CONCENTRATIONS OF
NITROGEN OXIDES . 7-1
7.1 INTRODUCTION . 7-1
7.2 AMBIENT AIR CONCENTRATIONS OF NITROGEN
OXIDES ... 7-2
7.2.1 Introduction 7-2
7.2.2 Ambient Air Concentrations of Nitric Acid
and Nitrate Aerosol . 7-4
7 2.3 Ambient Air Concentrations of Nitric Oxide
and Nitrogen Dioxide 7-6
7231 Data Availability and Exposure
Considerations 7-6
7232 Trends in Ambient Nitrogen Dioxide
Concentrations 7-8
7 2 3.3 Exposure Patterns Observed for Ambient
Nitrogen Dioxide and Nitric Oxide
Concentrations—Urban 7-8
7.2 3 4 Exposure Patterns Observed for Ambient
Nitrogen Dioxide and Nitric Oxide
Concentrations—Rural Forest and
Agriculture Areas 7-24
7.3 INDOOR ATR CONCENTRATIONS OF NITROGEN
OXIDES . 7-32
7.3.1 Background 7-32
7.3.2 Residences Without Indoor Sources 7-38
7.3.3 Residences with Gas Appliances 7-46
7.3 3 1 Average Indoor Concentrations and
Estimated Source Contributions 7-47
7332 Spatial Distributions 7-59
7333 Short-Term Indoor Concentrations 7-60
7.3 4 Unvented Space Heaters 7-65
7341 Unvented Kerosene Space Heaters 7-66
7.3 4 2 Unvented Gas Space Heaters . 7-71
7.3.5 Other Sources . 7-76
7.3.6 Modeling of Indoor Concentrations 7-76
7361 Physical/Chemical Models 7-77
7362 Statistical/Empirical Models 7-80
7.3.7 Reactive Decay Rate of Nitrogen Dioxide Indoors 7-83
7.4 NITRIC AND NITROUS ACIDS CONCENTRATIONS 7-88
7.5 SUMMARY . 7-90
7.5 1 Ambient Nitrogen Dioxide Levels 7-90
I-xii
-------
TABLE OF CONTENTS (cont'd)
7 5.2 Indoor Nitrogen Dioxide Levels 7-91
REFERENCES 7-95
ASSESSING TOTAL HUMAN EXPOSURE TO
NITROGEN DIOXIDE 8-1
8 1 INTRODUCTION . . . . 8-1
8.2 DIRECT METHODS 8-4
821 Biomarkers 8-4
822 Personal Monitoring ... . . 8-5
8 3 INDIRECT METHODS . ... . .... 8-17
831 Personal Exposure Models 8-21
8 4 SUMMARY ... . . 8-24
REFERENCES . . 8-26
I-Xlll
-------
LIST OF TABLES
Number Page
1-1 Key Human Health Effects of Exposure to Nitrogen Dioxide—
Clinical Studies 1-20
1-2 Key Human Health Effects of Exposure to Nitrogen Dioxide—
Epidemiological Studies ... . . . 1-22
1-3 Key Animal Toxicological Effects of Exposure to
Nitrogen Dioxide 1-24
3-1 Theoretical Concentrations of Nitrogen Oxides and Nitrogen
Acids That Would Be Present at Equilibrium with Molecular
Nitrogen, Molecular Oxygen, and Water in Air at 25 °C,
One Atmosphere, 50% Relative Humidity 3-3
3-2 Some Physical and Thermodynamic Properties of the
Nitrogen Oxides 3-6
3-3 Theoretical Equilibrium Concentrations of Nitric Oxide and
Nitrogen Dioxide in Air (50% Relative Humidity) at Various
Temperatures 3-9
3-4 Theoretical Concentrations of Dimtrogen Tnoxide and Dimtrogen
Tetroxide in Equilibrium with Various Levels of Gaseous Nitric
Oxide and Nitrogen Dioxide in Air at 25 °C 3-13
4-1 Major Source Categories . 4-3
4-2 Global Budget of Nitrogen Oxides in the Troposphere 4-13
4-3 Estimates of Nitrogen Oxide Emissions from Anthropogenic and
Natural Sources in the United States and Canada 4-14
4-4 Total National Emissions of Nitrogen Oxides, 1940 to 1990 4-15
4-5 Transportation Contribution to United States Nitrogen Oxides
Emissions . 4-16
4-6 Breakdown of 1988 Transportation Nitrogen Oxides 4-17
4-7 Emissions of Nitrogen Oxides from Stationary Fuel
Combustion Sources, 1970 to 1990 . . 4-18
4-8 Total National Nitrogen Oxide Emissions, 1940 to 2010 . ... 4-19
I-xiv
-------
LIST OF TABLES (cont'd)
Number Page
4-9 Comparison of Anthropogenic and Natural Sources of Nitrogen
Oxides Emissions for 1990 4-22
4-10 Comparison of Annual United States Nitiogen Oxide Emissions
Estimates from Four Inventories 4-23
4-11 Emission Factors for Nitric Oxide and Nitrogen Dioxide from
Burners on Gas Stoves, After Himmel and Dewerth (1974) 4-30
4-12 Emission Factors for Nitric Oxide and Nitrogen Dioxide from
Pilot Lights on Gas Stoves, After Himmel and Dewerth (1974) 4-31
4-13 Emission Factors for Nitric Oxide, Nitrogen Dioxide, and
Nitrogen Oxides for Top Burners on Gas Stoves Measured with
a Sampling Hood and with a Chamber, After Cole et al (1983)
and Moschandreas et al (1985) 4-33
4-14 Emission Factors for Nitric Oxide, Nitrogen Dioxide, and
Nitrogen Oxides for Ovens, After Cole et al (1983) and
Moschandreas et al (1985) 4-34
4-15 Emission Factors for Nitric Oxide and Nitrogen Dioxide from
Pilot Lights on Gas Stoves, After Moschandreas et al (1985) 4-35
4-16 Emission Factors for Nitric Oxide and Nitrogen Dioxide from
Range-Top Burners of Improved Design, After Cole and
Zawacki (1985) .... 4-36
4-17 Emission Factors for Nitrogen Dioxide from 10 Gas Stoves in
Use in Residences, Measured Independently by Research
Groups 4-37
4-18 Average Emission Factors for Nitric Oxide, Nitrogen Dioxide,
and Nitrogen Oxides from Burners on Gas Stoves Based on Data
Reported in the Literature 4-37
4-19 Emission Factors for Nitrogen Oxide and Nitrogen Dioxide
for Unvented Space Heaters 4-42
4-20 Emission Factors for Nitrogen Oxide and Nitrogen Dioxide
for Convective and Infrared Heaters of Various Designs,
Using Natural Gas and Propane . . 4-44
I-xv
-------
LEST OF TABLES (cont'd)
Number
4-21 Average Emission Factors for Nitric Oxide and Nitrogen
Dioxide from Kerosene Heaters, After Leaderer (1982) and
Traynor et al (1983b) 4-46
4-22 Average Emission Factors for Nitric Oxide, Nitrogen Dioxide,
and Nitrogen Oxides from Various Sources Based on Data
Reported in the Literature . . 4-49
5-1 Major Reactions in the Nitrate Radical-Dimtrogen Pentoxide
System at Night . . 5-35
5-2 Average Afternoon Background Pollutant Concentrations
Measured at Kenosha, Wisconsin 5-38
6-1 Performance Specifications for Nitrogen Dioxide Automated
Methods . . . 6-35
6-2 Comparability Test Specifications for Nitrogen Dioxide 6-36
6-3 Reference and Equivalent Methods for Nitrogen Dioxide
Designated by the United States Environmental
Protection Agency . . . . 6-37
6-4 National Precision and Accuracy Probability Limit Values
Expressed as Percent for Continuous and Manual Methods for
Nitrogen Dioxide . 6-39
7-1 Average Nitrogen Oxides Concentrations Measured at United States
Nonurban Monitoring Locations 7-3
7-2 Average Mixing Ratios Measured at Isolated United States Rural
Sites and Coastal Inflow Sites . . 7-4
7-3 Average Concentrations of Nitoc Acid and Nitrate Ions Measured
at Rural Sites ..... 7-5
7-4 Average Concentrations of Nitric Acid and Nitrate Ions Measured
at Urban Sites 7-6
7-5 Characterization of Hourly Average Nitrogen Dioxide
Concentrations Near Selected Electrical Generating Plants 7-11
I-xvi
-------
LIST OF TABLES! (cont'd)
Number Page
7-6 Maximum Annual Average Nitrogen Dioxide Concentrations
Reported in United States Metropolitan Statistical Areas,
1988 to 1990 7-13
7-7 Maximum Hourly Average Nitrogen Dioxide Concentrations
Reported in United States Metropolitan Statistical Areas,
1988 to 1990 7-15
7-8 Maximum Hourly Average Nitac Oxide Concentrations Reported
in United States Metropolitan Statistical Areas, 1988 to 1990 7-16
7-9 Hourly Incidence of Nitrogen Dioxide Concentrations Greater
Than 0 2 ppm for Stations with More Than One Occurrence,
1988 7-20
7-10 Maximum Hourly Average Nitrogen Dioxide Concentrations for
Selected United States Rural Forested Sites 7-26
7-11 Characterization of Hourly Average Nitrogen Dioxide
Concentrations for Selected United States Forest Sites 7-27
7-12 Maximum Hourly Average Nitrogen Dioxide Concentrations for
Selected United States Rural Agricultural Sites 7-28
7-13 Characterization of Hourly Average Nitrogen Dioxide
Concentrations for Selected United States Agricultural Sites 7-29
7-14 Maximum Hourly Average Nitric Oxide Concentrations Reported
in Rural Areas, 1988 to 1990 7-33
7-15 Average Outdoor Concentrations of Nitrogen Dioxide and
Average Indoor/Outdoor Ratios in Homes Without Known Indoor
Sources from Field Studies of Private Residences 7-39
7-16 Indoor and Outdoor Concentrations of Nitrogen Dioxide in
Homes with Gas Appliances, and the Calculated Average
Contribution of Those Appliances to Indoor Residential
Nitrogen Dioxide Levels 7-50
7-17 Frequency Distribution for Type of Range from 1985 and
1991 Surveys 7-58
7-18 Average Number of Days of Range Use per Week for Cooking,
by Income 7-58
I-xvii
-------
LIST OF TABLES (cont'd)
Number Page
7-19 Reported Range/Oven Cooking Frequency in 1985 and 1991,
by Type of Other Cooking Appliance 7-59
7-20 Summary Statistics for Gas Range Nitrogen Dioxide Maxima
Over Several Averaging Times . 7-64
7-21 Two-Week Average Nitrogen Dioxide Levels by Location for
Homes in Six Principle Source Categories, New Haven,
Connecticut, Area Study, Winter, 1983 7-68
7-22 One-Week Average Nitrogen Dioxide Levels in Homes in
North Central Texas by Source Category, with and Without
Unvented Gas Space Heater 7-74
7-23 Empirical Statistical Models (Regression) for Residential
Nitrogen Dioxide Concentrations Reported from Field Studies
of Indoor Levels ... . 7-82
8-1 Electric-Range Home Least Squares Regression Coefficients
and T-Statistics . 8-22
8-2 Gas-Range Home Least Squares Regression Coefficients and
T-Statistics . 8-23
I-XVlll
-------
LIST OF FIGURES
Number
1-1 Nitrogen Cycle .... . . 1-10
3-1 Calculated steady states of the free troposphere as a function
of nitrogen oxides concentration 3-5
4-1 Production of hydrocarbons, carbon monoxide, and nitrogen oxide
as a function of arr-fuel ratio 4-5
4-2 National trend in United States nitrogen oxides emissions,
1981 to 1990 . 4-19
4-3 Emission of nitrogen oxides compared with nitrogen oxide
standards 4-21
4-4 Laminar blue-flame 4-25
4-5 Emission factors for nitric oxide as a function of gas flow rate 4-39
4-6 Emission factors for nitrogen dioxide as a function of gas
flow rate 4-40
5-1 Summary of the gas phase chemistry of nitrogen oxides in
the clean troposphere .... . . 5-2
5-2 Major chemical reactions affecting oxygen species in the
troposphere 5-3
5-3 Major chemical reactions affecting hydrogen species in the
troposphere 5-4
5-4 Schematic diagram of the combined read ions of nitrogen,
oxygen, and hydrogen 5-5
5-5 Volatile organic compound oxidation in the atmosphere 5-6
5-6 Calculated steady state concentrations in the free troposphere
as a function of nitrogen oxides concentration . 5-7
5-7 Summertime (June 1 to August 31) and wintertime (December 1
to February 28) ozone mixing ratio versus nitrogen oxides
mixing ratio during the morning and afternoon 5-14
5-8 Summertime ozone mixing ratio versus nitrogen oxides mixing
ratio measured during the afternoon houis 5-15
I-xrx
-------
LIST OF FIGURES (cont'd)
Number
5-9 Model calculated daytime change in ozone values (from
sunrise to 1630 hours) for summer clear sky conditions is
compared to the observed difference between the afternoon
(1400 to 1900 hours) and the morning (0700 to 1100 hours)
for clear sky conditions 5-16
5-10 Oxygen production per unit nitrogen oxides per day from
the NMHC-PO model is plotted as function of nitrogen oxides
mixing ratios . 5-17
5-11 Total reactive odd nitrogen species shortfall 5-26
5-12 Formation and decay of diethylnitrosamine in the dark and in
the sunlight from diethylamine and from tnethylamine 5-29
5-13 Pollutant levels at the Kenosha, Wisconsin, sampling site
before and after passage of the lake-breeze front 5-38
6-1 Absolute error in nitrogen dioxide for 10 seconds in the dark
sampling line 6-13
7-1 National trend in the composite annual average nitrogen dioxide
concentrations at both National Air Monitoring Station sites and
all sites with 95% confidence intervals, 1980 to 1989 7-9
7-2 United States metropolitan area trends in the composite annual
average nitrogen dioxide concentration, 1980 to 1989 7-10
7-3 Distribution of peak annual nitrogen dioxide averages in
103 Metropolitan Statistical Areas, 1988 to 1989, as derived by
the United States Environmental Protection Agency from
Aerometnc Information Retrieval System (1991) 7-14
7-4 Monthly 50th, 90th, and 98th percentiles of one-hour nitrogen
dioxide concentrations at selected stations, 1986 to 1989, as
derived by the United States Environmental Protection Agency
from Aerometnc Information Retrieval System (1991) 7-18
7-5 Annual average nitrogen dioxide versus second-high one-hour
concentration, 1988, 1989, and 1990 7-19
I-xx
-------
LIST OF FIGURES (cont'd)
Number
7-6 Hourly relative frequency distributions of one-hour nitrogen
dioxide values at four selected stations for 1988, with
numbers of values greater than 0 2 ppm, as derived by the
United States Environmental Protection Agency from
Aerometnc Information Retrieval System (1991) 7-21
7-7 Percent of one-hour nitrogen dioxide values above 0 03 and
0 05 ppm versus annual averages greater than 0 03 ppm, 1988,
as derived by the United States Environmental Protection
Agency from Aerometnc Information Retrieval System (1991) 7-23
7-8 Relative distributions of one-hour nitrogen dioxide values
at selected stations, 1988, as derived by the United States
Environmental Protection Agency from Aerometnc Information
Retrieval System (1991) . . . 7-24
7-9 Seasonal pattern for nitrogen dioxide concentrations at
rural and forested Aerometnc Information Retrieval System
monitoring sites . 7-30
7-10 Diurnal pattern for nitrogen dioxide at rural and forested
Aerometnc Information Retrieval System monitoring sites 7-31
7-11 Diurnal patterns for nitrogen dioxide monthly average
concentrations at selected rural forested Aerometnc
Information Retrieval System monitoring sites 7-31
7-12 Diurnal patterns for nitrogen dioxide monthly average
concentrations at selected rural agricultural Aerometnc
Information Retrieval System monitoring sites 7-32
7-13 Winter nitrogen dioxide concentrations by site and sampling
location . 7-35
7-14 Ratio of average indoor nitrogen dioxide to ambient nitrogen
dioxide concentrations by season and location in homes without
a nitrogen dioxide source 7-42
7-15 Cumulative frequency distribution of nitrogen dioxide
concentrations (one-week sampling penod) by location foi homes
with no known gas appliances for a winter period in Southc-m
California . . 7 41
I xxi
-------
LIST OF FIGURES (cont'd)
Number
7-16 Cumulative frequency distribution and arithmetic means of
nitrogen dioxide concentrations (two-week sampling period) by
location for homes with no kerosene heater and no gas range for
a winter period in the New Haven, Connecticut, area 7-43
7-17 Cumulative frequency distribution of nitrogen dioxide
concentrations (one-week sampling period) by location for homes
with no gas appliances for a summer period in Southern
California 7-44
7-18 Indoor/outdoor nitrogen dioxide concentration ratios (two-week
sampling periods) as a function of time for three homes in the
United Kingdom without indoor mtiogen dioxide sources . . 7-45
7-19 Concentrations of nitrogen dioxide from October through
March during 1988 and 1989, Albuquerque, New Mexico 7-48
7-20 Indoor versus outdoor nitrogen dioxide in five housing
developments in Chattanooga, Tennessee 7-49
7-21 Nitrogen dioxide concentrations across seasons in Albuquerque,
New Mexico, bedrooms that do and do not use the gas stove for
space heating . 7-56
7-22 Verticle distribution of average nitrogen dioxide concentrations
(48-hour sampling periods) measured in nine New York City
apartments 7-61
7-23 Mean nitrogen dioxide concentrations (one-week sampling periods)
for eight sampling periods by location in the home and type of
cooking fuel . . 7-62
7-24 Nitrogen dioxide hourly levels in one home with gas
appliances ... . . 7-65
7-25 Cumulative frequency distribution and arithmetic means by
location, of average nitrogen dioxide levels (two-week sampling
periods) during kerosene heater use for residences with one
kerosene heater and no gas range, New Haven, Connecticut,
area study, winter 1983 7-70
7-26 Cumulative frequency distributions and summary statistics for
integrated nitrogen dioxide measurements in two locations
(152 study homes) . .... 7-72
I-xxii
-------
LIST OF FIGURES (cont'd)
Number Page
7-27 Cumulative frequency distributions and summary statistics for
indoor nitrogen dioxide concentrations in three groups of
monitored homes 7-73
7-28 Nitrogen dioxide box plots for 12 continuously monitored
homes . . . 7-75
7-29 Bar graph of nitrogen dioxide removal rate for various materials
evaluated in a 1 64-cubic meter test chamber at 50% relative
humidity . . . 7-85
7-30 The deposition rates in air changes per hour for nitrogen as a
function of percent relative humidity for two surface areas of
three materials 7-87
7-31 Concentration distributions for gas-phase species in
Boston nitrous acid and nitric acid 7-89
8-1 Average personal nitrogen dioxide exposure for each household
compared with outdoor concentrations for summer and winter 8-7
8-2 Average personal nitrogen dioxide exposure for each home
compared with average indoor concentrations for summer
and winter 8-8
8-3 Comparison of the house average two-week nitrogen dioxide
concentrations with the total personal nitrogen dioxide
levels measured over the same tune period for one adult
resident in each house, New Haven, Connecticut, area,
winter 1983 .... 8-10
8-4a Proportion of tune spent by women who are full-tune
homemakers in indoor, outdoor, and in-transit
microenvironments 8-20
8-4b Proportion of time spent by employed persons in indoor,
outdoor, and in-transit microenvironmenls 8-20
I-XXlll
-------
-------
AUTHORS
Chapter 1. Executive Summary of Air Quality Criteria for Oxides of Nitrogen
Dr Dennis J Kotchmar
Environmental Criteria and Assessment
Office
U S Environmental Protection Agency
Research Triangle Park, NC 27711
Mr William G Ewald
Environmental Criteria and Assessment
Office
U S Environmental Protection Agency
Research Triangle Park, NC 27711
Dr J H.B. Garner
Environmental Criteria and Assessment
Office
U S. Environmental Protection Agency
Research Triangle Park, NC 27711
Dr Judith A Graham
Environmental Criteria and Assessment
Office
U.S Environmental Protection Agency
Research Triangle Park, NC 27711
Dr Lester D. Grant
Environmental Criteria and Assessment
Office
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Ms Beverly Tilton
Environmental Criteria and Assessment
Office
U.S Environmental Protection Agency
Research Triangle Park, NC 27711
Chapter 2 Introduction
Dr Dennis J. Kotchmar
Environmental Criteria and Assessment
Office
U S Environmental Protection Agency
Research Triangle Park, NC 27711
Chapter 3: General Chemical and Physical Properties of Oxides of Nitrogen and
Oxides of Nitrogen-Derived Pollutants
Dr Robert W. Elias
Environmental Criteria and Assessment
Office
U S Environmental Protection Agency
Research Triangle Park, NC 27711
Mr. William G Ewald
Environmental Criteria and Assessment
Office
U S. Environmental Protection Agency
Research Triangle Park, NC 27711
I-xxv
-------
AUTHORS (cont'd)
Chapter 4- Ambient and Indoor Sources and Emissions of Nitrogen Oxides
Dr. Charles A. Amann Dr Gregory J McRae
KAB Engineering 702 Woodland Dnve
984 Satlerlee Road Ohara Township
Bloomfield Hills, MI 48304-3152 Pittsburgh, PA 15238
Dr. Cliff I Davidson
Department of Civil Engineering
Carnegie Mellon University
Pittsburgh, PA 15213
Chapter 5 Transport and Transformation of Nitrogen Oxides
Dr. Halvor Westberg
Laboratory for Atmospheric Research
Washington State University
Pullman, WA 99164-2730
Chapter 6. Sampling and Analysis for Nitrogen Oxides and Related Species
Dr. Joseph E. Sickles n
Atmospheric Research and Exposure Assessment
Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Chapter 7 Ambient and Indoor Concentrations of Nitrogen Oxides
Dr. Robert W Elias Dr Allen S LeFohn
Environmental Criteria and Assessment Office A S L Associates
U.S. Environmental Protection Agency 111 Last Chance Gulch
Research Triangle Park, NC 27711 Helena, MT 59601
Dr. Brian Leaderer Mr Tom McMullen
Pierce Foundation Laboratory Environmental Criteria and Assessment
290 Congress Avenue Office
New Haven, CT 06519 U S Environmental Protection Agency
Research Triangle Park, NC 27711
I-xxvi
-------
AUTHORS (cont'd)
Chapter 8 Assessing Total Human Exposure to Nitrogen Dioxide
Dr Brian Leaderer
Pierce Foundation Laboratory
290 Congress Avenue
New Haven, CT 06519
I-xxvii
-------
-------
CONTRIBUTORS AND REVIEWERS
Dr A Paul Altshuller
Atmospheric Research and Exposure
Assessment Laboratory
U S Environmental Protection Agency
Research Triangle Park, NC 27711
Dr Michael Berry
Environmental Criteria and Assessment Office
U S Environmental Protection Agency
Research Triangle Park, NC 27711
Dr IrwinH Billick
Gas Research Institute
8600 West Byrn Mawr Avenue
Chicago, IL 60631
Dr Steven D Colome
Integrated Environmental Services
University Tower, Suite 1090
4199 Campus Drive
Irvine, CA 92715
Dr. Cliff Davidson
Department of Civil Engineering
Carnegie Mellon University
Pittsburgh, PA 15213
Dr. Marcia Dodge
Air and Energy Engineering Research
Laboratory
U S Environmental Protection Agency
Research Triangle Park, NC 27711
Dr Robert Elias
Environmental Criteria and Assessment Office
U S Environmental Protection Agency
Reseach Triangle Parl£ NC 27711
Dr Don Fox
School of Public Health
ENVR-CB #7400
University of North Carolina
Chapel Hill, NC 27599-7400
Dr Vic Hasselblad
Center for Health Policy Research
Duke University
Durham, NC 27713
Dr Brian Heikes
Graduate School of Oceanography
University of Rhode Island
Narrangansett Bay Campus
Narrangansett, RI 02882-1197
Ms Pamela Johnson
Office of Air Quality Planning and Standards
U S Environmental Protection Agency
Research Triangle Park, NC 27711
Dr Dennis J Kotchmar
Environmental Criteria and Assessment Office
U S Environmental Protection Agency
Research Triangle Park, NC 27711
Mr Douglas Latimer
Latimer & Associates
505 27th Way #301
Boulder, CO 80303
Dr Brian Leaderer
Pierce Foundation Laboratory
Yale University School of Medicine
290 Congress Avenue
New Haven, CT 06519
Mr Frank McElroy
Atmospheric Research and Exposure
Assessment Laboratory
U S Environmental Protection Agency
Research Triangle Park, NC 27711
Dr J David Mobley
Air and Energy Engineering Research
Laboratory
U S Environmental Protection Agency
Research Triangle Park, NC 27711
I-xxix
-------
CONTRIBUTORS AND REVIEWERS (cont'd)
Mr. Warren Porter
U.S. Consumer Products Safety Commission
5401 Westbard Avenue
Room 724
Bethesda, MD 20816
Dr. P. Barry Ryan
Department of Environmental Science and
Physiology
Harvard School of Public Health
677 Huntington Avenue
Boston, MA 02115
Dr. Joseph E. Sickles n
Atmospheric Research and Exposure Assessment
Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Dr. Thomas Stock
University of Texas
School of Public Health
P.O. Box 20186
Houston, TX 77225
Ms. Beverly Tilton
Environmental Criteria and Assessment Office
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Mr. John H Wasser
Air and Energy Engineering Research
Laboratory
U S Environmental Protection Agency
Research Triangle Park, NC 27711
Dr HalWestberg
Laboratory for Atmospheric Research
Washington State University
Pullman, WA 99164-2730
Mr James White
Air and Energy Engineering Research
Laboratory
U S Environmental Protection Agency
Research Triangle Park, NC 27711
Dr Warren White
6840 Waterman Avenue
St Louis, Missouri 63130
Dr Ron Wyzga
Electric Power Research Institute
3412 Hillview Avenue
P O Box 10412
Palo Alto, CA 94303
I-xxx
-------
Former Chairman
US ENVIRONMENTAL PROTECTION AGENCY
SCIENCE ADVISORY BOARD
CLEAN AIR SCIENTIFIC ADVISORY COMMITTEE
Oxides of Nitrogen Review
Chairman
Dr Roger O McClellan
Chemical Industry Institute of Toxicology
PO Box 12137
Research Triangle Park, NC 27709
Dr George T Wolff
General Motors Research Laboratories
Environmental Science Department
Warren, MI 48090
Members
Dr GlenR Cass
Environmental Engineering Science
Department
Mail Code 138-78
California Institute of Technology
Pasadena, CA 91125
Dr Jean Ford, Medical Director
Harlem Hospital Center
506 Lenox Avenue
New York, NY 10037
Dr Benjamin Liu
University of Minnesota
125 Mechanical Engineering
111 Church Street, S E
Minneapolis, MN 55455-0111
Consultants
Dr William C Adams
Human Performance Laboratory
Department of Physical Education
University of California
Davis, CA 95616
Dr Joseph Mauderly
Inhalation Toxicology Research Institute
PO Box 5890
Albuquerque, NM 87185
Dr Marc B Schenker
Division of Occupational and Environmental
Medicine
IEHR Building
University of California
Davis, CA 95616
Dr MarkJ Utell
Pulmonary Disease Unit
Box 692
University of Rochester Medical Center
601 Elmwood Avenue
Rochester, NY 14642
Dr. John Balmes
San Francisco General Hospital
Occupational Health Clinic
Building 9, Room 109
San Francisco, CA 94110
I-xxxi
-------
CLEAN AIR SCIENTIFIC ADVISORY COMMITTEE (cont'd)
Consultants fcont'd)
Dr. Douglas Dockery
Harvard School of Public Health
Department of Environmental Science and
Physiology
665 Huntington Avenue
Boston, MA 02115
Dr. James Fenters
IIT Research Institute
10 West 35th Street
Chicago, IL 60616
Dr. Gareth Green
Harvard School of Public Health
677 Huntington Avenue
Boston, MA 02115
Dr. Robert Mercer
Center for Extrapolation Modeling
Box 3177
Duke University Medical Center
Department of Medicine
Durham, NC 27710
Dr JohnSkelly
Department of Plant Pathology
212A Buckhout Laboratory
Pennsylvania State University
University Park, PA 16802
Dr Michael J Symons
School of Public Health
Room 3104D
McGavran Greenberg Hall
University of North Carolina at Chapel Hill
Chapel Hill, NC 27599
Dr Warren White
8840 Waterman Avenue
St Louis, MO 63130
Designated Federal Official
Mr. Randall C. Bond
U.S. Environmental Protection Agency
Science Advisory Board (A-101F)
401 M Street, S.W
Washington, DC 20460
Staff Secretary
Ms Janice Jones
U S Environmental Protection Agency
Science Advisory Board (A-101F)
401 M Street, S W
Washington, DC 20460
I-xxxii
-------
PROJECT TEAM FOR DEVELOPMENT OF
AIR QUALITY CRITERIA FOR OXIDES OF NITROGEN
Scientific Staff
Dr Dennis J Kotchmar, Project Manager
Environmental Criteria and Assessment Office
(MD-52)
U S Environmental Protection Agency
Research Triangle Park, NC 27711
Ms Beverly Comfort
Environmental Criteria and Assessment Office
(MD-52)
U S Environmental Protection Agency
Research Triangle Park, NC 27711
Dr Robert W Elias
Environmental Criteria and Assessment Office
(MD-52)
U S Environmental Protection Agency
Research Triangle Park, NC 27711
Mr William G Ewald
Environmental Criteria and Assessment Office
(MD-52)
U S Environmental Protection Agency
Research Triangle Park, NC 27711
Dr J H B Garner
Environmental Criteria and Assessment Office
(MD-52)
U S Environmental Protection Agency
Research Triangle Park, NC 27711
Mr Thomas B McMullen
Environmental Criteria and Assessment Office
(MD-52)
U S Environmental Protection Agency
Research Triangle Park, NC 27711
Ms Ellie R Speh, Office Manager
Environmental Criteria and Assessment Office
(MD-52)
U S Environmental Protection Agency
Research Triangle Park, NC 27711
Ms Beverly Tilton
Environmental Criteria and Assessment Office
(MD-52)
U S Environmental Protection Agency
Research Triangle Park, NC 27711
Technical Support Staff
Mr Douglas B Fennell, Technical
Information Specialist
Environmental Criteria and Assessment Office
(MD-52)
U S Environmental Protection Agency
Research Triangle Park, NC 27711
Mr Allen G Hoyt, Technical Editor and
Graphic Artist
Environmental Criteria and Assessment Office
(MD-52)
U S. Environmental Protection Agency
Research Triangle Park, NC 27711
Ms Diane H Ray, Technical Information
Manager (Public Comments)
Environmental Criteria and Assessment Office
(MD-52)
U S Environmental Protection Agency
Research Triangle Park, NC 27711
Mr Richard N Wilson, Clerk
Environmental Criteria and Assessment Office
(MD-52)
U S Environmental Protection Agency
Research Triangle Park, NC 27711
I-xxxiu
-------
PROJECT TEAM FOR DEVELOPMENT OF
AIR QUALITY CRITERIA FOR OXIDES OF NITROGEN (cont'd)
Document Production Staff
Ms. Marianne Earner, Graphic Artist
ManTech Environmental Technology, Inc
P.O. Box 12313
Research Triangle Park, NC 27709
Mr. John R. Barton, Document Production
Coordinator
ManTech Environmental Technology, Inc
P.O. Box 12313
Research Triangle Park, NC 27709
Ms. Lynette D. Cradle, Lead Word
Processor
ManTech Environmental Technology, Inc
P.O. Box 12313
Research Triangle Park, NC 27709
Ms. Jorja R. Followill, Word Processor
ManTech Environmental Technology, Inc
P.O. Box 12313
Research Triangle Park, NC 27709
Ms Wendy B Lloyd, Word Processor
ManTech Environmental Technology, Inc.
PO Box 12313
Research Triangle Paik, NC 27709
Mr J Derrick Stout, Graphic Artist
ManTech Environmental Technology, Inc
P O. Box 12313
Research Triangle Paik, NC 27709
Mr Peter J Winz, Technical Editor
ManTech Environmental Technology, Inc
PO Box 12313
Research Triangle Park, NC 27709
Technical Reference Staff
Mr. John A. Bennett, Bibliographic Editor
ManTech Environmental Technology, Inc
P.O. Box 12313
Research Triangle Park, NC 27709
Ms. Susan L. McDonald, Bibliographic
Editor
Research Information Organizers
P.O. Box 13135
Research Triangle Park, NC 27709
Ms. Blythe Hatcher, Bibliographic Editor
Research Information Organizers
P.O. Box 13135
Research Triangle Park, NC 27709
Ms Deborah L Staves, Bibliographic
Editor
Research Information Organizers
P O Box 13135
Research Triangle Park, NC 27709
Ms Patricia R Tierney, Bibliographic
Editor
ManTech Environmental Technology, Inc
P O Box 12313
Research Triangle Park, NC 27709
I-xxxiv
-------
1. EXECUTIVE SUMMARY OF AIR QUALITY
CRITERIA FOR OXIDES OF NITROGEN
1.1 PURPOSE OF THE DOCUMENT
• The purpose of this document is to present air quality criteria for oxides of
nitrogen (NOX) in accordance with Sections 108 and 109 of the Clean Air Act
Section 108 (U S Code, 1991) directs the Administrator of the U S
Environmental Protection Agency (EPA) to list pollutants that may reasonably be
anticipated to endanger public health and welfare, and to issue air quality criteria
for them These air quality criteria are to reflect the latest scientific information
useful in indicating the land and extent of all identifiable effects on public health
and welfare that may be expected from the presence of the pollutant in the
ambient air
1.2 INTRODUCTION
• The present document, Air Quality Criteria for Oxides of Nitrogen, discusses the
latest scientific information useful in deriving criteria to serve as scientific bases
for EPA decisions regarding National Ambient Air Quality Standards (NAAQS)
for nitrogen dioxide (NO2) and/or other NOX compounds The document is
comprised of 16 chapters This Executive Summary concisely summarizes key
conclusions from the document The following subsections follow the chapter
organization
1.3 CHEMICAL AND PHYSICAL PROPERTIES OF NITROGEN
OXIDES AND NITROGEN OXIDE-DERIVED POLLUTANTS
• Discussion of the general chemical and physical properties of NOX and
NOx-denved pollutants is necessary for ml reduction to the complex chemical and
physical interactions that may occur in the atmosphere and other media In this
document, NOX is the sum of NO2 and nitric oxide (NO), and NOy refers to the
sum of NOX and other oxidized nitrogen compounds, except nitrous oxide (N2O)
These other compounds include nitric acid (HNO3), nitrogen tnoxide (NO3),
dimtrogen tnoxide (N2O3), dimtrogen tetroxide (N2O4), and dimtrogen pentoxide
(N2O5) Peroxyacetylmtrate (PAN) is nominally included with the NOy group of
compounds
• There are seven oxides of nitrogen that may be present in the ambient air NO,
NO2, N2O, NO3, N2O3, N2O4, and N2O5 Of these, NO and NO2 are generally
1-1
-------
present in highest concentrations in the lower troposphere Their
interconvertibihty in photochemical smog reactions has frequently resulted in
their being grouped together under the designation NOX, although analytic
techniques can distinguish clearly between them Of the two, NO2 has the
greater impact on human health
• Nitrous oxide is ubiquitous even in the absence of anthropogenic sources because
it is a product of natural biologic processes in soil It is not known, however, to
be involved in any photochemical smog reactions Although N2O is not
generally considered to be an air pollutant, it participates in upper atmospheric
reactions involving the ozone (O3) layer
• Although N03, N2O3, N2O4, and N2O5 are present in the lower atmosphere only
in very low concentrations, even in polluted environments, they play a role in
atmospheric chemical reactions leading to the transformation, transport, and
ultimate removal of nitrogen compounds from ambient air
• Ammonia (NH3) is generated, on a global scale, during the decomposition of
nitrogenous matter in natural ecosystems, and it may also be produced locally in
larger concentrations by human activities such as the maintenance of dense
animal populations It is discussed because, through its reaction with HNO3,
resulting in the formation of aerosol nitrate, it plays an important role in
determining the atmospheric fate of nitrogen oxides
• Other NOx-denved compounds that may be found in polluted air include nitrites,
nitrates, nitrogen acids, 2V-mtroso compounds, and organic compounds such as
the peroxyacyl nitrates (RC(O)OONO2, where R represents any one of a large
variety of possible organic groups)
1.4 EMISSIONS OF NITROGEN OXIDES FROM AMBIENT AND
INDOOR SOURCES
• Anthropogenic sources of NO2 emissions include transportation, stationary
source fuel combustion, various industrial processes, solid waste disposal, and
others, such as forest fires Natural sources of NOX are lightning, biological and
abiological processes in soil, and stratospheric intrusion
• Estimates for 1990 indicate that more than 80 % of United States NOX emissions
are emitted by highway vehicles, electric utilities, and industrial boilers
Quantitative estimates of the total amount of NOX emitted to the ambient global
atmosphere are available These estimates suggest that 122 to 152 x 10 metric
tons of NOX are emitted annually, with about 18 to 19 x 106 metric tons emitted
in the United States alone
1-2
-------
• The important indoor sources of NOX are gas stoves, unvented space heaters,
kerosene heaters, wood stoves, and tobacco products Total emissions and the
ratio of NO/NO2 from gas stoves and space heaters differ according to fuel flow
rate and flame adjustment Additional factors, such as the load (e g , cold pot of
water), heater type (convective versus radiant), and fuel type (natural gas,
propane, or kerosene) may also be important Only limited information is
available for wood stoves and tobacco products
1.5 TRANSPORT AND TRANSFORMATION OF NITROGEN OXIDES
• Nitrogen oxides are important chemical species in the planetary boundary layer,
as well as in the free troposphere and the stratosphere Nitrogen oxides play
important roles in the control of concentrations of radicals in the troposphere, in
the production of troposphenc O3, as an aerosol precursor, and in the production
and deposition of acidic species, directly or indirectly
• Combustion processes emit a variety of nitrogen compounds, but chiefly NO,
which can be oxidized to NO2 in ambient air in the presence of O3 or in a
photochemically reactive atmosphere Photolytic decomposition of NO2 leads to
regeneration of NO, producing also an excited oxygen atom that reacts with
molecular oxygen to form O3 Free radicals generated from the oxidative
degradation of volatile organic compounds (VOCs) oxidize NO to NO2 without
destroying O3 Thus, the amount of O3 formed in ambient air is dependent upon
the concentration of NOX present as well as the concentrations and reactivities of
VOC species
• Photochemical processes that include the coupled reactions of NOX, oxygen
species, and free radicals produce not only O3, but nitrogen-containing products
as well These oxidation products include HNO3, peroxynitnc acid, nitrous acid
(HONO), RC(O)O2NO2, N2O5, and inorganic and organic nitrates
• Nitric acid is a major sink for active nitrogen and is a contributor to acidic
deposition It has been estimated to account for roughly one-third of the total
acidity deposited in the eastern United States Potential physical and chemical
sinks for HNO3 include wet and dry deposition, photolysis, reaction with
hydroxyl (OH) radicals, and two processes that lead to aerosol production
neutralization with gaseous NH3 and reactions with alkaline soil particles
• Peroxyacyl nitrates are formed from the combination of organic peroxy radicals
with NO2 Peroxyacetylmtrate is the most abundant member of this homologous
series of compounds in the lower troposphere It can serve in the troposphere as
a temporary reservoir for reactive nitrogen species and can be regionally
transported, but it cannot function as a true smk in the lower troposphere because
of its thermal instability In the upper troposphere, where temperatures are
colder, the lifetime of PAN is longer
1-3
-------
• The NO3 radical is a short-lived radical that is formed in the troposphere
primarily by the reaction of NO2 with O3 In daylight, NO3 undergoes rapid
photolysis or reaction with NO After sunset, accumulation of NO3 can occur
and is expected to be controlled by the availability of NO2 and O3 plus chemical
destruction mechanisms involving the formation of N2O5 and HNO3
• Dimtrogen pentoxide, the anhydride of HNO3, is primarily a nighttime
constituent of ambient air because it is formed from the reaction of NO3 (itself a
nighttime species) and NO2 Dimtrogen pentoxide is thermally unstable, but at
the lower temperatures of the upper troposphere, it can serve as a temporary
reservoir of NO3 In the boundary layer, N2C>5 reacts heterogeneously with
water to form HNO3, which in turn is deposited out
• Amines, nitrosamines, and mtramines are thought to exist in ambient air, but at
low concentrations Both nitrosamines and mtramines have short lifetimes in
ambient air because they are photolytically decomposed or react with OH
radicals and O3
• The transport and dispersion of the various nitrogenous species are dependent on
both meteorological and chemical parameters Advection, diffusion, deposition,
and chemical transformations combine to dictate the atmospheric residence times,
in turn, atmospheric residence tunes help determine the geographic extent of
transport of a given species Surface emissions are dispersed vertically and
horizontally through the atmosphere by tubulent mixing processes that are
dependent to a large extent on the vertical temperature structure and wind speed
• As the result of meteorological processes, NOX emitted in the early morning
hours in an urban area will disperse vertically and horizontally (downwind) as
the day progresses On sunny summer days, most of the NOX will have been
converted to HNO3 and PAN by sunset Much of the HNO3 is removed by
deposition as the air mass is transported, but HNO3 and PAN carried in layers
aloft (above the nighttime inversion layer, but below a higher subsidence
inversion) can potentially be transported long distances
• Transport of reactive NOX in regional air masses can occur via several
mechanisms (1) mesoscale phenomena, such as mountain-valley wind flow or
land-sea breeze circulations (transport for tens to hundreds of kilometers),
(2) synoptic weather systems, such as the migratory highs that cross the eastern
United States in the summertime (transport for many hundreds of kilometers),
and (3) mesoscale phenomena coupled with slow-movmg high-pressure systems
having weak pressure gradients In the latter interrelated phenomena, mountain-
valley or land-water breezes can govern pollutant transport in the immediate
vicinity of sources, but the ultimate fate of reactive NOX species will be
distribution into the synoptic system Information remains sparse on NOX
species and their concentrations in synoptic transport systems
1-4
-------
• Except for N2O, the reactive nitrogen species comprising the NOX and NOy
families in the atmosphere do not absorb infrared radiation and, therefore, do not
contribute directly to radiative "greenhouse" forcing They can, however,
contribute indirectly to greenhouse processes through the photochemical
production of O3 in the troposphere Nitrous oxide, which is chemically inert in
the troposphere, readily absorbs infrared radiation and is among the more
significant non-carbon dioxide greenhouse gases Absorption of visible radiation
by NO2 could make this compound a possible source of other climatic influences
if atmospheric concentrations become sufficiently higher
• Both wet and dry deposition of NOX and other nitrogen species occur, but wet
deposition is not a significant removal mechanism for NO or NO2 because both
gases are minimally soluble in water Transformation to more highly oxidized
forms is necessary for effective wet deposition of NOX, and the reaction of NO2
with the OH radical to form HNO3 appears to be the main source of nitrate ions
in precipitation About one-third of the emissions of NOX in the United States
are estimated to be removed by wet deposition
• Dry deposition fluxes for NOX are highly uncertain, mainly because of analytical
problems and the simultaneous occurrence of emission and deposition of NOX
Available data indicate, however, that NO emissions exceed NO deposition and
that NO2 deposition exceeds NO deposition
1.6 SAMPLING AND ANALYSIS FOR OXIDES OF NITROGEN AND
RELATED SPECIES
• In 1987, EPA designated a Reference Method and Equivalent Methods for NO2,
which specify a measurement principle and calibration procedures, namely gas-
phase chemiluminescence (GP-CLM) with calibration using either gas-phase
titration of NO with O3 or an NO2 permeation device The Sodium Arsenite
Method in both the manual and continuous forms and the Tnethanolamine-
Guaiacol-Sulfite Method have been designated as Equivalent Methods
Subsequently, commercial GP-CLM instruments were designated as Reference
Methods The sensitivity of these devices was in the low parts-per-billion range,
and, although the GP-CLM instruments were recognized as being susceptible to
interferences by other nitroxy species, it was believed that the atmospheric
concentrations of these compounds were generally low relative to NO2
• Information from air quality monitoring networks is now readily available and
has shown the GP-CLM instruments to have nominal precision and accuracy of
±10 to 15% and 20%, respectively, and to have replaced manual methods to a
large extent in network applications Although the basic design and performance
of the commercial instruments have remained essentially unchanged, researchers
have improved GP-CLM measurement technology and have refined other
1-5
-------
instrumental methods to permit the determination of NO, NO2, and NOy in the
low parts-per-tallion range
A continuous liquid phase CLM device for sensitively detecting NO2 has been
developed and may be suitable to measure NO2 Passive samplers for NO2 have
been used primarily for workplace and indoor applications, but hold promise for
ambient measurements as well Gas chromatography with electron capture
detection is useful in the determination of PAN, other organic nitrates, and N2O
Laser-induced fluorescence has been introduced to detect NO, NO2, and HONO
with high sensitivity and specificity Tunable-diode laser spectroscopy has been
used to detect NO, NO2, and HNO3 Long-path spectroscopy has also been used
to detect NO, NO2, HONO, and NO3 Two-tone frequency modulated
spectroscopy holds promise for the sensitive measurement of NO, NO2, PAN,
HNO3, and N2O
Interest in acidification of the environment has resulted in the development of
methods for HONO and HNO3 Integrative methods using denuders have been
introduced to permit sensitive determination of these and other species
1.7 AMBIENT AND INDOOR CONCENTRATIONS OF OXIDES OF
NITROGEN
1.7.1 Ambient Nitrogen Dioxide Levels
• Nitrogen oxides concentrations in isolated rural sites and coastal inflow areas in
the United States generally range from a few tenths to 1 ppb The concentrations
in the atmospheric boundary layer and lower free troposphere in remote maritime
locations are in the range 0 02 to 0 04 ppb, and concentrations of NOX in remote
tropical forests have been reported to range from 0 02 to 0 08 ppb
• The average concentrations of HNO3 and nitrate ions (NO3") are generally in the
range 0 1 to 20 ppb and 0 1 to 10 ppb, respectively Because there are
conflicting reports on the ability of filters to accurately separate HNO3 from
NO3" aerosol, it may be more appropriate in some cases to focus on the total
NO3 (HNO3 + NO3") than on the individual components
• Analysis of NO2 data in the Aerometnc Information Retrieval System data base
for the period 1981 to 1990 indicates a downward trend for the composite United
States annual average NO2 concentration The 1990 composite NO2 average was
8% less than the 1981 average, and the difference was statistically significant
• The highest hourly and annual ambient NO2 levels, which are reported from
stations in Southern California, can exceed the annual NO2 NAAQS of
0.053 ppm The seasonal patterns at California stations are usually quite marked
1-6
-------
and reach their highest levels during the fall and winter months For most of the
other urban sites characterized, the highest monthly average NO2 concentrations
were obtained in the months of November, December, January, or February
• The diurnal patterns of NO2 for the urban sites showed that, on the average, the
highest concentrations occur in the late afternoon and evening hours (1700 to
2200 hours) For those urban areas experiencing hourly NO2 concentrations
> 0 2 ppm, the episodic occurrences are experienced usually in the midmorning
and afternoon/evening hours
• Based on data collected at rural locations for the period 1979 to 1991, the hourly
average NO2 concentrations for selected U S forest and agricultural sites were
< 0 10 ppm in most cases As observed for urban locations, a consistent
seasonal pattern was distinguishable for both the rural forested and agricultural
sites In general, the NO2 monthly average values were at their highest during
the fall and winter months A consistent diurnal pattern was also observed for
the rural forested and agricultural sites, late afternoon and evening hours
(approximately 1700 to 2200 hours) contained the highest NO2 concentrations
1.7.2 Indoor Nitrogen Dioxide Levels
• Indoor concentrations of NO2 are a function of outdoor concentrations, indoor
sources (source type, condition of source, source use, etc ), infiltration
ventilation, air mixing within and between rooms, reactive decay by interior
surfaces, and air cleaning or source venluig In homes without indoor sources of
NO2, concentrations are lower than outdoor levels due to removal by the
building envelope and interior surfaces
• Gas appliances (gas range/oven, water heater, etc ) are the major indoor source
category for indoor residential NO2 Nitrogen dioxide levels in homes with gas
appliances are higher than those without such appliances and are often higher
than levels encountered outdoors Within this catagory, the gas range/oven is a
major contributor, especially when used as a supplemental heat source Average
indoor concentrations in bedrooms (over a 1- to 2-week measurement period)
range from 20 to 120 jtcg/m3 (0 010 to 0 064 ppm) in some homes with gas
ranges Homes with gas ranges with pilot lights have higher NO2 levels than
homes that have gas ranges without pilol lights
• Very limited data exist on short-term (3-h or less) average indoor concentrations
of NO2 associated with gas appliance use The limited data suggest that short-
term indoor averages of NO2 are higher than those recorded for outdoors
• Unvented kerosene and gas space heaters are important sources of NO2 in homes
because of both the NO2 production rate of the heaters and the length of tune the
heaters are used Field studies indicate that average residential concentrations
1-7
-------
(1- or 2-week average levels) exhibit a wide distribution, varying primarily with
the amount of heater use and type of heater
• Efforts to model indoor NO2 levels have employed both (1) physical/chemical
and (2) empirical/statistical models These models have been used, with varying
success, in explaining measured indoor levels of NO2 for predicting NO2 for
specific indoor settings, and to estimate indoor concentration distributions
Various empirical/statistical models have also been developed from large field-
study data bases These employ questionnaire responses and measured physical
data (house volume, etc ) as key independent variables
• Indoor concentrations of HONO appear to be higher than outdoors, even when
indoor NO2 concentrations do not exceed outdoor levels In homes where
unvented combustion sources are used, elevated HONO levels may be associated
with direct emissions of HONO from the flame as well as with heterogeneous
reactions of the produced NO2 with water Nitric acid has been measured
indoors during a summer period at concentrations lower than ambient
1.8 ASSESSING TOTAL HUMAN EXPOSURE TO NITROGEN
DIOXIDE
• Exposure to NO2 occurs across a number of microenvironments or settings
An individual's integrated exposure (E) is the sum of all of the individual NO2
exposures (£z) over all time intervals for all microenvironments, weighted by the
multiplicative product of the time (jQ in each microenvironment times the NO2
concentration (q) in the microenvironment The assessment of human exposures
to NO2 can be represented by the following simplified basic model
E* V1 "C1 V1 -P s*
& - Li &i ~ LiJici
Accurate assessments of total NO2 exposure and the environments in which
exposures take place are essential to minimize misclassification errors in
epidemiologic studies, in defining population exposure distributions in risk
assessment, and in developing effective mitigation measures in risk management
• Personal NO2 exposures can be assessed A limited number of studies have been
conducted in which personal exposures to NO2 were measured using passive
monitors These studies generally indicate that outdoor levels of NO2, although
related to and contributing substantially to both personal levels and indoor
concentrations, are by themselves poor predictors of personal exposures for most
populations Average indoor residential concentrations (e g , whole-house
average or bedroom level) tend to be the best predictor of personal exposure,
typically explaining 50 to 60% of the variation in personal exposures
In selected populations, the indoor residential environment may not be a good
1-8
-------
predictor of total exposure because of the higher percentages of time spent in
different environments and/or the potential for unusual NO2 concentrations
1.9 EFFECTS OF NITROGEN OXIDES ON VEGETATION
• Nitrogen oxides emitted into the atmosphere have the potential for influencing
plant growth through either the leaves or the roots of plants Of the various
nitrogen oxides found in the ambient air, only NO and NO2 are considered
phytotoxic However, the concentrations of gaseous nitrogen oxides in the
atmosphere, by themselves or in combination with O3 or sulfur dioxide, rarely
are high enough to influence plant growth
1.10 EFFECTS OF NITROGEN OXIDES ON ECOSYSTEMS
• Nitrogen is an element essential for plant growth Most soils, however, are low
in nitrogen, therefore, soil is usually the growth-limiting factor in both
agricultural and natural ecosystems Nitrogen governs, to a large extent, the
utilization of phosphorus, potassium, and other nutrients Most plants growing
in natural ecosystems are adapted to living in soils with low nitrogen levels
• Now, however, because of the nitrogen deposition that has been occurring over
many years, forest ecosystems in the temperate regions of the world that at one
tune were considered to be nitrogen limited are experiencing increased nitrogen
deposition The traditionally held view that the growth of forests is nitrogen
limited, undoubtedly, has delayed acceptance of the idea that increased nitrogen
deposition is a potential source of stress in ecosystems
• Agricultural soils are usually so heavily fertilized that any effects of atmospheric
nitrogen deposition cannot be readily observed, therefore, crop (agroecosystem)
responses are not considered in the discussion of ecosystem effects because any
addition would be considered beneficial
• The mean annual wet nitrate and ammonia deposition in the eastern United States
ranges from 2 9 kg/ha at low elevations to 7 to 22 kg/ha at high elevations
• Nitrogen, whether added to the soil from atmospheric deposition, as fertilizer, or
formed by nitrification, may (1) be taken up by plants, (2) be taken up by
microorganisms, (3) be lost by runoff, or (4) escape as a gas into the
atmosphere
• Chronic nitrogen deposition to natural ecosystems in terrestrial habitats alters the
following plant and soil processes (1) plant uptake and allocation, (2) litter
production and decomposition, (3) immobilization (includes ammonification, the
release of ammonium, and nitrification, its conversion to nitrate, during the
1-9
-------
decay of litter and soil organic matter), (4) nitrate ion leaching, and (5) trace gas
emissions (e.g , nitrous oxides) and can lead to nitrogen saturation (see
Figure 1-1)
Process altered by
nitrogen deposition
Figure 1-1. Nitrogen cycle (dotted lines indicate processes altered by chronic nitrogen
deposition).
Patterns of nitrogen uptake and carbohydrate allocation directly influence plant
growth rates Growth responses are observable over tune only if measurements
are begun at the tune deposition begins The forests in eastern North America
today are responding to continuous nitrogen depositions that began no later than
the 1930s Therefore, there are probably no pristine forests
— Nitrogen uptake influences plant photosynthetic capacity and carbohydrate
production because approximately 75% of the nitrogen accumulated in a
leaf is used during the process of photosynthesis Carbohydrate allocation
influences plant growth Plants shift allocation to the shoot or root
depending on whether the need is for greater leaf or root growth Excess
1-10
-------
soil nitrate shifts carbohydrate allocation to the shoots from the roots and
provides nitrogen in a form difficult for plants to metabolize
— The high nitrogen content of leaves associated with increased soil nitrogen
is advantageous only if light and water are not limiting because the lack of
sunlight limits photosynthetic production of the large amounts of
carbohydrates required in the metabolic conversion of nitrates
— Reduced carbohydrate allocation to the roots of plants growing in soil high
in nitrogen is associated with increased ammo acids in foliage, decreased
root biomass, and the loss of the mycorrhizal root fungi necessary for
uptake of water and minerals such as nitrogen and phosphorus Foliage of
conifers in high nitrogen deposition areas of the United States are low in
phosphorus
— The loss of mycorrhizal fungi and nutrient imbalances tend to make
conifers more susceptible to drought, other pollutants, and pathogens
Drought has been implicated as either triggering or increasing tree decline
in forests in both North America and Europe Symptoms of magnesium
deficiency and drought have been associated with increased amounts of
soil nitrate
• Litter production (leaves, twigs, flowers, fruit, bark fragments, and dead plants)
and decomposition of soil organic matter by microorganisms are the most
important sources of usable nitrogen in unfertilized ecosystems
— The nitrogen content of plant tissues influences the size of microbial
populations and influences decomposition rates in both terrestrial and
aquatic habitats Increased nitrogen in litter increases the amount of
nitrogen entering soil and, in turn, alters soil microbial decomposition
processes resulting in increased ammonium formation Increases in
ammonium levels are expected to result in increased nitrate formation (see
Figure 1-1)
— Increases in the nitrogen content of litter and in litter decomposition rates
as well as a change in nitrogen cycling have been observed in the more
highly polluted areas of the San Bernardino Mountains of Southern
California
• Nitrogen saturation results when continuous additions of nitrogen (nitrogen
loading) to the soil exceed the capacity of plants and microorganisms to utilize
nitrogen Ecosystems no longer function as a nitrogen sink Saturation implies
that some resource other than nitrogen (e g, water and phosphorus for plants,
carbon for microorganisms) is limiting foiotic function
• Nitrogen saturation in itself need not have a negative impact on ecosystem
functioning; brief periods of nitrogen saituration, a result of commercial
1-11
-------
fertilization, generally have increased short-term productivity and often produced
greater long-term growth
• Nitrate soil levels in excess of plant and microbial demand always result in
increased nitrate leaching (i e , runoff [see Figure 1-1])
— Recent measurements indicating increased nitrogen leaching from certain
high-elevation forests in the southern Appalachian Mountains suggest that
these forests have reached saturation, cumulative additions have exceeded
the capacity of these systems to accumulate nitrogen
— Nitrate leaching has the potential for mobilizing aluminum and acidifying
soils and waters Pulses of nitrate and aluminum (A13+) have been
reported from the Smoky Mountains of North Carolina
— Nitrate leaching and runoff are probably more important sources of
nitrogen in wetlands and aquatic habitats than nitrogen deposition
• Changes in the nitrogen supply can have a considerable impact on the nutrient
balance of an ecosystem The nitrogen cycle, the source of all nitrogen required
by plants and animals, is mediated in both terrestrial and aquatic habitats almost
entirely by microorganisms, making it more sensitive to alterations in available
nitrogen
— Ecosystem response to environmental perturbations is determined by the
response of its constituent organisms Response involves alteration of
functions (the movement of energy and nutrients) between (1) the
individual and its environment, (2) the population and its environment, and
(3) the biological community and its environment, the ecosystem (see
Figure 1-1)
— Intense competition among plants for light, water, nutrients, and space,
along with the recurrent natural climatic (temperatuie, wind, rain, and
fire) and biological (herbivory, disease) stresses, can alter the species
composition of communities by eliminating those individuals sensitive to
specific stresses Those organisms able to cope with the stresses survive
— Competition for nutrients exists between the various species of plants and
between plants and soil microorganisms
— An increase in the nitrogen supply in both wetlands and soil habitats alters
the competitive relationships among plant species Fast growing species
that have a high nitrogen requirement are favored
— In the Netherlands, excess nitrogen deposition has been postulated for
(1) the replacement of heathlands by grasses, (2) a shift in composition of
the herb layer in forests toward species more commonly found in nitrogen-
1-12
-------
rich areas, and (3) the decrease in the past decades of fruiting bodies of
mycorrhizal fungi (those growing on tree roots)
— Terrestrial ecosystems suggested as being at risk from deposition of
nitrogen-based compounds are (1) heathlands, (2) low meadow vegetation
types used for extensive grazing and haymaking, and (3) coniferous
forests, especially those at high .altitudes
— Ombrotrophic bogs (those dependent on wet and dry atmospheric
deposition for their nutrients) are likely to be converted to forested
wetlands because the plants normally growing there are unable to compete
under increasing nitrogen deposition Increased nitrogen deposition has
resulted in bogs in western Europe being converted to grasslands (as
indicated above in the Netherlands)
— Wetlands harbor a disproportionate (relative to habitat area) share of the
flora that are threatened by extinction Eighteen species of plants, such as
the Venus fly trap, formally listed in the Code of Federal Regulation as
endangered and an additional 284 species listed as potentially threatened
are found principally in wetlands of the conterminous United States
Several species on this list are adapted to nitrogen-poor environments and,
therefore, are poor competitors in nitrogen-rich habitats
— In the United States, bogs inhabited by sensitive plants (e g , sun dews,
pitcher plants, and the endangered Venus fly trap) are found chiefly in the
southeast.
— Under anaerobic conditions, such as those found in wetlands, increased
denitnfication of large nitrate pools could lead to increased release of
nitrous oxide to the atmosphere Nitrous oxides have been implicated as
being among the gases mvolved in global warming (see Figure 1-1)
• The effects of nitrogen oxides in aquatic ecosystems fall into three general
categones (1) acidification, both chronic and episodic, (2) eutrophication of
both fresh water and estuaries, and (3) directly toxic effects
— Episodic acidification is far more common than chronic, and is well
documented in streams and lakes in the Adirondack Mountains, for
streams in the Catskill Mountains, and in a small proportion of lakes in
Vermont and parts of Canada Episodic acidification has been associated
with (1) seasonal snow melt and (2) seasonably of snow melt plus
increasing nitrate concentrations
— Eutrophication is generally due to the limited availability of phosphorus
Toxic effects of nitrogen on biota are associated with the presence of
un-ionized ammonia (NH3) at high pH and not with nitrates
1-13
-------
Concern regarding the possible effects of nitrogen saturation has lead to attempts
to develop critical loads A critical load is defined as a quantitative estimate of
exposure to one or more pollutants below which significant harmful effects on
sensitive elements of the environment do not occur according to present
knowledge. The cumulative nitrogen deposition (critical load) required to
saturate an ecosystem is a critical unknown
1,11 EFFECTS OF NITROGEN OXIDES ON VISIBILITY
• The primary effects of NOX on visibility are twofold (1) discoloration,
producing a brownish color seen in plumes, layered hazes, and uniform hazes,
and (2) reductions in visual range (increases in light extinction), especially in
urban areas in the western United States
• Nitrogen dioxide and ammonium nitrate, which are the optically active species of
NOX, can contribute significantly to visibility impairment in the form of plumes
and hazes
• Nitrogen dioxide causes yellow-brown discoloration, especially in power plant
plumes in the western United States
• Nitrogen dioxide and nitrate aerosol are significant contributors to urban haze,
especially in urbanized California and other western areas, where their combined
share of total light extinction can be 20 to 40% Because of atmospheric
conditions in those areas, NOX are relatively small contributors to light extinction
in nonurban areas
* The effects on visibility of NO2 in plumes can be accurately predicted via
models, but the effects of aerosol particles cannot be
• The effects of NOX emission contiols on nitrate aerosol concentrations and
resulting visibility effects are nonlinear Also, the large-scale reduction of
sulfate, which competes with nitrate for available NH3, may result in increases in
nitrate aerosol concentrations
• Economic studies have not focused specifically on NOx-associated changes in
visibiity for the most part, but some studies have considered the types of
visibility effects that are associated with NOX
• The aesthetic effects of air pollution-induced changes in atmospheric visibility
have been the focus of most economic studies The few studies of effects of
visibility changes on commercial operations (such as airports) suggest a very
small economic impact from NOX
1-14
-------
Two studies of visitors to the Grand Canyon National Park show that people
would be willing to pay $2 to $3 or $3 to $6 per visitor or per party per day,
respectively, to ensure that a thin, dark plume is not visible from a popular
observation point A third study indicates a willingness to pay much more than
this to preserve and protect for others good atmospheric visibility in the Grand
Canyon, but considerable uncertainty exists in the quantitative results of the third
study
The best economic information available for visibility effects associated with
NOX is for changes in visual range in urban areas affected by uniform hazes
These values fall between $10 and $100 per year per local household for a 10%
change in visual range in major urban areas in California and throughout the
eastern United States
For layered hazes in recreational or residential settings, initial results from one
study suggest annual household values of about $30 for a noticeable improvement
in visibility conditions in the Denver area, where layered hazes are common
1.12 EFFECTS OF NITROGEN OXIDES ON MATERIALS
• Both NOX (NO, NO2) and NOy (e g , HNO3) have been shown to cause or
accelerate damage to anthropogenic materials exposed to the atmosphere
• Strong evidence exists for the deleterious effects of NOX on dyes and fabrics
The effects observed are mainly fading, discoloration, and loss of strength of
textile fibers
• Nitrogen oxides attack metals, but attack by sulfur dioxide (SO2) is more
aggressive because of surface deposition characteristics Damage to metals from
NOX can generally be discounted, except perhaps indoors, where NO2 may react
synergistically with SO2 or where NOX deposition on electronic components and
magnetic recording equipment may lead to component or system failure
• Although NOX and NOy have been reported to play a role in damage to paints
and stone, SO2 and O3 are thought to be more directly damaging than NOX and
NOy in typical polluted atmospheres
• Archival and artistic materials, such as paper and artists' paints, have been
shown to be susceptible to damage by NOX, especially NO2 These are typically
materials used indoors
• The highest NOX levels are found indoors wherever unvented combustion
systems are used (e g , gas stoves) Because the widest variety of materials are
in indoor use, the principal effects of NOX on materials may thus occur indoors,
but few data are available
1-15
-------
Economic estimates of the monetary value of NOX damage to anthropogenic
materials are few and are inadequate because of the unavailability of reliable
damage functions, because they are out of date, or both
1.13 STUDIES OF THE EFFECTS OF NITROGEN COMPOUNDS ON
ANIMALS
• Effects of NO2 observed in laboratory animals (see below) have been confirmed
in several animal species, resulting in a conclusion that these effects could occur
in humans, if the appropriate exposures were encountered In addition,
mathematical dosunetry models suggest that the greatest dose of NO2 is delivered
to the same location in both animal and human lungs (the centriacinar region)
Thus, there is general support for a qualitative extrapolation of the data
However, the data base is currently inadequate to develop a quantitative animal-
to-human extrapolation (i e , the identification of the exposures that would
actually evoke similar responses in humans) Exercise increases the total uptake
of NO2 in the respiratory tract of humans and alters the distribution of dose
Generally, increased ventilation decreases the percent uptake in the upper
respiratory tract and increases the percent uptake in the lower respiratory tract
In people who exercise during exposure, this would tend to increase their
susceptibility to effects
• Nitrogen dioxide increases susceptibility to both bacterial and viral pulmonary
infections in animals This effect is probably due primarily to effects on alveolar
macrophages and possibly to changes in the immune system, but other
mechanisms cannot be ruled out The lowest observed concentration that
increases susceptibility to bacterial lung infections after acute exposure is
2 0 ppm NO2 (a 3-h exposure study in mice) Acute (17-h) exposures to
S:2 3 ppm NO2 also decrease pulmonary bactericidal activity in mice After
long-term exposures (e g , 3 to 6 mo) to 0 5 ppm NO2, mice have decreased
resistance to bacterial lung infections Nitrogen dioxide also increases
susceptibility to viral lung infections in mice Subchronic (7-week) exposures to
concentrations as low as 0 25 ppm NO2 can alter the systemic immune system in
mice
• When the relationship of NO2 exposure concentration and duration was studied,
concentration had more influence than duration on the health outcome This
conclusion is primarily founded on extensive investigations of lung antibacterial
defenses of mice, which also indicate that the exposure pattern (e g , baseline
level with daily peaks of NO2 or exposure 24 h/day versus 6 to 7 h/day) has an
impact on the study results Another report suggests that lung structural changes
are more dependent on exposure concentration than duration
• Nitrogen dioxide exposure causes lung structural alterations in several animal
species (e.g., mice, rats, monkeys). Acute exposure effects are of little interest
1-16
-------
because of high concentrations required to cause changes Prolonged exposures
produce changes in the cells lining the region of the lung where the conducting
airways and gas exchange area meet (i e , the centriacinar region) In addition,
the tissue in this region (i e , alveolar iriterstitium) becomes thicker Urban
patterns of NO2 (e g , 0 5-ppm baseline with brief peaks to 1 5 ppm for 6 weeks
or exposures to 0 5 ppm NO2 for 4 to 6 mo) cause these effects in rats
Several animal studies clearly demonstrate that chronic exposure to NO2 can
cause emphysema of the type seen in human lungs However, to date, only very
high, environmentally unrealistic concentrations of NO2 (^50 ppm) have been
reported to cause such effects
The effects of exposure to mixtures of NO2 and other pollutants are dependent
on the exposure regimen, species, and end point measured Most mixture
research involves NO2 and O3 and shows that additivity and synergism can
occur A similar conclusion can be drawn from the more limited research with
NO2 and sulfunc acid Findings of either additivity or synergism are of concern
because of the ubiquitous co-occurring nature of NO2 and O3 However, precise
extrapolation of these findings to ambient scenarios is confounded because the
mixtures used in laboratories (by concentration, concentration ratio, and pattern)
do not mimic the ambient air
1.14 EPIDEMIOLOGY STUDIES OF NITROGEN DIOXIDE
• Results from several of a number of indoor air epidemiology studies suggest that
increased respiratory symptoms in 5- to 12-year-old children are associated with
estimated exposure to NO2 The associations reported in the majority of the
studies did not reach statistical significance at p < 0 05 The consistency of
these studies was examined and the evidence synthesized in a combined
quantitative analysis (meta-analysis) of the subject studies Subject to
assumptions made for the combined analysis, the main conclusion from that
analysis was that an increased risk of about 20% for respiratory symptoms and
disease corresponded to each increase of 0 015 ppm (28.3 /xg/m3) in estimated
2-week average NO2 exposure, where mean weekly concentrations in bedrooms
in studies reporting NO2 levels were predominately between 0 008 and
0 065 ppm NO2 The measured NO2 studies gave a higher estimated odds ratio
than the surrogate estimates, which is consistent with a measurement error effect
The effect of having adjusted for covanates such as socioeconomic status,
smoking, and gender was that those stu
-------
rather, estimating actual exposure requires knowledge of both pollutant levels
and related human activity patterns The effects studied may be related to peak
exposures, average exposures, or a combination of the two To the extent that
health effects depend on peak exposures rather than average exposures, the
exposure estimates used in the subject studies and meta-analyses introduce
exposure measurement error These studies cannot distinguish between the
relative contributions of peak and average exposures and their relationship with
the observed health effects Additionally, a by-product of NO2, HONO, may be
a factor in observed health effects However, only very limited health and
aerometnc data are available that examine such possibilities Also, although the
level of similarity and common elements between the health outcome measures in
the NO2 studies provide some confidence in their use in the quantitative analysis,
the symptoms and illnesses combined are to some extent different and could
indeed reflect different underlying processes Thus, caution is necessary in
interpreting the meta-analysis results
• Although there is evidence that suggests that increased estimated NO2 exposure
is associated with increased respiratory symptoms in children aged 5 to 12 years,
the exposure estimated may be inadequate to determine a quantitative relationship
between estimates of exposure and symptoms The studies with measured NO2
exposure did so only for periods of 1 to 2 weeks and reported the values as
averages. None of the studies attempted to relate the effects seen to the pattern
of exposure, such as short-term peaks Furthermore, the extrapolation to
possible patterns of ambient exposure is difficult
• In individual indoor studies of infants 2 years of age and younger, no consistent
relationship was found between estimates of NO2 exposure and the prevalence of
respiratory symptoms and disease Based on a meta-analyses of these indoor
infant studies, subject to the assumptions made for the meta-analysis, the
combined odds ratio for the increase in respiratory disease per increase of
0.015 ppm NO2 was 1 09 with a 95% confidence interval of 0 95 to 1 26, where
mean weekly concentrations in bedrooms were predominately between 0 005 and
0.050 ppm NO2 in studies reporting levels Thus, although the overall combined
estimate is positive, it clearly contains the no-effect value of 1 0, (i e , is not
statistically significant), and so we cannot conclude that the evidence suggests an
effect in infants comparable to that seen in older children
• Other epidemiology studies have attempted to relate some measure of indoor
and/or outdoor NO2 exposure to long-term changes in pulmonary function
These changes were marginally significant Most studies did not find any
effects, which is consistent with controlled human exposure study data (see
Chapter 15) However, there is msufficient epidemiological evidence to make
any conclusion about the long- or short-term effects of NO2 on pulmonary
function
1-18
-------
1.15 CONTROLLED HUMAN EXPOSURE STUDIES OF OXIDES OF
NITROGEN
• Nitrogen dioxide causes decrements in lung function, particularly increased
airway resistance in healthy subjects at 2-h concentrations exceeding 2 0 ppm
• Nitrogen dioxide exposure results in increased airway responsiveness in healthy,
nonsmoking subjects exposed to concentrations exceeding 1 0 ppm for exposure
durations of 1 h or longer
• Nitrogen dioxide exposure at levels above 1 5 ppm may alter numbers and types
of inflammatory cells in the distal airways or alveoli, but these responses depend
upon exposure concentration, duration, and frequency Nitrogen dioxide may
alter function of cells within the lung and production of mediators that may be
important in lung host defenses
• Nitrogen dioxide exposure of asthmatics causes, in some subjects, increased
airway responsiveness to a variety of provocative mediators, including
cholinergic and histaminergic chemicals, SO2, and cold air However, the
presence of these responses appears to be influenced by the exposure protocol,
particularly whether or not the exposure includes exercise
• Modest decrements in spirometnc measures of lung function (3 to 8%) may
occur in some asthmatics and patients with chronic obstructive pulmonary disease
(COPD) from brief exposure to concentrations of NO2 greater than 2 ppm and
may also be observed with longer exposures to lower concentrations
• Nitric acid levels in the range of 100 to 200 ppb may cause some pulmonary
function responses in adolescent asthmatics, but not in healthy adults Other
commonly occurring NOX species do not appear to cause any pulmonary function
responses at concentrations expected in the ambient environment, even at higher
levels than in worst-case scenarios However, not all nitrogen oxides acid
species have been studied sufficiently
• No association between lung function responses and respiratory symptom
responses were observed Furthermore, there is little evidence of a
concentration-response relationship for changes in lung function, airway
responsiveness, or symptoms at the NO2 levels that are reviewed here
1-19
-------
1.16 NITROGEN DIOXIDE HEALTH EFFECTS:
CONCENTRATION-RESPONSE RELATIONSHIPS AND
SUBPOPULATIONS POTENTIALLY AT RISK
1.16.1 Concentration-Response Relationships
• Table 1-1 summarizes key health effects observed in controlled human exposure
(clinical) studies with NO2 exposure durations of 0 5 to 3 h The physiological
end point that, to date, appears to be the most sensitive indicator of response is a
change in airway responsiveness to bronchoconstnctors in asthmatics This
increase in airway responsiveness has been observed in some, but not all studies,
and only at relatively low NO2 concentrations within the range 0 2 to 0 3 ppm
TABLE 1-1. KEY HUMAN HEALTH EFFECTS OF EXPOSUBE TO
NITROGEN DIOXIDE—CLINICAL STUDIES81
NO2 (ppm) (Exposure Duration) Observed Effects
0 2-0 3 (0 5-2 0 h) Trend toward increased airway responsiveness to challenges in asthmatics
However, no significant effects observed by same or other investigators at
NO2 levels up to 4 ppm Small (4-6%) decreases in FEV1 or FVC in
adult or adolescent asthmatics, in response to NO2 alone
0 3 (3.75 h) Small decreases (5-9%) in FVC and FEVj in COPD patients with mild
exercise No effects seen by other investigators for COPD patients at
0 5-2 0 ppm NO2
1 5-2 0 (2-3 h) Increased airway responsiveness to bronchoconstnctors in healthy adults
However, effects not detected by other investigators at 2-4 ppm
^2.00 (1-3 h) Lung function changes (e g , increased airway lesistance) in healthy
subjects Effects not found by others at 2-4 ppm
aNO2 = Nitrogen dioxide
FEVj = Forced expiratory volume in 1 s
FVC = Forced vital capacity
COPD = Chronic obstructive pulmonary disease
Additionally, small decreases in forced expiratory volume mis (FEVX) or
forced vital capacity (FVC) in adult or adolescent asthmatics have been observed
in response to the same levels of NO2. However, NO2 concentration-response
relationships are not evident for either airway responsiveness or pulmonary
function changes A second category of sensitive subjects is patients with
COPD. Although small decreases have been observed in FVC and FEVX in
1-20
-------
COPD patients exposed to 0 3 ppm in one study, no effects were seen in other
studies at higher exposure levels At higher exposure levels (more than
1 5 ppm), NO2 exposure results in increased airway responsiveness and
increased airway resistance in healthy adults However, some researchers have
not observed any NO2-induced changes in airway resistance at NO2 levels
between 2 and 4 ppm
• The collective, combined evidence from epidemiology studies examining
relationships between estimates of exposure to NO2 and lower respiratory
symptoms and disease in children aged 5 to 12 years (as evaluated by an EPA
meta-analysis yielding quantitative estimates of effects) tends to demonstrate that
mcreased risk for respiratory illness among children is associated with exposure
to NO2, as summarized in Table 1-2 In individual indoor studies of infants
2 years of age and younger, no consistent relationship was found between
estimates of NO2 exposure and the prevalence of respiratory symptoms and
disease Based on a meta-analyses of these infant studies, the combined odds
ratio for the increase in respiratory disease per increase of 0 015 ppm NO2 was
1 09 with a 95 % confidence interval of 0 95 to 1 26 Thus, although the overall
combined estimate is positive, it clearly contains the no-effect value of 1 0, (i e ,
is not statistically significant), and so we cannot conclude that the evidence
suggests an effect in infants comparable to that seen in older children (see
Table 1-2) Higher levels (>0 3 ppm during a shift at work) in an occupational
_ setting were related to an elevated prevalence of acute respiratory symptoms in
adults Also, episodic exposures occurring over a period of 1 h or longer at
levels possibly as high as 1 5 ppm or higher have resulted in the occurrence of
acute respiratory symptoms Lastly, exceptionally high acute occupational
exposures of 25 to 100 ppm NO2 result in bronchial pneumonia, bronchitis, or
bronchiohtis, and very extreme occupational NO2 exposures (> 200 ppm) have
been associated with effects that range from hypoxemia and transient obstruction
of the airways to death
• Numerous concentration-response studies have been conducted with animals
using a wide range of exposure durations and end points, all of which influence
the outcome The major classes of effects observed at concentrations less than
1 0 ppm include decrements in host defenses, alterations in lung metabolism
(e g , mcreased lipid peroxidation and antioxidant metabolism), epithelial
remodeling of the lower respiratory tract, thickening of the centnacmar
interstitium, and a variety of extrapuhnonary changes Such findings can be
qualitatively extrapolated to humans, but major uncertainties in respiratory tract
dosimetry and species sensitivity currently preclude a quantitative extrapolation
Substantially higher NO2 concentrations (>12 ppm) have caused emphysema as
defined by National Institutes of Health cntena
1-21
-------
TABLE 1-2. KEY HUMAN HEALTH EFFECTS OF EXPOSURE TO
NITROGEN DIOXIDE—EPJDEMIOLOGICAL STUDIES'1
NO2 (pPm) (Exposure Duration)
Observed Effects
0 015-ppm increase, where mean weekly
concentrations in bedrooms in studies reporting
levels were predominately between 0 008 and
0.065 ppm NO2 (in 1- and 2-week integrated
average NO2 concentration estimating an
unspecified long-term average)
0.015-ppm increase in annual average of 2-week
NO2 levels, where mean weekly concentrations in
bedrooms were predominately between 0 005 and
0 050 ppm NO2
A meta-analysis shows an increased risk of lower
respiratory symptoms/disease in children 5 to 12 years
old associated with exposure estimates of NO2 levels
The 95% confidence interval of the odds ratio was 1 1 to
1 3 (see Chapter 14) Predominant source of exposure
contrast is homes with gas stoves vs homes with electric
stoves
In individual indoor studies of infants 2 years of age and
younger, no consistent relationship was found between
estimates of NO2 exposure and the prevalence of
respiratory symptoms and disease Based on a meta-
analyses of these infant studies, the combined odds ratio
for the increase in respiratory disease per increase of
0 015 ppm NO2 was 1 09 with a 95% confidence interval
of 0 95 to 1 26 Thus, although the overall combined
estimate is positive, it clearly contains the no-effect value
of 1 0, (i e , is not statistically significant), and so we
cannot conclude that the evidence suggests an effect in
infants comparable to that seen m older children (see
Chapter 14)
>0.3 ppm (average exposure during work shift) Elevated prevalence of acute respiiatory symptoms
Episodic exposure during hockey game to NO2 Occurrence of acute respiratory symptoms (cough, chest
levels of 1 5 ppm or higher pain, dyspnea)
25 to 100 ppm (episodic occupational exposure) Bronchial pneumonia, bronchitis, and bronchiohtis
induced by exceptionally high NO2 exposure
>200 ppm (extreme episodic exposures)
Extreme exposure health outcomes range from
hypoxemia/transient airway obstruction to death
8NO2 = Nitrogen dioxide
Results from infectivity studies examining concentration (C) >< duration (T, tune)
of exposure and pattern of exposure indicate that concentration exerted more
influence than time of exposure in increasing susceptibility to respiratory
bacterial infection in mice Furthermore, the exact pattern of exposure played a
major role m experimental outcomes Even so, duration is still important
For example, as exposures proceed from weeks to months at a given
concentration, structural changes in the lung become more severe Also, at
longer exposure durations, lower NO2 concentrations cause effects Due to the
1-22
-------
large number of animal toxicological studies and the variety of exposure regimes,
it is not possible to succinctly display the full range of concentration responses
Therefore Table 1-3 lists a few key studies showing the lowest concentrations
that caused several types of effects
1.16.2 Subpopulations Potentially at Risk
• Certain groups within the population may be more susceptible to the effects of
NO2 exposure, including persons with preexisting respiratory disease, children,
and the elderly The reasons for paying special attention to these groups is that
(1) they may be affected by lower levels of NO2 than other subpopulations or
(2) the impact of an effect of given magnitude may be greater Some causes of
heightened susceptibility are better understood than others Subpopulations that
already have reduced ventilatory reserves (e g , the elderly and persons with
asthma, emphysema, and chronic bronchitis) will be more impacted than other
groups by decrements in pulmonary function For example, a healthy young
person may not even notice a small percentage change in pulmonary function,
but a person whose activities are already limited by reduced lung function may
not have the reserve to compensate for the same percentage change
• Approximately 10 million persons in the United States have asthma In the
general population, asthma prevalence rates increased by 29% from 1980 to
1987 For those under 20 years old, asthma rates increased from approximately
35 to 50 per 1,000 persons, a 45% increase The airways of asthmatics may be
hyperresponsive to a variety of inhaled materials, including pollens, cold-dry air,
allergens, and air pollutants Asthmatics have the potential to be among the most
susceptible members of the population with regard to respiratory responses to
NO2 On the average, asthmatics are much more sensitive to inhaled
bronchoconstrictors such as histaniine, methacholine, or carbachol The potential
addition of an NO2-induced increase in anyway response to the already heightened
responsiveness to other substances raises the possibility of exacerbation of this
pulmonary disease by NO2
• Other potentially susceptible groups include patients with COPD, such as
emphysema and chronic bronchitis Some of these patients have airway
hyperresponsiveness to physical and chemical stimuli A major concern with
COPD patients is the absence of an adequate ventilatory reserve, a susceptibility
factor described above In addition, the poor distribution of respiratory tract
ventilation in COPD may lead to a greatei delivery of NO2 to the segment of the
lung that is well ventilated, thus resulting in a greater regional tissue dose Also,
NO2 exposure may alter already impaired defense mechanisms, making this
population potentially susceptible to respiratory infection It is estimated that
14 million persons in the United States (« 6%) suffer from COPD
1-23
-------
TABLE 1-3. KEY ANIMAL TOXICOLOGICAL EFFECTS OF EXPOSURE TO
NITROGEN DIOXIDE
Nitrogen Dioxide (ppm)
(Exposure Duration) Species
Observed Effects
0 04 ppm (continuous, 9 mo) Rat
0.2 ppm (continuous base for 1 year) Mouse
plus 0 8 ppm (1-h peak, 2x/day,
5 days/week)
0 25 ppm Mouse
(7 h/day, 5 days/week, 7 weeks)
0 3 ppm (2 h/day, 2 days) Rabbit
0.4 ppm (continuous, 4 weeks) Mouse
0 4 ppm (continuous, 9 mo) Rat
0.4 ppm (continuous, up to 27 mo) Rat
0 5 ppm (continuous) 3 mo Mouse
0.5-28 ppm (6 mm to 1 year) Mouse
0 5 ppm (continuous base, 6 weeks) Rat
plus 1 5 ppm (1-h peak, 2X/day,
5 days/week)
Increased hpid peroxidation (ethane in exhaled
breath)
Increased susceptibility to respiratory infection
and decreased vital capacity and respiratory
system compliance, compared to control or
baseline only
Systemic effect on cell-mediated immunity
Decreased phagocytosis of alveolar macrophages
Decreased systemic humoral immunity
Increased antioxidants and antioxidant
metabolism
Slight increase in thickness of air blood barrier
at 18 mo, becoming significant by 27 mo, also
alterations in bronchiolar and alveolar epithelium
by 27 mo
Increased susceptibility to respiratory infection
Linear increase in susceptibility to respiratory
infection with time, increased slope of curve
with increased concentration, concentration more
important than time
Alterations in Type 2 cells and increased
interstitial matrix of proximal alveolar region, no
changes in terminal bronchiolar region of adults
Based on epidemiology studies, children aged 5 to 12 years constitute a
subpopulation potentially susceptible to an increase in respiratory morbidity
associated with NO2 exposure (Chapter 14) In the United States, approximately
18 million children are in the age group 5 to 9 years, whereas around 17 million
children are in the age group 10 to 14 years However, the fractions of the
numbers of potentially at risk children in various age groups that are actually
exposed to NO2 concentrations/patterns sufficient to induce respiratory morbidity
have not been determined
Another potential susceptible subpopulation group is immunocompromised
individuals, who would have an increased susceptibility for infectious pulmonary
disease as well as other health effects Such people would hypothetically be
more susceptible to agents, such as NO2, that further compromise host defenses
Immunocompromised groups could include those people with abnormalities in
polymorphonuclear leukocyte number or function and those with humoral and/or
cell-mediated immunity dysfunctions This would include people with reduced
1-24
-------
immune function related to kidney transplants, acquired immune deficiency
syndrome (AIDS), and chemotherapy Although these immunocompromised
groups represent potentially at-risk susceptible populations for NO2 effects, no
human research has examined NO2 exposure in these groups Thus, there only
now exists a hypothesized association with increased susceptibility to NO2
Although it is clear that NO2 can affect alveolar macrophages, humoral
immunity, and cell-mediated immunity in otherwise normal animals
(Chapter 13), the ammal-to-human extrapolation cannot yet be made
quantitatively Nevertheless, it may be pradent to consider including such
reduced immune function groups as suscepf ible subpopulations at potentially
increased risk for NO2-induced health effects
1-25
-------
-------
2. INTRODUCTION
The purpose of this document is to present air quality criteria for oxides of nitrogen
(NOX) in accordance with Sections 108 and 109 of the Clean Air Act (CAA) Section 108
(U S Code, 1991) directs the Administrator of the U S Environmental Protection Agency
(EPA) to list pollutants that may reasonably be anticipated to endanger public health and
welfare, and to issue air quality criteria for them These air,quality criteria are to reflect the
latest scientific information useful in indicating the kind and extent of all identifiable effects
on public health and welfare that may be expected from the presence of the pollutant in the
ambient air
Section 109(a,b) (U S Code, 1991) directs the EPA Administrator to propose and
promulgate "primary" and "secondary" National Ambient Air Quality Standards (NAAQS)
for pollutants identified under Section 108 Section 109(b)(l) defines a primary standard as a
level of air quality, the attainment and maintenance of which in the judgment of the
Administrator, based on the criteria and allowing for an adequate margin of safety, is
requisite to protect the public health Section 109(d) of the Act (U S Code, 1991) requires
periodic review and, if appropriate, revision of existing criteria and standards In addition,
Section 109(c) specifically requires the Administrator to promulgate a primary standard for
nitrogen dioxide (NO2) with an averaging tune of not more than 3 h, unless no significant
evidence is found that such a short-term standard is required to protect health Under
Section 109(b) of the CAA, the Administrator must consider available information to set
secondary NAAQS that are based on the criteria and are requisite to protect the public
welfare from any known or anticipated adverse effects associated with the presence of such
pollutants The welfare effects included in the criteria are effects on vegetation, crops, soils,
water, animals, manufactured materials, weather, visibility, and climate, as well as damage
to and deterioration of property, hazards to transportation, and effects on economic values,
personal comfort, and well-being
A variety of NOX compounds and their transformation products occur naturally in the
environment and also result from human activities In addition to NO2, mtnc oxide (NO),
nitrous oxide, gaseous nitrous acid, gaseous nitric acid, and both nitrite and nitrate particles
2-1
-------
have all been found in the ambient air The formation of nitrosamines in the atmosphere by
reaction of NOX with amines has also been suggested, but not yet convincingly demonstrated
Available scientific research on the potential health and welfare effects of NOX compounds
provides the strongest evidence linking specific adverse effects to near-ambient concentrations
of NO2. Therefore, EPA has focused its criteria reviews primarily on health and welfare
effects reported to be associated with exposure to NO2 Nitrogen dioxide is an air pollutant
generated mainly by the photochemically initiated oxidation of NO, which is emitted from a
variety of mobile and stationary sources At elevated concentrations, NO2 can adversely
affect human health, vegetation, materials, and visibility Nitrogen oxide compounds can
also contribute to increased rates of acidic deposition and high ozone concentrations
2.1 REGULATORY AND SCIENTIFIC BACKGROUND
On April 30, 1971, EPA first promulgated the NAAQS for NO2 under Section 109 of
the CAA (Federal Register, 1971) Identical primary and secondary standards for NO2 were
o
set at 0.053 ppm (100 /*g/m ), averaged over 1 year The scientific bases for these standards
are contained in the original criteria document, Air Quality Criteria for Nitrogen Oxides
(U.S. Environmental Protection Agency, 1971) The primary standard set in 1971 was based
mainly on community epidemiology studies (Shy et al, 1970a,b, Pearlman et al, 1971)
conducted in Chattanooga, TN, which reported respiratory effects in children exposed to
low-level NO2 concentrations over a long-term period
In response to the August 1977 Clean Air Act Amendments, EPA developed the
document Health Effects of Short-Term Exposures to Nitrogen Dioxide Air Quality Criteria
(U.S. Environmental Protection Agency, 1978b) to serve as the basis for evaluating the need
to promulgate a NAAQS for short-term concentrations of NO2 On December 12, 1978
(Federal Register, 1978), EPA announced that this document on short-term effects of NO2
would be incorporated into the full revision of the 1971 Air Quality Criteria for Oxides of
Nitrogen, which EPA was then in the process of reviewing and updating in accordance with
Section 109(d)(l) of the CAA as amended This process resulted in the production of the
revised 1982 document, Air Quality Criteria for Oxides of Nitrogen (U S Environmental
Protection Agency, 1982a)
2-2
-------
In accordance with Sections 108 and 109 of the CAA, in 1985 EPA completed the
review of criteria upon which the existing primary arid secondary NO2 NAAQS were based
Reevaluation of the Chattanooga studies in view of later information (especially regarding the
accuracy of the air quality monitoring method for NO2 used in the studies) indicated that
these studies provide only limited qualitative evidence for an association between health
effects and ambient exposures to NO2 In reviewing the scientific basis for an annual
standard, EPA found that evidence showing the most serious health effects associated with
chronic NO2 exposures (e g , emphysematous-like alterations in the lung and increased
susceptibility to infection) came from animal studies conducted at concentrations well above
those permitted in the ambient air by the current annual standard Major factors impeding
use of these studies for standard-setting purposes included limitations of methods for
quantitatively extrapolating exposure-response relationships from these animal studies directly
to humans However, the seriousness of these effects, coupled with biological similarities
between humans and test animals, suggested that there was some risk to human health from
long-term exposure to elevated NO2 levels Other evidence from community epidemiology
and gas-stove epidemiology studies provided some qualitative support for concluding that
there is a relationship between adverse health effects and repeated, acute exposures to
elevated (>0 2 to 1 0 ppm) NO2 concentrations or longer term human exposure to near-
ambient levels of NO2 However, concern at that tune for limitations associated with these
studies (e g , unreliable or insufficient monitoring data and inadequate treatment of potential
confounding factors such as humidity and pollutants other than NO2) then precluded
derivation of quantitative exposure-response relationships
Although it is not possible to quantify the margin of safety provided by the existing
annual standard, two observations are relevant (1) a 0 053-ppm standard was consistent
with the Clean Air Science Advisory Committee's (CASAC's) recommendation (Friedlander,
1982) to set the annual standard at the lower end of the range (0 05 to 0 08 ppm) cited in the
Office of Air Quality Planning and Standards (OAQPS) Staff Paper (U S Environmental
Protection Agency, 1982b) to ensure an adequate margin of safety against long-term effects
and to provide some measure of protection against possible short-term health effects, and
(2) a 0 053-ppm standard would keep annual NO2 concentrations considerably below the
long-term exposure levels for which serious chronic eifects have been observed in animals
2-3
-------
Maintaining the current annual primary standard thus represented, at that time, a prudent
public health policy choice aimed at preventing any increased chronic health risk in large,
populated U S. urban areas that attain the annual standard On July 19, 1985, EPA
announced the final decision to retain the existing annual primary and secondary standards
The decision on the need, if any, for a separate short-term standard (less than 3 h) was
deferred, pending results from additional research focused on reducing uncertainties
associated with evaluating short-term health effects of NO2
2.2 CRITICAL ISSUES
Based on the available scientific evidence, several critical data bases and associated
issues are addressed in this document Some of the key issues addressed are as follows
1. An evaluation of the clinical studies data examining short-term exposure
(1 to 3 h) to NO2 in the range of 0 2 to 0 5 ppm and increased bronchial
reactivity in asthmatics is presented
2 The strength and consistency of the epidemiologic data base that relates
NO2 estimates of exposure and an increased rate and/or seventy of
respiratory symptoms and disease are reviewed Further, a meta-analysis
of selected studies is provided that evaluates the combined results of the
studies and examines their strength as a whole
3. Controlled human exposure (clinical) and animal lexicological studies that
examine the effects of NO2 on aspects of the respiratory host defense
system related to respiratory infection are assessed Also, support from the
animal toxicological data for the biologically plausible hypothesis that
relates respiratory symptoms and morbidity to NO2 exposure in
epidemiologic studies as being due to respiratory infection is discussed
4. A discussion reviewing the emphysematous potential in humans from
exposure to long-term chronic NO2 exposure is presented
5 Subpopulation groups that may be more susceptible to effects from ambient
NO2 exposure and are potentially at heightened risk are discussed
6. The ecological effects of critical nitrogen loading are also assessed
Concentrations of NO and NO2 in the atmosphere do not often reach
phytotoxic levels. Ecosystem exposure to nitrogen compounds, therefore,
is mainly through the soil Crops are excluded because they are usually
2-4
-------
heavily fertilized Nitrogen in forest soils, however, is usually growth
limiting Concern links nitrogen deposition with possible ecological
impacts arising from nitrogen saturation Saturation results when
continuous additions to soil background nitrogen exceeds the capacity of
plants and microorganisms in an ecosystem to utilize and retain nitrogen
Increases in soil nitrogen can play a selective role When nitrogen
becomes readily available, species or varieties adapted to living in soil with
low levels of nitrogen will be replaced by vegetation capable of utilizing
the increased supply The concept of critical load, "a quantitative estimate
of one or more pollutants below which significant harmful effects on
sensitive elements of the environment do not occur according to present
knowledge," was developed in Europe This concept, however, has not
been widely accepted in the United States because not enough is known
concerning the functioning of ecosystems to be able to set critical loads in
a completely objective fashion
2.3 ORGANIZATION OF THE DOCUMENT
The present document consists of 16 chapters The Executive Summary for the entire
document is contained in Chapter 1, followed by this general introduction in Chapter 2
Chapters 3 through 8 provide background information on physical and chemical properties of
NO2 and related compounds, sources and emissions, atmospheric transport, transformation,
and fate of NO2, methods for the collection and measurement of NO2, and ambient air
concentrations and factors affecting exposure of the general population Chapter 9 evaluates
NO2 effects on crops and natural vegetation, and Chapter 10 discusses effects on terrestrial
and aquatic ecosystems Chapter 11 describes effects on visibility, and Chapter 12 describes
damage to materials attributable to NO2 Chapters 13 through 16 evaluate information
concerning the health effects of NO2 More specifically, Chapter 13 discusses respiratory
tract deposition of NO2 and information derived from experimental toxicological studies of
animals Chapter 14 discusses epidemiological studies, and Chapter 15 discusses clinical
studies Chapter 16 integrates information on critical health issues derived from studies
reviewed in the prior three chapters (Chapters 13, 14, and 15)
Neither control techniques nor control strategies for the abatement of NOX are discussed
in this document, although some of the topics included are relevant to abatement strategies
Technologies for controlling NOX emissions and emissions of volatile organic compounds are
discussed in documents issued by OAQPS (e g , U S Environmental Protection Agency,
2-5
-------
1978b, 1983). Likewise, issues germane to the scientific basis for control strategies, but not
pertinent to the development of criteria, are addressed in numerous documents issued by
OAQPS.
In addition, certain issues of direct relevance to standard setting are not explicitly
addressed in this document, but are instead analyzed in documentation prepared by OAQPS
as part of its regulatory analyses Such analyses include (1) discussion of what constitutes an
"adverse effect" and delineation of particular adverse effects that the primary and secondary
NAAQS are intended to protect against, (2) exposure analyses and assessment of consequent
risk, and (3) discussion of factors to be considered in determining an adequate margin of
safety. Key points and conclusions from such analyses are summarized in a Staff Paper
prepared by OAQPS and reviewed by CASAC. Although scientific data contribute
significantly to decisions regarding the above issues, their resolution cannot be achieved
solely on the basis of experimentally acquired information Final decisions on items (1) and
(3) are made by the Administrator, as mandated by the CAA
A fourth issue directly pertinent to standard setting is identification of populations at
risk, which is basically a selection by EPA of the subpopulation(s) to be protected by the
promulgation of a given standard This issue is addressed only partially in this document
For example, information is presented on factors, such as preexisting disease, that may
biologically predispose individuals and subpopulations to adverse effects from exposures to
NOX. The identification of a population at risk, however, requires information above and
beyond data on biological predisposition, such as information on levels of exposure, activity
patterns, and personal habits Such information is included in the Staff Paper developed by
OAQPS.
The present document includes review and critical evaluation of relevant literature on
NOX through 1993. The material selected for review and comment in the text generally
comes from the more recent literature published since 1982, with emphasis on studies
conducted at or near NOX pollutant concentrations found in ambient air Oldei literature
cited in the previous criteria document for NOC (U S Environmental Protection Agency,
1982a) is generally not discussed. However, as appropriate, some limited discussion is
included of older studies judged to be significant because of their potential usefulness in
deriving a NAAQS An attempt has been made to discuss key literature in the text and
2-6
-------
present it in tables as well Reports of lesser importance for the purposes of this document
are typically only summarized in tables
Generally, only published material that has undergone scientific peer review is
included In the interest of admitting new and important information, however, some
material not yet published in the open literature but meeting other standards of scientific
reporting may be included Emphasis has been placed on studies in which NO2 exposure
concentrations were ^5 ppm On this basis, studies in which the lowest concentration
employed exceeded this level have been included if they contain unique data, such as
documentation of a previously unreported effect or of mechanisms of effects, or if they were
multiple-concentration studies designed to provide information on concentration-response
relationships In the areas of emphysema, mutagenesis, teratogenesis, and reproductive
effects, results of studies conducted at higher levels have been included because of the
potential importance of these effects to public health
In reviewing and summarizing the literature, an attempt is made to present alternative
points of view where scientific controversy exists As warranted, considerations bearing on
the quality of studies are noted
The general policy of EPA is to express concentrations of air pollutants in metric units
(e g , in micrograms per cubic meter [jug/m ]), as well as in the more widely used units,
such as parts per million (ppm) or parts per billion (ppb), which are neither metric nor
English units That policy has been followed in those chapters in which most of the data
have been obtained from laboratory studies conducted at room temperature (e g , Chapters 13
and 16) Data are presented to a large extent in the units reported by the original
researchers Some data, however, are presented both in micrograms per cubic meter and in
parts per million to facilitate comparison of data (NO2 conversions ppm x 1,882 = /*g/m3
[e g , 0 053 ppm X 1,882 = 100 j«g/m3] and /ig/m3 X 0 00053 = ppm [e g , 100 /*g/m3 x
0 00053 = 0 053 ppm]) Data reported in parts per million for studies conducted outdoors,
such as field and open-top chamber vegetation studies, ambient air monitoring, and research
on atmospheric chemistry, have not been converted In Ihese cases, conversion of reported
parts per million and parts per billion units is questionable because it assumes standard or
uniform temperatures and pressures Deposition studies are reported in micrograms per
hectare (/ig/ha)
2-7
-------
REFERENCES
Federal Register (1971) National primary and secondary ambient air quality standards F R (April 30)
36 8186-8201
Federal Register. (1978) Air quality criteria document for oxides of nitrogen F R (December 12)
43 58117-58118
Friedlander, S K (1982) CASAC review and closure of the OAQPS staff paper for nitrogen oxides
[memorandum to Anne M Gorsuch] July 6
Pearlman, M E ; Finklea, J F , Creason, J P , Shy, C M , Young, M M , Horton, R J M (1971) Nitrogen
dioxide and lower respiratory illness Pediatrics 47 391-398
Shy, C M , Creason, J P , Pearlman, M E , McClain, K E , Benson, F B , Young, M M (1970a) The
Chattanooga school children study effects of community exposure to nitrogen dioxide 1 Methods,
description of pollutant exposure, and results of ventilatory function testing J Air Pollut Control Assoc
20 539-545
Shy, C M , Creason, J P , Pearlman, M E , McClain, K E , Benson, F B , Young, M M (1970b) The
Chattanooga school children study effects of community exposure to nitrogen dioxide n Incidence of
acute respiratory illness J Air Pollut Control Assoc 20 582-588
U.S. Code. (1991) Clean Air Act, §108, air quality criteria and control techniques, §109, national ambient air
quality standards U S C 42 §§7408-7409
U.S. Environmental Protection Agency (1971) Air quality criteria for nitrogen oxides Washington, DC U S
Environmental Protection Agency, Air Pollution Control Office, EPA report no AP-84 Available from
NTIS, Springfield, VA, PB-197333/BE
U.S Environmental Protection Agency (1978) Health effects of short-term exposures to nitrogen dioxide (air
quality criteria) Research Triangle Park, NC Health Effects Research Laboratory, EPA report
no EPA-600/8-78-009
U.S Environmental Protection Agency (1982a) Air quality criteria for oxides of nitrogen Research Triangle
Park, NC. Office of Health and Environmental Assessment, Environmental Criteria and Assessment
Office, EPA report no EPA-600/8-82-026 Available from NTIS, Springfield, VA, PB83-131011
U.S. Environmental Protection Agency (1982b) Review of the national ambient air quality standards for nitrogen
oxides' assessment of scientific and technical information, OAQPS staff paper Research Triangle Park,
NC. Office of Air Quality Planning and Standards, EPA report no EPA-450/5-82-002 Available from
NTIS, Springfield, VA, PB83-132829
2-8
-------
3. GENERAL CHEMICAL AND PHYSICAL
PROPERTIES OF OXIDES OF NITROGEN AND
OXIDES OF NITROGEN-DERIVED POLLUTANTS
3.1 INTRODUCTION AND OVERVIEW
In this chapter, some general chemical and physical properties of oxides of nitrogen
(NOX) and NOx-denved pollutants are discussed as an introduction to complex chemical and
physical interactions that may occur in the atmosphere and other media The information
presented mainly summarizes the most salient points drawn from the predecessor NOX
criteria document (U S Environmental Protection Agency, 1982), with appropriate updating
on certain important aspects This overview is further augmented throughout the present
document as related topics are discussed in depth elsewhere
In this document, NOX is the sum of nitrogen dioxide (NO2) and nitric oxide (NO), and
NOy refers to the sum of NOX and other oxidized nitrogen compounds, except nitrous oxide
(N2O) These include nitric acid (HNO3), nitrogen trioxide (NO3), dimtrogen tnoxide
(N2O3), dimtrogen tetroxide (N2O4), and dimtrogen pentoxide (N2O5) Peroxyacetylnitrate
(PAN), although not discussed at great length m this document, is normally included with the
NOy group of compounds
There are seven oxides of nitrogen that may be present m the ambient air NO, NO2,
N2O, NO3, N2O3, N2O4, and N2O5 Of these, NO and NO2 are generally present m highest
concentrations in the lower troposphere (Chapter 7) Their interconverfabihty in
photochemical smog reactions (Chapter 5) has frequently resulted m their being grouped
together under the designation NOX, although analytic techniques can distinguish clearly
between them (Chapter 6) Of the two, NO2 has the greater impact on human health
(Chapter 16)
Nitrous oxide is ubiquitous even in the absence of anthropogenic sources because it is a
product of natural biologic processes in soil It is not, however, involved in any chemical
reactions m the lower atmosphere to any appreciable extent Although N2O is not generally
considered to be an air pollutant, it participates in upper atmospheric reactions involving the
stratospheric ozone (O3) layer, and it is a greenhouse gas (Chapter 5).
3-1
-------
Although NO3, N2O3, N2O4, and N2C>5 are present only ui very low concentrations,
even in polluted environments, they play a lole in atmospheric chemical reactions leading to
the transformation, transport, and ultimate removal of nitrogen compounds from ambient air
Ammonia (NH3) is generated, on a global scale, during the decomposition of
nitrogenous matter in natural ecosystems, and it may also be produced locally in larger
concentrations by human activities such as the maintenance of dense animal populations
It is discussed because, through its reaction with HNO3, resulting in the formation of aerosol
nitrate, it plays an important role in determining the atmospheric fate of mtiogen oxides
Other NOx-denved compounds that may be found in polluted air include nitrites,
nitrates, nitrogen acids, 2V-mtroso compounds, and organic compounds such as the
peroxyacyl nitrates (RC(O)OONO2, where R represents any one of a large variety of possible
organic groups) (Chapter 6)
The peroxyacyl nitrates, of which PAN (CH3C(O)OONO2) is of most concern in terms
of atmospheric concentrations, have been thoroughly reviewed in the recent EPA document
Air Quality Criteria for Ozone and Other Photochemical Oxidants (U S Environmental
Protection Agency, 1986), and are discussed only briefly in this chapter and elsewhere in this
document.
The discovery of AT-mtroso compounds in air, water, and food has led to concern about
possible human exposure to this family of compounds, some of which have been shown to be
carcinogenic in animals Health concerns also have been expressed about nitrates, which
occur as a component of pardculate matter in the respirable size range, suspended in ambient
air (Chapter 13). Particulate nitrate is produced in atmospheric reactions (Chapter 5)
Nitrates may also occur in significant concentrations in drinking water supplies, but this
occurrence is not believed to be the result of atmospheric production
Photochemical models predict that up to one-half of the original nitrogen oxides emitted
may be converted on a daily basis to nitrates and HNO3 This atmospheric production of
HNOs is an important component of acid rain (see Chapter 5)
Table 3-1 summarizes theoretical estimates of the concentrations of the various nitrogen
oxides and acids that would be present in an equilibrium state assuming initially only
molecules of nitrogen and oxygen at 1 atm pressure, 25 °C, and 50% relative humidity.
3-2
-------
TABLE 3-1. THEORETICAL CONCENTRATIONS OF NITROGEN OXIDES AND
NITROGEN ACIDS THAT WOULD BE PRESENT AT EQUILIBRIUM WITH
MOLECULAR NITROGEN, MOLECULAR OXYGEN, AND WATER IN AIR
AT 25 °C, ONE ATMOSPHERE, 50% RELATIVE HUMIDITY
Concentrations m Hypothetical Atmosphere (ppm)
Compound
Molecular oxygen
(02)
Molecular nitrogen
(N2)
Water (H2O)
Nitac oxide (NO)
Nitrogen dioxide
(N02)
Nitrogen tnoxide
(N03)
Dinitrogen tnoxide
(N203)
Dmitrogen tetroxide
(N204)
Dinitrogen pentoxide
(N205)
cw-Nitrous acid
(HONO)
trans-Nitrous acid
(HONO)
Nitric acid (HONO^
At Equilibnuma
2 06 X 105
7 69 X IO5
1 56 x 104
2 69 x 10"10
1 91 x 10"4
3 88 x 10~16
2 96 X 10"20
2 48 X 10'13
3 16 x 10'17
7 02 X 10"9
1 60 x 10'8
1 33 X 10"3
In Typical Sunlight-Irradiated,
Smoggy Atmospherea'
2 06 x 105
7 69 X 105
1 56 X 104
lo-1
ID'1
10"8 - IO"9
10'8 - 10'9
10'7 - 10'8
10'3 - 10'5
io-3
io-3
10'2 - 10'3
Assumes initially only molecules of nitrogen and oxygen at 1 atm pressure, 25 °C, and 50% relative humidity
The simulations assume that the sunlight is of fixed intensity, with a solar zenith angle of 40° Photochemical
aerosol formation has not been considered here
Theoretical estimates made using computer simulations of the chemical reactions rates in a synthetic smog
mixture with hydrocarbons present
Source Demerjian et al (1974)
3-3
-------
In polluted, sunlight-irradiated atmospheres concentrations of NOy species are far above
thermodynamic equilibrium concentrations Rather, expected concentrations of pollutants are
influenced by emissions and photochemically initiated reactions Table 3-1 lists one set of
estimated concentrations of nitrogen oxides and acids expected under more realistic
conditions, derived from computer simulations of photochemical smog reactions that might
occur in more or less typical urban environments White and Dietz (1984) have postulated
that more than one steady state is possible at certain rates of NOX emissions The calculated
steady states of the free troposphere are shown on Figure 3-1 for a range of NOX
concentrations taking into account CE^ (see White and Dietz, 1984, for all conditions
considered). Photolysis rates represent equinoctial diurnal averages at 45° N
3.2 NITROGEN OXIDES
Table 3-2 summarizes some important physical properties of nitrogen oxides under
standard temperature and pressure conditions of 25 °C and 1 atm, respectively The
remainder of this section descnbes chemical and physical properties of individual nitrogen
oxide species
3.2.1 Nitric Oxide
Nitric oxide is an odorless gas It is also colorless because its absorption bands are all
at wavelengths less than 230 nm, well below the visible wavelengths Nitric oxide is only
slightly soluble in water (0 006 g/100 g of water at 24 °C and 1 atm pressure) It has an
uneven number of valence electrons, but, unlike NO2, it does not dimenze in the gas phase
Nitac oxide is a principal by-product of combustion processes, arising from the
oxidation of molecular nitrogen (N^ in combustion air and of organically bound nitrogen
present in certain fuels such as coal and heavy oil The oxidation of nitrogen in combustion
3-4
-------
Steady State
g
10
10°
NOX Concentration (ppb)
10s
Figure 3-1. Calculated steady states of the free troposphere as a function of NOX
concentration.
Source White and Dietz (1984)
air occurs primarily through a set of reactions known as the extended Zeldovitch mechanism
(Zeldovitch, 1946)
O2 + M -* 2O + M
N2 + O -> NO + N
N + O2 -* NO + O,
(3-1)
(3-2)
(3-3)
3-5
-------
TABLE 3-2. SOME PHYSICAL AND THERMODYNAMIC PROPERTIES
OF THE NITROGEN OXIDES
Thermodynamic Functions
Oxide
Nitric oxide
(NO)
Nitrogen dioxide
(NOz)
Nitrous oxide
(N20)
Dinitrogen
tnoxide (N2O3)
Duutrogen
tetroxide (N2O^)
Dinitrogen
pentoxide
(N205)
Molecular
Weight
(g/mol)
3001
4601
4401
7601
9202
10801
Melting
Point
(°C)a'b
-1636
-112
-908
-102
-113
30
Boiling
Point
(°C)a>b
-1518
212
-885
47
(Decomposes)
212
324
(Decomposes)
Henry's-law
Coefficient at
25 °C (M/atm)c
1 93 X 10"3
1 2 X 10"2
2 47 X 10"2
06 + 02
14 + 08
Solubdity in
0 °C Water
(cm3 at STP/100 g)a
734
Reacts with water
(H2O) fonmng mtnc
acid (HONOj) and
nitrous acid (HONO)
13052
Reacts with H2O,
forming HONO
Reacts with H2O,
forming HONO2 and
HONO
Reacts with H2O,
fonmng HONO2
(Ideal Gas,
Enthalpy of
Formation
(kcal/mol)
2158
791
1961
1980
2 17
27
1 atm, 25 °C)
Entropy
(cal/mol-deg)
50347
5734
5255
7391
7272
828
aMatheson Gas Data Book (Matheson Company, 1966)
bHandbook of Chemistry and Physics (Weast et al, 1986)
°Schwartz and White (1981)
-------
with the additional equation (extended mechanism)
N + OH -* NO + H (3-4)
The high activation energy of reaction 3-2 (75 kcal/mol), coupled with its essential function
of breaking the strong N2 triple bond, make this the rate limiting step of the Zeldovitch
mechanism Due to the high activation energy, this mechanism for NO production proceeds
at a somewhat slower rate than the reactions of fuel constituents and is extremely temperature
sensitive (Bowman, 1973) Moreover, the production of atomic oxygen (O) required for the
first step is also highly temperature sensitive In the immediate vicinity of a flame, the high
temperatures, coupled with the kinetics of the hydrocarbons in the fuel, can drive the
O concentration to several tunes its equilibrium level The local ratio of fuel to air also has
a first order effect on the concentration of O (Bowman, 1973)
The reaction kinetics of thermal NO formation is further complicated by the fact that
certain hydrocarbon radicals can be effective in splitting the N2 bond through reactions such
as (Femmore, 1971)
CH + N2 -> CHN + N (3-5)
The rate of oxidation of the fuel (and intermediate hydrocarbon radical fragments) is usually
sufficiently rapid that only negligible quantities of the fuel radicals are available to attack the
N2 However, under fuel-rich conditions, this can become the dominant mode of breaking
the N2 bond and, in turn, can be responsible for significant NO formation (Engleman et al,
1976) Such reactions appear to have a relatively low activation energy and can proceed at a
rate comparable to oxidation of the fuel Because of the early formation of NO by this
mechanism, relative to that formed by the Zeldovitch mechanism, NO thus formed is often
referred to as "prompt NO" The importance of this mechanism has not been quantified for
practical systems
In fossil fuels such as coal and residual fuel oil, nitrogen compounds are bound within
the fuel matrix Typically, Number 6 residual oil contains 0 2 to 0 8% bound nitrogen by
weight, and coal typically contains 1 to 2 % If this 1 % nitrogen were quantitatively
3-7
-------
converted to NOX, it would account for about 2,000 ppm NOX in the exhaust of a coal-fired
unit. In practice, only a portion of these nitrogen compounds is converted to NOX, with the
remainder being converted to N2 Tests designed to determine the percent of the NOX
emissions due to oxidation of bound nitrogen (Pershing and Wendt, 1976) show that upward
of 80% of the NOX from a coal-fired boiler originates from fuel-bound nitrogen Details of
the kinetic mechanisms involved in fuel nitrogen oxidation are uncertain, due in part to the
variability of molecular composition among the many types (and sources) of coal and heavy
oils and to the complex nature of the heterogeneous processes occurring Experimental
evidence does, however, lend some insight into the processes involved A number of fuel-
bound nitrogen compounds have been cited (Axworthy and Schuman, 1973, Martin et al,
1971; Turner and Siegmund, 1972), but the degree of conversion to NOX does not seem to
be significantly affected by the compound type Oxides of nitrogen conversions arising from
fuel sources seem also to be relatively insensitive to temperature in diffusion flames The
most important parameters in determining fuel-bound nitrogen conversion appear to be the
local conditions prevailing when the nitrogen is evolved from the fuel Under fuel-rich
conditions, this nitrogen tends to form N2, whereas under fuel-lean conditions, significant
amounts of NOX are formed
Nitric oxide formation kinetics in typical furnaces are not fast enough to reach
equilibrium levels in the high temperature flame zone, and the NO destruction mechanisms
are far too slow to allow the NO, once formed, to reach equilibrium at typical stack
temperatures. This is to say that the NO formation process is kinetically controlled
Nitric oxide and NO2 produced in relatively large concentrations at high temperatures
in combustion processes would revert to lower concentrations characteristic approximately of
the equilibrium values shown in Table 3-3 were it not for the fact that combustion equipment
rapidly converts a large fraction of the thermal energy available to useful woik This results
in a rapid cooling of the combustion gases and a "freezing-in" of the produced NO and NO2
near concentrations characteristic of the high temperature phase of the process
A major implication of the fact that NOX emissions are defined by the kinetics of the
process rather than being an equilibrium phenomenon is that NOX emissions can be
effectively modified by changes in the details of the combustion process For clean fuels
such as natural gas or Number 2 distillate oil with no bound nitrogen, the NO formation is
3-8
-------
TABLE 3-3. THEORETICAL EQUILIBRIUM CONCENTRATIONS OF
NITRIC OXIDE AND NITROGEN DIOXIDE IN AIR
(50% RELATIVE HUMIDITY) AT VARIOUS TEMPERATURES
Temperature (K [°C])
298 (24 85)
500 (226 85)
1,000 (726 85)
1,500 (1,226 85)
2,000 (1,726 85)
Concentration (/*§
Nitric Oxide
3 29 X 10"10
(2 63 x 10"10)
8 18 x 10"4
(6 54 x 10"4)
43
(344)
1,620
(1,296)
9,946 25
(7,957)
;/m3 [ppm])
Nitrogen Dioxide
3 53 X 10"4
(1 88 X 10"4)
7 26 X 10"2
(3 86 X 10'2)
338
(180)
1235
(657)
23 88
(12 70)
Source National Research Council (1977)
dominated by the Zeldovitch mechanism Thus, combustion modifications that reduce peak
flame temperature, limit the gas residence time at peak temperatures, and/or reduce the
amount of O available at high temperatures will reduce the NOX emissions Examples of
such modifications include flue gas recirculation, reduced load, reduced combustion air
preheat temperature, water injection, and reduced excess air (Bowen and Hall, 1976a,b,c,
1977a,b,c,d,e)
In furnaces fired with coal or heavy oil, the major portion of the NOX emissions is from
fuel-bound nitrogen conversion. Thus, combustion modifications that reduce the availability
of oxygen when the nitrogen compounds are evolved will reduce the NOX produced
Examples of such modifications are reduction of the amount of excess air during firing,
establishing fuel-rich conditions during the early stages of combustion (staged combustion),
or new burner designs that tailor the rate of mixing between the fuel and air streams (Bowen
and Hall, 1976a,b,c)
3-9
-------
3.2.2 Nitrogen Dioxide
Nitrogen dioxide is a reddish-orange-brown gas with a characteristic pungent odor
Although its boiling point is 21 1 °C, the low partial pressure of NO2 in the atmosphere
prevents condensation Nitrogen dioxide is corrosive and highly oxidizing It has an uneven
number of valence electrons and forms the dimer N2O4 at higher concentrations and lower
temperatures, but the dimer is not important at ambient concentrations In the atmosphere,
NO can be oxidized to NO2 by the thermal reaction
2NO + O2 -* 2NO2 (3-6)
However, this reaction is of minor importance in most ambient situations because other
chemical processes are faster Reaction 3-6 is mainly responsible for the NO2 present in
combustion exhaust gases About 5 !o 10% by volume of the total emissions of NOX from
combustion sources is in the form of NO2, although substantial variations from one source to
another have been observed Under more dilute ambient conditions, photochemical smog
reactions involving hydrocarbons convert NO to NO2 (Chapter 5)
Nitrogen dioxide's principal involvement in photochemical smog stems from its
absorption of sunlight and subsequent decomposition (photolysis) to NO and O Nitrogen
dioxide is an efficient absorber of light over a broad range of ultraviolet (UV) and visible
wavelengths. Only quanta with wavelengths less than about 430 rim, however, have
sufficient energy to cause photolysis It should also be noted that photons having
wavelengths less than about 290 nm are largely absorbed in the upper atmosphere The
effective range of wavelengths responsible for photolysis of NO2 at ground level is,
therefore, 290 nm to 430 nm Because of its absorption properties, NO2 produces
discoloration and reduces visibility in the polluted lower troposphere
3.2.3 Nitrous Oxide
Nitrous oxide is a colorless gas with a slight odor at high concentrations Nitrous oxide
in the atmosphere arises as one product of the reduction of nitrate by a ubiquitous group of
bacteria that use nitrate as their terminal electron acceptor in the absence of oxygen
3-10
-------
(demtnfication) (Brezomk, 1972, Delwiche, 1970, Focht and Verstraete, 1977, Keeney,
1973)
Although N2O does not participate in the photooxidation reactions that produce
undesirable pollutants in the lower troposphere, it is an important atmospheric constituent
because it absorbs long- wave length radiation (therefore, serving as a greenhouse gas), and it
photolyzes in the stratosphere
Nitrous oxide transported to the stratosphere undergoes photolysis by absorbing
UV radiation at wavelengths below 300 nm to produce N2 and singlet oxygen (Johnston and
Selwyn, 1975)
N2O + hi> -* N2 + O('D), where hj> is a unit of radiant energy (3-7)
The singlet oxygen, formed in reaction 3-7, reacts with more N2O to produce two sets
of products
N2 + O2
N2O + O('D) -* and (3-8)
NO + NO
The NO produced enters a catalytic cycle, the net result of which is the regeneration of
NOX and the destruction of O3
NO + O3 -* N02 + O2 (3-9)
03 + hv -* O? + O (3-10)
O2 (3-11)
These reactions are of concern because of the possibility that increased N2O resulting
from demtrification of excess fertilizer may lead to a decrease of stratospheric O3 (Council
for Agricultural Science and Technology, 1976, Crutzen, 1976), with consequent potential
for adverse human health effects
3-11
-------
3.2.4 Nitrogen Trioxide
Nitrogen tnoxide has been identified in laboratory systems containing NO2/O3, NO2/O,
and N2O5 as an important reactive transient (Johnston, 1966) The reaction of NO2 and
O3 provides the primary atmospheric source of NO3
O3 + NO2 -> NO3 + O2 (3-12)
Because it is readily photolyzed by sunlight, NO3 is only important during the nighttime
hours. It reacts rapidly with both inorganic and organic species The reaction with NO2
produces N2C>5, which is the anhydride of HNO3 Nitrogen tnoxide is extremely reactive
with aldehydes, mercaptans, and olefinic hydrocarbons It may be that the nitrate radical
acts as a major sink for these types of organic compounds in the nighttime atmosphere
Details of NO3 radical chemistry are provided in Chapter 5, and a very comprehensive
review concerning the NO3 radical has been published recently (Wayne, 1991)
3.2.5 Dinitrogen Trioxide
In the atmosphere, N2O3 (also known as nitrogen sesquioxide) is in equilibrium with
NO and NO2 according to the following equation
NO + NO2 & N2O3 (3-13)
The equilibrium concentrations at typical urban levels of NO and NO2 range from about
10"4 /tg/m3 (»10~7 ppm) to 10~6 /*g/m3 («10"9 ppm) (Table 3-4) Dinitrogen tnoxide
concentrations of this magnitude (i e , «10"8 ppm = 10"5 ppt) would have little influence on
atmosphenc chemistry. The rate constants for reaction of N2O3 with organics are not large
enough to compensate for the low N2O3 concentration Dinitrogen tnoxide is the anhydnde
of nitrous acid (HONO) and reacts with liquid water to form the acid
N2O3 + H2O -> 2HONO (3-14)
3-12
-------
TABLE 3-4. THEORETICAL CONCENTRATIONS OF DINITROGEN TRIOXTOE
AND DINITROGEN TETROXIDE IN EQUILIBRIUM WITH VARIOUS LEVELS
OF GASEOUS NITRIC OXIDE AND NITROGEN DIOXIDE IN AIR AT 25 °C
Concentration (ppm)
Nitric Oxide
005
010
050
100
Nitrogen Dioxide
005
010
050
100
Dimtrogen Tnoxide
1 3 X 10"9
5 2 x 10'9
1 3 x 10"7
5 2 x 10"7
Dimtrogen Tetroxide
1 7 X 10"8
6 8 X 10'8
1 7 X 10"6
6 8 x 10'6
Source National Research Council (1977)
However, due to the low concentrations of N2O3 expected in the atmosphere, this reaction
would serve as a minimal source of HONO
3.2.6 Dinitrogen Tetroxide
Dimtrogen tetroxide (also known as nitrogen tetroxide) is the dimer of NO2 formed by
the association of NO2 molecules It also readily dissociates to establish the equilibrium.
2NO2 & N2O4 (3-15)
Table 3-4 presents theoretical predictions of concentrations of N2O3 and N2O4 in equilibrium
with various NO and NO2 concentrations
3.2.7 Dinitrogen Pentoxide
Dimtrogen pentoxide is a nighttime component of the atmosphere because it is formed
in the following reaction between NO2 and NO3, the latter of which can exist in appreciable
quantities only in the absence of sunlight
NO2 + NO3 + M ?* N2O5 (3-16)
3-13
-------
Nitrogen dioxide, NO3, and N2O5 are in equilibrium, however, the value of the equilibrium
constant is somewhat uncertain (Finlayson-Pitts and Pitts, 1986) Even though this
uncertainty exists, there should be sufficient quantities of N2C>5 present to have an impact on
nighttime chemistry Dinitrogen pentoxide combines with liquid water to form HNO3, and
in so doing, the equilibrium in reaction 3-16 is shifted to the right, and NO3 is consumed
The overall result of these reactions is the removal of a reactive radical (NO3) and the
formation of an acid (HNO3) that will be readily removed from the atmosphere by
depositional processes
3.3 NITRATES, NITRITES, AND NITROGEN ACIDS
3.3.1 Nitric Acid
Nitric acid in the gaseous state is colorless and photochemically stable in the
troposphere. The major pathways for atmospheric formation of HNO3 involve reaction 3-17
during the daylight period when hydroxyl radicals are present and reaction 3-18 at night
The N2C>5 is formed during the nighttime period as described in Section 324
NO2 + HO -* HNO3 (3-17)
N2O5 + H2O0) -» 2HNO3 (3-18)
Due to its volatility, HNO3 does not condense into aerosol at concentrations piesent in the
atmosphere. However, HNO3 can react with NH3 gas or aerosols containing alkaline crustal
material to produce particulate nitrates (Wolff, 1984) Wet and dry deposition of HNO3 are
primary removal mechanisms Neutralization by alkaline gaseous and aerosol materials
serves as a sink, as well.
3.3.2 Nitrous Acid
Nitrous acid is an important atmospheric constituent because it photolyzes, yielding a
hydroxyl radical Ambient concentrations up to 8 ppb have been recorded in urban
atmospheres (Harris et al, 1982) At the present time, the source(s) of HONO is highly
3-14
-------
uncertain As indicated in Section 325, the reaction of NO and NO2 in the presence of
water will produce HONO However, at the concentrations of NO and NO2 normally found
in the atmosphere, this is a negligible source When hydroxyl radicals are present (daytime),
HONO can be formed by reaction 3-19 However, because HONO is rapidly destroyed by
photolysis, only a very small steady-state concentration of HONO could be produced by this
pathway Alternate mechanisms involving heterogeneous reactions have been proposed, but
these are difficult to verify. It has been suggested that direct emission of HONO from
automobiles may serve as a source (Pitts et al, 1984)
NO + OH + M ^ HONO (3-19)
3.3.3 Organic Nitrates
Organic nitrates are formed when organic compounds are oxidized in the presence of
oxides of nitrogen A large number of alkyl nitrates could be formed by reactions such as
3-20 and 3-21, but few ambient measurements have been reported By far the most studied
organic nitrate is PAN, which results from the combination of a peroxyacetyl radical with
NO2 (reaction 3-22)
R-CH2-O + NO2 -» RC'H2-ONO2 (3-20)
RCH2O-O + NO -> RCH2ONO2 (3-21)
CH3C(O)OO + NO2 -* CH3C(O)OONO2 (3-22)
Peroxyacetyl nitrate was first identified as a component of Los Angeles smog Since that
tune, it has been shown to be ubiquitous in the atmosphere Peroxyacetyl nitrate is thermally
labile, with a lifetime of only minutes at temperatures above 25 °C However, at the lower
temperatures aloft in the troposphere, PAN has a sufficiently long lifetime to act as a storage
reservoir for NO2 Peroxyacetyl nitrate slowly decomposes through reactions with the
hydroxyl radical and sunlight (see Chapter 5)
3-15
-------
3.3.4 Aerosol Nitrates
Particulate ammonium nitrate is produced when NH3 reacts with HNO3 The reaction
is reversible under atmospheric conditions, with a temperature- and relative-humidity-
dependent equilibrium constant Stelson and Seinfeld (1982) have calculated the equilibrium
concentrations of gaseous NH3 and HNO3 and the resulting solid or aqueous ammonium
nitrate using fundamental thermodynamic principles The system is very complex and must
also include sulfate, because NH3 will preferentiaEy react with sulfuric acid to give
ammonium bisulfate and ammonium sulfate A summary of the physics and chemistry of the
NH3-sulfuric acid-HNO3 system has been provided by Seinfeld (1986)
3.4 AMMONIA
Ammonia is a colorless gas with a pungent odor It is extremely soluble in water,
forming the ammonium and hydroxy (OH") ions Ammonia is the only gaseous basic
compound present in appreciable concentrations in the atmosphere It reacts rapidly with
sulfuric acid to form ammonium bisulfate and sulfate aerosols Ammonia is removed from
the atmosphere by wet and dry deposition It will react slowly with hydroxyl radical
As shown in reaction 3-23, this produces amide (NH^ radicals, which can be oxidized to
NOX products or serve as a sink for NO and NO2 via reactions of the types shown in
reactions 3-24 and 3-25 The atmospheric fate of the NH2 radical is not well understood at
present.
NH3 + OH -* NH2 + H2O (3-23)
NH2 + NO -> N2 + H2O (3-24)
NH2 4 NO2 -> N2O + H2O (3-25)
3-16
-------
3.5 W-NITROSO COMPOUNDS
Organic mtroso compounds contain a mtroso group (-N = O) attached to a nitrogen or
carbon atom According to Magee (1971), W-nitroso compounds generally can be divided
into two groups—one group includes the dialkyl, alkylaryl, and diaryl nitrosamines, and the
other includes the alkyl and aryl mtrosamides
The principal chemical reaction involved in the formation of Af-nitrosamines is that of
the secondary amines with HONO Nitrosation is effected by agents having the structure
ONX, where X = 0-alkoxyl, nitrite ion (NO2"), nitrate ion, halogen, tetrafluoroborate,
hydrogen sulfate, or OH" The equilibrium reaction of nitrosonium ion (ON+), HONO, and
NO,",
ON+ + OH' <* HNO2 ^ H+ + NO2", (3-26)
is shifted to the right at pH > 7 The simplest form of mtrosation of amines involves
electrophilic attack by the ON+ and subsequent deprotonation
Mirvish (1970) studied the kinetics of dimethylnitrosamine mtrosation and pointed out
that the chief mtrosatuig agent at pH 1 is N2O3, the anhydride of HONO, which forms
reversibly from two HONO molecules The formation of nitrosamines is dependent on the
pK of the amine
Nitroso compounds are characteristically photosensitive and the mtroso group is split by
UV radiation Gaseous nitrosamines may be denitrosated by visible light Absorption
spectra of several nitrosamines are given in the literature (Rao and Bhaskar, 1969), the
characteristic spectra show a low intensity absorption maximum around 360 nm and an
mtense band around 235 nm Nitrosamines show threie relatively intense bands in the
infrared region of 7 1 to 7 4 ^m, 7.6 to 8 6 pm, and 9 15 to 9 55 pm Nuclear magnetic
resonance, infrared, UV, and mass spectrometry spectra have been reviewed by Magee et al
(1976)
Atmospheric reactions involving nitrosamines are discussed in Chapter 5
3-17
-------
3.6 SUMMARY
There are seven nitrogen oxides that may be present in the ambient air NO, NO2,
N2O, NO3, N2O3, N2O4, and N2C>5 Of these, NO and NO2 are generally present in highest
concentrations in polluted atmospheres Their interconvertibility in photochemical smog
reactions has frequently resulted in their being grouped together under the designation NOX,
although analytic techniques can distinguish cleaily between them Of the two, NO2 is the
more toxic and irritating compound
Nitrous oxide is ubiquitous even in the absence of anthropogenic sources because it is a
product of natural biologic processes in soil Nitrous oxide is inert in the lower troposphere,
so it is not a participant in photochemical smog leactions Although N2O is not generally
considered to be an air pollutant, it is a greenhouse gas and a reactant in upper atmospheric
reactions involving the stratospheric O3 layer
Nitrogen trioxide, N2O3, N2O4, N2O5, and HNO3 all play a role in atmospheric
chemical reactions leading to the transformation, transport, and ultimate removal of nitrogen
compounds from ambient air
Ammonia is emitted during the decomposition of nitrogenous matter in natural
ecosystems, but it may also be produced locally by human activities such as the maintenance
of dense animal populations It reacts with HNO3 in the troposphere to produce ammonium
nitrate aerosol Deposition of this aerosol then serves as a sink for the oxides of nitrogen
Compounds derived from NOX, including nitrites, nitrates, nitrogen acids, N-nitroso
compounds, and organic compounds such as the peroxyacyl nitrates (RC(O)OONO2, where
R represents any one of a large variety of possible organic groups), may also be found in
polluted air.
The peroxyacyl nitrates, of which PAN (CH3C(O)OONO2) is of most concern in terms
of atmospheric concentrations, have been thoroughly reviewed in the recent EPA document
Air Quality Criteria for Ozone and Other Photochemical Oxidants (U S Environmental
Protection Agency, 1986)
The discovery of Af-nitroso compounds (some of which have been shown to be
carcinogenic in animals) in air, water, and food has led to concern about possible human
exposure to this family of compounds Health concerns also have been expressed about
HNO3 vapor and particulate nitrates, occurring as a component of particulate matter in the
3-18
-------
respirable size range, suspended in ambient air These nitrates are produced in atmospheric
reactions Nitrates may also occur in significant concentrations in public and private
drinking water, but this occurrence is not believed to be the result of atmospheric production
Photochemical models predict that up to one-half of the original nitrogen oxides emitted
may be converted on a daily basis to nitrates and HNO3 This atmospheric production of
HNO3 is an important component of acidic ram and is a major contributor to the dry
deposition flux of acid gases to the earth's surface
3.6.1 Nitrogen Oxides
Nitric oxide is an odorless and colorless gas It is a major by-product of the
combustion process, arising both from the oxidation of N2 in the combustion air and of
nitrogen compounds bound in the fuel molecule The amount of NO formed from the
oxidation of N2 is dependent upon such parameters as peak flame temperature, quantity of
combustion air, and gas residence tune in the combustion chamber The amount of NO
arising from oxidation of fuel-bound nitrogen does not seem to depend significantly on either
the type of nitrogen compound involved or the flame temperature, but instead depends on the
specific air-to-fuel ratio at various stages in combustion and the nitrogen content of the fuel
Nitrogen dioxide is produced in minor quantities m the combustion process (5 to 10%
of the total oxides of nitrogen) In terms of significant atmospheric loading in populated
areas, NO2 arises mainly from the photochemically initiated conversion of NO to NO2 by a
variety of chemical processes in the atmosphere Nitrogen dioxide is corrosive and highly
oxidizing Its reddish-orange-brown color arises from its preferential absorption of blue
light Because of its strong absorption in this range (and also in the UV spectrum), NO2 can
cause visibility reduction and affect the spectral distribution of solar radiation m the polluted,
lower atmosphere
3.6.2 Nitrates, Nitrites, and Nitrogen Acids
Other compounds derived from NOX by means of atmospheric chemical processes
include nitrites, nitrates, nitrogen acids, organic compounds, such as the peroxyacyl nitrates,
and, possibly, the JV-mtroso compounds
3-19
-------
Nitric acid, a strong acid and powerful oxidizing agent, is colorless and
photochemically stable in the gaseous state Its high volatility prevents condensation into
droplets in the atmosphere It can, however, react with NH3 and alkaline aerosol materials
to form aerosol nitrates.
3.6.3 ^V-Nitroso Compounds
The 2V-nitroso family comprises a wide variety of compounds, all containing a nitroso
group (-N = O) attached to a nitrogen or carbon atom Their formation in the atmosphere
has been postulated to proceed through chemical reaction of amines with NOX and
NOx-derivatives in gas phase reactions and/or through atmospheric reactions involving
aerosols. Nitroso compounds are characteristically photosensitive and the nitroso group is
split by the UV radiation in sunlight Gaseous mtrosamines may also be demtrosated by
visible light.
3-20
-------
REFERENCES
Axworthy, A E , Schuman, M (1973) Investigation of the mechanism and chemistry of fuel nitrogen conversion
to nitrogen oxides in combustion In Proceedings, coal combustion seminar, June, Research Triangle
Park, NC Research Triangle Park, NC U S Environmental Protection Agency, Office of Research and
Development, pp 9-41, EPA report no EPA-650/2-73-023 Available from NTIS, Springfield, VA,
PB-224210
Bowen, J S , Hall, R E , eds (1976a) Proceedings of the stationairy source combustion symposium Volume I
Fundamental research, September 1975, Atlanta, GA Washington, DC U S Environmental Protection
Agency, Office of Research and Development, EPA report no EPA-600/2-76-152a Available from
NTIS, Springfield, VA, PB-256320
Bowen, J S , Hall, R E , eds (1976b) Proceedings of the stationaoy source combustion symposium Volume n
Fuels and process research and development, September 1975, Atlanta, GA Washington, DC U S
Environmental Protection Agency, Office of Research and Development, EPA report no
EPA-600/2-76-152b Available from NTIS, Springfield, VA, PB-256321
Bowen, J S , Hall, R E , eds (1976c) Proceedings of the stationary source combustion symposium Volume ITT
Field testing and surveys, September 1975, Atlanta, GA Washington, DC U S Environmental
Protection Agency, Office of Research and Development, EPA report no EPA-600/2-76-152c Available
from NTIS, Springfield, VA, PB-257146
Bowen, J S , Hall, R E , eds (1977a) Proceedings of the second stationary source combustion symposium
Volume I Small industrial, commercial, and residential systems, August-September, New Orleans, LA
Research Triangle Park, NC U S Environmental Protection Agency, Office of Research and
Development, EPA report no EPA-600/7-77-073a Available from NTIS, Springfield, VA, PB-270923.
Bowen, J S , Hall, R E , eds (1977b) Proceedings of the second stationary source combustion symposium
Volume n Utility and large industrial boilers, August-September, New Orleans, LA Research Triangle
Park, NC U S Environmental Protection Agency, Office of Research and Development, EPA report no
EPA-600/7-77-073b Available from NTIS, Springfield, VA, PB-271756/9
Bowen, J S , Hall, R E , eds (1977c) Proceedings of the second stationary source combustion symposium
Volume HI Stationary engine, industrial process combustion systems, and advanced processes,
August-September, New Orleans, LA Research Triangle Pcirk, NC U S Environmental Protection
Agency, Office of Research and Development, EPA report mo EPA-600/7-77-073c Available from
NTIS, Springfield, VA, PB-271757/7
Bowen, J S , Hall, R E , eds (1977d) Proceedings of the second stationary source combustion symposium
Volume IV Fundamental combustion research, August-September, New Orleans, LA Research Triangle
Park, NC U S Environmental Protection Agency, Office of Research and Development, EPA report no
EPA-600/7-77-073d Available from NTTS, Springfield, VA, PB-274029
Bowen, J S , Hall, R E , eds (1977e) Proceedings of the second stationary source combustion symposium
Volume V Addendum, August-September, New Orleans, LA Research Triangle Park, NC U S
Environmental Protection Agency, Office of Research and Development, EPA report no
EPA-600/7-77-073e Available from NTIS, Springfield, VA, PB-274897
Bowman, C T (1973) Kinetics of nitric oxide formation in combustion processes In Fourteenth symposium
(international) on combustion, August 1972, University Park, PA Pittsburgh, PA The Combustion
Institute, pp 729-738
3-21
-------
Brezonik, P L (1972) Nitrogen sources and transformations in natural waters In Allen, H E , Kramer, J R ,
eds Nutrients in natural waters New York, NY John Wiley & Sons, Inc , pp 1-50
Council for Agricultural Science and Technology (1976) Effect of increased nitrogen fixation on stratospheric
ozone Ames, IA Iowa State University, Department of Agronomy, report no 53, January 19
Crutzcn, P. J (1976) Upper limits on atmospheric ozone ieductions following increased application of fixed
nitrogen to the soil Geophys Res Lett 3 169-172
Delwiche, C C (1970) The nitrogen cycle Sci Am 223 137-147
Demerjian, K L , Kerr, J A , Calvert, J G (1974) The mechanism of photochemical smog formation
In1 Pitts, J N , Jr , Metcalf, R L , Lloyd, A C , eds Advances in environmental science and
technology, v 4. New York, NY John Wiley & Sons, pp 1-262
Englcman, V S , Sirrunski, V J , Bartok, W (1976) Mechanism and kinetics of the formation of NOX and other
combustion pollutants Phase n Modified combustion Research Triangle Park, NC U S Environmental
Protection Agency, Industrial Environmental Research Laboratory, EPA report no EPA-600/7-76-009b
Available from NTIS, Springfield, VA, PB-258875
Fenimore, C P. (1971) Formation of nitric oxide in prenuxed hydrocarbon flames In Thirteenth symposium
(international) on combustion, August 1970, Salt Lake City, UT Pittsburgh, PA The Combustion
Institute; pp 373-380
Finlayson-Pitts, B J , Pitts, J N , Jr (1986) Atmospheric chemistry fundamentals and experimental techniques
New York, NY John Wiley & Sons, pp 961-1007
Focht, D. D , Verstraete, W (1977) Biochemical ecology of nitrification and demtnfication Adv Microb Ecol
1 135-214
Harris, G W , Carter, W P L , Winer, A M , Pitts, J N , Jr , Platt, U , Perner, D (1982) Observations of
nitrous acid in the Los Angeles atmosphere and implications for predictions of ozone-precursor
relationships Environ Sci Technol 16 414-419
Johnston, H S (1966) Experimental chemical kinetics In Gas phase reaction rate theory New York, NY The
Ronald Press Company, pp 14-34
Johnston, H S , Selwyn, G S (1975) New cross sections for the absorption of near ultraviolet radiation by
nitrous oxide (N2O) Geophys Res Lett 2 549-551
Keeney, D R (1973) The nitrogen cycle in sediment-water systems J Environ Qual 2 15-29
Magee, P. N (1971) Toxicity of nitrosamines their possible human health hazards Food Cosmet Toxicol
9 207-218
Magee, P. N., Montesano, R , Preussmann, R (1976) N-mtroso compounds and related carcinogens In Searle,
C. E , ed Chemical carcinogens Washington, DC American Chemical Society, pp 491-625 (ACS
monograph 173)
Martin, G B , Pershing, D W , Berkau, E E (1971) Effects of fuel additives on air pollutant emissions from
distillate-oil-fired furnaces Research Triangle Park, NC U S Environmental Protection Agency, Office
of Air Programs, report no AP-87 Available from NTIS, Springfield, VA, PB-213630
3-22
-------
Matheson Company, Inc (1966) Matheson gas data book 4th ed East Rutherford, NJ The Matheson Company,
Inc
McConnell, J C (1973) Atmospheric ammonia J Geophys Res 78 7812-7821
Mirvish, S S (1970) Kinetics of dimethylamine mtrosation in relation to nitrosamine carcinogenesis J Natl
Cancer Inst 44 633-639
National Research Council (1977) Nitrogen oxides Washington, DC National Academy of Sciences
Pershing, D W , Wendt, J O L (1976) Pulverized coal combustion the influence of flame temperature and
coal composition on thermal and fuel NOX In Sixteenth symposium (international) on combustion,
August, Cambridge, MA Pittsburgh, PA The Combustion Institute, pp 389-399
Pitts, J N , Jr , Biermann, H W , Winer, A M , Tuazon, E C (1984) Spectroscopic identification and
measurement of gaseous nitrous acid in dilute auto exhaust Atmos Environ 18 847-854
Rao, CNR, Bhaskar, K R (1969) Spectroscopy of the nitroso group In Feuer, H , ed The chemistry of
the mtro and nitroso groups part 1 New York, NY Interscience Publishers, pp 137-163
Schwartz, S E , White, W, H (1981) Solubility equilibria of the nitrogen oxides and oxyacids in dilute aqueous
solution Adv Environ Sci Eng 4 1-45
Seinfeld, J H (1986) Atmospheric chemistry and physics of air pollution New York, NY John Wiley & Sons
Stelson, A W , Seinfeld, J H (1982) Thermodynamic prediction of the water activity, NEySTC^ dissociation
constant, density and refractive index for the NH^NC^^NH^SO^-H^O system at 25 °C Atmos
Environ 16 2507-2514
Turner, D W , Siegmund, C W (1972) Staged combustion and flue gas recycle potential for minimizing NOX
from fuel oil combustion Presented at The American Flame Research Committee Flame Days,
September, Chicago, IL
U S Environmental Protection Agency (1982) Air quality criteria for oxides of nitrogen Research Triangle
Park, NC Office of Health and Environmental Assessment, Environmental Criteria and Assessment
Office, EPA report no EPA-600/8-82-026 Available from NTIS, Springfield, VA, PB83-13 1011
U S Environmental Protection Agency (1986) Air quality criteria for ozone and other photochemical oxidants
Research Triangle Park, NC Office of Health and Environmental Assessment, Environmental Criteria
and Assessment Office, EPA report nos EPA-600/8-84-Q20aF-eF 5v Available from NTIS,
Springfield, VA, PB87-142949
Wayne, R P , Barnes, I , Biggs, P , Burrows, J P , Canosa-Mas, C E , Hjorth, J , Le Bras, G , Moortgat,
G K , Perner, D , Poulet, G , Restelli, G , Sidebottom, H (1991) The nitrate radical physics,
chemistry, and the atmosphere Atmos Environ Part A 25 1-203
Weast, R C , Astle, M J , Beyer, W H , eds (1986) CRC handbook of chemistry and physics
a ready-reference book of chemical and physical data 67th ed Boca Raton, FL CRC Press, Inc ,
pp B-lll - B-112
Wolff, G T (1984) On the nature of nitrate in coarse continental aerosols Atmos Environ 18 977-981
3-23
-------
White, W. H , Dietz, D. (1984) Does the photochemistry of the troposphere admit more than one steady state?
Nature (London) 309 242-244
Zcldovich, J. (1946) The oxidation of nitrogen in combustion and explosions Acta Physicochim URSS
21: 577-628
3-24
-------
4. AMBIENT AND INDOOR SOURCES AND
EMISSIONS OF NITROGEN OXIDES
4.1 INTRODUCTION
Sources of emissions of nitrogen oxides (NOX) in the ambient environment and indoors
play a major role in the determination of nitrogen dioxide (NO^) levels that are observed in
the ambient air and indoors Nitrogen dioxide levels are discussed in Chapter 7 Ambient
emission sources include both anthropogenic and natural categories Anthropogenic sources
primarily consist of emissions from various transportation vehicles, fuel combustion from
stationary sources, industrial processes, solid waste disposal and other sources such as forest
fires Natural sources of NO2 emissions include lightning, stratospheric injection and release
from soils and the oceans Indoor sources include gas stoves used for cooking, unvented
space heaters fueled with natural gas and propane, kerosene heaters, wood stoves and
tobacco products Knowledge of emission sources and patterns for NOX is important for air
quality planning
The present chapter, in Section 4 2, first discusses ambient sources of NOX and breaks
these down into anthropogenic and natural categories (Sections 421 and 422,
respectively) Next, a brief discussion of global estimates of NOX emissions is presented in
Section 423 Then an analysis of U S NOX emission sources, levels, and trends is
discussed in Section 424, followed by Section 4 2 5, a comparison of NOX emission
estimates The chapter then describes indoor NOX emission sources in Section 4 3, including
discussion of the formation of NOX in combustion in gas-fueled household appliances
(Section 432) Then, specific indoor sources are discussed in the following sections
(1) gas stoves used for cooking (Section 433), (2) unvented space heaters fueled with
natural gas and propane (Section 434), (3) kerosene heaters (Section 435), (4) wood
stoves (Section 436), and (5) tobacco products (Section 437) Next, a comparison of
emissions from these indoor sources and their influence on indoor air quality is discussed in
Section 438 Finally, the chapter is summarized in Section 4 4
4-1
-------
4.2 AMBIENT SOURCES OF NITROGEN OXIDES
Ambient sources of NOX can be classified into anthropogenic and natural categories
Most of the anthropogenic data for Section 421 was derived from the National Air Pollutant
Emissions Estimate 1940 to 1990 (U S Environmental Protection Agency, 1991a), and a
majority of the material for the natural source category for Section 422 was adopted from
the National Acid Precipitation Assessment Program Technology Report on Emissions
Involved in Acidic Deposition Processes (Placet et al, 1991)
Ideally, national emission estimates should result when the emissions of each individual
source in the country are added together However, this "bottom-up" approach is not
feasible for a national estimate, and, therefore, emission estimates presented here are based
on the "top-down" calculation approach When using the "top-down" estimating procedure,
emissions are determined through the application of the formula
Emissions = Activity Level x Emission Factor x (1 — Removal Efficiency)
The activity levels (e g., kilograms of coal consumed) are activities that are associated with
pollutant releases The emission factors (e g , grams of NOX emission per kilogram of coal
consumed) are estimated values that relate the quantity of a pollutant to some measure of
activity. The removal efficiency represents the fraction of emissions removed through the
application of control devices.
The methods used for preparing the data presented m this section are as similar as
possible to those used for the Aerometnc Information Retrieval System (AIRS) data
preparation (U S. Environmental Protection Agency, 199Ib) The source files include the
results of investigations of millions of sources of air pollution Sources are categorized into
two main classifications' (1) "point" sources (e g , petroleum refineries and utility boilers)
are facilities, plants, or activities for which individual records are maintained in the source
file; and (2) "area" sources (e.g , motor vehicles) are those activities for which aggregated
source and emission information are maintained for entire source categories
National activity data for individual source categories are obtained from many different
publications. Emission factors, however, are generally obtained from the U S
Environmental Protection Agency's (EPA's) Compilation of Air Pollutant Emission Factors,
4-2
-------
AP-42 (U S Environmental Protection Agency, 1985), and from the EPA's mobile source
emission factor model (U S Environmental Protection Agency, 1989)
4.2.1 Anthropogenic Sources of Nitrogen Oxides
Table 4-1 provides the major source categories and subcategones of anthropogenic NOX
emissions The following sections briefly describe the methodology for estimating the annual
emissions by major source categories
TABLE 4-1. MAJOR SOURCE CATEGORIES3
Category
Subcategory
Anthropogenic Sources
Transportation
Stationary Source Fuel
Combustion
Industrial Processes
Solid Waste Disposal
Miscellaneous
Highway Vehicles (Gasoline- and Diesel-Powered)
Aircraft
Railroads
Vessels
Off-Highway Vehicles and Machinery
Electric Utilities
Industrial Boilers
Commercial and Institutional Boilers and Furnaces
Residential Furnaces and Space Heaters
Chemical Manufacturing
Petroleum Refining
Primary and Secondary Metals
Iron and Steel Mills
Mineral Products
Food Production and Agriculture
Industrial Organic Solvent Use
Petroleum Production and Marketing
Incineration
Open Burning
Forest Fires
Other Burning (Agricultural Burning, Coal Refuse Burning,
and Structure Fires)
Miscellaneous Organic Solvent Evaporation
aU S Environmental Protection Agency (199 la)
For the purpose of this report, forest fires are considered anthropogenic sources, although some fires may be
caused by nature
4-3
-------
4.2.1.1 Transportation
This category includes gasoline- and diesel-powered motor vehicles, aircraft, railroads,
vessels, and off-highway vehicles
The dominant gasoline-fueled powerplant is a homogeneous-charge spark-ignition
engine. "Homogeneous charge" implies that an effort is made to provide a uniformly mixed
charge of fuel and air in the cylinder prior to combustion This necessitates introduction of
the fuel into the air upstream of the engine cylinder
In contrast, the diesel is a stratrfied-charge compression-ignition engine The fuel is
injected into the cylinder beginning near the end of the compression stroke This produces a
nonuniform mixture that ranges from 100% fuel at the exit from the injector nozzle to
0% fuel in other regions of the cylinder
Both of these engines require a higher average cylinder-gas temperature during the
expansion stroke than during the compression stroke in order to produce useful output work,
and the combustion process provides that difference At the high temperatures reached
during combustion, a small fraction of the cylinder air forms NO If chemical equilibrium
existed at all tunes in the cylinder, that NO would disappear as the cylinder gas cooled
during expansion. The nitrogen atoms in the NO would revert to molecular nitrogen and the
oxygen atoms would either revert to molecular oxygen or appear in water or an oxide of
carbon. The engine cycle is completed so quicldy, however, that chemical equilibrium
cannot be established and some of the NO is discharged with the exhaust gas
Ultimately, the NO exhausted from the tailpipe oxidizes over tune to NO2 In fact, a
small amount of the NO formed during combustion may already experience that oxidation
before exiting the tailpipe. Because emission standards are expressed on a mass basis, all
oxides of nitrogen exhausted from the tailpipe are assigned the molecular weight of NO2 and
are referred to as NOX.
Federal regulation of NOX from U S automobiles and light trucks began in 1973 and
regulation of NOX from heavy-duty trucks began in 1984 Regulation of unburned
hydrocarbons (HCs) and carbon monoxide (CO) emitted from these vehicles began even
earlier. The fuel economy of automobiles and light trucks (i e , light-duty vehicles [LDVs])
is also subject to federal standards
4-4
-------
Because of the significance of all these regulations to engine and vehicle design, it is
helpful to understand their interactions Emission trends for NOX, HCs, and CO, all
expressed in terms of mass per unit of engine output energy, are illustrated in Figure 4-1 for
a homogeneous-charge engine with ignition timing set for best fuel economy Thermal
efficiency, which is proportional to fuel economy, is also shown
o
i
u
I
«
I
Rich
Air-Fuel Ratio
Lean
Figure 4-1. Production of hydrocarbons, carbon monoxide, and nitrogen oxide as a
function of air-fuel ratio.
Source Heywood (1988)
Nitrogen oxide production is seen in Figure 4-1 to peak just to the lean side of the
chemically correct, or stoichiometnc, air-fuel ratio It is evident that NOX can be kept low
by running with a rich mixture, but the associated effects on HCs, CO, and fuel economy are
evident Similarly, operating with a very lean mixture lowers NOX
4-5
-------
To meet current emissions standards, gasoline-fueled LDVs employ a catalyst-based
exhaust aftertreatment system that, when the mixture is tightly controlled about the
stoichiometric ratio, is able to effect substantial reductions in all three regulated emissions
Unfortunately, the catalysts used are ineffective in reducing engine-out NOX in lean mixtures
To date, efforts to develop a lean-mixture NOX catalyst system with both high efficiency and
good durability have been unsuccessful Therefore, the only viable aftertreatment strategy
today necessitates using oxidizing (for HCs and CO) and reducing (for NOX) catalysts and,
for most engine operating conditions, a stoichiometric mixture The catalysts used are now
typically packaged together to make what is known as a three-way catalytic converter
Two other techniques have been used to decrease NOX emissions, but they are adjuncts
to, rather than replacements for, the three-way catalyst One is to delay the tuning of
combustion in the cycle by retarding the spark This lowers peak cylinder-gas temperature,
hence NO production, but is undesirable for its adverse effect on fuel economy The second
is to use exhaust-gas recirculation (EGR) back into the engine intake With a stoichiometric
air-fuel ratio, EGR serves as a diluent to decrease combustion temperature With a fast-burn
A
combustion-chamber design, a little EGR can actually improve the fuel economy of a
stoichiometnc engine somewhat, but excessive EGR deteriorates dnvability
Because of its charge stratification, the diesel engine is able to operate at much leaner
overall air-fuel ratios than the homogeneous-charge engine In fact, at a given operating
speed, diesel load is controlled by varying the overall air-fuel ratio rather than by
manipulating an intake throttle as on the homogeneous-charge engine However, the low
NOX production that might be expected from Figure 4-1 with such a lean mixture does not
occur. As injected fuel vaporizes and diffuses into the surrounding cylinder air, from the
spectrum of local air-fuel ratios established in the diesel, conditions for autoigmtion and
combustion are most favorable around the stoichiometnc ratio Thus, much of the fuel burns
locally in a mixture-ratio range near that for peak NOX production in Figure 4-1, irrespective
of the lean cylinder-average mixture
As is true of the homogeneous-charge engine, NOX from the diesel can be decreased by
delaying the timing of combustion through retardation of injection tuning, but with an
adverse effect on fuel economy Exhaust-gas recirculation is also effective in lowering NOX
at partial engine loads, but normally that increases particulate matter (PM)
4-6
-------
Heavy-duty diesels typically consume most of their fuel at high loads, where the
cylinder-average air-fuel ratio is usually already set at the rich end of the acceptable range
for diesel combustion (although still leaner than stoichiometric), as limited by the tune
available for proper mixing of fuel and air Under these conditions, additional combustion
diluent in the form of EGR is unwelcome, so EGR is generally not a useful NOx-reduction
option in heavy-duty applications
The federal government is examining introduction of alternative fuels in metropolitan
areas with the poorest air quality Included on the list of options are methanol, natural gas,
reformulated gasoline, ethanol, liquefied petroleum gas (LPG), and hydrogen A discussion
of alternative fuels, however, is outside the scope of this document
Highway Vehicles
Emissions from gasoline- and diesel-powered motor vehicles are based on vehicle miles
traveled and emission factors Eight vehicle categories are considered gasoline-powered
automobiles, diesel-powered automobiles, light-duty gasoline trucks (< 6,000 Ib), light-duty
gasoline trucks (6,000 to 8,500 Ib), light-duty diesel trucks, heavy-duty gasoline trucks and
buses, heavy-duty diesel trucks and buses, and motorcycles Emission factors were obtained
from the MOBILE4 model (U S Environmental Protection Agency, 1989) This model was
designed to be used as a tool for estimating exhaust and running loss emissions from highway
vehicles in nonattamment areas and in urban air sheds The revised model, MOBILESa,
updated and corrected May 20, 1993 (Federal Register, 1993), estimates higher NO2
emissions
Aircraft
Emissions from aircraft are based on the number of take-offs and landings reported by
the Federal Aviation Administration (U S Department of Transportation, Annual a) and on
AP-42 emission factors for various types of aircraft
4-7
-------
Railroads
Emissions for railroads are based on diesel and residual fuel oil consumption by
railroads as reported by the Energy Information Administration (U S Department of Energy,
Monthly). Average emission factors were used as applicable for each fuel type
Vessels
The consumption of diesel fuel, residual oil, and coal by vessels operating inside the
United States boundaries was obtained from the U S Department of Energy (Monthly,
Annual' a, Annual, b) Gasoline consumption is based on national boat and motor
registrations, together with usage factors (gallons per motor per year) (U S Department of
Energy, Monthly) and marine gasoline sales, as reported by the U S Department of
Transportation (Annual b) The estimates of fuel consumption are multiplied by AP-42
emission factors In the case of coal-fired vessels, an average emission factor for coal
combustion in boilers was used
Off-Highway Vehicles
This source category includes farm tractors, other farm machinery, construction
equipment, industrial machinery, small general utility engines, such as lawn mowers and
snowmobiles; and motorcycles Fuel use is estimated for each subcategory from equipment
population data and an annual fuel-use factor (Hare and Springer, 1973a,b,c), together with
fiiel deliveries of diesel fuel reported by the U S Department of Energy (Monthly) and
gasoline sales reported by the U S Department of Transportation (Annual b) for off-
highway use
4.2.1.2 Stationary Source Fuel Combustion
This category includes electric utilities (including boilers) combined cycle combustion
turbines, combustion (gas) turbines, cogeneration units, and internal combustion (diesel),
industrial boilers, commercial and mstitutional boilers and furnaces, and residential furnaces
and space heaters. However, because the emission factors for stationary sources depend on
the fuel used by each source, this section is divided into the following sections coal, fuel
4-8
-------
oil, natural gas, wood, and other fuels The NOX emissions summary, presented in
Section 4 2 3, is presented both by fuel type and by stationary source
Coal
The consumption of bituminous coal, lignite,
-------
AIRS (U.S. Environmental Protection Agency, 1991b) or AP-42 (U S Environmental
Protection Agency, 1985) As with natural gas, the "thermal mechanism" is generally the
principle source of NOX emissions (Bartok and Sarofim, 1991)
4.2.1.3 Industrial Processes
Production data for industries that produce the majority of emissions were obtained
from available publications Generally, the Minerals Yearbook (U S Department of the
Interior, Annual) and the Current Industrial Reports (U S Department of Commerce,
Annual: a), published by the Bureau of Census, provided most of the necessary data
Average emission factors were applied to the various production data Average nationwide
control efficiency values for various processes were obtained either fiom published reports
(Shannon et al., 1971, Vandegnft et al, 1971a,b), the 1985 National Acid Precipitation
Assessment Program (NAPAP) emission inventory (Saeger et al, 1989), or AIRS (U S
Environmental Protection Agency, 199 Ib)
Petroleum product storage and petroleum marketing operations (including gasoline,
crude oil, and distillate fuel oil storage and transfer, gasoline bulk terrains and bulk plants,
and retail gasoline service stations) are included as industrial processes Also included are
industrial surface coating and degreasing operations, graphic arts (printing and publishing),
and dry cleaners
4.2.1.4 Solid Waste Disposal
The emissions from this category are based on an assumed solid waste generation rate
of 5.5 Ib/capita/day This value was originally based on a study of solid waste collection and
disposal practices (U S Department of Health, Education, and Welfaie, 1968) This value is
adjusted each year based on information contained in AIRS (U S Environmental Protection
Agency, 199 Ib) Average emission factors are applied to the estimated quantities of solid
waste disposal.
4.2.1.5 Miscellaneous Sources
This major source category includes forest fires, agricultural burning, coal refuse
burning and structure fires
4-10
-------
Forest Fires
The U S Forest Service of the Department of Agriculture and the U S Department of
the Interior publish information on the number of forest fires, their location, and the acreage
burned each year The amount of forest biomass burned and controlled burning of forest
areas each year are estimated per acre by Yamate (1974) Average emission factors were
applied to the estimated quantities of materials burned
Agricultural Burning
A study was conducted by the EPA to obtain local agricultural and air pollution control
agency estimates of the number of acres and quantity of material burned per acre in
agricultural burning operations (Yamate, 1974) These data have been updated and used to
estimate emissions based on average emission factors
Coal Refuse Burning
Estimates of the number of burning coal-refuse piles existing in the United States are
reported by the Bureau of Mines (McNay, 1971) This publication presents a detailed
discussion of the nature, origin, and extent of this source of pollution Rough estimates of
the quantity of emissions were made using this information by applying average emission
factors for coal combustion It should be noted that the number of coal-refuse piles had
become negligible by 1975
Structure Fires
The U S Department of Commerce (Annual b) publishes information on the number
and type of structures damaged by fires each year Emissions are estimated by applying
average emission factors for wood combustion to the.se statistics
4.2.2 Natural Sources of Nitrogen Oxides;
Nitrogen oxides can be naturally produced by lightning, biological and abiological
processes in soil, stratospheric intrusion, and chemical or photochemical processes in the
oceans There are four source categories for natural NOX emissions lightning, soils,
stratospheric injection, and oceans
4-11
-------
Lightning
The release of energy generated by lightning produces the extremely high temperatures
required to convert atmospheric nitrogen and oxygen to NO Quantitative estimates of NOX
production have been derived from different methods One method (Borucki and Chameides,
1984) uses three general factors (1) the frequency of lightning flashes, (2) energy dissipated
per flash, and (3) the NOX production per unit of energy dissipated Another method
(Albntton et al, 1984) used nitrate deposition data from remote oceanic and polar sites
where NOX from intracloud lightning is expected to be the dominant precursoi Estimates of
Ughtning-based NOX emissions for North America (Placet et al , 1991) range from 1 2 to
1.7Tg/yearof NO2
Soils
At this time, the biological or abiological pathway to NOX production of the soils is still
being debated (Placet et al, 1991) In spite of the uncertainties, four local variables emerge
as possible factors that influence NOX emissions soil temperature, soil moisture content, soil
vegetation cover, and soil nutrients (Placet et al, 1991) The NAPAP has estimated national
emissions of NOX from soils to be 1 2 Tg/year
Stratospheric Injection
Nitrous oxide (N2O) is released into the troposphere from various sources on the
Earth's surface However, because there are no significant loss processes for N2O in the
troposphere, this N2O is transported to the stratosphere, where it is photodissociated or
oxidized. A portion of the NO produced by these processes in the stratosphere subsides mto
the troposphere. Because it has been extensively modeled, the magnitude of the NOX
emissions from this source is probably more well known than from any other natural source
(Placet et al, 1991) Crutzen and Schmailz (1983) estimated global NOX emissions from
stratospheric injection to be 0 5 Tg/year For the United States, stratospheric injection is
probably the source of less than 0 1 Tg/year of NOX (Placet et al, 1991)
4-12
-------
Oceans
Production of NO from the ocean is attributed to the photolysis of nitrite dissolved in
seawater (Zafinou and McFarland, 1981) On a global basis, oceans are a small natural
source of NOX Given the lifetime of NOX in the atmosphere—a few days or less (Liu et al ,
1987)—the transport of NOX into the United States is negligible (Placet et al, 1991), with the
possible exception of some local scale transport of NOX that may be measurable
4.2.3 Global Estimates of Nitrogen Oxides Emissions
Global sources for NOX appear to be on the order of 150 Tg of NOX per year (see
Table 4-2) (Logan, 1983) Similar estimates for global sources of NOX are given in Bauer
(1982) and Stedman and Shetter (1983)
TABLE 4-2. GLOBAL BUDGET OF NITROGEN OXIDES IN
THE TROPOSPHERE*1
Source
Production
Fossil-Fuel Combustion
Biomass Burning
Release from Soils
Lightning Discharges
Ammonia Oxidation
Ocean Surface (biologic)
High-Frying Aircraft
Stratosphere
Total Production
Nitrogen Oxides
Ehhalt and Drummond
42 (26-58)
36 (18-52)
17 (3-30)
15 7 (6-25)
10 (4-15)
—
1 (0 6-1 2)
2 (0 9-2 8)
122 (60-186)
(Tg/year)b
Logan
63 (44-88)
36 (13-76)
25 (13-50)
25 (6-63)
(0-31)
3
~
15
152 (79-310)
Derived from estimates according to Ehhalt and Drummond (1982) and Logan (1983)
Mean (low range—high range)
Note Values may not sum due to independent rounding
As with the emission estimates for the United States, emissions from anthropogenic
sources (fossil fuel combustion and biomass fires associated with agriculture) appear to
exceed the emissions from natural sources (lightning and soils) For the global estimate, the
ratio of emissions from anthropogenic sources to the emissions from natural sources is on the
4-13
-------
order of 2. For the emission estimates for the United States (see Table 4-3) this ratio is
approximately 9.
TABLE 4-3. ESTIMATES OF NITROGEN OXIDE EMISSIONS FROM
ANTHROPOGENIC AND NATURAL SOURCES IN THE UNITED STATES AND
CANADA (MILLIONS OF METRIC TONS/YEAR OF NITROGEN DIOXIDE
EQUIVALENT EMISSIONS)
Emissions
Emission Source
Anthropogenic sources0
Highway vehicles
Power plants
Industrial combustion
Other
Total anthropogenic
Natural sources6
Lightning
Soil
Stratospheric injection
Total natural*1
Totald
Best Estimate
8 1
60
41
43
226
1
2
<0 1
3
26
Rangea
6 2 - 10 5
50-72
34-49
34-54
18 0 - 28 0
03-2
03-5
NAf
06-7
19-35
Percent of Total b
32
23
16
17
88
4
8
12
100
Based on uncertainties in Table 4-5 for anthropogenic source categories
Percentages are based on "best estimate" values
"includes Canada and the United States Values for the United States are the "best estimate" for 1985 shown in
rfTable 4-5
Values may not sum to totals due to independent rounding
'Values are for North America from Table 4-6
NA » Not available
Source. U S Environmental Protection Agency (1991a)
4.2.4 Analysis of United States Nitrogen Dioxide Emission Sources,
Levels, and Trends
Table 4-4 presents the total emissions of NOX in the United States from 1940 to 1990
Preliminary estimates for 1990 indicate that over 80% of the national ISTOX emissions are
emitted by highway vehicles, electric utilities, and industrial boilers Nitrogen oxides
emissions for stationary fuel sources have grown steadily from 1940 through 1990 Between
4-14
-------
TABLE 4-4. TOTAL NATIONAL EMISSIONS OF NITROGEN OXIDES,
1940 TO 1990 (teragrams/year)a
Source Category
Transportation
Highway Vehicles
Aircraft
Railroads
Vessels
Other Off-Highway Vehicles
Transportation Total
Stationary Source Fuel Combustion
Electric Utilities
Industrial
Commercial-Institutional
Residential
Fuel Combustion Total
Industrial Processes
Petroleum Refining
Chemicals
Iron and Steel Mills
Pulp Mills
Mineral Products
Industrial Processes Total
Solid Waste Disposal
Incineration
Open Burning
Solid Waste Total
Miscellaneous
Forest Fires
Other Burning
Miscellaneous Total
Total of All Sources
1940
14
00
06
01
02
23
06
23
02
03
34
01
00
00
00
01
02
00
0.1
01
07
02
09
69
1950
22
00
09
0 1
04
36
12
29
03
03
47
01
00
0 1
00
01
03
01
0 1
02
04
02
06
94
1960
38
00
07
0 1
05
5 1
23
37
03
04
67
02
01
0 1
00
01
05
01
02
03
02
02
04
130
1970
63
0 1
06
0.1
08
80
44
39
09
04
9 1
02
02
01
00
02
07
01
03
04
02
01
03
185
1980
79
01
08
02
10
98
64
3 1
03
04
101
02
02
01
00
02
07
00
0 1
0 1
02
00
02
209
1990
56
0 1
05
02
1 1
75
73
33
02
04
11 2
02
01
00
00
02
06
00
01
01
02
00
03
196
aU S Environmental Protection Agency (1991a)
4-15
-------
the years 1940 and 1980, NOX emissions from transportation sources nearly quintupled, but
decreased by 30% from 1980 to 1990
In the recent past, the U S transportation system has been responsible for 40 to 50% of
the annual national NOX emissions, as indicated in Table 4-5 (U S Environmental Protection
Agency, 1991a) From 1982 through 1989, that percentage contribution has been decreasing
at an average rate of 2 5 %/year
TABLE 4-5. TRANSPORTATION CONTRIBUTION TO
U.S. NITROGEN OXIDES EMISSIONS
Year Millions of Metric Tons Percent of U S Total
1978 9~8464
1979 10 1 46 8
1980 98 46 9
1981 10 0 47 8
1982 94 47 0
1983 89 46 1
1984 88 44 4
1985 89 44 7
1986 83 43 5
1987 81 41 8
1988 81 40 5
1989 79 3_9_7
Source U S Environmental Protection Agency (1991a)
In 1988, the highway sector produced 76 % of the NOX from the transportation system,
as shown in Table 4-6 (U S Environmental Protection Agency, 1990) This amounted to
31 % of the national NOX emissions for that year Among the contributors to the balance
from the transportation system were air, marine, and rail vehicles As seen from Table 4-6,
in 1988, 62% of the NOX from highway vehicles came from gasoline engines and 38% came
from diesel engines
Table 4-7 shows the emissions of NOX from stationary fuel combustion sources
categorized by fuel type For the 1990 data, the combustion of coal accounted for 63 % of
the NOX emissions from stationary sources The majority of the emissions came from coal
4-16
-------
TABLE 4-6. BREAKDOWN OF 1988 TRANSPORTATION NITROGEN OXIDES
Typea
Gasoline LDV
Gasoline HDV
Motorcycle
Diesel LDV
Diesel HDV
Nonhighway
Total
Millions of Metric Tons
355
025
001
003
230
1 95
809
Percent of Highway Total
578
4 1
02
05
375
____
aLDV = Light-duty vehicle
HDV = Heavy-duty vehicle
Source U S Environmental Protection Agency (1990)
combustion in electee utilities Coal combustion accounted for 88 % of the NOX emissions
for electric utilities, while accounting for 61 % of the fossil-fuel electnc utility-generating
capacity (U S Department of Energy, 1991) Natural gas accounts for 29% of the NOX
emissions from stationary sources In the industrial boiler category, natural gas accounts for
nearly 72% of the NOX emissions Nitrogen oxides related to the combustion of coal
increased 82% from 1970 to 1990 During the same tune, NOX from fuel oil and natural gas
decreased 45 and 20%, respectively
Table 4-8 presents U S NOX emissions from 1940 to 1990 and current estimates of
future NOX emissions from highway vehicles, industrial sources, electee utilities, and all
other sources These emission trends are shown in Figure 4-2 The projections account for
the expected net effect of all provisions of the Clean Air Act (CAA) as amended in 1990
These include the NOX emission limits prescribed for utility boilers under the acid rain
provisions, the Tier I automobile tailpipe standards, and application of technology-based
requirements to nonutility boilers (generally greater than 100 tons/year) in ozone (O3)
nonattainment areas and the Northeast Ozone Transport Region The estimates do not fully
incorporate New Source Review requirements, such as offsets and lowest achievable emission
rates in nonattainment areas, nor do they incorporate additional controls required based on
attainment demonstration modeling They also do not attempt to estimate the extent
4-17
-------
TABLE 4-7. EMISSIONS OF NITROGEN OXIDES FROM STATIONARY FUEL
COMBUSTION SOURCES, 1970 TO 1990 (teragrams/yeair)3
Source
Coal
Electric Utilities
Industrial
Commercial-Institutional
Residential
Coal Total
Fuel Oil
Electric Utilities
Industrial
Commercial-Institutional
Residential
Fuel Oil Total
Natural Gas
Electric Utilities
Industrial
Commercial-Institutional
Residential
Natural Gas Total
Wood
Industrial
Residential
Wood Total
Other Fuels
Industrial
Residential
Other Fuels Total
Fuel Combustion Total
1970
3 17
070
002
002
391
039
03
019
Oil
099
088
277
0 11
022
398
009
004
013
005
006
Oil
9 12
1980
5 15
040
002
<001
557
044
022
014
008
085
078
224
0 12
022
336
0 12
008
020
007
003
010
1008
1990
642
067
003
<001
712
027
010
008
009
054
059
237
0 12
020
328
015
007
022
003
003
006
11 22
Note Values may not sum due to independent rounding
aU S. Environmental Protection Agency (199 la)
4-18
-------
TABLE 4-8. TOTAL NATIONAL NITROGEN OXIDE EMISSIONS,
1940 TO 2010 (teragrams/yearf
Source
Highway Vehicles
Electric Utilities
Industrial Sourcesb
Other
Total
1940
14
06
25
24
69
1950
22
12
32
28
94
1960
38
23
42
27
130
1970
63
44
46
32
185
1980
79
64
38
28
209
1990
56
73
39
28
196
2000
30
61
35
30
156
2010
29
74
41
3.2
176
aU S Environmental Protection Agency (1991a)
Includes industrial fuel combustion and processes
30
25-
20 —
Source Category:
m Transportation
W Fuel Combustion
CH Industrial Processes
• Solid Waste & Misc
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
Figure 4-2. National trend in U.S. nitrogen oxides emissions, 1981 to 1990 (106 metric
tons per year).
4-19
-------
to which any areas might be exempted from NOX stationary source controls under CAA
Section 182(f).
Projections of NOX emissions from highway vehicles are based on projected vehicle
miles traveled and MOBILE4 1 emission factors These emission factors reflect current
emission control standards and Tier I motor vehicle emission standards of the CAA (Tier n
standards are not reflected because these are discretionary ) As a result of these standards,
NOX emissions from highway vehicles are expected to decrease by almost 50% from 1990 to
2000.
Federal Tier I standards for 1994 and California's series of low-emission vehicle (LEV)
standards call for further reductions in NOX, regardless of the fuel used Specifically, these
standards are 0.4 g/mi NOX for Tier I and California's transitional low-emission vehicle, and
0.2/mi for the LEV and ultra-low-emission vehicle
An indication of the effectiveness of NOX regulation on the NOX per highway vehicle is
provided in Figure 4-3. Nitrogen oxide standards for both LDVs and diesel HDVs are
traced by the stairsteps The estimate plotted for the average highway vehicle was made in
the following way. Available NOX data from the entire transportation sector were multiplied
by 0 778, the average ratio of highway-vehicle energy consumption to total transportation
energy consumption over the 12-year period illustrated (U S Environmental Protection
Agency, 1991a). This product was then used to calculate the ratio to the total annual miles
traveled by all highway vehicles The resulting estimate, which involves an assumption that
NOX mass produced is proportional to energy consumption for cars, diesel tracks,
locomotives, jet liners, and pipelines alike is obviously not quantitatively correct, but the
trend is believed to be valid This estimated average is seen to track the trends in NOX
emission standards remarkably well
Future reductions in NOX standards, therefore, suggest a likely continued downtrend in
NOX grams per mile for highway vehicles The total NOX contributed will depend, of
course, on the product of grams per mile of NOX per vehicle and the total annual miles
driven. Table 4-5 suggests that since 1979, the decrease in NOX per vehicle-mile has more
than compensated for any increase in total miles driven
Additional measures being scheduled to decrease the average grams NOX per mile from
passenger cars are (1) advanced on-board diagnostics systems to signal the driver of engine
4-20
-------
IU
= 8
% 6
CO
"E 4
LU
CO
1 2
Z 1
0.8
0.6
0.4
0.2
-
-
I
3.1
l
1975
i
o J^x
2.0
I
i i
^XX^Wj,
^S^/^//?7/
a
i
!_!Q..7_g/bh£-h Diesel HOT std.
Gasoline LDV std. j 1 .0 g/mi
i
i i
5
i 4
0.4
I
1980 1985 1990 1995 2000
Year
Figure 4-3. Emission of nitrogen oxides compared with nitrogen oxide standards.
Source- U S. Environmental Protection Agency (1990)
and emissions-control system malfunctions, and (2) expanded and improved inspection and
maintenance procedures These steps will augment the impact of emissions standards Most
cars on the road today have control systems required to meet emission standards at
50,000 mi As of 1990, approximately 60% of these cars had been driven more than
50,000 mi, and about a quarter of registered cars were built before the NOX reducing catalyst
and electronic control had seen widespread application (U.S Environmental Protection
Agency, 1991a)
By 2000, all electric utility units with capacities {greater than 25 MW are expected to
meet new emission limits imposed by the CAA Also, new or modified electric power units
will be subject to revised performance standards As a result, NOX emissions from electric
utilities are expected to decrease by 16% in the next 10 years The analysis for utilities was
performed under the assumption that low-NOx combustion technology would be employed to
4-21
-------
meet the NOX provisions of Title IV The 6 1 Tg estimate for electric utilities in 2000 is
approximately 1.8 Tg (2 million short tons) less than what would have been emitted by
utilities without controls implemented as a result of the CAA Amendments of 1990
Estimates of future NOX emissions from industrial sources are based on state-level
growth factors and the expected application of reasonable available control technology where
required. As a result, a 10% reduction is expected in NOX emissions fiom industrial sources
from 1990 to 2000 This reduction may be more than offset by increases in emissions
between 2000 and 2010 The future trend of stationary source NOX emissions is presently
uncertain because it is not known whether O3 nonattainment areas will be exempt from the
proposed New Source Review policy that requires lowest achievable emission reductions and
offsets for new major sources
Table 4-9 presents the estimated emissions of NOX from anthropogenic and natural
sources for 1990 As shown, anthropogenic sources account for nearly 90 % of the NOX
emissions
TABLE 4-9. COMPARISON OF ANTHROPOGENIC AND NATURAL SOURCES OF
NITROGEN OXIDES EMISSIONS FOR 1990a
Sources
Anthropogenic
Transportation
Stationary
Industrial Processes
Solid Waste
Misc.
Natural
Lightning
Soils
Stratosphere
Oceans
Total
Emissions (Tg)
75
112
06
01
JQ_3
197
12
1 2
<01
.
-------
4.2.5 Comparison of Nitrogen Oxide Emissions Estimates
Table 4-10 compares several sources of NOX emission estimates These include the
NOX emission estimates of the 1985 NAPAP Emissions Inventory (Version 2) (Saeger et al ,
1989), the EPA long-term trends estimates (U S Environmental Protection Agency, 1990),
NAPAP month and state current emission trends (MSCET) estimates (Kohout et al, 1990),
and the 1982 Electric Power Research Institute (EPRI) Inventory (Heisler et al, 1988) For
1985 data, total emissions from the NAPAP, EPA, and MSCET inventories differ by less
than 9% The EPRI inventory estimate of 20 7 Tg is higher than the 1982 EPA and MSCET
estimates, which are 19 6 and 18 8 Tg, respectively
TABLE 4-10. COMPARISON OF ANNUAL U.S. NITROGEN OXIDE
EMISSIONS ESTIMATES FROM FOUR INVENTORIES8
Emissions (Tg/year)
Source Category
Electric Utilities
Nonutility Combustion
Transportation
Other Sources
Total
NAPAP Inventory15 EPA Trends0 MSCEl4 EPRI6
60
35
80
1 0
186
68
34
89
07
198
62
36
76
_Q_8
182
72
44
79
12
207
aNote Values may not sum to totals due to independent rounding
NAPAP = National Acid Precipitation Assessment Program
EPA = U S Environmental Protection Agency
MSCET = Month and state current emission trends
EPRI = Electric Power Research Institute
bSaegeretal (1989)
°U S Environmental Protection Agency (1990)
dKohoutetal (1990)
"Heisleretal (1988)
4.3 INDOOR EMISSION SOURCES OF NITROGEN OXIDES
4.3.1 Introduction
This section summarizes emissions of NOX from combustion sources commonly found
in the indoor residential environment that affect indoor air quality There are several reasons
for considering these emissions First, such information is needed to understand the
4-23
-------
fundamental physical and chemical processes influencing emissions This undei standing can
be used to help develop strategies for reducing emissions Second, examining emissions
from several types of sources and source categories can help identify the relative importance
of each source in affecting indoor air quality This information can guide decisions by house
occupants on combustion appliance purchases and methods of using such appliances Finally,
studying emissions from indoor sources can provide source strength input data needed for
indoor air quality modeling Predicting indoor airborne concentrations is important for
estimating the total exposure of individuals to NOX
This section begins with a brief discussion of the physics and chemistry of nitrogen
oxides formation m flames Several major source categories are then considered These
include gas stoves used for cooking, unvented gas space heaters, unvented kerosene heaters,
wood stoves primarily used for heating, and tobacco products For each category, several
studies will be discussed, including the measurement methods used and the resulting data
Following these presentations, the emissions of NOX for all categories will be compared
Note that several types of vented appliances commonly found indoors usually emit NOX
to the outdoors. Examples include furnaces, water heaters, and clothes dryers using gas, as
well as stoves and furnaces using wood, coal, and other fuels Under some circumstances,
these sources may contribute to elevated NOX levels indoors, for example, Hollowell et al
(1977) reported high NO and NC>2 concentrations in a house where a vented forced-air gas-
fired heating system was used Elevated concentrations may also be a problem with
malfunctioning vented appliances Other data (e g , Fortmann et al , 1984), however,
suggested that fugitive emissions of NOX from vented appliances are small Because the
importance of unvented appliances to indoor NOX levels is well-documented, this chapter
focuses on emissions from such appliances
4.3.2 Formation of Nitrogen Oxides in Combustion in Gas-Fueled
Household Appliances
Many household appliances incorporate laminar flames in which the input fuel is
premixed with a known amount of air before being subjected to combustion As an example,
Figure 4-4 shows a diagram of a single-port flame fueled with natural gas, taken from a
review by the Institute of Gas Technology (Zawacki et al, 1986) The figure is somewhat
4-24
-------
CD
00
1__
CD
C
CO
CD
s
.Q
<
I
CO
b
16
14
12
10
8
4
2
0
Outer
Cone
Secondary
Air
Fuel Flow
4scf/h
Primary
Air 70%
Excess
Air 80%
Radial
Profile
Post-Combustion
Reaction Zone
Combustion Zone
(Bright blue)
Secondary
Air
1 0 1
Burner Base (cm)
Figure 4-4. Laminar blue-flame. Reproduced from Zawacki et al. (1986). Permission
to be obtained.
4-25
-------
simplified in that most appliances include burners with several flame ports, the combustion
product distributions are affected by interference from adjacent flames Nevertheless, the
single flame model has been used successfully to approximate the behavior of more complex
systems.
The primary mixture consists of gas and 40 to 70% of the air required for complete
combustion. Excess air needed for complete combustion is provided through an annular
region surrounding the central port The combustion takes place in a thin layer surrounding
the inner cone, termed the combustion zone A number of intermediate products are formed
in this zone, including molecular hydrogen, CO, and radical species such as hydroxyl,
atomic oxygen, and atomic nitrogen Significant amounts of the major products, carbon
dioxide (CO^ and water, are also formed here
As the combustion products flow outward, other reactions occur in a luminous
secondary reaction zone, termed the postcombustion reaction zone The maximum
temperatures in the flame occur at the boundary of the outer cone defined by these reactions
Beyond this boundary, the combustion products mix with secondary air, resulting in a rapid
temperature decrease
Many investigators have considered NO and NO2 production in this type of combustion
process, as reviewed by Zawacki et al (1986) The NO concentration peaks at the outer
cone boundary, where production is assisted by high temperatures The low molecular
oxygen content there prevents production of NO2 The latter species is produced at a slightly
greater radial distance from the flame center, where oxygen provided by secondary air reacts
with the high NO concentration Overall, the production of NO is highest in regions of
maximum temperature Thus, NO production may be particularly high at "hot spots" in the
flame (Hayhurst and Vince, 1980)
It is of interest to consider the chemical reactions involved in the formation of NO and
NO2 in this system At high temperatures (> 1,600 K), NO is produced primarily by the
reactions first proposed by Zeldovich (1946)
O + N2 -> NO + N
N + O2 -» NO + O
4-26
-------
Coutant et al (1982) observed that NO also appears to form around the base of the flame
where the temperatures are too low for the Zeldovich mechanism to occur (about 1,350 K)
They hypothesize reactions of the form
CN, NH + O2 -* NO + products,
which may be important at high as well as low temperatures in the flame These authors use
the work of Fenimore (1971) to further hypothesize a number of specific reactions to produce
CNandNH
CH4 + O -> CH3 + OH
CH3 + OH -» CH2 f H2O
CH3 + N2 -> HCN f NH2
CH2 + N2 -* HCN f NH
After NO is produced from the reaction of CN and NH with O2 as shown above, oxidation
to NO2 occurs via the reaction.
NO + HO2 -> NO2 + OH
With sufficient oxygen present, the HO2 would come from the reaction of oxygen with
methane
CH4 + O2 -> HO2 •+ CH3,
or from reactions with formaldehyde or CHO as proposed by Peeters and Mahnen (1973)
CH2O, CHO + O2 -* HO2 + CHO, CO
Knowledge of these reactions has been used to design burners that may have smaller
NOX emissions For example, reducing the flame temperature decreases the production of
4-27
-------
NO, use of infrared tiles to lower the temperature has been shown to be an effective strategy
in decreasing NO emissions (Zawacki et al , 1986) The emissions of CO, however, tend to
increase under these conditions because of reduced oxidation of CO to CO2 Fuel type may
affect temperature and hence NOX emissions propane combustion is much hotter than that
of natural gas, resulting hi greater NO emissions Other factors such as the level of primary
aeration, use of recirculation of combustion products, and control of air currents near the
burner have been shown to affect NOX emissions, the reader is referred to Zawacki et al
(1986) for more detail
It is important to note that our understanding of NOX emissions even from simple
combustion systems is far from complete Reuther et al (1988) summarize data from
12 investigations of premixed stoichiometac air/methane combustion, and conclude that wide
variations in reported emissions are probably due to differences in measurement protocol
Reuther et al recommend that standardized measurement techniques be established for
further investigation of NOX emissions
Besides studies of simple well-controlled combustion, a number of investigators have
measured NOX emissions from common household appliances in the laboratory and in the
field. Data resulting from these measurements are briefly reviewed below.
4.3.3 Gas Stoves Used for Cooking
Several research programs have investigated NOX emissions from stoves fueled with
natural gas and propane Most of these studies have included a number of other pollutants as
well, such as CO, aldehydes, and unburned HCs This summary only addresses the NOX
emission data. Furthermore, only studies using fuels of composition commonly found in the
United States are included For additional information on emissions from other types of gas,
the reader is referred to Yamanaka et al (1979) and Caceres et al (1983)
One of the earliest studies of gas stove emissions was conducted by Himmel and
DeWerth (1974) at the American Gas Association Laboratories A total of 18 commercially
available residential stoves were examined The authors estimated that the population of
stoves tested was representative of at least 90% of the total population of gas stoves in use
within the United States at that time
4-28
-------
The emissions were sampled using a 28-cm diameter, 25-cm high quartz hood The
combustion exhaust gases were drawn through the top of the hood, passed through a cold
trap to reduce moisture content, and directed into pollutant analyzers These investigators
used the standard American National Standard Institute (ANSI) pot filled with water as a
load The ANSI pot is chrome-plated brass, having dimensions 19 cm in diameter by
15 cm in height The quartz hood was positioned over the burner and pot such that the CO2
concentration in the emissions was approximately 2% This value was selected to provide a
reasonably concentrated sample that did not have an excessive correction factor when
calculating the air-free pollutant concentration
Air-Free Pollutant Concentration = Ultimate CO2 x Sample Pollutant Concentration, (4-1)
Sample CO2
where the air-free and sample pollutant concentrations are given on a dry-weight basis
Ultimate CO2 refers to the concentration expected based! on stoichiometnc considerations
The combustion process was allowed to reach steady-state before recording data
This procedure was applied to each of four top burners on 16 stoves, and to three top
burners on each of 2 additional stoves Oven and broiler burners were also sampled, as were
the pilot lights for all types of burners Tests were conducted using well-adjusted blue
flames, and also poorly adjusted yellow flames In limited additional testing, emissions from
the burners were sampled with water-filled cooking pots made of different materials quartz,
pyrex, aluminum, copper, steel, iron, and Pyroceram
An overall summary of the data for burner operation is presented in Table 4-11 The
top burners were operated at maximum heat input rate, which was close to the rated
158 kJ/min (9,000 BTU/h) for some of the stoves and 211 kJ/min (12,000 BTU/h) for
others The oven and broiler burners were operated at heat input rates ranging from
193 kJ/min (11,000 BTU/h) to 422 kJ/min (24,000 BTU/h) The data show that emissions
of NO are generally in the range of 16 to 24 /tg/kJ (0 037 to 0 056 lb/106 BTU), whereas
emissions of NO2 are in the range of 5 to 14 /«g/kJ (0 012 to 0 033 lb/106 BTU)
Exceptions include an infrared burner with a very low NO emission factor and a pyrolytic
self-cleaning oven with a relatively high NO emission factor Table 4-11 also shows
emission factors as a function of heat input rate for a top burner on one of the stoves
4-29
-------
TABLE 4-11. EMISSION FACTORS FOR NITRIC OXIDE
AND NITROGEN DIOXIDE FROM BURNERS ON GAS STOVES,
AFTER HIMMEL AND DEWERTH (1974)
Top Burners
Top Burners
Ovens and Broilers
Ovens and Broilers
Top Burners with
Thermostat
Top Burners, 142 kJ/min
Top Burners, 190 kJ/min
Infrared Burners
Ovens and Broilers with
Catalytic Clean
Ovens and Broilers with
Catalytic Clean
Pyrolytic Self Clean
Oven
Top Burner of One
Stove, 150 kJ/min
Top Burner of One
Stove, 79 kJ/min
Top Burner of One
Stove, 39 kJ/min
Number
of
Burners
70
70
27
27
6
35
35
2
8
8
1
1
1
1
Flame
Type
Blue
Yellow
Blue
Yellow
Blue
Blue
Blue
Blue
Blue
Yellow
750 °Fa
Yellow
Yellow
VTellow
Emission Factor
for Nitric Oxide
O^g/kJ)
20
16
22
16
20
20
20
3
24
18
38
22
9
1
4
0
2
6
7
5
6
5
0
5
1
5
+ 44
±50
±55
±78
±30
± 40
± 44
±71
±108
42
45
Emission Factor
for Nitrogen
Dioxide
0*g/kJ)
8
13
5
11
10
7
8
5
5
7
14
10
14
15
4
5
5
5
5
5
6
2
1
44
4
2
0
3
±
±
±
±
±
±
±
±
±
2
5
1
8
3
1
2
2
2
22
4
91
1
2
30
35
74
48
"Temperature setting as given in the original reference
As the heat input rate increases, the data show that the emission factoi for NO increases, but
the emission factor for NO2 decreases
Emission factors for pilot lights associated with the top burners and with the ovens and
broilers are shown in Table 4-12 Three types of top burner pilot lights and two types of
oven and broiler pilot lights were tested The data show that emission factors for NO and
NO2 average roughly 12 9 and 7 3 /*g/kJ (0 030 and 0 016 lb/106 BTU), respectively, for
4-30
-------
TABLE 4-12. EMISSION FACTORS FOR NITRIC OXIDE AND NITROGEN
DIOXIDE FROM PILOT LIGHTS ON GAS STOVES,
AFTER HEMMEL AND DEWERTH (1974)
Top Burners
Top Burners
Top Burners
Ovens/Broilers
Ovens/Broilers
Pilot Typea
I
n
m
i
n
Emission Factor
for Nitric Oxide
fctg/kJ)
13.7
132
119
0265
101
Emission Factor
for Nitrogen Dioxide
G«g/kJ)
783
683
735
144
107
aTypes of pilot lights on top burners
I A free standing single flame that is surrounded by a flashtube assembly
II Same as above with a shield or baffle around the flame The baffle typically comes up to about one-half of
the flame height, protects the flame from drafts and tends to channel the combustion air supply
Combustion air supply openings may be in the baffle, or insi.de or outside the baffle on the range top
HI Same as n with the addition of a shield above the flame This shield can be a flat baffle or an arch-type
baffle The baffle appears to be for the purpose of keeping fhe range top cool Type TTT seems to be the
most popular of the three pilot types
Types of pilot lights on ovens and broilers
I A constant input pilot which typically is a horizontally oriented flame positioned directly below a flame
sensing element
n Similar to I, but operated in two fuel input modes (1) a standby pilot mode and (2) ignition mode, where
a secondary fuel supply ignites the burner The standby pilot mode is not directed onto a flame sensing
element, whereas the ignition flame is Generally this type of pilot is primary aerated
the top burner pilots For the oven and broiler pilots, the NO emission factor is very small,
the original data show high variability, with individual measurements ranging from near zero
to 2 2 /ig/kJ (0 005 lb/106 BTU) The NO2 emission factor is not as variable
Himmel and DeWerth (1974) also conducted statistical tests with the emission factor
data to determine the influence of various parameters on the total NOX emissions. Results
showed that the heat input rate and the number of burners in operation each had a significant
effect on emissions at the 99% confidence level Cooking utensil material had a significant
effect at the 95 %, but not at the 99 %, level The supporting grate material, grate height,
4-31
-------
and whether or not the oven was in operation had no significant effect on the top burner NOX
emissions There was no significant difference in NO2 emitted from stove to stove, although
differences in total NOX were observed The front burners had 13 % higher emission factors
for NO2 than the rear, attributed to differences in entertainment of secondary air to feed the
flame. The authors also discussed experimental burners that may be effective in reducing
NOX emissions from top burners They concluded that controlling the ingress of secondary
air to the flame could reduce NOX emissions by about 50% Reducing flame temperature
with a screen or other heat absorber in the flame could reduce NOX emissions by as much as
68%, whereas operating a burner at 100% plus primary aeration could reduce NOX emissions
by as much as 66%.
Cole et al. (1983) and Moschandreas et al (1985) conducted a variety of emissions
tests with top burners, ovens, broilers, and pilot lights The tests for the top burners were
conducted using two methods The first method was similar to that of Himmel and DeWerth
(1974) involving a quartz hood placed directly over the burner The second method involved
o -a
placing a stove in a 33-m (1,150-ft) all-aluminum chamber with controlled air exchange
(1 to 5 air changes per hour) and temperature characteristics, emissions were determined by
measuring the ambient concentrations in the chamber and using a one-compartment mass
balance model to calculate the source strength, as in the experiments of Cote et al (1974)
Four top burners on each of three ranges were tested using the quartz hood Both lean
gas with a heating content of 36,600 kJ/m3 (983 BTU/ft3) and rich gas with a heating content
of 38,100 kJ/m3 (1,022 BTU/ft3) were used Some of the tests were conducted with a
properly adjusted blue flame, and others used a poorly adjusted yellow flame The tests
were conducted at maximum heat inputs, which were in the range of 140 to 175 kJ/min
(8,000 to 10,000 BTTJ/h).
The results of these tests are shown in Table 4-13 The data show average values for
both lean and rich gas Note that both the NO and the NO2 emission factors agree well with
those of the Himmel and Dewerth study obtained using similar sampling procedures All of
these values refer to steady-state conditions A limited amount of data were also reported
showing variations in emission factor with tune before reaching steady-state, results showed
that emissions of NO increase during the approach to steady-state, whereas NO2 emissions
decrease until a steady condition is achieved
4-32
-------
TABLE 4-13. EMISSION FACTORS FOR NITRIC OXIDE, NITROGEN DIOXIDE,
AND NITROGEN OXIDES FOR TOP BURNERS ON GAS STOVES MEASURED
WITH A SAMPLING HOOD AND WITH A CHAMBER, AFTER COLE ET AL.
(1983) AND MOSCHANDREAS ET AL. (1985)a
Sampling Hood
Sampling Hood
Chamber, Mass Balance
Chamber, Mass Balance
Top Burner of One
Stove, 145 kJ/min
Top Burner of One
Stove, 121 kJ/min
Top Burner of One
Stove, 24 kJ/min
Top Burner of One
Stove, 13 kJ/min
Flame
Type
Blue
Yellow
Blue
Yellow
Blue
Blue
Blue
Blue
Emission
Factor
for Nitric
Oxide
Otg/kJ)
18 ± 1 1
16 ± 07
16 ± 1 1
11 + 1 8
15 ± 04
13 ±04
20 + 09
0 86 + 0.4
Emission
Factor
for Nitrogen
Dioxide
Otg/kJ)
99 + 11
14 + 1 1
10 + 20
16 ± 35
73 + 04
86 + 04
86 ± 04
14 ± 2 6
Emission
Factor
for Nitrogen
Oxides
0*g/kJ)
36 ± 20
39 ± 10
35 + 26
33 + 30
29 + 09
29 + 04
23 ± 1 3
16 ± 2 6
aThe first four rows refer to tests conducted at maximum heat input rate, values of average and standard
deviation in these tests are computed based on three values, each representing the average of 9 to
24 measurements on a single stove Values given for the chamber mass balance tests are for sampling 12 to
29 nun after turning on the stove
The chamber tests involved operating one burner on each of the same three stoves,
using blue and yellow flames Results are also shown in Table 4-13 The data from these
tests agree reasonably well with the results of direct sampling using the quartz hood,
suggesting the viability of either method
Moschandreas et al (1985) also examined the influence of heat input rate on emissions
Measurements were made with one stove operated in the chamber at four heat input rates
The results, shown in Table 4-13, are in qualitative agreement with Himmel and Dewerth
increasing the heat input rate increases NO, but decreases NO2 emission factors Additional
4-33
-------
tests conducted by Moschandreas et al (1985) showed that NO and NO2 emissions decreased
with increasing relative humidity in their chamber
As with the burner emissions, NO and NO2 emissions from ovens were measured using
two different techniques The first method involved use of a sampling probe positioned at
the oven flue outlet on the back of the range The second method involved placing the stove
in the chamber, as before Average heat input rates ranged from 90 to 340 kJ/min (5,100 to
19,500 BTU/h). Table 4-14 shows the results of the oven emissions tests using the first
method. The original data show that results for the second method of testing agree well with
those in Table 4-14 and therefore are not shown
TABLE 4-14. EMISSION FACTORS FOR NITRIC OXIDE,
NITROGEN DIOXIDE, AND NITROGEN OXIDES FOR OVENS,
AFTER COLE ET AL. (1983) AND MOSCHANDREAS ET AL. (1985)a
Range 1, Bake
Range 1, Broil
Range 2, Broil
Range 2, Broil
Range 3, Bake
Range 3, Broil
Range 3, Broil
Range 3, Self-clean
Type of
Gas
Rich
Lean
Lean
Rich
Lean
Lean
Rich
Rich
Emission
Factor
for Nitnc Oxide
(jug/kJ)
14 ±32
29 ±03
22 ± 0 1
22 ±03
14 ±0 1
23 ±03
23 ± 02
27 ± 10
Emission
Factor
for Nitrogen Dioxide
G*g/kJ)
11 + 15
7 1 ±08
97 ± 03
99 ±0 1
13 ±07
12 ±0 1
11 ± 04
12 ±02
Emission
Factor
for Nitrogen Oxides
G*g/kJ)
41 ± 54
53 ±09
44 ± 03
41 ±05
35 ± 15
47 ±05
46 ±06
54 ± 1 8
aBake test. Thermostat at 500 °F, burner cycles normally
Broil test' Thermostat at broil, burner on continuously
Self-clean test Thermostat at clean, programmed burner sequential operation
Finally, Moschandreas et al (1985) determined emissions from three pilot lights in one
of the ranges using the chamber method Results are shown in Table 4-15 The two top
pilots had a combined heat input rate of 4 4 kJ/min (250 BTU/h), compared with the single
bottom pilot heat input rate of 15 kJ/min (850 BTU/h) Despite the lower heat input rate of
the top pilots, both NO and NO2 emissions were substantially greater from the top pilots
Overall, the emissions in Table 4-15 are similar to those of the pilot lights tested by Himmel
and DeWerth (1974).
4-34
-------
TABLE 4-15. EMISSION FACTORS FOR NITRIC OXIDE
AND NITROGEN DIOXIDE FROM PILOT LIGHTS
ON GAS STOVES, AFTER MOSCHANDREAS ET AL. (1985)
All three pilots
All three pilots
All three pilots
Two top pilots
Bottom pilot
Air Exchange
Ratea
10
25
50
1 0
10
Emission Factor
for Nitric Oxide
0*g/U)
73 ±09
73 ±13
90 ± 26
17 ±17
39 ±13
Emission Factor
for Nitrogen Dioxide
G*g/kJ)
86 + 13
90 + 09
12 + 17
11 ± 30
82 ± 1 3
Air exchange rates refer to chamber operating conditions
Tests conducted by Fortmann et al (1984) involved two top burners on each of two
stoves, operated with a blue flame at maximum heat input rate A sheet metal sampling hood
covering all four burners was used, following the procedures of Yamanaka et al (1979)
A water-filled stainless steel pot (22-cm diameter by 17-cm high) was used as a load. These
tests gave average steady-state emission factors of 17 ^rg/kJ (0 040 lb/10 BTU) for NO and
12 /xg/kJ (0 028 lb/106 BTU) for NO2 Later tests by the same research group involved one
of these stoves sampled with a Teflon-coated hood (Borrazzo et al, 1987) Experiments
were performed during initial start-up and at steady-state Results for steady-state operation
showed average emission factors of 17 j&g/kJ (0 039 lb/106 BTU) for NO and 12 jwg/kJ
0 028 lb/106 BTU) for NO2 For initial start-up, emissions of NO increased and those of
NO2 decreased until steady-state was achieved, in qualitative agreement with the data of Cole
et al (1983) Both Fortmann et al (1984) and Borraz/o et al (1987) also sampled at several
heat input rates Results showed that although there is considerable variability, NO
emissions generally increase and NO2 emissions decrease with increasing heat input rate
Cole and Zawacki (1985) prepared a literature survey of emissions from gas-fired appliances,
including gas stoves Within the summary, they report emissions of NOX from two Gas
Research Institute studies of advanced design burners The American Gas Association design
involves stainless steel inserts applied to conventional burners The Shukla and Hurley
(1983) design incorporates a new infrared jet burner Preliminary results are presented in
Table 4-16 Reductions are seen in the unproved burners
4-35
-------
TABLE 4-16. EMISSION FACTORS FOR NITRIC OXIDE
AND NITROGEN DIOXIDE FROM RANGE-TOP BURNERS
OF IMPROVED DESIGN, AFTER COLE AND ZAWACKI (1985)
American Gas Association
Standard Burner
American Gas Association
2-Ring Insert
American Gas Association
3-Ring Insert
Shukla and Hurley Infrared
ShuMa and Hurley Infrared
ShuWa and Hurley Infrared
Heat Input
Rate
(kJ/min)
158
158
158
118
72 1
457
Emission Factor Emission Factor
for Nitric Oxide for Nitrogen Dioxide
(/ig/mkJ) (ng/mkJ)
202
904
861
26
22
1.7
146
125
108
13
1 3
086
Measurements of emissions from ten gas stoves currently in use in residences were
obtained by Tikalsky et al (1987) The emissions were measured using the hood method of
Himmel and DeWerth (1974) Data from each stove were obtained independently by two
research groups in order to obtain comparative data The results are shown in Table 4-17
Overall, the emission factors are similar to those reported in the literature from other
studies. Although the overall mean values reported by the two groups are in agreement,
results of the individual tests for each burner showed significant differences between the two
groups. The original data also showed greater variability in the results of these field tests
compared with results of laboratory tests reported in the literature The emissions did not
appear to vary with gas flow rate for the conditions of this study.
Summary of Emissions from Gas Stoves
Table 4-18 lists average emission factors for range-top burners and for oven and broiler
burners operated at maximum heat input rate Data are shown for both well-adjusted blue
flames and for poorly adjusted yellow flames Each of the averages is based on the total
number of stoves tested for that category using data from the above studies For top burners
with blue flames, a total of 27 values are represented, for yellow flames, there are a total of
4-36
-------
TABLE 4-17. EMISSION FACTORS FOR NITROGEN DIOXIDE FROM
10 GAS STOVES IN USE IN RESIDENCES, MEASURED INDEPENDENTLY BY
RESEARCH GROUPS (TTKALSKY ET AL., 1987)
(Values shown are arithmetic averages and standard deviations.)
Emission Factors Emission Factors
Measured by Group 1 Measured by Group 2
Right Front Burner
Right Front Burner
Right Front Burner
Left Rear Burner
Left Rear Burner
Left Rear Burner
Oven, Bake
Oven, Broil
Gas Flow Rate
High
Medium
Low
High
Medium
Low
G«g/kJ)
15 ± 3
15 + 5
14 ± 3
16 + 3
16 + 3
18 + 4
13 ±9
19 ± 14
(jKg/kJ)
14 ± 6
15 ± 8
26 ± 35a
15 ± 6
15 + 6
17 + 6
12 + 6
16 ± 7
aAverage influenced significantly by one extreme value
TABLE 4-18. AVERAGE EMISSION FACTORS FOR NITRIC OXIDE, NITROGEN
DIOXIDE, AND NITROGEN OXIDES FROM BURNERS ON GAS STOVES
BASED ON DATA REPORTED IN THE LITERATURE
Top Burners
Top Burners
Ovens and Broilers
Ovens and Broilers
Flame Type
Blue
Yellow
Blue
Yellow
Factor for Factor for Factor for
Nitric Oxide Nitrogen Dioxide Nitrogen Oxides
Otg/kJ) 0*g/kJ) (/*g/kJ)
20 0 ± 4 5
16 9 + 4 5
21 9 + 6 3
19 8 ± 9 6
10 2 +31
150 ±48
7 23 ± 3 01
114 ±57
41 0 ± 8 2
42 0 ± 9 1
40 9 ± 8 6
39 0 ± 10 8
23 values (24 for NOX) Averages for the oven and broiler burners represent 20 blue-flame
and 16 yellow-flame values Where data are reported for both oven and broiler burners for a
single stove, the values have been averaged to produce one emission factor for the oven and
broiler category for that stove Values are generally very similar for emissions from these
4-37
-------
two types of burners on the same stove Overall, the results show that well-adjusted blue
flames emit more NO but less NO2 than poorly adjusted yellow flames Emission factors
from range-top burners are comparable to those from oven and broiler burneis Note that
the emission factors from Tikalsky et al (1987) have not been included in these averages
because flame type (blue or yellow) was not specified
A recent literature survey, which includes most of the studies cited in this section, has
examined data from range-top burners for the purpose of identifying factors that significantly
influence emissions of NO, NO2, and total NOX (Davidson et al , 1987) The data were used
with several statistical tests First, analysis of variance was used to investigaie the
importance of three binary factors in explaining the observed variations in emissions The
factors considered were (1) type of combustion (poorly adjusted or well adjusted), (2) burner
position (front or rear), and (3) method of sampling (direct with a hood or indirect with a
chamber). The results showed that roughly one-third of the variance in the base 10
logarithm of the emission factor (log EF) for NO2 can be explained by noting whether the
combustion is poorly adjusted or well adjusted For NO and total NOX, the fraction of total
variance in log EF explained by this factor depends on the subset of the data chosen Values
of the fraction range from 0 088 to 0 56, depending on whether the subset of data involved
front or rear burners, and whether the measurements were conducted with a hood or in a
chamber. Burner position and method of sampling were both relatively unimportant in
explaining the observed variance for NO, NO2, or NOX
The emission factor data were then used to estimate coefficients in various multivanate
regression models. The first regression model incorporated several factors type of
combustion, burner position, method of sampling, the three two-way interactions among
these factors, and (M-l) binary factors corresponding to the M stoves for which data were
available Subsequent multivanate regression models were constructed by sequentially
eliminating a factor or factors from the previous model Results of these tests showed that
stove differences were significant at the 95 % level in explaining the variance in NO2 and
NOX emission factor Type of combustion was significant for NO and NO2 Burner position
had a smaller but still statistically significant effect in explaining variance in NO2 emissions
Similarly, the method of sampling had a small but statistically significant effect for NOX
emissions.
4-38
-------
The influence of gas flow rate (heat input rate) on emission factors was investigated
separately Statistical tests were not run for this factor due to lack of data and due to the
presence of detailed data for only one study Results of plotting all of the data for NO and
NO2 are shown in Figures 4-5 and 4-6, respectively. The graphs illustrate wide variations in
emission factors, apparently due to varying stove chiiracteristics, testing conditions, and other
nonumformities among the data sets However, theie are general trends toward increasing
emissions for NO and decreasing emissions for NO2 as flow rate increases Note that the
detailed data of Borrazzo et al (1987) in these figures suggest that the emission/gas flow rate
relationships are complex, despite the general trends
ui
O
240
220
200
18.0
16.0
14.0
12.0
100
80
60
40
20
nn
.
_
-
_
.
.
x
-
-
x
\
v
xx
x x
X X
Xx
x+
A
*£ XXX
x
A
O
X
X '
X
A
A
A
A
O
x x
^x"
x x ^\
+A
A
A
A
O
A
o Himmel and DeWerth (1974)
A Fortmann et al (1 984)
+ Moschandreas et al (1985)
x Borrazzo etal (1987)
00 200 400 600 800 1000 1200 1400 1600
I gas (kJ/min)
Figure 4-5. Emission factors for nitric oxide as a function of gas flow rate.
4-39
-------
&
1
17.5
15.0
12.5
10.0
75
5.0
X
*
x
x
x
x
A **
XX
xx >
V
X
+
0 X
< XX
X
X
XXA
X
xx x 4
x x x
A
o Himmel and DeWerth (1974)
A Fortmann et al (1984)
+ Moschandreas et al (1985) A
x Borrazzo et al (1987)
< XX
X
A
X
xx
X
^V
A
A
X
$
A A
O
A
00 200 400 600 800
1000 1200 1400 1600
gas
Figure 4-6. Emission factors for nitrogen dioxide as a function of gas flow rate.
4.3.4 Unvented Space Heaters Fueled with Natural Gas and Propane
A number of studies have considered emissions of NOX from unvented space heaters
As with the stove emissions studies, several different types of heaters and a variety of
measurement techniques were used This section summarizes these studies Much of the
information has been taken from the literature review of Cole and Zawacki (1985) It is
important to note that the emission factors for heaters must be interpreted differently from
those for stoves, due to differences in use profiles
Thrasher and DeWerth (1979) studied emissions from five heaters by measuring NO
and NO2 concentrations in the flue products during well-adjusted blue-flame operation All
of the heaters had cast iron Bunsen burners with drilled ports Three of the heaters had
suspended radiant tiles above the flame, whereas two did not The heat input rates varied
from 86.1 to 661 kJ/min (4,900 to 37,600 BTU/h) Results of these tests are given in
4-40
-------
Table 4-19 The data show that the emission factors are not monotomcally related to heat
input rate, the influence of the radiant tiles is unclear based on these tests These
investigators also examined emissions during poorly adjusted yellow-flame operation
No definite conclusions could be reached when results of these tests were compared with
blue-flame emissions
o o
Traynor et al (1983a) examined eight heaters using a 27-m (953-ft) environmental
chamber operated at 0 5 air changes per hour Emifjsions were computed based on airborne
concentrations in the chamber using the mass balance method Heaters with cast iron Bunsen
burners of both drilled- and slotted-port designs were used Heat input rates varied from
188 to 830 kJ/min Emission factors for NO, NO2, and NOX are shown in Table 4-19
In addition to these chamber tests, Traynor et al (1983a) operated four of the heaters in
an experimental research house with a volume of 240 m3 The tests in the house showed
greater total NOX emissions than were measured in the chamber with the same heaters
However, the ratios of NO2 emissions to NOX emissions measured m the house with each
heater were much smaller than the corresponding ralios measured in the chamber These
differences were attributed to the much longer time periods of heater operation in the house
as compared with the chamber tests
Previous work by Traynor et al (1984a) involved heaters without oxygen depletion
sensors (ODSs) which were all fueled by natural gas This subsequent set of tests involved
ODS-equipped heaters that were fueled by both natural gas and propane Infrared as well as
convective heaters were used The measurements were conducted m the same manner as
employed previously All tests were run under well-tuned conditions at full heat input
Results are shown in Table 4-19 Nitrogen dioxide emissions from the infrared heaters
average about one-third of those from convective heaters, whereas NOX emissions from the
infrared heaters are an order of magnitude smaller Nitric oxide emissions from the infrared
heaters are very small, below the limits of the measurement methods in some cases
No significant differences are observed between propane and natural gas heaters
In other tests, Traynor et al (1984a) considered emissions during short-term use before
the heaters were able to warm up completely Compared with the longer-term emissions in
Table 4-19, the tests showed slightly lower NO2 and NOX emissions for the infrared heaters
4-41
-------
^.
I
TABLE 4-19. EMISSION FACTORS FOR NITROGEN OXIDE (pg/kj) AND NITROGEN DIOXIDE (jtg/kj)
FOR INVENTED SPACE HEATERS
Study
Thrasher and
DeWerth
(1979)a
Traynor et al
(1983a)b
Traynor et al
(1984a)c
Bilhck et al
(1984),
Moschandreas
etal (1985),
and Zawacki
etal (1984)d
Distinguishuig
Features
With radiant tiles
No radiant tiles
No radiant hies
With radiant tiles
With radiant tiles
Drilled port
Drilled port
Drilled port
Slotted port
Slotted port
Infrared, P
Infrared, N
Infrared, N
Convective, P, drilled port
Convective, N, nbbon port
Convective, P, slotted port
Bunsen, no radiant tiles
Bunsen, no radiant tiles
Bunsen, with radiant tiles
Catalytic, no radiant tiles
Catalytic, no radiant tiles
Radiant (ceramic), no radiant tues
Radiant (ceramic), no radiant tiles
Heat Input
Rate
(kJ/min)
661
861
176
311
156
188
656
830
424
592
245
263
317
335
486
626
186
186
255
207
207
260
260
Emission Factor
for Nitrogen
Oxide
47
35
24
27
34
95
22
16
16
19
01
01
01
287
178
282
18
15
22
009
0
039
0
Emission Factor
for Nitrogen Dioxide
Qig/KS)
52
73
22
60
60
20
12
20
11
95
59
52
62
124
129
10
99
15
90
01
13
38
47
Emission Factor
for Nitrogen Oxides
Otg/kJ)
77
60
39
47
56
34
43
47
36
39
59
52
62
565
40 1
532
35
37
42
03
13
43
47
aFor connective heaters with drilled ports using natural gas
For convective heaters with radiant tiles using natural gas
°For convective heaters with radiant tiles and for infrared heaters fueled with natural gas (N) and propane (P)
For heaters with bunsen, catalytic, and ceramic tile burners using natural gas
-------
Similarly, the short-term emissions of NOX for the connective heaters were slightly lower
than the longer term emissions
Billick et al (1984), Moschandreas et al (1985), and Zawacki et al (1984) examined
four heaters of different burner designs Two of the heaters had ribbon-port Bunsen burners,
one had a catalytic burner, and one had a ceramic-tile burner The Bunsen burners were
used in the well-adjusted blue-flame mode The units were operated at their respective
maximum heat input rates, which varied from 186 to 260 kJ/min (10,600 to 14,800 BTU/h)
The tests involved direct measurements of the exhaust gases for all heaters, three of the
3 3
heaters were also tested in a 33-m (1,150-ft) chamber using the mass balance method The
results, shown in Table 4-19, indicate that the NO2 emission factors measured in the chamber
are much greater than those determined using the direct testing for each of the three heaters
The opposite is true for the NO emission factor
These researchers also investigated ways of reducing emissions by using various inserts
in the flame of one of the Bunsen burner heaters They found that ceramic rod inserts used
to reduce the temperature of the flame reduced NO emissions by 44%, although NO2
emissions were unchanged and CO emissions increased
In more recent work, Zawacki et al (1986) considered natural gas and propane heaters
These included three convective and four infrared heaters fueled with natural gas, two
propane convective heaters, and one propane infrared heater Tests were conducted using
three measurement methods a probe, a sampling hood, and a chamber Some of the heaters
were the same ones used previously by Traynor et al (1984a)
Results are shown in Table 4-20 Statistical analyses were performed on these data by
the investigators, the differences were generally insignificant between emissions determined
with the probe and with the hood for NO, NO2, and NOX However, NO emissions obtained
with the chamber method were smaller than those obtained with the other two methods This
was attributed to the vitiated atmosphere maintained within the chamber at the low air
exchange rates used Overall, the results were in general agreement with the data of Traynor
etal (1984a)
4-43
-------
TABLE 4-20. EMISSION FACTORS FOR NITROGEN OXIDE AND
NITROGEN DIOXIDE (^tg/kj) FOR CONVECTIVE AND
INFRARED HEATERS OF VARIOUS DESIGNS, USING
NATURAL GAS AND PROPANE (ZAWACKI ET AL., 1986)
Rated Input
Distinguishing Rate
Features" Test Method (kJ/min)
CBR, no RT, N
CBR, with RT.N
Infrared, N
Infrared, N
Convective,
slotted port, with
RT, P
Convective,
drilled port, with
RT, P
CBR, with RT, N
Infrared P
Probe 176
Hood
Chamber
Probe 264
Hood
Chamber
Probe 264
Hood
Chamber
Probe 304
Hood
Chamber
Probe 703
Hood
Chamber
Probe 352
Hood
Chamber
Probe 527
Hood
Chamber
Probe 316
Hood
Chamber
Emission Emission Emission Factor for
Factor for Factor for Nitrogen Oxides (as
Nitrogen Oxide Nitrogen Dioxide Nitrogen Dioxide)
Gig/kJ) (jig/kJ) (/tg/kJ)
177
16
97
21 7
207
11
06
0005
0005
17
05
03
398
393
283
444
43 9
31 8
13
145
53
09
05
04
98
109
84
84
9 1
104
1 8
28
35
26
4 1
53
76
76
8
84
8 1
105
233
20
19
16
3
3 3
353
354
233
419
408
272
27
28
35
5 1
48
57
686
679
512
764
754
593
433
43 1
27 1
3
3 9
4
CBR = Convective ribbon burners
RT = Radiant tiles
N — Natural gas
P = Propane.
Summary of Emissions from Unvented Gas Space Heaters
This section has summarized the findings of five separate investigations, with the data
given in Tables 4-19 and 4-20 The tables show that, on the average, conveclive space
heaters have emissions of NO roughly three tunes the emissions of NO2 The influence of
radiant tiles on emissions is not clear Heaters with catalytic burners, radiant ceramic-tile
4-44
-------
burners, and improved-design steel burners (radiant and Bunsen) have much smaller NO and
NO2 emissions than heaters with conventional cast-iron Bunsen burners
These data also indicate that measurement of emissions with a probe, sampling hood, or
chamber can all yield equivalent results under certain conditions Zawacki et al (1986)
suggest the use of a sampling hood as the preferred method, because of its versatility and
ease in use
4.3.5 Kerosene Heaters
In this section, the results of three studies reporting NOX emissions for portable
kerosene heaters are examined Yamanaka et al (1979) examined emissions from six radiant
and five convective kerosene heaters Emissions were sampled using a hood positioned over
the heater Results were presented only for total NOX (as NO2) The emission factors for
the radiant heaters averaged 13 ± 1 8 /xg/kJ, whereas the emission factors for the convective
heaters averaged 70 + 6 8 /ig/kJ Overall heat input rates for the 11 heaters were in the
range 110 to 200 kJ/min
Leaderer (1982) measured emissions of NO, NO2, and other pollutants from one
radiant and one convective kerosene heater The heaters were rated at 169 kJ/min
(9,600 BTU/h) and 153 kJ/min (8,700 BTU/h), respectively Measurements were performed
in a 34-m3 (1,200-ft3) chamber operated at 100 air changes per hour Emission factors were
determined by mass balance The data were obtainesd at three different heat input rates for
each heater, there were three sets of runs for each heat input rate
Results are shown in Table 4-21 Emission factors for NO from the radiant heater are
very small, whereas those from the convective heater are more than an order of magnitude
greater For NO2, the emission factors are also greater from the convective heater than from
the radiant heater, but only by factors of 1 5 to 3
Traynor et al (1983b) tested two radiant and two convective kerosene heaters
Emission factors were determined by the mass balance method, using airborne concentrations
measured in a 27-m chamber operated at 0 4 air changes per hour Two types of tests were
conducted In the first type, the heater was fired in the chamber and allowed to run for 1 h
In the second type, the heater was fired outside the chamber and allowed to warm up for
10 nun The heater was then brought into the chamber for a 1-h run
4-45
-------
TABLE 4-21. AVERAGE EMISSION FACTORS FOR NITRIC OXIDE AND
NITROGEN DIOXIDE FROM KEROSENE HEATERS, AFTER LEADERER (1982)
AND TRAYNOR ET AL. (1983b)
Leaderer
Leaderer
Leaderer
Leaderer
Leaderer
Leaderer
Traynor
ctal
Traynor
ctal.
Traynor
ctal
Traynor
ctal
Type of Heater
Radiant, new
Radiant, new
Radiant, new
Convective,
new
Convective,
new
Convective,
new
Radiant, new
Radiant,
1 year old
Convective,
new
Convective,
5 years old
Heat Input
Rate
(kJ/min)
144
113
844
158
979
373
137
111
131
948
Emission Factor for
Nitnc Oxide
0*g/kJ)
0 45 ± 0 05
0 08 + 0 05
0
17 ±03
12 ±06
11 ± 09
13 ±07
21
25 ±07
11 ±0 1
Emission Factor for
Nitrogen Dioxide
(pg/V)
44 ± 02
50 ± 02
59 ± 03
70 ± 04
15 ±03
17 ± 10
46 ± 08
5 1
13 ± 08
32 ± 28
Emission Factor for
Nitrogen Oxides
(Aig/kJ)
5 1 ± 02
5 1 ±02
59 ±03
33 ± 06
33 ± 10
34 ± 17
66 ± 13
83
51 ± 1 3
49 ± 28
Results are shown in Table 4-21 The data are the averages of both types of tests for
the two convective heaters and for the new radiant heater. The 1-year-old radiant heater was
studied using only the second type of test The original data show little difference in
emission factor between the two types of tests Overall, the results are similar to those of
Leaderer (1982), showing smaller emission factors for radiant as compared with convective
heaters.
The data reported by Traynor et al (1983b) in Table 4-21 refer to operation at the
maximum wick length In additional tests, these investigators reduced the length of the wick
to half of the full setting of the wick control knob (radiant heater) or until the flame was half
the length as it was previously (convective heater) Results showed slightly smaller NO
emissions for both heater types, for NO2, the emissions were comparable to those of the full
wick for the radiant heater, but about double those of the full wick for the convective heater
4-46
-------
Apte and Traynor (1986) reviewed data obtained by a Lawrence Berkeley Laboratory
group on emissions from two-stage kerosene heaters The two-stage burners resemble those
of radiant kerosene heaters, except that there is a second chamber above the radiant element
where additional combustion ear is introduced In thus region, the flame temperature is
allowed to rise and the combustion process is more complete Emissions of NO from the
two-stage heater are slightly greater than those from radiant heaters due to the higher flame
temperature, although emissions of NO2 as well as of CO and unburned HCs are lower The
data presented in this section show that emission factors of NO and NO2 for radiant kerosene
heaters are generally much smaller than those for convective kerosene heaters Emissions of
NO from two-stage heaters are only slightly greater than those from radiant heaters, whereas
emissions of NO2 are the lowest of the three heater types Most of the NOX emissions from
radiant heaters are in the form of NO2, for convective heaters that are two-stage heaters, the
emissions of NO and NO2 are of comparable magnitude There are insufficient data to
evaluate changes in emissions as kerosene heaters age
4.3.6 Wood Stoves
The use of wood stoves for residential space heating has become increasingly popular in
recent years Because of this, a number of studies have examined pollutant emissions from
wood stoves Some of these studies have measured emission factors based on concentrations
in the flue gases, such information would be useful for assessing the contribution of wood
stove emissions to ambient air quality Very little information is available, however, on
fugitive emissions from wood stoves into the indoor living space
In a detailed literature survey, Smith (1987) reports that emissions of pollutants from
wood stoves are highly variable, depending on the type of wood used, stove design, the way
the stove is used, and other factors He reports emission factors for NOX and other
pollutants for wood stoves used in developing countries Many of these stoves are unvented,
resulting in excessive indoor concentrations as the combustion products are exhausted into the
room This information is not applicable to the United States, where virtually all wood
stoves are vented to the outdoors
Traynor et al (1984b) have studied wood stoves used in a house (three airtight and one
nonaurtight) For each experiment, airborne concentrations of several pollutants were
4-47
-------
measured inside and outside the house during operation of one of the stoves The results
showed that all indoor and outdoor concentrations of NO and NO2 were below 0 02 ppm
Indoor airborne concentrations of some of the other pollutants were high during use of the
nonairtight stove, however The airtight stoves had little influence on indoor concentrations
of any pollutants In another study, Traynor et al (1982) found elevated airborne
concentrations of NO and NO2 in three occupied houses during operation of wood stoves and
a wood furnace The concentrations were highly variable, however, and the authors caution
that additional tests would be needed to determine the influence of wood stoves on indoor
concentrations of NOX and other pollutants
Because of the paucity of data, it is difficult to reach quantitative conclusions regarding
the importance of wood stoves However, the Limited information available suggests that
wood stoves are not a major contributor to NOX exposures indoors This is consistent with
the small NO emission rates expected from the low temperature combustion processes
characteristic of wood stoves
4.3.7 Tobacco Products
A number of studies have compared concentrations of NOX and other pollutants in
houses with smokers and houses without smokers In general, these studies have shown that
concentrations are greater in the homes of smokers There are few data available, however,
on NOX emission factors of tobacco products
A few studies have reported emissions of NOX from cigarettes while sampling both
sidestream and mainstream smoke together Woods (1983) report 0 079 mg/cigarette for
NO2, and the University of Kentucky (undated) reports 1 56 mg/cigaiette for NOX (as NO)
Moschandreas et al. (1985) lists emissions of 2 78 mg/cigarette for NO and
0.73 mg/cigarette for NO2 The National Research Council (1986) reports total NOX
emissions of 100 to 600 /tg/cigarette for mainstream smoke, with values 4 to 10 times greater
for sidestream smoke According to this reference, virtually all of the emitted NOX is in the
form of NO, once emitted, the NO is gradually oxidized to NO2 Thus, enviionments
containing cigarette smoke may have higher concentrations of both NO and NO2 than
environments without such smoke
4-48
-------
4.3.8 Comparison of Emissions from Sources Influencing Indoor Air
Quality
This section has considered emissions of NOX fiom gas stoves, heaters using natural gas
or propane, kerosene heaters, wood stoves, and tobacco products A significant number of
apphances in the first three categories have been tested, emission factors in these categories
have been averaged and are shown in Table 4-22 Note that all data for a single appliance
have been averaged before being used as an input to compute the grand averages shown in
the table This procedure has been followed even when a single appliance has been tested by
more than one group (e g , Tables 4-19 and 4-21) Some of the reported data are given as
values below a specified limit of detection, these values have been taken as zero in this
analysis Only data for the hood measurement technique have been used in Table 4-22 For
kerosene heaters, the NOX values include the data of Yamanaka et al (1979), whereas the
NO and NO2 values include only the data of Leaderer (1982) and Traynor et al (1983b)
TABLE 4-22. AVERAGE EMISSION FACTORS FOR NITRIC OXIDE,
NITROGEN DIOXIDE, AND NITROGEN OXIDES FROM VARIOUS SOURCES
BASED ON DATA REPORTED IN THE LITERATURE
Range-Top Burners
Ovens and Broilers
Unvented Gas Heaters
(Cast-iron Bunsen Burners)
Kerosene Heaters
(Convective)
Kerosene Heaters (Radiant)
Number of
Appliances
27
20
15
5
5
Emission Factor Emission Factor
for Nitric Oxide for Nitrogen Dioxide
(Aig/kJ) ftig/kJ)
20 0 + 4 5
219 ±63
229 ±96
152 ±602
079 ±090
102 ±
723 d
108 ±
168 ±
3 1
: 301
53
93
5 00 ± 0 58
Emission Factor
for Nitrogen Oxides
(pg/kl)
41 0 ± 8 2
40 9 ± 8 6
45 4 ± 11 4
55 1 ± 17 7a
9 69 + 3 68b
aTotal number of convective kerosene heaters tested for NOX = 10
Total number of radiant kerosene heaters tested for NOX =11
These data show that emissions of NOX are about 65 to 75% NO and 25 to 35% NO2
for range-top burners, for ovens and broilers fueled with natural gas, and for convective
heaters fueled with natural gas and propane In contrast, convective kerosene heaters have
emissions of NO and NO2 that are roughly comparable Radiant heaters using natural gas,
4-49
-------
propane, or kerosene all have emissions of NO that are negligible compared with those of
N02.
For gas stoves, emission factors of NO and NO2 from range-top burners operating at
maximum heat input rate are comparable to those from the oven and broiler burners
Well-adjusted blue flames emit slightly more NO and slightly less NO2 than poorly adjusted
yellow flames. The emission factor of NO increases as the heat input rate increases on top
burners, and the emission factor of NO2 decreases as the heat input rate increases When
first starting a range-top burner with a cold load (e g , a water-filled pot), the emission factor
of NO is initially small, but steadily increases as the load warms up The emission factor of
NO2, on the other hand, is initially high, but steadily decreases Pilot lights associated with
range-top burners have emission factors comparable to or slightly smaller than emission
factors for the burners Pilot lights associated with oven and broiler burners have NO
emission factors much smaller than those of the burners themselves, although NO2 emission
factors are comparable
Emission factors for NO and NO2 from convective natural gas and propane heaters with
Bunsen burners are similar to emission factors from natural gas stoves Emission factors of
NO from convective kerosene heaters are slightly smaller than those from gas stoves and
convective gas heaters, whereas emission factors of NO2 are slightly greater Emission
factors of NO from radiant heaters using natural gas, propane, or kerosene are very small,
whereas those for NO2 are about one-third of the respective convective heater emissions
There appears to be little difference in emission factor from a warmed kerosene heater
compared with emissions from a cold start The length of the wick can affect emissions
a smaller wick yields smaller NO emissions, but possibly greater NO2 emissions Two-stage
kerosene heaters have smaller emissions of NO2 than do either convective or radiant
kerosene heaters
Emissions of NOX to the indoor environment from wood stoves are not expected to be
significant given the low combustion temperatures involved No data are available to allow
quantification of such emissions, however
Only limited data are available on NOX emissions from tobacco Nearly all of the
emitted NOX from cigarettes is in the form of NO, although oxidation to NO2 occurs over
4-50
-------
time periods of several minutes following emission Sidestream smoke from a cigarette
contains up to an order of magnitude more NO than mainstream smoke
The emission factors given in Table 4-22 can be used with indoor air quality models to
predict indoor airborne concentrations of NOX, provided input data for other parameters
included in the model are available Examples of such parameters include air exchange
characteristics of the house, outdoor airborne concentrations of NOX, and appliance usage
patterns The last parameter is especially important very little information is available on
the frequency of use of stoves and other combustion sources Information on the way
occupants influence air exchange, such as by opening windows and doors, is also very
limited For many situations, our ability to predict indoor airborne concentrations is limited
by our lack of understanding of occupant behavior, rather than by lack of data on emission
factors
4.4 SUMMARY OF EMISSIONS OF NITROGEN OXIDES FROM
AMBIENT AND INDOOR SOURCES
Anthropogenic sources of NO2 emissions include transportation, stationary source fuel
combustion, various industrial processes, solid waste disposal, and others, such as forest
fires Natural sources of NOX are lightning, biological and abiological processes in soil, and
stratospheric intrusion Estimates for 1990 indicate lhat over 80% of the United States NOX
emissions are emitted by highway vehicles, electric utilities, and industrial boilers
Quantitative estimates of the total amount of NOX emitted to the ambient global
atmosphere are available These estimates suggest that 122 to 152 x 10 metric tons of NOX
are emitted annually, with about 18 to 19 x 10 metric tons emitted in the United States
alone
The important indoor sources of NOX are gas stoves, unvented space heaters, kerosene
heaters, wood stoves, tobacco products, and infiltration of ambient air containing NOX
Total emissions and the ratio of NO/NO2 from gas stoves and space heaters differ according
to fuel flow rate and flame adjustment Additional factors, such as the load (e g , cold pot
of water), heater type (convective versus radiant), and fuel type (natural gas, propane, or
4-51
-------
kerosene) may also be important Only limited information is available for wood stoves and
tobacco products.
Two factors, ambient and indoor NOX emissions, form the primary bases that determine
air concentrations and exposure in the human environment
4-52
-------
REFERENCES
Albntton, D L , Lm, S C , Kley, D (1984) Global nitrate deposition from kghtmng In Aneja, V P , ed
Environmental impact of natural emissions proceedings of an Air Pollution Control Association specialty
conference, March, Research Triangle Park, NC Pittsburgh, PA Air Pollution Control Association
Apte, M G , Traynor, G W (1986) Comparison of pollutant emission rates from unvented kerosene and gas
space heaters la Proceedings of the ASHRAE conference IAQ '86 managing indoor air for health and
energy conservation, April, Atlanta, GA Atlanta, GA American Society of Heating, Refrigerating and
Air-Conditioning Engineers, Inc , pp 405-416
Bartok, W , Sarofim, A F (1991) Fossil fuel combustion a source book New York, NY John Wiley & Sons,
Inc
Bauer, E (1982) Natural and anthropogenic sources of oxides of nitrogen (NOX) for the troposphere
Washington, DC U S Department of Transportation, Federal Aviation Administration, report no
FAA-EE-82-7 Available from NTTS, Springfield, VA, AD-A134516
Billick, I, Johnson, D , Moschandreas, D , Relwam, S (1984) Art investigation of operational factors that
influence emission rates from gas appliances In Berglund, B , Lindvall, T , Sundell, J , eds Indoor air
proceedings of the 3rd international conference on indoor air quality and climate, v 4, chemical
characterization and personal exposure, August, Stockholm, Sweden Stockholm, Sweden Swedish
Council for Building Research, pp 181-187 Available from NTIS, Springfield, VA, PB85-104214
Borrazzo, J E , Osborn, J F , Fortmann, R C , Keefer, R L , Davidson, C I (1987) Modeling and
monitoring of CO, NO and NO2 in a modern townhouse Atmos Environ 21 299-311
Borucki, W J , Chameides, W L (1984) Lightning estimates of the rates of energy dissipation and nitrogen
fixation Rev Geophys Space Phys 22 363-372
Caceres, T , Soto, H , Lissi, E , Cisternas, R (1983) Indoor house pollution appliance emissions and indoor
ambient concentrations Atmos Environ 17 1009-1013
Cole, J T , Zawacki, T S (1985) Emissions from residential gas-fired appliances Chicago, IL Institute of Gas
Technology, report no GRI-84/0164 Available from NTIS, Springfield, VA, PB85-174928
Cole, J T , Zawacki, T S , Macnss, R A , Moschandreas, D J (1983) Constituent source emission rate
characterization of three- gas fired domestic ranges Presented at 76th annual meeting of the Air
Pollution Control Association, June, Atlanta, GA Pittsburgh, PA Air Pollution Control Association,
paper no 83-64 3
Cote, W A , Wade, W A , m, Yocom, J E (1974) A study of indoor air quality Washington, DC U S
Environmental Protection Agency, Office of Research and Development, EPA report no
EPA-650/4-74-042 Available from NTIS, Springfield, VA, PB-238556
Coutant, R W , Merryman, E L , Levy, A (1982) Formation of NO^ in range-top burners Environ Lit
8 185-192
Crutzen, P J , Schmailz, U (1983) Chemical budgets of the stratosphere Plant Space Sci 31 1009-1032
Davidson, C I , Borrazzo, J E , Hendrickson, C T (1987) Pollutant emission factors for gas stoves
a literature survey Research Triangle Park, NC U S Environmental Protection Agency, Air and Energy
Engineering Research Laboratory, EPA report no EPA-600/9-87-005 Available from NTIS,
Springfield, VA, PB87-171328
4-53
-------
Ehhalt, D H , Drummond, J W (1982) The troposphenc cycle of NOX In Georgu, H W , Jaeschke, W ,
eds Chemistry of the unpolluted and polluted troposphere proceedings of the NATO Advanced Study
Institute, September-October 1981, Corfu, Greece Boston, MA D Reidel Publishing Company,
pp 219-251 (NATO Advanced Study Institutes series C—mathematical and physical sciences v 96)
Federal Register (1993) Air pollution control, motor vehicle emission factors—notice of model availability F R
(May 20) 58 29409-29410
Fcmmore, C. P (1971) Formation of nitric oxide in premixed hydrocarbon flames In Thirteenth symposium
(international) on combustion, August 1970, Salt Lake City, UT Pittsburgh, PA The Combustion
Institute, pp 373-380
Fortmann, R C , Borrazzo, J E , Davidson, C I (1984) Characterization of parameters influencing indoor
pollutant concentrations In Berglund, B , Lindvall, T , Sundell, J , eds Indoor air proceedings of the
3rd international conference on indoor air quality and climate, v 4, chemical characterization and
personal exposure, August, Stockholm, Sweden Stockholm, Sweden Swedish Council for Building
Research, pp. 259-264 Available from NITS, Springfield, VA, PB85-104214
Gschwandtner, G , Barnard, W , Carlson, P (1990) Feasibility of including regional and tempo al VOC
emissions estimates in the EPA emissions trends report Research Triangle Park, NC U S
Environmental Protection Agency, Office of Air Quality Planning and Standards, EPA report no
EPA/450/4-91/005A Available from NTTS, Springfield, VA, PB92-216910/XAB
Hare, C. T., Springer, K J (1973a) Exhaust emissions from uncontrolled vehicles and related equipment using
internal combustion engines part 3—motorcycles Ann Arbor, MI U S Environmental Protection
Agency, Office of Mobile Source Air Pollution Control, EPA report no APTD-1492 Available from
NTIS, Springfield, VA, PB-224884/7
Hare, C T , Springer, K J (1973b) Exhaust emissions from uncontrolled vehicles and related equipment using
internal combustion engines Final report Part 5 heavy-duty farm, construction, and industrial engines
Washington, DC U S Environmental Protection Agency, Office of Air Quality Planning and Standards,
report no APTD-1494 Available from NTIS, Springfield, VA, PB-232507
Hare, C T , Springer, K J (1973c) Exhaust emissions from uncontrolled vehicles and related equipment using
internal combustion engines part 4—small air-cooled spark ignition utility engines Ann Arbor, MI U S
Environmental Protection Agency, Office of Air and Water Programs, publication no APTD-1493
Available from. NTIS, Springfield, VA, PB-224885
Hayhurst, A N , Vince, I M (1980) Nitric oxide formation from nitrogen flames The importance of "prompt"
nitric oxide Prog Energy Combust Sci 6 35-51
Heywood, J B (1988) Internal combustion engine fundamentals New York, NY McGraw-Hill Book Co ,
Inc ; p. 571
Heisler, S. L , Collins, H , Collins, J , Hayden, P , Lurmann, F , Yeransian, T , Young, J (1988) Interim
emissions inventory for regional air quality studies Palo Alto, CA Electric Power Research Institute,
report no EPRIEA-6070
Himmel, R L , DeWerth, D W (1974) Evaluation of the pollutant emissions from gas-fired ranges Cleveland,
OH American Gas Association Laboratories, report no 1392
4-54
-------
Hollowell, C D , Budnitz, R J , Traynor, G W (1977) Combustion-generated indoor air pollution
In Kasuga, S , Suzuki, N , Yamada, T , Kimura, G , Inagaki, K , Onoe, K , eds Proceedings of the
fourth, international clean air congress, May, Tokyo, Japan Tokyo, Japan Japanese Union of Air
Pollution Prevention Associations, pp 684-687
Kohout, E J , Miller, D J , Nieves, L A , Rothman, D S , Sancks, C L , Stodolsky, F , Hanson, D A ,
(1990) Current emission trends for nitrogen oxides, sulfur dioxide, and volatile organic compounds by
month and state methodology and results Argonne, IL AJ gonne National Laboratory, Environmental
Assessment and Information Sciences Division, report no ANL/EAIS/TM-25 Available from NTIS,
Springfield, VA, DE92008666
Leaderer, B P (1982) Air pollutant emissions from kerosene space/ heaters Science (Washington, DC)
218 1113-1115
Liu, S C , Trainer, M , Fehsenfeld, F C , Parnsh, D D , Williams, E J , Fahey, D W , Huebler, G ,
Murphy, P C (1987) Ozone production in the rural troposphere and the implications for regional and
global ozone distributions J Geophys Res [Atmos ] 92 4191-4207
Logan, J A (1983) Nitrogen oxides in the troposphere global and regional budgets J Geophys Res C
Oceans Atmos 88 10785-10807
McNay, L M (1971) Coal refuse fires, an environmental hazard Washington, DC U S Department of the
Interior, Bureau of Mines, information circular 8515
Moschandreas, D J , Relwani, S M , O'Neill, H J , Cole, J T , Elkins, R H , Macnss, R A (1985)
Characterization of emission rates from indoor combustion sources Chicago, IL Gas Research Institute,
report no GRI 85/0075 Available from NTIS, Springfield, VA, PB86-103900
National Research Council (1986) Environmental tobacco smoke measuring exposures and assessing health
effects Washington, DC National Academy Press
Peeters, J , Mahnen, G (1973) Structure of ethylene-oxygen flames Reaction mechanism and rate constants of
elementary reactions In Weinberg, F J , ed Combustion Institute European Symposium 1973 papers
for presentation at the symposium, September, University of Sheffield, United Kingdom New York, NY
Academic Press, pp 53-58
Placet, M , Battye, R E , Fehsenfeld, F C , Bassett, G W (1991) Emissions involved in acidic deposition
processes In Irving, P M , ed Acidic depostion state of science and technology, volume I, emissions,
atmospheric processes, and deposition Washington, DC The U S National Acid Precipitation
Assessment Program (State of science and technology report no 1)
Reuther, J J , Billick, I H , Gaynor, A J (1988) The measurement of NC<2 from gas flames Combust Flame
71 331-335
Saeger, M , Langstaff, J , Walters, R , Modica, L , Zimmerman, I) , Fratt, D , Dulleba, D , Ryan, R ,
Demmy, J , Tax, W , Sprague, D , Mudgett, D , Werner, A S (1989) The 1985 NAPAP emissions
inventory (version 2) development annual data and modelers' tapes Research Triangle Park, NC U S
Environmental Protection Agency, Air and Energy Engineering Research Laboratory, EPA report no
EPA/600/7-89/012A Available from NTIS, Springfield, VA., PB91-119669/XAB
Shannon, L J , Gorman, P G , Reichel, M (1971) Particulate pollutant system study, volume U—fine particle
emissions Durham, NC U S Environmental Protection Agency, Air Pollution Control Office, EPA
report no APTD-0744 Available from NTIS, Springfield, VA, PB-203531
4-55
-------
Shukla, K C ; Hurley, J R (1983) Development of an efficient, low NOX domestic lange cooktop Thermo
Electron Corporation research, report, GBI contract no 5081-241-0544
Smith, K R (1987) Biofuels, air pollution, and health a global review New York, NY Plenum Press (Lester,
R K , Andelin, J , eds Modern perspectives in energy)
Statutes-at-Large (1988) Alternative motor fuels act of 1988, PL 100-494, October 14, 1988 Stat 102
2441-2453
Stcdman, D H , Shelter, R E (1983) The global budget of atmosphenc nitrogen species In Schwartz, S E ,
ed Trace atmosphenc constituents properties, transformations, and fates New York, NY John Wiley &
Sons, pp 411-454 (Nnagu, J O , ed Advances in environmental science and technology v 12)
Thrasher, W. H , DeWerth, D W (1979) Evaluation of the pollutant emissions from gas-fired room heaters
Cleveland, OH American Gas Association Laboratories, research report no 1515
Tikalsky, S , Reisdorf, K , Flickinger, J , Totzke, D , Haywood, J , Annen, L , Kanarek, M , Kaarakka, P ,
Prins, E (1987) Gas range/oven emissions impact analysis [final report (July 1985-December 1987)]
Chicago, EL Gas Research Institute, report no GRI-87/0119 Available from NTIS, Springfield, VA,
PB88-232756
Traynor, G W , Allen, J R , Apte, M G , Dillworth, J F , Girman, J R , Hollowell, C D , Koonce,
J F , Jr (1982) Indoor air pollution from portable kerosene-fired space heateis, wood-burning stoves,
and wood-burning furnaces In Proceedings of the Air Pollution Control Association specialty conference
on residential wood and coal combustion, March, Louisville, KY Pittsburgh, PA Air Pollution Control
Association, pp 253-263
Traynor, G. W , Girman, J R , Allen, J R , Apte, M G , Carruthers, A R , Dillworth, J F , Martin, V M
(1983a) Indoor air pollution due to emissions from unvented gas-fired space heaters Presented at 76th
annual meeting of the Air Pollution Control Association, June, Atlanta, GA Pittsburgh, PA Air
Pollution Control Association, paper no 83-9 6
Traynor, G W , Allen, J R , Apte, M G , Girman, J R , Hollowell, C D (1983b) Pollutant emissions from
portable kerosene-fired space heaters Environ Sci Technol 17 369-371
Traynor, G W , Apte, M G , Carruthers, A R , Dillworth, J F , Gnmsrud, D T (1984a) Pollutant emission
rates from unvented infrared and convective gas-fired space heaters Berkeley, CA U S Department of
Energy, Lawrence Berkeley Laboratory, report no LBL-18258 Available from NTIS, Springfield, VA,
DE85 010647/XAB
Traynor, G W , Apte, M G , Carruthers, A R , Dillworth, J F , Gnmsrud, D T , Gundel, L A (1984b)
Indoor air pollution to emissions from wood burning stoves Presented at 77th annual meeting of the Air
Pollution Control Association, June, San Francisco, CA Pittsburgh, PA Air Pollution Control
Association, paper no 84-33 4
U.S. Department of Commerce (Annual a) Current industrial reports Washington, DC Bureau of the Census
U S Department of Commerce (Annual b) Statistical abstract of the United States Washington, DC Bureau of
the Census
U S. Department of Energy (1982) Estimates of U S wood energy consumption from 1949 to 1981
Washington, DC Energy Information Administration, Office of Coal, Nuclear, Electricity, and Alternate
Fuels; publication no DOE/EIA-0341 Available from NTIS, Springfield, VA, DE82020578
4-56
-------
U S Department of Energy (1984) Estimates of U S wood energy consumption 1980-1983 Washington, DC
Energy Information Administration, Office of Coal, Nuclear, Electric and Alternate Fuels, publication
no DOE/EIA-0341(83) Available from NTIS, Springfield, VA, DE85003669/XAB
U S Department of Energy (1991) Inventory of power plants in the United States 1990 Washington, DC
Energy Information Administration, report no DOE/EIA-0095(90) Available from NTIS, Springfield,
VA, DE92-002711
U S Department of Energy (Annual a) Petroleum supply annual Washington, DC Energy Information
Administration
U S Department of Energy (Annual b) Coal distribution January-December Washington, DC Energy
Information Administration
U S Department of Energy (Annual c) Electric power annual Washington, DC Energy Information
Administration
U S Department of Energy (Annual d) Natural gas annual Washington, DC Energy Information
Administration
U S Department of Energy (Monthly) Petroleum marketing montlily Washington, DC Energy Information
Administration
U S Department of Health, Education, and Welfare (1968) National survey of community solid waste practices
Cincinnati, OH Public Health Service, PHS publication no 1867
U S Department of the Interior (Annual) Minerals yearbook Washington, DC Bureau of Mines
U S Department of Transportation (Annual a) FAA air traffic activity Washington, DC Federal Aviation
Administration
U S Department of Transportation (Annual b) Highway statistics Washington, DC Federal Highway
Administration
U S Environmental Protection Agency (1985) Compilation of air pollutant emission factors, volume I stationary
point and area sources, volume n mobile sources 4th ed Research Triangle Park, NC Office of Air
Quality Planning and Standards, EPA report no AP-42-ED-4-VOL-1 and AP-42-ED-4-VOL-2 Available
from NTIS, Springfield, VA, PB86-124906 and PB87-205266
U S Environmental Protection Agency (1989) User's guide to MOBILE4 (Mobile Source Emission Factor
Model) Ann Arbor, MI Office of Mobile Sources, EPA report no EPA-AA-TEB-89-01 Available
from NTIS, Springfield, VA, PB89-164271
U S Environmental Protection Agency (1990) National air pollutant emission estimates, 1940-1988 Research
Triangle Park, NC Office of Air Quality Planning and Standards, EPA report no EPA/450/4-90/001
Available from NTIS, Springfield, VA, PB90-199266/HSU
U S Environmental Protection Agency (1991a) National air pollutant emission estimates 1940-1990 Research
Triangle Park, NC U S Environmental Protection Agency, Office of Air Quality Planning and
Standards, EPA report no EPA/450/4-91/026 Available fiom NTIS, Springfield, VA, PB92-152859
U S Environmental Protection Agency (1991b) AIRS facility subsystem Research Triangle Park, NC Office of
Air Quality and Standards, National Air Data Branch
4-57
-------
U.S. Environmental Protection Agency (1992) National air pollutant emission estimates, 1900-1991 Research
Triangle Park, NC Office of Air Quality Planning and Standards, EPA report no EPA/454/R-92/013
Available from NITS, Springfield, VA, PB93-157808/XAB
University of Kentucky (n d ) Physical and analytical data on IR4F reference cigarettes from Tobacco and Health
Research Institute Lexington, KY [as cited in Moschandreas et al (1985)]
Vandegrift, A E , Shannon, L J , Gorman, P G , Lawless, E W , Sallee, E E , Reichel, M (1971a)
Particulate pollutant system study, volume I—mass emissions Durham, NC U S Environmental
Protection Agency, Air Pollution Control Office, EPA report no APTD-0743 Available from NTIS,
Springfield, VA, PB-203128
Vandegnft, A E , Shannon, L J , Lawless, E W , Gorman, P G , Sallee, E E , Reichel, M (1971b)
Particulate pollutant system study, volume HI—handbook of emission properties Durham, NC U S
Environmental Protection Agency, Air Pollution Control Office, EPA report no APTD-0745 Available
from NTTS, Springfield, VA, PB-203522
Woods, J E (1983) Sources of indoor air contaminants ASHRAE Trans 89 462-497
Yamanaka, S., Hirose, H , Takada, S (1979) Nitrogen oxides emissions from domestic kerosene-fired and
gas-fired appliances Atmos Environ 13 407-412
Yamate, G. (1974) Emissions inventory from forest wildfires, forest managed burns, and agricultural burns
Research Triangle Park, NC U S Environmental Protection Agency, Office of Air Quality Planning and
Standards, EPA report no EPA-450/3-74-062 Available from NTIS, Springfield, VA, PB-238766
Zafiriou, O. C ; McFarland, M (1981) Nitric oxide from nitrite photolysis in the central equatorial Pacific
J Geophys Res. C- Oceans Atmos 86 3173-3182
Zawacki, T. S ; Cole, J T , Huang, VMS, Banasiuk, H , Macnss, R A (1984) Efficiency and emissions
improvement of gas-fired space heaters Task 2 Unvented space heater emission reduction [final report]
Chicago, D> Gas Research Institute, report no GRI-84/0021 Available from NTIS, Springfield, VA,
PB84-237734.
Zawacki, T. S ; Cole, J T , Jasionowski, W J , Macnss, R A (1986) Measurement of emission rates from
gas-fired space heaters Chicago, IL Gas Research Institute, report no GRI-86/0245
Zeldovich, J. (1946) The oxidation of nitrogen in combustion and explosions Acta Physicochim URSS
21 577-628
4-58
-------
5. TRANSPORT AND TRANSFORMATION OF
NITROGEN OXIDES
5.1 BACKGROUND
Even in the clean, unpolluted troposphere, nitrogen oxides (NOX) play an important role
in natural atmospheric processes They regulate the oxidizing power of the free troposphere
by controlling the buildup and fate of free radical species Consequently, the concentration
of NOX is a critical factor in determining ozone (O3) production and volatile organic
compound (VOC) chemistry under natural circumstances, and even more so under
circumstances of added anthropogenic emissions
Much of the troposphenc chemistry discussed in this chapter is based on the cyclic
reactions of nitrogen (Figure 5-1), oxygen (Figure 5-2), and hydrogen (Figure 5-3) These
are schematically combined in Figure 5-4 to show the complexity of interactions among these
three groups of gaseous compounds Collectively, these four diagrams, along with
Figure 5-5, provide a visual reference for the series of individual reactions discussed
throughout the chapter White and Deitz (1984) demonstrated the complexity of these
interactions and the calculated steady state concentrations of other species of gaseous
compounds in the troposphere as a function of NOX concentration (Figure 5-6)
In urban environments, the oxides of nitrogen react with VOCs in the presence of
sunlight to produce oxidants such as O3 and peroxyacetylmtrate (PAN) There are many
urban areas in the United States that are not in compliance with the National Ambient Air
Quality Standards for O3 Therefore, the involvement of NOX in this photochemical process
must be critically examined when considering O3 contiol strategies On a global basis,
increasing O3 levels in the troposphere are of concern because of the ability of O^ to absorb
outgoing radiation and, thus, contribute to the greenhouse warming of the earth's
atmosphere In recent years, acidic deposition has created a great deal of concern in the
eastern United States The oxides of nitrogen contribute to atmospheric acidity through their
conversion to nitric acid (HNO3)
5-1
-------
hv, M
Heterogeneous
loss
Figure 5-1. Summary of the gas phase chemistry of NOX in the clean troposphere.
Source Finalyson-Pitts and Pitts (1986)
On a global scale, the chemistry of NOX compounds is important because it plays a
major role in controlling free radical concentrations The buildup and fate of species such as
HO2', 'OH, and RO2 (where R is an organic moiety) is very dependent on the concentration
of nitric oxide (NO) that is present As will be discussed later in this chapter, the levels of
NOX present can regulate the oxidizing power of the free troposphere
5-2
-------
H20
02
rt^^"
OH
03
hv
O2+M
O(3P)
OH
NO\. j
X
(M)
HO2, CH3O2 + NO
\
Odd oxygen
Figure 5-2. Major chemical reactions affecting oxygen species in the troposphere.
Molecules acting as third bodies are demoted as *M' (e.g., N2, O2, Ar,
H20).
Source Finalyson-Pitts and Pitts (1986)
In summary, the atmospheric chemistry of the oxides of nitrogen has an important
impact on (1) the production of O3, (2) acidification in the atmosphere, and (3) control of
radical concentrations in the clean troposphere
In the ensuing portions of this chapter, the chemical mechanisms that relate to each of
these impacts will be described An important complement to the chemistry is the dispersion
and geographical movement of the oxides of nitrogen and their oxidation products This
chapter will address the factors that control the transport of these species within the
atmosphere
5.2 THE ROLE OF NITROGEN OXIDES IN OZONE PRODUCTION
Solar radiation triggers a series of reactions in the atmosphere between gaseous organic
molecules and NOX This chemistry involves a variety of unstable excited molecules and
5-3
-------
H2O2
f V\\Ji
' OH
H20
Heterogeneous
removal
Heterogeneous
removal
Figure 5-3. Major chemical reactions affecting hydrogen species (OH, H, H2O, H2O2)
in the troposphere.
Source- Finalyson-Pitts and Pitts (1986)
molecular fragments that lead to the production of secondary pollutants In urban plumes,
O3 is the predominant product of these reactions The amount of NOX present can determine
whether the photochemical reactions lead to the production or consumption of O3
Nitrogen oxides are emitted in most combustion processes, with NO being the major
constituent. Upon entering the ambient atmosphere, the NO can be oxidized quite rapidly in
the presence of O3 or in a photochemically reactive atmosphere to mtiogen dioxide (NO2)
The conversion of NO to NO2 can occur via reactions shown in Equations 5-1 and 5-2, as
follows:
5-4
-------
M
Figure 5-4. Schematic diagram of the combined reactions of nitrogen, oxygen, and
hydrogen.
Source Carmichael and Peters (1984)
2ND + O2 -» 2NO2
NO + O3 -* NO2 + O2
(5-1)
(5-2)
Reaction with molecular oxygen as the oxidizing agent is relatively slow, however, and
is only important in the immediate vicinity of a source where NO concentrations are
elevated Thus, in ambient air, Equation 5-1 is negligible At typical ambient levels of NO,
Equation 5-2, which destroys a molecule of O3, represents a more important mechanism by
which NO is converted to NO2 During the daytime, NO2 absorbs sunlight at wavelengths
less than 430 nm and decomposes to NO and a tnplet-P oxygen atom as represented by
Equation 5-3 The highly reactive oxygen atom forms O3 through collisions with oxygen
molecules In Equation 5-4, the M represents a third molecule (such as molecular nitrogen,
5-5
-------
o/ N>
O , _ CHSCH - CHCHS »• CH SCHO + CH ,CHO 2
CH3CH-CHCHa
3 3
I OH
CH,CH(OH)CHCHg \ o
CH»CH(OH)CH(CHS)OO % I \
[ NO ^ CH,CHCHCH,
CH,CH(OH)CH(CH,)0 + NO2
CH,CH(OH) + CH»CHO
HOj+CH.CHO CH»CH(OH)OO
I NO | NO
OH + N02 CH,CH(OH)0 + N02
CH,+ HCOOH
figure 5-5. Volatile organic compound oxidation in the atmosphere.
Source. Atkinson and Carter (1984)
molecular oxygen, etc ) that absorbs excess vibrational energy from the newly formed
O3 molecules.
NO2 + h? ^ NO + O(3P) (5-3)
O(3P) + O2 + M -> O3 + M (5-4)
The net effect of Equations 5-2 through 5-4 is an equilibrium in which NO, NO2, and
O3 concentrations are interdependent
NO2 + O2 + hv * NO + O3
In the absence of competing reactions, the NO, NO2, and O3 are expected to reach a
steady-state condition, with their concentrations defined by the following relationship
5-6
-------
Steady State
10
10°
NOx Concentration (ppb)
Figure 5-6. Calculated steady state concentrations in the free troposphere as a function
of nitrogen oxides concentration. Conditions described by White and Dietz
(1984).
Source White and Dietz (1984)
5-7
-------
JN02 [N02]
L 3j*s ~ T2—prop
where JNO2 is the photolysis rate for Equation 5-3 and #2 is the rate constant for
Equation 5-2.
Even in the cleanest daytime atmospheres, however, sufficient quantities oi HO2 and
various organic RO2' (where R is an organic moiety) radicals will be present that can
compete with O3 for converting NO to NO2 Thus, Equations 5-5 and 5-6 increase the
[NOJ/ENO] ratio, which leads to an increase in O3 levels
NO + HO 2 -* NO2 + OH (5-5)
NO + RO 2 -» NO2 + RO (5-6)
In these reactions, NO is converted to NO2 without destroying an O3 molecule, as occurs in
Equation 5-2; and consequently, O3 levels will build up in the atmosphere The peroxy
radicals in Equations 5-5 and 5-6 can be initially formed by photolysis of aldehydes and,
subsequently, from other reactions associated with the photoxidation of VOCs
These peroxy radical reactions oxidize NO to NO2 without destroying O3 In effect,
NO and NO2 are cycled catalytically, with the original NOX concentration remaining
essentially unchanged; but the O3 concentration builds up The amount of O3 formed is
dependent on the concentration of NOX present as well as the amounts and reactivity of
VOCs available. In urban plumes, O3 concentrations in excess of 200 ppb are not
uncommon in many areas of the United States However, in NOx-nch plumes, such as those
emanating from fossil-fuel burning power plants, O3 buildup is not observed until the NOX is
diluted with ambient air.
5.2.1 Urban Plume Chemistry
During the early morning hours, urban plumes typically contain a multitude of VOCs
and oxides of nitrogen (pnmanly NO) In the larger cities, during 0600 to 0900 hours,
nonmethane VOC (NMHC) levels generally average between 250 and 1,000 ppbC at the
5-8
-------
surface During the same period, NOX concentrations fall in the range 20 to 150 ppb, which
leads to typical VOC to NOX ratios of 7 to 15 Ozone levels in the early morning urban air
mass are near zero due to scavenging by NO (Equation 5-2) Because the reaction in
Equation 5-2 is very rapid, O3 concentrations cannot increase until most of the NO has been
converted to NO2 As indicated earlier, this will occui as sunlight energy becomes sufficient
to generate HO2 and RO2 Reactions such as Equations 5-3 and 5-4 then increase the NO2
to NO ratio
It is inappropriate in this document to discuss in detail the VOC chemistry involved in
oxidant formation Volatile organic compound reactions have been reviewed by many
authors (Atkinson, 1990, Atkinson and Carter, 1984) The VOCs serve as a source of free
radicals that propogate the series of chemical reactions leading to oxidant production The
process is initiated by the reaction of a VOC with an osudizing species present in the
atmosphere An example of this process is depicted in Figure 5-5
Of the three oxidizing species shown in Figure 5-5, the OH, radical is considered to be
by far the most important Several different production mechanisms for the OH radical have
been recognized Nitrous acid, which accumulates in urban areas during the nighttime hours
(Hams et al, 1982), will photolyze at sunrise, yielding OH radicals as illustrated in
Equation 5-7
HONO + h? -» OH f NO (5-7)
The photolysis of aldehydes will also lead to OH production Formaldehyde has been
shown to be present in significant quantities during the morning hours Photolysis of
formaldehyde generates OH radicals through the sequence shown below (Fmlayson-Pitts and
Pitts, 1986)
CH2O + ht> -* H + HCO (5-8)
H + O2 + M H» HO2 + M (5-9)
HCO + O2 -* HO2 + CO (5-10)
HO2 + NO ^ N02 + OH (5-11)
5-9
-------
The third source of OH radicals is via photolysis of O3 and subsequent reaction of the
excuted atomic oxygen atoms (O[ D]) that are produced with atmospheric watei vapor
O3 + by -* O(*D) + O2 (5-12)
O^D) + H2O -> 2 OH (5-13)
This source is probably not very important during the early hours of urban plume chemistry
because O3 levels are typically very low However, as the day progresses and O3 levels
build up, it certainly becomes more important especially in the free troposphere and in
pristine areas.
The amount of O3 produced in an urban plume is dependent on the absolute amounts of
VOCs, oxides of nitrogen, and sunlight In a particular urban area where the VOC to NOX
ratio remains relatively constant under conditions of high solar intensity, the greater the
concentrations of VOCs and NOX, the greater will be the production of O3 However, care
must be exercised when comparing oxidant production in different urban areas Simulations
of the atmospheric chemistry of O3 formation have shown that the VOC to NOX ratio can be
quite important in O3 formation At very low ratios where NO is present in relatively high
concentrations, O3 cannot build up because of scavenging by the NO (Equation 5-2) At the
high VOC to NOX ratios, there is insufficient NOX to propogate the radical reactions that lead
to O3 production. Urban VOC to NOX ratios are generally in the range of 5 to 15 In urban
areas with higher VOC to NOX ratios, NOX control will aid in reducing O3 production The
opposite is true when urban VOC to NOX ratios are low In this latter case, reducing NOX
emissions will likely result in increased O3 production
5.2.2 Ozone Production in Rural Environments
Ozone producing reactions identical to those described in urban plumes will occur in
rural environments, provided there are sufficient quantities of VOCs and oxides of nitrogen
present. Because the magnitude of natural VOC emissions is considerably larger than natural
NOX emissions, there are generally sufficient quantities of VOCs present with the result
being that the production of O3 is limited by the oxides of nitrogen Rural environments can
be characterized as follows, according to then NOX concentrations.
5-10
-------
(1) regions in which NOX levels average less than 50 ppt—this condition is generally
limited to clean maritime areas, most of f he free troposphere, and possibly some
remote continental locations in the southern hemisphere,
(2) areas with NOX concentrations ranging from about 50 to 200 ppt—clean
continental regions in the northern hemisphere comprise this category;
(3) semipolluted environments with average NOX concentrations ranging between
1 and 10 ppb—this condition is typical of much of the eastern United States and
probably rural areas in the industrialized countries of Europe and Asia
In the cleanest atmospheres, methane and carbon monoxide (CO) are the species present
in highest concentrations Consequently, they, along with the oxides of nitrogen, are
important participants in the photochemical reactions that control trace gas concentrations
Crutzen (1988) and Penkett (1991) have pointed out that O3 can be produced and destroyed
by the methane and CO oxidation cycles, depending on the concentrations of NO present
For example, the carbon monoxide oxidation cycles can proceed by either of the reaction
sequences shown below
CO + OH -* C02 + H (5-14a)
H + O2 + M -* HO2 + M (5-14b)
HO2 + NO ^ HO + NO2 (5-14c)
NO2 + ht> -* NO + O (5-14d)
O + O2 + M -» Oj + M (5-14e)
*Net CO + 202 -» CO2 + O3 (5-14f)
CO -H OH ^ CO2 + H (5-15a)
H + O2 + M -* H02 + M (5-15b)
HO2 + O3 ^ OH + 2O2 (5-15c)
*Net CO + O3 -* CO2 + O2 (5-15d)
Whether reaction series 5-14 or 5-15 predominates is dependent on the NO concentration
the rate constant for Equation 5-14c is about 5,000 times larger than that for Equation 5-15c
Thus, at [O3]/[NO] ratios less than 5,000, the O3 production sequence (5-14) is more
5-11
-------
important, whereas the O3 destruction sequence (5-15) predominates when NO levels are
extremely low At background O3 levels of 20 to 40 ppb, the 5,000 ratio corresponds to NO
levels of 10 ppt or less Similarly, for the oxidation of methane, the following net reactions
can be derived depending on the NO levels (Crutzen, 1988)
CH4 + 4O2 -> CH2O + H2O + 2O3 (5-16)
CH2O + 4O2 -» CO + 2 OH + 2O3 (5-17)
CH2O + 2O2 -» CO + H2O + O3 (5-18)
Equation 5-17 results from a series of reactions that are initiated by photodissociation of the
formaldehyde and subsequent oxidation of NO to NO2 via HO2 radicals Equation 5-18
involves nearly the same sequences of steps, but is initiated by reaction of the formaldehyde
with OH radicals. It is apparent from net Equations 5-16, 5-17, 5-18, and 5-14f that
oxidation of methane to CO and then to carbon dioxide (CO2) will yield a gain in
O3 molecules in environments with sufficient NO present
In NO-depleted environments, a loss of O3 is expected to occur due to a series of
reactions represented by the net Equations 5-19 to 5-21 and 5-15d
CH4 + OH + OH2 -> CH2O + 2H2O (5-19)
CH2O + 2O3 -» CO + 2O2 + 2 OH (5-20)
CH2O + O3 -» CO + H2O + O2 (5-21)
Thus, it is expected that in all but NO-deficient environments (NO < 10 ppt), O3 production
will occur during the daytime hours when conditions are conducive for photochemical
processes. In the context of this document, it is of greatest interest to study the
environments with adequate NO in order to see if there is a quantitative relationship between
NOX levels and the amount of O3 that is photochemically produced
Research studies at a mountain site in Colorado (Niwot Ridge) have provided the best
understanding of NOX-O3 relationships in rural areas (Parnsh et al, 1986) When winds are
from the west, the Niwot site is fumigated by clean air that is devoid of recent anthropogenic
emissions. Nitrogen dioxide levels in these westerly air masses are typical of those
5-12
-------
associated with clean continental environments (Category 2)—namely between 10 and
200 ppt However, the Niwot site is less than 100 Ion from the Denver-Boulder metropolitan
areas Occasionally, upslope winds advect polluted air that originates in this urban region to
the mountain site Under upslope flow conditions, NOX levels generally exceed 800 ppt and
range up to a few parts per billion Figures 5-7a anid 5-7b show the relationship between
low NOX levels and O3 at Niwot Ridge The two lines in each figure represent morning and
afternoon relationships The best fit lines during the morning hours include data collected
during the 0700 to 1100 hours tune periods, when the nocturnal inversion has burned off but
photochemical O3 production has not fully developed It is evident that there is little
dependence of O3 on NOX during the winter and only a slight dependence during the morning
hours in the summer The slope of summer-morning best fit line is 1 9 + 2 8 In contrast
to the morning behavior, a large dependence of O3 on NOX is seen in the afternoon
summertime data The slope of the least-squares fit is 16 8 ± 2 6. Thus, the daily
photochemical production of O3 amounts to about 17 ppbv O3/day/ppbv NOX
Kelly et al (1984) made similar measurements at sites in South Dakota, Virginia and
Louisana Their data provided a value of 6 ppbv O}/ppbv NOX for the daily photochemical
production of O3 This is considerably less than the value of 17 found at Niwot Ridge and
could be due to the fact that the Kelly et al data covered the time period between 1000 and
1400 hours, when the daily solar flux was not at its maximum Also, the average NOX levels
were higher (2 to 9 ppbv) at the three sites monitored by Kelly et al (1989). Based on
Niwot Ridge data, O3 production per unit NOX becomes smaller at NOX concentrations above
1 ppb Figure 5-8 shows the summertime O3 mixing ratio measured during the afternoon
hours (1500 to 2000 hours) versus the concurrently measured NOX mixing ratio Starting at
about 0 5 ppb NOX, O3 values exhibit a general inciease up to approximately 3 ppb Above
and below this range of NOX levels, there is no apparent dependence of O3 on NOX levels in
the ambient data from Niwot Ridge Recent photochemical modeling results of Liu et al
(1987) agree fairly well with the O3-NOX relationships derived from ambient data at Niwot
Ridge
Figure 5-9 compares calculated change in O3 values (Q) with ambient measurement
data (open circles with error bars) at the Colorado site The model included nonmethane
VOC chemistry, surface deposition of trace gases, and the dilution effect of trace gases due
5-13
-------
30
60
SO
40
30
(b)
WINTER
I
I
02 04 06
NOX (ppbv)
08
10
Figure 5-7. (a) Summertime (June 1 to August 31) and (b) wintertime (December 1 to
February 28) ozone mixing ratio versus nitrogen oxides mixing ratio during
the morning and afternoon. Filled circles are used for the morning values;
open circles are used for the afternoon values. Each point is an average
concentration of ozone for all nitrogen oxides concentration values in a
parts-per-billion-volume interval. The vertical error bars give the 95%
confidence limits for the average deduced from the standard deviation of
the measurements and the number of measurements in each nitrogen oxides
interval. The lines give the linear, least squares fit to the daia averages.
For clarity, the morning points and the linear fits have been offset
horizontally as indicated by the second abscissas.
Source. Pamsh et al. (1986)
5-14
-------
100
SO-
IL 60H
g 40
20 J
3:00-8:00 P.M.
'I,,'
11
!j|l|i'Hlh
•h
(Mi
I ill I I 1 I J T I T t T I IT |
0.01 0.1 1
I I I I T I
10
[NOJ (ppbv)
Figure 5-8. Summertime ozone mixing ratio versus nitrogen oxides mixing ratio
measured during the afternoon hours.
Source Parnsh et al (1986)
to changes in the daily inversion height The solid lines in Figure 5-9 represents the model
t i
calculated dependence of O3 on NOX Several scenarios were examined in the model, with
that represented by the NMHC-PO line being deemed the mbst appropriate for the Niwot
site Below 1 ppb NOX, the model overestimates the O3 buildup by a factor of 2 This is
suspected to result from an overestimation of odd hydrogen radical concentrations Although
the modeled results always exceed the measured values, the agreement becomes better at
higher NOX concentrations and the general shape of the calculated and measured curves are
very comparable An important feature of the net daily O3 change shown in Figure 5-9 is
the nonlinear relationship with NOX Both calculations and measurements indicate that
5-15
-------
120
100
80
S* 60
.a
&
a 40
20
0
-20
NMHC-FO
CO-CH4
345
NOX (ppbv)
6
8
Figure 5-9. Model calculated daytime change (Q) in ozone values (from sunrise to
1630 hours) for summer clear sky conditions is compared to the observed
difference between the afternoon (1400 to 1900 hours) and the morning
(0700 to 1100 hours) for clear sky conditions. Open circles with error
bars are ambient measurement data at the Colorado site. The dashed line
is calculated from a model without nonmethane VOCs. The shaded area
represents calculated values from a model with anthropogenic nonmethane
VOCs. The lower envelope of the shaded area is calculated by assuming
no overnight retention of secondary VOCs (NMHC-FO), whereas the
upper envelope assumes buildup of secondary VOCs to then* steady state
values (NMHC-FO).
Source Liuetal (1987)
O3 production increases more rapidly at low concentrations of NOX This is demonstrated in
more detail in Figure 5-10 The two curves in Figure 5-10 show the calculated average daily
03 production per unit concentration of NOX (AP) versus the NOX concentration for summer
and winter conditions The shape of the two curves is similar for the two seasons, however,
the summertime daily O3 production values are approximately a factor of 10 larger This is
ue to the higher photochemical activity in the summer The decline in daily O3 production
5-16
-------
50
5. 40
ox
I 30
jt
O* 20
a.
? 10
WINTER X
SUMME
10
10-2
10-1
100
101
102
Figure 5-10. Oxygen production per unit nitrogen oxides per day (AP) from the
NMHC-PO model is plotted as function of nitrogen oxides mixing ratios.
A constant nonmethane volatile organic compound to nitrogen oxides ratio
is assumed; see text for details. The solid line gives summer values. The
dashed line gives the winter values multiplied by 10.
Source Liu et al (1987)
rate at NOX concentrations larger than 1 ppb reported by Liu et al (i e , Figure 5-9) is
consistant with other modeling results Photochemical smog models suggest that the degree
of nonhneanty is a function of the ratio of NMHC to NOX and the relative abundance of
various VOCs
As pointed out by Liu et al (1987), the nonlinearity of the O3 production-NOx
dependence may have important implications for regional and global O3 budgets Clearly, as
atmospheric turbulence and advection dilute NOX emissions, the efficiency of O3 production
will be enhanced For the United States, Liu and co workers have estimated an average
summer column O3 production rate due to the reactions involving anthropogenic NOX and
NMHCs that is 20 tunes larger than the downward flux from the stratosphere If the
O3 production from natural NOX emissions is also considered, the proportion of
O3 production in the eastern and central regions of the United States that is associated with
human activities amounts up to 50 to 80% These findings are supported in a recent report
by Trainer et al (1987) that compared model predicted O3 buildup with observed values at a
5-17
-------
rural site in central Pennsylvania These authors concluded that photochemistry of NOX,
which is mainly of athropogenic origin, and isoprene from biogemc sources can lead to
elevated O3 levels
5.3 ODD NITROGEN SPECIES
Up to this point, there has been little discussion of the nitrogen containing products that
are produced from NOX in troposphenc photochemical reactions These oxidation products,
which are commonly referred to as odd nitrogen species, include HNO3, peroxymtnc acid
(HO2NO2), nitrous acid (HONO), peroxyacylmtrates (RC(O)O2NO2), dimtrogen pentoxide
(N2O5), nitrate radical (NO3), and organic nitrates
5.3.1 Nitric Acid
Nitric acid is a strong mineral acid that contributes to acidic deposition problems in the
United States According to recent estimates, HNO3 accounts for roughly one-third of the
total acidity deposited in the eastern United States (Calvert and Stockwell, 1983) In terms
of atmospheric photochemistry, HNO3 is a major sink for active nitrogen. During the
daytime hours, HNO3 is formed by the reaction of NO2 with the OH radical
NO2 + OH + M -> HN03 + M (5-22)
Equation 5-22 serves as a chain terminating step in the photochemistry that produces urban
smog This is a relatively fast reaction that can produce significant amounts of HNO3 over a
period of a few hours During nighttime, the heterogenous reaction between gaseous N2O5
and liquid water is thought to be a source of HNO3 The sequence of reactions that produce
N2C>5 and, subsequently, HNO3 are as follows
NO2 + O3 -* NO3 + O2 (5-23)
NO3 + NO2 -* N2O5 (5-24)
N2O5 + H2O -> HNO3 (5-25)
5-18
-------
This pathway to HNO3 is not viable in the daytime because the NO3 photolyzes rapidly
and, therefore, is not present in sufficient quantities 1o react with NO2 The NO3 will
abstract a hydrogen atom from VOCs, aldehydes, and mercaptans, which in theory provides
another nighttime source of HNO3
R-H
RC(O) -H + NO3 -» HNO3 + other products (5-26)
RS-H
The importance of this reaction pathway to HNO3 is not well understood at the present tune
(see NO3 discussion)
Logan (1983) has estimated a lifetime of 1 to 10 days for HNO3 in the lower
troposphere This variability results from the fact that the primary removal mechanism is
deposition Kelly (1987) indicates that HNO3 disappears faster than explained by dry
deposition, and suggests that HNO3 reactions with coarse soil particles and subsequent dry
deposition may be important The loss of HNO3 by rainout is subject to precipitation
frequency, whereas the dry depositional loss vanes with the surface cover and dispersion
characteristics within the boundary layer Chemical destruction mechanisms for HNO3 do
exist, however, their importance is not well understood or is suspected to be minor in the
lower troposphere For example, HNO3 can be destroyed through photolysis or reaction
with OH
HNO3 + OH ^ NO3 + H2O (5-27)
HNO3 + hv -» NO2 + OH (5-28)
Equations 5-27 and 5-28 are slow and, therefore, probably can't compete with the
depositional losses of HNO3 in the boundary layer Neutralization of HNO3 through reaction
with gaseous ammonia (NH3) is another potential sink for HNO3
HNO3 + NH3 -* NH4 NO3 (5-29)
5-19
-------
The importance of Equation 5-29 as a removal mechanism for HNO3 is not well understood
The interaction with NH3 has been reported to influence surface fluxes of HNO3 (Huebert
et al., 1988) Furthermore, it has proven difficult to separate the deposition of aerosol
nitrate and ammonium ion from deposition due to HNO3, NO2, and NH3 (Hanson and
Lindberg, 1991)
5.3.2 Nitrous Acid
Little is known about the distribution and concentration of HONO in various ambient
atmospheres. There have been a few measurements in urban environments (Harris et al,
1982). During the daytime, HONO levels are expected to be low because it photolyzes
rapidly.
HONO + hv -» HO + NO (5-30)
This reaction likely serves as a source of OH radicals during the morning in urban regions
where HONO may accumulate during the nighttime hours The most likely production
mechanisms for HONO include
HO + NO + M -> HONO + M (5-31)
NO + NO2 + H2O -> 2HONO (5-32)
Equation 5-31 will only lead to a buildup of HONO during the late afternoon and evening
hours, when sunlight intensities are low but some OH radicals are still present
Equation 5-32 can produce HONO throughout the nighttime hours The reaction probably
involves a multistep sequence analogous to Equations 5-23 and 5-24 that produce HNO3
Dinitrogen trioxide is the anhydride of HONO and reacts with liquid water to form the acid
5.3.3 Peroxynitric Acid
Although this oxidized form of HNO3 has never been measured, it is expected to be
present in the upper troposphere Models suggest concentrations in the 10 to 100 ppt range
at altitudes above 6 km (Singh, 1987, Logan, 1983) Peroxynitric acid is thermally unstable,
5-20
-------
consequently, boundary-layer concentrations are expected to be extremely low (< 1 ppt)
Peroxymtac acid is formed through the combination of a HO2 radical with NO2
HO2 + NO2 + M ^ HO2NO2 + M (5-33)
In the upper troposphere, HO2NO2 is destroyed by photolysis or by reaction with OH
HO2NO2 + ht> -* HO2 + NO2 (5-34)
HO2NO2 + OH ^ products (5-35)
5.3.4 Peroxyacylnitrates
Peroxyacetylmtrate is the most abundant of this family of nitrates The next higher
homolog, peroxypropionyl nitrate (PPN), is generally less than 10% of the PAN
concentration, with higher molecular weight species such as peroxybenzoyl nitrate expected
to be present at even lower levels Peroxyacetylmtrate is the only member of this family of
compounds that has been extensively studied It is a strong oxidant and, therefore, can have
adverse effects on human health and can cause plant damage if ambient concentrations
become high enough Of greatest interest to this chapter is the role PAN plays in
atmospheric chemistry Based on its primary means of formation,
CH3C(O)OO + NO2 -* CH3C(O)O2NO2, (5-36)
it might be expected that PAN would serve as a sink for oxides of nitrogen This is not true,
however, because PAN is thermally unstable and is much more likely to produce NO2
through the reverse of Equation 5-36 than to be removed by depositional process If the
lifetime of PAN was determined by its thermal decomposition, the lifetime would be
approximately 1 h at 25 °C, 2 days at 0 °C, 5 mo al -23 °C, and 42 years at -43 °C
Based on these estimated lifetimes, it was suggested that PAN could be the principal form of
reactive nitrogen in the upper troposphere (Singh and Hanst, 1981) In reality, because PAN
reacts with OH radical and photolyzes, its mean lifetime cannot exceed 3 mo
5-21
-------
CH3C(O)O2NO2 + hv -» products (5-37)
CH3C(O)O2NO2 + OH ^ products (5-38)
Equations 5-37 and 5-38 are slow compared to the thermal degradation of PAN at
temperatures above 0 °C and, therefore, are unimportant in determining the summertime
boundary-layer fate of this reactive nitrogen compound
There is no question that PAN can serve as a storage reservoir for NO2 In the
summertime boundary layer, PAN concentrations will be decreased somewhat by dry
depositional losses over land (deposition velocity of approximately 0 25 cm/s), but it is very
likely that a significant fraction of the PAN produced in urban plumes can be transported into
the regional environment For example, PAN lifetimes of about 5 and 20 h have been
calculated at 20 and 10 °C, respectively In transport layers above a nighttime surface
inversion, PAN could be transported several hundred kilometers Even during transport
conditions where mixing occurs down to the surface, PAN is expected to persist because it is
continually being produced
5.3.5 Nitrate Radical
The NO3 radical is a short-lived oxide of nitrogen that is formed by the reaction of
NO2 with O3
NO2 + O3 ^ N03 + O2 (5-39)
Other sources of NO3 exist (Wayne et al, 1991), however, Equations 5-3 to 5-39 serve
as the primary troposphenc production mechanism for NO3 Photolysis of NO3 is rapid,
resulting in a lifetime of about 5 s at midday Furthermore, NO3 reacts rapidly with NO,
which limits its lifetime both during the daylight and nighttime hours At NO concentrations
of 320 pptv, the lifetime of NO3 due to reaction with NO is similar to that for photolysis
(«*5 s) Thus, in urban regions where NO concentrations normally exceed 300 pptv, the
reaction with NO will control the NO3 lifetime
At night, NO3 concentrations on the order of 0 3 ppt have been measured in clean
tropospheric air, with recorded levels ranging up to 430 ppt in urban areas (Wayne et al,
5-22
-------
1991) After sunset, the buildup of NO3 is expected to be controlled by the availability of
NO2 and O3 plus the main chemical destruction mechanisms
NO3 + NO2 -> NO + NO2 + O2 (5-40)
NO2 + NO3 -* N2O5 (5-24)
N2O5 + H2O -* 2HNO3 (5-25)
In clean background environments, it has been reported that measured NO3 levels are
significantly less than predicted from a consideration of Equations 5-39, 5-40, 5-24, and 5-25
alone This implies that some additional loss mechanism must be occurring Speculation has
centered around four different loss processes (Platt et al , 1981, Noxon et al, 1980)
(1) heteorogenous losses of NO3 and/or N2O5 on particle surfaces,
(2) reactions with water vapor,
(3) reaction of NO3 with NO, and
(4) NO3 reaction with organic compounds
Based on modeling results, Heikes and Thompson (1983) suggest that the low NO3
concentrations could result from the reaction of NO3 with NO, provided sufficient quantities
of NO are present at night In the absence of NO, heterogenous loss of NO3 and N2O5
could account for lower than expected NO3 levels, provided that their sticking coefficients
3
are greater than 10"
Wayne et al (1991) have studied the nighttime chemistry of NO3 and conclude that
simple analyses are useful, but generally insufficient for interpreting NO3 behavior They
suggest that a numerical simulation is required in order to accurately assess NO3 observations
in individual data sets Perner et al (1991) have conducted such a modeling exercise, and
their results suggest that nighttime concentrations of NO3 can be reduced by the presence of
naturally emitted monoterpene Thus, in regions when reactive organic compounds are
present in nighttime air masses, lower than anticipated NO3 concentrations may be due to
scavenging by organic species
5-23
-------
Reactions of NO3 with organic species have garnered considerable interest in recent
years. Kinetic studies have shown NO3 to be very reactive toward a variety of organic
compounds. For example, at NO3 levels of approximately 100 ppt, the lifetime of
monoterpene VOCs due to NO3 oxidation will be less than 10 min This, then, might be an
important nighttime sink for both biogemc VOCs and the NO3 radical (Winer et al, 1984)
At this time, little is known about the mechanisms or products that result from the reaction
of organics with NO3. It is expected that hydrogen abstraction reactions will yield HNO3
and organic radicals.
R-H + NO3 -* HNO3 + RO2 (5-26)
RCHO + NO3 -> HNO3 + RCO (5-26)
With olefinic or aromatic VOCs, organic nitrates are expected to predominate
R-CH = CH2 + NO3 -> RCH-CH2ONO2 -* products (5-41)
Ar-OH + NO3 -* ArO + HNO3 -> Ar(OH)NO2 (5-42)
5.3.6 Dinitrogen Pentoxide
Dmitrogen pentoxide is the anhydride of HNO3 As indicated in the previous section,
it is formed from NO3 and NO2 Because NO3 is present only at night, N2O5 is primarily a
nighttime species as well and is thermally unstable, decomposing to NO3 and NO2 At high
altitudes in the troposphere, where temperatures are low, N2O5 can act as a temporary
reservoir for NO3 Dinitrogen pentoxide photolyzes at wavelengths less than 330 mm to
give, once again, NO3 and NO2 This provides the major source of NO3 production in the
stratosphere
Dinitrogen pentoxide reacts heterogeneously with water on the surface of hydrometers
to form HNO3. This serves as the main nighttime production mechanism for HNO3, and
because HNO3 is readily deposited by dry and wet deposition, it provides an important
mechanism for removal of NOX from the atmosphere The importance of the gas-phase
reaction between N2O5 and water vapor is not well understood (Logan, 1983)
5-24
-------
N2O5 + H2O(g) -* 2HNO3 (5-43)
An upper limit of 13 to 21 cm /molecule/s has been reported for the rate coefficient
(Tuazon et al, 1983) Wayne et al (1991) has pointed out that even with this low rate
coefficient, the gas-phase reaction of N2O5 and water vapor could contribute significantly to
the atmospheric formation of HNO3 However, until the rate constant is established with
greater certainty, the prominence of this reaction as a source of HNO3 will remain obscure
There are a few reports of N2O5 reacting with aromatic VOCs such as naphthalene and
pyrene (Pitts et al , 1985, Atkinson et al , 1986), and mtroarenes appear to be the product of
the reaction
5.3.7 Total Reactive Odd Nitrogen Species
The total reactive odd nitrogen species are refeiTed to as NOy It is expected that NO,
NO2, PAN, and HNO3 comprise the bulk of the NOy present in the ambient atmosphere
This has been tested by measuring these individual compounds at the same tune as a total
NOy measurement is performed (Fahey et al, 1986) The sum of the individual species
should equal the NOy concentration if NO, NO2, HNO3, and PAN are the only nitrogen
compounds present Figure 5-11 shows a plot of the (NOy)j/NOy ratio versus NOy, (NOy)x
is the sum of measured odd nitrogen compounds A ratio of 1 0 implies that the sum of the
individual species is equal to the total NOy These all were summertime measurements at
Point Arena on the California coast, Niwot Ridge in Colorado, and Scotia Range in central
Pennsylvania Both at the Point Arena and Niwot sites, there is a significant odd nitrogen
shortfall Approximately 45 % of the odd nitrogen species are unaccounted for at the Niwot
site During the winter months, the (NOy)t shortfall is not nearly as large («5%), which
implies that the unknown component is most likely photochemically produced Organic
nitrates, in particular methyl nitrate and higher homologs of PAN, have been suggested as
the missing component Calvert and Madromch (1987) have recently reported that organic
nitrates should be important products of photochemistry Although the importance of the
organic nitrates is recognized, evidence for their existence in the atmosphere is sparse and
often only circumstantial Therefore, it is unknown whether or not they constitute the
missing NOy fraction at Niwot Ridge and elsewhere
5-25
-------
1.2
1.0
08
O* 0.6
0.4
0.2
0
O Point Arena
D Niwot Ridge
• Scotia Range
I I I I I I I I
I I I I I
102
103
N0y (pptv)
104
figure 5-11. Total reactive odd nitrogen species shortfall.
Source Fahey et al (1986)
5.3.8 Amines, Nitrosamines, and Nitramines
The concentration of amines in the atmosphere is thought to be low, although there are
few data available to confirm this hypothesis The highest concentrations would be expected
in the vicinity of various sources (e g , cattle feed lots, sewage treatment facilities, waste
incinerators, industrial plants) that utilize or produce amines
Nitrosamines and nitramines are produced in the troposphere through photochemical
reactions involving alkyl amines and the oxides of nitrogen (NO and NO2) Because both
nitrosamines and nitramines have proven to be carcinogenic in animals, considerable interest
has centered around the troposphenc sources and distribution of these organomtrogen
compounds. In the proceeding NOX Criteria Document (U S Environmental Protection
Agency, 1982), three chemical mechanisms were described for the formation of nitrosamines
in the atmosphere
5-26
-------
(1) reaction of gaseous amines with NOX and HONO (nonphotochemical),
(2) photochemical reactions of amines with NOX ID the gas phase, and
(3) heterogenous formation processes involving atmospheric aerosols
All three of these pathways involve reactions of amines with the oxides of nitrogen
and/or HNO3 The dark reaction (nonphotochemical) and heterogenous mechanism are
poorly understood In the former case, there are conflicting reports concerning reaction rates
and yields of mtrosamine products formed It is quite likely that at least part of the dark
Equations 5-32 and 5-44 that are believed to convert amines to nitrosamines occur on the
walls of chambers used to study these processes (Finlayson-Pitts and Pitts, 1986)
NO + NO2 + H2O -> 2HONO (5-32)
R2NH + HONO -» R2NNO + H2O (5-44)
Therefore, caution must be exercised when extrapolating laboratory smog chamber studies to
the real atmosphere Laboratory studies conducted in the late 1970s (Hanst et al, 1977,
Grosjean et al., 1978, Pitts et al, 1978) indicated low yields (3%) of nitrosamine production
in the dark If this is true, boundary layer concentrations of nitrosamines should be low
during the nighttime hours
Because there do not appear to be any recent studies that clarify the nonphotochemical
conversion of amines to nitrosamines, the reader is referred to the 1982 NOX Criteria
Document for a more thorough discussion of this subject
As is the case with most heterogenous chemical transformations in the atmosphere,
nitrosation in aerosols is highly speculative The absorption of basic amines by acidic
aerosol droplets followed by reaction with nitrite, HONO, or other species could,
theoretically, lead to the formation of nitrosamines (U S Environmental Protection Agency,
1976) Whether or not the nitrosamines so produced could withstand photodecomposition
and/or further oxidation is unknown
5-27
-------
There is good evidence that photolysis of gaseous amines in the presence of NOX will
produce nitrosamines and nitramines Figure 5-12 shows the concentration tune profiles for
diethylmtrosamine in photooxidation experiments involving diethyl- and tnethylamuie in the
presence of NO and NO2 Diethylmtrosamine appears shortly after the reactants are mixed
in a darkened chamber This is presumably due to the dark Equations 5-32 and 5-44
discussed previously The diethymitrosamine that formed in the dark rapidly decayed,
although there was clearly additional generation of this compound from tnethylamuie
followed by its decomposition after continued photolysis Pitts et al (1978) derived the
following sequence of reactions to explain the photochemical transformations
(C2H5)3N + OH -» (C2H5)2NCHCH3 + H2O
* O2
I (NO -* NO2)
HO2 + (C2H5)2NC(O)-CH3 <- (C2H5)2NC(O)HCH3 •* (C2H5)2NCHO + CH3
02
NO2
CH3CHO + (C2H5)2N -> (C2H5)2NNO2
1, NO
(C2H5)2NNO
In the case of diethylamine, hydrogen abstraction can occur from the nitrogen as well This
probably accounts for the significantly higher yield of nitramines observed from secondary
amines compared to tertiary amines
R2NH + OH -» R2N + H2O (5-45)
R2N + NO2 -> R2NNO2 (5-46)
5-28
-------
60
.1
€ 20
DFrom(C2H5)2NH
OFrom(C2H5)3N
2
Time (h)
Dark
Sunlight
Figure 5-12. Formation and decay of diethylnitrosaniine in the dark and in the sunlight
from diethylamine (open squares) and from triethylamine (open circles).
Source U S Environmental Protection Agency (1982)
In the case of diethylamine, hydrogen abstraction can occur from the nitrogen as well This
probably accounts for the significantly higher yield of mtramines observed from secondary
amines compared to tertiary amines.
The mtrosammes and nitramines have limited lifetimes in the atmosphere due to
photolytic decomposition and/or reactions with OH radical and O3 Nitrosamines absorb
light in the ultraviolet region (325 to 375 mm) efficiently and are rapidly photolyzed
Tuazon et al (1984) estimated that dimethylnitrosamine has a half-life of about 5 mm at
Los Angeles latitudes during the midsummer daytime Photolysis will control the fate of
mtrosammes in the troposphere because the reactions with OH and O3 are relatively slow
(Tuazon et al, 1984) The lifetime of dimethylnitrosamine due to reaction with OH
ft -2
([ OH] = 1 x 10 cm") has been estimated to be 4 days If the reaction with O3 was the
lifetime determining process at 100 ppb O3, the lifetime of dimethylnitrosamine would
exceed 1 year
5-29
-------
The situation is somewhat different with mtramines They have low light absorption
cross sections and do not photolyze readily Nitramines react very slowly with O3 (lifetime
of dimethylnitramine »4 years) but will be removed from the atmosphere through reaction
with OH radical For example, the estimated lifetime for dimethylnitramine is about 3 days
at an OH concentration of 1 x 106 cm"3 (Tuazon et al, 1984) Little is known about the
mechanism or products formed in the reaction of mtramines with OH
5.4 TRANSPORT
The transport and dispersion of the various odd nitrogen species are dependent on both
meteorological and chemical parameters Advection, diffusion, deposition, and chemical
transformations combine to dictate the atmospheric residence tune of a particular trace gas
Nitrogenous species that undergo slow chemical changes in the troposphere, and are not
readily removed by depositional processes, can have atmosphenc lifetimes of several months
Gases with lifetimes on the order of months can be dispersed over continental scales and
possibly even over an entire hemisphere At the other extreme are gases that undergo rapid
chemical transformation and/or depositional losses that limit their atmosphenc residence
times to a few hours or less Dispersion of these short-lived species may be limited to only a
few kilometers from their point of emission
Surface emissions are dispersed vertically and horizontally through the atmosphere by
turbulent mixing processes that are dependent to a large extent on the vertical temperature
structure and wind speed On the vertical scale, transport can occur in three separate layers
(1) The daytime and/or nighttime mixed layer—this layer can extend from the surface
up to a few hundred meters at night or several thousand meters during the daytime
(2) A layer that exists during the nighttime above a low level surface inversion and
below the daytime mixing height—this layer will generally fall in the 200 to
2,000 m altitude (above ground level) band
(3) The free troposphere—this transport zone is above the boundary layer mixing
region
5-30
-------
During the warm, summertime period when the impact of reactive nitrogen species is
the greatest, vertical mixing follows a fairly predictable diurnal cycle A surface inversion
normally develops during the evening hours and persists throughout the nighttime and
morning period, until broken by surface heating While the inversion is in place, surface
NOX emissions can lead to relatively high, local concentrations because of the restricted
vertical dispersion Following the breakup of a nighttime surface inversion, vertical mixing
will increase and surface-based emissions will disperse to higher altitudes The extent of
vertical mixing during the daytime is often controlled by synoptic weather features. Elevated
temperature inversions associated with high pressure systems are common in many parts of
the United States An elevated inversion in the Los Angeles Basin traps pollutants in the
lower 600 m of the atmosphere In the midwestern and northeastern regions of the United
States, summertime afternoon mixing levels normally range between 1,600 and 1,800 m
(Holzworth, 1967) Horizontal dispersion of trace gases in the mixing layer is caused by
horizontal turbulence and vertical wind shear For mesoscale and synoptic scale transport,
mean wind shear is the dominant cause for dispersion
The dispersion processes described above coupled with chemical transformations of a
particular reactive nitrogen compound dictate transport distances in the troposphere
A reasonable understanding exists concerning the short-term (daylight hours) fate of NOX
emitted in urban areas during the morning hours As described in detail in Section 521,
NOX emitted in the early morning hours in an urban area will disperse vertically and move
downwind as the day progresses On sunny summer days, most of the NOX will have been
converted to HNO3 and PAN by sunset Much of the HNO3 will be removed by depositional
processes as the air mass moves along After dusk, an upper portion of the daytime mixed
layer will be decoupled from the surface due to formation of a low-level radiation inversion
Transport will continue m this upper level during the nighttime hours and although
photochemical processes will cease, other dark phase chemical reactions can proceed There
are no reports of plume measurement studies that have tracked plumes for more than one
daylight period Thus, nothing is known concerning the fate of the remaining nitrogenous
species that become entrapped in the layer above the nighttime surface inversion and below a
higher subsidence inversion Peroxyacetylnitrate and HNO3, if carried along in this layer,
could be transported long distances
5-31
-------
5.4.1 Transport of Reactive Nitrogen Species in Urban Plumes
The most extensive studies of the fate of nitrogen species in urban plumes have been
reported by Spicer and co-workers They have examined the behavior of reactive nitrogen
compounds in plumes emanating from Los Angeles, CA (Spicer et al, 1979), Phoenix, AZ
(Spicer et al, 1978); Boston, MA (Spicer, 1982a), and Philadelphia, PA (Spicer and
Sverdrup, 1981). The nitrogen budget denved from Boston plume measurements provides a
good example of the fate of reactive nitrogen compounds in urban plumes Nitrogen oxides
concentrations at short distances from Boston varied from about 30 to 130 ppb After travel
times of 4 to 7 h, NOX concentrations in the plume were in the 5- to 10-ppb range Removal
rates of NOX were calculated from measured NOX, PAN, and HNO3 concentrations
Dilution effects wereiaccounted for by using tracers of opportunity such as CO, acetylene,
and tnchlorofluromethane The plume was monitored by aircraft to distances as far as
150 km east of Boston This corresponded to reaction tunes of as much as 7 5 h The NOX
removal rate ranged from 0 14 to 0 24 h" for four different days This corresponds to NOX
lifetimes (I/removal rate) of 4 2Jx> 7 1 h (Altshuller, 1986) These lifetimes apply to sunny,
summertime, moderately polluted plume conditions with transport mainly over water
Nitrogen oxides depositional losses during over-water transport should be very small In the
Boston plume, the chemical loss of NOX is equal to the appearance of nitrate pioducts
(HNO3, PAN, and NO3) On an August 18, 1978, flight, an overall NOX loss rate of
0.24 h" was obtained During the same measurement period, a value of 0 23 h" was
calculated for the conversion rate of NOX to nitrate products
Somewhat lower NOX loss rates have been reported from data collected in Los Angeles
(Chang et al , 1979). Chang and coworkers denved a value of 0 04 h as a lower limit for
the yearly average daytime NOX removal rate Calvert (1976) estimated the NOX removal
rate to be approximately 0 09 h" during the midmormng to early afternoon hours It has
been suggested that the values denved from Los Angeles data probably repiesent only a
portion of the true NOX loss rate because of NOX measurement interfeiences by HNO3 and
PAN. However, the 0 09 h"1 value denved by Calvert agrees well with recent estimates in
the Detroit metropolitan area (Kelly, 1987) Using a combination of captive outdoor
irradiation experiments, photochemical modeling, and ambient measurements, Kelly obtained
an NOX removal rate of approximately 0 1 h" It was determined that HNO3 accounted for
5-32
-------
67 to 84% of the nitrogen-containing products This led to the conclusion that HNO3
formation will control the chemical lifetime of NOX on photochemically active days
Based on a calculated formation rate of HNO3, Kelly was able to estimate the expected
ambient HNO3 concentrations in the Detroit plume The predicted concentrations always
exceeded the measured ambient HNO3 concentrations by a significant amount (3 to 4 tunes)
In order to reconcile this difference, Kelly hypothesized that once formed, HNO3 was rapidly
removed from the urban plume Several removal mechanisms were considered, with
incorporation into coarse atmospheric aerosol deemed to be most likely Based on Kelly's
analysis, approximately 70% of the gaseous NOX in Detroit's morning atmosphere is
photochemically converted to products by sundown In the absence of sinks, the product
distribution would be «75% HNO3, «20% PAN, and the remaining 5% as other organic
nitrates However, removal of HNO3 as coarse nitrate leads to a maximum HNO3
concentration that is only 20 to 30% of that expected in the absence of the aerosol sink
Because the coarse nitrate aerosol is quickly removed by sedimentation, the majority of the
nitrogen-containing species emitted and produced in the Detroit urban area are not
transported long distances downwind
It should be kept in mind that the studies just described in Boston, Los Angeles, and
Detroit addressed only the daytime fate of reactive nitrogen species The nighttime
chemistry of odd nitrogen compounds in urban environs is poorly understood Nighttime
emissions of NO will react with O3 to produce NO2 as long as there is sufficient O3 present
Often the O3 reservior that exists aloft over urban areas is decoupled from the surface layer
at night by a low-level radiation inversion Under these conditions, the O3 supply is not
replenished, and once it has been used up, NO will no longer be oxidized to NO2 via
reaction with O3 As described earlier, during nighttime periods when O3 is present, it can
react with NO2 to form NO3 The NO3 radical is very reactive, consequently, its
concentration remains low (»1 to 500 ppt) It reacts rapidly with NO so as the night
progresses and NO levels increase, NO3 concentrations will fall
/
NO3 + NO -* 2NO2 (5-47)
5-33
-------
Other sinks for NO3 include the reactions with organic species (Equations 5-26, 5-41, and
5-42) and NO2, which produces N2O5 (Equation 5-24) The N2O5 will react with water to
form HNO3 (Equation 5-25). It can also react with VOCs, which may produce significant
quantities of mtro-substituted polycyclic aromatics (Pitts, 1987)
Chemical models have been employed to try to sort out the nighttime NOX chemistry
(Stockwell and Calvert, 1983, Jones and Seinfeld, 1983, Russell et al, 1985) Russell et al
(1985) were able to get reasonable agreement between predicted and measured time profiles
of NO3, NO2, and O3 during nighttime hours in the Los Angeles Basin The reactions of
importance to nighttime chemistry and the rate constants as employed in the Russell et al
model are listed in Table 5-1 In order to accurately model this nighttime NOX chemistry,
atmospheric concentrations of NO, NO2, NO3, O3, water vapor, and organic species must be
known. In addition, meteorological factors (such as mixing conditions, temperature, etc ) are
important An entire data set such as this is not currently available for modeling purposes
Consequently, concentrations must be estimated for unmeasured species From the rate
constants shown in Table 5-1, it is obvious that NO can play a very critical role in nighttime
chemistry. Nitric oxide very rapidly scavenges NO3 For example, Russell et al calculated
an NO3 concentration of 12 ppt m the Los Angeles surface layer when 1 ppb NO is present
and over 200 ppt NO3 when NO is negligible Dimtrogen pentoxide is another species that
can significantly influence nighttime chemistry The magnitudes of the various N2©5 loss
processes (see Table 5-1) are not well understood Due to the transient character of N2O5, it
has been difficult to determine the homogenous gas-phase reaction rate constant with water
vapor, the deposition velocity, and heterogenous interactions with ambient particulate matter
5.4.2 Transport and Chemistry in Combustion Plumes
Interest ui the NOX chemistry of power plant plumes increased significantly following
the 1974 report by Davis et al that O3 could be generated through a series of reactions
involving sulfur and nitrogen constituents within this type of plume In subsequent studies, it
was pointed out that the well-known photochemical reactions involving oxides of nitrogen
and VOCs are the more likely mechanisms of O3 buildup in power-plant plumes (Miller
et al., 1978). Because the VOC to NOX ratio in these plumes is very low, the VOCs must
5-34
-------
TABLE 5-1. MAJOR REACTIONS IN THE NITRATE RADICAL-DINITROGEN
PENTOXTOE SYSTEM AT NIGHT
Reaction
NO2 + O3
NO + NO3
N02 + N03
N2°5
N2O5 + H2O
N03 + HCHO
N03 + RCHO
N03 + OLE
N02 + N03
N2©5 + aerosol
NO3 + aerosol
7
— >
8
— >
44
— >
45
— >
46
— >
53
— >
54
— >
56
— >
57
— >
— >
— >
NO3 + O2
2NO2
N2O5
NO2 + NO3
2HNO3
HNO3 + HO2 + CO
RC03 + HN03
RPN
NO + NO2 + O2
2HN03
aerosol
Bate Constant 298 K
(ppm mm units)
k7 = 0 05
k8 = 29560
k44 = 2510
k45 = 29
k4(5 = 1 9 X 10-6
ks, = 0 86
k54 = 36
k5(5 = 12 4
k5/ = 0 59
k*
N205
k*
N03
Reference and
Comments*
1
1
2
1,3
4
5
5,6
5, 7, 8, 9
10
11
11
a(l) Baulch et al (1982), (2) Tuazon et al (1984), (3) Malko and Troe (1982), (4) Tuazon et al (1983),
(5) Atkinson and Lloyd (1984), (6) the rate constant used for the NO3 reaction with high aldehydes is that
measured for acetaldehyde, (7) the value used for the rate constant of the NO3 reaction with olefins is that
measured for the nitrate radical reaction with propene, (8) the ultimate products of reaction (56) are reported to
be mtroxyperoxyalkyl nitrates and dinitrates (Bandow et al , 1980), (9) Bandow et al (1980), (10) Atkinson
and Lloyd (1984), (11) Russell et al (1985)
Source Russell et al (1985)
be mixed into the power-plant plume as it moves downwind Excess O3 concentrations of
20 to 50 ppb above ambient have been reported in plumes after several hours of downwind
transport However, an O3 buildup is not found in all power-plant plumes (Hegg et al,
1977, Ogren et al, 1977, White, 1977) The single most important ingredient appears to be
the availability of reactive VOCs in the dilution air Aircraft measurements have shown little
enhancement of inorganic and particulate nitrate concentrations in power-plant plumes (Hegg
and Hobbs, 1979)
5-35
-------
Due to the difficulties associated with tracking and making accurate measurements in a
narrow power-plant plume, little information is available concerning the fate of reactive
nitrogen species in these plumes Hegg et al (1977) have reported measurements in four
different NOx-rich power-plant plumes Out to distances of 90 km and travel times up to
4 h, no O3 enhancements were observed in any of these plumes Measured NO2/NO ratios
in these plumes were generally in the 1 to 2 range This is considerably below the value of
10, which has been shown in laboratory experiments to be the minimum ratio at which
appreciable O3 generation can occur As a consequence, smog chamber and modeling
studies have been employed to study NOX transformation rates Smog chamber experiments
under a variety of conditions expected in real atmospheres yield NOX lifetimes varying from
about 1 to 7 h (Spicer et al, 1981) The fastest conversion times were observed when VOC
to NOX ratios were high ( = 15) and the VOC mix included those species typically found in
urban atmospheres Nitric acid and PAN were the major nitrogen-containing products
observed in the chamber reactions Generally, they accounted for between 70 and 90 % of
the nitrogenous species present at the end of the final irradiation period The PAN/HNO3
ratio varied depending on the initial VOC to NOX ratio Less PAN was produced in cases
where organic levels were initially low
The smog chamber studies imply that the NOX lifetime in a power-plant plume can vary
from a few hours to more than a day, depending on environmental conditions Under
conditions of low VOC levels (e g , rural areas or aloft above a surface inversion), the NOX
lifetime will be sufficiently long to allow NOX input to regional air masses
5.4.3 Regional Transport
Transport of reactive nitrogen species in regional air masses can involve several
mechanisms Mesoscale phenomena such as land-sea breeze circulations or mountain-valley
wind flows will transport pollutants over distances of tens to hundreds of kilometers On a
larger scale, synoptic weather systems such as the migratory highs that cross the eastern
United States in the summertime influence air quality over many hundreds of kilometers
The accumulation and fate of nitrogen compounds will differ somewhat between the
mesoscale and synoptic systems Mountain-valley and land-water transport mechanisms have
dual temporal scales due to their dependence on solar heating However, in the larger scale
5-36
-------
synoptic systems, reactive nitrogen species can build up over multiday periods The
residence tune of air parcels within a slow moving high-pressure system can be as long as
6 days (Vukovich et al , 1977)
In many cases, the transport mechanisms mentioned above are interrelated For
example, slow-moving high-pressure systems that migrate across the eastern United States
are characterized by weak pressure gradients Thus, mountain-valley or land-water breezes
can dictate pollutant transport in the immediate vicinity of sources, but the eventual fate of
reactive nitrogen species will be distribution into the synoptic system
Combined studies of air quality and meteorology .along the western shore of Lake
Michigan have clearly documented this relationship (Lyons and Cole, 1976, Westberg et al,
1981) Data shown in Table 5-2 and Figure 5-13 were collected near Kenosha, WI, during
the period August 14 to 22, 1976 During this tune, a strong, slow-moving high-pressure
system traveled across the Great Lakes Region Degradation of air quality in southeastern
Wisconsin was clearly associated with both synoptic transport and mesoscale (lake breeze)
advection during this 9-day period August 14 was the first day during which the effects of
the advancing high-pressure system were observed in southeastern Wisconsin Northerly
flow associated with the leading edge of the anticyclone persisted through August 16 As can
be seen in Table 5-2, pollutant levels were low during lhat period with northerly winds.
From August 18 to 22, however, meteorology along the western shore of Lake Michigan was
controlled by synoptic features characteristic of the trailing edge of an anticyclone and the
local lake-breeze phenomenon Thus, gradient winds were from the southwest, but during
the afternoon hours, a shift to southeasterly flow occurred as the lake-breeze front moved
inland Figure 5-13 shows pollutant profiles recorded about 5 mi inland on the afternoons of
August 18 and 19 Pollutant levels increased dramatically following passage of the
lake-breeze front on both of these days These high pollutant levels were most likely the
result of emissions from the Chicago-Hammond-Gary urban complex During the night and
early morning hours, the plume from this industrial region drifts in a northerly direction over
the lake Morning sunlight serves to initiate photochemical processes in the contaminated air
mass over Lake Michigan High levels of secondary pollutants such as O3 and NO2
developed by early afternoon, when the air mass is transported onshore by the lake-breeze
5-37
-------
TABLE 5-2. AVERAGE AFTERNOON BACKGROUND POLLUTANT
CONCENTRATIONS MEASURED AT KENOSHA, WISCONSIN
North wind
South wind
03
(ppb)
37
94
NO
(Ppb)
2
3
NO2
(ppb)
3
7
NMTHC
(ppbC)
123
213
CO
(ppm)
04
06
CFC13
(PPt)
163
229
Source Westbergetal (1981)
600
Lake-breeze
front past trailer
Lake-breeze
front past trailer
1200
14 16
August 18
181200
Time of day
14 16
August 19
18
Figure 5-13. Pollutant levels at the Kenosha, WI, sampling site before and after passage
of the lake-breeze front.
Source' Westbergetal (1981)
5-38
-------
The pollutants associated with the local, mesoscale lake-breeze system certainly get
incorporated into the larger scale synoptic circulation and contribute to the increased levels
associated with southwesterly flow (see Table 5-2) The distribution of odd nitrogen species
within these anticyclones will depend on rates of chemical conversion and deposition
The studies just described provide little detailed information about oxides of nitrogen
behavior under various regional transport scenarios At the tune those studies were
conducted, sophisticated NOX and NOy monitoring instrumentation was not generally
available Consequently, we only know that NOX levels were near the detection limit of the
instrumentation (»1 to 2 ppb) in synoptic air masses advected into the United States from
Canada and that NOX concentrations increased significantly after a period of residence over
industrialized regions of the United States The composition of odd nitrogen species in the
aged air mass is not well known However, more recent studies may be able to provide
better insight into oxides of nitrogen chemistry in regional air masses (Fahey et al, 1986;
Luke and Dickerson, 1987) For example, Luke and Dickerson (1987) have reported NOX
and NOy measurements off the east coast of the United States They subdivided the
atmosphere along their flight tracks into two-dimensional boxes and calculated the flux of
nitrogen species through each box A calculated gross nitrogen flux of 0 5 Tg/year was
derived using this methodology Luke and Dickerson emphasize that the 0 5 Tg/year flux
should be viewed with caution because it results from the combination of an annually
averaged wind field and an NOy data base of limited time resolution Sampling flights were
conducted during the period January 3 to 11, 1986
Even though the Luke and Dickerson flux numbei is subject to considerable
uncertainty, it is interesting to compare it to earlier flux estimates that were derived by less
direct methods Logan (1983) calculated a nitrogen flux of 1 7 Tg/year by balancing the
regional nitrogen budget of eastern North America Galloway et al (1984) estimated that
1 1 to 3 2 Tg/year are transported eastward off the Atlantic coast More recently, Galloway
and Whelpdale (1987) have reduced their flux estimate downward to 0 8 to 1 2 Tg/year
This flux corresponds to approximately 25% of the NOX emitted to the atmosphere of eastern
North America based on Logan's (1983) NOX emission estimate of 4 5 Tg nitrogen/year
5-39
-------
5.5 OXIDES OF NITROGEN AND THE GREENHOUSE EFFECT
The oxygenated nitrogen species typically assumed to comprise the NOX and NOy
families, as discussed in previous sections of this chapter, do not absorb infrared radiation,
and, therefore, do not contribute to direct radiative "greenhouse" forcing Nitrogen dioxide
is an efficient absorber of visible radiation, and it has been proposed as a possible source of
additional climatic influences, assuming atmospheric concentrations were to become
sufficiently large (Wuebbles, 1989) The previously described NOX species can, however,
contribute indirectly to the greenhouse process through the photochemical production of O3,
a known greenhouse gas Additionally, nitrous oxide (N2O), which is chemically inert in the
troposphere, readily absorbs longwave radiation and is among the more significant non-CO2
greenhouse gases.
5.5.1 Ozone Greenhouse Effects Related to Nitrogen Oxides
On a per mole basis, Rodhe (1990) estimated troposphenc O3 to be more effective at
absorbing infrared radiation than CO2 Measured background troposphenc
O3 concentrations, particularly in the northern hemisphere, have shown an apparent increase
over the past few decades, although the uncertainties are generally large (Logan, 1985;
Oltmans and Kohmyr, 1986, Angell, 1988) Rodhe estimates the current annual rate of
increase in global troposphenc O3 as 0 5 % Using a one-dimensional model, Lacis et al
(1990) showed the direction of the O3-induced radiative forcing to be sensitive to the vertical
O3 distnbution. Troposphenc O3 increases, as well as stratospheric decreases, could both
lead to surface warming Ozone concentration changes in the upper troposphere and lower
stratosphere, where temperatures are at a minimum compared to surface temperatures, are
the most effective in producing surface layer temperature changes (Wuebbles, 1989)
It should be noted, however, that Lacis et al (1990) calculated a net O3-induced 0 05 °C
(±0.05 °C) surface cooling for mid-latitude regions during the 1970s Based oe limited
observational O3 column data, the modeled surface cooling caused by decreases in
stratospheric O3 outweighed warming effects brought on by troposphenc O3 increases
The photochemical relationship of NOX to the formation of troposphenc O3 has been
previously described in Section 52 It is generally regarded that due to the relatively short
atmospheric lifetime of O3, local (urban) sources of O3 do not contnbute significantly to the
5-40
-------
upper troposphenc global O3 levels (Machta, 1983) Sources of greenhouse-important
troposphenc O3 are presumably an equal combination of downward injection of stratospheric
O3 and O3 precursors (NOX) during intrusion episodes and upper troposphenc photochemical
production (Liu et al, 1980, Fishman, 1985, Wuebbles et al, 1989) Of particular interest
then, is the mechanisms by which NOX species can be transported and dispersed sufficiently
throughout the mid to upper troposphere to result in the formation of upper-level O3 The
processes of mesoscale and synoptic transport (see Section 5 4 3) in combination with
transformation of NOX to reservoir species, such as peioxy- and organic nitrates, could serve
to relay the O3 precursor species to the remote troposphere Singh et al (1985) submitted
that PAN, in particular, would be an effective long-range transport mechanism for boundary
layer NOX, thereby influencing background levels of Oj, as well as other important oxidation
compounds (i e , OH radicals)
Another proposed anthropogenic source of mid and upper troposphenc, as well as
stratospheric, O3 precursors (NOX and VOCs) is via the exhaust of high-flying jet aircraft
(Liu et al, 1980, Kinnison et al, 1988) Liu et al estimated a potential 7 to 15% increase
in upper troposphenc O3 over the northern hemisphere due to high-flying subsonic aircraft
for the decade of the 1970s They compared this to an average observed 8% increase over
the same hemisphere from 1966 to 1977 Nitrogen oxides are also known to be produced in
conjunction with lightning discharges (Logan, 1983) While investigating lightning, followed
by upward transport, as a potential source for stratospheric NOy, Ko et al (1986) showed
that significant levels, above background, of troposphenc NOy can be produced, especially in
the tropical latitudes
In a recent World Meteorological Organization Assessment Document (World
Meteorological Organization, 1991) dealing with the status of atmosphenc O3, an
international group of scientists provided the following conclusions concerning troposphenc
O3-NOX relationships
(1) Increases in NOX and other O3 precursor species can lead to increases in
troposphenc O3
(2) The increase in these precursors could be the reason for the approximately
10% per decade increase in O3 measured at northern mid-latitudes over the past
two decades
5-41
-------
(3) Estimates of O3 production from NOX emissions vary depending on the model
employed; thus, the usefulness of such calculations is limited
(4) Given the key role NOX plays in troposphenc O3 chemistry, high priority should be
given to observations of NOX compounds
As briefly mentioned earlier, O3 perturbations within the stratosphere, particularly the
lower regions, can produce surface temperature changes of equal or greater magnitude than
O3 changes within the troposphere (Lacis et al, 1990) In the lower to mid-stratosphere, at
middle latitudes, the destruction of stratospheric O3 proceeds via a senes of complex,
catalyzed reactions involving the HOX and C1X families of free radicals These reactions, as
summarized by Johnston (1982), are as follows
NO •*• O3 -* N02 + O2
direct
catalyzed by HOX (5"48)
catalyzed by C1X
O3 + h? -> O + O2 (5-49)
NO2 + O ^ NO + O2, (5-50)
where, because Equation 5-50 is the rate determining step, the gross rate of stratospheric
O3 depletion through reactions of NOX can be approximated by
-d[03]/dt = 2 ft^ [O] [N02] (5-51)
In polar (high-latitude) regions, especially in the Antarctic where favorable conditions
often exist, additional NOy/ClX heterogenous reactions on the particulate surfaces of polar
stratospheric clouds can further enhance stratospheric O3 depletion (see Section 5 6), thereby
increasing the likelihood of greenhouse forcing
5.5.2 Nitrous Oxide Greenhouse Contributions
Chemically unreactive in the troposphere, N2O readily absorbs infrared radiation and is
estimated to be responsible for approximately 4 to 5 % of the theorized greenhouse effect
5-42
-------
(Hansen et al, 1989, Rodhe, 1990) At present atmosphenc mixing ratios (307 to
310 ppbv), N2O, on a per mole basis, is given to be 200 tunes more effective than CO2 as
an absorber of heat radiation (Wuebbles, 1989, Rodhe, 1990) Nitrous oxide in the
troposphere is thought to originate predominately through soil demtnfication (McElroy,
1980, Machta, 1983) Additionally, anthropogenic sources, especially high-temperature
combustion, may also release significant amounts of N2O Wuebbles (1989) estimated that
as much as 40% of atmosphenc N2O to be a product of anthropogenic processes (fossil fuel
combustion, 21%, biomass burning, 5%, fertilized soils, 5%, cultivated natural soils, 10%)
Other researchers (Hao et al, 1987, Muzio and Kramlich, 1988) have proposed that the
contribution of fossil fuel combustion may by significantly less Thiemens and Trogler
(1991) calculated that the commercial manufacture of nylon releases, on a global basis,
o
approximately 6 6 x 10 kg of N2O annually This would account for about 0 03 % of the
current troposphenc levels, or about 10% of the observed annual increase
Although there are as yet some questions in the partitioning of N2O sources, the
atmosphenc levels of N2O have been observed to be increasing at an annual rate of 0 2 to
03% (Machta, 1983, Wuebbles et al , 1989, Rodhe, 1990) With an atmosphenc lifetime of
150 years (Rodhe, 1990, Levander, 1990), it can be seen that given an increase in
atmosphenc loading, N2O will become an increasingly more important greenhouse gas
Levander (1990) using a 0 2% rate of increase in N2O emissions, predicted that after
50 years, the resulting buildup in N2O would be 350 times more effective at absorbing
infrared radiation than an equivalent amount of CO2 emissions The Levander estimate is
somewhat larger than that listed for N2O in the Scientific Assessment of Ozone Depletion
1991 (World Meteorological Organization, 1991) document A global warming potential of
270 was calculated for 50 years in the future An N2O lifetime of 132 years was employed
in the World Meterological Organization calculation
Nitrous oxide does not decompose until it is transported to the stratosphere The main
sink for stratospheric N2O is the reaction with O(1D) (Johnston, 1982, Logan, 1983,
Wuebbles, 1989) The products of such reactions are the primary sources for stratospheric
NOX, which, as previously mentioned, catalytically react to deplete stratospheric O3 and
partially supply NOX species to the upper troposphere during intrusion episodes
5-43
-------
5.6 STRATOSPHERIC OZONE DEPLETION BY OXIDES OF
NITROGEN
Oxides of nitrogen, from anticipated high-altitude supersonic aircraft exhaust, were first
proposed as a destruction mechanism for stratospheric O3 in the early 1970s (Crutzen, 1970,
Johnston, 1971). Since then, the relationship between stratospheric odd nitrogen species,
NOy (NO + NO2 + NO3 + HNO3 + chlorine nitrate [C1NO3] + N2O5 + HO2NO2), has
received considerable attention
The primary source of stratospheric NOy is thought to be via the reaction of O( D) and
N2O (Johnston, 1982, Logan, 1983, Wuebbles, 1989),
O(*D) + N2O -» 2 NO (5-52)
Significant sources of N2O were discussed in Section 552 Jackman et al (1980) list other
possible, less significant, sources of stratospheric odd nitrogen as NOy production by
lightning in the troposphere followed by upward transport, nuclear bomb blasts,
thermosphenc NOy production with downward transport, and NOy pioduced during ionizing
events (i.e , solar proton episodes) Additionally, recent renewed interest in stratospheric,
intercontinental passenger jet aircraft has likewise restunulated research into the sensitivity of
stratospheric O3 to potential aircraft exhaust (Kmnison et al, 1988, Kinnison and Wuebbles,
1989).
In the mid-latitude and lower- to mid-stratosphenc regions, as mentioned in
Section 5.5.1 (Equations 5-48 to 5-51), NO cycles through NO2 for a net destruction of two
molecules of O3
NO + O3 -* NO2 + O2 (5-48)
O3 + hv -* O + O2 (5-49)
NO2 + O -* NO + 02 (5-50)
Net 2O3 -» 3O2
It is known, however, that catalytic reactions involving chlorinated compounds within
the stratosphere can be even more effective as an O3 depletion mechanism These reactions,
although not specifically involving NOX in the O3 destruction reaction, are an important
5-44
-------
component of the overall reaction system The homogeneous CIO dimer reaction sequences,
which are most important, as proposed by Molina and Molina (1987), are as follows
2 CIO + M -> C12O2 + M (5-53)
C12O2 + hv -* Cl + C1OO (5-54)
C10O + M ^ Cl + O2 + M (5-55)
2C1 + 2O3 -* 2C1O + 2O2 (5-56)
Net 2O3 -* 3O2
Fahey et al (1989) showed a similar CIO dimer reaction sequence, differing only by the self-
decomposition of dichlonne dioxide and the photolysis of diatomic chloride The actual
O3 depletion mechanism (Equation 5-56) is the same in both reaction schemes.
2C1O + M -> C12O2 + M (5-53)
C12O2 + M -* C12 H- O2 + M (5-57)
C12 + hf -> 2C1 (5-58)
2C1 + 2O3 -» 2C1O + 2O2 (5-56)
Net. 2O3 -> 3O2
McElroy et al (1986a) also proposed the following chlorine/bromine oxidation cycle as
an important stratospheric O3 depletion mechanisms
Br + O3 ^ BrO + O2 (5-59)
Cl + O3 -> CIO + O2 (5-60)
CIO + BrO ^ Cl + Br + O2 (5-61)
Net 2O3 -> 3O2
The roles of Equations 5-53 to 5-61 in the stratospheric O3 depletion are generally limited in
the mid-latitude regions by the reaction of CIO and BrO with NO2, which forms the
unreactive C1NO3 and bromine nitrate (BrNO3) (McElroy et al, 1986a)
5-45
-------
CIO + NO2 + M -> C1NO3 + M (5-62)
BrO + N02 + M ^ BrNO3 + M (5-63)
In the atmospheric regions mentioned above, where oxygenated nitrogen compounds are
generally much more prevalent than chlorine compounds, Equations 5-62 and 5-63 act as
important sinks for CIO and BrO, but as an insignificant sink for NO2 (McElroy and
Salawitch, 1989) Therefore, in order for Equations 5-53 to 5-61 to become significant, NO2
must be removed from the reaction sequences or transformed into a less reactive species
Of more importance is that after the reaction sequences occur, the initial scavengers,
NO or CIO, are reformed so the sequence can continue unabated until something removes
them from the sequence and terminates the reactions One CIO molecule can destroy
100,000 O3 molecules under normal conditions
Several heterogeneous reactions, believed to occur on the ice surfaces of polar
stratospheric clouds (PSCs), have been proposed that act to sequester or remove
(by deposition) reactive odd nitrogen This, then, initiates the more effective O3 depletion
chlorine and chlorine/bromine cycles The proposed heterogeneous reactions involve the
reactions of C1NO3 and N2O5 with hydrogen chloride (HC1) and water, presumably in the
solid phase, on PSCs particulate surfaces (Molina et al, 1987; Tolbert et al, 1987, Tolbert
et al., 1988a).
C1NO3 + HC1 •* C12 + HNO3 (5-64)
C1NO3 + H2O -> HOC1HNO3 (5-65)
N2O5 + HC1 -* C1NO2 + HNO3 (5-66)
N2O5 + H2O -» 2 HNO3 (5-25)
Leu (1988) also suggested recombination of CIO on the surface of PSCs,
CIO + CIO -* C12 + O2 (5-67)
However, much uncertainty still remains regarding the mechanism and the importance of
Equation 5-67 to the heterogeneous reactions associated with stratospheric O3 depletion
5-46
-------
Polar stratospheric clouds are commonly formed at altitudes in the 10- to 20-km range
over the Antarctic, and, to a lesser extent, over the Arctic, during the respective winter
months (McCormick et al , 1982) Model calculations that have included both heterogeneous
(Equations 5-64 to 5-66) and homogeneous mechanisms have simulated the observed
Antarctic O3 depletion patterns reasonably well (McElroy et al , 1986b, Solomon et al ,
1986, Wofsy et al, 1988, Fahey et al, 1989) Additionally, the model calculations of
Douglass and Stolarski (1989) have demonstrated that such heterogeneous reactions may have
a noticeable impact on the O3/NOy/ClX chemistry of the Arctic stratosphere, even though
arctic PSCs are less common and less persistent
McElroy et al (1986a) and others have suggesled that the HNO3 formed in the above
heterogeneous reactions could condense, along with water, resulting in the formation of PSCs
at temperatures warmer than the ice point of water The embellished formation of PSCs
would, in turn, enhance the efficiency of the heterogeneous reactions of the polar
stratospheric O3 depletion mechanisms Laboratory experiments by Hanson and
Mauersberger (1988) showed the HNO3 trihydrate formed at stratospheric conditions
condensed approximately 7 K above the ice point
Recent investigations have suggested there is still considerable uncertainty in the precise
mechanisms of the theorized polar stratospheric heterogeneous chemistry Wolff et al
(1989) presented data that contradicted the findings of earlier investigators (Molina et al ,
1987, Wofsy et al , 1988) pertaining to the incorporation and movement of HC1 within ice
crystals Wolff et al (1989) found HC1 is not readily incorporated into ice crystals, but
rather strongly partitioned along gram boundaries They proposed that the HC1 may be
present in some form other than solid (i e , liquid surface film, grain boundary liquid,
chemisorbed to the ice surface, reactant on super-cooled droplet), thereby supplying the HC1
as required by current theories
Other mechanisms for the release of active chlorine from the reservoir species (C1NO3
and N2O5), which would initiate the homogeneous O3 depletion cycle, have been presented
by Tolbert et al (1988b) and Finlayson-Pitts et al (1989) Laboratory studies by Tolbert
et al (1988b) suggested that the heterogeneous reactions suspected to occur on PSCs may
also occur on atmospheric sulfunc acid aerosols Tins would have the effect of extending the
latitude and lowering the altitude at which the proposed heterogeneous reactions could occur
5-47
-------
Finlayson-Pitts et al (1989) found that C1NO3 and N2O5 react with sodium chloride particles
at 298 K similarly to the previously discussed polar stratospheric reactions
C1NO3 + NaCl(s) -* C12 + NaNO3(s) (5-68)
N2O5 + NaCl(s) -* C1NO2 + NaNO3(s) (5-69)
These additional mechanisms, which result in photochemically active chlorine (O3-depleting)
species, indicate that reactions similar to the stratospheric heterogeneous reactions may also
impact troposphenc chemistry. Additional research may also result in the inclusion of these
mechanisms in polar, and possibly global, stratospheric chemistry
5.7 DEPOSITION OF NITROGEN OXIDES
Oxides of nitrogen and related nitrogen-containing species can be removed from the
atmosphere by dry and/or wet deposition Dry deposition consists of transfer of a gaseous
species from the atmosphere to an underlying surface, where the gas is chemically or
biologically assimilated Wet deposition requires incorporation of NOX into cloud and/or
precipitation particles, followed by delivery to the earth's surface Wet deposition rates are
highly variable because they depend on atmospheric advection and mixing processes, storm
dynamics, atmospheric chemical transformations, and physiochemical processes in the cloud
environment
5.7.1 Dry Deposition of Nitrogen Oxides
The general procedure for calculating dry deposition fluxes (F) is to multiply deposition
velocity (Vd) for a particular trace gas by its air concentration (C) at some reference height
above the surface.
= VdC
The deposition velocity is an experimentally determined parameter, which depends on
meteorological conditions, surface, and trace gas characteristics In order to apply deposition
5-48
-------
velocity to a diversity of conditions, Vd can be broken down into the reciprocal sum of three
individual resistances
Vd = l/(Ra + Rb + Re)
Ra is an aerodynamic resistance related to the atmospheric turbulence above the surface,
Kb accounts for resistance associated with the thin boundary layer that exists very close to
the surface, and Re defines the sink capacity of the surface itself
The aerodynamic resistance (Ra) is a function of a number of physical and
meteorological parameters, including friction velocity, height above the surface, surface
roughness, length, and turbulence class In most cases, Ra cannot be calculated directly, and
approximations must be invoked Various formulations have been described by Hicks et al
(1987)
Molecular diffusion becomes the primary transport pathway in the thin layers close to
the depositional surface Therefore, Rb is a function of the molecular diffusivity of the
depositing gas As with Ra, normally it is difficult to evaluate Rb precisely because surface
roughness is highly variable and, consequently, difficult to parameterize in deposition
models
For many gases, the surface uptake resistance (Re) is the most difficult to evaluate and
is dependant on the sink capacity of the depositional surface, which is a function of numerous
physical, chemical, and biological processes Transport through stomatal openings on leaf
surfaces, for example, is a function of solar radiation, leaf temperature, leaf water potential,
etc
5.7.2 Methods for Determining Deposition Velocities
Deposition velocities (Vd) are defined as the ratio of surface fluxes to air concentrations
at some height above the surface (e g , 2 m) In order to determine Vd, the vertical flux
(rate of transport per unit area) and fluctuations of a trace gas must be measured The most
common methods employed include eddy correlation, vertical gradients, and enclosure-based
systems
5-49
-------
5.7.2.1 Eddy Correlation
This is the most direct method for measuring vertical fluxes It requires fast response
and sensitive detection of trace gases, along with the simultaneous measurement of vertical
velocity Vertical velocity is generally determined with a sonic anemometer The limiting
feature associated with the eddy correlation technique is availability of fast-response chemical
sensors. Of the nitrogen-containing trace gases, only NO and NO2 have detection systems of
sufficient speed to utilize the eddy correlation method
The fast response limitation is eliminated in a variation of eddy correlation known as
eddy accumulation Vertical air movement is still monitored with a sonic anemometer, but
instead of having a colocated fast response chemical sensor, a pump is used to fill two air
sampling devices Air is pumped into one of the containers when movement is upward
(as sensed by the sonic anemometer) and into the second container when air is subsiding
The mass of the trace gas of interest in each container is then determined using conventional
analytical techniques As long as sample integrity is maintained in the collection device,
eddy accumulation could be used to measure the flux of such nitrogenous gases as PAN and
PPN; organic nitrates, and, possibly, HNO3
5.7.2.2 Vertical Gradient Methods
These methods involve the measurement of trace gas concentrations at several levels
above the surface The vertical concentration profile is proportional to the flux of the trace
gas of interest
F = Kz(dc/dz)
The proportionality constant (Kz) is normally estimated from concurrently measured
meteorological parameters or by assuming that it is the same for another quantity as for the
trace gas of interest For example, sensible heat flux can be measured by eddy correlation
This allows calculation of the transfer (diffusion) coefficient (Kz), which can then be
combined with concurrent vertical profile measurements of a trace gas to determine its flux
The main limitation of this latter procedure is the requirement that sources and sinks of the
trace gas and heat flux are similarly distributed
5-50
-------
5.7.2.3 Chamber Methods
A volume of air is enclosed above a deposition.il surface (soil, water, etc ) by one of
two types of chambers In closed circulation chambers, trace gas fluxes are determined by
periodically collecting samples from the chamber and calculating the change in concentration
with tune In the closed or static mode, the chamber remains over the surface of interest
only long enough to make the measurement By contrast, dynamic or open chambers
generally are kept in place for several hours or even days In this latter case, a continuous
flow of air passes through the enclosure When chamber methods are utilized for
determining NOX depositional fluxes, special care must be exercised to eliminate the
following problems (Mosier, 1989)
(1) Uptake or production of NO and NO2 by chamber wall material—corrections must
be made when fluxes of NO and NO2 are very small
(2) Changes in aerodynamic mixing close to the surface—airflows through the chamber
should mimic the natural environment as closely as possible
(3) If reaction tunes for species of interest are similar to the enclosure residence tune,
corrections must be made This is generally true for open chambers where NO,
NO2, and O3 are present
(4) For NO, the net exchange above the surface has been shown to be dependent on
the concentration Thus, a "compensation point" exists above which there will be
uptake of NO and below which NO emission will occur Therefore, in order to get
representative information, NO concentrations in chamber air must be as close as
possible to those in ambient air
5.7.3 Deposition of Nitrogen Oxides
Nitrogen oxides dry deposition fluxes are still very uncertain Results from many
micrometeorological studies exhibit large scatter in NOX fluxes This is due both to
analytical problems (i e , interference from other nitrogen-containing species) and to the fact
that NOX exchange includes simultaneous emission and deposition It is generally believed
that NO emission exceeds NO deposition, and that NO2 deposition is greater than NO
deposition (Johansson, 1989) Deposition velocities (Vrf) for NO have been reported to range
from less than 0 1 to approximately 0 2 cm/s For NO2, reported Vd values generally fall in
the range 0 3 to 0 8 cm/s (Hanson and Lindberg, 1991) Because NO normally constitutes a
5-51
-------
small fraction (»10%) of the atmospheric NOX concentration, the dry deposition of NO to
terrestrial surfaces can be neglected as a sink for removal of atmospheric NOX
5.7.4 Nitric Acid Deposition
The surface uptake resistance for HNO3 deposition to terrestrial surfaces has been
shown to be very small (Huebert and Robert, 1985) Therefore, when aerodynamic and
diffusional processes bring gaseous HNO3 in contact with a surface, the HNO3 molecules
will deposit at nearly 100% efficiency Consequently, deposition velocities are larger than
those for NO and NO2 Terrestrial values for Vd reportedly range between 0 5 and
3.0 cm/s. Nitric acid deposition velocities over water surfaces are somewhat lower, falling
in the 0.3 to 0 7 cm/s range
5.7.5 Deposition of Peroxyacetylnitrate
Very little information exists concerning PAN deposition rates A Vd of 0 25 cm/s has
been reported over a grass and soil surface (Garland and Penkett, 1976) In their study of
the photochemistry of biogenic emissions in the Amazon Basin, Jacob and Wofsy (1988)
assumed for PAN a Vd value of 2 cm/s A high value was selected because of the large
surface area associated with the tropical vegetation canopy It is expected thai the deposition
velocity over water surfaces would constitute the other extreme, with values as low as
0.01 cm/s having been proposed (Andreae et al, 1988) Deposition velocities for PAN will
probably remain uncertain due to the difficulties associated with making accurate PAN
measurements
5.7.6 Wet Deposition of Nitrogen Oxides
Wet deposition is not a significant atmospheric removal mechanism for NO and NO2
These two gases are minimally soluble in water and, therefore, must be transformed to more
highly oxidized forms for wet removal to become effective The reaction of NO2 with OH
radical to produce HNO3 appears to be the main source of the nitrate ion measured in
precipitation It is estimated that about one-third of United States NOt emissions are
removed by wet deposition processes (Hicks et al , 1991)
5-52
-------
5.8 SUMMARY AND CONCLUSIONS
Nitrogen oxides are important chemical species in the planetary boundary layer, as well
as in the free troposphere and the stratosphere Nitrogen oxides play important roles (1) in
the control of concentrations of radicals in the clean troposphere, (2) in the production of
troposphenc O3, (3) as an aerosol precursor, and (4) in the production and deposition of
acidic species, directly or indirectly
5.8.1 Ozone Production
Combustion processes emit a variety of nitrogen compounds, but chiefly NO, which is
rapidly oxidized to NO2 in ambient air, primarily by O3 Photolytic decomposition of NO2
then leads to regeneration of NO, producing also an excited oxygen atom that reacts with
molecular oxygen to form O3 In the absence of competing reactions, NO, NO2, and
O3 reach an equilibrium described by the steady-state equation
Competing reactions exist, however, so that free radicals (HO2 , RO2) generated from
the oxidative degradation of volatile organic compounds (VOCs) oxidize NO to NO2 without
destroying O3 Thus, the amount of O3 formed in ambient air is dependent upon the
concentration of NOX present as well as the concentrations and reactivities of VOC species
5.8.2 Production of Odd Nitrogen Species
Photochemical processes that include the coupled reactions of NOX, oxygen species, and
free radicals produce not only O3, but mtrogen-contamuig products as well These oxidation
products include HNO3, HO2NO2, HONO, RC(O)O2NO2, N2O5, and inorganic and organic
nitrates
Nitric acid is a major sink for active nitrogen and is a contributor to acidic deposition
It has been estimated to account for roughly one-third of the total acidity deposited in the
eastern United States (Calvert and Stockwell, 1983) Potential physical and chemical sinks
for HNO3 include wet and dry deposition, photolysis, reaction with OH radicals, and
neutralization by gaseous NH3, which leads to aerosol production
Peroxyacyl nitrates are formed from the combination of RO2 radicals with NO2
Peroxyacetylmtrate is the most abundant member in the lower troposphere of this
homologous series of compounds It can serve in the troposphere as a temporary reservoir
5-53
-------
for reactive nitrogen species and can be regionally transported, but it cannot function as a
true sink in the lower troposphere because of its thermal instability In the upper
troposphere, where temperatures are colder, the lifetime of PAN is longer, but is only about
3 mo because PAN is photolyzed and also reacts with OH radicals
The NO3 radical is a short-lived radical that is formed in the tioposphere primarily by
the reaction of NO2 with O3 In daylight, NO3 undergoes rapid photolysis or reaction with
NO. After sunset, accumulation of NO3 can occur and is expected to be controlled by the
availability of NO2 and O3 plus chemical destruction mechanisms involving the formation of
N2O5 and HNO3.
Dinitrogen pentoxide, the anhydride of HNO3, is primarily a nighttime constituent of
ambient air because it is formed from the reaction of NO3 (itself a nighttime species) and
NO2. Dinitrogen pentoxide is thermally unstable, but at the lower temperatures of the upper
troposphere, it can serve as a temporary reservoir of NO3 In the boundary layer, N2O5
reacts heterogeneously with water to form HNO3, which in turn is deposited out
Amines, mtrosamines, and nitramrnes are thought to exist in ambient air, but at low
concentrations. Both mtrosamines and nitramrnes have short lifetimes in ambient air because
they are photolytically decomposed (nitrosamines) and/or react with OH radicals and
O3 (nitramines and mtrosamines)
5.8.3 Transport
5.8.3.1 General Features
The transport and dispersion of the various nitrogenous species are dependent on both
meteorological and chemical parameters Advection, diffusion, deposition, and chemical
transformations combine to dictate the atmospheric residence time of a particular trace gas
In turn, atmospheric residence times help determine the geographic extent of transport of a
given species Surface emissions are dispersed vertically and horizontally through the
atmosphere by tubulent mixing processes that are dependent to a large extent on the vertical
temperature structure and wind speed
As the result of meteorological processes, NOX emitted in the early morning hours in an
urban area will disperse vertically and horizontally (downwind) as the day progresses
On sunny summer days, most of the NOX will have been converted to HNO3 and PAN by
5-54
-------
sunset Much of the HNO3 is removed by deposition as the air mass is transported, but
HNO3 and PAN earned in layers aloft (above the nighttime inversion layer, but below a
higher subsidence inversion) can potentially be transported long distances
5.8.3.2 Transport of Reactive Nitrogen Species in Urban Plumes
Studies of the fate of reactive nitrogen species in daytime urban plumes indicate
removal rates ranging from 0 04 h" in Los Angeles (Chang et al, 1979), to 0 1 h" in
Detroit (Kelly, 1987), to 0 14 to 0 24 h"1 (for four different, nonconsecutive days) in Boston
(Spicer, 1982a) In the Detroit study, HNO3 accounted for 67 to 84% of the nitrogenous
transformation products, but still fell short of predicted HNO3 levels Removal by
incorporation into coarse atmospheric aerosol was postulated as a major sink for HNO3 and
as the cause of the discrepancy between measured and predicted levels (Kelly, 1987)
The nighttime chemistry of NOy is poorly understood Nighttime concentrations of
NO3 show a typical pattern of increase until O3 is no longer available, followed by a
decrease as NO emissions cannot be oxidized by O3 to NO2, but react instead with the NO3
5.8.3.3 Transport and Chemistry in Combustion Plumes
Ozone buildup in power-plant plumes appears to be the result of mixing of VOCs into
the plume as it moves downwind because the VOC/NOX ratio in these plumes is quite low
An O3 buildup is not found in all power-plant plumes, however (e g , Hegg et al, 1977,
Ogren et al , 1977, White, 1977), and the most important factor in the m-plume formation of
O3 appears to be the availability of reactive VOCs in the dilution air
Little information is available on the fate of reactive nitrogen species in NOx-nch
plumes Aircraft measurements have shown little increase in inorganic and particulate nitrate
concentrations in power-plant plumes (Hegg and Hobbs, 1979) Chamber and modeling
studies indicate that in NOx-nch but VOC-poor plumes, the NOX lifetime will be long
enough to allow NOX to be incorporated into regional air masses (Spicer et al , 1981)
5.8.3.4 Regional Transport
Transport of reactive NOX in regional air masses can occur via several mechanisms
(1) mesoscale phenomena, such as mountain-valley wind flow or land-sea breeze circulations
5-55
-------
(transport for tens to hundreds of kilometers), (2) synoptic weather systems, such as the
migratory highs that cross the eastern United States in the summertime (transport for many
hundreds of kilometers), and (3) mesoscale phenomena coupled with slow-moving high-
pressure systems having weak pressure gradients In the latter interrelated phenomena,
mountain-valley or land-water breezes can govern pollutant transport in the immediate
vicinity of sources, but the ultimate fate of reactive NOX species will be distribution into the
synoptic system
Information remains sparse on NOX species and their concentrations in synoptic
transport systems Calculated fluxes for the northeastern, Atlantic coast area (Luke and
Dickerson, 1987; Galloway et al, 1984, Galloway and Whelpdale, 1987) correspond to
about 25% of the NOX emitted to the atmosphere of eastern North America, using Logan's
(1983) emission estimate of 4 5 x 1012 g nitrogen/year
5.8.4 Oxides of Nitrogen and the Greenhouse Effect
Except for N2O, the reactive nitrogen species comprising the NOX and NOy families in
the atmosphere do not absorb infrared radiation and, therefore, do not contribute directly to
radiative "greenhouse" forcing They can, however, contribute indirectly to greenhouse
processes through the photochemical production of O3 in the troposphere Ozone absorbs
infrared radiation more effectively per mole than CO2 (Rodhe, 1990)
Nitrous oxide, which is chemically inert in the troposphere, readily absorbs infrared
radiation and is among the more significant non-CO2 greenhouse gases Absorption of
visible radiation by NO2 could make this compound a possible source of other climatic
influences if atmospheric concentrations become sufficiently higher (Wuebbles, 1989)
5.8.4.1 Nitrous Oxide Greenhouse Contributions
Nitrous oxide is thought, on a per mole basis, to be 200 tunes more effective than CO2
as an absorber of heat radiation (Wuebbles, 1989, Rodhe, 1990) It is estimated at present
levels to be responsible for about 4 to 5 % of the theorized greenhouse effect (Hansen et al,
1989; Rodhe, 1990) Assuming a 02% per year increase, Levander (1990) predicted that
the increased atmospheric mixing ratio of N2O after 50 years would result in an even greater
efficiency (350 times) in infrared radiation absorption compared to^espected CO2 levels
5-56
-------
5.8.4.2 Stratospheric Ozone Depletion by Oxides of Nitrogen
In mid-latitudes and lower to mid-stratosphenc regions, cyclic reactions initiated by the
oxidation of NO by O3 lead to the net destruction of two molecules of O3 per cycle Among
the stratospheric O3-depletion mechanisms that have been proposed, however, are much more
important reactions involving the dimenzation of CIO in the presence of a third body, M,
and subsequent sequences in which the monomer is regenerated and two O3 molecules are
destroyed (Fahey et al , 1989, Molina and Molina, 1987) McElroy et al (1986a) also
proposed chlorine (Cl) and bromine (Br) oxidation cycles as an important stratospheric
O3-depletion mechanism In this mechanism, the dunenzation and regeneration of CIO,
coupled with the oxidation of Br and Cl by O3, are limited by the reaction of CIO and BrO
with NO2 to form the unreactive C1NO3 or BrNO3 (McElroy et al, 1986a) Reactions of
CIO and BrO with NO2 are important sinks for the halogen oxides, but are insignificant sinks
for NO2 (McElroy and Salawitch, 1989)
Sequestering of reactive NOy by heterogeneous reactions on the ice surfaces of polar,
PSCs has been proposed as a means of removing NO2 Its removal allows other
O3-depleting cycles to proceed (Molina et al, 1987, Tolbert et al , 1987, Tolbert et al ,
1988a) Dmitrogen pentoxide has been implicated in these heterogeneous reactions
5.8.5 Deposition of Nitrogen Oxides
Both wet and dry deposition of NOX and other nitrogen species occur, but wet
deposition is not a significant removal mechanism foi NO or NO2 because both gases are
minimally soluble in water Transformation to more highly oxidized forms is necessary for
effective wet deposition of NOX, and the reaction of NO2 with the OH radical to form
HNO3 appears to be the main source of nitrate ion m precipitation About one-third of the
emissions of NOX in the United States is estimated to be removed by wet deposition (Hicks
et al , 1991)
Dry deposition fluxes for NOX are highly uncertain, mainly because of analytical
problems and the simultaneous occurrence of emission and deposition of NOX Available
data indicate, however, that NO emissions exceed NO deposition and that NO2 deposition
exceeds NO deposition Reported Vd values for respective nitrogen species are < 0 1 to
«0 2 cm/s for NO, 0 3 to 0 8 cm/s for NO2, and 0 5 to 3 0 cm/s for HNO3 over land and
5-57
-------
0.3 to 0 7 cm/s for HNO3 over water (Huebert and Robert, 1985) The few data that exist
show deposition rates for PAN of 0 01 cm/s over water (Andreae et al, 1988) and 0 25 cm/s
(Garland and Penkett, 1976) to 2 0 cm/s (Jacob and Wofsy, 1988) over land
5-58
-------
REFERENCES
Altshuller, A P (1986) The role of nitrogen oxides in nonurban ozone formation in the planetary boundary layer
over N America, W Europe and adjacent areas of ocean Atmos Environ 20 245-268
Andreae, M O , Browell, E V , Garstang, M , Gregory, G L , Harass, R C , Hill, G F , Jacob, D J ,
Pereira, M C , Sachse, G W , Setzer, A W , Silva Dias, P L , Talbot, R W , Torres, A L , Wofsy,
S C (1988) Biomass-burning emissions and associated haze layers over Amazonia J Geophys Res
[Atmos] 93 1509-1527
Angell, J K (1988) An update through 1985 of the variations in global total ozone and north temperate
layer-mean ozone ' J Appl Meteorol 27 91-97
Atkinson, R (1990) Gas-phase troposphenc chemistry of organic compounds a review Atmos Environ Part A
24 1-41
Atkinson, R , Carter, W P L (1984) Kinetics and mechanisms of the gas-phase reactions of ozone with organic
compounds under atmospheric conditions Chem Rev 84 437-470
Atkinson, R , Lloyd, A C (1984) Evaluation of kinetic and mechanistic data for modeling of photochemical
smog J Phys Chem Ref Data 13 315-344
Atkinson, R , Winer, A M , Pitts, J N , Jr (1986) Estimation of night-time ^05 concentrations from ambient
NC<2 and NC<3 radical concentrations and the role of ^05 in night-time chemistry Atmos Environ
20 331-339
Eandow, H , Okuda, M , Akimoto, H (1980) Mechanism of the gas-phase reactions of C^Hg and NO3 radicals
J Phys Chem 84 3604-3608
Baulch, D L , Cox, R A , Crutzen, P J , Hampson, R F , Kerr,, J A , Troe, J , Watson, R T (1982)
Evaluated kinetic and photochemical data for atmospheric chemistry supplement 1 J Phys Chem Ref
Data 11 327-496
Calvert, J G (1976) Hydrocarbon involvement in photochemical smog formation in Los Angeles atmosphere
Environ Sci Technol 10 256-262
Calvert, J G , Madromch, S (1987) Theoretical study of the initial products of the atmospheric oxidation of
hydrocarbons J Geophys Res [Atmos ] 92 2211-2220
Calvert, J G , Stockwell, W R (1983) Acid generation in the troposphere by gas-phase chemistry Environ
Sci Technol 17 428A-443A
Carmichael, G R , Peters, L K (1984) An Eulenan transport/trarisformation/removal model for SO2 and
sulfate—I Model development Atmos Environ 18 937-951
Chang, T Y , Norbeck, J M , Weinstock, B (1979) An estimate of the NOX removal rate in an urban
atmosphere Environ Sci Technol 13 1534-1537
Crutzen, P J (1970) The influence of nitrogen oxides on the atmospheric ozone content Q J R Meteorol
Soc 96 320-325
5-59
-------
Crutzen, P J (1988) Tropospheric ozone an overview In Isaksen, I S A , ed Tropospheric ozone—regional
and global scale interactions proceedings of the NATO advanced workshop on regional and global ozone
interaction and its environmental consequences, June 1987, Lillehammer, Norway Dordrecht, The
Netherlands D Reidel Publishing Company, pp 3-32 (NATO advanced science institutes studies
scries C, mathematical and physical sciences v 227)
Davis, D D., Smith, G , Klauber, G (1974) Trace gas analysis of power plant plumes via aircraft measurement
O3, NOX, and SO2 chemistry Science (Washington, DC) 186 733-736
Douglass, A R , Stolarski, R S (1989) Impact of heterogeneous reactions on stratospheric chemistry of the
Arctic Geophy. Res Lett 16 131-134
Fancy, D. W ; Hubler, G , Parnsh, D D , Williams, E J , Norton, R B , Ridley, B A , Singh, H B , Liu,
S C , Fehsenfeld, F C (1986) Reactive nitrogen species in the troposphere measurements of NO,
NO2, HNO3, particulate nitrate, peroxyacetyl nitrate (PAN), O3, and total reactive odd nitrogen (NOy) at
Niwot Ridge, Colorado J Geophys Res [Atmos ] 91 9781-9793
Fahey, D. W ; Murphy, D M , Kelly, K K , Ko, M K W , Proffitt, M H , Eubank, C S , Ferry, G V ,
Loewenstein, M , Chan, K R (1989) Measurements of nitric oxide and total icactive nitrogen in the
Antarctic stratosphere observations and chemical implications J Geophys Res [Atmos ]
94 16665-16681
Fmlayson-Pitts, B J , Ezell, M J , Pitts, J N , Jr (1989) Formation of chemically active chlorine compounds
by reactions of atmospheric NaCl particles with gaseous N2O5 and C1ONO2 Nature (London) 337
241-244
Fmlayson-Pitts, B J j Pitts, J N , Jr (1986) Atmospheric chemistry fundamentals and experimental techniques
New York, NY John Wiley & Sons, pp 961-1007
Fishman, J. (1985) Ozone in the troposphere In Whitten, R C , Prasad, S , eds Ozone in the free atmosphere
New York, NY Van Nostrand Reinhold, pp 161-194
Galloway, J N., Whelpdale, D M , Wolff, G T (1984) The flux of S and N eastwaid from North America
Atmos Environ 18 2595-2607
Galloway, J N , Whelpdale, D M (1987) WATOX-86 overview and western North Atlantic Ocean S and N
atmospheric budgets Global Biogeochem Cycles 1 261-281
Garland, J A , Penkett, S A (1976) Absorption of peroxy acetyl nitrate and ozone by natural surfaces Atmos
Environ 10 1127-1131
Grosjean, D ; Van Cauwenberghe, K , Schmid, J , Pitts, J N , Jr (1978) Formation of mtrosamines and
mtramines by photooxidation of amines under simulated atmospheric conditions In Proceedings of the
fourth joint conference on sensing of environmental pollutants, November 1977, New Orleans, LA
Washington, DC: American Chemical Society, pp 196-199
Hanscn, J., Lacis, A , Prather, M (1989) Greenhouse effect of chlorofluorocarbons aad other trace gases
J Geophys Res [Atmos ] 94 16417-16421
Hanson, D , Mauersberger, K (1988) Solubility and equilibrium vapor pressures of hydrogen chloride dissolved
in polar stratospheric cloud materials ice and the tnhydrate of nitric acid Geophys Res Lett
15 1507-1510
5-60
-------
Hanson, P J , Lindberg, S E (1991) Dry deposition of reactive nitrogen compounds a review of leaf, canopy
and non-foliar measurements Atmos Environ Part A 25 1615-1634
Hanst, P L , Spence, J W , Miller, M (1977) Atmospheric chemistry of ^V-mtroso dimethylamine Environ
Sci Technol 11 403-405
Hao, W M , Wofsy, S C , McElroy, M B , Beer, J M , Toqan, M A (1987) Sources of atmospheric nitrous
oxide from combustion J Geophys Res [Atmos ] 92 3098-3104
Hams, G W , Carter, W P L , Winer, A M , Pitts, J N , Jr , PJatt, U , Perner, D (1982) Observations of
nitrous acid in the Los Angeles atmosphere and implications for predictions of ozone-precursor
relationships Environ Sci Technol 16 414-419
Hegg, D A , Hobbs, P V (1979) Some observations of particulate nitrate concentrations in coal-fired power
plant plumes Atmos Environ 13 1715-1716
Hegg, D , Hobbs, P V , Radke, L F , Harrison, H (1977) Ozone and nitrogen oxides in power plant plumes
In Dimitnades, B , ed International conference on photochemical oxidant pollution and its control
proceedings, v I, September 1976, Raleigh, NC Research Triangle Park, NC U S Environmental
Protection Agency, Environmental Sciences Research Laboratory, pp 173-183, EPA report no
EPA-600/3-77-001a Available from NTJS, Springfield, VA, PB-264232
Heikes, B G , Thompson, A M (1983) Effects of heterogeneous piocesses on NC»3, HONO, and HNO3
chemistry in the troposphere J Geophys Res C Oceans Atmos 88 10883-10895
Hicks, B B , Baldocchi, D D , Meyers, T P , Hosker, R P , Jr , Matt, D R (1987) A preliminary multiple
resistance routine for deriving dry deposition velocities from measured quantities Water Air Soil Pollut
36 311-330
Hicks, B B , Draxler, R R , Albntton, D L , Fehsenfeld, F C , Dodge, M , Schwartz S E , Tanner, R L.,
Hales, J M , Meyers, T P , Vong, R J (1991) Atmospheric processes research and process model
development In Irving, P M , ed Acidic deposition state of science and technology, volume I,
emissions, atmosphenc processes and deposition Washington, DC The U S National Acid Precipitation
Assessment Program (State of science and technology report no 2)
Holzworth, G C (1967) Mixing depths, wind speeds and air pollution potential for selected locations in the
United States J Appl Meteorol 6 1039-1044
Huebert, B J , Luke, W T , Delany, A C , Brost, R A (1988) Measurements of concentrations and dry
surface fluxes of atmosphenc nitrates in the presence of ammonia J Geophys Res [Atmos ]
93 7127-7136
Huebert, B J , Robert, C H (1985) The dry deposition of nitric acid to grass J Geophys Res [Atmos ]
90 2085-2090
Jackman, C H , Frederick, J E , Stolarski, R S (1980) Production of odd nitrogen in the stratosphere and
mesosphere an intercompanson of source strengths J Geophys Res C Oceans Atmos 85 7495-7505
Jacob, D J , Wofsy, S C (1988) Photochemistry of biogenic emissions over the Amazon forest J Geophys
Res [Atmos ] 93 1477-1486
5-61
-------
Johansson, C (1989) Fluxes of NOX above soil and vegetation In Andreae, M O , Schimel, I) S , eds
Exchange of trace gases between terrestrial ecosystems and the atmosphere report of the Dahlem
workshop, February, Berlin, Federal Republic of Germany New York, NY John Wilep & Sons,
pp 229-246
Johnston, H. (1971) Reduction of stratospheric ozone by nitrogen oxide catalysts from supersonic transport
exhaust Science (Washington, DC) 173 517-522
Johnston, H S (1982) Odd nitrogen processes In Bower, F A , Ward, R B , eds Stratospheric ozone and
man volume I Boca Raton, FL CRC Press, Lie , pp 87-140
Jones, C. L , Seinfeld, J H (1983) The oxidation of NO2 to nitrate—day and night Atmos Environ
17 2370-2373
Kelly, N A (1987) The photochemical formation and fate of nitric acid in the metropolitan Detroit area
ambient, captive-air irradiation and modeling results Atmos Environ 21 2163-2177
Kelly, N A , Wolff, G T , Ferman, M A (1984) Sources and sinks of ozone in rural areas Atmos Environ
18 1251-1266
Kelly, T J., McLaren, S E , Kadlecek, J A (1989) Seasonal variations in atmospheric SOX and NOy species in
the Adirondacks Atmos Environ 23 1315-1332
Kinmson, D. E., Wuebbles, D J , Johnston, H S (1988) A study of the sensitivity of stratospheric ozone to
hypersonic aircraft emissions Presented at First international conference on hypersomcs flight in the 21st
century, September, Grand Forks, ND Livermore, CA U S Department of Energy, Lawrence
Livermore National Laboratory, Atmospheric and Geophysical Sciences Division, report no
UCRL-98314 Available from NITS, Springfield, VA, DE89-001240
Kinnison, D E , Wuebbles, D J (1989) Preventing depletion of stratospheric ozone—implications on future
aircraft emissions Livermore, CA Lawrence Livermore National Laborabory, report no
UCRL-99926-Rev 1 Available from NTIS, Springfield, VA, DE89-013779
Ko, M. K. W , McEIroy, M B , Weisenstein, D K , Sze, N D (1986) Lightning a possible source of
stratospheric odd nitrogen J Geophys Res [Atmos ] 91 5395-5404
Lacis, A A , Wuebbles, D J , Logan, J A (1990) Radiative forcing of climate by changes in the vertical
distribution of ozone J Geophys Res [Atmos ] 95 9971-9981
Leu, M -T (1988) Heterogeneous reactions of ^05 with H/^O and HC1 on ice surfaces implications for
Antarctic ozone depletion Geophys Res Lett 15 851-854
Levander, T (1990) The relative contributions to the greenhouse effect from the use of different fuels Atmos
Environ Part A 24 2707-2714
Liu, S C ; Kley, D , McFarland, M , Mahlman, J D , Levy, H , n (1980) On the origin of tropospheric
ozone J Geophys Res C Oceans Atmos 85 7546-7552
Liu, S C ; Trainer, M , Fehsenfeld, F C , Parnsh, D D , Williams, E J , Fahey, D W , Huebler, G ,
Murphy, P C (1987) Ozone production in the rural troposphere and the implications for regional and
global ozone distributions J Geophys Res [Atmos ] 92 4191-4207
Logan, J A. (1983) Nitrogen oxides in the troposphere global and regional budgets J Geophys
Res C Oceans Atmos 88 10785-10807
5-62
-------
Logan, J A (1985) Troposphenc ozone seasonal behavior, trends, and anthropogenic influence J Geophys
Res [Atmos ] 90 10463-10482
Luke, W T , Dickerson, R R (1987) The flux of reactive nitrogen compounds from eastern North America to
the western Atlantic Ocean Global Biogeochem Cycles 1 329-343
Lyons, W A , Cole, H S (1976) Photochemical oxidant transport mesoscale lake breeze and synoptic-scale
aspects J Appl Meteorol 15 733-743
Machta, L (1983) Effects of non-CO2 greenhouse gases In Changing climate report of the Carbon Dioxide
Assessment Committee Washington, DC National Academy Press, pp 285-291
Malko, M W , Troe, J (1982) Analysis of the ummolecular reaction N2O5 + M = NO2 + NO3 + M Int
J Chem Kinet 14 399-416
McCormick, M P , Steele, H M , Hamill, P , Chu, W P , Swissler, T J (1982) Polar stratospheric cloud
sightings by SAM H J Atmos Sci 39 1387-1397
McElroy, M B (1980) Sources and sinks for nitrous oxide Washington, DC U S Department of
Transportation, report no FAA-EE-80-20
McElroy, M B , Salawitch, R J , Wofsy, S C , Logan, J A (198.6a) Reductions of Antarctic ozone due to
synergistic interactions of chlorine and bromine Nature (London) 321 759-762
McElroy, M B , Salawitch, R J , Wofsy, S C (1986b) Antarctic O3 chemical mechanisms for the spring
decrease Geophys Res Lett 13(suppl ) 1296-1299
McElroy, M B , Salawitch, R J (1989) Stratospheric ozone impact of human activity Planet Space Sci
37 1653-1672
Miller, D F , Alkezweeny, A J , Hales, J M , Lee, R N (1978) Ozone formation related to power plant
emissions Science (Washington, DC) 202 1186-1188
Molina, L T , Molina, M J (1987) Production of chlorine oxide (C^Oj) from the self-reaction of the CIO
radical J Phys Chem 91 433-436
Molina, M J , Tso, T -L , Molina, L T , Wang, F C -Y (1987) Antarctic stratospheric chemistry of chlorine
nitrate, hydrogen chloride, and ice release of active chlorine Science (Washington, DC)
238 1253-1257
Mosier, A R (1989) Chamber and isotope techniques In Andreae, M O , Schimel, D S , eds Exchange of
trace gases between terrestrial ecosystems and the atmosphere New York, NY John Wiley & Sons,
pp 175-187 (Bernhard, S , ed Life sciences research reports 47)
Muzio, L J , Kramlich, J C (1988) An artifact in the measurement of N2O from combustion sources Geophys
Res Lett 15 1369-1372
Noxon, J F, Norton, R B , Marovich, E (1980) NC^ in the troposphere Geophys Res Lett 7 125-128
Ogren, J A , Blumenthal, D L , Vanderpol, A H (1977) Oxidant measurements in western power plant
plumes Volume I technical analysis Volume II data Palo Alto, CA Electric Power Research Institute,
report no EPRI EA-421 Volume H available from NTIS, Springfield, VA, DE82-901024
5-63
-------
Oltmans, S J , Komhyr, W D (1986) Surface ozone distributions and variations from 1973-1984 measurements
at the NOAA geophysical monitoring for climatic change baseline observatories J Geophys Res
[Atmos]91 5229-5236
Parrish, D D ; Fahey, D W , Williams, E J , Liu, S C , Trainer, M , Murphy, P C , Albntton, D L ,
Fehsenfeld, F C (1986) Background ozone and anthropogenic ozone enhancement at Niwot Ridge,
Colorado. J Atmos Chem 4 63-80
Penkctt, S. A (1991) Changing ozone evidence for a perturbed atmosphere Environ Sci Technol
25 631-635
Perncr, D , Greenberg, J , Noxon, J F , Schmeltekopf, A , Winkler, P , Zimmerman, P (1991) The interaction
of natural hydrocarbons and NO3 above the canopy J Atmos Chem in press
Pitts, J. N , Jr (1987) Nitration of gaseous polycychc aromatic hydrocarbons in simulated and ambient urban
atmospheres' a source of mutagemc mtroarenes Atmos Environ 21 2531-2547
Pitts, J N , Jr ; Grosjean, D , Van Cauwenberghe, K , Schmid, J P , Fitz, D R (1978) Photooxidation of
aliphatic amines under simulated atmospheric conditions formation of mtrosamm.es, mtramines, amides,
and photochemical oxidant Environ Sci Technol 12 946-953
Pitts, J N , Jr , Winer, A M , Aschmann, S M , Carter, W P L , Atkinson, R (1985) Experimental protocol
for determining hydroxyl radical reaction rate constants estimation of atmospheric reactivity Research
Triangle Park, NC U S Environmental Protection Agency, Atmospheric Sciences Research Laboratory,
EPA report no EPA-600/3-85-000 Available from NTIS, Springfield, VA, FB85-238558
Plntt, U ; Perner, D., Sohroeder, J , Kessler, C , Toenmssen, A (1981) The diurnal variation of NOg
J Geophys Res. C Oceans Atmos 86 11965-11970
Rodhe, H. (1990) A comparison of the contribution of various gases to the greenhouse effect Science
(Washington, DC) 248- 1217-1219
Russell, A G , McRae, G J , Cass, G R (1985) The dynamics of nitric acid production and the fate of
nitrogen oxides. Atmos Environ 19 893-903
Singh, H. B. (1987) Reactive nitrogen in the troposphere chemistry and transport of NOX and PAN Environ
Sci Technol 21 320-327
Singh, H B ; Hanst, P L (1981) Peroxyacetyl nitrate (PAN) in the unpolluted atmosphere an important
reservoir for nitrogen oxides Geophys Res Lett 8 941-944
Singh, H B ; Salas, L J , Ridley, B A , Shelter, J D , Donahue, N M , Fehsenfeld, F C , Fahey, D W ,
Pamsh, D D , Williams, E J , Liu, S C , Hubler, G , Murphy, P C (1985) Relationship between
peroxyacetyl nitrate and nitrogen oxides in the clean troposphere Nature (London) 318 347-349
Solomon, S., Garcia, R R , Rowland, F S , Wuebbles, D J (1986) On the depletion of Antarctic ozone
Nature (London) 321 755-758
Spicer, C W (1982) Nitrogen oxide reactions in the urban plume of Boston Science (Washington, DC)
215 1095-1097
Spicer, C. W., Joseph, D W , Ward, G F (1978) Investigations of nitrogen oxides within the plume of an
isolated city New York, NY Coordinating Research Council, Ihc , report no
CRC-APRAC-CAPA-9-77 Available from NTIS, Springfield, VA, PB-290107
5-64
-------
Spicer, C W , Joseph, D W , Sticksel, P R , Ward, G F (1979) Ozone sources and transport in the
northeastern United States Environ Sci Technol 13 975-985
Spicer, C W , Sverdrup, G M (1981) Trace nitrogen chemistry during the Philadelphia oxidant data
enhancement study (1979) Research Triangle Park, NC U S Environmental Protection Agency, Office
of Air Quality Planning and Standards, contract no 68-02-3338
Spicer, C W , Sverdrup, G M , Kuhlman, M R (1981) Smog chamber studies of NOX chemistry in power
plant plumes Atmos Environ 15 2353-2365
Stockwell, W R , Calvert, J G (1983) The mechanism of NO3 and HONO formation in the nighttime chemistry
of the urban atmosphere J Geophys Res C Oceans Atmos 88 6673-6682
Thiemens, M H , Trogler, W C (1991) Nylon production an unknown source of atmospheric nitrous oxide
Science (Washington, DC) 251 932-934
Tolbert, M A , Rossi, M J , Malhotra, R , Golden, D M (1987) Reaction of chlorine nitrate with hydrogen
chloride and water at Antarctic stratospheric temperatures Science (Washington, DC) 238 1258-1260
Tolbert, M A , Rossi, M J , Golden, D M (1988a) Antarctic ozone depletion chemistry reactions of N2O5
with H2O and HC1 on ice surfaces Science (Washington, DC) 240 1018-1021
Tolbert, M A , Rossi, M J , Golden, D M (1988b) Heterogeneous interactions of chlorine nitrate, hydrogen
chloride, and nitric acid with sulfunc acid surfaces at stratospheric temperatures Geophys Res Lett
15 847-850
Trainer, M , Williams, E J , Pamsh, D D , Buhr, M P , Allwine, E J , Westberg, H H , Fehsenfeld, F C ,
Liu, S C (1987) Models and observations of the impact of natural hydrocarbons on rural ozone Nature
(London) 329 705-707
Tuazon, E C , Atkinson, R , Plum, C N , Winer, A M , Pitts, J N , Jr (1983) The reaction of gas phase
N2O5 with water vapor Geophys Res Lett 10 953-956
Tuazon, E C , Carter, W P L , Atkinson, R , Winer, A M , Pitts, J N , Jr (1984) Atmospheric reactions of
JV-nitrosodimethylamme and dimethylmtramine Environ Sci Technol 18 49-54
U S Environmental Protection Agency (1976) Assessment of scientific information on mtrosamines report of an
ad hoc study group of the Science Advisory Board Executive Committee
U S Environmental Protection Agency (1982) Air quality criteria lor oxides of nitrogen Research Triangle
Park, NC Office of Health and Environmental Assessment, Environmental Criteria and Assessment
Office, EPA report no EPA-600/8-82-026 Available from NTIS, Springfield, VA, PB83-131011
Vukovich, F M , Bach, W D , Jr , Cnssman, B W , King, W J (1977) On the relationship between high
ozone in the rural surface layer and high pressure systems A.tmos Environ 11 967-983
Wayne, R P , Barnes, I , Biggs, P , Burrows, J P , Canosa-Mas, C E , Hjorth, J , Le Bras, G , Moortgat,
G K , Perner, D , Poulet, G , Restelh, G , Sidebottom, H (1991) The nitrate radical physics,
chemistry, and the atmosphere Atmos Environ Part A 25 1-203
Westberg, H , Sexton, K , Roberts, E (1981) Transport of pollutants along the western shore of Lake Michigan
J AirPollut Control Assoc 31 385-388
5-65
-------
White, W. H. (1977) NOX-O3 photochemistry in power plant plumes comparison of theory with observation
Environ Sci Technol 11 995-1000
White, W. H , Dietz, D (1984) Does the photochemistry of the troposphere admit more than one steady state''
Nature (London) 309 242-244
Winer, A M , Atkinson, R , Pitts, J N , Jr (1984) Gaseous nitrate radical possible nighttime atmospheric sink
for biogemc organic compounds Science (Washington, DC) 224 156-159
Wofsy, S C ; Molina, M J , Salawitch, R J , Fox, L E , McElroy, M B (1988) Interactions between HC1,
NOX, and H2O ice in the Antarctic stratosphere implications for ozone J Geophys Res [Atmos ]
93 2442-2450.
Wolff, E. W., Mulvaney, R , Gates, K (1989) Diffusion and location of hydrochloric acid in ice implications
for polar stratospheric clouds and ozone depletion Geophys Res Lett 16 48 7-490
World Meteorological Organization (1991) Scientific assessment of ozone depletion 1991 Geneva, Switzerland
World Meteorological Organization, global ozone research monitoring project i eport no 25
Wuebbles, D J (1989) On the mitigation of non-CO2 greenhouse gases Washington, DC U S Department of
Energy, report no UCRL-101523 Available from NTIS, Springfield, VA, DE89016659
Wuebbles, D J , Grant, K E , Connell, P S , Penner, J E (1989) The role of atmospheric chemistry in
climate change JAPCA39 22-28
5-66
-------
6. SAMPLING AND ANALYSIS FOR
NITROGEN OXIDES AND RELATED SPECIES
6.1 INTRODUCTION
This chapter addresses various methods to measure selected airborne species containing
nitrogen and oxygen The focus is on methodologies currently available or in general use for
in situ monitoring of airborne concentrations in both ambient and indoor environments
Methods for measuring the species of interest at their respective sources are not considered,
and remote sensing technologies are mentioned in only a few cases
Although the primary focus in this document is nitrogen dioxide (NO2), other species
containing nitrogen and oxygen are also considered This chapter is organized into several
sections, with each section devoted to a different species The species under consideration
are nitric oxide (NO), NO2, nitrogen oxides (NOX), total reactive odd nitrogen oxides (NOy),
peroxyacetyl nitrate (PAN) and other organic nitrates, rutnc acid (HNO3), nitrous acid
(HONO), dimtrogen pentoxide (N2O5), nitrate radical (NO3), paniculate nitrate ion (NO3"),
and nitrous oxide (N2O). In this chapter, NOX represents the sum of NO and NO2 (i.e.,
NOX = NO + NO2), whereas the term oxides of nitrogen is used as a generic name for
mtroxycompounds (those compounds containing both nitrogen and oxygen)
Where possible, discussions of sampling and analysis methods for each species address
pertinent characteristics for each method Topics discussed include method type (i e , in
situ, remote, active, passive), description, status (i e , concept, laboratory prototype,
commercially available), interferences, time resolution, sensitivity, and precision and
accuracy A good overview of many of the currently available methods for measuring
nitrogen-containing species is the proceedings of a recent National Aeronautics and Space
Administration (NASA) workshop (National Aeronautics and Space Administration, 1983).
Methods development usually progresses through several stages concepts, laboratory
prototypes, laboratory evaluations, field tests, field evaluations and comparisons against other
"proven" methods, and finally, consensus acceptance by the user community At each stage,
modifications may be implemented to improve or resolve weaknesses that have been
revealed This is usually a winnowing process As a result of limitations discovered during
6-1
-------
this process, many candidate methods may be abandoned in favor of other methods
At some stage near the end of the process, commercialization may occur In the current
document, those methods that have successfully progressed to the final stages of development
are emphasized.
6.2 NITRIC OXIDE
Although NO2, rather than NO, is the primary focus of this document, the most
commonly used method of measuring NO2 does not detect the NO2 molecule directly
Instead, the method relies on a chemiluminescent reaction of the NO molecule after NO2 has
been converted to NO Thus, to provide a background for subsequent discussions of
measurement methods for NO2 and other nitrogen-containing species, NO rather than NO2 is
the first species that is addressed Airborne concentrations of NO can be determined by
various methods As noted previously, the most commonly used method is
chemiluminescence (CLM). Other methods include laser-induced fluorescence (OF),
absorption spectroscopy, iomzation spectroscopy, and passive collection with subsequent wet
chemical analysis
6.2.1 Chemiluminescence
Chemttuminescence can be used to detect several airborne rntrogen-contarning species
(i.e., NO, NO2, HONO, HNO3, N2O5, PAN, NOX, NOy, ammonia [NH3], and NO3")
Among these compounds, only NO is detected directly, whereas the other compounds must
be converted in some manner to NO prior to detection
The principle is based on the detection of the light emitted following the reaction of NO
with ozone (O3) Excess O3 is added to an air sample containing NO that is passing through
a darkened reaction vessel with infrared-reflective walls and a window for viewing by a
photomultiplier (PM) tube The light-emitting species is an electronically excited NO2
radical, a product of the reaction of NO and O3, which relaxes by photon emission that
ranges in wavelength well beyond 600 nm and is centered near 1,200 run. Light is detected
by a red-sensitive PM tube fitted with optical filters to prevent interference by radiation
below 600 nm produced by ozonalysis of other materials The intensity of the measured
6-2
-------
light is proportional to the concentration of NO in the air sample, and the concentration can
be determined by calibration with atmospheres of known composition
In applications to detect other airborne nitrogen-containing species, the air sample is
preconditioned prior to entering the reaction vessel to convert some or all of these species to
NO, and that signal is compared to the signal for an unconditioned sample The signal from
the unconditioned sample represents NO, whereas that from the preconditioned sample
represents the sum of the originally present NO along with the NO resulting from conversion
of the other nitrogen species Signal differencing permits determination of the other
nitrogen-containing species The specificity of the preconditioning process may be
controversial and is discussed in subsequent sections
Chemiluminescence is designated by the U S. Environmental Protection Agency (EPA)
as the Reference Method for determining NO2 in ambient air (see Section 6.3) As a result,
commercial instruments for measuring NO and NO2 are available Detection limits of
approximately 5 ppb with response tunes on the order of minutes are claimed by suppliers
Although these performance parameters are adequate for monitoring NO and NO2 in
relatively polluted urban and suburban environments, they may be inadequate in less polluted
remote areas Efforts have been reported by several researchers to improve the sensitivity
and response of CLM NO measurement technology to permit deployment in remote locations
both on ground-based and airborne platforms Delany et al (1982), Dickerson et al. (1984),
Tanner et al (1983), and Kelly (1986) reported techniques for modifying commercially
available NOX detectors to achieve unproved sensitivity and response times Modifications
that can be employed include (1) operation at a low pressure, high flow rate, and increased
O3 supply, (2) addition of a prereactor where sample air and O3 flows are mixed out of view
of the PM tube to obtain a more stable background signal, (3) use of a larger, more efficient
reaction vessel, with highly reflective walls, that promotes the reaction close to the PM tube,
(4) use of pure oxygen as the O3 source, (5) cooling the PM tube to reduce noise in the dark
current, and (6) change of the electronics to employ photon counting techniques rather than
analog signal processing
Kelly (1986) has provided instructions for application of the first three modifications to
the Thermo Electron Model 14-B and the Monitor Labs Model 8840 Postmodification
detection limits of 0 1 to 0 2 ppb and 90% response tunes of 5 to 10 s were claimed for
6-3
-------
these instruments Dickerson et al (1984) apphed modifications (1), (2), (3), and (5) noted
above to a Thermo Electron Model 14 B/E and reported minimum detection limits (MDLs)
for NO of 10 ppt, with a 1/e response tune of 20 s
Other workers have developed highly sensitive research-grade instruments for the CLM
determination of NO (Ridley and Hewlett, 1974; Kley and McFarland, 1980, Belas et al,
1981; Drummond et al, 1985, Carroll et al, 1985, Torres, 1985, Kondo et al, 1987)
Such devices have been used to measure NO at the earth's surface, from airborne platforms
in the troposphere, and from balloon-borne platforms in the stratosphere These instruments
generally employ those features listed above Minimum detection limits of 5 ppt or less,
response times of 2 to 60 s, and accuracy of 10 to 20% have been claimed
ChemUuminescence NO instruments appear to be specific for NO, Water vapor may
act to quench excited NO2 efficiently (Matthews et al, 1977, Folsom and Courtney, 1979)
Operation at reduced pressure reduces this problem With a commercial analyzer, a 7%
reduction in the NO signal was reported for 81 % relative humidity (RH) versus dry air
(MacPhee et al., 1976) Recent tests of eight commercial analyzers have not shown a water
vapor interference with the NO2 signal (Michie et al, 1983) With various research-grade
instruments, interference due to varying humidities has been reported to be negligible below
20 ppm water (H2O), increase by less than 10% with RH up to 2 5%, and show no change
between 2 and 100% RH (Fahey et al, 1985a, Drummond et al, 1985) Using commercial
CLM instruments, no or very small (i e , less than 2%) interferences have been reported for
6 chlorine-containing species (Joshi andBufalnn, 1978), 14 sulfur-containing species (Sickles
and Wright, 1979), 7 nitrogen-containing species, and 3 sulfur-containing species (Grosjean
and Harrison, 1985b) Zafinou and True (1986), however, do report interferences from
hydrogen sulfide (H2S) and from gases purged from anoxic waters that may have contained
sulfur compounds Using research-grade CLM instruments, Fahey et al (1985a) found no
NO interference for NO2, HNO3, N2O5, and PAN and negligible responses foi NH3,
hydrogen cyanide (HCN), N2O, methane (CH4), and nine chlonne-contaimng and three
sulfur-contauiing compounds These findings are consistent with those of Drummond et al
(1985), who report no or negligible NO interferences from NO2, HNO3, PAN, HO2NO2,
hydrogen peroxide (H^O^, propylene, H2O, and aerosols using a resezirch-gracle instrument
with a humidified O3 source
6-4
-------
From an operational perspective, aerosols can accumulate on the glass filter separating
the reaction chamber from the PM tube, causing a reduction in sensitivity (Klapheck and
Wmkler, 1985) Cleaning the filter is reported to restore the original sensitivity
Whereas most of the CLM methods discussed above are continuous, CLM has also
been used to analyze nitrogen species collected in integrated samples Gallagher et al (1985)
have taken cryosamples (4 K) of whole stratospheric air Samples were analyzed following
desorption in the laboratory for NO and NO2 usmg modified commercial CLM instruments
Braman et al (1986) have employed a series of hollow denuder tubes coated with chemicals
chosen to preconcentrate various oxides of nitrogen The collected nitrogen species are
thermally desorbed, and detected as NO with a commercial CLM instrument The coating
materials used to preconcentrate the various target species in sequence are tungstic acid
(removes HNO3), potassium-iron oxide (removes HONO), copper (I) iodide (removes NO2),
and cobalt (HI) oxide (removes NO) Future field testaig is needed to demonstrate the
adequacy of this method
6.2.2 Laser-Induced Fluorescence
Laser-induced fluorescence techniques may incorporate smgle-photon (SP), two-photon
(TP), or photofragmentation (PF) schemes Although SP-LIF has been used to measure NO
(Bradshaw et al, 1982), TP-IIF represents an advancement in the state of the art (Bradshaw
et al, 1985) and is discussed here The TP-LIF detection principle requires that a molecule
have more than one bonding excited state and can be sequentially pumped into the highest
state If the lifetime of the excited state is short compared to colhsional deactivation, the
excited molecule will decay to a more stable state by a fluorescence process The
fluorescence wavelength is shifted relative to that of the pumping wavelengths and thus
overcomes noise problems associated with background nonresonant fluorescence For
application to NO, pulsed ultraviolet (UV) and infrared (IR) laser light sources are used
Ground state X n NO is excited to the A E electronic level usmg UV light of 226-nm
wavelength Then usmg IR wavelengths of 1 06 to 1 15 j«m, the molecule is further pulsed
to the D2E level. The fluorescence resulting from the D2S to X2!! transition is monitored at
187 to 220 nm By usmg long-wavelength blocking filters with solar-blind PM tubes, this
type of detector discriminates against noise and becomes signal, rather than signal-to-noise,
6-5
-------
limited. Photon counting and grated-charge integrators are used for signal processing The
intensity of the light is related to the concentration of NO in the air sample by calibration
with atmospheres of known concentration
Because the TP-IJF instrument is signal limited, the sensitivity is defined by the
integration time (eg., 1 ppt for 5 nun and 10 ppt for 30 s) (Davis et al, 1987) Propagation
of error analysis has been used to place 90% confidence limits of ±16% on the accuracy of
TP-UF NO field measurements performed on an aircraft Also, the TP-OF technique is
expected to be highly specific because it has two levels of spectroscopic selectivity If some
trace atmospheric compound were to produce an NO molecule by interaction with the
226-nm beam, then an opportunity for interference exists Potential interference from
EDNfO3, nitromethane, mtroethane, CH3ONO2, NO2, PAN, HONO, sulfur dioxide (SO^), and
CH3ONO has been evaluated Only the last compound was found to show potential
interference, and arguments have been given to neglect its influence when sampling
tropospheric air (Davis et al, 1987)
6.2.3 Absorption Spectroscopy
Absorption spectroscopy encompasses techniques that measure the change in radiance
from a source that occurs as a result of absorption by analyte molecules over a known path
length. Several techniques, including Fourier transform IR spectroscopy (FUR), long-path
absorption, and IR tunable-diode laser spectroscopy (TOLAS), have been employed for
measuring the concentration of various NOX in the atmosphere (National Aeronautics and
Space Administration, 1983) Among these techniques, the TDLAS is a well-developed
technique that has been applied to NO as well as NO2 and HNO3 Similar sensitivities have
been reported for both remote sensing applications using open air path lengths and in situ
application using multipass cells (Cassidy and Reid, 1982) The latter configuration has
found broader application for ambient measurements of NOX, and the use of the White cell
avoids atmospheric turbulence-related errors that can affect open air application As a result,
in situ TDLAS is the primary focus of this section
Tunable-diode laser spectroscopy employs a tunable-diode laser to scan over a narrow
wavelength region around a particular absorption line or feature of the gas of interest High
sensitivity is achieved by the high spectral radiance of the diodes and the rapid tunability of
6-6
-------
the laser With rapid scanning back and forth across an absorption line, the absorption
appears as an AC signal at twice the tuning frequency And can be sensitively detected by
sychronous demodulation System sensitivities sufficient to measure signal changes of
10" permit the detection of concentrations of less than 10 ppt with a 1-km path length For
analyte molecules that have resolved absorption spectra that are not coincident with other
atmospheric constituents, TOLAS is highly specific Additional information on the operating
principles and hardware for TDLAS is provided both by Schiff et al (1983) and by National
Aeronautics and Space Administration (1983)
For optically thin systems, Beer's Law suggests that the fraction of the power
transmitted through an absorbing medium is proportional to the concentration of the
absorbing molecule However, because the total laser power is not measured, it is usually
necessary to calibrate the TDLAS by introducing a known concentration of the target gas and
determining the proportionality between signal and concentration
Using a 40-m path length near 1,850 cm"1, the MDL for NO is 0 5 ppb (Schiff et al
1983) At a sampling rate giving a 4-s residence time in the White cell, stable NO signals
are achieved in approximately 1 mm Linearity has been demonstrated between the signal
and NO concentration at levels between 7 and 175 ppb Because the NO calibration gas is
introduced directly into the sampling line, surface losses are compensated for automatically
As noted previously, the measurement of NO using TDLAS is highly specific
A newly developed method, two-tone frequency modulated spectroscopy (TTFMS), has
shown great promise in the laboratory for the measurement of NO, NO2, PAN, HNO3, ^O,
and other atmospheric trace gases (Hansen, 1989) Two-tone frequency modulated
spectroscopy uses a diode laser light source that is modulated simultaneously at two
arbitrary, but closely spaced frequencies The beat tone between these two frequencies is
monitored as the laser carrier and associated sidebands are tuned through an absorption line
The method is fast, specific, and extremely sensitive Using a low pressure (20-torr)
multiple-reflection optical cell with a 100-m path length and 1-mui signal averaging tune, the
projected MDL for NO is 4 ppt, and the projected MDLs for NO2, PAN, HNO3, and N2O
are lower Additional development of this laboratory prototype is needed to demonstrate its
performance in the field
6-7
-------
6.2.4 Passive Samplers
Whereas the previous methods are focused primarily on low NO concentrations
representative of ambient air, passive samplers are focused on atmospheres having higher
concentrations, such as those found indoors or in the workplace They are used to obtain
data at a large number of sites averaged ovei a long period of tune IHie Palmes tube is a
passive sampler that relies on diffusion of an analyte molecule through a quiescent diffusion
path of known length and cross-sectional area to a reactive surface where the molecule is
captured by chemical reaction (Palmes et al, 1976) After exposure durations ranging from
hours to days, the reactive surface is analyzed and the integrated loading of the reaction
product is used to infer the average gas concentration A quiescent diffusion path is required
to ensure that sampling is diffusion controlled, and as a result, is relatively constant This
permits the average ambient concentration to be related directly to the ratio of the reaction
product loading to the exposure time This proportionality factor is analogous to the
reciprocal of a sampling rate, where sampling rate is the product of the diffusivity of the
analyte gas and the area of the opening through which the molecules diffuse divided by the
distance they must travel to be collected
Palmes tubes are fabricated in a range of measurements from tubing (Palmes and
Tomczyk, 1979). The dimensions are chosen to provide a ratio of sampling aiea to diffusion
distance of 0.1 and thus ensure diffusion-controlled sampling. Reactive grids aire secured and
sealed at one end of the tubing segment using a plastic cap. The opposite end of the tube is
sealed with a similar cap The capped sampler is stored until the sample is to be collected
A sample is collected by placmg the tube in the appropriate location (e g , for personal
sampling the tube may be attached to a worker's lapel), removing the end cap opposite the
gndded end with the open end facing down, sampling for the appropriate period, recording
the time, recapping the tube, and returning the sampler to the laboratory for analysis
The Palmes tube passive sampler does not measure NO directly Two tubes are
required: one has reactive grids coated with tnethanolamine (TEA) to collect NO2 The
second tube is similar, but has an additional leactive surface coated with chromic acid to
convert NO to NO2, which is in turn collected by the TEA-coated grids The NO
concentration is determined by subtraction aftei correction for differences in sampling rates
caused by differences in diffusivities of the two molecules To ensure reliable results,
6-8
-------
contact between the chromic acid coated surface and the TEA-coated grids for longer than
24 h must be avoided.
Analysis is accomplished by extracting the grids in solution and analyzing the extract
for nitrite ion (NO2") This analysis may be performed by adding a solution of water,
sulfanilamide, and Af-^naphthylethylene-diamme-dmydrochloride (NEDA) reagent directly
into the tube and determining the concentration using a spectrophotometer at 540 run (Palmes
et al, 1976) Increased sensitivities are claimed by analyzing the solution using ion
chromatography (1C) with a concentrator column (Miller, 1984) The colonmetric analysis
is calibrated by dilution of gravimetncaHy prepared nitrite solutions
The sampling rate (i e , 0 02 cm3/s) for NO is reported to be independent of pressure,
to increase approximately 1 % for each 5 5 °C increase in temperature, and to increase by
3 % for each 5 cm/s increase in wind velocity (Palmes and Tomczyk, 1979) Linear response
was found for loadings between 2 and 120 ppm-h This method was proposed for sampling
occupational exposures where the dosage is not to exceed 25 ppm for 8 h (i e , 200 ppm-h)
This method cannot be used for sampling periods longer than 24 h. The reliability of this
method in the field at both parts-per-billion and parts-per-million levels remains to be
demonstrated
A badge-type sampler similar to the Palmes tube has been devised by Yanagisawa and
Nishimura (1982) Their device uses a series of 12 kyers of chromium tnoxide (CrO3)
impregnated glass fiber to oxidize NO to NO2 The filters also act as a diffusion barrier
between the ambient air and a TEA-coated cellulose fiber filter Nitric oxide is oxidized to
NO2 on the oxidizing filters and collected along with NO2 that has diffused from ambient air
through the filters to the TEA-coated collection surface The TEA-coated filter is extracted
and analyzed for NO2" Either a colonmetnc or 1C analytical finish may be employed The
analytical finish is calibrated by dilution of gravunetocally prepared nitrite solutions
An effective proportionality factor (i e , calibration factor) for the badge is provided by the
supplier This technique is claimed to be more sensitive than the Palmes tube and to have a
lower detection limit equivalent to a dosage of 0 07 ppm-h
6-9
-------
6.2.5 Calibration
ChemUuminescence, TP-LIF, and TDLAS NO measurement systems all employ
calibration cylinders containing known concentrations of NO in molecular nitrogen (N2) at
nominal concentrations of 1 to 50 ppm (Carroll et al, 1985, Bradshaw et al, 1985, Schiff
et al, 1983) Calibrations are performed using dynamic dilution with air
In the calibration procedure for the measurement of NO2 by CLM, the EPA specifies
the use of an NO concentration standard (Code of Federal Regulations, 1987a) This
standard is a cylinder of compressed gas containing between 50 and 100 ppm NO in
nitrogen. The concentration must be traceable according to a certification protocol to a
National Institute of Standards and Technology (NIST, formerly National Bureau of
Standards) Standard Reference Material (SRM) or an NIST/EPA-appioved commercially
available Certified Reference Material The National Institute of Standards and Technology
provides 10 NO SRMs at nominal concentrations between 5 and 3,000 ppm (National Bureau
of Standards, 1988) Shores and Smith (1984) have demonstrated that aluminum calibration
cylinders containing 10 to 150 ppm NO in nitrogen were stable over tune, and for 103 such
cylinders, the average change was less than 1% over an 18-mo period Commercially
supplied cylinders from 11 producers containing certified concentrations at nominal values of
70 and 400 ppm were evaluated for accuracy (Wright et al, 1987) In all cases, the certified
and auditor-measured concentrations were within 5 %, and in over two-thirds of the cases the
agreement was within 2%
Passive NO samplers do not employ full calibration of sampling and analysis operations
(Palmes et al, 1976) Only the analysis portion of the procedure is calibrated Calibration
standards for colonmetnc or 1C determination of nitrite are prepared similarly Dilution of
gravimetricaUy prepared liquid solutions of nitrite is used to produce calibration standards
that cover the working range of analysis
6.2.6 Intercomparisons
Several intercompansons of the performance of research-grade NO instrumentation
have been conducted recently (Walega et al, 1984, Hoell et al, 1985, Hoell et al, 1987,
Fehsenfeld et al., 1987). Walega et al (1984), for example, reported comparisons of NO
measurements made with a highly sensitive CLM instrument and a TDLAS system
6-10
-------
Measurements of NO-spiked synthetic air were made' in the laboratory and field In addition,
measurements were made of ambient and downtown Los Angeles air Good agreement was
found for all test conditions
Tests to compare the performance of several instruments at measuring trace gases in the
troposphere have been performed as part of NASA's Global Troposphenc Experiment/
Chemical Instrumentation Test and Evaluation 1 and 2 (GTE/CITE 1 and 2) Hoell et al
(1985 and 1987) and Gregory et al (1990a) have reported comparisons of NO measurements
made with two highly sensitive CLM instruments and a TP-LEF instrument The first
intercompanson was a ground-based study performed at Wallops Island, VA The second
intercompanson was an airborne study comprised of two missions performed on a Convair
CV-990 flown out of California and Hawaii The third intercompanson involved 13 flights
sampling tropical, nontropical, maritime, and continental air masses at altitudes between
150 and 5,000 m The two CLM instruments were of similar design (Kley and McFarland,
1980), with the mam differences being the injection of water vapor to the airstream entering
the reaction chamber of one instrument to minimize (he background variability caused by
changing ambient humidity and to suppress an O3-related background signal In the first
study, measurements of ambient NO concentrations ranged from 10 to 60 ppt and of
NO-enriched ambient air ranged from 20 to 170 ppt Agreement among the techniques at the
95% confidence level was ±30%, and no artifact or species interferences were identified
In the second and third studies, NO concentrations ranged from below 5 ppt to above
100 ppt, with the majority below 20 ppt At NO concentrations below 20 ppt, measurements
agreed to within stated instrument precision and accuracy (i e , to within 15 to 20 ppt)
Good correlation was observed between CLM and TP-LLF measurements The authors
concluded that equally valid measurements of ambient NO can be expected from either
instrument
A field intercompanson of instruments designed to measure NO, NOX, and NOy was
conducted near Boulder, CO (Fehsenfeld et al, 1987) The study was performed to compare
the performance of instruments that employed different approaches to reduce NOX or NOy to
NO prior to detection by CLM In several tests, both zero air and ambient air spiked with
NO were measured Excellent agreement was found among the measurements of the three
tested instruments These results confirm the equivalence of CLM NO detection systems
6-11
-------
6.2.7 Sampling Considerations for Nitric Oxide and Other
Nitrogen-Containing Species
Nitric oxide reacts rapidly with O3 to form NO2 In the presence of sunlight, NO2 will
photolyze to form NO and atomic oxygen, which will combine with atmospheric molecular
oxygen to form O3 Thus, under daylight conditions NO, O3, and NO2 can all exist
simultaneously in ambient air in a condition known as a photostationary state, where the rate
of photolysis of NO2 is nearly equal to the rate of reaction between NO and O3 to form
NO2. The relative amounts of the three species at any tune are influenced by the intensity of
the sunlight present at that moment When a sample is drawn into a dark sampling line,
photolysis ceases, but NO continues to react with O3 to form NO2 As a result, long
residence times in sampling lines must be avoided to ensure a representative sample
Sampling requirements for a given error in tolerance were discussed by Butcher and Ruff
(1971). Figure 6-1 shows the absolute error in NO2 introduced for a 10-s residence time in
a dark sampling line in the presence of NO and O3 at various concentrations
In addition to sampling time considerations, sampling surfaces should be considered
Oxides of nitrogen are, in general, reactive species As a result, the most nearly inert
materials (i e , glass and Teflon™) are recommended for use in sampling trams If water
molecules accumulate on sampling train surfaces and influence sample integrity, then species
solubility may be one indicator of the susceptibility of a species to surface effects.
Solubilities at 25 °C, expressed as Henry's Law coefficients (M/atm), for selected nitrogen
containing species are NO, 2 X 10~3, NO2, 1 x 10~2, N2O, 3 x 10"2, PAN, 4, HONO, 50,
NH3, 60; and HNO3, 2 X 105 (Schwartz, 1983) This suggests that of the NOX species, NO
may be the least susceptible to surface effects, whereas surface effects may be very important
in the sampling of HNO3
6.3 NITROGEN DIOXIDE
Among the NOX species, NO2 is the only criteria pollutant and the only species to have
sampling and analysis methodologies specified by the EPA for determining ambient airborne
concentrations. As a result, methods for sampling and analysis of NO2 are emphasized in
this document Airborne concentrations of NO2 can be determined by several methods,
6-12
-------
1.
Q.
CL
O
H
LU
O
O
O
CO
O
0.002 _
0.001
0.001
001 0.10
NO CONCENTRATION (ppm)
Figure 6-1. Absolute error in nitrogen dioxide (A NO2) for 10 s in the dark sampling
line
Source Butcher and Ruff (1971)
including CLM, LEF, absorption spectroscopy, and bubbler and passive collection with
subsequent wet chemical analysis
6.3.1 Chemiluminescence, Nitric Oxide Plus Ozone
Instruments discussed MI this section sample continuously and employ the CLM reaction
of NO and O3, but do not detect NO2 directly Instead, they rely on the direct detection of
NO, the conversion of some or all of the NO2 in the air sample to NO, reaction with
O3, and the appropriate signal processing to infer the NO2 concentration The CLM NO
detection principle and hardware are described in Section 621 To measure NO2, a CLM
6-13
-------
NO detector, a converter, plumbing modifications, and changes in signal processing are
required
Several methods have been employed to reduce NO2 to NO (Kelly, 1986) They
include catalytic reduction using heated molybdenum or stainless steel, reaction with CO over
a gold catalyst surface, reaction with iron sulfate (FeSO4) at room temperatuie, reaction with
carbon at 200 °C, and photolysis of NO2 to NO at 320 to 400 nm
Because CLM is designated by the EPA as the Reference Method for NO2 in ambient
air (Code of Federal Regulations, 1987a), CLM instruments for the determination of NO2 are
readily available commercially As noted previously, these instruments are used to measure
both NO and NO2 Nominal detection limits of approximately 5 ppb and response tunes on
the order of minutes are claimed by suppliers Field evaluation of nine instruments has
shown the MDLs to range from 5 to 13 ppb (Michie et al, 1983, Holland and McElroy,
1986). Recent field and laboratory evaluation of two commercial instruments operated on a
0.00- to 0.05-ppm range revealed detection limits of between 0 5 and 1 0 ppb and operating
precision estimates of ±0 3 ppb (Rickman et al, 1989) Although these performance
parameters are adequate for monitoring NO and NO2 in urban and suburban environments,
they may be inadequate in less polluted remote areas As noted in Section 621, efforts
have been reported by several researchers to improve the sensitivity and response of CLM
NO measurement technology to permit deployment in remote locations on ground-based and
airborne platforms. Because the research-grade instruments employed by these workers
included NO2-to-NO converters and were designed to measure both NO and NO2, instrument
performance for the determination of NO2 is also improved substantially over that of
commercially available instruments Typically reported performance parameters for NO2
response using research-grade CLM instruments are MDLs of 10 to 25 ppt, response times
of 1 to 100 s, and accuracy of 30 to 40% (Helas et al, 1987, Fehsenfeld et al, 1987)
Different converters may not be specific for NO2 and may convert several nitrogen-
containing compounds to NO, giving rise to artificially high values for NO2 Early in the
day, in urban areas, NO and NO2 make up most of the airborne oxides of nitrogen As the
day proceeds, these compounds are oxidized by atmospheric chemical reactions to other
species (e.g , PAN, HNO3, and other NOy compounds [see Section 6 5]) Thus the potential
for appreciable interference depends on (1) reaction time (i e , greater in the afternoon than
6-14
-------
in the morning, and greater after days of reaction than on the day of emission), (2) sampling
location and/or air mass (i e , greater in rural than in urban areas, and greater in maritime
than continental air masses), and (3) the presence of a specific interferant in a freshly
released plume
Using commercial instruments, Winer et al (1974) found over 90% conversion of
PAN, ethyl nitrate, and ethyl nitrite to NO with a molybdenum converter and similar
responses to PAN and n-propyl nitrate with a carbon converter With a stainless steel
converter at 650 °C, Matthews et al (1977) reported 100% conversion for NO2, 86% for
NH3, 82% for methylamine (CH3NH2), 68% for HCN, 1% for N2O, and 0% for N2 Using
commercial instruments, Cox (1974) reported quantitative response to HONO, and Joseph
and Spicer (1978) found quantitative conversion of HNO3 to NO with a molybdenum
converter at 350 °C Similar responses to PAN, methyl nitrate, n-propyl nitrate, n-butyl
nitrate, and HNO3, substantial response to mtrocresol, and no response to peroxybenzoyl
nitrate (PBzN) were reported with a commercial instrument using a molybdenum converter at
450 °C (Grosjean and Harrison, 1985b) These results were confirmed for PAN and HNO3
by Rickman and Wright (1986) using commercial instruments with a molybdenum converter
at 375 °C and a carbon converter at 285 °C
Interferences from species that do not contain nitrogen have also been reported Joshi
and Bufalini (1978), using a commercial instrument with a carbon converter, found
significant apparent NO2 responses to phosgene, tachloroacetyl chloride, chloroform,
chlorine, hydrogen chloride, and photochemical reaction products of a perchloroethylene-
NOX mixture Grosjean and Harrison (1985b) reported substantial responses to
photochemical reaction products of molecular chloride (C12)-NOX and Cl2-methanethiol
mixtures and small negative responses to methanethiol, methyl sulfide, and ethyl sulfide
Sickles and Wright (1979), using a commercial instrument with a molybdenum converter at
450 °C, found small negative responses to 3-methylthiophene, methanethiol, ethanethiol,
ethyl sulfide, ethyl disulfide, methyl disulfide, H2S, 2,5-dimethylthiophene, methyl sulfide,
methyl ethyl sulfide, and negligible responses to thiophene, 2-methylthiophene, carbonyl
sulfide, and carbon disulfide
With a research-grade instrument, Bollinger et al (1983) reported that NO2, HNO3,
n-propyl nitrate, and N2O5 are reduced to NO by a gold-catalyzed reaction with carbon
6-15
-------
monoxide (CO). Fahey et al (1985a), using a sunilar uistrument wilh 3,000 ppm CO over a
gold converter at 300 °C, reported conversion efficiencies exceeding 90% foi NO2, HNO3,
N2O5, and PAN Although negligible response to HCN and NH3 was found in the presence
of water vapor at a converter temperature of 300 °C, complete conversion was noted at
700 °C.
A room temperature NO2-to-NO converter using FeSO4 has been suggested by Wmfield
(1977) and adopted in research-grade instruments by Helas et al (1981), Kondo et al
(1983), and Dickerson et al (1984). A reduction in conversion efficiency has been reported
under dry conditions, and conversion of PAN, HONO, and other nitrogen-containing species
to NO has been noted (Fehsenfeld et al, 1987) Nonspecificity of the FeSO4 converter has
been observed by Fehsenfeld et al (1987) in measuring NO and NO2 in a remote
environment. At NOX levels below 1 ppb, results from the FeSO4 converter were biased
high in the measurement of NO2 (i e , a factor of 2 at 0 1 ppb) Airborne measurements of
N02 at concentrations below 0 2 ppb showed high biases of factors of 2 to 3 for a CLM
instrument with an FeSO4 converter (Gregory et al, 1990b)
In another research-grade instrument, Kley and McFarland (1980) used a xenon (Xe)
arc lamp to photolyze a portion of the NO2 in sampled air to NO and determine the NO2
concentration from the increase in NO A fractional conversion was established using a
calibration source Interferences with the photolytic converter (CLM-PC) approach are
expected from HONO, NO3, HO2NO2, and N2O5, but not from HNO3, n-propyl nitrate, and
PAN A detailed description of the operation (including minimization of mterferent
decomposition and both homogeneous and heterogeneous oxidation ol NO) and performance
of a CLM-PC instrument is given by Ridley et al (1988) An artifact identified with this
method is caused by nitrate-containing aerosols deposited on the surface of the photolysis
tube that release NO and NO2 upon irradiation (Bollinger et al, 1984) This interference is
eliminated by filtering sampled air and periodic cleaning of tube surfaces
The methods discussed above employ CLM detection of NO and are continuous Other
researchers have employed various methods of integrated sampling followed by a CLM
instrument for measuring NO and NO2 in the desorbed sample Gallagher et al (1985) have
used cryosampling of stratospheric whole air samples, and Braman et al (1986) have used
copper (I) iodide coated denuder tubes to sample NO2 in ambient air
6-16
-------
6.3.2 Chemiluminescence, Luminol
A method for the direct CLM determination of NO2 was reported by Maeda et al
(1980) This method is based on the CLM reaction of gaseous NO2 with a surface wetted
with an alkaline solution of luminol (5-ammo-2,3-dihydro-l,4-phthalazinedione) The
emission is strong at wavelengths between 380 and 520 nm. The intensity of the measured
light is proportional to the NO2 concentration in the sampled air, and the concentration can
be determined by calibration with atmospheres of known concentration
Since the introduction of the luminol method by Maeda et al (1980), improvements
have been made to develop an instrument suitable foi use in the field (Wendel et al, 1983),
and additional modifications have recently been made to produce a continuous commercial
instrument (Schiff et al, 1986) Detection limits of 5 to 30 ppt and a response time of
seconds have been claimed based on laboratory tests (Wendel et al, 1983, Schiff et al,
1986) Recent laboratory evaluation of two instruments has revealed a detection limit (i e ,
twice the standard deviation of the clean-air response) of 5 ppt and 95 % rise and fall tunes of
110 and 15 s (Rickman et al, 1988) Field tests of Ihe same instruments have shown an
operating precision of +0 6 ppb
The original method showed no interferences from NO, N2O, NH3, CO, propylene,
and 1,2-dichloroethylene, but positive interferences from O3 and SO2 and a negative
interference from carbon dioxide (CO2) (Maeda et al , 1980). More recently, the luminol
solution has been reformulated, containing H2O, lumcnol, sodium hydroxide (NaOH), sodium
sulfite, and alcohol in proportions chosen to enhance the sensitivity and minimize
interferences from O3, SO2, and CO2 (Wendel et al, 1983, Schiff et al, 1986)
At concentrations below 100 ppb, no interferences were reported for HNO3, NH3, HCN,
H2O2, CO, CO2, and SO2 Negative interferences of 1 2 to 5 % have been reported for
NO (Schiff et al, 1986, Spicer et al, 1991) Also, the instrument has shown sensitivity to
PAN (Wendel et al, 1983, Sickles, 1987), different sensitivities to HONO (Rickman et al,
1989, Spicer et al, 1991), and nonlinear response to NO2 (Schiff et al , 1986, Kelly et al,
1990, Spicer et al, 1991) Furthermore, the method has shown appreciable sensitivity to an
operating temperature that can be resolved by controlling the temperature of the reaction cell
or by signal processing (Schiff et al, 1986, Bubacz et al, 1987) Recent tests of the CLM
(luminol) instrument have demonstrated the need to correct results for pressure (as might be
6-17
-------
seen in airborne applications), nonlineanty of response below 3 ppb NO2, interferences from
O3 and PAN, and the age-dependent sensitivity of the luminol solution (Kelly et al, 1990)
Lastly, a manufacturer-supplied O3 scrubber designed to eliminate the O3 interference was
also found to remove appreciable amounts of NO2
6.3.3 Photofragmentation/Two-Photon Laser-Induced Fluorescence
Two NO2 sensors based on measurement of fluorescence of excited NO2 have been
reported by Fincher et al (1978) One device employs a small, high-pressure xenon arc
flash lamp to excite NO2 This device has a sensitivity of 10 ppb for an 80-s (i e ,
1,024 flash-lamp pulses) integration time The other device uses a 442-nm IIP and a
PM tube with photon counting of light above 600 nm The sensitivity of this device is 1 ppb
for a comparable integration time A major drawback of these devices for broad ambient
applications is limited sensitivity associated with background signals This has been
overcome with PF/TP-LIF instrumentation
An NO2 sensor incorporating PF/TP-LIF has recently been developed and deployed in
the field (Rodgers et al., 1980, Davis, 1988) With this method, NO ts measured in one cell
using TP-UF (see Section 622) A xenon fluoride excimer laser with output at 353 nm is
used in a second cell to photolyze NO2 The total NO signal in the second cell resulting
from ambient and photofragment NO is measured as NO using TP-OF The NO2
concentration is determined from the difference in signals of the two reaction cells and the
fractional photolysis of NO2 The NO2 fluorescence cell is calibrated using calibration
sources of NO and NO2
Because the PF/TP-LEF instrument is signal-limited, the sensitivity is defined by
integration time The detection limit for NO2 for a 2-min integration f ime is 12 ppt (Davis,
1988). The accuracy of PF/TP-UF NO2 determinations is likely to be similar to the +16%
reported for TP-UF NO measurements (Davis et al, 1987) At 15 ppt NO and 50 ppt NO2,
the precision of NO2 determinations is given as ±17% (Gregory et al, 1990b)
The PF/TP-UF technique is expected to be highly specific for N02 In addition to
those potential NO interferents with TP-OF that are discussed in Section 622, other species
that could photolyze or otherwise decompose to produce NO or NO2 have been considered
6-18
-------
(Davis, 1988) Arguments were given to dismiss HNO3, HO2NO2, N2O5, CH3ONO, and
CH3ONO2 as possible interfering species
6.3.4 Absorption Spectroscopy
Absorption Spectroscopy is discussed in a previous section for NO (Section 623)
Absorption methods may measure the absorption of light in the UV, visible, or IR regions of
the electromagnetic spectrum They may employ closed cells for in situ measurements (e g ,
TDLAS) or open paths for remote sensing Absorption methods require a source of
radiation Active methods utilize artificial light from a source such as an incandescent lamp
or a laser, whereas passive methods use natural light from the sun or moon Laser sources
offer advantages of scanning, narrow spectral width, and high intensity, and as a result,
usually provide better sensitivity than nonlaser sources In this section, several absorption
techniques are addressed, including the in situ methods, TDLAS, photometry, and ITEMS,
as well as remote sensors employing long-path absorption differential optical absorption
Spectroscopy (DOAS) and differential absorption lidar (DIAL)
Tunable-diode laser Spectroscopy is a well-developed technique that has been used to
measure NO2 as well as other species in the atmosphere Descnptive information about the
operating principle is given in Section 623 and the references therein With a 150-m path
length near 1,600 cm"1, the MDL is 0 1 ppb and the accuracy is ±15% (Mackay and Schiff,
1987) For a 40-m path length, the MDL for NO2 is 0 5 ppb (Schiff et al, 1983) At a
sampling rate giving a 4-s residence tune in the White cell, stable NO2 signals are achieved
in approximately 1 mm Linearity has been demonstrated between the signal and NO2
concentration at levels between 35 and 175 ppb Surface losses are compensated for
automatically because the NO2 calibration gas, typically from a permeation tube, is
introduced directly into the sampling line Tunable-diode laser Spectroscopy is a
spectroscopic technique, as a result, the measurement of NO2 using this method is highly
specific
A prototype instrument using an in situ absorption technique to measure NO2 was
recently reported (Jung and Kowalski, 1986) This technique employs a modified
commercial O3 photometric analyzer to measure the absorption of visible light by NO2 at
wavelengths longer than 400 nm The signal obtained m a 1 12-m absorption cell from
6-19
-------
unscrubbed ambient air is compared with that from ambient air scrubbed of NO2
A microcomputer uses Beer's Law and an absorption coefficient derived from an NO2
calibration source to determine the NO2 concentration Interferences from NH3, NO,
O3, SO2, and PAN have been shown to be negligible Comparisons with commercially
available CLM analyzers monitoring smog chamber experiments and ambienl air have shown
good agreement when NO2 was expected to be present The CLM signal was found to
exceed that of the photometer when photochemical reaction products such as PAN were
believed to be present Although noise of less than 3 ppb and linear response have been
demonstrated between 100 and 700 ppb, additional development and evaluation are needed to
permit routine use of this technique to ambient monitoring applications
Cryogenic sampling at 77 K combining the matrix-isolation technique (i e , solid CO2)
with FUR. spectroscopy has shown promise for the sensitive determination of NO2, PAN,
HNO3, HONO, and N2O5 (Griffith and Schuster, 1987) A theoretical MDL of 5 ppt was
claimed for NO2 in 15-L integrated samples of ambient air
A laboratory prototype method, TTFMS, has been developed recently and is described
in Section 6.2.3 (Hansen, 1989) This method has a projected MDL for NOl of 0 3 ppt
Long-path absorption is a remote sensing technique that is typically operated over open
terrain with optical path lengths of up to tens of kilometers The absorption cross section for
each species of interest is determined in the laboratory This information is used to convert
optical densities measured in the field to concentration data For 1-h observation periods, an
MDL of 20 ppt has been reported using an artificial light source and a 9 2-km path length
(Johnston and McKenzie, 1984) Noxon (1978) used the sun as a light source and the
structure in the NO2 absorption spectrum near 440 nm to obtain measures of the abundance
of NO2 in both the troposphere and the stratosphere Platt and Perner (1983) have reported
the application of DOAS to the determination of several nitrogen-containing species
A xenon high-pressure lamp or an incandescent lamp was used with a 1- to 10-km path
length Selected applicable compounds, detection limits, and target wavelengths are NO,
400 ppt, and 226 nm, NO2, 100 ppt, and 363 nm, HONO, 20 ppt, and 354 nm, and NO3,
0.5 ppt, and 662 nm The DOAS technique has been recently adapted to employ a 25-m
multipass open reflection system with a path length of up to 2 km (Biermann et al, 1988)
Using a 0.8-km path length and 12-min averaging tunes, MDLs and accuracies for NO2,
6-20
-------
HONO, and NO3 of 4 ppb (±10%), 0 6 ppb (+30%), and 20 ppt (+15%), respectively,
have been reported
Common remote sensing techniques employ light detection and ranging (hdar)
Differential absorption hdar is a long-path absorption technique This method employs light
of two wavelengths propagated over a given distance at a given intensity The concentration
of the gas species of interest is related to the difference ui intensities of the two wavelengths
at the receiver Differential absorption lidar techniques have been applied to the ambient
measurement of both NO (Alden et al, 1982) and NO2 (Frednksson and Hertz, 1984; Edner
et al, 1987) Baumgartner et al (1979) reported a 5-ppb sensitivity for NO2, and Staehr
et al (1985) reported a 10-ppb sensitivity for NO2 using a laser source and a 6-km path
length Differential absorption hdar methods are in the development stages for monitoring
NO2 in ambient atmospheres
6.3.5 Wet Chemical Methods
Most wet chemical methods for measuring NO2 involve the collection of NO2 in
solution followed by a colorimetnc finish using an azo dye Many variations of this method
exist, including both manual and automated versions Szonntagh (1979) traced the history of
azo dye methods for NO2 sampling and analysis Nitrogen dioxide, first collected in aqueous
solution, is thought to form HONO An aromatic amine is used in the presence of an acid to
react with HONO and form a diazonium salt The salt then rearranges and couples with
another organic amine that has been added to form a red azo dye The intensity of the color
is proportional to the NO2 collected and is measured using a spectrophotometer A good
overview of wet chemical methods for sampling and analysis of NO2 is given by Purdue and
Hauser (1980)
6.3.5.1 Griess-Saltzman Method
In this method, air is sampled for no longer than 30 mm through a fritted bubbler that
contains the Griess-Saltzman reagent (Saltzman, 1954) This reagent is a solution of
sulfanilic acid, NEDA, and acetic acid Color development is complete within 15 mm and is
measured at 550 nm within an hour. Interferences fiom SO2 and PAN have been noted, but
are usually too low in ambient air to cause significant error Concentrations of NO2 ranging
6-21
-------
from 20 to 800 ppb for 30-min sampling periods may be determined using this method
An MDL of 2 ppb and a precision of ±11%, as well as a positive bias of 18% for spiked
ambient air, have been reported (Saltzman, 1980, Purdue and Hauser, 1980)
Calibration is usually performed statically using dilute solutions of sodium nitrite
Saltzman (1954) reported that 0 72 moles of nitrite were formed for each mole of NO2
absorbed. Values of this "stoichiometnc factor" ranging from 0 5 to 1 0 have been reported
(Crecelius and Forwerg, 1970). The method can, however, be cahb rated dynamically using
calibrated NO2 gas standards
6.3.5.2 Continuous Saltzman Method
The measurement principle used in the Continuous Saltzman Method is based on the
Griess-Saltzman reaction Ambient air is sampled continuously through a gas-liquid
contactor, and NO2 is collected on contact with an absorbing solution containing diazotizing-
coupling reagents After the color has developed, the absorbance of the solution is measured
continuously with a spectrophotometer Because O3 was found to act as a negative
interferent, this method was eliminated as a candidate for designation as an Equlvalent
Method by EPA. The MDL is 10 ppb, the bias ranges from +3 to +15% at NO2
concentrations between 30 and 150 ppb, and the precision is approximately ± 12 % (Purdue
and Hauser, 1980). Although calibration can be performed using nitrite solutions or
calibrated NO2 gas standards, the latter approach is recommended
6.3.5.3 Alkaline Guaiacol Method
Various alternatives to the Griess-Saltzman Method have been proposed Recently,
Baveja et al. (1984) proposed a method using alkaline o-methoxyphenol (guaiacol) as both
the absorbing medium and coupling reagent Samples are collected using fatted bubblers
containing an alkaline guaiacol solution After sampling, ^-nitroaniline is added, the pH is
adjusted with hydrochloric acid (HC1) and later with NaOH, the resulting dye is extracted in
amyl alcohol, and the absorbance is read at 545 nm Collection efficiency of 98 % and a
stoichiometric factor of 0.74 were reported
6-22
-------
6.3.5.4 Jacobs-Hochheiser Method
This method is a modified version of the Gness-Saltzman method to permit 24-h
sampling and delay in analysis times (Jacobs and Hochheiser, 1958) Samples are collected
by bubbling ambient air through a 0 1 N aqueous NaOH solution using a fatted bubbler
The collected nitrite is then reacted with sulfamlamide and NEDA in acid media to form an
azo dye, which is measured with a spectrophotometei at 540 nm As with the Gness-
Saltzman method, dilute nitrite solutions are used for calibration This method was the
original Reference Method designated by the EPA foi NO2 (Purdue and Hauser, 1980)
However, testing of the method showed that the originally specified NO2 collection efficiency
of 35 % was not constant and that it varied nonlinearly with NO2 concentration In addition,
interferences from NO and combinations of NO and NO2 were found As a result, this
method was withdrawn in 1973 and is considered unacceptable for air sampling and analysis
6.3.5.5 Sodium Arsenite Method (Manual and Continuous)
This method has been designated by the EPA as an Equivalent Method in both the
manual and continuous forms (Federal Register, 1986) The manual method is a 24-h
integrated method similar to the Jacobs-Hochheiser method, except that sodium arsenite is
added to the aqueous NaOH absorbing solution, and atn orifice bubbler is used The nitrite is
reacted with sulfamlamide and NEDA in acid media to form an azo dye, which is determined
with a spectrophotometer The continuous method employs the same measurement principle,
but uses hardware to permit continuous determination of NO2 in a manner similar to that for
the Continuous Saltzman Method
The overall NO2 recovery is 82% Although NO and CO2 may act as interferents,
their impact is minimal at typical ambient levels Sulfur dioxide has not been tested as an
interferent The MDL is 5 ppb, the bias is —3%, independent of concentration, and the
precision at NO2 concentrations between 30 and 150 ppb is +6 ppb (Purdue and Hauser,
1980) Recently, HONO was found to respond equivalently to NO2 (Braman et al, 1986)
This interference is likely to be appreciable in urban environments during nighttime hours,
where concentrations above 5 ppb have been observed (Rodgers and Davis, 1989, Appel
et al , 1990)
6-23
-------
6.3.5.6 Triethanolamine-Guaiacol-Sulfite Method
This method, designated by the EPA as an Equivalent Method (Federal Register, 1986),
is a manual 24-h integrated method Samples are collected using orifice bubblers and a
solution of TEA, guaiacol, and sodium metabisulfite The resulting nitrite is reacted with
sulfanilamide and 8-amino-l-naphthalene-sulfonic acid ammonium salt (ANSA), and the
resulting azo dye is determined at 550 nm with a spectrophotometer No interferences were
found in tests with NH3, CO, formaldehyde, NO, phenol, O3, and SO2 The overall NO2
recovery is 93%; the MDL is 8 ppb, the bias is —5%, independent of concentration, and the
precision at NO2 concentrations between 30 and 150 ppb is ±6 ppb (Purdue and Hauser,
1980).
6.3.5.7 Triethanolamine Method
This method is a manual 24-h integrated method (Ellis and Margeson, 1974) Samples
are collected using an aqueous solution of TEA and fritted bubblers As with the Equivalent
Methods, the resulting nitrite may be determined with a spectrophotometer after reaction with
sulfanilamide and either NEDA or ANSA Although recoveries of 80 to 90% were found at
NO2 concentrations between 20 and 350 ppb using fritted bubblers, only 50% recovery was
found using orifice bubblers. Because the requirement of a fatted rather than an orifice
bubbler was considered to be a major disadvantage by the EPA, the development of this
method as an Equivalent Method was terminated (Purdue and Hauser, 1980)
Triethanolamuie has been used as the collection medium for many active and passive
techniques to sample NO2 Although colonmetry may be used as the analytical finish, 1C
recently appears to be the method of choice (Miller, 1984) Vinjamoori and Ling (1981)
have used an aqueous solution of TEA, ethylene glycol, and acetone on 13X molecular sieves
to sample air in the workplace for NO2 Passive devices employing TEA have been used for
industrial hygiene and indoor air quality sampling (Palmes et al, 1976, Wallace and Ott,
1982) as well as for ambient applications (Sickles and Michie, 1987, Muhk and Williams,
1987). Recently, a method using Whatman GF/B filters coated with an aqueous solution of
TEA, ethylene glycol, and acetone has been developed for extended sampling of both NO2
and SO2 from ambient air (Sickles et al, 1990) This method, using 1C to determine the
collected NO2 as nitrite and nitrate, showed no interferences from NO, NH3, 03, H2S,
6-24
-------
methanethiol (CH3SH), and carbon disulfide (CS^, bul showed major uiterferences from
PAN and HONO Recovery of NO2 averaged 87% in laboratory tests at concentrations
between 5 and 400 ppb Using two coated 47-mm diameter filters in series, filter
temperatures above 5 °C, and flow rates below 2 L/mm are required to insure good NO2
collection efficiency This method has been incorporated into the design of a prototype
sampler known as a transition flow reactor (TFR) system for measuring acidic deposition
components (Knapp et al, 1986)
6.3.6 Other Active Methods
Several other methods reported for the determination of NO2 include lomzation
spectroscopy, mass spectrometry, photothermal detection, denuders, and solid sorbents
However, because they are in the early stages of development or are not being used widely,
they are only briefly mentioned here
lomzation spectroscopy is a new and sensitive in situ laser technique that is currently
under investigation for troposphenc measurements of NO and NO2 (National Aeronautics and
Space Administration, 1983). This method is in the early stages of development
Atmospheric pressure lonization mass spectrometry has been investigated for the
continuous measurement of NO2 and SO2 in ambient air (Benoit, 1983) An MDL of
approximately 0 5 ppb was reported.
Methods employing photothermal detection of NO2 have been reported (Poizat and
Atkinson, 1982, Higashi et al, 1983, Adams et al, 1986) Detection is accomplished by
selectively exciting transitions of NO2 with a chopped continuous wave or pulsed laser
source At pressures near atmospheric, collisional de-excitation converts the absorbed energy
into translational energy, leading to a temperature rise along the beam and expansion of the
gas The resulting change in refractive index of the thermal lens can be detected by
spectroscopic means (Higashi et al, 1983), or the resulting pressure change can be detected
with a microphone (Poizat and Atkinson, 1982) Minimum detection limits of 2 to 5 ppb
have been reported. Development of photoacoustic methods for continuous application may
be limited by acoustic noise associated with vibration and flow
A denuder is a tube or channel with its walls coated with a chemical chosen to remove
a gaseous species of interest from a sample drawn through the tube under laminar conditions
6-25
-------
The concentration of the species of interest is determined by measuring the amount of the
species collected on the walls or by comparing the signal strengths in Ihe presence and
absence of the denuder Possanzim et al (1984) recommended using a coating containing KI
to collect NO2. Subsequent tests showed the collection efficiency of this material to be
dependent on the humidity of the sample (SicHes, 1987) An alkaline guaiacol coating on
annular denuders has shown high collection efficiency for NO2 (Buttim et al, 1987) After
extraction using deiomzed water, 1C analysis showed quantitative recovery as NO2" The
3
reported MDL was 0 13 ppb for a 1-m sample (i e , 16 L/min for 1 h), and the median
precision for 14 paired samples was 3 9 % expressed as relative standard deviation (RSD)
A direct interference was noted with HONO, none was found with NO or PAN, and no
humidity effects were observed between 20 and 80% RH No interference tesls were
performed with O3, however, comparison of 4-h results with those of a commercial CLM
(NO + O3) instrument sampling air near Rome showed good correlation (i e , r = 0 92)
Although alkaline guaiacol solutions degrade with tune, no degradation effects were found
for 72-h presampling or 24-h postsampling denuder storage Longer duration storage tests
(e.g., 2 weeks pre- and postsampling) and additional field evaluations are needed before this
method is ready for routine application
Recently, Adams et al (1986) found activated manganese dioxide to be effective under
unspecified RH for removing NO2 in denuder applications Additional tests are needed
under conditions representative of the ambient environment before this approach is ready for
routine application.
Lipari (1984) has used a commercially available cartridge containing the sorbent Flonsil
(magnesium silicate) coated with diphenylamine to sample NO2 in ambient and indoor air
The NO2 reacts with the diphenylamine to form mtro- and mtroso-denvatives, which are
subsequently detected by high performance liquid chromatography (HPLC)-UV Although
no interference from NO, O3, SO2, HNO3, and water vapor were found, PAN produced a
50% positive interference, and HONO was expected to interfere quantitatively Sorbent
temperature must be held below 32 °C to prevent volatilization of the diphenylamine and
nonquantitative sampling. An MDL of 0 1 ppb for a 2 0-m air sample was claimed The
method shows promise for the sensitive determination of NO2 under conditions where the
noted interferents and temperature sensitivity do not pose problems
6-26
-------
6.3.7 Passive Samplers
Passive samplers are frequently used in industrial hygiene, indoor air, and personal
exposure studies, and are less frequently used in ambient air sampling Passive NO samplers
were described in Section 624 Namiesnik et al (1984) have provided a good overview of
passive samplers The basis for all passive samplers is the same The gaseous analyte
molecule is transported from the bulk air to a reactive surface, where the molecule impinges
and is captured by chemical reaction After exposure periods ranging from hours to days,
the reactive surface is analyzed and the integrated loading of the reaction product is used to
infer the average gas concentration present during the sampling period When the transport
of analyte molecules to the reactive surface is diffusion-controlled, the average ambient
concentration may be related directly to the ratio of the product loading to the sampling
duration This proportionality is defined as a sampler calibration factor or alternatively as
the reciprocal of the sampler sampling rate
One type of passive NO2 sampler for ambient applications is the nitration plate It is
essentially an open petri dish containing TEA-impregnated filter paper Thus, there is no
diffusion barrier between the ambient air and the NO2 collection surface Nitrogen dioxide
reacts with the TEA and is retained primarily as NO2 , which can be extracted and
determined with a spectrophotometer or by 1C A smgle calibration factor is provided by the
manufacturer A recent study indicated that the calibration factors determined experimentally
for the device are extremely sensitive to wind speed, NO2 concentration, and temperature
(Sickles and Michie, 1987) Triethanolamine is expected to collect not only NO2 but
HONO, HNO3, and PAN (Sickles, 1987) These results suggest that nitration plates may be
useful only as qualitative indicators of ambient levels of NOX
Another open-surface device has been proposed by Kosmus (1985) for ambient
applications This device uses chromatographic paper in the shape of a candle that is coated
with diphenylamine and is continually impregnated with a potassium thiocyanate-glycenn
catalyst Nitrogen oxides, presumably NO2 and NO to some extent, are collected by a
catalyzed reaction with diphenylamine to form the nitrosamine After extraction, the
nitrosamine may be determined with a spectrophotometer or by differential pulse
polangraphy Interference by iron oxide particles was noted, and interference from both
PAN and HONO is expected (Lipari, 1984) Sensitivity to wind speed, as noted previously
6-27
-------
for the nitration plate, is also expected Collocated sampling was performed using four
candles and a CLM NO/NO2 analyzer at each of 14 stations (Kosmus, 1985) The
nitrosamine loadings were highly correlated, but in a nonlinear manner, with the sum of
100% NO2 plus 90% NO from the CLM instruments These results also suggest that open-
surface passive samplers may be useful as qualitative indicators of ambient levels of NOX
Mulik and Williams (1986) have adapted the nitration plate concept by adding diffusion
barriers in their design of a passive sampling device (PSD) for NO2 in ambient and personal
exposure applications The physical configuration employs a TEA-coated cellulose filter that
uses two 200-mesh stainless steel diffusion screens and two stainless steel perforated plates
on each side of the coated filter to act as diffusion barriers This design permits NO2
collection on both faces of the filter After sampling, the filter is removed from the PSD,
extracted in water, and analyzed for NO2~ by 1C A sensitivity of 0 03 ppm-h and a
sampling rate of 2 6 cm /s were claimed Comparison of PSD results with CLM
determinations of NO2 in laboratory tests at concentrations between 10 and 250 ppb were
linearly related and highly correlated (i e , r = 0 996) The device exhibited increased
sampling rates of approximately 50% as the wind speed increased from 20 to 45 cm/s, but
displayed a relatively constant sampling rate at wind speeds between 45 and 300 cm/s (Malik
and Williams, 1987) Interferences from PAN and HONO are expected (Sickles, 1987)
Results of TDLAS and triplicate daily PSD NO2 measurements in a recent 13-day field study
showed good agreement between the study average values, but a correlation coefficient for
daily results of only 0.47 (Mulik and Williams, 1987, Sickles et al, 1990) Further
development and testing of the PSD appears warranted
The Palmes tube is a passive device that has been used to sample air in the workplace
and indoor environments to assess personal exposure to NO2 (Palmes et al, 1976, Wallace
and Ott, 1982) This device and its operation were described in detail in Section 624
It consists of a tube, open at one end with a TEA-coated interior surface on the closed end
Nitrogen dioxide diffuses through the 7 1-cm diffusion length, where it is captured by TEA
Analysis is accomplished by extracting the TEA-coated surface and analyzing the extract for
N02~. This may be done directly by adding an aqueous solution of sulfamlamide and NEDA
to the tube and determining the NO2" concentration using a spectrophotometer at 540 nm
(Palmes et al., 1976) A stoichiometnc factor of unity, a linear response for dosages
6-28
-------
3
between 1 and 30 ppm-h, and a sampling rate of 0 02 cm /s are reported An improvement
in sensitivity from 0 3 ppm-h to 0 03 ppm-h is claimed if the aqueous TEA extract is
analyzed by 1C using a concentrator column (Mulik and Williams, 1986). Absorption and
desorption of NO2 by the internal walls of the acrylic tube have been reported to limit
applications to exposures of 0 1 ppm-h (Miller, 1988) This sensitivity can be unproved to
0 03 ppm-h by using stainless steel rather than acrylic tubes The device exhibited sampling
rates increased by up to 15 % as wind speed was increased from 50 to 250 cm/s (Palmes
et al., 1976) and decreased by 15% as temperature was reduced from 27 to 15 °C (Girman
et al, 1984) Interferences from PAN and HONO are expected (Sickles, 1987) The
calibration factor for the Palmes tube is theoretically derived, and the analytical finish is
calibrated by dilution of gravimetncally prepared nitrite solutions
The performance of the Palmes tube has been compared with that of two commercially
available passive personal samplers, the DuPont Pro-Tek and the MDA Chronotox System
(Woebkenberg, 1982) The Palmes tube displayed greater sensitivity than either of the
commercial samplers and displayed adequate precision and accuracy at loadings between
1 and 80 ppm-h Because the commercial devices may be used at only moderate to high
loadings (i e , above 5 ppm-h), they are not sufficiently sensitive for most ambient or
personal exposure applications They are thusly not discussed further in this document
A badge-type NO2 personal sampler has been devised by Yanagisawa and Nishirnura
(1982) Their device uses a series of five layers of hydrophobic Teflon™-type filter material
as a diffusion barrier between ambient air and a TEA-coated cellulose fiber filter Nitrogen
dioxide diffuses through the hydrophobic filters to the TEA-coated surface, where it is
collected Following extraction of the TEA-coated filtei in a solution of sulfanilic acid,
phosphoric acid, and NEDA, a colonmetric finish at 540 nm is employed A sensitivity of
q
0 07 ppm-h, a sampling rate of 1 4 cm /s, and an accuracy of ±20% are claimed The
device exhibited increased sampling rates of up to 30% as the wind speed was increased from
15 to 400 cm/s Interferences from PAN and HONO are expected (Sickles, 1987) The
calibration factor for the sampler is provided by the supplier, and the analytical finish is
calibrated by dilution of gravimetiically prepared nitrite solutions
A variation on the above approach has been proposed by Cadoff and Hodgeson (1983)
The sampler is comprised of a Nuclepore 47-mm filter holder with a capped base containing
6-29
-------
a TEA-coated glass fiber filter and a 0 8-/on pore size polycarbonate filter The
polycarbonate filter and the air space between this filter and the TEA-coated filter act as a
diffusion barrier. A colonmetnc analytical finish is employed The performance was tested
at NO2 loadings between 0 06 and 1 ppm-h, and a sampling rate of 1 9 cm3/s was claimed
West and Reiszner (1978) proposed a passive NO2 sampler using a silicone membrane
as a diffusion barrier between ambient air and an alkaline thymol blue NO2 collection
solution. Collected NO2 is converted to NO2" and determined colonmetncally Results of a
field comparison with the EPA-designated TGS method were not favorable and showed the
proposed device to yield results approximately a factor of three higher than the TGS method
As a result, this method is not recommended
6.3.8 Calibration
Two methods have been designated by the EPA (Code of Federal Regulations, 1987a)
as alternative calibration procedures for the measurement of NO2 in the atmosphere These
methods use permeation tubes or gas phase titration (GPT) to generate known amounts of
NO2. Calibrations are performed using dynamic dilution with air
A permeation tube is a porous, inert tube usually made of Teflon™ that has been
partially filled with liquid NO2 and sealed Permeation of NO2 through the porous tube will
occur at a constant rate if the temperature of the tube and the NO2 concentration gradient
across the tube are held constant The tube is maintained at a constant temperature
(±0.1 °C), and a measured flow of a dry earner, usually nitrogen, is passed over it The
NO2 permeating through the porous wall and entering the earner stream is diluted with zero
air to produce calibration NO2 atmospheres of known concentrations The permeation tube
is calibrated gravimetncally by measuring the weight loss of the tube over time The
National Institute of Standards and Technology (NIST) provides SRM permeation tubes that
emit NO2 at a nominal rate of 1 /ig/min (National Bureau of Standards, 1988) Additional
information on the performance of NO2 permeation tubes is given by Hughes et al (1977)
A recent report by Braman and de la Cantera (1986) indicates that permeation sources of
NO2 can produce atmospheres that are contaminated with other oxides of nitrogen, including
HNO3, HONO, and NO Further work appears warranted to define the conditions where
permeation devices may be used to provide an unambiguous source of NO2
6-30
-------
Gas phase titration employs the rapid, quantitative gas phase reaction between NO,
usually from a standard gas cylinder, and O3, from a stable O3 generator, to produce one
NO2 molecule for each NO molecule consumed by reaction When O3 is added to excess
NO in a titration system, the decrease in NO (and O3) is equivalent to the NO2 produced
Different amounts of NO2 may be produced by adding different amounts of O3 When the
NO concentration and the flow rates entering the dynamic titration system are known
accurately, the NO2 concentration leaving the system can be determined accurately The
accuracy and stability of NO standard gas cylinders are described in Section 625
A third source of NO2 sometimes used for calibration is a cylinder of compressed gas
containing NO2, usually in N2 (Fehsenfeld et al, 1987, Davis, 1988) Calibrations are
subsequently performed using dynamic dilution with zero air These cylinders are
commercially available, and the NO2 concentration should be referenced to an accepted
standard Bennett (1979) has shown that, of 26 aluminum cylinders initially containing
supplier-certified concentrations of NO2 in N2 between 100 and 300 ppb, 10 showed modest
declines in NO2 concentration during the first 3 mo after preparation The NO2 levels in all
26 cylinders declined substantially over the 10-mo study period Schiff et al (1983) have
noted problems handling trace concentrations of NO2 from a cylinder A cylinder containing
9 ppm NO2 in N2 gave 15 % higher readings for NO2 when analysis was performed by CLM
than by TOLAS (Walega et al, 1984) This discrepancy may have been due to an impurity
(e g , HNO3) in the cylinder that could act as an interference with the CLM, but not the
TOLAS, determination of NO2 Davis (1988) examined a cylinder containing 44 ppm NO2
in air at regular intervals over 3 years and observed a 16% change in concentration. In view
of these findings, caution should be exercised if a cylinder containing NO2 in N2 or air is to
be employed as a calibration source of NO2
6.3.9 Intel-comparisons
Several intercomparisons of research-grade NO2 instrumentation have been conducted
(Helas et al, 1981, Walega et al, 1984, Sickles et al, 1990, Fehsenfeld et al, 1987, 1990,
Gregory et al, 1990b) and are described in this section Also, the performance of EPA-
Designated Methods, based on intercompansons and other studies, is discussed in
Section 6 3 10
6-31
-------
Helas et al (1981) report the results of a field intercompanson of several NO2 methods
conducted in April 1979 at Deuselbach, Germany Good agreement between a highly
sensitive CLM instrument and long-path absorption was found over the 1- to 8 ppb range of
observed NO2 concentrations
Walega et al (1984) report comparisons of NO2 measurements from a highly sensitive
CLM instrument using a thermal NO2 to NO converter with NO2 measurements from a
TOLAS system. Measurements of NO2-spiked synthetic air conducted both in the laboratory
and in the field showed good agreement Measurements were made of ambient and captive
air in downtown Los Angeles and showed maximum respective concentrations of 100 and
600 ppb. Chemiluminescence results were appreciably higher than those of the TDLAS
this difference averaged 18% in the ambient-air studies and 15% in the captive-air studies
In the latter studies, the agreement was generally within 10% in the morning, but by the end
of the day, could be as large as 80% This behavior was attributed to the leaction of NO2
and the accumulation of photochemically produced CLM interferents such as PAN that
occurred during the day.
Daily NO2 concentrations determined by TDLAS, CLM (luminol), and PSDs were
reported recently from a 13-day study conducted at Research Triangle Park, NC, in the fall
of 1986 (Bubacz et al, 1987; Mulik and Williams, 1987, Sickles et al, 1990) Collocated
sampling was performed using a TDLAS system, two CLM (luminol) instruments, and
triplicate daily PSDs. Daily average results were computed for the TDLAS, each CLM
(luminol) instrument, and the PSDs The 13-day average values from the CLM (luminol)
instruments and the TDLAS system agreed to within 2 ppb, the average daily ratios of NO2
by CLM (luminol) to TDLAS were 1 01 + 011 (standard deviation) and 1 19 ± 0 17, the
respective correlation coefficients were high, 0 94 and 0 91, and although the results of one
CLM (luminol) instrument showed no bias, results of the other were biased higher than those
of the TDLAS system The 13-day average values from the PSDs and TDLAS system
agreed to within 1 ppb; the average daily ratio of NO2 by PSD to TDLAS was 1 08 ± 0 32
There was no apparent bias, but the correlation coefficient was only 0 47
A field intercomparison of instruments designed to measure NO, NOX, and NOy was
conducted near Boulder, CO (Fehsenfeld et al, 1987) In addition, an intercomparison of
NO2 measurements was performed using two different NO2-to-NO converters prior to NO
6-32
-------
detection by CLM. The two CLM detection systems were tested and found to be equivalent
One instrument used a photolytic NO2-to-NO converter, whereas the other employed a
FeSO4 7H2O surface converter In spiking tests, the instrument with the FeSO>4 converter
responded to NO2, PAN, and n-propyl nitrate, but not to HNO3 or NH3 The CLM-PC
instrument responded to NO2, but not significantly to HNO3, NH3, w-propyl nitrate, or PAN
For measurements of NO2 + NO in ambient air, results from the two instruments agreed at
concentrations above 1 ppb However, results from the instrument with the surface converter
were biased higher than those from the photolytic converter at lower NO + NO2
concentrations This discrepancy was a factor of 2 at 0 1 ppb These results suggest that
surface converters sufficiently active to convert NO2 to NO can convert other NOX species
such as PAN to NO Although the use of CLM with surface NO2-to-NO converters may not
pose a problem in many urban and suburban areas where NO and NO2 are expected to be the
dominant NOX, results cited here and elsewhere in this section suggest that surface converters
are unsuitable for the mterference-free measurement of NO2 in ambient air containing PAN
and similar compounds
Fehsenfeld et al. (1990) performed a ground-based intercompanson of NO2
measurements using CLM-PC, CLM (luminol), and TDLAS research-grade instruments near
Boulder, CO Ambient concentrations ranged from 0 02 to 4 ppb The potential
interferences of H2O2, HNO3, n-propyl nitrate, PAN, and O3 were examined in spiking
tests. Only the CLM (luminol) instrument displayed appreciable interferences, and they were
with O3 (0 6%) and PAN (24%) At ambient NO2 concentrations above 2 ppb, all three
instruments gave similar results Below 2 ppb, interferences from O3 and PAN provided
high biases to the CLM (luminol) results, but they could be corrected with measured O3 and
PAN results at NO2 levels above 0 3 ppb An O3 scrubber added to a second CLM
(luminol) instrument removed the O3 interference, but failed to remove PAN and appeared to
remove substantial amounts (i e , 50%) of NO2 Removal of NO2 in the manufacturer-
supplied O3 scrubber has also been reported by Kelly et al (1990) Tunable-diode laser
spectroscopy results compared favorably with CLM-PC at relatively high NO2 levels (i e ,
>0 4 ppb), but displayed a high bias (i e , factor of 5) at lower NO2 concentrations
(Fehsenfeld et al, 1987) No interferences or artifacts were found for the CLM-PC results
6-33
-------
An airborne intercompanson (i e , CITE 2) of NO2 measurements was conducted by
NASA using TDLAS, PF/TP-LEF, CLM-PC, and CLM (with FeSO4 converter) research-
grade instruments (Gregory et al, 1990b) Sampling flights were performed primarily in the
free troposphere, and NO2 concentrations were below 200 ppt and generally below 100 ppt
High biases (i.e., factors of 2 to 3) apparently resulting from PAN interferences were present
in results from the CLM instrument with the FeSO4 converter, and results from this
instrument were not considered in subsequent analyses At concentrations below 200 ppt,
results from the remaining three instruments were highly correlated (i e , correlation
coefficients ranged from 0.84 to 0 95) and displayed a general level of agreement to within
30 to 40%. The PF/TP-LEF results were higher than those of the CLM-PC, and the TDLAS
results were the lowest. At concentrations below 50 ppt, the results were poorly correlated,
although the PF/TP-LEF and CLM-PC results agreed to within 20 ppt Below 50 ppt,
TDLAS results were much higher than those of the other two instruments. This bias, similar
to that observed in the ground-based intercomparison of Fehsenfeld el al (1990), was
enhanced at low NO2 concentrations by an error in the data reduction protocol employed in
both studies.
6.3.10 Designated Methods
Acceptable sampling and analysis methodologies for NO2 have been specified by the
BPA (Code of Federal Regulations, 1987b) These designated methods are termed
"Reference" or "Fx|uivalent" In 1973, the original Federal Reference Method for NO2, the
Jacobs-Hochheiser Technique, was withdrawn because of technical deficiencies (Purdue and
Hauser, 1980) In 1976, the measurement principle and the associated calibration procedure
on which Reference Methods for NO2 must be based were specified The measurement
principle is gas-phase chemiluminescence and the calibration proceduie may employ either
GPT of an NO standard with O3 or an NO2 permeation device (Code of Federal Regulations,
1987a). Because only the measurement principle and calibration procedures applicable to
NO2 Reference Methods were specified, different analyzers can be built and designated as
Reference Methods, provided they meet the performance specifications shown in Table 6-1
(Code of Federal Regulations, 1987b)
6-34
-------
TABLE 6-1. PERFORMANCE SPECIFICATIONS FOR
NITROGEN DIOXIDE AUTOMATED METHODS3
Performance Parameter Units NO2
Range ppm 0-0 5
Noise
0% upper range limit ppm 0 005
80 % upper range limit ppm 0 005
Lower detectable limit ppm 0 01
Interference equivalent
Each interferant (SO2, NO, NH3, H2O) ppm ±0 02
Total interferant ppm ^0 04
Zero drift, 12 and 24 h ppm ±0 02
Span drift, 24 h
20% of upper range limit
80% upper range limit
Lag tune
Rise time
Fall time
Precision
20% of upper range limit
80 % of upper range limit
%
%
min
mm
nun
ppm
ppm
±200
±50
20
15
15
002
003
aNO2 = Nitrogen dioxide
SO2 = Sulfur dioxide
NO = Nitric oxide
NHg = Ammonia
H2O = Water
Source Code of Federal Regulations (1987b)
To be designated as an Equivalent Method, the candidate method must be based on
measurement principles different from the Reference Method and meet certain performance
specifications (Code of Federal Regulations, 1987b) An Equivalent Method may be either
6-35
-------
manual or automated To be designated as Equivalent, a candidate manual method must
demonstrate comparability, as shown in Table 6-2, with the Reference Method when applied
simultaneously to a real atmosphere A candidate automated method must meet the
performance specifications shown in Table 6-1 and demonstrate comparability as shown in
Table 6-2 with the Reference Method when apphed simultaneously to a real atmosphere
TABLE 6-2. COMPARABILITY TEST SPECIFICATIONS
FOR NITROGEN DIOXIDE
Nitrogen Dioxide Maximum Discrepancy
Concentration Range Specification
(ppm) (ppm)
Low 0.02 to 0 08 0 02
Medium 0 10 to 0 20 0 02
High 0.25 to 0 35 003
Source- Code of Federal Regulations (1987b)
Methods designated by the EPA as Reference and Equivalent are identified in Table 6-3
(Federal Register, 1986) Detailed descriptions of these and other methods for NO2 are
presented in previous subsections Studies were conducted to provide a basis for the
designation of methods by the EPA Tests were performed to compare the performance of
CLM, continuous colonmetric, manual sodium arsenite, and manual TGS methods (Purdue
and Hauser, 1980). The methods were compared by measuring NO2 in spiked and unspiked
ambient air simultaneously Quadruplicate samples were taken for the two manual methods
and duplicate analyzers were used for the two continuous methods The NO2 spikes were
varied randomly from day to day over the sampling schedule and ranged from 0 to 430 ppb
Agreement both within and between methods was good the average difference was never
greater than 4 ppb Correlation coefficients for between-method comparisons exceeded
0.985 in all cases No between-method differences could be attributed to concentrations of
NO, CO2, O3, total sulfur, or total suspended paniculate matter in the ambient air
Significant negative interference in the continuous colonmetnc method was found at NO2
6-36
-------
TABLE 6-3. REFERENCE AND EQUIVALENT METHODS FOR NITROGEN
DIOXIDE DESIGNATED BY THE U.S. ENVIRONMENTAL PROTECTION AGENCY
Method
Manual Methods (Equivalent Methods')
Sodium aresmte
Sodium aresmte/Technicon n
TGS-ANSA
Analyzers (Reference Methods)
Beckman 952A
Bendix 8101-B
Bendix 8101-C
CSI 1600
Meloy NA53OR
Monitor Labs 844OE
Monitor Labs 8840
Philips PW9762/02
Thermo Electron 14B/E
Thermo Electron 14D/E
Designation Number Method Code
EQN-1277-026
EQN-1277-027
EQM-1277-028
RFNA-0179-034
RFNA-0479-038
RFNA-0777-022
RFNA-0977-025
RFNA-1078-031
RFNA-0677-021
RFNA-0280-042
RFNA-0879-040
RFNA-0179-035
RFNA-0279-037
026
027
028
034
038
022
025
031
021
042
040
035
037
Source Federal Register (1986)
concentrations of 40 and 53 ppb in the presence of 180 and 340 ppb 0^ However,
at O3 concentrations of 50 ppb, no interference was detected Also, no interference was
detected with the manual sodium arsemte method at NO concentrations as high as 250 ppb
The performance of the CLM analyzers was judged to be superior to that of the continuous
colonmetac analyzers with respect to zero drift, span drift, response tunes, and overall
operation Of the two manual methods, the performance of the sodium arsemte method was
judged superior to the TGS method
Eight of the Reference Methods have undergone extensive postdesignation testing in the
laboratory and field (Michie et al, 1983) Performance test results have been reported and
were found to meet the specifications shown in Table 6 1 Based on the field test results,
minimum detection limits were defined as three times the precision These MDL results
ranged from 5 to 13 ppb with an average of 9 ppb An independent analysis of this data by
Holland and McElroy (1986) also showed similar results
6-37
-------
Interrogation of the National Aerometnc Data Bank records for 1985 revealed that NO2
data were archived from 40 states (Hustvedt, 1987) Of the 335 data sets, CLM was
employed in 291 cases, and manual methods were employed in the remainder Of the
manual methods, 42 employed the sodium arsemte method with either orifice or fritted
bubblers. Interrogation of the Precision Accuracy Reporting System (PARS) data base for
the State and Local Air Monitoring Stations network for fourth quarter 1986 and first quarter
1987 records revealed that data were archived from tests of 114 CLM analyzers (Rhodes,
1987). Of these, 43% were Bendix 8101C, 40% were CSI1600, 15% were Monitor
Labs 8840 and 8440E, 2% were Meloy NA 530R, and 1% were Beckman 952A
To illustrate the precision and accuracy of the designated methods in field applications,
PARS data were examined for 1983 through 1986 (Rhodes and Evans, 1988) The results,
shown in Table 6-4 as 95 % probability limits, suggest that the precision of continuous NO2
analyzers falls in the range of ±10 to 15%, whereas the manual methods are much worse, at
±20 to 50%. It should be noted that the manual precision results show a recent worsening
This trend may reflect the phasing out of manual methods in the network that was completed
by the end of 1986
The tabulated probability limits for accuracy of continuous NO2 analyzers are ±20%,
whereas for the manual methods, they are ±3 to 7% (Rhodes and Evans, 1988) The
accuracy results reflect audits of the analysis portion of the manual melhods and audits of
both sampling and analysis for the continuous methods Thus, the apparent difference in
accuracy may be reflecting differences in the auditing procedures employed
6.4 NITROGEN OXIDES
For the purposes of this document, NOX is considered to be the sum of NO and NO2
No widely accepted methods are available for determining NOX except by determining NO
and NO2 individually and summing, or by converting NO2 to NO and determining NOX as
the total NO Sections 6 2 and 6 3 describe methods for the determination of NO and NO2,
respectively.
Commercial CLM NOX analyzers catalytically convert NO2 to NO and measure NOX as
the sum of the originally present NO and the converted NO As noted in Section 6 3.1,
6-38
-------
TABLE 6-4. NATIONAL PRECISION AND ACCURACY
PROBABILITY LIMIT VALUES EXPRESSED AS PERCENT
FOR CONTINUOUS AND MANUAL METHODS FOR NITROGEN DIOXIDE
Nitrogen Dioxide
Method
Continuous
Precision
Accuracy
Manual
Precision
Accuracy
1983a
-13 + 12
(9,299)b
-19 + 15
(680)°
-19 + 21
(l,324)d
-5 + 6
(348)c
1984
-14 + 13
(8,653)b
-21 + 20
(613)°
-21 + 27
(691)d
-6 + 7
(175)c
1985
-12 + 12
(7,695)b
-20 + 21
(573)c
-27 + 29
(469)d
-3+5
(161)c
1986
-11 + 11
(6,686)b
-21 + 20
(529)c
-48 + 45
(174)d
-4 + 5
(92)c
Calculated differently for 1983 than for 1984 through 1986
Number of precision checks
°Number of audits, manual at 0 074 to 0 083 ppm, continuous at 0 03 to 0 08 ppm
Number of collocated samples
Source Rhodes and Evans (1988)
NO2-to-NO converters used may not be specific for NO2 Heated molybdenum converters,
typically used in commercial analyzers, have been shown to convert PAN, HNO3, and other
nitroxy compounds to NO, giving rise to artificially high values for NO2 and NOX
In research-grade CLM NOX analyzers, FeSO4 converters have been shown to overestimate
NOX by a factor of 2 to 3 at concentrations of 0 2 ppb (Fehsenfeld et al, 1987, Gregory
et al, 1990b)
The catalytic conversion approach will permit an accurate measure of NOX as long as
the nitroxy compounds present in the sampled atmosphere are limited to NO and NO2
Atmospheric concentrations of potential interferences aie generally low relative to NO2
(Code of Federal Regulations, 1987a) There are cases, however, where compounds other
than NO and NO2 contribute substantially to the atmospheric nitroxy burden Examples
include urban atmospheres such as Los Angeles, where both PAN and HNO3 levels may
6-39
-------
reach appreciable levels (Tuazon et al, 1981), and remote environments, where PAN may
comprise a significant fraction of the airborne mtroxy reservoir (Fehsenfeld et al, 1987,
Gregory etal., 1990c).
A prototype method employing CLM has been suggested to measure NOX (Fontijn
et al., 1980). This method uses the reaction between atomic hydrogen and NO2 to give NO
along with the subsequent CLM reaction between atomic hydrogen and NO llie emission
occurs between 628 and 800 nm, and the intensity is measured by a PM tube at 640 to
740 nm At a constant atomic hydrogen concentration, the light intensity is proportional to
the NOX concentration The instrument was developed for application to automotive exhaust
gas. Significant interferences were noted for O2 and ethene, but not for H2O, toluene,
isopentane, CO, CO2, NH3, and HCN Response was linear within 2% from 6 to
3,000 ppm. Significant development is needed if the limit of detection for this technique is
to be extended from 6 ppm to the parts-per-tnllion to parts-per-billion range appropriate for
ambient air monitoring applications.
6.5 TOTAL REACTIVE ODD NITROGEN OXIDES
In the present document, total reactive odd nitrogen oxides are represented by NOy
Individual components comprising NOy are NO, NO2, NO3, N2O5, HONO, HNO3,
HO2NO2, PAN, other organic nitrates, and particulate NO3"
Although no single instrument has been devised to measure NOy, researchers have
combined highly sensitive research-grade CLM NO detectors with catalytic converters that
are sufficiently active to reduce most of the important gas phase NOy species to NO for
subsequent detection (Helas et al, 1981, Dickerson, 1984, Fahey et al, 1986) Calibrations
are performed using dynamic dilution with air Two standards are usually employed
(1) a cylinder of compressed gas containing NO in nitrogen at an NIST-traceable
concentration, and (2) an NO2 permeation tube The NO cylinder is used to calibrate the
instrument for NO, and NO2 from the permeation tube is used as a surrogate to calibrate the
instrument for NOy
Two types of heated converters have been employed molybdenum and gold As noted
in Section 6.3.1, heated molybdenum has been shown to convert NO2, HNO3, PAN, methyl
6-40
-------
nitrate, ethyl nitrate and nitrite, n-propyl nitrate, and n-butyl nitrate to NO with high
efficiency Dickerson (1984) also reports that NO3 and N2O5 are converted to NO on heated
molybdenum, whereas acetomtnle, HCN, and NH3 axe not Dickerson (1984) has coupled
this converter with a sensitive CLM NO detector and reported a detection limit for NOy of
25 ppt for a 20-s integration time and an accuracy of ±40% at levels well above the
detection limit (Fehsenfeld et al, 1987)
A gold catalyst operated at 300 °C in the presence of 3,000 ppm CO has been reported
to reduce NOy to NO (Bellinger et al, 1983, Fahey et al, 1985a) Converter efficiencies
near 100% were found for NO2, HNO3, N2O5, and PAN Interferences in the presence of
water vapor were found to be negligible for O3, NH3, N2O, HCN, CH4, and various
chlonne-and sulfur-containing compounds Fahey et al (1986) coupled this converter with a
sensitive CLM NO detector and reported a detection limit of 10 ppt for a 10-s integration
tune and an accuracy of ±15%
A field intercompanson of the two instruments described above was conducted near
Boulder, CO (Fehsenfeld et al, 1987). In this study, ambient NOy concentrations ranged
from 400 ppt to over 100 ppb Both instruments gave similar estimates of NOy
concentrations under conditions that varied from representing urban to continental
background air
Using the instrument described above with a gold converter, Fahey et al (1986)
compared NOy measurements with the sum of the component species measured individually
The NOy levels systematically exceeded the sum The difference was attributed to the
presence of one or more unmeasured organic nitrate species that are similar to PAN and may
be of photochemical origin
6.6 PEROXYACETYL NITRATE
Several methods have been used to measure the concentration of PAN in ambient air
Stephens (1969) and Roberts (1990) have provided a good overview of many of these
methods Peroxyacetyl nitrate was first measured by using long-path infrared spectrometry,
however, insufficient sensitivity by this technique prompted the development of other
methods (Darley et al, 1963) A ground-based FTER system with a 1-km cell has reported
6-41
-------
detection limits of 4 ppb for PAN near 790 and 1,160 cm"1 (Tuazon et al, 1978) The
limited sensitivity and the complexity of FTER systems have generally limited ambient
applications of the FUR to the relatively high concentrations associated with the Los Angeles
basin More recently, cryogenic sampling and matrix-isolation FT1R has been used to
measure PAN in 15-L integrated samples of ambient air with a theoretical MDL of 50 ppt
(Griffith and Schuster, 1987) A laboratory prototype method, TTFMS, has a projected
MDL for PAN of 2 ppt (Hansen, 1989) Gas chromatography with name lomzation
detection (GC-EED) may be employed to measure PAN, but this method is only practical for
concentrated mixtures above 10 ppm using a 1-mL sample loop (Meyrahn et al, 1987) The
most common method is gas chromatography using electron capture detection (GC-ECD)
(Darley et al., 1963, Smith et al, 1972, Stephens and Price, 1973, Singh and Salas, 1983)
6.6.1 Gas Chromatography-Electron Capture Detection
Both manual and automated integrated sampling methods using GC-ECD have been
employed (Stephens and Price, 1973, Lonneman et al, 1976) Relatively low column and
detector temperatures (below 50 and 100 °C, respectively) have been used to minimize
thermal decomposition of PAN Short packed columns coated with polyethylene glycol-type
stationary phases (e g , Carbowax 400) have normally been used Recently, improved
precision and sensitivity have been achieved using silica capillary columns (Helmig et al,
1989; Roberts et al, 1989) Although sampling intervals are limited by the elution tunes of
the chromatographic system, intervals of 10 to 15 mm have been employed (Helmig et al,
1989; Nieboer and Van Ham, 1976) Using packed columns, detection limits of 10 ppt have
been reported using direct sampling with a 20-mL sample loop (Vierlcorn-Rudolph et al,
1985), and detection limits of 1 to 5 ppt have been reported using cryogenic enrichment of
samples (Vierkorn-Rudolph et al, 1985, Singh and Salas, 1983) Capillary columns offer
the potential for considerable (i e., factor of 20) enhancement in sensitivity (Roberts et al,
1989). Accuracy estimates of ±20 to 30% have been claimed
A comparison of two similar GC-ECD methods for airborne PAN measurements was
performed (Gregory et al, 1990c) Both methods employed cryogenic enrichment of
samples, used packed GC columns, and claimed detection limits below 5 ppt Results of this
study showed that at PAN concentrations below 100 ppt, agreement was approximately
6-42
-------
17 ppt, and at higher concentrations (i e , 100 to 300 ppt) the measurements agreed to within
25% (expressed as a percent difference) These findings are generally consistent with
accuracy claims noted earlier
6.6.2 Alkaline Hydrolysis
Alkaline hydrolysis in 5 % NaOH has been shown by Nicksic et al (1967) to convert
PAN quantitatively to nitrite and acetate This permits sampling with a bubbler containing
5 % NaOH and subsequent analysis for nitrite or acetate Nitrogen dioxide is usually present
in ambient air with PAN It can interfere with PAN determination as nitrite because NQ2
may be collected as nitrite in alkaline solution Acetate particles or acetic acid can interfere
with PAN determination as acetate A method involving alkaline hydrolysis followed by 1C
determination of acetate has been used to measure PAN in photochemical systems (Grosjean
and Harrison, 1985a) Results compare favorably with those of a CLM method employing
the difference in NOX signals measured upstream and downstream of an alkaline bubbler
In addition to NaOH, other alkaline salts (e g , potassium hydroxide and sodium carbonate
[Na2CO3]) have been used to coat filters, cartridges, and annular denuders (Grosjean and
Parmar, 1990, Williams and Grosjean, 1990) Peroxyacetyl nitrate collection efficiencies
ranged from 10 to 100%, depending on the type and amount of the alkaline salt, the flow
rate, and the collection device employed
6.6.3 Gas Chromatography—Alternate Detectors
As noted in Section 631, PAN is readily reduced to NO Meyrahn et al (1987) have
coupled a GC to separate PAN, NO, and NO2, a molybdenum converter, and a CLM NO
analyzer to measure PAN as NO Using a 10 mL sample loop, a detection limit of 10 ppb
was reported
The luminol-based detector has shown sensitivity to PAN, as discussed in
Section 632 Burkhardt et al (1988) used gas chromatography and a commercially
available luminol-based instrument (i e , Scintrex LMA-3 Lummox) to detect both NO2 and
PAN Using a sampling interval of 40 s, linear response was claimed from 0 2 to 170 ppb
NO2 and from 1 to 65 ppb PAN Although the PAN calibration was nonlinear below 1 ppb,
an MDL of 0 12 ppb was reported Drummond et al (1989) have slightly modified the
6-43
-------
above approach by converting the PAN from the GC column to NO2 and measuring the
resulting NO2 with a CLM (luminol) instrument
6.6.4 Peroxyacetyl Nitrate Stability
Peroxyacetyl nitrate is an unstable gas and is subject to surface-related decomposition as
well as thermal instability Peroxyacetyl nitrate exists in a temperature-sensitive equilibrium
with the peroxyacetyl radical and NO2 (Cox and Roffey, 1977) Incieased temperature
favors the peroxyacetyl radical and NO2 at the expense of PAN Added NO2 should force
the equilibrium toward PAN and enhance its stability In the presence of NO, peroxyacetyl
radicals react rapidly to form NO2 and acetoxy radicals, which decompose in O2 to radicals
that also convert NO to NO2 As a result, the presence of NO acts to reduce PAN stability
and enhance its decay rate (Lonneman et al, 1982) Stephens (1969) reported that
appreciable PAN loss in a metal sampling valve was traced to decomposition on a silver-
soldered joint Meyrahn et al (1987) reported that PAN decayed according to first order
kinetics at a rate of 2 to 4%/h in glass vessels that had been previously conditioned with
PAN. They employed 200 ppm PAN in glass vessels and the noted first-order decay as the
basis for one proposed method of in-field PAN calibration In contrast, Holdren and Spicer
(1984) found that without NO2 added, 20 ppb PAN decayed in Tedlar bags according to first
order kinetics at a rate of 40%/h The addition of 100 ppb NO2 acted to stabilize the PAN
(20 ppb) m the Tedlar bags
A humidity-related difference in GC-ECD response has been reported (Holdren and
Rasmussen, 1976) Low responses observed at humidities below 30% and PAN
concentrations of 10 and 100 ppb, but not 1,000 ppb, were attributed to sample-column
interactions. This effect was not observed by Lonneman (1977) Watanabe and Stephens
(1978) conducted experiments at 140 ppb and did not conclude that the reduced response was
from faults in the detector or the instrument They concluded that there was no column-
related effect, and they observed surface-related sorption by PAN at 140 ppb in dry acid-
washed glass flasks. They recommended that moist air be used to prepare PAN calibration
mixtures to avoid potential surface-mediated effects
Another surface-related effect has been reported for PAN analyses of remote marine ear
(Singh and Viezee, 1988) Peroxyacetyl nitrate concentrations were found to increase by
6-44
-------
20 to 170 ppt, an average factor of 3 2, when the sample was stored in a glass vessel for
1 to 2 nun prior to analysis This effect remains to be explained
6.6.5 Calibration
Because PAN is unstable, the preparation of reliable calibration standards is difficult
Several methods have been employed The original method used the photolysis of ethyl
nitrite in pure oxygen (Stephens, 1969). When pure PAN is desired, the reaction mixture
must be purified, usually by chromatography, to remove the major by-products, acetaldehyde
and methyl and ethyl nitrates (Stephens et al, 1965) For GC calibration, purification is
unnecessary, the PAN concentration in the reactant matrix is established from the IR
absorption spectrum and subsequently diluted to the parts-per-billion working range needed
for calibration purposes (Stephens and Price, 1973)
Static mixtures of molecular chlorine, acetaldehyde, and NO2 in the ratio of 2 4 4 can
be photolyzed in the presence of a slight excess NO2 to give a near stoichiometnc yield of
PAN (Gay et al, 1976) This method was adapted by Singh and Salas (1983) and later by
Grosjean et al (1984) using photolytic reactors to provide continuous PAN calibration units
at concentrations between 2 and 400 ppb In the former approach, the PAN concentration is
established by measuring the change in acetaldehyde concentration across the reactor In the
latter, the PAN concentration is established by measuring the acetate in an alkaline bubbler
where PAN is hydrolyzed
A static technique involving the photolysis of acetone in the presence of NO2 and air at
250 nm has been reported to produce a constant concentration of PAN (Meyrahn et al,
1987) A Penray mercury lamp is inserted into a mixture of 10 ppm NO2 and 1 % acetone
and irradiated for 3 mm to yield 89 + 03 ppm PAN
Peroxyacetyl nitrate can be synthesized in the condensed phase by the nitration of
peracetic acid in hexane (Helmig et al , 1989), heptane (Nielsen et al , 1982), octane
(Holdren and Spicer, 1984), or n-tndecane (Gaffney et al, 1984) Purification of PAN in
the liquid phase is needed using the first two methods The resulting PAN-organic solution
can be stored at —20 to —80 °C with losses of less lhan 3 6%/mo and can be injected
directly into a vessel containing air to produce a calibration mixture The PAN concentration
is normally established by FUR analysis of the solution or the resulting PAN-air mixture
6-45
-------
As noted in Section 631, PAN is readily reduced to NO, and CLM NOX analyzers
have near quantitative response to PAN Thus under some circumstances, CLM NOX
response can be used for PAN calibration One method uses the difference in NOX signal
measured upstream and downstream of an alkaline bubbler (Grosjean and Harrison, 1985a)
Joos et al (1986) have coupled a CLM NOX analyzer with a GC system to permit calibration
of the BCD response by reference to the CLM NOX analyzer that has been calibrated by
traditional methods.
As noted previously, NO in the presence of PAN is converted to NO2 Approximately
four molecules of NO can react per molecule of PAN Lonneman et al (1982) have devised
a PAN calibration procedure based on the reaction of PAN with NO in the presence of
benzaldehyde, which is added to control unwanted radical chemistry and improve precision
Using this approach and an initial NO-to-PAN ratio of between 10 and 20 to 1, the change in
NO concentration is monitored with a CLM NO analyzer, the change in PAN GC-ECD
response is monitored, and the resulting ratio (i e , ANO/APAN) is divided by the
stoichiometric factor of 4.7 to arrive at a calibration factor for the BCD
Peroxyacetyl nitrate and n-propyl nitrate (NPN) have similar BCD responses Serial
dilution of the more stable compound, NPN, has been used for field operations (Vierkorn-
Rudolph et al, 1985) This approach is not recommended for primary cahbration, however,
because it does not permit verification of quantitative delivery of PAN to the detector
(Stephens and Price, 1973)
6.6.6 Other Organic Nitrates
Other organic nitrates (e g , alkyl nitrates, peroxypropionyl nitrate [PPN], and PBzN)
are also present in the atmosphere, but usually at lower concentrations than PAN (Fahey
et al., 1986) In general, similar methods for sampling, analysis, and calibration may be
used for other organic nitrates as are used for PAN (Stephens, 1969) Both FUR and
GC-ECD may be used to measure these compounds
With MDLs of 0 1 to 0 4 ppb, inspection of 3,000 GC-ECD chromatograms recorded
at five to nine sites during the 1987 Southern California Air Quality Study yielded only seven
possible (but nonprobable) observations of methyl nitrate (Grosjean el al, 1990) Roberts
et al. (1989) have reported separation of PAN, PPN, and Cx to C4 alkyl nitrates and the
6-46
-------
potential increase in sensitivity by a factor of 20 using fused silica-coated capillary columns
rather than the more conventional coated packed columns Atlas (1988) has used two 5-mg
charcoal traps in series to collect C$ to C7 alkyl nitrates from 12- to 300-L samples at 200 to
400 mL/min in remote atmospheres The traps are extracted in small volumes of benzene
and analyzed using capillary GC-ECD Concentrations as low as 1 ppt were reported
Peroxybenzoyl nitrate may be collected as methyl benzoate using bubblers containing
methanol-NaOH solutions (Appel, 1973) The resulting methyl benzoate is solvent extracted
and analyzed by packed column GC-FID with an MDL of 70 ppt Recently, a collection
method using aqueous alkaline hydrolysis of PBzN to the benzoate ion followed by IC-UV
analysis was reported to have a detection limit of 30 ppt in a 60-L sample (Fung and
Grosjean, 1985) Using this method, a median PBzN level of 0 32 ppb was reported for
Los Angeles air samples
6.7 NITRIC ACID
Several methods are available for the determination of airborne concentrations of
HNO3 Among them are filtration (Okita et al , 1976, Spicer et al , 1978b), denuder tubes
(Forrest et al, 1982, DeSantis et al, 1985, Perm, 1986), CLM (Joseph and Spicer, 1978),
absorption spectroscopy (Tuazon et al, 1978, Schiff et al, 1983, Biermann et al, 1988),
and nucrocoulometry (Spicer et al, 1978b) Filtration and denuder techniques involve
collection of HNO3 onto a media and subsequent analytical determination As a result of its
2-ppb detection limit and long response tune, microcoulometry has been largely replaced by
other methods Consequently, only the first four methods listed above are described here
6.7.1 Filtration
Filtration techniques generally employ dual filtei s that rely on the collection of
particulate NO3" on the first filter and gaseous HNO3 as NO3" on the second filter This
method is sometimes called the filter pack (FP) method Typically, filtration is used in
conjunction with instrumental detection or subsequent chemical analysis of the material
collected on the filter media Filter extracts are usually analyzed for NO3" using 1C
Efficient HNO3 collection has been found with nylon filters (Spicer et al, 1978b) and with
6-47
-------
filters impregnated with sodium chloride (NaCl) or sodium fluoride (TSTaF) (Okita et al,
1976; Forrest et al, 1980; Fuglsang, 1986). The HNO3 capacity of 47-mm diameter
NaCl-coated filters (500 /*g/cm) far exceeds that of nylon (30 jug/cm) (Anlauf et al, 1986)
This advantage may be offset because the presence of the chloride ion in the NaCl-coated
filter extract may hamper 1C determination of NO3" With a 47-mm diameter nylon filter
sampling at 1 m3/h at a nominal HNO3 level of 5 /jg/m (2 ppb), the capacity is sufficient for
just over 4 days of sampling. The sensitivity of filtration and other integrative methods
depends on the detection limit of the analytical finish, the variability and magnitude of the
blank level, collection and extraction efficiencies, and the volume of air sampled As an
example, under an assumed 1C detection limit for NO3" of 0 05 /tg/mL, a filter extraction
volume of 10 mL, negligible blank, quantitative collection and extraction, and a sampled air
3 3
volume of 24 m (i e., flow rate 1 m /h for 1 day), the minimum sensitivity is 0 02 /*g/m
(8 ppt). A precision of ±10% and an accuracy of +20 to —40% are claimed for FPs
containing Teflon™ and nylon filters (Fahey et al, 1986)
Although the HNO3 determination by FP methods is desirable due to simplicity, high
sensitivity, and low cost, there is great difficulty in distinguishing between gaseous and
particulate forms of nitrate Errors in the measurement of gaseous HNO3 may be in the
form of positive artifacts due to volatilization of collected aerosol nitrates on the prefilter to
form gaseous HNO3 (i.e , NEySTC^ <* NH3 + HNO3) (Appel et al, 1980), reaction of
collected particulate nitrates on the prefilter with strong acids, resulting in the release of
HNO3 (i e., H2SO4 + 2NH4NO3 -> (NH4)2SO4 + 2HNO3) (Appel and Tokiwa, 1981), or
formation of HNO3 on the collection medium by interaction with other NOX species (e g ,
HONO or NC>2) (Eatough et al, 1988, Spicer and Schumacher, 1979) Negative HNO3
artifacts may result from retention of HNO3 by the prefilter collection medium (Appel et al,
1984), retention of HNO3 by collected particles on the prefilter (Appel et al, 1980), a low
capacity for HNO3 on the collection medium, or losses of HNO3 by volatilization or by
displacement by other acids
Inert prefilter materials, such as Teflon™, should not collect appreciable amounts of
HNO3 (Appel et al, 1979), this, however, does not preclude the possibility of HNO3
reaction with aerosol particles collected on Teflon™ prefilters In addition, some types of
Teflon™ may sorb HNO3 to a larger extent than others (Appel et al, 1988) This
6-48
-------
underscores the importance of using "inert" materials for all surfaces coming into contact
with HNO3 to insure representative sampling
6.7.2 Denuders
To avoid some of the artifact problems associated with the use of filters, denuder tube
samplers were introduced In general, a denuder is a tube or channel that has its walls
coated with or fabricated from a substance that removes the gaseous species of interest, in
this case HNO3 (also see Section 636) The HNO3 molecules diffuse to and impact the
surface while the sample is drawn through the channel The flow conditions are usually
laminar (Re < 2,000), and by taking advantage of differences in diffusivities, permit
particles to pass through the denuder relatively undisturbed Using the sampled air volume,
the concentration of HNO3 is calculated from the measured amount of NO3" collected on the
denuder walls (Perm, 1986) or from the difference of NO3" collected downstream in the
presence and absence of the denuder (Shaw et al., 1982, Forrest et al, 1982)
Denuder tubes have employed magnesium oxide (MgO) (Shaw et al, 1982), Na2CO3
(Perm, 1986), nylon (Mulawa and Cadle, 1985), aluminum sulfate (A12[SO4]3) (Lindqvist,
1985), magnesium sulfate (MgSO4), barium sulfate (BaSO4) (Klockow, 1989), and tungstic
acid (TA) (McClenny et al, 1982) to retain HNO3 A coating of MgO is frequently used
with the denuder difference (DD) approach, where one coated denuder is followed by nylon
or NaCl-coated filters and a parallel arrangement uses a nylon or NaCl-coated filter The
difference in NO3" on the two parallel filters is attributed to HNO3 (Shaw et al , 1982) The
Na2CO3 coating is used in methods employing the DD or direct analysis approach In the
latter, the denuder is extracted, and the extract is analyzed for NO3", which is attributed to
HNO3 (Perm, 1986)
To maintain laminar flow, the flow rate through a single conventional open channel
denuder of reasonable dimensions is limited to approximately 1 to 2 L/min As a result,
long duration samples or numerous open channel denuders located in parallel may be
required to provide sufficient analyte for quantitation A new type of denuder, the annular
denuder (AD), has been developed where the same equivalent diameter permits the flow rate
to be increased by a factor of 12 (Possanzini et al, 1983) In the AD, ambient air is passed
through the annular space of two concentnc tubes The outside of the inner tube and the
6-49
-------
inside of the outer tube are coated with a specific gas-absorbing substance For collecting
HNO3, Na2CO3 has been used (DeSantis et al , 1985) In cases where appreciable HONO is
present and may be cocollected on Na2CO3 and the resulting NO2" may be oxidized to NO3"
over extended sampling periods by atmospheric oxidants (e g , O3), two or more denuders
are used to permit resolution of HNO3 and HONO (Febo et al , 1986, Pernno et al , 1990)
The first denuder is coated with NaCl or NaF to collect HNO3 as NO3" and the downstream
denuder(s) is coated with an aqueous solution of Na2CO3 and glycerol to collect HONO as
3
the sum of NO2" and NO3" For a 1 m /h 1-day sample under the same assumptions given
earlier for filtration, the MDL for the AD is 0 02 /tg/m (8 ppt) Median precision estimates
of 8 and 5 % RSD have been reported for 13 22-h and 12 1-week duration samples (Sickles
et al., 1989; Sickles, 1987).
Partial denuders have been fabricated of nylon filter material (Mulawa and Cadle,
1985). These denuders, operated under laminar-flow conditions, have relied on a
mathematical description of molecular diffusion to a perfect wall sink along with HNO3
deposition measured along the length of the denuder to infer the sampled, ambient HNO^
concentration A refinement in the data treatment has been offered recently that considers
interferent nitrate on nylon partial denuders (Febo et al , 1988) Although these denuders are
operated under laminar-flow conditions, a recently introduced technique employs a nylon
partial denuder that is operated under transition-flow conditions (Re « 2,600) (Knapp et al ,
1986). This approach, TFR, uses a piece of nylon filter material rolled into a cylindrical
shape and placed in a Teflon™ tube The sample is drawn through the tube and denuder
under transition-flow conditions, where a constant fraction of the HNO3 is claimed to be
collected. The tube is followed by a Teflon™ and a nylon filter The HNO3 is calculated by
analyzing the denuder extract for NO3" and applying the constant collection fraction The
particulate NO3" is determined algebraically using the NO3" measured in the extracts of the
Teflon™ and nylon filters. For 85% collection efficiency and 1 m3/h 1-day sample under
the same assumptions given earlier for filtration, the MDL for the TFR is 0 2
(0.1 ppb). A median precision of 7% RSD has been determined for seven 1-week duration
samples (Knapp et al , 1986)
Automated systems using coated denuders with thermal desorption employ A12(SO4)3,
MgSO4, BaSO4, or TA to preconcentrate HNO3 for subsequent delivery to an instrumental
6-50
-------
detection system In the first case, HNO3 from a 30 L sample is collected on an
Al2(SO4)3-coated denuder, thermally desorbed, thermally converted to NO, and analyzed by
gas chromatography with a photoiomzation detector (Lindqvist, 1985). A nominal MDL of
5 ppt and precision estimate of ±10% were claimed
Klockow et al (1989) have used MgSO4- and BaSO4-coated denuders to collect HNO3
The sample is thermally desorbed and measured with a CLM NOX analyzer For a 30-min
sample at 5 L/min, a nominal MDL of 0 1 /*g/m3 (40 ppt) and precision estimate of +5 %
were claimed
A TA-coated denuder has been used to collect HNO3 for analysis on an automated basis
with a 40-min cycle time (McClenny et al, 1982) The collected HNO3 is thermally
desorbed as NO2, thermally converted to NO, and measured with a commercial CLM NOX
analyzer A nominal MDL of 70 ppt and a precision estimate of ±10% were claimed
Recent claims for a similar device with a 20-min cycle tune have included MDL of 20 ppt,
accuracy of 15 to 20%, and precision of 8% (Gregory et al, 1990d) Tungstic acid-coated
denuders have drawbacks they are difficult to prepare, have low capacities, and are subject
to unknown atmospheric interferences (Fellin et al, 1984, Eatough et al, 1985, Roberts
et al , 1987)
6.7.3 Chemiluminescence
As noted in Section 631, HNO3 is readily reduced to NO in NO2-to-NO converters
used in commercial and many research-grade CLM NOX analyzers (Joseph and Spicer, 1978,
Bellinger et al, 1983, Fahey et al, 1985a, Grosjean and Harrison, 1985b, Rickman and
Wright, 1986) Nitac acid measurements that employ CLM generally use an NOX analyzer
to measure the NOX in a sampled air stream in the presence and absence of an HNO3
scrubber The difference in these NOX signals is attributed to HNO3 Nylon filters have
been used as an HNO3 scrubber with both commercial and research-grade CLM NOX
analyzers The instrumental performance for HNO3 is similar to that for NO2 with the same
instruments (Joseph and Spicer, 1978, Kelly et al, 1979) Losses of HNO3 from the
sampled air to the interior surfaces of the sampling Line and instrument may lead to
nonquantitative responses or increases in response time (Rickman and Wright, 1986, Appel
et al, 1988)
6-51
-------
In other methods employing CLM and described in Section 6 7 2, TA, MgSO4, or
BaSO4 are used as regenerable HNO3 scrubbers (McClenny et al, 1982, Klockow et al,
1989). With these methods, HNO3 is collected on a coated denuder, the collected HNO3 is
thermally desorbed as NO2 (regenerating the scrubber), the desorbed NO2 is thermally
converted to NO, and the resulting NO is measured with a commercial CLM analyzer
6.7.4 Absorption Spectroscopy
Absorption spectroscopy is discussed in Sections 623 and 634 Although FUR,
TDLAS, and, potentially, TTFMS techniques may be used to measure ambient levels of
HN03, poor sensitivity limits ambient applications of FTTR A ground-based 23-m multipass
FT1R system with a 1-km path length has reported detection limits of 4 ppb near 900 cm"1
(Tuazon et al , 1978; Biermann et al, 1988) A theoretical MDL for HNO3 of 10 ppt has
been claimed for 15-L integrated samples of ambient air using cryogenic sampling and
matrix-isolation FTIR (Griffith and Schuster, 1987)
Cassidy and Reid (1982) report an expected MDL of 0 4 ppb using TDLAS near
1,330 cm"1. For a 40-m path length near 1,720 cm"1, the MDL is 0 4 ppb (Schiff et al,
1983). With a 150-m path length, Mackay and Schiff (1987) report an MDL of 0 1 ppb and
an accuracy of ±20%. Although the volumetric residence time in the White cell of the
TDLAS is 4 s, sample-surface interactions limit the response tune to changes in HNO3
concentration to about 5 nun As described in Section 6 2 3, a laboratory prototype method,
TTFMS, has been developed (Hansen, 1989) The projected MDL for HNO3 is 0 3 ppt
6.7.5 Calibration
Nitric acid is a highly polar material and consequently interacts readily with many
surfaces (Goldan et al, 1983, Appel et al, 1988) This reactivity prevents the preparation
of stable calibration mixtures in cylinders of compressed gases Two methods, permeation
devices and diffusion tubes, are generally employed to generate calibration atmospheres of
HNO3 (Schiff et al, 1983; Goldan et al, 1983) Permeation tubes are described for NO2 in
Section 6 3.8 Permeation tubes for HNO3 with various emission rates are available from
commercial suppliers An alternate permeation device may be fabricated in the laboratory by
6-52
-------
passing earner gas through a length of Teflon™ tubing that is immersed in a reservoir of
HNO3 and sulfuric acid (H2SO4) (Mackay and Schiff, 1987)
Diffusion tubes are generally fabricated in the laboratory (Schiff et al, 1983) A liquid
mixture of HNO3 and H2SO4 is held in a reservoir that is connected to a clean air dilution
manifold by a capillary tube The HNO3 diffusion rate depends on the length and area of the
capillary as well as the temperature of the reservoir
Nominal HNO3 emission rates for permeation tubes are provided by the supplier and
are calculated for diffusion tubes (Nelson, 1971) Although it is common to calibrate
permeation tubes gravimetncally, it has been reported that non-HNO3 species (i e , NO2) are
also released and may account for 10 to 15 % of the observed weight loss (Goldan et al,
1983) Because the emission rate estimate for diffusion tubes is also an approximation, the
independent measurement of HNO3 emission rates from permeation and diffusion tubes is
recommended
This measurement may be accomplished by pH titration or by using nylon filters, NaCl-
coated filters, or caustic bubblers to collect and quantify the HNO3 as NO3" Because caustic
bubblers may also collect NO2 to some extent, their use could overestimate the HNO3
emission rate, and a filtration technique is preferred Alternative, but more elaborate,
methods of confirming the HNO3 emission rate are FUR, TDLAS, and the CLM NOX
analysis that uses photolysis to convert NO2 to NO
6.7.6 Intercomparisons
Several field studies have been conducted that have permitted the comparison of
different techniques for the measurement of HNO3 (Spicer et al, 1982, Walega et al, 1984,
Anlauf et al, 1985, Roberts et al, 1987, Hering et al, 1988, Solomon et al, 1988, Benner
et al, 1987, Tanner et al, 1989, Sickles et al, 1990, Gregory et al, 1990d, Dasch et al,
1989) Results from these studies suggest that the FP overestimates the HNO3 concentrations
and that coated denuder thermal desorption techniques in various tested configurations may
not provide reliable measurements of HNO3
An mtercompanson of HNO3 measurement methods was conducted in Claremont, CA,
in August and September of 1979 (Spicer et al, 1982) Ten methods were compared
6-53
-------
five FP, two DD, two CLM, and one FTTR The results of five methods (i e , two FP,
one DD, one CLM, and one FTER) were in excellent agreement with median results
Walega et al (1984) report comparisons of CLM and TOLAS HNO3 measurements of
ambient and captive air performed during October and November of 1981 in Los Angeles,
CA The CLM gave erratic HNO3 results for ambient air Although CLM and TOLAS
measurements of HNO3 in captive air samples were highly correlated, lineai regression
analysis indicated significant biases
Measurements of HNO3 were made during June 1982 at a rural site in Ontario using
FP, TOLAS, and TA techniques (Anlauf et al, 1985) For daytime measurements, the FP
and TA measurements were 16% lower than the TOLAS results Nighttime TA results
exceeded those from the FP by a factor of 2 Roberts et al (1987) compared FP and TA
measurements of HNO3 made at a rural site in the Colorado mountains The TA results
were a factor of 3 higher than those of the FP It was concluded that there are unknown
atmospheric species that interfere with TA measurements of HNO3
Another HNO3 intercomparison study was conducted in Claremont, CA, in September
1985 (Hering et al., 1988) The methods compared include FP, DD, AD, TFR, TA, FTIR,
and TDLAS. For the whole study, comparison of method means against mean of methods
showed the FP to be 36% high, the DD to be 1 % low, the AD to be 21 % low, and the
TDLAS to be 13 % low Comparison of TFR means against DD means showed the TFR to
be 9 % high. Tunable-diode laser spectroscopy gave lower daytime and higher nighttime
readings than the DD In those few cases where the HNO3 concentrations were sufficiently
high to be detected by the FTTR, agreement within reported uncertainties was observed
between the FT3R and the FP, DD, AD, and TDLAS Results from the TA technique were
high at night and low during the day, and in view of large systematic differences, they were
not included in many of the reported analyses
During 1986, HNO3 data were collected using DD and FP techniques for 24-h periods
every 6 days at eight sites in the Los Angeles basin and at one background site (Solomon
et al., 1988). The annual average DD basin-wide estimate of HNO3 was 4 6 /*g/m The
corresponding FP estimate exceeded that of the DD by 3 4 /tg/m, or by approximately 80%
A study was conducted in January and February 1986 near Page, AZ (Benner et al,
1987). Twelve-hour gaseous HNO3 concentrations were measured with FP and AD The
6-54
-------
mean HNO3 concentration measured with the FP, 1 1 jwg/m, exceeded that measured with the
AD by 10%
A study was conducted in July 1986 on Long Island, NY, to compare HNO3
measurements resolved to a 6-h basis using high-volume FP, DD,> real-tune two-channel
(i e , nylon filter versus no nylon filter) CLM, and Al2(SO4)3-coated denuder thermal
desorption-to-CLM (Tanner et al, 1989) The FP results were highly correlated with those
of the DD The daytime real-tune CLM results were correlated with those of the DD, but
nighttime real-time CLM results exceeded DD results This may have been caused by the
retention of nighttime HONO on the nylon filter Results with the Al2(SO4)3-coated denuder
were scattered, mostly lower, and poorly correlated with the other methods
Daily measurements of HNO3 were made in the Research Triangle Park, NC, during
13 days in September and October 1986 (Sickles et al, 1990) Comparisons of the TOLAS
results with those of the AD, FP, and TFR revealed significant differences at the 0 05 level
for the comparison between the TDLAS and TFR results Significant differences were not
apparent in the other two cases Comparisons of the study wide means of daily ratios of AD,
FP, and TFR to TDLAS results showed the AD to be 5% low, the FP to be 8% high, and
the TFR to be 36 to 76% high
6.8 NITROUS ACID
The measurement of HONO in ambient atmospheres is receiving increased recent
attention Currently available techniques employ denuder (Ferm and Sjodin, 1985), AD
(DeSantis et al, 1985), CLM (Cox, 1974, Braman et al, 1986), PF/UF (Rodgers and
Davis, 1989), and absorption spectroscopy (Biermann et al, 1988, Tuazon et al, 1978)
6.8.1 Denuders
See Sections 636 and 6 7 2 for additional discussions of denuders As noted in
Section 621, Braman et al (1986) have employed a senes of open-channel denuders coated
with materials that act to preconcentrate HNO3, HONO, NO2, and NO from sampled
ambient air Nitrous acid is collected using a potassium iron oxide coated denuder located
downstream of a TA-coated denuder that removes HNO3 The HONO is thermally desorbed
6-55
-------
from the potassium iron oxide coated denuder and detected as NO with a CLM NO analyzer
Although sub-parts-per-billion sensitivity is claimed, field testing is needed to demonstrate
the adequacy of this method
Nylon filter material has also been used as an open-channel denuder to collect HONO
(Benner et al., 1988) Recent studies, however, have indicated that HONO may not be
retained quantitatively by nylon filters (Sickles and Hodson, 1989, Perrino et al , 1988)
Perm and Sjodin (1985) have used two conventional open-channel Na2CO3-coated
denuders in series for the determination of HONO in ambient air Nitrous acid is collected
quantitatively on the first denuder, whereas interferent artifacts from PAN and other NOX
species (i e., NO^ are collected in approximately equal amounts on both denuders Each
denuder is extracted and the extract is analyzed for NO2" using spectrophometry or 1C
To correct for interferent artifacts, the difference in NO2" found on the two denuders is
attributed to HQNO
Annular denuders have also been used to measure HONO using a similar approach
(DeSantis et al., 1985, Sickles, 1987, Sickles et al , 1988, Eatough et al , 1988, Vossler
et al., 1988; Koutrakis et al., 1988, Dasch et al , 1989, Appel et al , 1990, Perrino et al ,
1990). The MDL for a 1-day AD sample operating at 1 m3/h, assuming an extract volume
of 10 mL, negligible blank, and an 1C detection limit of 0 05 /ng NO27mL, is 0 02
(10 ppt). Estimates of precision for 1-day AD samples range from 5 to 15% (Sickles et al.,
1989; Vossler et al., 1988) In those cases where denuder sampling is performed over
extended periods in the presence of oxidants (i e , O3), the collected NO2" may be oxidized
to NO3" (Febo et al , 1986, Sickles et al , 1989, Sickles and Hodson, 1989, Permno et al ,
1988) To avoid this potential for sampling artifacts, an initial denuder coated with NaCl or
NaF is added to collect HNO3 as NO3" and pass HONO The difference in the sums of
NO2" and N03~ on the two downstream Na2CO3-coated denuders is attributed to HONO
(Febo et al., 1986, Perrino et al , 1990)
6.8.2 Chemilmninescence
It has been shown that HONO may be measured nonspecifically as NOX with a CLM
NOX analyzer (Cox, 1974, Sickles and Hodson, 1989, Spicer et al , 1991) As a result,
HONO can be determined using a CLM NOX analyzer to measure the NOX in a sampled air
6-56
-------
stream in the presence and absence of a HONO scrubber by attributing the difference to
HONO (Cox, 1974) Dilute aqueous NaOH solutions (Cox, 1974), filters coated with
Na2CO3 (Sickles and Hodson, 1989, Rickman et al, 1989, Kanda and Taira, 1990), and
denuders coated with Na2CO3 (Brauer et al, 1990) have been employed as HONO
scrubbers Because these scrubbers are not specific foi HONO, this approach cannot be used
to measure HONO in atmospheres containing other oxides of nitrogen that can be removed
by the scrubber (e g , HNO3) without the appropriate corrections As noted in the previous
section, Braman et al (1986) have used a system of selective denuders to collect HONO as
well as HNO3, NO2, and NO for subsequent thermal desorption and detection as NO with a
CLM NO analyzer
6.8.3 Photofragmentation/Laser-Induced Fluorescence
Photofragmentation/laser-induced fluorescence is discussed in Section 6 3 3 for the
measurement of NO2. In its present application, HONO is photofragmented to NO and
hydroxyl (OH) radical using radiation at 355 nm from a Nd YAG laser (Rodgers and Davis,
1989) Appreciable amounts of NO are also produced by the photolysis of NO2, which is
generally present along with HONO in ambient air As a result, the current method is based
on the detection of OH radical using SP-LJF With this technique, the resulting OH radical
2. -4-
is excited to the A £ state using laser radiation at 282 nm, and the fluorescence at 310 nm
that accompanies the A to X transition of the excited OH radical is monitored Detection
limits in the low tens of parts per trillion for 15-min integration times are claimed
6.8.4 Absorption Spectroscopy
Although HONO is potentially detectable (i e , MDL of 4 ppb) using a 23-m multipass
FTIR system with a 1-km path length, FUR has not been used to measure the concentration
of HONO in ambient air (Tuazon et al, 1978) A theoretical MDL for HONO of 10 ppt has
been claimed for 15-L integrated samples of ambient air using cryogenic sampling and
matrix-isolation FUR (Griffith and Schuster, 1987) Long-path UV/visible DOAS has been
used to determine HONO as well as other trace atmospheric constituents (see Section 634)
Using a 25-m multipass open system with a 0.8-km path length at wavelengths near 354 nm,
6-57
-------
an MDL of 0 6 ppb is claimed (Biermann et al, 1988) Using a single-pass open system
with a 10-km path length, an MDL of 20 ppt has been reported (Platt and Perner, 1983)
6.8.5 Calibration
The preparation of reliable calibration mixtures containing known concentrations of
HONO is difficult Atmospheres containing HONO as well as NO2 and NO may be
produced by acidifying solutions of sodium nitrite (NaNO2) with H2SO4 (Cox, 1974) The
method of Braman and de la Cantera (1986) uses a sublimation source where HONO is
produced by subliming oxalic acid onto solid NaNO2 at 30 to 60% RH Small
concentrations of HNO3, NO2, and NO may also be generated using the latter technique
Both methods require independent and periodic determination of the HONO concentration
because the source strengths are not necessarily constant A recent refinement in the method
of Cox (1974) has employed a flow generation system to produce stable concentrations of
HONO at the parts-per-biUion level (Kanda and Taira, 1990)
6.8.6 Inter-comparisons
Concentrations of HONO were determined in an indoor air quality study conducted in
two research houses using ADs and the CLM difference method (Brauer et al, 1990)
Reported concentrations (n = 60) were below 90 ppb, the results were highly correlated
(r = 0 86), and the slope was close to unity (CLM = 0 92 AD) In November and
December 1987, an outdoor study was conducted in Long Beach, CA, where simultaneous
HONO measurements were made using AD and DOAS on 6 days (Appel et al, 1990) The
AD samples were integrated over 4 and 6 h, and the 15-min DOAS results were averaged to
permit comparison with the AD results The HONO concentrations ranged from less than
1 to approximately 15 ppb, and the AD results were highly correlated with those of the
DOAS. Except at the low HONO levels that occurred during the midday periods, where the
AD results exceeded the DOAS results, the AD results were 7% lower than the DOAS
results. This difference is within the +30% uncertainty of the DOAS results in the study
6-58
-------
6.9 DINITROGEN PENTOXIDE AND NITRATE RADICALS
The NO3 radical photolyzes rapidly, and as a result, ambient concentrations are low
during daylight hours Dimtrogen pentoxide exists in a thermally sensitive equilibrium with
NO2 and the NO3 radical and can also react heterogeneously with water vapor to produce
HNO3 In addition, the NO3 radical reacts rapidly with NO to produce NO2 In spite of
their low ambient concentrations, N2O5 and the NO3 radical may have important roles in
both troposphenc and stratospheric NOX chemistry
Although N2O5 has not been measured in the troposphere, it has been observed in the
stratosphere using spectroscopic methods (Roscoe, 1982) In the troposphere, nighttime
N2O5 concentrations of up to 15 ppb have been inferred under the assumption of equilibrium
using measured NO2 and NO3 radical concentrations (Atkinson et al, 1986)
At concentrations above 5 ppb, measurement of N2O5 with FUR spectrometry appears
feasible using a 1-km path length near 1,250 cm" A theoretical MDL for N2O5 of 20 ppt
has been claimed for 15-L integrated samples of ambient air using cryogenic sampling and
matrix-isolation FUR (Griffith and Schuster, 1987)
Dimtrogen pentoxide is readily reduced to NO at temperatures above 200 °C and, as
noted in Section 631, may be measured nonspecifically as NOX with CLM NOX analyzers
(Bellinger et al, 1983, Fahey et al, 1985a) An N2O5 calibration system has been devised
using a crystalline sample at —80 °C, thermal dissociation of gaseous N2O5, scavenging of
the dissociation product (i e , the NO3 radical) with added NO to produce NO2, and a CLM
NO detector (Fahey et al, 1985b) This calibration technique focuses on the loss of NO,
and an accuracy of ±15% is claimed
Ambient concentrations of the NO3 radical have been made using DOAS, and
concentrations between 1 and 430 ppt have been observed (Atkinson et al, 1986)
Additional information on absorption spectroscopy is given rti Section 634 Using a 25-m
multipass open system with 0 8-km path length, an MDL of 20 ppt is claimed (Biermann
et al, 1988) Usmg an optical path length of 17 km and a wavelength of 662 nm, the
reported detection limit for the NO3 radical is 1 ppt (Platt et al, 1984) Noxon (1983),
using a passive absorption spectroscopic method with the moon as the light source, reports an
NO3 concentration of 0 25 ppt measured at a 3-km altitude from Mauna Loa, Hawaii
6-59
-------
6.10 PARTICULATE NITRATE
Atmospheric aerosols are chemically heterogeneous and occur in sizes ranging
nominally from <0 01 to 100 jwm Many methods are available for sampling ambient
aerosols, including unpactors, filtration, and filtration coupled with devices to remove
particles larger than a specified size (e g , elutnators, unpactors, and cyclones) The method
of choice usually depends on the particle size range and the chemical composition of the
aerosol of interest As an example, ambient concentrations of particles are subject to
National Ambient Air Quality Standards (Code of Federal Regulations, 1987a) These
standards focus on the concentration of particulate mass for all particles less than 10 jum in
equivalent aerodynamic diameter, rather than on the individual chemical species (e g ,
nitrates) comprising the collected particles
The particle size distribution of ambient particulate nitrate (PN) is bimodal (Kadowaki,
1977; Wolff, 1984, Yoshizumi, 1986, Wall et al, 1988) Particulate nitrate'is concentrated
in the coarse size (i e , greater than 2 5 jtim) in marine environments, where ambient HNO3
reacts with the coarse suspended sea salt (i e , NaCl) to form sodium nitrate (NaNO3)
Under other circumstances, the size distribution of PN will be determined by environmental
conditions and the relative presence of precursors, including HNO3, NH3, and acidic
aerosols. For example, in the eastern United States, during the summer, when the
concentration of acidic sulfates is high, the temperature is high, and the NH3 emissions are
low, the NH4NO3 ^ NH3 + HNO3 equilibrium is shifted to the right This and
metathetical reactions with acidic aerosols and gases make gaseous HNO3 available for
reaction with and retention by coarse soil-derived particles, giving rise to high concentrations
of PN in the coarse size range In contrast, some western urban areas (e g , Los Angeles
and Denver) have low SO2 emissions and adequate NH3 emissions to neutralize acidic
aerosols. These conditions favor the concentration of PN in the fine size range, presumably
as ammonium nitrate (NH4NO3)
6.10.1 Filtration
Particulate nitrates are generally collected by filtration techniques for subsequent
analysis. Using ambient dust, John and Reischl (1978) found the filtration efficiencies of
Nuclepore (polycarbonate, 0 8-/*m pore) and Whatman 41 (cellulose) filters to be less than
6-60
-------
1
90% Efficiencies exceeded 99% for Gelman GA" (ceEulose acetate), Gelman Spectrograde
(glass fiber), Gelman A (glass fiber), MSA (glass fiber), and EPA-grade (glass fiber) filters
For polytetrafluoroethylene (Teflon™) membrane filters, the efficiencies exceeded 99% for
Fluoropore 1-jwm pore, Ghia 1- to 3-fun pore, and Ghia 2- to 4-jnm pore filters, but did not
exceed 99 % for some tests with the Fluoropore 3-jttm pore and Ghia 3- to 5-jtim pore filters
The integrity of PN collected on filters may depend on storage conditions and other
factors Highsmith et al (1986) have attributed weight loss observed on quartz high-volume
and Teflon™ dichotomous filters to particle loss and volatilization during handling and
shipment Smith et al (1978) report a 73 % loss of NO3" from Gelman AE (glass fiber)
high-volume filters stored for 15 mo in the open at room temperature In contrast, filters
stored for 2 years in containers at -28 °C showed no loss of NO3" Witz et al (1990) have
observed 19% loss of nitrates on PM-10 samples collecled on Whatman QM-A (quartz and
glass fiber) filters after 1 week of room temperature storage Nitrate losses ranged from
28 to 50% after 1 mo Dunwoody (1986), using acid-tieated Whatman QM-A (quartz and
glass fiber) high-volume filters, found NO3" losses of 86% after 6 to 8 mo of dry room
temperature storage, whereas refrigerated filter extracts were stable over this period Witz
et al (1990) found nitrate losses to increase with decreasing filter alkalinity, increasing
acidity of the aerosol deposit, and increasing storage temperature Dunwoody (1986) found
that filters spiked with solutions containing potassium nitrate and other salts showed no NO3"
losses over 60 days of storage In contrast, filters spiked with HNO3 lost 70 to 90% of the
NO3" over a period of 3 days
Using filters to collect PN can also result in both positive and negative biases that occur
during the sampling process Some of the difficulties encountered with filtration techniques
for distinguishing between particulate and gaseous nitrate are also discussed in Section 671
Gas-filter interactions may lead to one type of positive bias Glass fiber filters have been
employed to collect particles including PN from ambient air, and at one point, glass fiber
filters were specified by the EPA for sampling total suspended particulate matter (Code of
Federal Regulations, 1987a) Glass fiber filters can retain gaseous HNO3 and to a lesser
extent promote the oxidation of gaseous NO2, leading to the formation of artifact nitrates and
the resulting positive biases (Appel et al, 1979, Spicer and Schumacher, 1979) Substantial
positive biases from HNO3 have been reported by Appel et al (1979) for Gelman A, GA"1,
6-61
-------
and Spectrograde; Whatman 41, MSA 1106 BH (glass fiber), and "EPA. grade" filters, by
Spicer and Schumacher (1979) for Millipore Nylon, Gelman E, A, AE, AA, and
Spectrograde (glass fiber), MSA 1106 BH, Millipore (cellulose acetate), Gelman Microquartz
(quartz fiber); and Pallflex E 70-2075W (quartz fiber) filters, and by Appel et al (1984) for
Gelman "EPA grade", Schleicher and Schuell (glass fiber), Schleicher and Schuell 1 HV
(glass fiber); Whatman EPM 2600 and EPM 1000 (glass fiber), Whatman QM-A, Pallflex
2500 QAST (quartz fiber), Gelman Microquartz, and Gelman ADL (quartz fiber) filters
Witz and Wendt (1981) report that the magnitudes of artifact nitrates on high-volume sampler
filters were ordered as follows Whatman EPM 1000 > Gelman AE (acid washed glass
fiber) > Gelman Microquartz > Pallflex TX40H120 (Teflon™-coated glass fiber) > Pallflex
2500 QAO (quartz fiber) Artifact nitrates on Gelman A and Pallflex TX40H120 filters
based on laboratory tests exceeded the amount found on Pallflex QAST filters by factors of
8.6 and 3.4, respectively (Mueller and Hidy, 1983) Substantial amounts of artifact nitrates
have been reported based on field studies using Gelman A and Pallflex TX40H120 filters
(Pierson et al., 1980) Higher ambient PN measurements were reported using S & S and
Whatman QM-A filters than with Gelman Microquartz, Pallflex 2500 QAST, or
Membrana/Ghia Zefiuor (Teflon™) filters (Rehme et al, 1984) Small biases were also
reported for EPA/ADL (quartz fiber) and Pallflex QAST filters (Spicei and Schumacher,
1979). Negligible artifact nitrates were reported for Fluoropore (Teflon™) (Appel et al,
1979, Mueller and Hidy, 1983) and Ghia Zefiuor filters (Appel et al, 1984), and no artifact
nitrates were reported for Nuclepore (0 8-/tm pore) and Millipore Mitex (Teflon™) filters
(Spicer and Schumacher, 1979) Good agreement was reported for PN collected on acid-
treated Pallflex 2500 QAO and Fluoropore filters (Forrest et al, 1982), in contrast to tests
where PN on Pallflex QAST exceeded that on Ghia Zefluor filters by 33% (Appel et al,
1984).
A second source of positive bias in using filtration for the collection of PN is the
retention of gaseous HNOs by particulate matter collected on the filter Appel et al (1980)
have reported increased retention of HNO3 with mass loading of particulate matter on Ghia
Zefluor filters
Negative biases may arise from at least two sources Particulate nitrates may react with
cocollected acidic aerosols or gases to release HNC>3 from the particulate catch, leading to
6-62
-------
one type of negative bias In the laboratory, the separate introduction of H2SO4 aerosols and
gaseous HC1 each resulted in appreciable losses and downstream recovery as nitrate of
preloaded NK^NC^ from Ghia Zefluor filters (Appel and Tokiwa, 1981) Harker et al
(1977) reported that nitrates collected during chamber experiments on Gelman Spectrograde
filters were displaced by sulfate-contaimng, and presumably acidic, aerosols according to a
metathetical reaction The introduction of H2SO4 aerosols to ambient particles preloaded on
acid-treated Pallflex 2500 QAO filters resulted in appieciable PN losses (Forrest et al,
1980) Pierson et al (1980) reported similar observations for PN on both glass and quartz
filters Negative correlations have been reported between the fraction of PN measured on
acid gas-denuded Teflon™ filters and both NH3-denuded measurements of strong acid on
Teflon™ filters (Appel and Tokiwa, 1981) and measurements of strong acid on acid-treated
Pallflex 2500 QAO filters (Forrest et al, 1982)
A second source of a negative bias with filtration for the collection of PN is the
volatilization of NH4NO3 Nominal 40 to 50% losses of nitrate due to volatilization have
been reported where laboratory air free of HNO3 and NH3 was drawn through Ghia Zefluor
filters preloaded with NH4NO3 Ambient air drawn through acid-treated Pallflex 2500 QAO
filters preloaded with NH4NO3 has shown nitrate volatilization losses ranging between 0 and
72% (Forrest et al, 1980)
Artifact nitrate formation depends to a large extent on the composition of the filter
material (e g , glass versus Teflon™) Artifact nitrate formation also increases with relative
humidity and decreases with temperature (Appel et al, 1979, Forrest et al, 1980)
To insure efficient particle collection and mirnmize artifact nitrate formation, Teflon™
membrane or selected quartz fiber filters are preferred over glass, Teflon™-coated glass,
cellulose, cellulose acetate, or polycarbonate filters Quartz fiber filters permit sampling at
high flow rates with modest pressure drops, however, they are fragile, require care in
handling (Rehme et al, 1984), are subject to some positive bias from artifact nitrate
formation (Appel et al, 1984), and may require pretreatment to insure low blank levels
(Leahy et al, 1980) Teflon™ membrane filters are the most nearly inert, but in contrast, are
subject to clogging with increased mass loadings (Rehitne et al, 1984)
6-63
-------
6.10.2 Denuders/Filtration
Many of the previously mentioned biases may be eliminated by deploying a
combination of denuders and filters (Appel et al, 1981) For additional information on
denuders, see Section 672 Biases involving interaction of HNO3 with the filter or
collected sample have been circumvented to some degree through the use of denuders
During sampling, gaseous HNO3 diffuses to the surface of the denuder and is collected by
reaction with the denuder surface, whereas particles pass through the device uncollected
An inert filter material (e g., Teflon™ membrane or quartz fiber) is then used to collect the
particulate matter Because the inert particulate filter is still subject to the negative biases of
liberated HNO3 and volatilized NH4NO3, the resulting NO3" must be collected on a backup
filter. Nylon, Na2CO3-coated, NaCl-coated, or NaF-coated filters may be used as backup
filters. Although pressure drop and capacity considerations favor coated filters for high flow
rate applications, due to the presence of the chloride or fluoride ions, this collection method
may not be compatible with an 1C finish for determining NO3"
As noted in Section 6 7 1, a minimum sensitivity of 0 02 jwg/m may be calculated under
the assumptions of 0.05 jwg/mL analytical detection limit, 10 mL extraction volume,
negligible blank, quantitative collection and extraction, and a sampled air volume of 24 m3
(i.e., 1 m3/h for 1 day) Thus, for a combination of a denuder and two filters, the PN
sensitivity should be approximately 0 04 j^g/m Median precision estimates of 4 to 16%
RSD have been reported for 22-h duration samples of fine PN (Vossler et al, 1988, Sickles,
1987), and a median precision of 4% RSD was reported for 12 1-week duration samples of
fine PN (Sickles, 1987)
6.10.3 Impactors
When size-resolved samples or size distribution information are needed, impactors may
be preferred over filters for sampling ambient aerosols The dichotomous sampler (Wolff,
1984, Wall et al, 1988) has employed a virtual impactor to provide individual aerosol
samples of ambient aerosol above and below a given size (e g , 2 5 pm) Selected high-
volume samples have employed the cascade impactor to collect ambient aerosols below a
given size (e g , 10 jum) Both cyclones and impactors have been used as the initial element
in AD samplers to exclude particles nominally larger than 2 5 jttm from air samples
6-64
-------
containing PN (Sickles et al, 1988, Vossler et al, 1988, Koutrakts et al, 1988) Cascade
impactors have been used to obtain size distribution information on ambient PN (Kadowaki,
1977, Yoshizumi, 1986, Wall et al, 1988, Sloane et al,, 1991)
Impactors may have as many as 10 stages, providing for the collection and size
discrimination of aerosols with aerodynamic diameters between 0 03 and 20 /*m Backup
filters are usually used following the final impactor stage to collect the remaining particles
Sampling artifacts using unpactors include particle bounce, HNO3 sorption, volatilization of
NH4NO3, and changes in size distribution from EH changes in the lower stages (Wall et al,
1988, Sloane et al., 1991) In a recent study using the Berner cascade impactor, the
presence of grease on impactor stages had no apparent effect, comparisons of size
distributions made with and without an HNO3 denuder preceding the impactor revealed no
HNO3 sorption effects, and PN losses from NH4NO3 volatilization were 8 % in comparison
with results from a dichotomous sampler (Wall et al., 1988) An evaluation of sampling
artifacts occurring for aerosol samples collected using unpactors versus filters has indicated
that volatilization losses depend on several parameters, including the surface area of the
aerosol deposit (Zhang, 1991) The relatively small volatilization loss with an impactor
(e g , 8%) observed by Wall et al (1988), in comparison to the 50% loss with filters noted
in Section 6 10 1 (Forrest et al, 1980), may be reflecting differences in the surface areas of
the aerosol deposited in the two types of samples
6.10.4 Analysis
After the collection of PN by the techniques discussed in the previous sections, samples
are analyzed directly or indirectly for nitrate Several methods have been used, including 1C
(Mulik et al, 1976), colonmetry (Mullin and Riley, 1955), derivatization/GC (Tesch et al,
1976), HPLC (Kamiura and Tanaka, 1979), voltametry (Bodim and Sawyer, 1977), ion
specific electrode (Dnscoll et al, 1972), FTTR (Bogard et al., 1982), and CLM (Yoshizumi
et al, 1985) Ion chromatography and colonmetry are the methods most commonly used as
analytical finishes for the determination of PN
Many of the methods for determining PN require the extraction of nitrates prior to
analysis Extraction of nitrate spiked onto nylon filters showed quantitative recovery using
1C eluent solution or basic (i e , 0 003N NaOH) solution, but not using water (Henng et al,
6-65
-------
1988). Similar tests with spiked Teflon™ membrane filters showed essentially quantitative
recovery in each extraction medium Other tests have shown good NO3" recoveries from
spiked and ambient nitrates on cellulose, glass fiber, and Teflon™ filters using 1C eluent and
ultrasonication, boiling deiomzed water, and sequential extraction in warm 1C eluent and
deionized water (Jenke, 1983)
In recent years, 1C has become a method of choice for the determination of many
anions and cations in solution. Ion chromatography uses conductimetnc detection and a
combination of resin columns to separate the ions of interest and stop or suppress the eluent
from the background (Small et al , 1975, Muhk et al , 1976) The biomide and phosphate
interferences noted by Mulik et al (1976) generally do not present problems with
environmental samples of PN In some cases where filter or denuder extracts are analyzed
by 1C, to permit good resolution of various peaks, care must be taken to prevent excessive
concentrations of chloride or H2O2 must be added to oxidize sulfite to sulfate One recent
study reported a precision estimate of 1 % for replicate NO3" measurements in extracts of
ambient samples where the concentration was above 0 15 jwg/mL (Sickles et al , 1988a)
Detection limits for NO3" of 0.025 to 0 1 /tg/mL have been reported using 1C with a 0 5 mL
sample loop (Anlauf et al , 1988, Mulik et al , 1976) As noted in Section 6 7 1, a
0.05 /4g/mL analytical detection limit for NO3" corresponds to an ambient concentration of
0.02 jwg/m on a single filter sampling at 1 m3/h for 1 day Although nonsuppressed 1C has
poorer detection limits than the previously described suppressed approach, successful
application to the analysis of nitrates in ambient aerosols has also been reported for
nonsuppressed 1C (Willison and Clark, 1984)
Various colonmetric methods for NO3" have been used In one widely automated
method, NO3" in the extract is reduced to NO2", which is diazotized and determined at
550 nm using the Griess-Saltzman Method (see Section 635) (Saltzman, 1954) The
reduction may be accomplished usmg a copper-cadmium (Cu-Cd) redactor column
(Technicon, 1972) or using hydrazine sulfate with copper as a catalyst under slightly basic
conditions (Mullin and Riley, 1955; Kamphake et al , 1967) The detection limits of
0.001 to 0.006 fig/mL claimed with these methods are somewhat moie sensitive than those
previously cited for the 1C method A comparison of the performance of 1C with this
6-66
-------
colonmetnc method for PN collected on Fluoropore fillers showed excellent agreement
(Fung et al, 1979)
Another colonmetnc technique, the Brucine Method, involves the reaction of NO3" with
brucine under acidic conditions (Kothny et al, 1972) The color is measured at 410 nm, and
a detection limit of 0 4 jug/mL has been reported Other colonmetnc methods involve the
nitration of 2,4-xylenol in the presence of H2SO4, followed by steam distillation and
absorbance measurement at 435 nm (Saltzman et al, 1972), or the nitration of toluene in the
presence of H2SO4, followed by extraction into toluene and absorbance measurement of the
mtrotoluene-toluene complex at 284 nm (Bhatty and Townshend, 1971) The sensitivity of
both methods is marginal (i e , 1 /tg/mL), and they are subject to interferences (Saltzman
et al, 1972, Appel et al, 1977, Norwitz and Keliher, 1978, Kamiura and Tanaka, 1979,
Bhatty and Townshend, 1971)
Using reactions similar to those descnbed above, Tesch et al (1976) have reacted NO3"
with benzene or other aromatic compounds in the presence of H2SO4 and measured the
resulting nitroaromatic compound using GC-ECD A sensitivity of 0 1 jug/mL and
applicability to determining NO3" in saliva, blood, drinking water, and airborne particles
were claimed The above method has been modified by Tanner et al (1979) using electron
capture-sensitive fluoroaromatic denvatizmg agents and a more effective catalyst (i e ,
trifluoromethanesulfonic acid) With a sensitivity of 0 01 /ig/mL, this method has been
applied to microliter-sized samples and the analysis of PN
High-performance liquid chromatography coupled with UV detection at 210 nm has
been used to measure NO3" in the aqueous extracts of PN from glass fiber filters (Kamiura
and Tanaka, 1979) No interferences were reported, and a detection limit of 0 1 jwg/mL was
claimed
A voltametnc technique for the measurement of NO3" in solution has been reported by
Bodim and Sawyer (1977) The technique is based on the reduction of NO3" by a Cu-Cd
catalyst that is formed on the surface of a pyrolytic graphite electrode The detection limit is
0 06 jtcg/mL, but NO2 is a direct interferent Favorable comparison was reported between
the results of this method and those of the Techmcon (1972) colonmetnc method for
determining NO3" in extracts of PN
6-67
-------
Ion specific electrodes have been used to measure NO3" in extracts of PN with a
detection limit of 1 /jg/mL (Simeonov and Puxbaum, 1977) Ion specific electrodes suffer
from poor sensitivity, potential drifts caused by variable agitation speed, frequent need for
restandardization, and interferences by other ions (Dnscoll et al, 1972) Spicer et al
(1978a) evaluated a NO gas-sensing electrode for the indirect measurement of NO3" in
solution Because the electrode responds to NO2" in solution, the appioach was to measure
NO2" in solution before and after reducing the NO3" to NO2" and attribute the difference to
NC>3~. Although the gas-sensing electrode was both sensitive (i e , 0 1 /*g/mL) and specific
for NO2", difficulties in the reduction step prevented further development
A dry technique using FUR for measuring NO3" incorporated in a potassium bromide
matrix from samples of ambient PN has been reported (Bogard et al, 1982) Absorbance
bands for NO3" occur at 2,430, 1,384, and 840 cm"1 Using the 1,384 cm"1 band, a
detection limit of 0 1 /tig NO3" per sample was reported, although the nearby ammonium ion
band under same circumstances may not permit distinct resolution of NO3" Techniques have
been developed recently that permit FTIR detection of nitrates and other species in samples
of ambient aerosols collected by filtration on thin Teflon™ membrane iliters using direct
transmission or by unpaction using attenuated total internal reflection (Johnson and Kumar,
1987) This method is currently in the research prototype stage of its development
The decomposition of NO3" followed by the CLM detection of the resulting NOX (see
Sections 6.2.1 and 6.3 1) has been used to determine NO3" in PN samples Thermal
decomposition can be applied to NO3" either on filters or in liquid extracts (Spicer et al,
1985). With this technique, NO3" is decomposed by rapid heating to 425 °C in an
N2 atmosphere, and the resulting NOX (i e , NO and NO^ is measured (i e , integrated)
using a conventional CLM analyzer Particulate nitrite is a direct interferent A detection
limit of 0.7 /*g/mL is claimed Comparison of CLM and 1C analyses of spiked and ambient
samples showed good agreement, although the 1C was more precise, especially at low
concentration levels Yoshizumi et al (1985) have modified a method developed by Cox
(1980) to reduce NO3" and NO2" in solution and measure the evolving NO using a
commercially supplied CLM instrument The method of Yoshizumi et al (1985) uses a flow
system and does not distinguish NO2" from NO3" (i e , NO2" is a direct interferent), but has a
NO3" detection limit of 0.001 jtg/mL The method of Cox (1980), although using a batch
6-68
-------
approach, does distinguish NO2" from NO3" and has respective detection limits of
0 00005 and 0 05 ^g/mL
It has been suggested that volatile and nonvolatile nitrates may be distinguished by
taking advantage of their different temperatures of volatilization (Yoshizumi and Hoshi,
1985) Samples of atmospheric particles collected by filtration or impaction are heated in a
furnace to the optimum volatilization temperature of NB^NC^ (i e , 160 °C) The volatilized
nitrate is then collected in water for subsequent determination (e g , by 1C) A similar
principle has been used in thermal denuders (Klockow et al, 1989) In this case, HNO3 is
collected at ambient temperature on a MgSO4-coated AD, and NH4NO3 is collected at
150 °C on a similar downstream denuder After sampling, denuders are heated in turn to
700 °C to liberate NOX for determination by a CLM NOX analyzer Sturges and Harrison
(1988) have reported several potential interferences with the volatilization approach Nitric
acid from volatilized NH4NO3 m the presence of NaCl, for example, will cause displacement
of the chloride as HC1 and formation of nonvolatile NaNO3 Differences in the thermal
stabilities of ammonium sulfate/mtrate double salts were also demonstrated These
observations cast doubt on the feasibility of thermal speciation of PN
6.11 NITROUS OXIDE
Ambient N2O levels have been measured by several methods These methods include
infrared spectroscopy (both absorption and emissions spectra), mass spectrometry,
manometry, and gas chromatography coupled with thermal conductivity, flame lonization,
ultrasonic phase shift, helium lomzation, and electron capture detectors (Pierotti and
Rasmussen, 1977) The most commonly used method employs GC-ECD with a detection
limit of 20 ppb (Thijsse, 1978) and a precision of ±3 % at the background level of 330 ppb
(Cicerone et al, 1978) Cassidy and Reid (1982) also report an expected MDL of 20 ppb
for N2O using TOLAS near 1,150 cm"1 (see Section 623 for more on TOLAS) As was
descnbed in Section 6 2 3, a laboratory prototype method, TTFMS, has been developed with
a projected MDL for N2O of 3 ppt (Hansen, 1989)
Calibration can be performed using commercially supplied cylinders of compressed gas
(Thijsse, 1978), dilution of pure N2O, N2O permeation tubes (Cicerone et al, 1978), or
6-69
-------
gravimetric preparation of calibration mixtures (Komhyr et al 1988) Standard reference
material mixtures of N2O and CO2 in air are also available at nominal N2O concentrations of
300 and 330 ppb (National Bureau of Standards, 1988)
6.12 SUMMARY
Since the publication in 1971 of the original version of Air Qualify Criteria for Nitrogen
Oxides, changes have occurred in the technology associated with the sampling and analysis
for ambient NOX and related species During the 1970s, roughly the period between
publication of the original Criteria Document and its first update and revision, several events
occurred that focused on the determination of NO2 in ambient air In 1973, the original
Reference Method was withdrawn because of unresolvable technical difficulties Major
methods development efforts over the next 3 to 4 years yielded both automated and manual
methods that were suitable for the determination of NO2 in ambient air As a result, EPA
designated a new Reference Method and Equivalent Methods for NO2 The Reference
Method specifies a measurement principle and calibration procedures, namely gas-phase
CLM (GP-CLM) with calibration using either GPT of NO with O3 or an NO2 permeation
device. The Sodium Arsemte Method in both the manual and continuous forms and the TGS
Method were also designated as Equivalent Methods Subsequently, commercial GP-CLM
instruments were designated as Reference Methods The sensitivity of these devices was in
the low parts-per-billion range, and, although the GP-CLM instruments were recognized as
being susceptible to interferences by other nitroxy species, it was believed that the
atmospheric concentrations of these compounds were generally low relative to NO2
In the 1980s, additional developments occurred Information from air quality
monitoring networks is now readily available and has shown the GP-CLM instruments to
have nominal precision and accuracy of ±10 to 15% and 20%, respectively, and to have
replaced manual methods to a large extent in network applications Heightened interest in
the research community on the speciation of atmospheric trace gases and specifically
nitrogen-containing species has prompted a new wave of methods development Although
the basic design and performance of the commercial instruments have remained essentially
unchanged, researchers have improved GP-CLM measurement technology and refined other
6-70
-------
instrumental methods to permit the determination of NO, NO2, and NOy in the low parts-
per-tnllion range Although GP-CLM NO detectors coupled with catalytic NO2-to-NO
converters are still not specific for NO2, they have proven useful for measuring NOy, and
GP-CLM NO detectors coupled with photolytic NO2-to-NO converters have shown unproved
specificity for NO2
A continuous liquid phase CLM device for sensitively detecting NO2 has been
developed and may be suitable to measure NO2 if interference problems can be overcome
Passive samplers for NO2 have been used primarily for workplace and indoor applications,
but hold promise for ambient measurements as well Gas chromatography with electron
capture detection is useful in the determination of PAN, other organic nitrates, and N2O.
Laser-induced fluorescence has been introduced to detect NO, NO2, and HONO with
high sensitivity and specificity Tunable-diode laser spectroscopy has been used to detect
NO, NO2, and HNO3 Long-path spectroscopy has also been used to detect NO, NO2,
HONO, and NO3 Two-tone frequency modulated spectroscopy holds promise for the
sensitive measurement of NO, NO2, PAN, HNO3, and N2O These spectroscopic methods
are research tools and are not yet easily or economically suited for routine monitoring
Interest in acidification of the environment has resulted in the development of methods
for HONO and HNO3 Integrative methods using denuders have been introduced to permit
sensitive determination of these and other species In recent years, the potential for artifacts
in using filters for sampling particulate matter and specifically particulate nitrate has been
recognized This has given nse to careful characterization of filter media for potential
artifacts and the use of combinations of denuders and filters to permit more specific
determination of nitrogen-containing gases and particulate nitrates in ambient air.
6-71
-------
REFERENCES
Adams, K M , Japar, S M , Pierson, W R (1986) Development of a MnO2-coated, cylindnoal denuder for
removing NO2 from atmosphenc samples Atmos Environ 20 1211-1215
Alden, M , Edner, H , Svanberg, S (1982) Laser momtonng of atmosphenc NO using ultraviolet
differential-absorption techniques Opt Lett 7 543-545
Anlauf, K G ; Felhn, P , Wiebe, H A , Schiff, H I, Mackay, G I, Braman, R S , Gilbert, R (1985)
A comparison of three methods for measurement of atmosphenc nitric acid and aerosol nitrate and
ammonium Atmos Environ 19 325-333
Anlauf, K. G., Wiebe, H A , Felhn, P (1986) Characterization of several integratrve sampling methods for
nitnc acid, sulphur dioxide and atmosphenc particles J Air Pollut Control Assoc 36 715-723
Anlauf, K. G , MacTavish, D C , Wiebe, H A , Schiff, H I, Mackay, G I (1988) Measurement of
atmosphenc nitnc acid by the filter method and comparisons with the tuneable diode laser and other
methods Atmos Environ 22 1579-1586
Appel, B. (1973) A new and more sensitive procedure for analysis of peroxybenzoyl nitrate J Air Pollut
Control Assoc. 23 1042-1044
Appel, B R , Tokiwa, Y (1981) Atmosphenc particulate nitrate sampling errors due to reactions with particulate
and gaseous strong acids Atmos Environ 15 1087-1089
Appel, B R ; Hoffer, E M , Kothny, E L , Wall, S M (1977) Interference in 2,4 xylenol piocedure for
nitrate determination in atmosphenc aerosols Environ Sci Technol 11 189-190
Appel, B R , Wall, S M , Tokiwa, Y , Haik, M (1979) Interference effects in sampling particulate nitrate in
ambient air Atmos Environ 13 319-325
Appel, B R., Wall, S M , Tokiwa, Y , Haik, M (1980) Simultaneous nitnc acid, particulate nitrate and acidity
measurements in ambient air Atmos Environ 14 549-554
Appel, B R.; Tokiwa, Y , Haik, M (1981) Sampling of nitrates in ambient air Atmos Environ 15 283-289
Appel, B R ; Tokiwa, Y , Haik, M , Kothny, E L (1984) Artifact particulate sulfate and nitrate formation on
filter media Atmos Environ 18 409-416
Appel, B R , Povard, V , Kothny, E L (1988) Loss of nitnc acid within inlet devices intended to exclude
coarse particles during atmosphenc sampling Atmos Environ 22 2535-2540
Appel, B. R., Winer, A M , Tokiwa, Y , Biermann, H W (1990) Companson of atmosphenc nitrous acid
measurements by annular denuder and differential optical absorption systems Atmos Environ Part A
24: 611-616
Atkinson, R , Winer, A M , Pitts, J N , Jr (1986) Estimation of night-time N2O5 concentrations from ambient
NO2 and NQ$ radical concentrations and the role of N2O5 in night-time chemistry Atmos Environ
20 331-339
Atlas, E (1988) Evidence for SC3 alkyl nitrates in rural and remote atmospheres Nature (London) 331
426-428.
6-72
-------
Baveja, A K , Chaube, A , Gupta, V K (1984) Extractive spectiophotometric method for the determination of
atmospheric nitrogen dioxide Atmos Environ 18 989-993
Benner, C L , Eatough, D J , Eatough, N L , Bhardwaja, P (1987) Evaluation of an annular denuder method
for the collection of atmospheric nitrogenous species in the southwest desert Presented at 80th annual
meeting of the Air Pollution Control Association, June, New York, NY Pittsburgh, PA Air Pollution
Control Association, paper no 87-63 6
Benner, C L , Eatough, N L , Lewis, E A , Eatough, D J , Huang, A A , Ellis, E C (1988) Diffusion
coefficients for ambient nitric and nitrous acids from denuder experiments in the 1985 nitrogen species
methods comparison study Atmos Environ 22 1669-1672
Bennett, B I (1979) Stability evaluation of ambient concentrations of sulfur dioxide, nitric oxide, and nitrogen
dioxide contained in compressed gas cylinders Research Triangle Park, NC U S Environmental
Protection Agency, Environmental Monitoring and Support Laboratory, EPA report no EPA
600/4-79-006 Available from NITS, Springfield, VA, PB-292749
Benoit, F M (1983) Detection of nitrogen and sulfur dioxides in the atmosphere by atmospheric pressure
lomzation mass spectrometry Anal Chem 55 2097-2099
Bhatty, M K , Townshend, A (1971) Spectrophotometnc determination of small amounts of nitrate and nitrite
by conversion to rutrotoluene and extraction into toluene Anal Chun Acta 56 55-60
Biermann, H W , Tuazon, E C , Winer, A M , Wallington, T J , Pitts, J N , Jr (1988) Simultaneous
absolute measurements of gaseous nitrogen species in urban ambient air by long pathlength infrared and
ultraviolet-visible spectroscopy Atmos Environ 22 1545-1554
Bodim, M E , Sawyer, D T (1977) Voltammetnc determination of nitrate ion at parts-per-bilhon levels Anal
Chem 49 485-489
Bogard, J S , Johnson, S A , Kumar, R , Cunningham, P T (1982) Quantitative analysis of nitrate ion in
ambient aerosols by Fourier-transform infrared spectroscopy Environ Sci Technol 16 136-140
Bollinger, M J , Sievers, R E , Fahey, D W , Fehsenfeld, F C (1983) Conversion of nitrogen dioxide, nitric
acid, and n-propyl nitrate to nitric oxide by gold-catalyzed reduction with carbon monoxide Anal Chem
55 1980-1986
Bollinger, M J , Hahn, C J , Parnsh, D D , Murphy, P C , Albntton, D L , Fehsenfeld, F C (1984) NOX
measurements in clean continental air and analysis of the contributing meteorology J Geophys Res
[Atmos ] 89 9623-9631
Bradshaw, J D , Rodgers, M O , Davis, D D (1982) Single photon laser-induced fluorescence detection of
NO and SC<2 for atmospheric conditions of composition and pressure Appl Opt 21 2493-2500
Bradshaw, J D , Rodgers, M O , Sandholm, S T , KeSheng, S , Davis, D D (1985) A two-photon
laser-induced fluorescence field instrument for ground-based and airborne measurements of atmospheric
NO J Geophys Res [Atmos ] 90 12861-12873
Braman, R S , de la Cantera, M A (1986) Sublimation sources lor nitrous acid and other nitrogen compounds
in air Anal Chem 58 1533-1537
Braman, R S , de la Cantera, M A , Han, Q X (1986) Sequential, selective hollow tube preconcentration and
chemiluminescence analysis system for nitrogen oxide compounds in air Anal Chem 58 1537-1541
6-73
-------
Brauer, M.; Ryan, P B , Suh, H H , Koutrakis, P , Spengler, J D , Leslie, N P , Billick, I H (1990)
Measurements of nitrous acid inside two research houses Environ Sci Technol 24 1521-1527
Bubacz, D K ; Daughtrey, E H , Pled, J D , Kronmiller, K G (1987) Lummox measurements of ambient
NO2 In' Proceedings of the 1987 EPA/APCA symposium on measurement of toxic and related air
pollutants, May; Research Triangle Park, NC Pittsburgh, PA Air Pollution Control Association,
pp 398-403, EPA report no EPA-600/9-87-010 Available from NTIS, Springfield, VA, PB88-113402
(APCA publication VIP-8)
Burkhardt, M R , Maniga, N I, Stedman, D H , Paur, R J (1988) Gas chromatographic method for
measuring nitrogen dioxide and peroxyacetyl nitrate in air without compressed gas cylinders Anal
Chem 60 816-819
Butcher, S S , Ruff, R E (1971) Effect of inlet residence tune on analysis of atmospheric nitrogen oxides and
ozone Anal Chem 43 1890-1892
Buttim, P , DiPalo, V., Possanzmi, M (1987) Coupling of denuder and ion chromatographic techniques for NO2
trace level determination in air Sci Total Environ 61 59-72
Cadoff, B C , Hodgeson, J (1983) Passive sampler for ambient levels of nitrogen dioxide Anal Chem
55 2083-2085
Carroll, M. A , McFarland, M , Ridley, B A , Albntton, D L (1985) Ground-based mtnc oxide measurements
at Wallops Island, Virginia J Geophys Res [Atmos ] 90 12853-12860
Cassidy, D. T , Reid, J (1982) Atmospheric pressure monitoring of trace gases using tunable diode lasers Appl
Opt 21 1185-1190
Cicerone, R J., Shelter, J D , Stedman, D H , Kelly, T J , Liu, S C (1978) Atmospheric N2O
measurements to determine its sources, sinks, and variations J Geophys Res C Oceans Atmos
83. 3042-3050
Code of Federal Regulations (1987a) National primary and secondary ambient air quality standards C F R
40- §50
Code of Federal Regulations (1987b) Ambient air monitoring reference and equivalent methods C F R
40 §53
Cox, R A (1974) The photolysis of gaseous nitrous acid J Photochem 3 175-188
Cox, R. D (1980) Determination of nitrate and nitrite at the parts per billion level by chemiluminescence Anal
Chem 52 332-335
Cox, R A , Roffey, M J (1977) Thermal decomposition of peroxyacetylrutrate in the presence of mtnc oxide
Environ Sci Technol 11 900-906
Crecelius, H -J , Forwerg, W (1970) Investigations of the "Saltzman factor " Staub Reinhalt Luft 30 23-25
Darley, E F , Kettner, K A , Stephens, E R (1963) Analysis of peroxyacyl nitrates by gas chromatography
with electron capture detection Anal Chem 35 589-591
Dasch, J M , Cadle, S H , Kennedy, K G , Mulawa, P A (1989) Comparison of annular denuders and filter
packs for atmospheric sampling Atmos Environ 23 2775-2782
6-74
-------
Davis, D D (1988) Atmospheric nitrogen oxides, their detection and chemistry In Third year report to
Coordinating Research Council Atlanta, GA Georgia Institute of Technology, pp 1-13
Davis, D D , Bradshaw, J D , Rodgers, M O , Sandholm, S T , KeSheng, S (1987) Free troposphenc and
boundary layer measurements of NO over the central and eastern North Pacific Ocean J Geophys Res
[Atmos]92 2049-2070
De Santis, F , Febo, A , Perrino, C , Possanzim, M , Liberti, A (1985) Simultaneous measurements of nitric
acid, nitrous acid, hydrogen chloride and sulfur dioxide no. air by means of high-efficiency annular
denuders In Proceedings of the ECE workshop on advancements in air pollution monitoring and
procedures, June, Freiburg, Federal Republic of Germany Bonn, Federal Republic of Germany Federal
Ministry of the Interior, pp 68-75
Delany, A C , Dickerson, R R , Melchior, F L , Jr , Wartburg, A F (1982) Modification of a commercial
NOX detector for high sensitivity Rev Sci lustrum 53 1899-1902
Dickerson, R R (1984) Measurements of reactive nitrogen compounds in the free troposphere Atmos Environ
18 2585-2593
Dickerson, R R , Delany, A C , Wartburg, A F (1984) Further modification of a commercial NOX detector
for high sensitivity Rev Sci Instrum 55 1995-1998
Dnscoll, J N , Berger, A W , Becker, J H , Funkhouser, J T , Valentine, J R (1972) Determination of
oxides of nitrogen in combustion effluents with a nitrate ion selective electrode J Air Pollut Control
Assoc 22 119-122
Drummond, J W , Volz, A , Ehhalt, D H (1985) An optimized chemiluminescence detector for troposphenc
NO measurements J Atmos Chem 2 287-306
Drummond, J W , Schiff, H I, Karecki, D R , Mackay, G I (1989) Measurements of NO2, NOX, O3, PAN,
HNO3, H2O2, and H2CO during the Southern California air quality study Presented at 82nd annual
meeting of the Air and Waste Management Association, June, Anaheim, CA Pittsburgh, PA Air and
Waste Management Association
Dunwoody, C L (1986) Rapid nitrate loss from PM10 filters J Air Pollut Control Assoc 36 817-818
Eatough, D J , White, V F , Hansen, L D , Eatough, N L , Blhs, E C (1985) Hydration of nitric acid and
its collection in the atmosphere by diffusion denuders Anal Chem 57 743-748
Eatough, N L , McGregor, S , Lewis, E A , Eatough, D J , Huang, A A , Ellis, E C (1988) Comparison of
six denuder methods and a filter pack for the collection of ambient HNO^Cg), HNO2(g), and SO2(g) in
the 1985 NSMC study Atmos Environ 22 1601-1618
Edner, H , Frednksson, K , Sunesson, A , Svanberg, S , Uneus, L , Wendt, W (1987) Mobile remote sensing
system for atmospheric monitoring Appl Opt 26 4330 4338
Ellis, E C , Margeson, J H (1974) Evaluation of tnethanolamme procedure for determination of nitrogen
dioxide in ambient air Research Triangle Park, NC U S Environmental Protection Agency, Quality
Assurance and Environmental Monitoring Laboratory, EPA report no EPA-650/4-74-031 Available
from NTIS, Springfield, VA, PB-237348/8BE
Fahey, D W , Eubank, C S , Hubler, G , Fehsenfeld, F C (1985a) Evaluation of a catalytic reduction
technique for the measurement of total reactive odd-nitrogen NOV in the atmosphere J Atmos Chem
3 435-468
6-75
-------
Fahey, D W , Eubank, C S , Huebler, G , Fehsenfeld, F C (1985b) A calibrated source of N2O5 Atmos
Environ 19 1883-1890
Fahey, D W , Hubler, G , Parnsh, D D , Wilhams, E J , Norton, R B , Ridley, B A , Singh, H B , Liu,
S. C , Fehsenfeld, F C (1986) Reactive nitrogen species in the troposphere measurements of NO,
NO2, HNO3, particulate nitrate, peroxyacetyl nitrate (PAN), O3, and total reactive odd nitrogen (NOy) at
Niwot Ridge, Colorado J Geophys Res [Atmos ] 91 9781-9793
Febo, A , De Santis, F , Perrino, C (1986) Measurement of atmospheric nitrous and nitric acid by means of
annular denuders In Physico-chemical behavior of atmospheric pollutants proceedings of the fourth
European symposium, September, Stresa, Italy Boston, MA D Reidel Publishing Company,
pp 121-125.
Febo, A ; De Santis, F , Liberti, A , Pemno, C (1988) Nitric acid-nitrate aerosol measurements by a diffusion
denuder a performance evaluation Atmos Environ 22 2062-2064
Federal Register (1986) Part 58 - ambient air quality surveillance appendix A - quality assurance requirements
for state and local air monitoring stations (SLAMS) F R (March 19) 51 9595
Fehsenfeld, F. C , Dickerson, R R , Hubler, G , Luke, W T , Nunnermacker, L J , Williams, E J , Roberts,
J M , Calvert, J G , Curran, C M , Delany, A C , Eubank, C S , Fahey, D W , Fried, A ,
Gandrud, B. W , Langford, A O , Murphy, P C , Norton, R B , Pickering, K E , Ridley, B A
(1987) A ground-based intercompanson of NO, NOX, and NOy measurement techniques J Geophys
Res [Atmos] 92 14710-14722
Fehsenfeld, F C , Drummond, J W , Roychowdhury, U K , Galvin, P J , Williams, E J , Buhr, M P ,
Parnsh, D D , Hubler, G , Langford, A O , Calvert, J G , Ridley, B A , Grahek, F , Heikes, B G ,
Kok, G L , Shelter, J D , Walega, J G , Elsworth, C M , Norton, R B , Fahey, D W , Murphy,
P C , Hovermale, C , Mohnen, V A , Demerjian, K L , Mackay, G I, Schiff, H I (1990)
Intercompanson of NO2 measurement techniques J Geophys Res [Atmos ] 95 3579-3597
Fcllin, P ; Brown, P M , Caton, R B (1984) Development of a nitric acid sampler Downsview, ON, Canada
Concord Scientific Corporation
Ferm, M (1986) A Na2CO3-coated denuder and filter for determination of gaseous HNO3 and particulate NO3"
in the atmosphere Atmos Environ 20 1193-1201
Perm, M , Sjodin, A (1985) A sodium carbonate coated denuder for determination of mtious acid in the
atmosphere Atmos Environ 19 979-983
Fincher, C L , Tucker, A W , Birnbaum, M (1978) Laser and flash lamp fluoresc ence NO2 monitors
a comparison In Schlossberg, H , ed Laser spectroscopy applications and techniques, [proceedings],
August, San Diego, CA Bellingham, WA Society of Photo-Optical Instrumentation Engineers,
pp. 137-140 (Proceedings of the Society of Photo-Optical Instrumentation Engineers v 158)
Folsom, B A , Courtney, C W (1979) Accuracy of chemiluminescent analyzers measuring nitric oxide in stack
gases J AirPollut Control Assoc 29 1166-1169
Fontijn, A ; Volltrauer, H. N , Frenchu, W R (1980) NOX (= NO + NO^ monitor based on an H-atom direct
chemiluminescence method Environ Sci Technol 14 324-328
Forrest, J , Tanner, R L , Spandau, D , D'Ottavio, T , Newman, L (1980) Determination of total inorganic
nitrate utilizing collection of nitric acid on NaCl-impregnated filters Atmos Enviion 14 137-144
6-76
-------
Forrest, J , Spandau, D J , Tanner, R L , Newman, L (1982) Determination of atmospheric nitrate and nitric
acid employing a diffusion denuder with a filter pack Almos Environ 16 1473-1485
Frednksson, K A , Hertz, H M (1984) Evaluation of the DIAL technique for studies on NO2 using a mobile
hdar system Appl Opt 23 1403-1411
Fuglsang, K (1986) A filter pack for determination of total ammonia, total nitrate, sulfur dioxide and sulfate in
the atmosphere Miljoestyrelsen, Roskilde, Denmark Riso National Laboratory, report no
MST-LUFT-A-103 Available from NITS, Springfield, VA, DE87750099
Fung, K , Grosjean, D (1985) Peroxybenzoylmtrate measurements in smog chambers and in urban air Sci
Total Environ 46 29-40
Fung, K K , Heisler, S L , Pnce, A , Nuesca, B V , Mueller, P K (1979) Comparison of ion
chromatography and automated wet chemical methods for analysis of sulfate and nitrate in ambient
particulate filter samples In Mulik, J D , Sawicki, E , eds Ion chromatographic analysis of
environmental pollutants volume 2 Ann Arbor, MI Ann Arbor Science Publishers Lie , pp 203-209
Gafmey, J S , Fajer, R , Senum, G I (1984) An improved procedure for high purity gaseous peroxyacyl nitrate
production use of heavy lipid solvents Atmos Environ 18 215-218
Gallagher, C C , Forsberg, C A , Fieri, R V , Faucher, G A , Calo, J M (1985) Nitric oxide and nitrogen
dioxide content of whole air samples obtained at altitudes from 12'to 30 km J Geophys Res [Atmos ]
90 7899-7912
Gay, B W , Jr , Noonan, R C , Bufahni, J J , Hanst, P L (1976) Photochemical synthesis of peroxyacyl
nitrates in gas phase via chlorine-aldehyde reaction Environ Sci Technol 10 82-85
Girman, J R , Hodgson, A T , Robinson, B K , Traynor, G W (1984) Laboratory studies of the temperature
dependence of the Palmes NO2 passive sampler In Proceedings of the national symposium on recent
advances in pollutant monitoring of ambient air and stationary sources, May, Raleigh, NC Research
Triangle Park, NC U S Environmental Protection Agency, Environmental Monitoring Systems
Laboratory, pp 152-166, EPA report no EPA-600/9-84-019 Available from NITS, Springfield, VA,
PB85-144053/REB
Goldan, P D , Kuster, W C , Albntton, D L , Fehsenfeld, F C , Connell, P S , Norton, R B , Huebert,
B J (1983) Calibration and tests of the filter-collection method for measuring clean-air, ambient levels
of nitric acid Atmos Environ 17 1355-1364
Gregory, G L , Hoell, J M , Jr , Torres, A L , Carroll, M A , Ridley, B A , Rodgers, M O ,
Bradshaw, J , Sandholm, S , Davis, D D (1990a) An uitercompanson of airborne nitric oxide
measurements a second opportunity J Geophys Res [Atmos ] 95 10129-10138
Gregory, G L , Hoell, J M , Jr , Carroll, M A , Ridley, B A , Davis, D D , Bradshaw, J , Rodgers, M O ,
Sandholm, S T , Schiff, H I, Hastie, D R , Karecki, D R , Mackay, G I , Harris, G W , Torres,
A L , Fried, A (1990b) An uitercompanson of airborne- nitrogen dioxide instruments J Geophys Res
[Atmos] 95 10103-10127
Gregory, G L , Hoell, J M , Jr , Ridley, B A., Singh, H B , Gandrud, B , Salas, L J , Shatter, J (1990c)
An uitercompanson of airborne PAN measurements J Geophys Res [Atmos ] 95 10077-10087
Gregory, G L , Hoell, J M , Jr , Huebert, B J , van Bramer, S E , LeBel, P J , Vay, S A , Mannaro,
R M ; Schiff, H I; Hastie, D R , Mackay, G I , Karecki, D R (1990d) An uitercompanson of
airborne mtnc acid measurements J Geophys Res [Atmos ] 95 10089-10102
6-77
-------
Griffith, D W T., Schuster, G (1987) Atmospheric trace gas analysis using matrix isolation Fourier transform
infrared spectroscopy J Atmos Chem 5 59-81
Grosjean, D ; Harrison, J (1985a) Peroxyacetyl nitrate comparison of alkaline hydiolysis and
chemiluminescence methods Environ Sci Technol 19 749-752
Grosjean, D , Harrison, J (1985b) Response of chemiluminescence NOX analyzers and ultraviolet ozone
analyzers to organic air pollutants Environ Sci Technol 19 862-865
Grosjean, D., Parmar, S S (1990) Interferences from aldehydes and peroxyacetyl mtrate when sampling urban
air organic acids on alkaline traps Environ Sci Technol 24 1021-1026
Grosjean, D , Fung, K , Collins, J , Harrison, J , Breitung, E (1984) Portable generator for on-site calibration
of peroxyacetyl mtrate analyzers Anal Chem 56 569-573
Grosjean, D ; Parmar, S S , Williams, E L , II (1990) Southern California air quality study a search for
methyl mtrate Atmos Environ Part A 24 1207-1210
Hansen, D A (1989) Measuring trace gases with FM spectroscopy EPRIJ 14(4) 42-43
Harker, A B , Richards, L W , Clark, W E (1977) The effect of atmospheric SO2 photochemistry upon
observed nitrate concentrations in aerosols Atmos Environ 11 87-91
Helas, G ; Flanz, M , Warneck, P (1981) Improved NOX monitor for measurements in troposphenc clean air
regions Int J Environ Anal Chem 10 155-166
Helas, G ; Broil, A , Rumpel, K -J , Warneck, P (1987) On the origins of night-time NO at a rural
measurement site Atmos Environ 21 2285-2295
Helmig, D , Mueller, J , Klein, W (1989) Improvements in analysis of atmospheric peroxyacetyl mtrate (PAN)
Atmos Environ 23 2187-2192
Hering, S V.; Lawson, D R , Allegnm, I, Febo, A , Pernno, C , Possanzim, M , Sickles, J E , H, Anlauf,
K G , Wiebe, A ; Appel, B R , John, W , Ondo, J , Wall, S , Braman, R S , Sutton, R , Cass,
G R ; Solomon, P A , Eatough, D J , Eatough, N L , Ellis, E C , Grosjean, D , Hicks, B B ,
Womack, J D , Horrocks, J , Knapp, K T , Ellestad, T G , Paur, R J , Mitchell, W J ,
Pleasant, M , Peake, E , MacLean, A , Pierson, W R , Brachaczek, W , Schiff, H I , Mackay, G I,
Spicer, C W , Stedman, D H , Winer, A M , Biermann, H W , Tuazon, E C (1988) The nitric acid
Shootout field comparison of measurement methods Atmos Environ 22 1'519-1539
Higashi, T ; Imasaka, T , Ishibashi, N (1983) Thermal lens spectrophotometry with argon laser excitation source
for nitrogen dioxide determination Anal Chem 55 1907-1910
Highsmith, V. R , Bond, A E , Howes, J E , Jr (1986) Particle and substrate losses from Teflon® and quartz
filters Atmos Environ 20 1413-1417
Hoell, J. M , Jr , Gregory, G L , McDougal, D S , Carroll, M A , McFarland, M , Ridley, B A , Davis,
D. D , Bradshaw, J , Rodgers, M O , Torres, A L (1985) An intercompanson of nitric oxide
measurement techniques J Geophys Res [Atmos ] 90 12843-12851
Hoell, J M , Jr , Gregory, G L , McDougal, D S , Torres, A L , Davis, D D , Bradshaw, J , Rodgers,
M O ; Ridley, B A , Carroll, M A (1987) Airborne intercomparison of nitric oxide measurement
techniques J Geophys Res [Atmos ] 92 1995-2008
6-78
-------
Holdren, M W , Rasmussen, R A (1976) Moisture anomaly in analysis of peroxyacetyl nitrate (PAN)
Environ Sci Technol 10 185-187
Holdren, M W , Spicer, C W (1984) Field compatible calibration procedure for peroxyacetyl nitrate Environ
Sci Technol 18 113-116
Holland, D M , McElroy, F F (1986) Analytical method comparisons by estimates of precision and lower
detection limit Environ Sci Technol 20 1157-1161
Hughes, E E , Rook, H L , Deardorff, E R , Margeson, J H , Fuerst, R G (1977) Performance of a
nitrogen dioxide permeation device Anal Chem 49 1823-1829
Hustvedt, C (1987) SAROAD quick look report [personal communication] Research Triangle Park, NC U S
Environmental Protection Agency
Jacobs, M B , Hochheiser, S (1958) Continuous sampling and ultramicrodeternunation of nitrogen dioxide in
air Anal Chem 30 426-428
Jenke, D R (1983) Comparison of three methods for the extraction of selected anions from media used for the
collection of airborne particulates J Air Pollut Control Assoc 33 765-767
John, W , Reischl, G (1978) Measurements of the filtration efficiencies of selected filter types Atmos Environ
12 2015-2019
Johnston, P V , McKenzie, R L (1984) Long-path absorption measurements of troposphenc NC^ in rural New
Zealand Geophys Res Lett 11 69-72
Joos, L F , Landolt, W F , Leuenberger, H (1986) Calibration of peroxyacetyl nitrate measurements with an
NOX analyzer Environ Sci Technol 20 1269-1273
Joseph, D W , Spicer, C W (1978) Chemiluminescence method for atmospheric monitoring of nitric acid and
nitrogen oxides Anal Chem 50 1400-1403
Joshi, S B , Bufalim, J J (1978) Halocarbon interferences in chemiluminescent measurements of NOX Environ
Sci Technol 12 597-599
Jung, J , Kowalski, J (1986) Direct ambient nitrogen dioxide measurement by visible light absorption
In Proceedings of the 1986 EPA/APCA symposium on measurement of toxic air pollutants, April,
Raleigh, NC Pittsburgh, PA Air Pollution Control Association, pp 845-855, EPA report
no EPA-600/9-86-013 Available from NTIS, Springfield, VA, PB87-182713 (APCA publication
VIP-7)
Kadowaki, S (1977) Size distribution and chemical composition of atmospheric particulate nitrate in the Nagoya
area Atmos Environ 11 671-675
Kamiura, T , Tanaka, M (1979) Determination of nitrate in suspended particulate matter by high-performance
liquid chromatography with U V detection Anal Chim Acta 110 117-122
Kamphake, L J , Hannah, S A , Cohen, J M (1967) Automated analysis for nitrate by hydrazine reduction
Water Res 1 205-216
Kanda, Y , Taira, M (1990) Chemiluminescent method for continuous monitoring of nitrous acid in ambient air
Anal Chem 62 2084-2087
6-79
-------
Kelly, T. J. (1986) Modifications of commercial oxides of nitrogen detectors for improved response Upton, NY
U S Department of Energy, Brookhaven National Laboratory, report no BNL-38000 Available from
NTIS, Springfield, VA, DE86010536
Kelly, T. J ; Stedman, D H , Kok, G L (1979) Measurements of H2O2 and HNO3 in rural air Geophys Res
Lett. 6 375-378
Kelly, T J , Spicer, C W , Ward, G F (1990) An assessment of the luminol chemilummescence technique foi
measurement of NO2 in ambient air Atmos Environ Part A 24 2397-2403
Klapheck, K., Winkler, P (1985) Sensitivity loss of a NOx-chemiluminescence analyzer due to deposit
formation Atmos Environ 19 1545-1548
Kley, D , McFarland, M (1980) Chemilummescence detector for NO and NO2 Atmos Technol 12 63-69
Klockow, D ; Niessner, R , Malejczyk, M , Kiendl, H , von Berg, B , Keuken, M P , Wayers-Ypelaan, A ,
Slamna, J (1989) Determination of nitric acid and ammonium nitrate by means of a computer-controlled
thermodenuder system Atmos Environ 23 1131-1138
Knapp, K T , Durham, J L , Ellestad, T G (1986) Pollutant sampler for measurements of atmospheric acidic
dry deposition Environ Sci Technol 20 633-637
Komhyr, W D., Dutton, E G , Thompson, T M (1988) A general gravimetric dilution technique for preparing
trace calibration gases N2O calibration gas preparation Environ Sci Technol 22 845-848
Kondo, Y., Iwata, A , Takagi, M (1983) A chemiluminescent NOx-detector for the aircraft measurement
J. Meteorol Soc Jpn 61 756-762
Kondo, Y , Matthews, W A , Iwata, A , Monta, Y , Takagi, M (1987) Aircraft measurements of oxides of
nitrogen along the eastern run of the Asian continent winter observations J Atmos Chern 5 37-58
Kosmus, W (1985) Summation method for monitoring nitrogen oxides Int J Environ Anal Chem
22 269-279.
Kothny, E L , Cook, W A , Dimitnades, B , Ferrand, E F , Nifong, G D , McDamel, P W , Saltzman,
B E , Weiss, F T (1972) Tentative method of analysis for nitrate in atmospheric parficulate matter
(brucine method) Health Lab Sci 9 324-326
Koutrakis, P , Wolfson, J M , Slater, J L , Brauer, M , Spengler, J D , Stevens, R K , Stone, C L (1988)
Evaluation of an annular denuder/filter pack system to collect acidic aerosols and gases Environ Sci
Technol 22 1463-1468
Leahy, D. F , Phillips, M F , Garber, R W , Tanner, R L (1980) Filter material for sampling of ambient
aerosols Anal Chem 52 1779-1780
Lindqvist, F (1985) Determination of nitnc acid in ambient air by gas chromatography/photoiomzation detection
after collection in a denuder J Air Pollut Control Assoc 35 19-23
LJpari, F. (1984) New solid-sorbent method for ambient nitrogen dioxide monitoring Anal Chem
56 1820-1826
Lonneman, W A (1977) PAN measurement in dry and humid atmospheres Environ Sci Technol
11- 194-195
6-80
-------
Lonneman, W A , Bufalim, J J , Seila, R L (1976) PAN and oxidant measurement in ambient atmospheres
Environ Sci Technol 10 374-380
Lonneman, W A , Bufalim, J J , Namte, G R (1982) Calibration procedure for PAN based on its thermal
decomposition in the presence of nitric oxide Environ Sci Technol 16 655-660
Mackay, G I , Schiff, H I (1987) Reference measurements of HNO^ and NO2 by tunable diode laser
absorption spectroscopy In Proceedings of the 1987 EPA/APCA symposium on measurement of toxic
and related air pollutants, May, Research Triangle Park, NC Pittsburgh, PA Air Pollution Control
Association, pp 367-372, EPA report no EPA-600/9-87-010 Available from NTIS, Springfield, VA,
PB88-113402 (APCA publication VIP-8)
MacPhee, R D , Higuchi, J E , Leh, F K V (1976) Comparison of oxidant measurement methods, ultraviolet
photometry, and moisture effects In Speciality conference on ozone/oxidants-interactions with the total
environment, March, Dallas, TX Pittsburgh, PA Air Pollution Control Association, pp 166-179
Maeda, Y , Aoki, K , Munemon, M (1980) Chemiluminescence method for the determination of nitrogen
dioxide Anal Chem 52 307-311
Matthews, R D , Sawyer, R F , Schefer, R W (1977) Interfeiences in chemiluminescent measurement of NO
and NC<2 emissions from combustion systems Environ Sci Technol 11 1092-1096
McClenny, W A , Galley, P C , Braman, R S , Shelley, T J (1982) Tungstic acid technique for monitoring
nitric acid and ammonia in ambient air Anal Chem 54 365-369
Meyrahn, H , Helas, G , Warneck, P (1987) Gas chromatographic determination of peroxyacetyl nitrate two
convenient calibration techniques J Atmos Chem 5 405-415
Micme, R M , Jr , Sokash, J A , Fntschel, B P , McElroy, F F , Thompson, V L (1983) Performance test
results and comparative data for designated reference methods for nitrogen dioxide Research Triangle
Park, NC U S Environmental Protection Agency, Enviionmental Monitoring Systems Laboratory, EPA
report no EPA-600/4-83-019 Available from NTIS, Springfield, VA, PB83-200238
Miller, D P (1984) Ion chromatographic analysis of Palmes tubes for nitrite Atmos Environ 18 891-892
Miller, D P (1988) Low-level determination of nitrogen dioxide in ambient air using the Palmes Tube Atmos
Environ 22 945-947
Mueller, P K , Hidy, G M (1983) The sulfate regional experiment (SURE) report of findings Palo Alto, CA
Electric Power Research Institute, EPRI report no EA-1901 3v
Mulawa, P A , Cadle, S H (1985) A comparison of nitric acid and particulate nitrate measurements by the
penetration and denuder difference methods Atmos Environ 19 1317-1324
Mulik, J D , Williams, D (1986) Passive sampling devices for NO2 In Proceedings of the 1986 EPA/APCA
symposium on measurement of toxic air pollutants, April, Raleigh, NC Pittsburgh, PA Air Pollution
Control Association, pp 61-70, EPA report no EPA-600/9-86-013 Available from NITS, Springfield,
VA, PB87-182713 (APCA publication VJP-7)
Mulik, J D , Williams, D E (1987) Passive sampling device measurements of NO2 in ambient air In
Proceedings of the 1987 EPA/APCA symposium on measurement of toxic and related air pollutants,
May, Research Triangle Park, NC Pittsburgh, PA Air Pollution Control Association, pp 387-397, EPA
report no EPA-600/9-87-010 Available from NTIS, Springfield, VA, PB88-113402 (APCA publication
VIP-8)
6-81
-------
Mulik, J , Puckett, R , Williams, D , Sawicki, E (1976) Ion chromatograpbic analysis of sulfate and nitrate in
ambient aerosols Anal Lett 9 653-663
Mullin, J. B., Riley, J. P (1955) The spectrophotometric determination of nitrate in natural waters, with
particular reference to sea-water Anal Chun Acta 12 464-480
Namiesnik, J , Gorecki, T , Kozlowski, E , Torres, L , Mathieu, J (1984) Passive dosimeters - an approach to
atmospheric pollutants analysis Sci Total Environ 38 225-258
National Aeronautics and Space Administration (1983) Assessment of techniques for measuring tropospheric
NxOy- proceedings of a workshop, August 1982, Palo Alto, CA Hampton, VA Langley Research
Center, NASA conference publication NASA-CP-2292 Available from NHS, Springfield, VA,
N84-13706
National Bureau of Standards (1988) NBS standard reference materials catalog 1988 89 Washington, DC U S
Department of Commerce, Office of Standard Reference Materials, NBS spec lal publication 260
Available from GPO, Washington, DC, S/N 003-003-02841-0
Nelson, G. O (1971) Controlled test atmospheres principles and techniques Ann Aibor, MI Ann Arbor
Science Publishers, Ihc , pp 126-134
Nicksic, S. W ; Harkms, J , Mueller, P K (1967) Some analyses for PAN and studies of its structure Atmos
Environ 1. 11-18
Nieboer, H., Van Ham, J (1976) Peroxyacetyl nitrate (PAN) in relation to ozone and some meteorological
parameters at Delft in The Netherlands Atmos Environ 10 115-120
Nielsen, T , Hansen, A M , Thomsen, E L (1982) A convenient method for preparation of pure standards of
peroxyacetyl nitrate for atmospheric analyses Atmos Environ 16 2447-2450
Norwitz, G ; Kehher, P N (1978) Inorganic interferences in the 2,4-xylenol spectrophotometnc method for
nitrate and their elimination Anal Chun Acta 98 323-333
Noxon, J. F (1978) Troposphenc NO2 J Geophys Res C Oceans Atmos 83 3051-3057
Noxon, J F (1983) NO3 and NO2 in the mid-Pacific troposphere J Geophys Res C Oceans Atmos
88. 11017-11021
Okita, T., Monmoto, S., Izawa, M , Konno, S (1976) Measurement of gaseous and particulate nitrates in the
atmosphere Atmos Environ 10 1085-1089
Palmes, E D , Tomczyk, C (1979) Personal sampler for NOX Am Ind Hyg Assoc J 40 588-591
Palmes, E D , Gunmson, A F , DiMattio, J , Tomczyk, C (1976) Personal sampler for nitrogen dioxide
Am Ind Hyg Assoc J 37 570-577
Pernno, C , De Santis, F , Febo, A (1988) Uptake of nitrous acid and nitrogen oxide's by nylon surfaces
implications for nitric acid measurement Atmos Environ 22 1925-1930
Pcrrino, C , De Santis, F , Febo, A (1990) Criteria for the choice of a denuder sampling technique devoted to
the measurement of atmospheric nitrous and nitric acids Atmos Environ Part A 24 617-626
Pierotti, D , Rasmussen, R A (1977) The atmospheric distribution of nitrous oxide J Geophys Res
82- 5823-5832
6-82
-------
Pierson, W R , Brachaczek, W W , Kormski, T J , Truex, T J , Butler, J W (1980) Artifact formation of
sulfate, nitrate, and hydrogen ion on backup filters Allegheny Mountain experiment J Air Pollut
Control Assoc 30 30-34
Platt, U , Perner, D (1983) Measurements of atmospheric trace gases by long path differential UV/visible
absorption spectroscopy In Kilhnger, D K , Mooradian, A , eds Optical and laser remote sensing
New York, NY Spnnger-Verlag, pp 97-105 (Springer series m optical sciences v 39)
Platt, U F , Winer, A M , Biermann, H W , Atkinson, R , Pitts, J N , Jr (1984) Measurement of nitrate
radical concentrations in continental air Environ Sci Technol 18 365-369
Poizat, O , Atkinson, G H (1982) Determination of nitrogen dioxide by visible photoacoustic spectroscopy
Anal Chem 54 1485-1489
Possanzini, M , Febo, A , Liberti, A (1983) New design of a high-performance denuder for the sampling of
atmospheric pollutants Atmos Environ 17 2605-2610
Possanzini, M , Febo, A , Cecchini, F (1984) Development of a KI annular denuder for NO2 collection Anal
Lett 17 887-896
Purdue, L J , Hauser, T R (1980) Review of U S Environmental Protection Agency NO^ monitoring
methodology requirements In Lee, S D , ed Nitrogen oxides and their effects on health Ann Arbor,
MI Ann Arbor Science Publishers, Inc , pp 51-76
Rehme, K A , Smith, C F , Beard, M E , Fitz-Simons, T (1984) Investigation of filter media for use in the
determination of mass concentrations of ambient particulate matter Research Triangle Park, NC U S
Environmental Protection Agency, Environmental Monitoring Systems Laboratory, EPA report no
EPA-600/4-84-048 Available from NTIS, Springfield, VA, PB84-199876
Rhodes, R C (1987) [Personal communication, interrogation of PARS database] Research Triangle Park, NC
U S Environmental Protection Agency
Rhodes, R C , Evans, E G (1988) Precision and accuracy assessments for state and local air monitoring
networks, 1986 Research Triangle Park, NC U S Environmental Protection Agency, Environmental
Monitoring Systems Laboratory, EPA report no EPA-600/4-88-007 Available from NTIS, Springfield,
VA, PB88-180138
Rickman, E E , Jr , Wright, R S (1986) Interference of nitrogenous compounds on chemiluminescent
measurement of nitrogen dioxide Research Triangle Park, NC Research Triangle Institute, RTI report
no RTT/3180/24-01F
Rickman, E E , Jr , Green, A H , Wright, R S , Sickles, J E , H (1988) Laboratory and field evaluations of
extrasensitive sulfur dioxide and nitrogen dioxide analyzers for acid deposition monitoring Research
Triangle Park, NC U S Environmental Protection Agency, Atmospheric Research and Exposure
Assessment Laboratory, EPA report no EPA-600/3-90-034 Available from NTIS, Springfield, VA,
PB90-201062
Rickman, E E , Jr , Green, A H , Wright, R S , Sickles, J E , n (1989) Laboratory and field evaluations of
extrasensitive sulfur dioxide and nitrogen dioxide analyzers for acid deposition monitoring Research
Triangle Park, NC Research Triangle Institute, RTI report no RTI/3999/18-04F
Ridley, B A , Hewlett, L C (1974) An instrument for nitric oxide measurements in the stratosphere Rev Sci
Instrum 45 742-746
6-83
-------
Ridley, B A , Carroll, M A , Gregory, G L , Sachse, G W (1988) NO and NC^ in the tioposphere
technique and measurements in regions of a folded tropopause J Geophys Res [Atinos ]
93. 15813-15830
Roberts, J M (1990) The atmospheric chemistry of organic nitrates Atmos Environ Part A 24 243-287
Roberts, J M , Norton, R B , Goldan, P D , Fehsenfeld, F C (1987) Evaluation of the tungsten oxide
denuder tube technique as a method for the measurement of low concentrations of nitnc acid in the
troposphere J Atmos Chem 5 217-238
Roberts, J. M ; Fajer, R W , Spnngston, S R (1989) Capillary gas chromatographic separation of alkyl nitrates
and peroxycarboxykc nitric anhydrides Anal Chem 61 771-772
Rodgers, M O , Davis, D D (1989) A UV-photofragmentation/laser-induced fluorescence sensor for the
atmospheric detection of HONO Environ Sci Technol 23 1106-1112
Rodgers, M O , Asai, K , Davis, D D (1980) Photofragmentation-laser induced fluorescence a new method
for detecting atmospheric trace gases Appl Opt 19 3597-3605
Roscoe, H K (1982) Tentative observation of stratospheric N2O5 Geophys Res Lett 9 901-902
Saltzman, B E (1954) Colonmetnc microdetermination of nitrogen dioxide in the atmosphere Anal Chem
26- 1949-1955
Saltzman, B E (1980) Critique of measurement techniques for ambient nitrogen oxides In Lee, S D , ed
Nitrogen oxides and their effects on health Ann Arbor, MI Ann Arbor Science Publishers, Inc ,
pp 31-50
Saltzman, B E , Cook, W A , Dimitnades, B , Ferrand, E F , Kothny, E L , Levin, L , McDamel, P W
(1972) Tentative method of analysis for nitrate in atmospheric particulate matter (2,4-xylenol method (1))
In Methods of air sampling and analysis Washington, DC American Public Health Association,
pp 322-324
Schiff, H I; Hastie, D R , Mackay, G I, Iguchi, T , Ridley, B A (1983) Tunable diode laser systems for
measuring trace gases in troposphenc air a discussion of their use and the sampling and calibration
procedures for NO, NO2, and HNO3 Environ Sci Technol 17 352A-364A
Schiff, H. I , Mackay, G I, Castledine, C , Hams, G W , Tran, Q (1986) A sensitive diiect measurement
NO2 instrument In Proceedings of the 1986 EPA/APCA symposium on m« asuremenf of toxic air
pollutants, April, Raleigh, NC Pittsburgh, PA Air Pollution Control Association, pp 834-844, EPA
report no EPA-600/9-86-013 Available from NTIS, Springfield, VA, PB87-182713 (APCA publication
VIP-7)
Shaw, R W , Jr , Stevens, R K , Bowermaster, J , Tesch, J W , Tew, E (1982) Measurements of atmospheric
nitrate and nitric acid the denuder difference experiment Atmos Environ 16 845-853
Sickles, J E , n, Hodson, L L (1989) Fate of nitrous acid on selected collection surfaces .Atmos Environ
23 2321-2324
Sickles, J E., II, Michie, R M (1987) Evaluation of the performance of sulfation and nitration plates Atmos
Environ 21 1385-1391.
6-84
-------
Sickles, J E , n, Wright, R S (1979) Atmospheric chemistry of selected sulfur-containing compounds outdoor
smog chamber study - phase 1 Research Triangle Park, NC U S Environmental Protection Agency,
Environmental Sciences Research Laboratory, pp 45-49, EPA report no EPA-600/7-79-227 Available
from NTIS, Springfield, VA, PB81-141525
Sickles, J E , n, McClenny, W A , Paur, R J (1987) Sampling and analytical methods development for dry
deposition monitoring Research Triangle Park, NC U S Environmental Protection Agency,
Environmental Monitoring Systems Laboratory, EPA report no EPA-600/4-87-011 Available from
NTIS, Springfield, VA, PB87-233318
Sickles, J E , II, Perrino, C , Allegnm, I , Febo, A , Possanzini, M , Paur, R J (1988a) Sampling and
analysis of ambient air near Los Angeles using an annulai denuder system Atmos Environ
22 1619-1625
Sickles, J E , n, Hodson, L L , McClenny, W A , Paur, R J , Ellestad, T G , Mulik, J D , Anlauf, K G ,
Wiebe, H A , Mackay, G E , Schiff, H I , Bubacz, D K (1988b) Field comparison of methods for
the measurement of contributors to acidic dry deposition Atmos Environ (submitted)
Sickles, J E , n, Hodson, L L , Rickman, E E , Jr , Saeger, M L , Hardison, D L , Turner, A R , Sokol,
C K , Estes, E D , Paur, R J (1989) Comparison of the annular denuder system and the transition
flow reactor for measurements of selected dry deposition species JAPCA 39 1218-1224
Sickles, J E , n, Grohse, P M , Hodson, L L , Salmons, C A , Cox, K W , Turner, A R , Estes, E D
(1990) Development of a method for the sampling and analysis of sulfur dioxide and nitrogen dioxide
from ambient air Anal Chem 62 338-346
Simeonov, V , Puxbaum, H (1977) A comparative study on the nitrate determination in airborne dust
Mikrochim Acta 2 397-403
Singh, H B , Salas, L J (1983) Methodology for the analysis of peroxyacetyl nitrate (PAN) in the unpolluted
atmosphere Atmos Environ 17 1507-1516
Singh, H B , Viezee, W (1988) Enhancement of PAN abundance in the Pacific marine air upon contact with
selected surfaces Atmos Environ 22 419-422
Sloane, C S , Watson, J , Chow, J , Pntchett, L , Richards, L W (1991) Size-segregated fine particle
measurements by chemical species and their impact on visibility impairment in Denver Atmos Environ
Part A 25 1013-1024
Small, H , Stevens, T S , Bauman, W C (1975) Novel ion exchange chromatographic metho4 using
conductimetnc detection Anal Chem 47 1801-1809
Smith, R G , Bryan, R J , Feldstein, M , Levadie, B , Miller, F A , Stephens, E R (1972) Tentative method
of analysis for peroxyacetyl nitrate (PAN) in the atmosphere (gas chromatographic method) In Methods
of air sampling and analysis Washington, DC American Public Health Association, pp 215-219
Smith, J P , Grosjean, D , Pitts, J N , Jr (1978) Observation of significant losses of particulate nitrate and
ammonium from high volume glass fiber filter samples stored at room temperature J Air Pollut Control
Assoc 28 930-933
Solomon, P A , Fall, T , Salmon, L , Lin, P , Vasquez, F , Cass, G R (1988) Acquisition of acid vapor and
aerosol concentration data for use in dry deposition studies in the South Coast Air Basin volume I
Pasadena, CA California Institute of Technology, Environmental Quality Laboratory, EQL report 25 to
CAARB
6-85
-------
Spicer, C W ; Schumacher, P M (1979) Particulate nitrate laboratory and field studies of major sampling
interferences Atmos Environ 13 543-552
Spicer, C W , Schumacher, P M , Kouyounyian, J A , Joseph, D W (1978a) Sampling and analytical
methodology for atmospheric particulate nitrates final report Research Triangle Park, NC U S
Environmental Protection Agency, Environmental Sciences Research Laboratory, EPA report
no EPA-600/2-78-067 Available from NTTS, Springfield, VA, PB-281933
Spicer, C. W , Ward, G F , Gay, B W , Jr (1978b) A further evaluation of microcoulometry for atmospheric
nitnc acid monitoring Anal Lett All. 85-95
Spicer, C. W , Howes, J E , Jr , Bishop, T A , Arnold, L H , Stevens, R K (1982) Nitric acid measurement
methods an intercompanson Atmos Environ 16 1487-1500
Spicer, C W , Joseph, D W , Schumacher, P M (1985) Determination of nitrate in atmospheric particulate
matter by thermal decomposition and chemtluminescence Anal Chem 57 2338-2341
Spicer, C. W ; Ward, G F , Kenny, D V , Leslie, N P , Bilhck, I H (1991) Measurement of oxidized
nitrogen compounds m indoor air Presented at Measurement of toxic and lelated air pollutants, May,
Durham, NC Pittsburgh, PA Air & Waste Management Association
Staehr, W , Tahmann, W , Weitkamp, C (1985) Range-resolved differential absorption hdar optimization of
range and sensitivity Appl Opt 24 1950-1956
Stephens, E R (1969) The formation, reactions, and properties of peroxyacyl nitrates (PANs) in photochemical
air pollution In. Pitts, J N , Jr , Metcalf, R L , eds Advances in environmental science and
technology v 1 New York, NY Wiley-Interscience, pp 119-146
Stephens, E R., Price, M A (1973) Analysis of an important air pollutant peroxyacetyl nitrate J Chem
Educ 50 351-354
Stephens, E R , Burleson, F R , Cardiff, E A (1965) The production of pure peroxyacyl nitrates J Air
Pollut Control Assoc 15 87-89
Sturges, W T ; Harrison, R M (1988) Thermal speciation of atmospheric nitrate and chloride a critical
evaluation Environ Sci Technol 22 1305-1311
Szonntagh, E. L (1979) Colonmetnc azo dye methods for the atmospheric analysis of nitrogen dioxide,
historical development Period Polytech Chem Eng 23 207-215
Tanner, R. L ; Fajer, R , Gaffhey, J (1979) Determination of parts-per-bilhon concentrations of aqueous nitrate
by denvatization gas chromatography with electron capture detection Anal Chem 51 865-870
Tanner, R L , Daum, P H , Kelly, T J (1983) New instrumentation for airborne acid rain i esearch
Int J Environ Anal Chem 13 323-335
Tanner, R L ; Kelly, T J , Dezaro, D A , Forrest, J (1989) A comparison of filter, denuder, and real-time
chemilununescence techniques for nitric acid determination in ambient air Atmos Environ
23 2213-2222
Technicon. (1972) Industrial method no 158-71W/tentative nitrate and nitrite in. water and seawater Tarrytown,
NY: Technicon Industrial Systems
6-86
-------
Tesch, J W , Rehg, W R , Sievers, R E (1976) Microdetemunation of nitrates and nitrites in saliva, blood,
water, and suspended particulates in air by gas chromatography J Chromatogr 126 743-755
Tbijsse, Th R (1978) Gas chromatographic measurement of nitrous oxide and carbon dioxide in air using
electron capture detection Atmos Environ 12 2001-2003
Torres, A L (1985) Nitric oxide measurements at a nonurban eastern United States site Wallops instrument
results from My 1983 GTE/CITE mission J Geophys Res [Atmos ] 90 12875-12880
Tuazon, E C , Graham, R A , Winer, A M , Easton, R R , Pitts, J N , Jr , Hanst, P L (1978) A kilometer
pathlength Fourier-transform infrared system for the study of trace pollutants in ambient and synthetic
atmospheres Atmos Environ 12 865-875
Tuazon, E C , Winer, A M , Graham, R A , Pitts, J N , Jr (1981) Atmospheric measurements of trace
pollutants long path Fourier transform infrared spectroscopy Research Triangle Park, NC U S
Environmental Protection Agency, Environmental Sciences Research Laboratory, EPA report
no EPA-600/3-81-026 Available from NTIS, Springfield, VA, PB81-179848
Vierkorn-Rudolph, B , Rudolph, J , Diederich, S (1985) Determination of peroxyacetylmtrate (PAN) in
unpolluted areas Int J Environ Anal Chem 20 131-140
Vinjamoon, D V , Ling, C -S (1981) Personal monitoring method for nitrogen dioxide and sulfur dioxide with
solid sorbent sampling and ion chromatographic determination Anal Chem 53 1689-1691
Vossler, T L , Stevens, R K , Paur, R J , Baumgardner, R E , Bell, J P (1988) Evaluation of improved
inlets and annular denuder systems to measure inorganic air pollutants Atmos Environ 22 1729-1736
Walega, J G , Stedman, D H , Shelter, R E , Mackay, G I, Iguchi, T , Schiff, H I (1984) Comparison of a
chemilummescent and a tunable diode laser absorption technique for the measurement of nitrogen oxide,
nitrogen dioxide, and nitric acid Environ Sci Technol 18 823-826
Wall, S M , John, W , Ondo, J L (1988) Measurement of aerosol size distributions for nitrate and major ionic
species Atmos Environ 22 1649-1656
Wallace, L A , Ott, W R (1982) Personal monitors a state-of-lhe-art survey J Air Pollut Control Assoc
32 601-610
Watanabe, I , Stephens, E R (1978) Reexamtnation of moisture anomaly in analysis of peroxyacetyl nitrate
Environ Sci Technol 12 222-223
Wendel, G J , Stedman, D H , Cantrell, C A , Damrauer, L (1983) Lununol-based nitrogen dioxide detector
Anal Chem 55 937-940
West, P W , Reiszner, K D (1978) Personal monitor for nitrogen dioxide Research Triangle Park, NC U S
Environmental Protection Agency, Environmental Sciences Research Laboratory, EPA report
no EPA-600/2-78-001 Available from NTIS, Springfield, VA, PB-277437
Williams, E L , II, Grosjean, D (1990) Removal of atmospheric oxidants with annular denuders Environ Sci
Technol 24 811-814
Willison, M J , Clarke, A G (1984) Analysis of atmospheric aerosols by nonsuppressed ion chromatography
Anal Chem 56 1037-1039
6-87
-------
Winer, A M ; Peters, J W , Smith, J P , Pitts, J N , Jr (1974) Response of commercial chemiluminescent
NO-NO2 analyzers to other nitrogen-containing compounds Environ Sci Technol 8 1118-1121
Winfield, T W. (1977) A method for converting NC>2 to NO by ferrous sulfate prior to chemihiminescent
measurements In Proceedings of the American Chemical Society 173rd national meeting, March, New
Orleans, LA Washington, DC American Chemical Society, pp 372-374
Witz, S.; Wendt, J. G (1981) Artifact sulfate and nitrate formation at two sites in the South Coast Air Basin
A collaborative study between the South Coast Air Quality Management Disfrict and the California Air
Resources Board Environ Sci Technol 15 79-83
Witz, S.; Eden, R W , Wadley, M W , Dunwoody, C , Papa, R , Torre, K J (1990) Rapid loss of particulate
nitrate, chloride and ammonium on quartz fiber filters during storage J Air Waste Manage Assoc
40. 53-61
Woebkenberg, M L (1982) A comparison of three passive personal sampling methods foi NO2 Am Ind Hyg
Assoc J 43:553-561
Wolff, G T. (1984) On the nature of nitrate in coarse continental aerosols Atmos Environ 18 977-981
Wright, R S , Tew, E L , Decker, C E , von Lehmden, D J , Barnard, W F (1987) Performance audits of
EPA protocol gases and inspection and maintenance calibration gases JAPCA 37 384-385
Yanagisawa, Y , Nishimura, H (1982) A badge-type personal sampler for measurement of personal exposure to
NO2 and NO in ambient air Environ Lit 8 235-242
Yoshizunu, K (1986) Regional size distributions of sulfate and nitrate in the Tokyo metropolitan area in summer
Atmos Environ 20 763-766
Yoshizurni, K.; Hoshi, A (1985) Size distributions of ammonium nitrate and sodium nitrate in atmospheric
aerosols Environ Sci Technol 19 258-261
Yoshizurni, K , Aoki, K , Matsuoka, T , Asakura, S (1985) Determination of nitrate by a flow system with a
chemiluminescent NOX analyzer Anal Chem 57 737-740
Zafinou, O C , True, M B (1986) Interferences in environmental analysis of NO by NO plus 03 detectors
a rapid screening technique Environ Sci Technol 20 594-596
Zhang, X (1991) Measurements of size-resolved atmospheric aerosol chemical composition with impactors data
integrity and applications [Ph D dissertation] Minneapolis, MN University of Minnesota Available
from- University Microfilms International, Ann Arbor, MI, order no 9116541
6-88
-------
7. AMBIENT AND INDOOR CONCENTRATIONS
OF NITROGEN OXIDES
7.1 INTRODUCTION
The preceding chapters describe the fundamental chemistry of nitrogen oxides (NOX),
the sources (ambient and indoor), the transformations to other forms that take place during
transport in the atmosphere, and the methods of measurement This chapter describes the
NOX concentrations that have been measured in natural and human environments, with the
major emphasis in this chapter placed on characterizing and summarizing the extensive
nitrogen dioxide (NO2) monitoring data that have been collected under ambient and indoor
conditions
In the course of their daily activities, individuals spend varying amounts of tune in a
variety of environments (outdoors, residential, occupational, public access buildings,
transportation vehicles, etc ) While in these environments, exposures to NO2 occur that can
demonstrate considerable variability in magnitude, frequency, and duration In determining
or assessing the potential for adverse health risks associated with exposure to NO2, it is
necessary to consider exposures that occur across all environments This chapter reviews the
available data on NO2 concentrations in two important environments that account for the
major fraction of nitrogen dioxide exposures, ambient ,air and the indoor residential
environment Little data exist for other environments
The vast majority of data on NO2 exists for ambient air and the residential indoor
environment The ambient air is important both because of exposures that occur there and
because of its impact on indoor air quality. Significant ambient concentrations of nitrogen
oxides are usually confined to (1) urban areas and (2) urban and rural areas near major
sources of NOX Long-term (multiple-year) patterns and trends in NO2 concentrations are
available only from stationary continuous monitors Ambient data are reviewed here for
peak annual averages, trends, and seasonal, diurnal, and distributional patterns
The indoor residential environment is important because of the considerable amount of
tune individuals spend there (Szalai, 1972, Ott, 1989, Robinson, 1977, Wiley et al,
1991a,b) and the existence of sources of NO2 (combustion sources) Data on indoor nitrogen
7-1
-------
dioxide concentrations are collected predominantly in selected settings dunng comparatively
short-term (days or weeks) studies The indoor data are reviewed within the framework of
the mass balance model under equilibrium conditions Emphasis is placed on assessing the
contribution of outdoor NO2 concentrations on indoor levels and the impact of indoor
concentrations, reactive decay rates for NO2 and indoor reaction products (HONO)
Chapter 8 considers the available data on integrated NO2 exposures that occur across
different environments and are determined either by personal monitoring or through
modeling
7.2 AMBIENT AIR CONCENTRATIONS OF NITROGEN OXIDES
7.2.1 Introduction
As discussed in Chapter 6, most measurements of NOX have been made by devices that
convert NO2 to nitric oxide (NO), which is then measured by chemilummescence (National
Research Council, 1991) Comparison of these measurements with more specific techniques
suggests that all surface converters that can transform NO2 to NO also convert other reactive
NOX species, such as peroxyacetyl nitrate (PAN), to NO, thereby causing interference
In urban locations, where the local NO sources are typically large, NO and NO2 are
probably the dominant constituents of the total reactive nitrogen, which comprises NOX,
nitric acid (HNO3), nitrate radical (NO3), dimtrogen pentoxide, nitrous acid (HONO), PAN,
and other organic nitrogen compounds. Thus, interference from PAN and other NOX species
in urban areas is believed to be relatively small, in rural and remote locations, however,
interferences may be substantial (National Research Council, 1991)
Nitrogen oxide concentrations have been measured in numerous U S nonurban
monitoring locations (Table 7-1) Nonurban monitoring areas are not necessarily isolated
from local sources of pollution (e g , Kelly et al, 1982, Lefohn et al, 1991), many of the
monitoring locations listed in Table 7-1 may be affected by anthropogenic sources (National
Research Council, 1991) The NOX concentrations usually exceed 1 ppb and exhibit short-
term variability. For isolated rural sites and coastal inflow areas in the United States, the
NOX concentrations generally range from a few tenths to 1 ppb (Table 7-2) The
concentrations in the atmospheric boundary layer and lower free troposphere in remote
7-2
-------
TABLE 7-1. AVERAGE NITROGEN OXIDES CONCENTRATIONS MEASURED AT
U.S. NONURBAN MONITORING LOCATIONS3
Location
Fort McHenry, MD
Dubois, PA
McConnelsville, OH
Wilmington, OH
Wooster, OH
Bradford, PA
Creston, LA
Dendder, LA
Montague, MA
Scranton, PA
Indian River, DE
Research Triangle Park, NC
Lewisburg, WV
Duncan Falls, OH
Fort Wayne, IN
Rockport, IN
Giles County, TN
Jetmore, KS
Lamoure County, ND
Wnght County, MN
Traverse County, MN
Scotia, PA
Scotia, PA
NO
(Ppb)
ND
ND
ND
ND
ND
2
4
1
2
3
3
10
1
1
3
3
3
1
24
48
33
27
32
30
35
29
36
48
40
20
NO2
(ppb)
6b
10b
6b
6b
6b
3b
2b
3b
3b
llb
5b
13b
5b
8b
7b
?b
10b
4b
17b
15b
28b
21b
54b
67b
58b
47b
37b
36b
29b
22b
NOX
(ppb)
ND
ND
ND
ND
ND
5b
6b
4b
5b
14b
8b
23b
6b
9b
10b
10b
13b
5b
41b
63b
61b
48b
86b
97b
93b
76b
73b
84b
69b
42b
30b
3 lb
References
Research Triangle
Institute (1975)
Decker et al (1976)
Martinez and Singh
(1979)
Pratt etal (1983)
>
Parnsh et al (1986)
Pamshetal (1988)
aNO = Nitnc oxide
NO2 = Nitrogen dioxide
bUpper limit for NO2 and NOX
Source National Research Council (1991)
NOX = Nitrogen oxides
ND = No data
7-3
-------
TABLE 7-2. AVERAGE MIXING RATIOS MEASURED AT ISOLATED U.S. RURAL
SITES AND COASTAL INFLOW SITES3
Location
NO
(ppb)
NO2
(ppb)
NOX
(PPb)
References
Niwot Ridge, CO
Pierre, SDC
Schaeffer Observatory,
Whiteface Mountain, NY
Niwot Ridge, CO
Niwot Ridge, CO
Point Arena, CA
0-2b Kelly etal (1980)
1 2b Kelly et al (1982)
<;0 2 1 lb Kelly et al (1984)
0 80 Bellinger et al (1984)
056 Fehsenfeld et al (1987)
0 37 Parnsh et al (1985)
aNO = Nitric oxide
NO2 = Nitrogen dioxide
NOX = Nitrogen oxides
bUpper limit for NO2 and NOX
°Measurement site located 40 km WNW of Pierre
Source National Research Council (1991)
maritime locations are in the range 0 02 to 0 04 ppb, whereas concentrations of NOX in
remote tropical forests have been reported to be in the range from 0 02 to 0 08 ppb (National
Research Council, 1991) The higher concentrations experienced in the remote tropical
forests as compared with those observed in remote marine locations may be due to biogenic
NOX emissions from the soil (Kaplan et al , 1988, Torres and Buchan, 1988)
Elevated concentrations of NOX occur in or near urban areas because of the dominant
role of anthropogenic emissions in the budget of atmospheric NOX and the fact that the
sources of these emissions tend to be located in or near these areas (National Research
Council, 1991) Concentrations of NOX decline rapidly as pollution plumes travel away from
the urban core (Kelly et al, 1986)
7.2.2 Ambient Air Concentrations of Nitric Acid and Nitrate Aerosol
Photochemically initiated reactions oxidize NOX to HNO3 in the atmosphere (Spicer,
1977). In addition, nitrate ion (NO3") aerosols are formed as a result of reaction products of
7-4
-------
HNO3 Nitric acid is a major contributor to airborne strong acidity, the measurement of
which is intimately related to the accurate determination of particulate nitrate, a contributor
to visibility reduction. Aerosol NO3" is formed by the reaction of HNO3 with alkaline
aerosols (Wolff, 1984) and ammonia (NH3) The average concentrations of HNO3 and NO3"
are generally in the range 0 1 to 20 ppb and 0 1 to 10 ppb, respectively (Allegnni and
De Santis, 1989) The values measured at rural and urban sites, respectively, are shown in
Tables 7-3 and 7-4 Because there are conflicting reports on the ability of filters to
accurately separate HNO3 from NO3" aerosol (Lindberg et al, 1990), in some cases, it may
be more appropriate to focus on the total NO3 (HNO3 + NO3") than on the individual
components
TABLE 7-3. AVERAGE CONCENTRATIONS OF NITRIC ACID AND
NITRATE IONS MEASURED AT RURAL SITES3
Location
Bermuda
Berkshire Mtns , MA
Lewes, DE
Near Pierre, SDC
Oak Ridge, TN
Smoky Mountains, NC
Luray, VA
Whiteface Mtn , NY
Coweeta, NC
Thompson Forest, WA
Duke Forest, NC
Huntington Forest, NY
Rowland, ME
Whiteface Mtn , NY
HNO3
0 066 ppb
—
0 52 ppb
—
3 0 /xg/m3
0 33 ppb
0 36 ppb
0 20 ppb
1 8 /*g/m3
0 9 /xg/m3
2 8 /xg/m3
1 8 /xg/m3
OO «. /**»"^
o /ig/m
1 2 /xg/m3
NO3"
0 15 /xg/m3b
0 10 /xg/m3b
—
<0 10 /xg/m3
0 19 /tg/m3
0 31 /xg/m3
0 44 /xg/m3
0 25 /tg/m3
—
—
—
—
—
—
References
Wolff etal (1986a)
Wolff and Korsog (1989)
Wolff etal (1986b)
Kelly etal (1982)
Lindberg etal (1990)
Cadle and Mulawa (1988)
Cadleetal (1982)
Kelly etal (1984)
Hanson etal (1992)
Hanson etal (1992)
Hanson etal (1992)
Hanson etal (1992)
Hanson et al (1992)
Hanson etal (1992)
aHNO3 = Nitric acid
NO3" = Nitrate ion
Fine particle measurement for NO3"
Measurement site located 40 km WNW of Pierre, SD
7-5
-------
TABLE 7-4. AVERAGE CONCENTRATIONS OF NITRIC ACID AND
NITRATE IONS MEASURED AT URBAN SITES3
Location
Claremont, CA
Long Beach, CA
Denver, CO
Warren, MI
Warren, MI
Warren, MI
Boston, MA
HNO3
10 3 /tg/m3
1.9 /tg/m3
2 69 /*g/m3e
1 24 jKg/m3
1 14 /ig/m3
1 42 ppbf
0 59 ppbg
NO3"
12 4 /*g/m3b
7 9 jug/m30
20 2 /*g/m3b
14 6 Atg/m30
4 1 ,*g/m3<;
0 9 jtg/m3d
3 06 ^g/m3e
3 70 /tg/m3
4 20 /ig/m3
6 06 nmol/m3f
11 76 nmol/m3g
References
Wolff etal (1991)
Wolff etal (1991)
Wolff (1984)
Cadle (1985)
Dasch et al (1989)
Dasch and Cadle (1990)
Braueretal (1991)
8HN03 = Nitnc acid
NO3" = Nitrate ion
fraction
"TPine particle fraction
Coarse particle fraction
"Summer seasonal ambient concentration
Summer average ambient concentration
^Winter average ambient concentration
7.2.3 Ambient Air Concentrations of Nitric Oxide and Nitrogen Dioxide
7.2.3.1 Data Availability and Exposure Considerations
Most data on ambient concentrations of NO and NO2 in the United States are available
from a network of monitoring stations that was established to determine compliance with the
National Ambient Air Quality Standard (NAAQS) for NO2, which is 0 053 ppm or
o
100 /ig/m (annual average) Information is readily available from the data base supported
by this network through the U S Environmental Protection Agency's (EPA's) computerized
information system, Aerometric Information Retrieval System (AIRS) Although much of
this information is closely related to compliance and enforcement, the data can also be used
for determining patterns and trends and as inputs to exposure assessment (e g , Lefohn et al,
7-6
-------
1991) In some cases, the data can be used to augment existing epidemiological studies,
where only indoor air data have been collected
The National Air Monitoring Network consists of three types of sites The National
Air Monitoring Station (NAMS) sites are located in areas where the concentrations of NO2
and subsequent potential human exposures are expected to be high Criteria for these sites
have been established by regulation (Federal Register, 1979) to meet uniform standards of
siting, quality assurance, equivalent analytical methodology, sampling intervals, and
instrument selection to ensure consistency among the reporting agencies For NO2, NAMS
sites are located only in urban areas with populations exceeding 1 million The other two
types of sites are States and Local Air Monitoring Station and Special Purpose Monitor sites,
which meet the same rigid criteria for the NAMS sites but may be located in areas not
necessarily directed toward high concentrations and elevated human exposure
For NO2 and NO, the sampling interval is 1 h and the instrument method used for all
stations is chemilummescence These instruments operate continuously and produce a
measurement every hour In order to produce a valid annual average, at least half the
possible 8,760 hourly readings must be reported
In the following subsections, data from the AIRS network are extracted and analyzed to
provide background for specific exposure issues considered in later chapters The analyses
proceed from a national picture of peak annual averages in Metropolitan Statistical Areas
(MSAs) through national 10- and 3-year trends to characteristic seasonal and diurnal patterns
at selected stations and from a brief examination of the incidence of episodic 1-h levels and
associated annual averages No attempt was made to include information on ambient air
concentrations from monitoring sites other than those in the AIRS network Likewise, the
major emphasis is on NO2 information, with some discussion on NO concentrations Little
information on other NOX species is available
Hourly average concentration information (the absolute value of the highest NO2
average concentrations, and when these concentrations occurred) is summarized for urban,
rural forested, and rural agricultural areas in the United States The land use designation of
"rural" does not imply that a specific location is isolated from anthropogenic influences
Rather, the designation only implies the existing use of the land No attempt has been made
to select isolated sites The large variation among point-source strengths (e g, electric
7-7
-------
generating plants) makes it difficult to describe "typical" NO2 exposures A monthly average
was calculated for each hour of the day The monthly averages for each hour of the day
were compared with one another to characterize the seasonal and diurnal patterns occurring
at a specific site, in most cases, the increase in monthly average concentration was correlated
with the occurrence of higher hourly average concentrations
7.2.3.2 Trends in Ambient Nitrogen Dioxide Concentrations
In order to be included in the 10-year trend analysis in the annual National Air Quality
and Emissions Trend Report (U S Environmental Protection Agency, 199 Ib), a station must
have reported valid data for at least 8 of the last 10 years A companion analysis of the most
recent 3 years requires valid data in all 3 years Analyses in the above report cover the
periods 1981 to 1990 and 1988 to 1990, respectively, 166 sites met the 10-year requirement,
211 met the 3-year requirement Of the 166 10-year sites, 42 were NAMS sites
For the period 1981 to 1990, there were indications of a downward trend for the
composite annual average NO2 concentration for both the 166 sites and the 42 NAMS sites
subset (U.S. Environmental Protection Agency, 1991b) Using the full set of data, the 1990
composite NO2 average was 8 % less than the 1981 level, a statistically significant difference
A similar trend was observed with the NAMS sites that, for NO2, are located only in urban
areas with populations of 1 million or greater (Figure 7-1) Using the 1980 to 1989 period,
the composite annual average is strongly correlated to population size (U S Environmental
Protection Agency, 1991a) When sites in MSAs with 250,000 to 500,000 population and
500,000 to 1 million people were compared to sites with more than 1 million, there was
a regular pattern that persisted over the full 10 years The sites with populations over
1 million were 0 01 ppm higher than those with 250,000 to 500,000, with the mid-population
sites in between, as seen in Figure 7-2
7.2.3.3 Exposure Patterns Observed for Ambient Nitrogen Dioxide and Nitric Oxide
Concentrations—Urban
The NO2 hourly average concentrations tend to be less than 0 001 ppm in remote areas,
0.001 to 0.020 ppm in rural areas, 0 02 to 0 20 ppm in moderately polluted areas, and 0 2 to
0.5 ppm in heavily polluted areas (Legge et al, 1990) Information in Tables 7-2 and 7-3
7-8
-------
o.
o
U.UD
005 -
004 -
003 -
002 -
001 -
n nn
M A AO^
NAAUo
^ *$ li''T"^^~"iK77~T-r
z s 4 - j I I- jfe jfc — 1 3
I a * -_. -•--=» s. =e * T
_K J- 31 J ^- J. J J| J^
• NAMS SITES (36) ° ALL SITES (148)
i 1 1 1 1 1 1 1 r
1980 1981 1982 1983 1984 198& 1986 1987 1988 1989
Figure 7-1. National trend hi the composite annual average nitrogen dioxide
concentrations at both National Ah* Monitoring Station sites and all sites
with 95% confidence intervals, 1980 to 1989.
Source U S Environmental Protection Agency (1991a)
illustrate this observation A high-elevation site at Mt Mitchell, NC, provides an example
of low NO2 concentrations in rural areas In 1987, the maximum hourly average
concentration of NO2 was below 0 007 ppm, almost 99 % of the hourly average NO2 values
were below 0 005 ppm (Lefohn, 1989) In addition, electric generating stations do provide a
source of NO2 in both rural and urban areas A summary of the hourly average NO2
concentrations that were measured near selected electric generating stations is provided in
Table 7-5 Although the hourly average NO2 concentrations measured are a function of the
emitter's source strength, as well as other considerations such as meteorological and
geographic conditions, the infrequent occurrence of houily average concentrations of NO2
> 0 10 ppm is evident In general, for the sites listed irt the table, only 5 % of the hourly
average concentrations were >0 05 ppm For rural locations near electric generating plants,
the infrequent occurrence of hourly average concentrations of NO2 >0 05 ppm has been
reported previously in the literature (Lefohn and Tingey, 1984, Lefohn et al, 1987)
7-9
-------
003
0025
g- 0.02
o
1
0.015
0.01
0005
-w
t
...... -o
'
MSA Size
> 1,000,000 > 500,000 > 250,000
—H— ..—o-~ n
i
i
i
I
1980 1981 1982 1983 1984 1985 1986 1987 1988 1989
Figure 7-2. United States metropolitan area trends in the composite annual average
nitrogen dioxide concentration, 1980 to 1989.
Source U S Environmental Protection Agency (1991a)
Table 7-6 lists the highest annual average for 103 MSAs reporting at least one NO2
monitoring station with valid data in 1988, 1989, and 1990 (U S Environmental Protection
Agency, 1990, 1991a, 1991b) The MSAs are listed in alphabetical order Peak annual
averages in these MSAs range from 0 007 to 0 061 ppm Figure 7-3 shows that the
collective mode for the peak annual average in the 1988 to 1989 period was approximately
0.02 ppm. Table 7-7 lists the highest hourly NO2 average concentrations by MSAs reported
across the United States. The highest hourly NO2 average concentrations ranged from
0.040 to 0 540 ppm. Throughout the last 10 years, Los Angeles, CA, was the only urban
area to record violations of the annual NO2 NAAQS of 0 053 ppm (U S Environmental
Protection Agency, 1991b) For the 45 sites monitoring NO2 during the period 1980 to 1989
and experiencing at least 1 year's data capture ^75%, nine sites exceeded the NAAQS at
least 1 year during the monitoring period As expected, this area also experiences the
7-10
-------
TABLE 7-5. CHARACTERIZATION OF HOURLY AVERAGE NITROGEN DIOXIDE CONCENTRATIONS NEAR
SELECTED ELECTRICAL GENERATING PLANTS (concentrations in ppm)
Monitoring Site Generating Plant
Atlanta, GA J McDonough
131210048
Indianapolis, IN Stout
180970057
Indianapolis, IN Stout
180970073
Owensboro, KY Elmer Smith
210590005
Year
1984
1985
1986
1987
1988
1989
1990
1991
1983
1984
1985
1986
1987
1988
1989
1990
1991
1979
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
Mm
0002
0002
0002
0003
0003
0003
0003
0003
0003
0003
0003
0003
0003
0003
0003
0003
0003
0002
0003
0003
0003
0003
0003
0003
0003
0003
0003
0003
10%
0007
0011
0014
0012
0013
0013
0010
0009
0007
0009
0009
0008
0009
0010
0009
0008
0008
0002
0003
0003
0003
0003
0006
0005
0006
0005
0003
0003
30%
0015
0018
0021
0018
0019
0019
0017
0016
0012
0014
0013
0012
0015
0015
0015
0013
0012
0006
0008
0006
0005
0007
0009
0009
0010
0009
0007
0007
50%
0022
0024
0027
0025
0025
0025
0024
0022
0018
0019
0018
0017
0019
0021
0019
0017
0016
0010
0011
0009
0009
0010
0012
0013
0013
0012
0010
0010
Percentiles
70% 90%
0032
0032
0036
0034
0035
0034
0033
0031
0025
0025
0024
0022
0026
0028
0025
0023
0022
0014
0016
0014
0013
0014
0016
0018
0018
0016
0013
0014
0050
0047
0052
0049
0053
0050
0048
0045
0038
0037
0037
0034
0037
0042
0036
0034
0032
0025
0026
0023
0021
0022
0024
0028
0027
0024
0021
0021
95%
0060
0054
0062
0058
0062
0059
0056
0054
0046
0044
0045
0041
0046
0049
0044
0040
0037
0031
0033
0027
0027
0027
0028
0034
0032
0030
0025
0026
99% Max Number Obs
0090
0068
0079
0073
0078
0078
0071
0072
0063
0056
0058
0057
0061
0067
0060
0053
0051
0047
0048
0093
0039
0039
0037
0036
0046
0044
0041
0035
0 129
0 118
0 166
0 125
0 125
0 140
0 111
0 127
0088
0086
0087
0 101
0 118
0093
0088
0072
0078
0080
0093
0067
0060
0068
0052
0081
0074
0090
0059
0057
7,451
8,073
7,997
7,163
8,202
7,876
8,252
8,110
8,036
8,259
7,681
6,730
6,883
8,075
7,772
8,018
7,730
7,396
8,134
8,016
8,007
8,079
8,025
7,787
7,934
8,234
8,355
8,265
Annual Arith Mean
00269
00269
00306
00283
00295
00289
00270
00253
00206
00212
00207
00191
00220
00238
00214
00195
00183
00121
00136
00112
00106
00119
00137
00150
00153
00135
00113
00112
-------
N>
TABLE 7-5 (cont'd). CHARACTERIZATION OF HOURLY AVERAGE NITROGEN DIOXIDE CONCENTRATIONS
NEAR SELECTED ELECTRICAL GENERATING PLANTS (concentrations in ppra)
Monitoring Site Generating Plant Year
Henderson, KY Henderson
211010013
Falcon Heights, MN High Bridge
271230864
St Charles Co , MO Sioux
291831002
Monroeville, PA Cheswick
420030003
1983
1984
1985
1986
1987
1988
1989
1990
1991
1990
1991
1983
1985
1986
1987
1988
1989
1990
1991
1991
Min
0002
0002
0002
0002
0003
0003
0003
0003
0003
0003
0003
0002
0002
0002
0003
0003
0003
0003
0003
0003
10%
0006
0006
0002
0002
0003
0003
0003
0003
0003
0006
0003
0002
0002
0002
0003
0003
0003
0003
0003
0009
30%
0012
0012
0012
0010
0011
0010
0010
0008
0011
0010
0008
0002
0002
0005
0007
0003
0003
0003
0003
0014
Percenhles
50% 70% 90%
0019
0020
0020
0018
0019
0019
0017
0015
0019
0014
0014
0007
0008
0009
0012
0007
0012
0008
0007
0020
0.028
0029
0029
0026
0028
0029
0026
0024
0026
0022
0020
0014
0012
0013
0016
0012
0016
0016
0012
0027
0042
0042
0044
0038
0042
0044
0043
0035
0039
0032
0030
0022
0020
0023
0028
0023
0027
0023
0022
0040
95%
0050
0050
0052
0046
0050
0052
0051
0042
0047
0036
0036
0026
0024
0027
0032
0027
0031
0027
0028
0048
99%
0068
0066
0068
0061
0066
0071
0070
0054
0063
0046
0048
0038
0031
0037
0045
0039
0039
0035
0038
0063
Max Number Obs Annual Anth Mean
0097
0105
0 103
0098
0105
0103
0102
0090
0106
0059
0108
0078
0055
0 105
0078
0086
0063
0078
0077
0 102
8,349
8,330
8,230
8,183
8,354
8,268
8,259
8,215
7,872
8,468
8,249
8,251
7,794
7,401
8,127
7,341
8,096
7,662
8,200
8,389
00220
00225
00223
00199
00214
00219
00204
00178
00205
00168
00157
00103
00092
00107
00134
00099
00122
00106
00098
00228
Source AIRS (1991, 1992)
-------
TABLE 7-6. MAXIMUM ANNUAL AVERAGE NITROGEN DIOXIDE
CONCENTRATIONS REPORTED IN U.S. METROPOLITAN
STATISTICAL AREAS, 1988 TO 1990
Metropolitan Statistical Area
Albuquerque, MM
Allentown-Bethlehem, PA-NJ
Anaheim-Santa Ana, CA
Atlanta, GA
Austin, TX
Bakersfield, CA
Baltimore, MD
Baton Rouge, LA
Beaumont-Port Arthur, TX
Beaver Co , PA
Bergen-Passaic, NJ
Boston, MA
Bndgeport-Milford, CT
Buffalo, NY
Burlington, VT
Charleston, WV
Chicago, IL
Chico, CA
Cincinnati, OH-KY-IN
Cleveland, OH
Dallas, TX
Denver, CO
Detroit, MI
El Paso, TX
Ene, PA
Evansville, IN-KY
Ft Wayne, IN
Ft Worth-Arlington, TX
Fresno, CA
Greensboro et al , NC
Harnsburg-Lebanon-Carhsle, PA
Hartford, CT
Houston, TX
Huntmgton-Ashland, WV-KY-OH
Indianapolis, IN
Jacksonville, FL
Jersey City, NJ
Johnson City et al , TN-VA
Johnstown, PA
Kansas City, MO-KS
Kenosha, WI
Lancaster, PA
Little Rock-N Little Rock, AR
Los Angeles-Long Beach, CA
Louisville, KY-OH
Manchester, NH
Memphis, TN-AR-MS
Miami-Hialeah, FL
Middlesex-Somerset-Hunterdon, NJ
Milwaukee, WI
Minneapolis-St Paul, MN-WI
1988
(ppm)
0018
0020
0046
0030
-
0032
0034
0021
-
0020
0036
0033
0027
0022
0019
0024
0032
0016
0030
0031
0021
0039
0023
0021
0016
0022
0010
0014
0032
0018
0021
0020
0028
0016
0024
0019
0033
-
0019
0014
0014
0020
0010
0061
0023
0024
0034
0017
0025
0027
0020
1989
(ppm)
0019
0020
0047
0029
0017
0033
0035
0019
0007
0020
0035
0032
0026
0024
0019
0021
0034
0016
0030
0034
0021
0040
0026
0022
0015
0020
0011
0013
0032
0016
0022
0020
0028
0013
0023
0015
0031
0019
0019
0015
0016
0018
0009
0057
-
0022
0026
0018
0024
0029
0009
1990
(ppm)
0018
0017
0047
0027
0017
0032
0034
0018
0013
0020
0031
0032
0026
0023
0018
0020
0031
0015
0028
0029
0018
-
0024
0017
0015
0018
0009
0012
0026
0017
0020
0019
0029
0016
0020
0015
0030
0019
0025
0015
0010
0017
0009
0056
0030
-
0023
0016
0022
0024
0022
Metropolitan Statistical Area
Modesto, CA
Nashville, TN
Nassau-Suffolk, NY
New Haven-Menden, CT
New Orleans, LA
New Yoik, NY
Newark, NJ
Norfolk et al , VA
Oakland, CA
Oklahoma City, OK
Orlando, FL
Owensboro, KY
Oxnard-Ventura, CA
Philadelphia, PA-NJ
Pittsburgh, PA
Providence, RI
Provo-Oiem, UT
Raleigh-Durham, NC
Reading, PA
Redding, CA
Richmond-Petersburg, VA
Riversid< -San Bernardino, CA
Roanoke, VA
Sacramento, CA
Sagmaw-Bay City-Midland, MI
St LOUIF, MO-EL
Sahnas-Seaside-Monterey, CA
Salt Lab, City-Ogden, UT
San Diego, CA
San Francisco, CA
San Jose, CA
Santa Barbara et al , CA
Santa Cruz, CA
Santa Rosa-Petaluma, CA
Scranton-Wilkes-Barre, PA
Springfield, MO
Springfield, MA
Steubenville-Weirton, OH-WV
Stockton, CA
Tampa eit al , FL
Tucson, AZ
Tulsa, OK
Vallejo-Fairfield-Napa, CA
Visaha-Tulare-Porterville, CA
Washington, DC-MD-VA
West Palm Beach et al , FL
Wheeling, WV-OH
Wilmington, DE-NJ-MD
Worcesttr, MA
York, PA
1988
(ppm)
0027
0012
0033
0029
0024
0041
0040
0017
0026
0029
-
0015
0018
0039
0030
-
0028
-
0024
0013
0026
0047
0016
0025
-
0025
-
0035
0035
0026
0032
0017
0008
0016
0019
0010
-
0021
0026
0021
0017
0017
0019
0023
0030
0013
0018
0033
0029
0023
1989
(ppm)
0027
0012
0029
0028
0022
0049
0038
0020
0025
0015
0013
0014
0027
0040
0028
0024
0028
0012
0023
0014
0025
0045
0014
0025
0009
0026
0014
0034
0032
0026
0032
0027
0009
0015
0021
0010
0029
0023
0026
0022
0023
0020
0019
0021
0031
0013
0019
0034
0026
0022
1990
(ppm)
0026
0012
0028
0027
0020
0046
0035
0019
0023
0015
0012
0011
0025
0035
0031
0024
0023
0014
0022
-
0023
0041
0013
0024
0008
0026
0012
0029
0029
0022
0030
0022
0008
0015
0020
0008
0026
0,020
0026
0013
0022
0015
0018
0021
0030
0014
-
0033
0022
0022
Source U S Environmental Protection Agency (1990, 1991a, 1991b)
7-13
-------
!•» -
13 -
12 -
11 -
10 -
9 -
8 -
-
6 -
S -
4 -
3 -
2 -
1 _
o -I
nr
Hi
/ P
/ s '
/ /
/ /
/ /
r-jr-j
n
71
/
/
Q n
' ' ' " ' ' ' ' ' ' ' ' ' n Ann
''•''''''•''> ' «jtd
I i
,,,,,,,,/,,/,„ ^. , , _ | ,_ _
x ,,,,,,,,,, , , \ UL , J UU x U 1 U I
' 1 Mr 1 MM h f M r
001
002
003
004
005
006
Peak MSA annual average (ppm), 1988 89
figure 7-3. Distribution of peak annual nitrogen dioxide averages iin 103 Metropolitan
Statistical Areas, 1988 to 1989, as derived by the U.S. Environmental
Protection Agency from AIRS (1991).
highest hourly average concentrations Although the information is limited, Table 7-8
summarizes the maximum hourly average NO concentrations reported by MSAs in the United
States.
The seasonal and diurnal patterns in ambient NO2 concentrations were explored using
data from a majority of urban sites For the analysis, a subset of sites was selected from the
AIRS data base. Four years of continuous data (1986 through 1989) were required, with no
more than a month of incomplete data A month was considered incomplete if less than
75 % of the hourly measurements were reported There were 156 sites that mel these
requirements, and it was observed that the vast majority of sites had at least 90% of the data
present for each month The purpose of these criteria was to provide a general selection of
all types of stations that could yield some information about seasonal and diurnal patterns
There was no attempt to select a geographically or demographically representative set of
7-14
-------
TABLE 7-7. MAXIMUM HOURLY AVERAGE NITROGEN DIOXIDE
CONCENTRATIONS REPORTED IN U.S. METROPOLITAN
STATISTICAL AREAS, 1988 TO 1990
Metropolitan Statistical Area
Albuquerque, NM
Allentown-Bethlehem, PA-NJ
Anaheim-Santa Ana, CA
Atlanta, GA
Austin, TX
Bafcersfield, CA
Baltimore, MD
Baton Rouge, LA
Beaumont-Port Arthur, TX
Beaver Co , PA
Bergen-Passaic, NJ
Boston, MA
Bndgeport-Milford, CT
Buffalo, NY
Burlington, VT
Charleston, WV
Chicago, IL
Chico, CA
Cincinnati, OH-KY-IN
Cleveland, OH
Dallas, TX
Denver, CO
Detroit, MI
El Paso, TX
Erie, PA
Evansville, IN-K.Y
Ft Wayne, IN
Ft Worth-Arlington, TX
Fresno, CA
Greensboro et al , NC
Harnsburg-Lebanon-Carhsle, PA
Hartford, CT
Houston, TX
Huntmgton-Ashland, WV-KY-OH
Indianapolis, IN
Jacksonville, FL
Jersey City, NJ
Johnson City et al , TN-VA
Johnstown, PA
Kansas City, MO-KS
Kenosha, WI
Lancaster, PA
Little Rock-N Little Rock, AR
Los Angeles-Long Beach, CA
Louisville, KY-OH
Manchester, NH
Memphis, TN-AR-MS
Miami-Hialeah, FL
Middlesex-Somerset-Hunterdon, NJ
Milwaukee, WI
Minneapolis-St Paul, MN-WI
1988
(ppm)
0086
0 109
0280
0125
0080
0120
0126
0286
0050
0074
0151
0155
0107
0096
0078
0095
0200
0 100
0223
0138
0110
0205
0215
0 120
0089
0312
0076
0080
0210
0206
0 106
0094
0 190
0141
0096
0097
0212
0146
0077
0079
0058
0105
0092
0540
0 100
0239
0118
0092
0122
0113
0328
1989
(ppm)
0126
0116
0280
0140
0080
0130
0161
0 104
0060
0092
0151
0 178
0 121
0149
0066
0080
0380
0080
0115
0110
0140
0462
0136
0 140
0077
0 102
0095
0090
0 190
0113
0085
0075
0 170
0154
0091
0110
0 174
0125
0062
0090
0082
0082
0082
0340
0088
0 178
0 108
0096
0096
0117
0 106
1990
(ppm)
0 118
0098
0220
0111
0080
0140
0120
0097
0060
0076
0151
0250
0 147
0 107
0073
0067
0150
0080
0 135
0215
0090
-
0110
0 190
0 108
0 118
0060
0080
0 160
0075
0074
0091
0 160
0 120
0072
0086
0 138
0090
0075
0067
0061
0121
0070
0280
0 117
0 115
0 132
0088
0085
0094
0 130
Metropolitan Statistical Area
Modesto, CA
Nashville, TN
Nassau-Suffolk, NY
New Haven-Menden, CT
New Orleans, LA
NewYoik, NY
Newark, NJ
Norfolk et al , VA
Oakland, CA
Oklahoma City, OK
Orlando, FL
Owensboro, KY
Oxnard- Ventura, CA
Philadelphia, PA-NJ
Pittsburgh, PA
Providence, RI
Provo-Oiem, UT
Raleigh-Durham, NC
Reading, PA
Redding, CA
Richmond-Petersburg, VA
Riversidfc-San Bernardino, CA
Roanoke, VA
Sacramento, CA
Sagmaw-Bay City-Midland, MI
St Louis, MO-IL
Salinas-Seaside-Monterey, CA
Salt Lake City-Ogden, VT
San Diego, CA
San Francisco, CA
San Jose, CA '
Santa Bai bara et al , CA
Santa Cruz, CA
Santa Rosa-Petaluma, CA
Scranton Wilkes-Barre, PA
Springfield, MO
Springfield, MA
Steubenville-Weirton, OH-WV
' Stockton, CA
Tampa et al , FL
Tucson, AZ
Tulsa, OK
Vallejo-Fairfield-Napa, CA
Visaha-Tulare-Porterville, CA
Washington, DC-MD-VA
West Palm Beach et al , FL
Wheeling, WV-OH
Wilmington, DE-NJ-MD
Worcester, MA,.
York, PA
1988
(ppm)
0130
0070
0 129
0121
0097
0451
0273
0088
0 140
0318
0066
0074
0 110
0168
0117
0112
0130
-
0118
0100
0141
0210
0119
0 180
0083
0 108
-
0 160
0280
0 130
0 160
0 160
0050
0 120
0126
0075
0203
0 117
0 110
0088
0 125
0088
0090
0 170
0 155
0073
0077
0 110
0250
0 135
1989
(ppm)
0140
0095
0 124
0119
0109
0169
0272
0098
0 150
0125
0065
0090
0 120
0200
0159
0111
0150
0070
0117
0080
0094
0200
0075
0270
0057
0 131
0070
0 163
0230
0230
0150
0 120
0040
0090
0107
0082
0 150
0 167
0 130
0086
0093
0 106
0 130
0210
0 158
0062
0 134
0 130
0122
0 102
1990
(ppm)
0100
0 150
0 114
0122
0086
0213
0213
0083
0 130
0089
0076
0059
0160
0150
0173
0097
0 122
0078
0096
0070
0108
0200
0075
0 160
0044
0098
0060
0 132
0 180
0130
0150
0110
0050
0090
0134
0074
0 104
0 128
0 120
0080
0 116
0090
0080
0 120
0 149
0071
0067
0290
0088
0 129
Source AIRS (1992)
7-15
-------
TABLE 7-8. MAXIMUM HOURLY AVERAGE NITRIC OXIDE CONCENTRATIONS
(ppm) REPORTED IN U.S. METROPOLITAN STATISTICAL AREAS, 1988 TO 1990
Metropolitan Statistical Area
Austin, TX
Baton Rouge, LA
Beaumont-Port Arthur, TX
Boston, MA
Bndgeport-Milford, CT
Charleston, WV
Chicago, EL
Cincinnati, OH-KY-IN
Cleveland, OH
Dallas, TX
Denver, CO
Evansvdle, IN-KY
Ft. Worth-Arlington, TX
Hartford, CT
Houston, TX
Huntington-Ashland, WV-KY-OH
Indianapolis, IN
Kenosha, WI
Milwaukee, WI
Minneapolis-St Paul, MN-WI
Nashville, TN
New Haven-Menden, CT
New Orleans, LA
Norfolk et al , VA
Oklahoma City, OK
Owensboro, KY
Pittsburgh, PA
Raleigh-Durham, NC
Richmond-Petersburg, VA
Roanoke, VA
St Louis, MO-EL
Springfield, MA
Steubenville-Weirton, OH-WV
Washington, DC-MD-VA
Wheeling, WV-OH
Worcester, MA
Hourly Average Values
1988 1989 1990
0338
0671
0560
0305
0868
0500
0475
0760
0755
0466
0290
0438
0850
0236
0464
0298
0396
0463
0659
0488
0438
0 173
0559
0455
0514
0653
0266
0549
0320
0410
0621
0479
0328
0899
0500
0435
0760
0476
0240
0427
0640
0279
0448
0354
0335
0602
0419
0459
0416
0 144
0650
0436
0203
0453
0475
0640
0 159
0694
0400
0394
0 160
0658
0654
0346
1078
0500
0420
0610
0346
0250
0400
0840
0350
0 170
0648
0404
0265
0700
0461
0403
0348
0 170
0703
0425
0451
0248
0466
0512
0611
0648
Annual
1988
00140
00519
00293
00249
00561
00357
00327
00223
00517
00247
00112
00226
00284
00122
00148
00082
0.0196
00182
00457
00217
00089
00071
00305
00327
00145
00482
00150
00288
Average Values
1989 1990
00168
00120
00629
00245
00286
0561
00439
00359
00233
00225
00099
00204
00241
00120
00155
00200
00205
00430
00198
00130
00092
00069
00400
00180
00094
00292
00339
0.0318
00160
00285
00186
00145
00097
00665
00302
00351
00493
00396
00316
00205
00210
00101
00182
00287
00117
00082
00206
00244
00196
00441
00198
00145
00093
00065
00421
00131
00158
00097
00291
00371
00381
00286
Source AIRS (1992)
7-16
-------
sites Of the 156 sites, 70 were located in residential settings, 45 were in an industrial
setting, 20 were commercial, and 20 were miscellaneous, mostly agricultural
Seasonal patterns for the 156 sites were examined by compiling monthly distributions of
hourly values and plotting the 50th, 90th, and 98th percentiles over the period 1986 through
1989 The 50th percentile approximates the geometric mean In reviewing the NO2
monitoring data, a consistent seasonal pattern was distinguishable For most of the sites, the
highest monthly average concentrations occurred during the months of November, December,
January, or February However, there were some exceptions The sample plots in
Figure 7-4a through 7-4d illustrate the diversity of seasonal patterns that exist in the U S
cities The pattern at the Long Beach site shows a broad span of winter months with
elevated values—from about September into March or April The Denver site exhibits much
narrower winter peaks and broader summer troughs The pattern at the Cleveland site is
altogether lower, such peaks as there are occur in the spring and summer months, and are
discernable only in the 90th and 98th percentiles The data for the Richmond site exhibit no
discernable pattern These figures show that seasonal peaks do not occur at the same time
for all sites and, indeed, there are some locations with no prominent seasonal pattern
The diurnal patterns were explored using the monthly average concentrations for each
hour Generally, the highest monthly averages occur in the late afternoon and evening
(1700 to 2200 hours) Because of the interest in assessing the tune of day when the hourly
values above a specified threshold concentration occur, an additional analysis was performed
The hourly average concentration of 0.2 ppm was used as a threshold value because it is
described as a possible benchmark concentration above which human physiologic responses
may be detected In 1988, 216 NO2 monitoring stations reported to EPA's data bank a
"valid" year's data, that is, at least 75% of possible hourly values Figure 7-5 compares the
annual averages with the second-high 1-h values Four California stations reported annual
averages equaling or exceeding the annual standard of 0 053 ppm
The 18 stations reporting a second-high 1-h value greater than 0 2 ppm are identified by
state in Figure 7-5. The diurnal incidences of credible 1-h NO2 values greater than 0 2 ppm
for the other 16 stations are listed in Table 7-9, late morning is when these events are most
likely to occur
7-17
-------
O IS
A- LONG BEACH. CALIFORNIA "S6-'a&
•M CO OE2O) M A M .1 UAAONI
O IS
O
0.2
B. DENVER. C30L.ORADO
O OS
O- CLEVELAND. OHIO *86-'89
•M B7
A M J J A :
Figure 7-4. Monthly 50th, 90th, and 98th percentiles of 1-h nitrogen dioxide
concentrations at selected stations, 1986 to 1989, as derived by the U.S.
Environmental Protection Agency from AIRS (1991).
7-18
-------
OB -
OA —
O.3 -
Q2
01 -
o —
NO2 AT 21 4 STATIONS- 1 988
MN CA §A C
NH CACA c
+ * * £*** * *
* * :*:; :i,+***< *
+ + +i-***||l*^!):lll5* +** +
^^+^|*| ti|** **•*•*
*
CA
A ,
feA
< Annual Std
OO2 OO4
ANNUAL AVERAGE, ppm
OS —
O4 —
O3 -
01 —
o —
NO2 AT 244 STATIONS - 1 989
c
NJ CACA CA
CA CCA CA
CACA
* * * **+ *t ***i ** ***+ -i-
* * ^|jli**!ji*ii**^**^*Hf*
l*:**si**
A
CA
CA
< Annual Std
OO2 OO4
ANNUAL AVERAGE, ppm
NO2 AT 269 STATIONS - 1990
*± *
< Annual Std
OO2 OO4
ANNUAL AVERAGE, ppm
Figure 7-5. Annual average nitrogen dioxide versus second-high 1-h concentration,
1988, 1989, and 1990. (Second-high 1-h values >0.2 are identified by
state, as derived by the U.S. Environmental Protection Agency from AIRS,
1991, 1992.)
7-19
-------
TABLE 7-9. HOURLY INCIDENCE OF NITROGEN DIOXIDE CONCENTRATIONS
GREATER THAN 0.2 ppm FOR STATIONS WITH MORE THAN
ONE OCCURRENCE, 1988
1988 012
Anaheim, CA
Azusa, CA
Burbank, CA
Hawthorne, CA
La Habra, CA
Long Beach, CA
Los Angeles (0113), CA
Los Angeles (1103), CA
Lynwood, CA
Pico Rivera, CA
San Diego, CA
Whittier, CA
Worcester, MA
Minneapolis, MN
Manchester, NH
Bayonne, NJ
Clock Hour
3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
2211
112
12 1133211
12311
1 1
2342 1
2 3
67663 13
1 3 2
1 1
22 11111
1 2
1 1
1 1 1
11 1
1 1
Source AIRS (1991)
The 1-h data from the remaining 16 stations have been examined in detail, hour by
hour, to serve two purposes to place these high values in perspective with Ihe general
distributions of 1-h values, and to show the changing shape of the general distribution of all
1-h values through the 24-h period in response to the cumulative influences of local
emissions and meteorology
Figure 7-6 represents the hour-by-hour percent frequency distribution of 1-h NO2
values for four of the stations selected from the group of 16 in Table 7-9 Ihe San Diego,
CA, location evidently often receives midday ventilation from sea bieezes, causing the
distributions of 1-h values to contract and shift toward lower concentrations Ihrough the early
afternoon Toward sunset, the upper tails of the distributions begin to extend toward higher
concentrations, but the peaks remain about 0 01 ppm After midnight, a subset of higher
7-20
-------
|
I
#>02
n—i—i—i—i—i i T~1—I—i—I—i—i—i—i—i—i—r
000 002 004 006 008 010 012 014 016 018 020
1-h NO2 Concentration (ppm)
#>02
i—i—i—i—i—i—i i i i—i—i i i i i i r
000 002 004 006 008 010 012 014 016 018 020
1-h NO, Concentration (ppm)
#>02
\ i i I
000 002 004 006 008 010 0.12 014 016 018 020
Worchaster Massachusetts, 1988
1800 hours
1200 hours
0600 hours
0000 hours
#>02
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
0
0
0
0
0
0
0
1-h NO2 Concentration (ppm)
000 002 004 006 008 010 012 014 016 018 020
1-h NO 2 Concentration (ppm)
Figure 7-6. Hourly relative frequency distributions of 1-h nitrogen dioxide values at four selected
stations for 1988, with numbers of values greater than 0.2 ppm, as derived by the U.S.
Environmental Protection Agency from AIRS (1991).
7-21
-------
concentrations emerges and, around sunrise, has shifted the distribution peaks to around
0.04 ppm (The 0300- or 0400-hours tune slot is used for calibration checks at California
stations; no data are reported)
Distributions for the Burbank, CA, station, situated along the northern edge of the
valley near the mountains, are notably broader than those for the San Diego station They
exhibit their narrowest spread between midnight and sunrise Morning rush-hour emissions
extend the upper tail and shift the peak higher, briefly Through midday, the peaks shift
somewhat lower, but the distributions remain broad From late afternoon to midnight, the
peaks shift higher and the distributions broaden further
The Bayonne, NJ, and Worcester, MA, stations have similar patterns, their
distributions broaden during morning rush-hour, shift toward lower peaks through early
afternoon when ventilation usually improves, then broaden and shift toward higher
concentrations through the evening hours when winds characteristically subside and inversion
forms.
From this review of NO2 data, it is concluded that 1-h concentrations gieater than
0.2 ppm are infrequent events that lie well above the general distribution of 1-h values
These excursions presumably are produced by the rare coincidence of emissions and
meteorological conditions, a distal rather than a proximal feature on the very attenuated tail
of a station's main data distribution
From the group of 216 stations with valid data for 1988, discussed previously, the
subset of 43 stations with annual averages >0 03 ppm are examined next Twenty-three
stations are located in California, the other 20 are located in 13 other states For this group
of stations with annual averages above this concentration of potential interest, this question is
posed: What is the relationship between the annual average and incidence of 1-h values
above selected thresholds9 In Figure 7-7, the percentages of 1-h values greater than 0 03
ppm and 0.05 ppm are plotted versus the annual averages for these 43 stations For this
subset of stations in the upper portion of the distribution, the percentages of 1-h values
above the chosen thresholds bear a reasonably linear relationship to the stations' annual
averages for the 0 03-ppm group, R2= 0 805, and for the 0 05-ppm group, R2 = 0 924
Complete distributions of a year's 1-h values for four stations are compared in
Figure 7-8. Two stations were selected from the California basin area and two were selected
7-22
-------
1
j=
T-
•8
100%
90% -
80% -
70% -
60% -
50% -
40% -
30% -
20% -
10%
%>003
n n
an
%>005
0025
0035 0045
Annual Average NO2 (ppm)
0055
0065
Figure 7-7. Percent of 1-h nitrogen dioxide values above 0.03 and 0.05 ppm versus
annual averages >0.03 ppm, 1988, as derived by the U.S. Environmental
Protection Agency from AIRS (1991).
from the east coast, with annual averages approximately in the middle of the annual average
range depicted in Figure 7-7 Although the Los Angeles and Baltimore stations have similar
annual averages, the Baltimore station has a higher percentage of values around 0 04 ppm,
and its distribution slides under the Los Angeles distribution at around 0 07 ppm Likewise,
the Anaheim and New York City annual averages are similar, but the New York City station
has a higher percentage of 0 05-ppm values, then drops under the Anaheim distribution at
0 07 ppm
This very limited comparison suggests that, at fixed-site ambient monitoring locations
with annual averages above the national average (see Figure 7-2), percentages of values in
the middle of the 1-h distribution are rather consistenlly related to the annual average
However, as discussed earlier in this chapter, the NO2 elevated concentrations in the
California basin area are not experienced in most of the rest of the country The California
exposures are a result of a unique combination of sources and meteorology
7-23
-------
§
3
35%
30% -
25% -
20% -
-t; 15% -
10% -
Ann X, ppm
X Los Angeles, CA (0113) 0035
A Anaheim, CA (0001) 0046
O Baltimore, MD (0040) 0034
0 New York City, NY (0010) 0041
005 01 015 02
1-h NO2 Concentration (ppm)
O A B -t3—B—H-
025
03
Figure 7-8. Relative distributions of 1-h nitrogen dioxide values at selected stations,
1988, as derived by the U.S. Environmental Protection Agency from AIRS
(1991).
7.2.3.4 Exposure Patterns Observed for Ambient Nitrogen Dioxide and Nitric Oxide
Concentrations—Rural Forest and Agriculture Areas
Because of the interest in assessing the potential effects of NO2 and NO exposures on
vegetation, monitoring data from rural forested and agricultural stations are characterized
Because of the paucity of information, the main focus is on NO2 hourly average
concentration data. To be selected, a station that measured NO2 had to (1) be designated
rural forested or agricultural, (2) have collected data at least during the period 1990 to 1991,
and (3) have experienced data capture of at least 75 % or greater for the hourly average
values over an annual period Eight stations at rural forested sites, which measured hourly
average values of NO2, were selected for analysis Thirty-three site-years of data were used
in the analysis, where a site-year is defined as one year of data for a specific site For
example, for the Perry County station in Pennsylvania, 1983-1991 data were used in the
7-24
-------
analysis, thus, 9 site-years of information were used For the rural agricultural sites, data
from 25 stations were used in the analysis, for a total of 142 site-years of information
For the penod 1979 to 1991, the hourly average NO2 concentrations for selected forest
and agricultural sites were <0 10 ppm in most cases Table 7-10 summarizes the maximum
hourly average concentrations for the eight rural forested sites used in the analysis
An example of the percentile distribution information for the 1-h values, as well as the
annual arithmetic average concentration, for a specific year is provided in Table 7-11
Table 7-12 lists the maximum hourly average concentrations for the 25 rural agricultural
sites, Table 7-13 shows the percentile distributions and annual arithmetic means for some of
these sites Agricultural sites that were not located near major metropolitan areas
experienced few hourly average maximum concentrations > 0 10 ppm
As observed for urban locations, a consistent seasonal pattern for NO2 was
distinguishable for both the rural forested and the rural agricultural sites In general, the
NO2 monthly average values decreased during the spring, were at their lowest levels during
the summer, then rose during the fall and winter The months of November, December,
January, or February contained the highest monthly average concentrations (Figure 7-9)
A consistent NO2 diurnal pattern was observed for the rural forested and agricultural
sites The late afternoon and evening hours (approximately 1700 to 2200 hours) contained
the highest NO2 monthly average concentrations (Figure 7-10) After 2200 hours, the NO2
levels dropped until approximately 0600 hours The levels began to rise for a few hours
until approximately 0900 to 1000 hours After 1000 hours, levels began to drop again and
continued to drop throughout the afternoon The NO2 monthly average concentrations
decreased to their lowest levels between the hours of 1000 and 1700 hours Figures 7-11
and 7-12 illustrate the typical patterns observed for selected rural forested and agricultural
sites Upon reviewing the individual hourly average concentrations, the same pattern became
apparent
For the data set used, the seasonal and diurnal patterns were consistent across all sites,
with the exception of five sites in Montana and two in California Some of the Montana data
sets contained monthly average NO2 values so low that it was difficult to distinguish any
pattern However, the data for some of the site-years showed a slight increase in the average
7-25
-------
TABLE 7-10. MAXIMUM HOURLY AVERAGE NITROGEN DIOXIDE CONCENTRATIONS FOR SELECTED
U.S. RURAL FORESTED SITES (concentrations in ppm)
AIRSroa Site Name 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991
051191002 North Little Rock, AR 0039 0075 0141 0055 0062 0102 0080 0071 0074 0092 0.082 0070 0090
181090004 Morgan Co , IN 0.060
250154002 Ware Co, MA 0095 0064 0082 0091 0086 0081
300870761 Rosebud Co , MT 0 032
300870762 Rosebud Co , MT 0 041
350281002 Los Alamos Co , NM 0 010
420990301 Perry Co , PA 0 160 0 060 0 065 0 051 0 044 0 072 0 055 0 041 0 045
450190046 Mount Pleasant, SC 0031
-------
TABLE 7-11. CHARACTERIZATION OF HOURLY AVERAGE NITROGEN DIOXIDE CONCENTRATIONS
FOR SELECTED U.S. FOREST SITES (concentrations in ppm)
AIRSID*
051191002
181090004
250154002
300870761
300870762
350281002
420990301
450190046
Monitoring Site
N Little Rock, AR
Morgan Co , IN
Ware, MA
Rosebud Co , MT
Rosebud Co , MT
Los Alamos Co , NM
Perry Co , PA
Mount Pleasant, SC
Year
1990
1991
1990
1990
1990
1991
1989
1991
Mm
0003
0003
0003
0003
0003
0003
0003
0002
10%
0003
0003
0003
0003
0003
0003
0003
0002
30%
0005
0006
0003
0003
0003
0003
0003
0002
Percentiles
50% 70% 90%
0007
0007
0005
0003
0003
0003
0005
0002
0011
0010
0009
0003
0003
0003
0008
0002
0020
0015
0022
0003
0003
0005
0015
0005
95%
0025
0019
0031
0003
0003
0005
0020
0010
99%
0038
0026
0048
0010
0010
0007
0031
0010
Max
0070
0060
0086
0032
0041
0010
0055
0031
Number
Obs
8,240
7,239
7,848
8,359
8,299
8,677
7,068
6,147
Annual
Arith Mean
00094
00086
00090
00027
00027
00030
00069
00027
o aAIRS ID = Aerometric Information Retrieval System identification number
Source AIRS (1992)
-------
TABLE 7-12. MAXIMUM HOURLY AVERAGE NITROGEN DIOXIDE CONCENTRATIONS FOR SELECT
U.S. RURAL AGRICULTURAL SITES (concentrations in ppm)
V
Affisro*
060295001
061110005
061113001
080013001
181470002
181470006
181550001
260430901
260430902
270176316
271710007
290470005
291831002
300870701
300870702
300870704
340273001
350450014
360010012
380130001
380650002
401430174
471190106
540250001
550210008
Site Name
Kern Co , CA
Ventura Co , CA
Oxnard, CA
Welby, CO
Spencer Co , IN
Spencer Co , IN
Switzerland, IN
Dickinson Co , MI
Dickinson Co , MI
Carlton Co , MN
Wright Co , MN
Clay Co , MO
St Charles Co , MO
Rosebud Co , MT
Rosebud Co , MT
Rosebud Co , MT
Morns Co , NJ
San Juan Co , NM
Albany Co , NY
Burke Co , ND
Oliver Co , ND
Tulsa Co , OK
Maury Co , TN
Greenbner Co , WV
Columbia Co., WI
1979 1980 1981 1982 1983 1984 1985
0060 0
0 110
0 224 0 190 0 300 0 203 0
0054 0059 0049 0042 0046 0
0065 0049 0089 0040 0046 0
0
0
0 197 0 081 0 120 0
0078 0
0
0
0
0 108 0
0062 0080 0060 0
0047 0048 0046 0060 0037 0
050
305
049
038
044
037
057
055
054
042
045
118
111
047
1986
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
080
130
199
047
045
046
050
080
105
038
033
038
098
079
083
047
1987
0040
0080
0217
0357
0239
0039
0063
0 156
0078
0039
0030
0029
0081
0059
0052
1988
0080
0110
0205
0076
0118
0045
0035
0079
0086
0040
0033
0041
0 103
0076
0037
0040
1989
0060
0110
0049
0039
0035
0075
0076
0063
0044
0 142
0054
0076
0070
0017
0037
0090
0038
1990
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
070
050
100
035
045
049
035
052
061
054
078
035
039
045
094
063
192
026
053
041
038
1991
0048
0 185
0045
0042
0034
0072
0077
0 113
0029
0054
0082
0071
0083
0044
0044
aAIRS ID = Aerometac Information Retrieval System identification number
Source AIRS (1992)
-------
TABLE 7-13. CHARACTERIZATION OF HOURLY AVERAGE NITROGEN DIOXIDE CONCENTRATIONS
FOR SELECTED U.S. AGRICULTURAL SITES (concentrations in ppm)
AERSIDa
060295001
061110005
061113001
080013001
181470002
181470006
181550001
260430901
260430902
270176316
271710007
<, 290470005
^ 291831002
300870701
300870702
300870704
340273001
350450014
360010012
380130001
380650002
401430174
471190106
540250001
550210008
Monitoring Site
Kern Co , CA
Ventura CO , CA
Oxnard, CA
Welby, CO
Spencer CO , IN
Spencer CO , IN
Switzerland, IN
Dickinson Co , MI
Dickinson Co , MI
Carlton Co , MN
Wright Co , MN
Clay Co , MO
St Charles Co , MO
Rosebud Co , MT
Rosebud Co , MT
Rosebud Co , MT
Morns Co , NJ
San Juan Co , NM
Albany Co , NY
Burke Co , ND
Oliver Co , ND
Tulsa Co , OK
Maury Co , TN
Greenbner Co , WV
Columbia Co , WI
Year
1990
1989
1989
1987
1985
1985
1991
1990
1990
1990
1989
1990
1990
1988
1988
1988
1990
1990
1990
1990
1989
1989
1990
1989
1988
Mm
0003
0003
0003
0003
0003
0003
0003
0003
0003
0003
0003
0003
0003
0003
0003
0003
0003
0003
0003
0003
0003
0003
0003
0003
0002
10% 30%
0 010 0 010
0003 0010
0 010 0 010
0003 0010
0003 0006
0005 0007
0006 0008
0003 0003
0003 0003
0003 0003
0003 0003
0003 0003
0003 0003
0003 0003
0003 0003
0003 0003
0003 0007
0003 0003
0 005 0 009
0003 0003
0003 0003
0003 0003
0003 0003
0003 0003
0 002 0 002
Percentiles
50% 70% 90%
0010
0010
0020
0021
0008
0009
0011
0003
0005
0003
0006
0007
0008
0003
0003
0003
0011
0003
0015
0003
0003
0003
0005
0003
0002
0 010 0 020
0 010 0 010
0 020 0 030
0 035 0 053
0011 0016
0 012 0 017
0 014 0 019
0008 0016
0006 0010
0005 0009
0010 0019
0011 0020
0016 0023
0003 0011
0003 0005
0 005 0 006
0 016 0 029
0 008 0 018
0 024 0 036
0003 0003
0003 0003
0 009 0 019
0 008 0 012
0003 0007
0006 0013
95%
0030
0020
0040
0063
0020
0021
0023
0021
0013
0012
0025
0027
0027
0016
0008
0009
0035
0025
0041
0003
0005
0024
0015
0009
0017
99% Max Number Obs Annual Arith Mean
0030
0020
0050
0097
0026
0027
0029
0030
0020
0020
0036
0039
0035
0027
0016
0017
0047
0041
0054
0005
0013
0038
0023
0018
0027
0070
0060
0 110
0217
0049
0038
0045
0049
0035
0052
0075
0054
0078
0040
0033
0041
0094
0063
0 192
0026
0037
0090
0041
0038
0040
7,928
7,614
7,811
8,136
7,429
8,191
7,922
8,651
8,491
7,417
7,917
7,981
7,662
7,093
6,888
8,003
8,260
8,548
8,499
8,361
8,701
8,259
7,271
7,501
7,963
00134
00086
00180
00257
00091
00103
00117
00070
00054
00044
00086
00088
00106
00048
00033
00034
00138
00076
00182
00026
00029
00077
00064
00037
00056
AIRS ID = Aerometric Information Retrieval System identification number
Source AIRS (1992)
-------
Jan. Feb Mar. Apr May June July Aug Sept Oct Nov Dec
Month
Figure 7-9. Seasonal pattern for nitrogen dioxide concentrations at rural and forested
Aerometric Information Retrieval System monitoring sites.
Source: AIRS (1992)
values during the summer months There was no clear or consistent pattern in the California
data sets
Studies of the joint occurrence of gaseous NO2/sulfur dioxide (SO2) and NO2/ozone
(O3) have concluded that (1) the co-occurrence of two-pollutant mixtures lasted only a few
hours per episode, and (2) the time between episodes is generally large (i e , weeks,
sometimes months) (Lefohn and Tingey, 1984, Lane and Bell, 1984, Jacobson and
McManus, 1985, Lefohn et al, 1987) Lefohn et al (1987), using hourly averaged data
collected at rural sites, reported that the periods of co-occurrence represent a small portion of
the potential plant growing period For human ambient exposure considerations, Lefohn and
Tingey (1984) noted that, in most cases, the simultaneous co-occurrence of NO2/O3 was
infrequent. However, for several sites located in the southern California South Coast Air
Basin, the authors reported more than 450 simultaneous co-occurrences of each pollutant at
7-30
-------
12345678 9101112131415161718192021222324
Hour
Figure 7-10. Diurnal pattern for nitrogen dioxide at rural and forested Aerometric
Information Retrieval System monitoring sites.
Source AIRS (1992)
Morgan Co, IN
Ware, MA
N Little Rock, AR
Peny Co, PA
11 13 15 17 19 21
Hour
23
Figure 7-11. Diurnal patterns for nitrogen dioxide monthly average concentrations at
selected rural forested Aerometric Information Retrieval System
monitoring sites.
Source AIRS (1992)
7-31
-------
Spencer Co, IN
Clay Co, MO
Rosebud Co, MT
Morris Co, NJ
1
Figure 7-12. Diurnal patterns for nitrogen dioxide monthly average concentrations at
selected rural agricultural Aerometric Information Retrieval System
monitoring sites.
Source AIRS (1992)
hourly average concentrations equal to or greater than 0 05 ppm For Denver, CO, and San
Jose, CA, more than 100 co-occurrences were reported
Table 7-14 summarizes the maximum hourly average NO concentrations reported in
rural areas for the period 1988 to 1990
7.3 INDOOR AIR CONCENTRATIONS OF NITROGEN OXIDES
7.3.1 Background
Exposures to air contaminants occur across a number of microenvironments (residences,
industrial and nomndustnal workplaces, community air, automobiles, public access buildings,
etc.) in which people spend then- tune A major portion of their time is spent in the
residential indoor environment (Szalai, 1972) A microenvironment is a three-dimensional
7-32
-------
TABLE 7-14. MAXIMUM HOURLY AVERAGE NITRIC OXIDE
CONCENTRATIONS (ppm) REPORTED IN RURAL AREAS, 1988 TO 1990
Hourly Average Values
1988 1989 1990
North Little Rock, AR
Santa Maria, CA
Jalama, CA
Lompoc, CA
Los Padres NF, CA
Gaviota, CA
Isla Vista, CA
Vandenberg AFB, CA
Adams Co , CO
Ware, MA
Clay Co , MO
St Charles Co , MO
St Louis Co , MO
Morns Co , NJ
Burke Co , ND
Dunn Co , ND
Mercer Co , ND
Oliver Co , ND
Maury Co , TN
Greenbner Co , TN
Ft Winnebago, WI
0 138
0574
0064
0506
0370
0210
0 148
0066
0058
0265
0027
0 128
0067
0428
0348
0269
0 108
0013
0438
0061
0053
0041
0 163
0 143
0015
0005
0005
0053
0220
0014
0059
0375
0444
0225
0 140
0017
0083
0085
-
Annual Average Values
1988 1989 1990
00062
00346
00060
00130
00137
00092
00070
00051
00052
00052
00050
00058
00055
00071
00158
00097
00077
00050
00056
00051
00050
00052
00056
00065
00050
00050
00050
00059
00094
00051
00061
00088
00093
00089
00066
00050
00052
00064
Source AIRS (1992)
space having a volume such that the pollutant concentration during some specific tune
interval is considered to be spatially defined, and usually is assumed to be spatially uniform
(National Research Council, 1981) Ideally, the total exposure to a given air contaminant or
category of air contaminants should be assessed over all microenvironments in evaluating
adverse health or comfort effects and in formulating cost-effective mitigation efforts to
reduce or minimize the risks associated with exposure
The indoor residential environment is particularly important in assessing total air
contaminant exposure because (1) individuals spend the major portion of their time indoors
(Szalai, 1972), (2) the highest concentrations of several important air contaminants occur
indoors (National Research Council, 1981), and (3) the most susceptible segments of the
population (the old, the young, and the infirm) are indoors for long periods of time
In addition, weathenzation programs and the use of supplemental space heaters may have
7-33
-------
resulted in increased indoor levels of potentially hazardous air contaminants, further
highlighting the relative importance of the nomndustrial indoor settings
Nitrogen oxides are introduced to indoor environments through emissions from a
variety of combustion sources and through the infiltration or ventilation of air from outdoors
The resulting indoor concentration, both long- and short-term averages (averaging tunes of
seconds to weeks), is dependent on a complex interaction of several interrelated factors
affecting the introduction, dispersion, and removal of NOX These factors include, for
example, such variables as (1) the type, nature (factors affecting the generating rate of NO2),
and number of sources, (2) source use characteristics, (3) building characteristics,
(4) infiltration or ventilation rates, (5) air mixing between and within compartments in an
indoor space; (6) removal rates and potential remission or generation by indoor surfaces and
chemical transformations, (7) existence and effectiveness of air contaminant removal systems,
and (8) outdoor concentrations
The above factors interact to produce a range of indoor concentrations of NO2 The
variability of NO2 levels in residences is demonstrated in Figure 7-13 (Drye et al, 1989)
This figure summarizes data collected in five areas located in four distinct geographical
regions in the United States (Boston and Watertown, MA, Southern California, Portage, WI,
and St. Louis, MO) during the winter for 978 residences Nitrogen dioxide levels were
measured for 1-week periods using Palmes passive samplers (Palmes et al, 1976) in three
locations (outdoors, kitchen, and bedroom) over a summer (not shown here) and a winter
period The relationship between the means by sampling location is similai across all sites,
with kitchen means higher than bedroom means and ambient means lower than bedroom
means. The exception to this is Southern California, where the bedroom mean is slightly
lower than the ambient mean Indoor levels of NO2, particularly in the kitchen, exhibited
considerably more variability than outdoor levels, except in Southern California The data
collected for the summer period exhibited the same trends as shown for the winter data in
Figure 7-13. The data in Figure 7-13 highlight not only the variability in indoor NO2
concentrations, but also the importance of considering residential NO2 concentrations in
assessing exposures. The interaction of the factors to produce the resulting indoor
concentrations shown in Figure 7-13 is usually considered within the framework of the mass-
balance principle.
7-34
-------
0150-
0 140 -
0130-
0120-
0110-
0100-
0090-
0080-
0070-
0060-
0050-
0040 _
0030-
0020-
0010_
o-
i
!
1
I
•
1
*
•
I
"T
\
^
V
3
1
.
T
r
r
Prri |
[~j 1
•
i L
i
) +?
1
1
I
LEGEND
Percent) tes
. 95th
• mean
H SO*
Lj-1 2Sft
1 5th
•
AMB BR KIT AMB BR KIT AMB BR KIT AMB BFt KIT AMB BR KIT
Boston So Cal Portage St Louis Watertown
Figure 7-13. Winter nitrogen dioxide concentrations by site and sampling location.
Abbreviations
AMB = Outdoor
BR = Bedroom
KIT = Kitchen
Source Drye et al (1989)
In its simplest form, where equilibrium conditions are assumed for a single
compartment with complete mixing and no air cleaner, Ihe mass-balance model can be
represented by the following equation
where
A+K
t = Cl + C2,
= Outdoor Air Contnbution,
C, =
S/V
A+K
= Indoor Source Contnbution,
(7-1)
(7-2)
(7-3)
7-35
-------
*5
and where Ci = steady-state indoor concentration of NO2 (jug/m )
Cj = contnbution to indoor NO2 from outdoor air (j»g/m )
C2 = contnbution to indoor NO2 from indoor sources (jug/m )
P = fraction of outdoor NO2 that penetrates the building shell
A = air exchange rate in air changes per hour— ACH (h"1)
C0 = outdoor NO2
K = removal rate of NO2 by indoor chemicals
transformations— equivalent ACH (h" )
S = generation rate or source strength of NO2 (/tcg/h)
V = volume of the indoor space (m )
This simplified form of the model could be used to evaluate NO2 levels indoors
In actuality, however, indoor spaces are often multicompartments with incomplete mixing
where the source generation and contaminant removal rates and air contaminant
concentrations vary considerably in time Equation 7-1 is particularly useful for determining
the impact on indoor air contaminant concentrations from sources that are used over
relatively long periods of tune (e g , unvented kerosene or gas space heaters) where steady-
state or equilibrium conditions are approximated (on the order of hours) When applied to
sources that are intermittent in their use (e g , gas range or tobacco combustion),
Equation 7-1 averages over the off/on periods of the sources to determine average input
parameters for the model Short-term indoor concentrations (on the order of seconds or
minutes) of air contaminants associated with sources whose use vanes considerably with time
can be modeled with the differential version of Equation 7-1 (National Research Council,
1981) when detailed information is available on the time variability of the source use,
mixing, and removal terms Field data on short-term variability of contaminant
concentrations and associated variables regulating the introduction, dispersal, and removal of
NO2 exist for small numbers of "test-house" studies where test conditions were partially
controlled, but have not typically been collected for homes under normal occupancy use
conditions The test-house data, nevertheless, do provide insights into the dynamics of the
generation, dispersal, and removal of contaminants in homes
Outdoor levels of NO2 play an important role in determining indoor levels The
available ambient NO2 data recorded by stationary monitors are reviewed in Section 7 2
Penetration of NO2 through the building envelope (F) and removal by mdoor surfaces or
chemical transformations (K) are related to its chemical reactivity and to the building
7-36
-------
construction materials and furnishings and other factors No direct measurements of the
penetration factor for NO2 through building shells are available The removal rate, K, can
vary from less than 0 1 to longer than 2 0 h"1 The factors influencing the removal rate of
NO2 are discussed in Section 738 Infiltration rates (A) and house volumes (V) can vary by
an order of magnitude or more Infiltration rates from 0 1 to 2 0 h" and house volumes
from 100 to more than 700 m3 are within the range encountered (Gnmsrud et al, 1982, Grot
and Clark, 1979, Billick, 1991, Koutrakis et al, 1992) Factors impacting the source
strengths for NO2 emissions from indoor source are reviewed m Chapter 4
Section 7 3 summarizes the available data on the levels of NO2 indoors, Cv largely
within the framework of the simplified mass-balance model shown in Equation 7-1 Nitrogen
dioxide concentrations measured indoors in homes with no known sources, Q, are compared
to outdoor levels as a function of season of the year, housing type, and region Data on
average and peak NO2 levels indoors as a function of individual and combmations of indoor
sources, C2, are reviewed with an emphasis on attempting to approximate the average
contribution to indoor levels as a function of source type, housing type, and region Data on
the spatial distribution of NO2 between and within rooms indoors as a function of source type
are reviewed, as are recent data on the removal of NO2 by indoor surfaces A gaseous
product of NO2 reactivity, HONO, has recently been identified as a potentially important
indoor air contaminant Section 7 4 reviews the recent data on indoor levels of HONO and
the variables related to its formation Available results on efforts to model (both empirical
and physical/chemical models) indoor NO2 levels are presented and discussed, as are efforts
to use chamber-generated emission factors for major indoor sources to predict levels in
homes under actual use conditions The impact of infiltration rates, air exchange rates
between rooms, and house volumes are not addressed
Nitrogen dioxide is the major NOX species consideied m this chapter because a
considerable amount of indoor sampling data exists for il and exposure to it is of health
importance Indoor concentrations of HONO (Section 7 4) are also considered because of
speculation that it may be of health interest The residential indoor environment has been the
major indoor microenvironment for which NO2 levels have been measured There are few
data available on NO2 levels m other indoor microenviromments Section 7 3 focuses
primarily on findings of major field studies that have evaluated NO2 levels in residential
7-37
-------
indoor microenvironments in order to approximate the range of concentrations indoors
associated with indoor sources Data collected in test-house studies are drawn upon to
highlight various aspects of the dispersal and removal of NO2 indoors
7.3.2 Residences Without Indoor Sources
In the absence of any indoor source of NO2, indoor NO2 concentrations are a function
of the building envelope penetration factor (P), the air exchange rate (A), the reactivity rate
(K), and the outdoor concentration (C0) This condition is represented in Equation 7-1 for
steady-state conditions when S/V is set to zero as
(7-4>
There are no chamber or field studies that have measured the penetration factoi (P) for NO2
or field studies that have separate measures of both ventilation and the reactivity rate for
NO2. Indirect estimates of K from field ventilation data have been made (Wilson et al ,
1986). Limited field data for ventilation rates in residences exist (e g , Gnmsrud et al ,
1982; Grot and Clark, 1979, Wilson et al , 1986, Sheldon et al , 1989, New York State
Department of Health, 1989, Koutrakis et al , 1992) There have, however, been several
field studies that have investigated levels of NO2 in residences As part of these study
designs, indoor and outdoor levels of NO2 were monitored in subsamples of homes that had
no known indoor sources of NO2 The indoor/outdoor ratios of NO2 measured in these
studies provide general information on the role of the penetration factor, the air exchange
rate, and the reactivity rate (PA/[A+K\) in impacting indoor NO2 concentrations in
residences without known NO2 sources
Table 7-15 presents the average outdoor NO2 concentrations measured in several large
field studies and the corresponding average NO2 indoor/outdoor ratios by location in the
residences without indoor NO2 sources The table also presents a breakdown (when
available) of the data by geographical location, housing type, and season of the year
The average indoor/outdoor ratios for all the studies in Table 7-15 are less than 1, as
would be predicted by Equation 7-2 when indoor sources are not present The exceptions are
the winter Kingston, TN, ratio for the kitchen and bedroom and the kitchen summer ratio for
7-38
-------
TABLE 7-15. AVERAGE OUTDOOR CONCENTRATIONS OF NITROGEN DIOXIDE AND AVERAGE
INDOOR/OUTDOOR RATIOS IN HOMES WITHOUT KNOWN INDOOR SOURCES FROM
FIELD STUDIES OF PRIVATE RESIDENCES3
Location
Southern
California
New Haven,
CT
Albuquerque,
NM
California
Portage,
Tucson,
AZ
Boston,
MA
Northern
Central Texas
Suffolk County,
NY
Onondago County,
NY
Portage,
WI
Watertown, MA
Housing Type
Mixed
Single-family
unattached
Mixed
Mobile homes
Mixed
Mixed
Mixed
Single-family
unattached
Single-family
unattached
Single-family
unattached
Single-family
unattached
Not given
Averaging
Time Seasons
7 days
14 days
14 days
7 days
7 days
14 days
14 days
5 days
7 days
7 days
7 days
3-4 days
Summer
Spring
Winter
Winter
Winter 1
Wmter 2
Summer
Winter
Summer
Winter
Summer
Spring/Fall
Winter
Summer
Fall
Winter/Spring
Wmter
Wmter
Wmter
Average over
all seasons
November
December
Average NC>2
Number of Outdoors
Homes (ppm)
70
100
69
60
60
56
46
23
47
47
56
41
23
117
117
124
9
49
66
25
18
10
00381
00231
00483
00070
00141
00195
00137
00236
00081
00091
00106
00136
00195
00168
00200
00178
00285
00188
00115
00068
00196
00244
Indoor/Outdoor Ratios
Kitchen
080
072
056
056
0
061
027
091
065
086
071
064
076
043
053
047
070
065
065
039
Bedroom
075
060
047
055
050
032
054
026
072
045
076
055
052
075
040
047
—
051
051
030
Reference
Wilson et al
(1986)
Leaderer et al
(1986a)
Marbury et al
(1988)
Petreas et al
(1988)
Quackenboss
etal (1986)
Quackenboss
etal (1987)
Ryan et al
(1988)
Koontz et al
(1986)
Research Triangle
Institute (1990)
Spengler et al
(1983)
Clausing et al
(1984)
-------
TABLE 7-15 (cont'd). AVERAGE OUTDOOR CONCENTRATIONS OF NITROGEN DIOXIDE AND AVERAGE
INDOOR/OUTDOOR RATIOS IN HOMES WITHOUT KNOWN INDOOR SOURCES FROM
FIELD STUDIES OF PRIVATE RESIDENCES3
Location
Middlesbrough
Middlesbrough
Los Angeles
Portage, WI
Kingston, TN
Steubenville, OH
Topeka, KS
Watertown, MA
St Louis, MO
Housing Type
Not given
Not given
Mixed
Mixed
Mixed
Mixed
Mixed
Mixed
Mixed
Averaging
Time Seasons
7 days
7 days
2 days
7 days
7 days
7 days
7 days
7 days
7 days
Winter
Winter
Winter
Not Winter
Summer
Winter
Summer
Winter
Summer
Winter
Summer
Winter
Summer
Winter
Summer
Winter
Number of
Homes
87
15
47
124
176
205
291
306
53
71
171
208
70
77
176
202
Average NO2
Outdoors
(ppm)
00186
00184
00442
00305
00069
00085
00093
00121
00256
00212
00081
00109
00212
00204
00140
00167
Indoor/Outdoor Ratios
Kitchen
097
—
__
091
073
07
1 15
074
066
10
081
077
052
092
084
Bedroom
075
075
051
057
082
065
063
101
068
058
088
079
071
040
083
079
Reference
Goldstein et al
(1979)
Melia et al
(1982)
Spengleretal (1992a,b)
Butler et al
(1990)
aNO2 = Nitrogen dioxide
Mixed = Single-family attached, single-family unattached, condo:
munium apartment
-------
Topeka, KS The high ratios in these locations suggest that significant indoor sources existed
in some of the houses thought not to have sources Exceptionally high standard deviations
were reported in indoor concentrations for these studies (Butler et al, 1990) relative to the
outdoor levels observed
The indoor/outdoor ratios reported for the studies in Table 7-15 show general trends
related to season and location in the residence Average ratios are highest in the summer and
lowest in the winter, with the ratios in the spring and fall period falling between the winter
and summer values The highest ratios are found for the kitchen and lowest for the
bedroom, with the living room values in between There are not enough data to determine if
the average ratios exhibit any differences as a function of housing type or geographical
location, although these studies were conducted in widely different climate regions with
different housing and demographic characteristics
The ratios listed in Table 7-15 are calculated from the average indoor and outdoor
concentrations reported for each study listed. These studies typically did not report the
standard deviations or standard errors of the average indoor/outdoor ratios from which the
above observations are drawn, thus not allowing a test for statistical significance of the
trends The distributions of the ratios by season and location in the residence were also not
reported Some studies have reported on the distribution of the indoor/outdoor ratios for
residences without sources
Leaderer et al (1986a) reported an overall house NO2 indoor/outdoor ratio of
0 58 ± 0 31 (n = 123) for the winter sample period in the New Haven, CT, area,
demonstrating the variability of the ratio The distributions of 2-week indoor to ambient
NO2 ratios for homes without known indoor sources in three U S cities (Portage, WI,
Krngston/Harriman, TN, and Steubenville, OH) were recently reported (Spengler et al,
1992b) A box plot of the data is presented in Figure 7-14 The figure demonstrates the
lack of normality in the ratios and considerable variability in the ratios The variability
could be due to several factors, including failure to account for existence of indoor sources
(attached garages, backdrafting and faulty or disconnected flues, cigarette smoking, unvented
combustion sources, etc ), variations in infiltration rates, and differences in removal rates
related to ulterior furnishings
7-41
-------
1.4 -
12 -
10 -
08 -
06-
04 -
02 -
0.0
+
T
1
+
+
O
+
i
(-L-,
I I
Percentage
—Maximum
— 95th
75th
mean
—50%
Uj&
- 5th
I — minimum
+
T
7
O
+
M
38 42
S W
Portage
99
S
96
W
Kingston
Location
19 21
S W
Steubenville
Figure 7-14. Ratio of average indoor nitrogen dioxide to ambient nitrogen dioxide
concentrations by season and location in homes without a nitrogen dioxide
source.
Abbreviations
S — Summer
W ^ Winter
n = Number of homes observed
Source. Spengleretal (1992a)
Distributions of indoor and outdoor concentrations by season have been reported for
two large field studies These data can be evaluated for indoor and outdoor concentrations
for residences without known indoor NO2 sources The cumulative frequency distributions
of concentrations of NO2 for residences without known indoor sources by location in the
residence and for outdoors for two different geographic areas (Southern California and New
Haven, CT) during the winter season are shown in Figures 7-15 and 7-16 A similar
7-42
-------
100
I 80
£ 60
JANUARY 1985
- SOUTHERN CALIFORNIA
- GAS STUDY
ELECTRIC RANGE/
NO GAS APPLIANCES
5
40
D BEDROOMS
A KITCHEN
• OUTDOORS
JL
20 40 60 80
NO2 Concentration Gug/rn3)
100
120
Figure 7-15. Cumulative frequency distribution of nitrogen dioxide concentrations
(1-week sampling period) by location for homes with no known gas
appliances for a winter period in Southern California.
Source Wilson et al (1986)
100
80-
o>
a.
| 40
O
20
(n)
• OUTDOORS 132 (144)
A BEDROOMS 73 (146)
• KITCHEN 76 (147)
O LIVING ROOMS 73 (146)
10 15 20 25
NO2 Concentration (ng/m3)
30
35
Figure 7-16. Cumulative frequency distribution and arithmetic means of nitrogen
dioxide concentrations (2-week sampling period) by location for homes
with no kerosene heater and no gas range for a whiter period in the
New Haven, CT, area.
Source Leaderer et al (1984)
7-43
-------
cumulative frequency distribution for the summer period for homes monitored in the
Southern California Gas Company study is shown in Figure 7-17 The winter distributions
for the two studies (Figures 7-15 and 7-16) are similar despite the large differences in
outdoor concentrations Figures 7-15 and 7-16 highlight the substantial differences in indoor
and outdoor concentrations during the winter period Figures 7-15 and 7-17, for the same
population of homes in the Southern California Gas Company study, highlight the differences
in the distributions of NO2 concentrations indoors relative to the outdoor levels as a function
of season (winter versus summer) The distributions are different, with indoor levels much
closer in concentration to outdoor levels in the summer period
loop
I
s.
-------
o c
O O
§1
o c
•o o
.EO
o w
io9
cc^
1 0
09
08
07
06
05
04
03
02
01
• .••*•
* •
r !•:• .*•'"•• .*
• • • * •
• • * **• * ***• "t.
-• .-••** * •.:"••*..'
• • • •
"• c *
.
I I I I I I I I I I I
JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC
Month
Figure 7-18. Indoor/outdoor nitrogen dioxide concentration ratios (2-week sampling
periods) as a function of time for three homes in the United Kingdom
without indoor nitrogen dioxide sources.
Source Atkins and Law (1987)
seasonal variability in the airtightness of the residences. In the winter months, outside doors,
windows, and other openings are closed and air entering the residences infiltrates through the
budding envelope, with more effective removal of outdoor NO2 With lower air exchange
rates in the winter, the A/(A+K) term in Equation 7-2 is smaller, this term approaches 1 in
summer conditions with higher air exchange rates Duiing the summer period, windows and
doors are more typically open, minimizing removal by the building envelope
No information is available on the impact on the summer ratio for homes with air
conditioning Seasonal differences in the ratio and variability in the ratio within a season
from residence to residence are also hkely due to variations in the penetration factor and
reactivity rate for NO2 Variations in the penetration factor and reactivity rate, like
ventilation, can have a substantial impact on the indoor/outdoor ratios of NO2 There is,
however, little information on the variability of these two additional factors
Indoor concentrations of NO2 in residences without known indoor sources of NO2 are
typically dominated by outdoor levels Indoor levels are usually below outdoor levels in
such homes, thus providing some degree of protection irom outdoor concentrations. The
7-45
-------
indoor/outdoor NO2 ratios in these homes are typically lower in the winter than in the
summer and lower in the bedroom than in the kitchen The ratios by season and location in
a residence show considerable variability The factors impacting this variability are not well
characterized.
7.3.3 Residences with Gas Appliances
In the 1980 census, it was estimated that gas (natural gas and liquid propane) was used
for cooking, heating water, or drying clothes in approximately 45 1 % of all homes in the
United States (U S Bureau of the Census, 1982) The 1990 U S census asked the more
general question of what fuel is used most for heating without the specific inquiry for fuel
used for water heating and for cooking Thus the results for 1990 of 56 7% using natural
gas or liquid propane may not be directly comparable to the earlier result (U S Bureau of
the Census, 1989, 1992). In some other countries (e g , the Netherlands), nearly 100% of
the homes may have gas appliances Unvented, partially vented, and improperly vented gas
appliances, particularly the gas cooking range and oven, represent an important source
category of NO2 emissions into the indoor residential environment Emissions of NO2 from
these gas appliances (the source term, S, in Equation 7-1) are a function of a number of
variables related to source type (range top or oven, water heater, dryer, number of pilot
lights, burner design, etc.), source condition (age, maintenance, combustion efficiency, etc.),
source use (number of burners used, frequency of use, fuel consumption rate, length of use,
improper use, etc), and venting of emissions (existence and use of outside vents over
ranges, efficiency of vents, venting of gas dryers, etc ) The available data on emissions
rates from gas appliances and other indoor sources are presented in Section 4 3
The factors that affect NO2 emissions from gas appliances into residences in
combination with the residence factors (house volume, number of rooms, infiltration rate,
room and whole-house mixing rate, pollutant decay rate, etc ) result in indoor residential
NO2 concentrations associated with gas appliance use The contribution of NO2 emissions
from gas appliances to indoor concentrations in simple terms can be represented by the
source contribution term of Equation 7-1
7-46
-------
C2 - —S/Y— _ indoor Source Contribution (7-3)
A + K
This assumes complete mixing in and between rooms and representation of a highly
time-varying source as an equivalent constant source When the contribution of the outdoor
NO2 levels is added (Equation 7-2), the resultant indoor concentration is determined
(Equation 7-1)
7.3.3.1 Average Indoor Concentrations and Estimated Source Contributions
The presence and use of a gas range in a residence result in higher indoor levels of
NO2 than in homes with electric cooking ranges This is clearly demonstrated in data
collected as part of a study of respiratory illnesses in infants and NO2 exposure (Samet et al,
1992) In this study, NO2 concentrations were obtained over 2-week periods outdoors and in
infants' bedrooms using Palmes passive samplers for a sample of approximately 700 homes
with and approximately 315 homes without gas appliances The summary results averaged
across seasons are shown in Figure 7-19 Nitrogen dioxide concentrations in homes with gas
stoves are higher than those found outdoors and considerably higher than levels found in
homes with electric stoves Levels in homes with gas stoves are higher in the kitchen than in
the bedroom Bedroom concentrations in homes with electee stoves are less than outdoor
levels. Bedroom levels in homes with gas stoves are higher than those in homes with electee
stoves and are higher than outdoor concentrations Concentrations of NO2 in homes with gas
stoves also demonstrated more variability than levels in homes with electee stoves
The importance of presence and use of gas appliances on indoor NO2 concentrations is
also demonstrated in a study conducted Chattanooga, TN (Parkhurst et al, 1988) In this
study, weekly NO2 concentrations were measured over four periods during a 10-week
sampling period for 235 residences in five housmg developments Four of the developments
were served by natural gas heating (vented) and appliances, and one development was served
by all electee heating and cooking Measured average indoor and outdoor concentrations by
development are shown in Figure 7-20 Houses in the development served by electee
heating and cooking had indoor levels that were 80% of the outdoor levels On the other
hand, NO2 concentrations in homes in the developments served by natural gas for cooking
and heating were 2 6 to 5 8 tunes the outdoor levels In these homes, kitchen levels were
7-47
-------
0090 -1
0080 -
0.070 -
0060 -
Jooso -
0*0040 -
0030 -
0020 -
0010 -
0 -
Percentile
-t- >
? H ^
K
I:
? I
-| >
<
-r 95th
|-L, 75th
x mean
— 50%
LJ 25%
-L 5%
C
T
_ r^
- i J ?
Bed Bed Out Bed LR Kit Bed Out
Gas Elec Door Gas Gas Gas Elec Door
fiiimmnr Winter
Figure 7-19. Concentrations of nitrogen dioxide (ppm) from October through March
during 1988 and 1989, Albuquerque, NM. Concentrations are provided
for infants' bedrooms in homes with gas and electric stoves, outdoors, and
for the kitchen and living room for homes with gas stoves, by season.
Source- Samet et al (1992)
higher than activity room levels, and levels in homes that reported the use of their gas stove
for heating were about 85 ug/m3 (0 045 ppm) higher than in those that did not
There have been a large number of field studies, both in the United States and in
Europe, that have sought to determine the levels (averaged over several days or more) of
NO2 in residences associated with the use of gas appliances and, more specifically, gas
ranges and ovens. These field studies have been directed toward both assessing exposures to
complement epidemiologic studies and to determine the range and distribution of indoor NO2
levels in homes with gas cooking A summary of the findings of the major field studies
(those with large sample sizes) directed toward assessing residential indoor NO2 levels in
homes with gas appliances is shown in Table 7-16 The results of 18 such studies (15 U S
studies, 2 British studies, and a summary of several studies conducted in the Netherlands) are
7-48
-------
0150 -,
ED Outdoor
• Indoor
Electric
Gas 4
Gas 5
Gas 8
Gas 12
Figure 7-20. Indoor versus outdoor nitrogen dioxide in five housing developments in
Chattanooga, TN. One development was all electric and four had gas
heating and appliances (Numbers 4, 5, 8, and 12).
Source Parkhurst et al (1988)
presented as the average NO2 concentrations measured1 outdoors and at various locations
indoors by geographic location, housing type, sampling tune, and the type of cooking device
present Butler et al (1990) and Neas et al (1991) report data from six locations
Many of the homes included in Table 7-16 may have had other gas appliances, which
may or may not have been explicitly described or analyzed, or other sources (attached
garages, leaky flues, etc ) The presence of kerosene heaters was noted in varying
percentages of the residences sampled in the six locations reported by Butler et al (1990)
and Neas et al (1991). It should be emphasized that only the average NO2 concentrations
measured in each study are presented and that there was a broad variation of concentrations
associated with each mean The distributions are not normal This variation can be seen in
Figure 7-14
All measurements in Table 7-16 employed passive NO2 monitors (Palmes et al, 1976)
except for the Spengler et al (1992b) study in California, which used passive badges
(Yanagisawa and Nishimura, 1982, Lee et al, 1992) Table 7-16 also presents estimates of
7-49
-------
<]
o
TABLE 7-16. INDOOR AND OUTDOOR CONCENTRATIONS OF NITROGEN DIOXIDE IN HOMES WITH
GAS APPLIANCES, AND THE CALCULATED AVERAGE CONTRIBUTION OF THOSE APPLIANCES
TO INDOOR RESIDENTIAL NITROGEN DIOXIDE LEVELS3
Average Measured NO2
(ppro)
Housing Averaging
Location Type Time
Southern Mixed 7 days
California
New Haven, Single family 14 days
CT unattached
Albuquerque, Mixed 14 days
NM
California Mobile homes 7 days
Portage, Mixed 7 days
WI
Tucson, Mixed 14 days
AZ
Boston, Mixed 14 days
MA
Gas Appliances
Furnace
Oven/range
w/wo pilot lights
Oven/range
pilots
Oven/range
no pilots
Water heater
in home
Wall furnace
Floor furnace
Oven/range
w/wo pilot
Oven/range
w/wo pilot
Oven/range
w/wo pilot
Oven/range
w/wo pilot
Oven/range
w/wo pilot
Oven/range
w/wo pilot
Number of
Season Homes
Summer
Spring
Winter
Winter
Winter
Winter
Winter
Winter
Winter
Winter
Winter
Summer
Winter
Summer
Winter
Summer
Spring/Fall
Winter
Summer
Fall
Wint /Sprg
147
202
141
98
38
21
90
42
42
82
75
265
231
36
34
13
11
10
301 '<
277
298
Outdoors Kitchen Bedroom
0040
0026
0055
0057
0051
0049
0069
0063
0008
0019
0020
0011
0022
0006
0008
0012
0019
0,024
0023
0021
0049
0042
0054
0060
0039
0031
0085
0094
0024
—
0023
0028
0021
0037
0021
0024
0032
0035
0039
0039
0036
0027
0037
0040
0028
0027
0060
0067
0015
0036
0020
0016
0020
0011
0017
0014
0017
0023
0024
0025
0026
Other
—
—
—
—
—
—
0016
0041
0039
—
0016
0027
0016
0022
0027
0027
0028
0029
Indoor NC>2 Due to Source
(ppm)
Kitchen Bedroom
0016
0019
0025
0028
0011
0006
0026
0035
0020
—
0016
0022
0015
0032
0010
0011
0017
0018
0030
0028
0006
0012
0011
0014
0004
0006
0020
0023
0011
0024
0024
0010
0014
0007
0008
0004
0006
0011
0008
0016
0016
Other Comment Reference
—
—
—
—
—
—
0012
0031
0032
=
0011
0022
0006
0004
0013
0010
0018
0018
1,2
1,2
1,2
1,2
1,2
1,2,3
1,4
1,4
1,5
1,5,6
1,7
1,8
1,9
1,9
Wilson et al
(1986)
Leaderer et al
(1986a)
Marbury et al
(1988)
Petreas et al
(1988)
Quackenboss
etal (1986)
Quackenboss
etal (1987)
Ryan et al
(1988), Ryan and
Spengler (1992)
-------
-J
TABLE 7-16 (cont'd). INDOOR AND OUTDOOR CONCENTRATIONS OF NITROGEN DIOXIDE IN HOMES WITH
GAS APPLIANCES, AND THE CALCULATED AVERAGE CONTRIBUTION OF THOSE APPLIANCES
TO INDOOR RESIDENTIAL NITROGEN DIOXIDE LEVELS3
Average Measured NO2
(ppm)
Location
Northern
Central Texas
Suffolk Co,
NY
Onondago Co ,
NY
New York,
NY
Portage,
WI
Watertown,
MA
Middlesbrough,
UK
Middlesbrough,
UK
Arnet
Enschede
Ede
Vlagttwedde
Rotterdam I
Rotterdam n
Housing
Type
Single family
unattached
Single-family
unattached
Single-family
unattached
Apartments
Single-family
unattached
Not given
Not given
Not given
Not given
Not given
Rural area
Inner City
Inner City
Averaging
Tune
5 days
7 days
7 days
2 days
7 days
3-4 days
7 days
7 days
7 days
7 days
7 days
7 days
7 days
Gas Appliances
Furnace
Oven/range
w/wo pilot
Oven/range
w/wo pilot
Oven/range
w/wo pilot
Oven/range
w/wo pilot
Natural gas
oven/range
w/o pilots
LPgas
oven/range
W/o pilots
Gas cooking
Gas cooking
w/o pilots
Gas cooking
Gas cooking
w/o pilots
Water heaters
IMI
lid
n n
Season
Winter
Winter
Winter
Summer
Fall 1
Fall 2
Winter 1
Winter 2
Spring
All seasons
All seasons
November
December
Winter
Winter
Fall/Winter
II H
nil
Number of
Homes
22
42
56
14
15
9
8
18
13
36
76
60
51
428
183
294
173
162
228
102
Outdoors
0018
0020
0016
0058
0032
0039
0053
0040
0050
0008
0006
0020
0024
0019
0018
0019
0023
0015
0024
0024
Kitchen
—
0041
0033
0065
0051
0057
0064
0067
0064
0035
0035
0039
0046
0113
—
0063
0060
0057
0076
0076
Bedroom
—
0
0052
0034
0035
0040
0033
0043
0019
0020
0024
0024
0031
0032
0023
0013
0027
0034
Other
0029
0028
0027
0056
0038
0040
0050
0043
0052
-
—
0027
0032
_
0044
0051
0027
0027
0042
0039
Indoor NO2 Due to Source
Kitchen
—
0032
0022
0016
0028
0024
0032
0043
0029
0029
0031
0027
0036
0095
0051
0047
0048
0062
0062
Bedroom
—
—
0003
0012
0008
0008
0010
0008
0015
0016
0014
0017
0013
0021
0009
0004
0013
0020
Other
0020
0020
0014
0007
0015
0013
0019
0020
0018
—
0018
0023
—
0032
0020
0015
0018
0028
0024
Commentb Reference
1,10 Koontz et al
(1986)
Research
Triangle Institute
1,9 (1990)
Goldstein et al
(1985)
9,11,12
Spengler et al
(1983)
1,13
Clausing et al
1,9,14 (1984)
1,15 Goldstein et al
(1979)
1,16 Melia et al
(1982)
Noy et al
(1984)
9,17
-------
a
to
TABLE 7-16 (cont'd). INDOOR AND OUTDOOR CONCENTRATIONS OF NITROGEN DIOXIDE IN
HOMES WITH GAS APPLIANCES, AND THE CALCULATED AVERAGE CONTRIBUTION OF THOSE
APPLIANCES TO INDOOR RESIDENTIAL NITROGEN DIOXIDE LEVELS3
Average Measured NO2
Cppm)
Location
Watertown,
MA
Kingston,
TN
St Louis,
MO
Steubenville,
OH
Portage,
WI
Topeka,
KS
Southern
California
Chattanooga,
TN
Housing
Type
Mixed
Mixed
Mixed
Mixed
Mixed
Mixed
Mixed
Apartments in
a development
Averaging Gas Appliances
Time Furnace
7 days
7 days
7 days
7 days
7 days
7 days
2 days
7 days
Oven/range
w/wo pilot
Oven/range
w/wo pilot
Oven/range
w/wo pilot
Oven/range
w/wo pilot
Oven/range
w/wo pilot
Oven/range
w/wo pilot
Gas/oven
w pilot
Gas/oven
wo pilot
Oven/range
w/wo pilots
Oven/range used
for heating
Number of
Season Homes
Summer
Winter
Summer
Winter
Summer
Winter
Summer
Winter
Summer
Winter
Summer
Winter
Not Winter
Winter
Not Winter
Winter
Winter
Winter
162
91
208
93
110
87
221
93
115
34
-
-
Outdoors Kitchen Bedroom Other
0021
0022
0015
0015
0017
0019
0025
0023
0008
0009
0010
0013
0039
0049
0032
0043
0021
0021
0030
0041
0018
0055
0032
0045
0036
0048
0019
0027
0023
0039
-
-
-
-
-
~
0023 -
0025 -
0015 -
0043 -
0020 -
0033 -
0026 -
0031 -
0012 -
0016 -
0014 -
0027 -
0034 -
0034 -
0023 -
0023 -
0075
0120
Indoor NO2 Due to Source
(ppm)
Kitchen Bedroom Other Comment Reference
0014
0030
0016
0029
0017
0033
0012
0020
0013
0029
-
-
-
-
-
-
0007 -
0017 -
0006
0018
0008
0017
0006
0010
0005
0017
0011
0004
0005
0001
0058
0103
1
1,18
1,19
1,20
1,21
1,22
1
1
1,23
Butler et al
(1990), Neas
etal (1991)
Spengler et al
(1992b)
Parkhurst et al
(1988)
Nitrogen dioxide
Mixed = Single-family attached, single-family unattached, condominium apartment
tLP = Liquified petroleum
"The comment codes are as follows
1 Background correction determined by multiplying the indoor/outdoor ratio for homes in the study with no indoor NO2 sources for a given season times the outdoor NO2 concentration measured for
the home with sources and subtracting the product from the indoor level measured in the house
2 Homes contain forced-air gas furnace These homes are thought not to contribute significantly to indoor levels for this sample
3 Homes with electric range/oven, forced-air gas furnace, and gas water heater in home Comparison is made with electric range/oven, forced-air gas furnace, and gas water heater located outside
home
4 Homes have gas range/oven with source contribution calculated after correction of a gas range/oven Values are background corrected with gas stove
5 Living room or activity room
6 Sampling was done over two different periods for the same houses within the same winter period
-------
7 Outdoor values were obtained from five locations, housing types, mobile home
8 Other location is home average, bedroom refers to average of levels in one or more bedrooms in house
9 Other location is the main living room
10 Other location is point nearest center of home
11 48-h samples over 30 consecutive days
12 Indoor/outdoor (I/O) ratio is assessed to be 0 6, 0 7, and 0 85 for the Winter, Spring/Fall, and Summer periods, respectively, for all locations because no control home (no gas appliances)
measurements were available Using these I/O jcatios, the impact of sources was calculated as m footnote #1
13 Each home was sampled six times over a 1-year period
14 Outdoor levels are average for homes with and without gas appliances
15 Outdoor levels were recorded at 75 locations in the general sampling area and are not home-specific Bedroom levels were obtained for 107 of the 428 homes
16 Outdoor levels were recorded at 82 locations in the general sampling areas and are not home-specific Outdoor levels were recorded at the beginning and end of the study
17 Indoor/outdoor (I/O) ratio is assumed to be 0 6 for all locations because no control home (no gas appliances) measurements were available Using I/O ratio of 0 6, the impact of sources was calculated
as m footnote #1
18 90 of the 91 homes had kerosene heaters, thus contribution of gas stove cannot be estimated
19 13 of the 208 homes had kerosene heaters
20 29 of the 93 homes had kerosene heaters
21 10 of the 110 homes had kerosene heaters
22 13 of the 87 homes had kerosene heaters
23 The total number of houses sampled in four developments with gas stoves was approximately 188 No breakdown was given for the number of apartments in which the gas stove was or was not
used for heating The other location here is the whole-house average
-------
contributions of the various apparent indoor sources for all gas appliances to the average
indoor NO2 levels measured, in Equation 7-3 This was done by applying a background
correction factor (subtracting the contribution of outdoor concentrations) to the measured
indoor levels. The correction factor was determined by multiplying the mean indoor/outdoor
ratio for homes in the study with no indoor NO2 sources (Table 7-15) for a given season by
the mean outdoor NO2 concentration measured for the homes with sources and then
subtracting the product from the mean indoor levels Indoor/outdoor ratios of 0 6 for the
winter, 0.85 for the summer, and 0 7 for the fall and spring were assumed for those studies
that did not have a sample of homes without indoor sources
Although only average concentrations are presented in Table 7-16, they do allow for a
number of general observations to be made regarding both the measured levels of N(>2 ui
homes with gas appliances and the calculated contributions of the gas appliances to those
levels.
(1) There is considerable variation in the reported average indoor
concentrations among the U S and European studies (see Table 7-16)
The Parkhurst et al (1988) study and the European studies indicate that
there may be housing groups with gas stoves that have elevated indoor
levels of NO2 These variations are no doubt related to variations in a
number of factors, including outdoor concentrations, cooking patterns,
source characteristics, unreported sources, existence of different gas
appliances (in the Dutch homes, gas-fired, water tap heaters [geysers]),
house infiltration rates, use of local exhaust fans, house volumes, and
differences in removal rates by internal surfaces
(2) Both indoor and outdoor concentrations are generally higher in the
winter than in the summer The higher indoor levels in the winter are
reflected in the calculated contribution by gas stoves shown below The
higher indoor levels during the winter may be due to lower ventilation
rates, increased source use, and possible seasonal differences in removal
rates
(3) Within a typical house with an unvented gas cooking range in the
kitchen, a concentration gradient within a residence and between seasons
exists. The concentration is highest in the kitchen and lowest in the
bedroom, with other rooms (e g , living room) between the two
Bedroom levels can typically range from 50 to 75 % of those measured
in the kitchen This gradient is reflected in the calculated contribution
of gas cooking ranges and ovens to the average concentrations of NO2
m the kitchen, bedroom, and other rooms among the U S studies
7-54
-------
(excluding the Parkhurst et al [1988] study) The contribution of gas
cooking ranges and ovens to the average concentrations, by room, by
season, is very consistent, varying by about a factor of 2 Winter
contributions to the kitchen, bedroom, and other locations across all
studies averaged 52, 28, and 39 /*g/m3 (0 028, 0 015, and 0 021 ppm),
respectively, whereas summer contributions averaged 28, 13, and
16 jwg/m3 (0 015, 0 007, and 0 008 ppm), respectively The
contribution during the fall/spring period hes between the summer and
winter periods Although the contribution to the kitchen in the
European studies is higher than the U S studies, the contribution of gas
stoves to the other areas in the home is similar for the European and
U.S. studies It should be noted that these are only approximations and
that the variation associated with them is large The seasonal
differences may be related to seasonal differences in source use,
infiltration, and removal by interior surfaces and furnishings The
gradient may be due to differences in air mixing within and between
rooms in residences
(4) Indoor concentrations are higher in homes that have gas cooking ranges
with pilot lights than those without pilot lights The extensive Wilson
et al (1986) data indicate that pilot lighls in the gas cooking range add
about 20 /xg/m3 (0 Oil ppm) to the kitchen levels in the winter and less
to the other rooms Data from the Boston study (Ryan and Spengler,
1992) indicate that the presence of a comtinuously burning pilot light in
a gas range home increases the indoor NO2 concentration by
approximately 0 010 ppm above that which it would be if the gas range
used an alternative ignition system The water heater was found on
average to add approximately 12 jtg/m3 (0 006 ppm) to the kitchen
during the winter and less to the rest of the house This result,
however, was not found to be statistically significant due to high
variability in the measurements Also, no data examine peak
concentrations in homes with gas range with and without pilot lights
(5) The presence of wall or floor furnaces with gas appliances is more
associated with higher concentrations than just gas appliances Indoor
NO2 levels associated with wall or floor furnaces were thought to be
due to leaky flues (Wilson et al, 1986) In a follow-up study (Beals
et al, 1987), the homes with high levels of NO2 were investigated for
the source of high levels and to determine the role of leaky flues The
results of the follow-up study indicated that there was not a generic
problem with this system However, this system is more likely to have
an undetected leaky flue or unvented pilot light
(6) Elevated NO2 levels in apartments associated with the use of gas ranges
for space heating were observed in one study (Parkhurst et al, 1988)
and suspected in another (Beals et al, 1987) In the Parkhurst et al
(1988) study, NO2 levels were found to increase approximately
7-55
-------
85 /ig/m (0 045 ppm) due to stove use for space heating The Seals et
al (1987) study found that 20% of all residences with gas ranges use
the ranges as supplemental heat sources, 28 % of those residences with
wall or floor furnaces reported occasional use of the gas range for space
heating The impact on residential levels, however, was not estimated
In a recent study conducted in Albuquerque, NM (Spengler and Samet,
1992), it was estimated that approximately 10% of the homes with gas
stoves reported using them for space heating Figure 7-21 compares
NO2 levels across several seasons for residences with gas stoves that do
and do not use the stove for space heating This study estimated that
during the heating season, the use of gas stoves for space heating added
approximately 0 038 ppm to the bedroom Spengler et al (1992b)
report use of stove to heat on the average of 13 % but ranging fiom
3.8 to 33 3% across U S and Canadian communities Koontz et al
(1992) report a relatively small fraction of households, about 6% in the
United States use a range for residential heating It is used during
winter for 1 5 to 2 0 days
0060H
0050-
•£« 0040-
if
I
0030-
0020-
0010_
I I I I I I T I I I I
APR88 JUL88 OCT88 JAN89 APR89 JUL89 OCT89 JAN90 APR90 JUL90 OCT90 JAN91
Date
Figure 7-21. Nitrogen dioxide concentrations across seasons in Albuquerque, NM;
bedrooms that do and do not use the gas stove for space heating.
Source: Spengler and Samet (1992)
7-56
-------
It is remarkable that the contribution of gas cooking to indoor NO2 levels is as
consistent as it is among studies for locations in the residences and by season, given the great
variability of the factors that govern the emissions (source type, source condition, source use,
and source venting) and dilution and removal of NO2 indoors (house volume, infiltration,
mixing within and between rooms, decay rates, etc ) This consistency is not observed until
the impact of outdoor concentrations is corrected for These background levels can vary
considerably over time and geographic area The impact of gas cooking and possibly other
unvented or improperly vented combustion sources on indoor NO2 levels is superimposed
upon the indoor background level resulting from outdoor levels In areas where outdoor
levels are low, concentrations indoors from gas appliances will be higher than (and in many
cases, much higher than) outdoor levels (e g , Marbury et al, 1988, Quackenboss et al,
1987, 1988, Spengler et al, 1983, Leaderer et al, 1986a, Ryan et al, 1988, Ryan and
Spengler, 1992) If outdoor concentrations are high, then indoor levels in homes with gas
appliances will be closer to and even lower than the outdoor levels (Wilson et al, 1986)
In the Los Angeles study (Colome et al, 1992), exposures to NO2 away from home on
average exceed personal exposures encountered while at home due to higher outdoor NO2
concentrations In Los Angeles (Spengler et al , 1992a,b), outdoor concentrations are strong
predictors of personal exposure
Ryan et al (1992) observes that bedroom concentrations have a strong influence on
personal exposures in both Boston and Los Angeles A salient difference between Boston
and Los Angeles is that, in Los Angeles, the outdoor concentration appeared to play a
dominant role in influencing indoor concentrations, whereas in Boston this is not the case
The contributions of gas appliances to indoor levels of NO2 shown in Table 7-16 are
average concentrations calculated from the average levels reported by the investigators for
each field study It is important to note that the variability around the calculated
contributions in Table 7-16 is generally large
Koontz et al (1992) report the results of 1985 and 1991 surveys of type of cooking
facilities and frequency of cooking in the United States The frequency distribution for type
of range is shown in Table 7-17, which shows a 2 6% decrease in gas ranges with pilot
lights The statistical variation of these results is not reported The average amount of tune
in minutes spent using the range to cook by income is shown in Table 7-18 Changes in
7-57
-------
TABLE 7-17. FREQUENCY DISTRIBUTION FOR TYPE OF RANGE
FROM 1985 AND 1991 SURVEYS
Type of Range
Gas, with pilot light
Gas, pilotiess
Gas, not sure
Electric
Gas and electric
None
Percent Frequency for
1985 Survey 1991
293
74
02
61 0
19
01
Survey
267
99
07
608
19
0 1
Source. Koontzetal (1992)
TABLE 7-18. AVERAGE NUMBER OF DAYS OF RANGE USE PER WEEK
FOR COOKING, BY INCOME
Characteristic
Income of Respondent
Less than $12,500
$12,500 to $22,499
$22,500 to $34,999
$35,000 to $49,999
$50,000 or more
Days for Breakfast
28
28
24
19
22
Days for Lunch
26
24
22
16
1 8
Days for Dinner
46
48
49
5 1
5 1
Source Koontzetal (1992)
reported range/oven cooking frequency in 1985 and 1991 related to microwave use is shown
in Table 7-19. Use of gas ranges without pilot lights and changes in cooking practices, such
as increased use of microwave ranges, could result in lower NO2 levels in such homes
The Koontz et al. (1992) survey examines the change from 1985 to 1991 but provides
no data for earlier use of cooking stoves with and without pilot lights or the use of
microwave ranges The trend in this data suggests that, in the period before 1985, homes
with pilot lights would be more common than after 1985, and other cooking choices, such as
microwave, were used less frequently Thus, in epidemiology studies (see Chapter 14)
7-58
-------
TABLE 7-19. REPORTED RANGE/OVEN COOKING FREQUENCY IN 1985
AND 1991, BY TYPE OF OTHER COOKING APPLIANCE
Average Cooking Days per Week
Type of Other Cooking Appliance
Microwave
None
Microwave
None
Microwave
None
oven only
oven only
oven only
1985 Survey
Breakfast
35
37
Lunch
29
29
Dinner
55
55
1991 Survey
23
33
22
32
49
58
Source Koontzetal (1992)
conducted before 1985 that did not directly measure NO2 levels, NO2 estimates based on
average differences between homes with gas ranges with pilot lights and homes with electric
stoves may reflect gas stove type and other cooking practices in use at that time
7.3.3.2 Spatial Distributions
As demonstrated in Table 7-16, NO2 concentrations in residences with gas appliances
exhibit a pronounced variation by season and by location in a residence The calculated
contribution of gas appliances to indoor levels of NO2 (corrected for outdoor contributions—
Table 7-16) is highest in the winter and lowest in the summer, with the largest differences
seen in the kitchen The calculated total seasonal differences are on the order of a factor
of 2 (e g , Quackenboss et al, 1986, 1987, Ryan et al, 1988, Ryan and Spengler, 1992,
Wilson et al , 1986, Spengler et al, 1983, Goldstein et al, 1985) The seasonal effect is
related to variations in outdoor NO2 levels, source use, infiltration, and removal by interior
surfaces
7-59
-------
Spatial distributions of NO2 within and among rooms in a house where a gas range or
oven is used are a function of mixing in the space Goldstein et al (1985) reported the
vertical distributions of NO2 levels in nine apartments in New York City where gas ranges
were used. The concentrations were 48-h average values (Palmes tubes) measured at five
elevations in the kitchens and living rooms of each apartment Figure 7-22 shows the result
of that study. A pronounced vertical gradient was observed in the kitchen, with the highest
levels observed at the ceiling and lowest at the floor A similar, but less pronounced,
gradient was observed for the living area Wilson et al (1986) investigated the vertical
distribution of NO2 levels in gas-cooking homes The results showed a vertical gradient in
the kitchen for some, but not all, of the homes The potential for a strong spatial NO2
gradient in kitchens with gas ranges suggests that placement of monitors in the kitchen during
field studies could result in a larger standard deviation in kitchen concentrations than for
other rooms. In considering the results from such studies, the monitor placement issue has
to be considered in interpreting the results
All studies investigating NO2 concentrations in homes with gas appliances have found a
concentration gradient between rooms (Table 7-16), with the kitchen being highest and the
bedroom being lowest This gradient is highlighted in Figure 7-23 Figure 7-23 also
highlights the seasonal differences in indoor NO2 levels in homes with gas appliances
In this study (Spengler et al, 1983), season and location in the house were found to be
statistically significant predictors of NO2 levels in homes with gas appliances The within-
home spatial variations are related to such variables as air exchange rates among rooms, air
mixing within a room, volume of a house, location of the air sampler, and the frequency and
length of gas appliance use
7.3.3.3 Short-Term Indoor Concentrations
The majority of data on indoor NO2 levels associated with gas stove use is from
integrated monitors, sampling over periods of days The recorded NO2 values are averaged
over several on/off periods of gas stove use, and as such, do not directly measure short-term
indoor NO2 levels (levels on a time frame of minutes or hours) in homes, which occur
during source use. The short-term concentrations are associated with length of source use,
number of sources (e g., number of gas burners used), and location at which the
7-60
-------
6
5
I 4
o
o
£
5
3
2
1
LIVING ROOM
KITCHEN
10 20 80 90 100 110 120 130
NQ, (ug/nrf3)
Figure 7-22. Verticle distribution of average nitrogen dioxide concentrations (48-h
sampling periods) measured in nine New York City apartments. Plotted
from data by Goldstein et al. (1985).
measurement is taken relative to the source (e g , immediately over the source or several feet
away) Few studies have measured short-term NO2 levels in residences with gas appliances
In a study of the incidence of respiratory illness in households using gas and electricity
for cooking, Keller et al (1979) used continuous chemiluminescence monitoring over 3-day
periods in 46 homes in the Columbus, OH, area to measure continuous variations in indoor
NO2 and NO levels in relation to cooking tunes The study found that variations in
instantaneous peak NO2 levels (highest concentrations measured with a continuous monitor)
in gas-cooking households reached as high as eight times the 24-h average values In several
households, instantaneous peak NO2 concentrations exceeded 1,900 jwg/m3 (1 ppm) The
study did not report the location of the sampler relative to the source, the number of sources,
length of source use, or sample averaging time As part of a study of respiratory disease
7-61
-------
I
c
o
I
8
§
CO
CD
100
90
80
70
60
50
40
30
20
10
SUMMER
FALL
WINTER
O
OUTDOORS
LP-KIT
NG-KIT
E-KIT
LP-BED
NG-BED
E-BED
1 2345678
Period (July 1980 - June 1981)
figure 7-23. Mean nitrogen dioxide concentrations (1-week sampling periods) for eight
sampling periods by location in the home and type of cooking fuel.
Source Spengler et al (1983)
rates and pulmonary function in children associated with NO2 exposuie, Speizer et al (1980)
reported instantaneous peak NO2 exposures, measured by a continuous chemiluminescence
o
monitor, in excess of 1,100 jwg/m (0 583 ppm) in a kitchen within 3 ft of a gas range/oven
o
(gas oven on), with 1-h average peak exposure at approximately 665 /*g/m (0 352 ppm)
The peak 1-h average level associated with a range-top gas burner was 428 /*g/m
(0.227 ppm). Concentrations were monitored continuously by chemiluminescence in only
one house. Hosein and Bouhuys (1979), using monitoring by chemiluminescence, reported
o
peak 2-h NO2 levels of over 3,000 jttg/m (1 59 ppm) 3 ft from the source in a kitchen
during use of a gas range
In one of the first studies of the impact of gas ranges on indoor air quality, Wade et al
(1975) reported continuous NO2 measurements (chemiluminescence) in three locations in four
7-62
-------
houses with gas ranges Measurements were spread over all four seasons and outdoor levels
were recorded Nitrogen dioxide levels in the kitchen responded rapidly to gas range use,
with less rapid response in other locations in the house Peak 5-min concentrations 1 m from
the gas range exceeded 100 ^g/m3 (0 053 ppm) over 50% of the time in two of the houses
and 20% of the tune in one house The data were not sufficient to construct cumulative
frequency distributions for the fourth house measured Peak levels were considerably lower
in other locations in the houses The NO2 frequency distributions for longer averaging times
(e g , 1 h) for other locations in the houses were not reported
The most extensive data collected to date on peaik indoor levels of NO2 associated with
gas appliance use are reported by Harlos et al (1987) In this study, an electrochemical-cell
NO2 monitor was used to record time-averaged NO2 levels of 5 s and longer during cooking
for personal exposure for over 18 volunteers Subjects wore the electrochemical monitor
during their cooking activities and the sample was drawn at the breathing zone Table 7-20
shows the summary statistics reported from the study for personal exposures over averaging
times from 0 6 seconds to 1 h Peak 1-min concentrations reached as high as 1,880 jwg/m3
(1 ppm) The considerable variability in recorded exposures is due to source, source use,
subjects' activities, and air-mixing characteristics in the kitchen
Lebret et al (1987) conducted real-tune NO2 concentration measurements at three
locations (kitchen, living room, and bedroom) and oul doors for 12 Dutch homes using a
chemilumrnescence monitor Measurements were conducted over periods of 135 to 273 h
The homes sampled had gas ranges and geysers (demand gas water heaters) Maximum
1-min average concentrations in the kitchens ranged fiom 400 to 3,808 jttg/m (0 21 to
2 02 ppm) in the kitchen, whereas the living room and bedroom levels were typically on the
order of 30 and 18%, respectively, of the kitchen levels Maximum 1-h average
concentrations in the kitchen ranged from 230 to 2,055 /ig/m (0 12 to 1 09 ppm), whereas
living room levels were typically 50% of the kitchen levels and bedroom levels were about
30 % of the kitchen levels The relative contribution to peak levels from gas range use verses
geyser use was not determined
Using a chemiluminescent analyzer, Tikalsky et al (1987) measured NO2 levels
continuously in three locations (kitchen, main activity room, and outside) in 10 homes for
two 3- to 6-day sampling periods The sampling probes were placed 4 to 6 ft above the
7-63
-------
TABLE 7-20. SUMMARY STATISTICS FOR GAS RANGE NITROGEN DIOXIDE
MAXIMA (ppm) OVER SEVERAL AVERAGING TIMES
Time
Gas Cooking (n = 18)
Mean Maxima
Standard Error
Maximum
Minimum
5s
052
0.24
1.20
0196
15s
044
023
1 13
0175
1 nun
038
0215
10
013
3 mm
034
022
098
Oil
30 mm
022
013
060
007
Ih
018
009
042
006
Source Harlos et al (1987)
ground and, in the kitchen, 6 to 10 ft from the gas stove The homes had gas appliances
The continuously monitored NO2 concentrations were averaged over 1-, 3-, and 24-h
periods. Data were also gathered on source use One-hour average NO2 levels in the
kitchens and activity rooms as high as 1,000 and 750 /*g/m3 (0 53 and 0 40 ppm),
respectively, were recorded Hourly NO2 monitoring data covering one 24-h period for one
home are shown in Figure 7-24 Baseline levels in the home without oven and stove top use
are above outdoor levels and are related to pilot light and water heater use The highest 1-h
average concentrations occurred in the late morning and late afternoon and were associated
with oven use (morning peak) and stove-top use (late afternoon) Time plots of NO2 for the
other homes sampled demonstrate considerable variability within the course of a day, with
N02 levels in the homes varying with gas appliance use and, in some cases, with furnace
use.
There is only a very limited data base available on short-term indoor concentrations of
NO2 associated with gas appliance use In the absence of adequate field study data, it is
difficult to assess the short-term average levels indoors associated with gas appliances
The existing data, however, suggest that short-term indoor NO2 concentrations of concern
may be higher than those recorded for outdoors
7-64
-------
12 1 234587 89 10 11 12 12 345 878 9 10 11 12
Hour
AAAA AAAA
O n niMiiiniiru
w
00
I i I I I ] I I I I I I 1 [ T I I I I I I 1 I
12 1 234567 89 10 11 12 12 3466789 10 11 12
AM
O Furnace
A Water Heater
n Stove Top
PM
Hour
V Ovan
O Exhaust FaiVdolhas Dryar
AM
Figure 7-24. Nitrogen dioxide hourly levels in one home with gas appliances.
Source Tikalsky et al (1987)
7.3.4 Unvented Space Heaters
Unvented space heaters are used in the colder climates to supplement central heating
systems or in more moderate climates as the primary siource of heat During the heating
season, space heaters will generally be used for a number of hours during the day, resulting
in emissions over a relatively long period of tune
The actual number of unvented space heaters in use in residential and commercial
settings in the United States and then1 use patterns is not known In many countries,
kerosene heaters are an important heating source (Middle East, Japan, Korea, etc )
As many as 17 million such heaters have been sold through 1987, with current yearly
estimated sales of 1 million A residential energy survey conducted by the U S Bureau of
7-65
-------
Census (1982) estimated that 3 million residences use unvented gas space heaters (fueled by
natural gas or propane) The heaters are used in 3 to 4% of the houses in the United States,
with their use being more prevalent in the South Census region of the United States, where
they are the main heating system in about 10% of the housing units In recent years, the
numbers of new unvented kerosene and gas space heaters sold have declined The large
number of unvented space heaters sold m the United States and the potential for their use,
particularly during periods when energy costs rise quickly, make them an important source of
N02 indoors.
As discussed in Chapter 4, NO2 emissions from unvented kerosene and gas space
heaters can vary considerably and are a function of heater design (convective, radiant,
combination burner designs, etc ) and condition of heater and manner of operation (e g ,
flame setting) Levels of NO2 indoors resulting fiom heater use are a function of the heater
emission variables, along with heater use variables (number of hours of use, volume of house
heated, etc ) and the variables governing the dispersal and elimination of the NO2 emissions
indoors (infiltration, mixing, etc) The contribution of NO2 emissions from unvented
kerosene and gas space heaters to indoor concentrations in simple terms is given in the
source contribution term in Equation 7-3 Because kerosene and gas space heaters, unlike
other gas appliances, are typically used for several hours at a time, they approach being a
continuous source and hence their contribution to indoor concentrations is probably
reasonably represented by the steady state model in Equation 7-3
7.3.4.1 Unvented Kerosene Space Heaters
Average Indooi Concentrations
The most extensive data collected to date on residential levels of NO2 associated with
the use of unvented kerosene space heaters are reported by Leaderer et al (1986a) This
field study of 333 homes m the New Haven, CT, area was conducted during the 1982-1983
heating season to assess the range and distribution of air contaminants associated with
residential unvented combustion sources, with particular emphasis on NO2 levels related to
kerosene heaters, and to assess exposures to complement an epidemiologic study of the health
impact of heater use The study employed a nested design for exposure assessment that
utilized questionnaires and seveial levels of air monitoring Two-week average NO2 levels
7-66
-------
were recorded in three locations in each house (kitchen, living room, and bedroom) and
outdoors using Palmes tubes (Palmes et al , 1977)
The measured 2-week NO2 concentrations by location in the homes for six general
source categories are shown in Table 7-21 Also shown in Table 7-21 are the percent of
homes in which NO2 levels exceeded the primary ambient air quality standard The findings
indicate that the greater the number of sources, the higher the average concentrations of
NO2 Homes with one kerosene heater and no gas range/oven had NO2 levels four to five
tunes higher than the levels in homes without a heater or a gas range/oven Nitrogen dioxide
levels in homes with a kerosene heater but no gas range/oven were roughly comparable to
homes with a gas range/oven only The study also showed that homes with convective
heaters had higher NO2 levels than homes with radiant heaters It was noted that the average
concentrations measured may have been lower than would have been expected due to the
particularly mild winter encountered during the study— the median of daily hours of kerosene
heater use was only 6 h/day
The data in Table 7-21 represent the average levels over a 2- week period and do not
reflect the actual concentrations in the homes during heater use Using the measured
concentrations and questionnaire data on heater use during air sampling and correcting for>
outdoor levels, the authors calculated the indoor NO2 levels that may have existed during
actual heater use The resultant calculated cumulative frequency distribution of NO2 levels
by location in the homes during heater use for homes with one kerosene heater and no gas
range/oven is shown in Figure 7-25 The adjusted data show that over 49 % of the residences
with one kerosene heater had average NO2 concentrations in the house in excess of
100 /ig/m3 (0.053 ppm), with 8 4% in excess of 480 ^g/m3 (0 254 ppm) The levels would
be higher in homes with more than one kerosene heater and/or a gas range
In a study of two Vermont homes in which kerosene space heaters were used, Ryan
et al (1983) found average NO2 concentrations indoors (Palmes tubes) to range from 19 to
304 jitg/m3 (0 01 to 0 16 ppm) over two sampling periods from 81 to 174 h Nitrogen
dioxide levels were highest in the house that used two kerosene heaters as a primary heat
source Traynor et al (1984), in a field study of indoor air pollutants in residences with
suspected combustion-related sources, reported 1-week average NO2 levels in three homes
with kerosene heaters and five homes with a kerosene heater and gas range Nitrogen
7-67
-------
TABLE 7-21. TWO-WEEK AVERAGE NITROGEN DIOXIDE LEVELS BY
LOCATION FOR HOMES IN SIX PRINCIPLE SOURCE CATEGORIES,3
NEW HAVEN, CONNECTICUT, AREA STUDY, WINTER, 1983
Source Category,
Location
No Kerosene Heater or Gas Stove
Outdoors
House Average
Kitchen
Living Room
Bedroom
One Kerosene Heater, No Gas Stove
Outdoors
House Average
Kitchen
Living Room
Bedroom
No Kerosene Heater, Gas Stove
Outdoors
House Average
Kitchen
Living Room
Bedroom
One Kerosene Heater, Gas Stove
Outdoors
House Average
Kitchen
Living Room
Bedroom
n
144
145
147
146
145
95
95
96
96
95
42
42
42
42
42
18
18
18
18
18
Nitrogen
Mean
0007
0004
0004
0004
0004
0007
0020
0021
0020
0017
0008
0018
0024
0016
0015
0008
0035
0040
0030
0036
Dioxide (ppm)
SDb
0003
0002
0002
0002
0005
0002
0017
0019
0019
0016
0002
0014
0017
0013
0013
0003
0023
0028
0020
0030
%>0053ppm
0
0
0
0
0
0
2 1
42
52
53
0
48
48
48
48
0
167
222
11 1
167
7-68
-------
TABLE 7-21 (cont'd). TWO-WEEK AVERAGE NITROGEN DIOXIDE LEVELS
BY LOCATION FOR HOMES IN SIX PRINCIPLE SOURCE CATEGORIES,3
NEW HAVEN, CONNECTICUT, AREA STUDY, WINTER, 1983
Source Category,
Location
Two Kerosene Heaters, No Gas Stove
Outdoors
House Average
Kitchen
Living Room
Bedroom
Two Kerosene Heaters, Gas Stove
Outdoors
House Average
Kitchen
Living Room
Bedroom
n
13
13
13
13
13
3
3
3
3
3
Nitrogen
Mean
0009
0 037
0038
0039
0036
0012
0045
0050
0041
0045
Dioxide (ppm)
SDb
0005
0020
0017
0023
0024
0003
0013
0012
0020
0010
%>0053ppm
0
230
230
385
23 1
0
333
666
333
333
Repeat monitoring data (n = 19) are included Samples were lost for two homes in one home, the monitors
were capped early by the residents, and in the second home, repeated efforts by the interviewers to retrieve
the monitors failed
SD = Standard deviation
Source Leaderer et al (1986a)
*3
dioxide levels in the homes with kerosene heaters ranged from 48 to 222 j^g/m (0 025 to
0 118 ppm) The differences in concentrations reflected differences in usage
Spatial Distributions
As shown in Table 7-21, NO2 levels in homes wilh a kerosene heater only do not
exhibit a pronounced concentration gradient among rooms in a house Leaderer et al (1984)
found no strong spatial gradient among rooms, which contrasted with the strong gradient
7-69
-------
100r
80
S. 60
I
o
40
20-
98
Bedroom = 0 057 ppm
| Kitchen = 0 088 ppm
O Living Room = 0 081 ppm
01 02 03 04 05 06
NOs Concentration (ppm)
07
08
09
Figure 7-25. Cumulative frequency distribution and arithmetic means by location, of
average nitrogen dioxide levels (2-week sampling periods) during kerosene
heater use for residences with one kerosene heater and no gas range, New
Haven, CT, area study, winter 1983.
Source Leaderer et al (1986a)
observed for homes with gas ranges The relatively long operating periods for the heater, on
the order of several hours, and the strong convective heat output evidently foster rapid
mixing within the homes where they are used
Concentrations During Heater Use
Because the heaters are used for several hours at a time, unlike gas appliances,
equilibrium concentrations are more likely to be achieved, resulting in sustained high NO2
concentrations over periods of several hours rather than seconds or minutes as with gas
appliances. The variability in these equilibrium levels from heater use period to heater use
period for a given heater will be driven by variations in infiltration rates, NO2 deposition
rates and the volume of the house heated (e g , one room or several rooms)
7-70
-------
As part of the nested air sampling design protocol used by Leaderer et al (1987),
14 residences were monitored for periods of 43 to 209 h for NO2 levels in two locations m
the home (room with heater and a bedroom) and outdoors using a continuous
chemiluminescence monitor (Leaderer et al, 1984) ITurteen had kerosene heaters, of which
four had a gas range, whereas one house had a gas range but no kerosene heater
Equilibrium levels in the homes associated with kerosene heater use ranged from 19 3 to
847 /ig/m3 (0 010 to 0 45 ppm), and levels typically exceeded 100 /ig/m3 (0 053 ppm)
In this set of houses, levels were generally higher in the room where the heater was used
These levels were typically sustained over periods of several hours (outdoor levels
subtracted) and were sustained for a 24-h period m one of the houses monitored (Leaderer
et al, 1986a)
7.3.4.2 Unvented Gas Space Heaters
Average Indoor Concentrations
The most comprehensive study on indoor levels of NO2 associated with the use of
unvented gas space heaters (UVGSHs) was reported by Koontz et al 1986 and 1988 In this
study, 157^ residences in four Texas cities were monitored for NO2 concentrations and
frequency of UVGSH and gas range use from January to March 1985 Nitrogen dioxide was
monitored in two locations m all homes (central location and remote or peripheral location)
using Palmes tubes over sampling periods of approximately 5 days In a subsample of
16 homes, NO2 was measured continuously by chemiluminescence
The cumulative percent distributions of 5-day average NO2 levels measured m all
homes in both locations are shown in Figure 7-26 Approximately 70% of the homes
exceeded 100 jug/m3 (0 053 ppm), and 20% exceeded 480 jwg/m3 (0 25 ppm) The
cumulative frequency distributions for average NO2 by three categories of homes (primary
UVGSH, secondary UVGSH, and non-UVGSH) are shown in Figure 7-27 The highest
concentrations were measured m homes where the UVGSH was the primary source of heat
<> -3
(average of 472 /xg/m [0 25 ppm]) and lowest for non-UVGSH (average of 56 ftg/m
[0 03 ppm])
Table 7-22 presents a more detailed breakdown of the average levels recorded as a
o
function of source categories Gas stove use alone contributed 57 jwg/m (0 03 ppm) on
7-71
-------
100
so -
80 -
70 -
SO -
40 -
30 -
20 -
10 -
Mean + standard deviation
D Central location
+ Peripheral location
Central location - 0165 + 0198 ppm
Peripheral location -0163 + 0198 ppm
01
02
03
04 05
N02 (ppm)
06
07
08
09
Figure 7-26. Cumulative frequency distributions and summary statistics for integrated
nitrogen dioxide measurements in two locations (152 study homes).
Source Koontzetal (1986)
average to residential levels The use of UVGSHs in homes either as a primary or secondary
heat source results in high levels of NO2 in those homes In this study, associations between
indoor concentrations of NO2 and variables derived from questionnaires, activity logs, and
other recorded information (indoor and indoor temperatures, gas-meter readings, etc ) were
examined. The single most important variable accounting for variations in indoor NO2 levels
in homes using UVGSHs was the difference between indoor and outdoor temperatures
Li homes using the heaters as a primary heat source, variations in indoor/outdoor
temperature differences accounted for 64% of the variation in NO2 levels and 33% of the
variation for homes where the heaters are used as a supplemental heat source
In a study of 14 homes with one or more UVGSHs (primary source of heat) in the
Atlanta, GA, area, McCarthy et al (1987) measured NO2 levels by both chemiluminescence
and passive monitors in two locations in the homes (room with the heater and a remote room
in the house) and outdoors Chemiluminescence measurements were taken over 5-min
7-72
-------
Mean + Standard Deviation
D Non-UVGSH-0030 ± 0026 ppm
Secondary UVG'SH-0113 ± 0122 ppm
A Primary UVGSH-0 251 ± 0224 ppm
N02 (ppm)
Figure 7-27. Cumulative frequency distributions and summary statistics for indoor
nitrogen dioxide concentrations in three groups of monitored homes.
Source Koontz et al (1988)
periods in turn from each of the three sampling points for each house over a 96-h sampling
period The authors reported only the summary statistics for NO2 based on the continuously
collected data Eleven of the 14 UVGSH homes exceeded 100 pg/m (0 053 ppm) during
•^
the sampling period Mean values ranged from 40 to 1,460 /ig/m (0 02 to 0 77 ppm) and
varied as a function of the use pattern of the heater Only one of the homes used more than
one heater during an- sampling
The highest residential average NO2 concentrations observed for homes are associated
with UVGSHs Unvented gas space heaters are a major source of NO2 in residences The
reported average concentrations covered periods when the heaters were used as well as not
used Unless the gas heaters were in continuous operation during the air sampling penod,
actual NO2 levels during heater use will be higher than the average values reported
7-73
-------
TABLE 7-22. ONE-WEEK AVERAGE NITROGEN DIOXIDE LEVELS IN HOMES
IN NORTH CENTRAL TEXAS BY SOURCE CATEGORY, WITH AND WITHOUT
UNVENTED GAS SPACE HEATER3
Source Category.
No UVGSH, Gas Stove
UVGSH as Secondary
Heat Source, No Gas Stove
UVGSH as Secondary
Heat Source, Gas Stove
UVGSH as Primary Heat
Source, No Gas Stove
UVGSH as Primary Heat
Source, Gas Stove
aN02 = Nitrogen dioxide
N = Number
SD = Standard deviation
N
22
9
29
5
73
NO2 (ppm)
Mean
0030
0068
0099
0098
0280
SD
0019
0075
0094
0053
0235
UVGSH = Unvented gas space heater
Source Koontzetal (1986)
Concentrations During Heater Use and Spatial Variations
Results from the two field studies on indoor NO2 concentrations in residences with
UVGSHs (Koontz et al., 1986, 1988, McCarthy et al, 1987) indicate that there are no
significant differences in NO2 levels between the room with the heater and remote rooms in
the houses monitored Koontz et al (1986, 1988) demonstrated there was not a strong
spatial NO2 gradient in homes using UVGSHs Concentrations were highly correlated
between the two locations (r = 0 89), with a difference of less than 20% in more than half
the homes. This finding is similar to that for unvented kerosene space heaters and is in
contrast to the pronounced spatial gradient for NO2 observed for homes with gas ranges
The high convective heat output of the heaters and relatively long heater use tunes apparently
foster rapid and complete mixing in houses
7-74
-------
Although both field studies collected data on short term indoor concentrations of NO2
associated with UVGSH use, only Koontz et al (1988) have reported the results of those
measurements Of the 16 houses monitored continuously for NO2, 12 reported use of their
UVGSH during the monitoring period A box plot of the monitoring results (15-min
averages) for these 12 houses is shown in Figure 7-28 The wide variation in concentrations
o
is evident with a highest 15-min concentration of 2,716 /ig/m (1 44 ppm) recorded in house
A The higher concentrations may be the levels encountered for sustained periods of tune
(e g , hours) because they may approximate the equilibrium levels under periods of heater
use
1 o
£
CL
.O.
CXI
05
I
I
= f
f
I
^
]\
1
^
I + :
r *-
i i
3
ABCDEFGHIJKL
i — Maximum
I
7
-, — 75th percentile
Average
50th percentile
-J — 25th percentile
— Minimum
Home
A-FW18105
B-FW01713
C-FW06910
D-DL01115
E-DL01218
F-DL05303
G - DL08503
H-DL11624
I-WF11012
J-WF11615
K-GP15606
L-GP15917
Key
Figure 7-28. Nitrogen dioxide box plots for 12 continuously monitored homes.
Source Koontz et al (1988)
7-75
-------
7.3.5 Other Sources
The major sources of NO2 in residences are unvented gas and keiosene space heaters,
gas appliances, and outdoor NO2 levels Improper use of gas appliances (e g , using a gas
stove to heat living space) and improper operation of vented gas appliances (e g , improper
use or malfunctiomng gas appliances) can be important contributors to NO2 concentrations
measured indoors. Beals et al (1987) provides some data on the contribution to indoor NO2
levels from improper use of gas appliances The highest NO2 concentrations in the homes
studied were associated with the use of a gas range/oven as a supplemental heat source The
impact of improperly operating wall or floor furnaces (spilling a portion of the exhaust fumes
into the home) on indoor NO2 levels has not been assessed
The contribution of wood or coal burning stoves or fireplaces to indoor NO2 levels has
not yet been assessed To the extent that there is leakage of the exhaust gas into the living
space during stoking the fire or through spilling of a portion of the exhaust gas into the living
space, these sources will contribute to indoor levels of NO2
Using Palmes tubes, Good et al (1982) compared 7-day average NO2 levels in homes
with and without smokers and without gas ranges Concentrations weie measured in three
locations in each home (living room, bedroom, and kitchen) and outdoors There were a
total of 79 homes monitored (no gas range) over two seasons (winter and summer) Analysis
of the data indicates that the contribution to residential NO2 levels from cigarette smoking
-------
some form of the general mass balance equation (e g , Equation 7-1) The physical/chemical
modeling approach requires detailed information on the input parameters (source strengths,
infiltration rates, mixing, reaction rates, etc ) to predict the indoor concentrations The input
parameters are either measured ui chamber studies and in homes or are estimated The
second modeling approach is statistical in nature based upon empirical measurements These
models make simple assumptions with little or no transformations of the independent
variables that are input to the model The statistical models utilize as input parameters data
obtained in large field studies through both measurement and estimation (questionnaires)
The statistical models are typically simple linear models where the independent variables are
used as they are recorded from the questionnaires to explain variations in the concentrations
of the air contaminants measured Both modeling approaches have been utilized in
evaluating indoor concentrations of NO2
7.3.6.1 Physical/Chemical Models
The physical/chemical modeling approach has been used by a number of investigators
in chamber, test-house, and small field studies (involving a small number of homes) to
estimate emission rates of NO2 from combustion sources (e g , Traynor et al , 1982,
Moschandreas et al, 1984, Leaderer, 1982, Brauer et al, 1990), to estimate reactive decay
rates (e g , Yamanaka, 1984, Borrazzo et al, 1987a,b, Leaderer et al, 1986a,b, Spicer
et al, 1986, 1989, Ozkaynak et al, 1982), to estimate the impact of ventilation and mixing
on the spatial and temporal distribution of NO2 (e g , Borrazzo et al, 1987a; Ozkaynak
et al, 1982, Traynor et al, 1982), and to evaluate the applicability of emission rates
determined under controlled conditions in estimating indoor concentrations of NO2 (e g ,
Traynor et al, 1982, Borrazzo et al, 1987a) The physical/chemical approach is the bases
for a number of models or subcomponents of larger exposure models that calculate the air
concentrations of a single- or multrroom building (e g , Austin et al, 1988, Hayes, 1989,
Sparks, 1988, Nazaroff and Cass, 1986) More recently, three studies have been reported
that utilize distributions of the input variables to the mass balance equation (emission rates,
source use, decay rates, ventilation rates, etc ) determined from the published literature to
estimate the distributions of NO2 levels indoors for specific sources and combinations of
sources (Traynor et al, 1987, Hemphill et al, 1987, Drye et al, 1989) Use of the
7-77
-------
physical/chemical models to evaluate model input parameters (e g , source strength) in
explaining measured indoor levels of NO2, user-friendly computer models for piedicting NO2
for specific indoor settings and conditions, and efforts to use physical/chemical models to
estimate concentration distributions will be touched upon here Emphasis is placed upon
physical/chemical models that are used to predict population distributions of NO2
Borrazzo et al (1987a) applied a mass-balance model to NO2 levels measured in a town
house with gas appliances Nitrogen dioxide emission rates were determined from a portable
sampling hood, reactive decay rates were determined from a comparison of NO2 and sulfur
hexafluoride (SF6), and infiltration rates were determined from SF6 decay rates Comparing
model predictions with measured concentrations yielded a difference of 28 % for NO2
Differences in NO2 emission rates over tune of use of gas appliances and breakdown of the
well-mixed single-compartment model assumption were thought to account for the
discrepancies in the predicted versus measured concentrations In a study of the effects of
ventilation on residential air pollution from a gas-fired range (Traynor et al, 1982), a gas
range tested in a series of chamber studies was used in a test house and measured NOX levels
were compared to those predicted by a mass-balance model In this study, infiltration, gas
consumption, and NO2 reactivity rates were measured The results indicated good overall
agreement between the measured and predicted NOX levels over the full test periods,
although discrepancies in predicited and measuied concentration were observed in the start-up
phase of the sources The authors note that the main deficiency in the model is the
assumption that the house is a single cell and, as such, does not address the spatial variation
in concentrations.
In recent years a number of user-friendly indoor air quality models, based upon
physical/chemical approaches, have been developed (e g , SHAPE, PAQM) to evaluate the
impact of indoor sources and outdoor contaminant levels on air contaminant levels in single
rooms or multiple rooms (e g., Austin et al, 1988, Hayes, 1989, Sparks, 1988, Nazaroff
and Cass, 1986) These models are frequently a component of a larger model used to predict
personal exposure A brief review of these models is provided by Weil et al (1990) The
models utilize estimated or measured values for whole-house air exchange rates, mixing
within and between rooms, source strengths and use patterns, and contaminant removal
and/or chemical transformations to predict indoor concentrations over varying time frames
7-78
-------
The uncertainty associated with the input parameters and the use of these models to predict
indoor NO2 concentrations in single residences is not known No systematic effort has been
made to validate these models for predicting NO2 levels in single residences
Traynor et al (1987) have reported on efforts to develop a macromodel for assessing
indoor concentrations of combustion-generated pollutants The model is a single-chamber,
well-mixed, mass-balance model (Equation 7-1) that utilizes experimentally derived estimates
of emission factors, building penetration factors, and reactivity rates in combination with
existing regional and national data (e g , house volumes, market penetration of unvented
3
combustion sources) and source usage and infiltration models to estimate indoor pollutant
concentration distributions Deterministic and Monte Carlo simulation techniques are used to
combine all of the inputs to yield the concentration distributions The macro model will be
used to estimate the distribution of NO2 concentrations indoors In a parallel development of
a statistical and physical/chemical model, Hemphill et al (1987) developed a stochastic
model based on the physical model to predict indoor NO2 concentration distributions in
homes using UVGSHs as the primary source of heat
Billick et al (1988) used a macro modeling approach similar to that used by Traynor
et al (1987) This simulation model used parameters measured in field and laboratory
studies as inputs to construct distributions of air exchange rates, reactive decay rates, outdoor
NO2 levels, gas range NO2 emission rates, and gas range useage The model was used to
generate indoor distributions of NO2 for homes with gas or electric ranges for winter and
summer The computed distributions were then compared to measured concentrations in a
sample of over 600 homes in Southern California The measured and estimated NO2 means
differed by 5% for the summer For the winter,, modeled values of the means were 10%
lower than the measured values for gas range homes and 25 % lower for electric range
homes
An extensive data base on indoor levels of NO2 associated with unvented gas space
heating was reported by Koontz et al ,(1986, 1988) (see Section 7342) The authors
utilized the physical/chemical model to predict indoor NO2 levels and to compare the results
to indoor levels predicted from a regression analysis of the measured data The
physical/chemical model was used to predict steady-state concentrations for four cases
(no indoor sources, pilot lights only, pilot lights plus cooking, and pilot lights plus cooking
7-79
-------
and heating) The authors incorporated into the model a term for heat demand for a UVGSH
to maintain a given indoor/outdoor temperature differential Emission rates were derived
from the literature, source use and house volumes were measured, air exchange rates were
estimated, and a constant decay rate was assumed The model predicts a broad distribution
of indoor NO2 concentrations associated with use of UVGSHs where indoor/outdoor
temperature differentials are the best predictors This study also found good agreement
between the mass-balance model and a regression model developed from the collected data
The authors concluded that the steady-state physical/chemical model provides reliable
estimates for annual average NO2, but may underestimate the frequency of occurrence of
peak concentrations Statistical results to assess the accuracy of the model weie not
presented.
The use of physical/chemical (mass-balance) models (single compartment) to predict
indoor concentrations of NO2 indoors or distributions of concentrations in homes with
combustion sources requires accurate information on the input parameters Although data are
available for some of the input parameters under controlled experimental conditions (e g ,
emission rates), there are very limited data available on the variability of the input
parameters in actual homes or the factors that control the variability of those inputs (e g ,
variability of emission or decay rates) Obtaining field measurements or estimates of the
inputs in large numbers of homes would be expensive and time consuming. Such modeling
efforts, however, do help to identify the potential range of indoor NO2 concentrations and
factors that may result in high levels and the potential effectiveness of mitigation efforts
7.3.6.2 Statistical/Empirical Models
Field studies that have measured NO2 concentrations in residences and associated
outdoor levels for time periods of a week or more have typically obtained questionnaire
information on sources in the residences, source use, building characteristics (house volume,
number of rooms, etc ), building use, and meteorological conditions In some cases,
additional measurements such as temperatures have been recorded Several investigators
have attempted to fit simple regression models to their field-study data bases in an effort to
determine if the variations in NO2 levels seen among houses can be explained by variations
in the questionnaire responses and any additional measurements that may have been taken
7-80
-------
The goal has been to see how well questionnaire information or easily available information
(meteorological data) can predict indoor NO2 levels In most cases, a linear model has been
used, but several investigators have used log transformations of variables Table 7-23
presents a summary of the regression models that have been fitted to large field-study data
2
bases The independent variables entered into the analysis (p < 0 05), the R , and standard
error of the estimate reported by the investigators are shown in the table No standard errors
were reported for a number of the models, although several investigators reported standard
errors for the independent variables in their models Linear regression models, with the
exception of the Petreas et al (1988) model, explain from 40 to 70% of the variations in
residential NO2 levels and typically have large standard errors associated with their
estimates Although the log transformations of variables have always produced a higher
percent of explained variation due to the skewed distribution of the original variables,
interpretation of the coefficients in a nonlinear model can require special attention
The independent variables reported as being significant in each model are broken down
into four general categories in Table 7-23 (1) sources, (2) source use, (3) removal/dilution,
and (4) interactive terms The only independent variables that are common to all models are
those that deal with the identification of sources in the residences and outdoor concentrations
The identification of the sources accounted for the major portion of the explained variation in
indoor NO2 levels for all models Those models that incorporate source-use information or
proxies for source-use generally produce better fitting models Only one model, developed
from an extensive data base (Wilson et al, 1986), found a number of variables related to
the removal or dilution of NO2 indoors Three models found independent variable
interactive terms to be significant
Butler et al (1990) reported on the regression model developed from the most extensive
data base (up to 1,952 observations) It is an extension of a previous effort (Drye et al,
1989), but differs from it in that a larger data base is used, data from electric ranges are also
factored in, and additional independent variables are considered The data used are from
eight different locations and are drawn from several studies the Beals et al (1987) study,
the Indoor Air Quality Characterization Study (Wilson et al, 1986), the Boston Residential
NO2 Characterization Study (Ryan et al, 1988), and the Harvard Six City Study (Ferns
et al, 1979) The authors reported conducting a verification test (results not shown) of the
7-81
-------
TABLE 7-23. EMPIRICAL STATISTICAL MODELS (REGRESSION) FOR RESIDENTIAL NITROGEN DIOXIDE
CONCENTRATIONS REPORTED FROM FIELD STUDIES OF INDOOR LEVELS3
oo
ts)
House
Location
Bedroom
Kitchen
Bedroom
Kitchen
Living room
Activity room
Kitchen
Bedroom
Center of
house
Kitchen
Living room
Kitchen
Bedroom
Number of
Observations
400 to
578
318
114
215 to
262
29 to
82
173
1782 to
1952
R2
045 to
063
057 to
066
069
031 to
039
040 to
069
059 to
068
040 to
076
SE (ppm)
Sources
(0012 to 0022) Outdoor N02
Gas range/oven
Gas floor furnace
Range pilots
Oven pilots
Gas water heater
Age of gas oven
(0 010 to 0 01 1) Outdoor NO2
Convect /radiant
kerosene heater
Gas range/oven
Other gas appl
Cig smoking
(0075) Outdoor NO2
Gas stove with pilot light
Gas stove w/o pilot light
Gas dryer
Floor/wall furnace
Use of toaster/microwave
in gas stove house
— Gas/cooking fuel
Location variable
Cigarette smoking
— Unvented gas space heater
Condition of furnace
Number of pilot lights
Outdoor NO2
— Outdoor NO2
Presence of gas geyser
Cooking fuel
Type of space heating
Log of NO2 in rooms
— Outdoor NO2
Gas range pilot lights
— Microwave oven
Source Use
Oven use
Number of occupants
Oven cleaning
Max temp
Outdoor temp
Convect /radiant
kero heater use
Gas/oven use
Number of
cigarettes
Income level
Difference between
indoor/outdoor temp
Exposure to high wind
Use of gas range for heat
Presence or
absence
Removal/Dilution Interactive Terms
Total house volume Pilot lights,
Air exchange rate ventilation volume,
Open windows outdoor NO2
Bedroom area
House age
Number of bedrooms
Gas range hood use
Number of fireplaces Convective kerosene
heater use squared
,
House volume
Bedroom window open
Gas range exhaust fan
Kitchen volume
Number of doors
House volume
Multifarmly
Reference
Wilson et al
(1986)
Leaderer et al
(1987)
Marbury et al
(1988)
Petreas et al
(1988)
Koontz et al
(1986)
Noy et al
(1984)
Butler et al
(1990)
SE = Standard error NC^ = Nitrogen dioxide
-------
model by comparing modeled results to values measured in Watertown, MA, and reported
that in general, there was good correspondence between the modeled and observed values
Tests of the earlier version of this model (Drye et al , 1989) indicated good model
performance
The model developed from the second most extensive data base on NO2 levels indoors
associated with gas appliances (Beals et al, 1987) only explained approximately 60% of the
variation in indoor levels, with standard errors in the range of 40 jug/m3 (0 021 ppm) There
is little uniformity among the models in the form of the source use and removal/dilution
terms or their significance in the models
Regression models developed from field studies employing questionnaires to explain
variations in indoor levels of NO2 have met with only moderate success Better information
through additional measurements and better questionnaire design, is needed on source type
and condition, source use, contaminant removal (infiltration and reactive decay), and
between- and among-room mixing if the statistical/empirical models are to be used to
estimate indoor concentrations of NO2 in homes without measurements The unexplained
variance may be a function of factors that are difficult to address by questionnaire, such as
actual reactive decay in a home
7.3.7 Reactive Decay Rate of Nitrogen Dioxide Indoors
A number of field studies of NO2 levels in residences have reported that NO2 is
removed more rapidly than can be accounted for by infiltration alone (Wade et al, 1975,
Macnss and Elkins, 1977, Ozkaynak et al, 1982, Ryan et al, 1983, Traynor et al, 1982,
Leaderer et al , 1986a) Nitrogen dioxide indoors is removed by infiltration/ventilation and
by interior surfaces and furnishings The removal of NO2 by interior surfaces and
furnishings and reactions occurring in air is often referred to as the reactive decay rate of
NO2 Failure to account for the reactive decay rate (K in Equations 7-2 and 7-3) can
(1) lead to a serious underestimation of emission rate measurements in chamber and test-
house studies and a serious overestimate of indoor concentrations when using emission rates
to model indoor levels and (2) be a significant factor in the actual NO2 levels measured in
residences The NO2 reactive decay rate is typically determined by comparing the decay of
NO2, after a source is shut off, to that of a relatively nonreactive gas (e g , carbon monoxide
7-83
-------
[CO], carbon dioxide [COJ, SFg) The measured reactive decay rates in the above-
mentioned field studies typically ranged from 0 1 to 1 6 air changes per hour All studies
noted that the reactive decay of NO2 is as important and in some cases, more important than
infiltration in removing NO2 indoors Leaderer et al (1986a), in the continuous monitoring
of NO2, NO, CO, and CO2 in seven houses over periods ranging from 2 to 8 days, reported
that the NO2 decay rate was always greater than that due to infiltration alone and was highly
variable among houses and among tune periods within a house
In an effort to identify the factors that control the NO2 reactive decay rate, a number of
small chamber (Miyazaki, 1984, Spicer et al, 1986), large chamber (Moschandreas et al,
1985; Leaderer et al, 1986b), and test-house studies (Yamanaka, 1984, Borrazzo et al,
1987a,b; Fortmann et al, 1987) have been conducted The most extensive small chamber
work is reported by Spicer et al (1986), in which 35 residential materials were screened for
o
NO2 reactivity in a 1 64-m chamber, and in which a limited number of the materials were
tested for the impact of relative humidity on the reactivity rate Figure 7-29 shows the
relative rates of NO2 removal for the materials screened The figure indicates that many of
the materials used for building construction and furnishings are significant sinks for NO2 and
that their removal rate is highly variable Many of the matenals weie found to reduce a
significant fraction of the removed NO2 to NO In no cases was NO2 remitted, although
some materials emitted NO The authors noted that the matenals that removed NO^ most
rapidly fall hi two categories porous mineral matenals of high surface area and cellulosic
material denved from vegetable matter Higher relative humidities were found to enhance
the removal rate for some matenals (e g , wool carpet), reduce the removal rate for some
(e.g., cement block), and have little effect on others (e g , wallboard) In a subset of
experiments (Spicer et al, 1989), the mechanisms by which selected matenals removed NO2
were investigated In these experiments, known amounts of NO2 were passed over a packed
bed of granular or shredded matenals Nitrogen dioxide and NO2 reaction products were
measured and mass balances calculated The results indicated different removal mechanisms
for NO2, including removal and retention by matenals, reactions with adsorbed water
producing NO and HONO, and chemical reactions with organic constituents of the test
matenal
7-84
-------
01234567 8
Cement Block
Wool Carpet
Brick (Used)
Masomte
Cotton/Polyester Bedspread
Painted (Fiat Latex) Wallboard
Plywood
Acrylic Fiber Carpet
Nylon Carpet
Vinyl Wall Covering (Paperbacked)
Ceiling Tile
Polyester Carpet
Acrylic Carpet
Furnace Filters (New)
Dehumidrfier
Oak Paneling
Vinyl-Coated Wallpaper
Particle Board
Furnace Filters (Used)
Ceramic Tile
Wool (80%) Polyester (20%) Fabnc
Cotton Tern/doth
Spider Plants (With Soil Covered)
Walltex Covering
Waxed Asphalt Tiles
Window Glass
Used Furnace Heat Exchanger
Formica Counter Top
Polyethylene Sheet
Asphalt Floor Tiles
Vinyl Floor Tile
Galvanized Metal Duct
Plastic Storm Windows
01234567 8 9
Rate Constant for N0£ Removal (1/h)
Figure 7-29. Bar graph of nitrogen dioxide removal rate for various materials evaluated
in a 1.64-m3 test chamber at 50% relative humidity.
Source Spicer et al (1986)
7-85
-------
o
In a senes of small (0 69-m ) chamber studies (Miyazaki, 1984), reactive decay rates
for NO2 were found to vary as a function of material type and to increase with increasing
surface area of the material, degree of stirring in the chamber, temperature, and relative
humidity A saturation effect was noted on some of the carpets tested
*»
In a senes of large (34-m ) chamber studies, Leaderer et al (1986b) evaluated the
reactive decay rate of NO2 as a function of material type, surface area of material, relative
humidity, and air mixing (Figure 7-30) The reactive decay rate was found to vary as a
function of material surface roughness and surface area Carpeting was found to be most
effective in removing NO2, and painted wallboard was least effective Increases in relative
humidity were associated with increases in removal rates for all materials tested, but the
slope was a shallow one Of particular interest is the finding in this study that the degree of
air mixing and turbulence was a dominant variable in determining the reactive decay rate for
NO2. Moschandreas et al (1985) evaluated six materials in a 14 5-m chamber and found
variations in decay rates by material types and a positive impact on NO2 decay rates in an
empty chamber by relative humidity
Yamanaka (1984) assessed NO2 reactive decay rates in a Japanese living room and
found the decay to be comprised of both homogeneous and heterogeneous processes The
rates were found to vary as a function of surface property and sharply as a function of
relative humidity. During the decay, NO production was noted In a test-house study,
Fortmann et al (1987) noted that the NO2 decay rate tends to decrease as the concentration
increases It is not clear whether this is due to surface saturation or second-order kinetics
This study also noted a sharp increase in NO levels during the NO2 decay, indicating NO
production as a result of the NO2 decay In a test-house study conducted over a 7-mo
period, Borrazzo et al (1987a) found that reaction rates for NO2 in the test house were
sensitive to the location in the house where they were measured This indicates that reaction
losses during transport of NO2 from room to room in a house may be important
The reactive decay of NO2 in residences associated with interior surface materials and
furnishings is an important mechanism for removing NO2 in residences Nitrogen dioxide
reactive decay rates vary as a function of the type of material and surface area of the
material The impact of relative humidity on the decay rate is unclear, with some studies
showing a pronounced impact (Yamanaka, 1984) and others showing moderate or little
7-86
-------
175
150
125
100
075
050
025
Material (area, m )
• Painted board (53)
o Painted board (29)
• Wallpaper (48 3)
n Wallpaper (24 2)
A Rug (29) i
A Rug (14 5)
r-048
s-0005
r = 05
s-00034
r-083
s-00048
10 20 30 40 50 60
Relative Humidity (%)
70
80
90
Figure 7-30. The deposition rates in air changes per hour for nitrogen as a function of
percent relative humidity for two suiface areas of three materials.
Source Leaderer et al (1986b)
impact (e g , Spicer et al, 1986, Leaderer et al, 1986b) The degree of air mixing or
turbulence can have an important effect on the reactive decay rate A by-product of NO2
removal by materials is NO production and a saturation effect may occur for some materials
Reactive decay of NO2 in residences is highly variable between residences, within rooms in a
residence, and on a temporal basis within a residence The large number of variables
controlling the reactive decay rate makes it very difficult to assess in large field studies
through questionnaire or integrated air sampling
7-87
-------
7.4 NITRIC AND NITROUS ACIDS CONCENTRATIONS
Nitric acid and HONO may be formed in the gas phase during combustion and by
heterogenous hydrolysis of NO2 The rate of formation of nitrogen acids, particularly
HONO, is expected to vary with the NO2 concentration, humidity and temperature, light
intensity, and the various surfaces present in a home (Spengler et al, 1993)
Brauer et al (1991) used annular denuder filter pack sampling systems to measure the
gaseous pollutants HNO3, HONO, NO2, and NH3 during summer and winter periods in
Boston, MA Five homes were sampled during the winter period and SDC were sampled
during the summer. All homes in the winter period had gas ranges During the summer
period, four of the six homes had no unvented gas appliances Indoor samples were placed
in a room adjacent to the kitchen and at a height of approximately 15m above the floor
Outdoor samples were placed 3 to 5 m from the home Outdoor levels of HNO3 exceeded
indoor concentrations during both seasons (Figure 7-16b) Lower indoor concentrations are
due to the lack of measurable indoor production and the high surface reactivity of HNO^
The authors attribute the higher indoor levels in the summer to higher outdoor summer levels
of HNO3 and higher infiltration rates in the summer The authors conclude that the major
source of HNO3 indoors is outdoor concentrations In a recent paper, Weschler et al (1992)
argue that in the summer an indoor reaction between O3 and NO2 can be a significant source
of HNO3 and peroxy radicals High air-exchange rates could produce levels of O3 and NO2
high enough for such a reaction to be significant The major pathway for formation of
HNO3 indoors may be the NO3 abstracting a hydrogen atom from vapor-phase organic
compounds. A review of data collected by Brauer et al (1991) and Weschler et al (1992)
supports this conclusion
It is not known whether exposure to gaseous HONO is associated with health effects
An effective nitrosating agent, HONO may react with gaseous secondary amines in air to
form mtrosamines, which have been shown to be carcinogenic in animals (Pitts et al, 1978,
Magee, 1982). In Brauer et al (1991), indoor levels of HONO were higher indoors than
outdoors for both winter and summer for all of the homes (Figure 7-31), even those without
unvented gas appliances where indoor levels of NO2 were lower than outdoor levels Indooi
and outdoor levels of HONO were found to be correlated with NO2 concentrations Winer
and Bierman (1991) report ambient short-term (15-min) HONO levels of 7 ppb in Claremont,
7-i
-------
12 '
I 10-
1 8'
1 6-
o 4
O n
i 2-
0 .
90%
VKO/.
/O/O
• mean
c/w
-_ 25%
10%
(A) MONO
S"
o.
-a
c
i
1
o
O
0
"3L
bdjid ' | • |
3 -i
25
2
1 5 -
.
1 .
.
0
(B)
g
I
HN03
B
.-*-.
Outdoor Indoor Outdoor Indoor
N-29 N-31 N-24 N-29
SUMMER WINTER
Outdoor Indoor Outdoor Indoor
N-29 N-31 N-24 N-29
SUMMER WINTER
Figure 7-31. Concentration distributions (in ppb) for gas-phase species in Boston:
(A) nitrous acid and (B) nitric acid (N = number of valid observations).
Source Brauer et al (1991)
CA, and 16 ppb in Long Beach, CA The highest levels occurred just before sunrise
Nighttime concentrations between 2 and 4 ppb were not uncommon ui Los Angeles
Spengler et al (1993) studied levels of NO2 and HONO indoors in 10 homes in
Albuquerque, NM The indoor 24-h mean HONO concentration ranged from 2 to 8 ppb
Indoor HONO concentrations were found to be well correlated with indoor NO2 levels,
HONO concentrations ranged from 5 % to 15 % of the measured NO2 concentrations
Indoor concentrations of HONO appear to be higher indoors than outdoors, even when
indoor concentrations of NO2 are do not exceed outdoor levels A possible mechanism for
this is the heterogeneous reaction of NO2 with water (Sakamaki et al, 1983, Pitts et al,
1984, Svensson et al, 1987; Jenkin et al, 1988, Brauer et al, 1990, 1991). In homes
where unvented combustion sources are used, elevated HONO levels may be associated with
direct emission of HONO from the flame as well as with heterogeneous reactions with water
of the produced NO2 (Pitts et al, 1989, Brauer et al, 1990, 1991)
7-89
-------
7.5 SUMMARY
7.5.1 Ambient Nitrogen Dioxide Levels
Nitrogen oxides concentrations in isolated rural sites and coastal inflow areas in the
United States generally range from a few tenths to 1 ppb The concentrations in the
atmospheric boundary layer and lower free troposphere in remote maritime locations are in
the range 0 02 to 0.04 ppb, and concentrations of NOX in remote tropical forests have been
reported to range from 0 02 to 0 08 ppb (Kelly et al, 1982, Lefohn et al, 1991, National
Research Council, 1991)
Analysis of NO2 data in the AIRS data base for the period 1981 to 1990 indicates a
downward trend for the composite United States annual average NO2 concentration The
1990 composite NO2 average was 8% less than the 1981 average, and the difference was
statistically significant (AIRS, 1992)
The highest hourly and annual ambient NO2 levels are reported from stations in
Southern California, where the current annual standard of 0 053 ppm has been exceeded
The seasonal patterns at California stations are usually quite marked and reach their highest
levels during the fall and winter months For most of the other urban sites characterized, the
highest monthly average NO2 concentrations also were obtained in the fall or winter months
(U.S. Environmental Protection Agency, 1991a,b)
The diurnal patterns of NO2 for the urban sites showed that, on the average, the highest
concentrations occur in the late afternoon and evening hours (1700 to 2200 hours) For those
urban areas experiencing hourly NO2 concentrations > 0 2 ppm, the episodic occurrences are
experienced usually in the midmorning and afternoon/evening hours (AIRS, 1991)
Based on data collected at rural locations for the period 1979 to 1991, the hourly
average NO2 concentrations for selected U S forest and agricultural sites were < 0 10 ppm
in most cases. As observed for urban locations, a consistent seasonal pattern was
distinguishable for both the rural forested and agricultural sites In general, the NO2
monthly average values were at their highest during the fall and winter months A consistent
diurnal pattern was also observed for the rural forested and agricultural sites, late afternoon
and evening hours (approximately 1700 to 2200 hours) contained the highest NO2
concentrations There were some exceptions to these patterns (AIRS, 1992)
7-90
-------
Studies characterizing the joint occurrence of gaseous NO2/SO2 and NO2/O3 have
demonstrated that (1) the co-occurrence of two-pollutant mixtures lasted only a few hours per
episode, and (2) the time between episodes is generally long (i e, weeks, sometimes months)
(Lefohn and Tingey, 1984) The periods of co-occuicrence represent a small portion of the
potential plant growing period For human ambient exposure considerations, in most cases,
the simultaneous co-occurrence of NO2/O3 was infrequent However, for several sites
located in the Southern California South Coast Air Basui, more than 450 simultaneous
co-occurrences of each pollutant at hourly average concentrations equal to or greater than
0 05 ppm have been reported Besides considering the joint occurrence of gaseous
pollutants, it may be advisable to consider the joint occurrence of O3 with nitrogen via dry
deposition in forested landscapes (Taylor et al, 1992)
The average concentrations of HNO3 and NO3" are generally in the range 0 1 to 20 ppb
and 0 1 to 10 ppb, respectively (AUegnni and De Santis, 1989, Wolff et al , 1986a,b, 1991,
Kelly et al, 1982, 1984) Because there are conflicting reports on the ability of filters to
accurately separate HNO3 from NO3" aerosol, it may be more appropriate in some cases to
focus on the total NO3 (HNO3 + NO3~) than on the individual components
7.5.2 Indoor Nitrogen Dioxide Levels
Indoor concentrations of NO2 are a function of outdoor concentrations, indoor sources
(source type, condition of source, source use, etc ), infiltration/ventilation, arr mixing within
and between rooms, reactive decay by interior surfaces, and air cleaning or source venting
In homes without indoor sources of NO2, concentrations are lower than outdoor levels due to
removal by the building envelope and interior surfaces, thus providing some degree of
protection from outdoor concentrations Indoor/outdoor ratios for homes without sources
vary considerably by season of the year, with the lowest ratios occurring in the winter and
the highest occurring during the summer Considerable variability in the ratios within a
season exists The differences are probably due to seasonal differences in infiltration rates,
NO2 reactivity rates, penetration factors, and outdoor concentrations (Leaderer et al, 1986a)
Gas appliances (gas range/oven, water heater, etc ) are the major indoor source
category for indoor residential NO2 by virtue of the number of homes with such sources
(approximately 45% of all homes in the United States) (U S Bureau of the Census, 1982)
7-91
-------
Nitrogen dioxide levels in homes with gas appliances are higher than those without such
appliances and are often higher than levels encountered outdoors Within this category, the
gas range/oven is a major contributor, especially when used as a supplemental heat source
Average indoor concentrations in bedrooms (over a 1- to 2-week measurement period) range
3
from 20 to 120 jttg/m (0 010 to 0 064 ppm) in some homes with gas ranges Homes with
gas ranges with pilot lights have higher NO2 levels than homes that have gas ranges without
pilot lights (Wilson et al, 1986, Quackenboss et al, 1986, Leaderer et al, 1986a, Melia
et al., 1982; Butler et al, 1990, Neas et al, 1991)
Average NO2 concentrations in homes with gas ranges/ovens exhibit a spatial gradient
within and between rooms Kitchen levels are higher than other rooms and a pronounced
vertical concentration gradient in the kitchen has been observed in some homes, with
concentrations being highest nearest the ceiling (Goldstein et al, 1985, Wilson et al, 1986)
Average NO2 concentrations are highest during the winter months and lowest during the
summer months This seasonal temporal gradient is probably related to seasonal differences
ML infiltration, source use, NO2 reactivity rates indoors, and outdoor concentrations (Spengler
et al., 1983; Wilson et al, 1986) The impact of gas appliance use on indoor NO2 levels
may be superimposed upon the background level resulting from outdoor concentrations The
results of field studies of the impact of gas ranges on indoor NO2 are fairly consistent Once
corrected for the contribution of outdoor concentrations, the average contribution of gas
ranges to NO2 levels indoors is similar by locations within homes and by season across the
studies There is, however, considerable variability associated with these averages
Very limited data exist on short-term (3-h or less) average indoor concentrations of
NO2 associated with gas appliance use Harlos et al. (1987) reported a 1-h mean maximum
<1
of346/*g/m (0.18 ppm) in a study of 18 volunteers The limited data suggest that short-
term indoor averages of NO2 are higher than those recorded for outdoors
Unvented kerosene and gas space heaters are important sources of NO2 in homes both
because of the NO2 production rate of the heaters and the length of time they are used
Field studies indicate that average residential concentrations (1- or 2-week average levels)
exhibit a wide distribution, varying primarily with the amount of heater use and type of
heater. Average concentrations in homes using unvented kerosene heaters have been
measured well in excess of 100 /tg/m (0.053 ppm) In one study, calculations of NO2
7-92
-------
residential levels during heater use (in homes without gas appliances) indicate that
o
approximately 50 % of the homes had concentrations above 100 jttg/m (0 053 ppm) and 8 %
O
had concentrations above 480 /ttg/m (0 25 ppm) (Leaderer et al, 1986a) Nitrogen dioxide
2
levels of 847 jitg/m (0 45 ppm) over a 1-h period in a home during use of a kerosene heater
have been measured A large field study of mdoor NO2 concentrations in homes using
UVGSHs (most also had gas ranges) found that approximately 70% of the homes had
average concentrations in excess of 100 /-cg/m3 (0 053 ppm) and 20% had levels in excess of
480 /ig/m3 (0 025 ppm) (Koontz et al, 1986, 1988) This study found that the
indoor/outdoor temperature difference was the best indicator of indoor NO2 levels during the
colder winter periods when heating demands are greatest The highest concentration
-3
recorded for a home with a UVGSH was 2,716 ug/m (1 44 ppm) (15-min averaging period)
(Koontz et al, 1988) The higher concentrations may be the levels encountered for sustained
periods of time (e g , hours) because they may approximate the equilibrium levels under
periods of heater use No pronounced spatial gradient of NO2 was found in homes with
unvented kerosene space heaters, contrary to the strong spatial gradient noted for homes with
gas appliances. This is probably due to the strong convective heat output and the long
operating hours of the heaters, which result in rapid mixing within the homes
Improper use of gas appliances (e g , using a gas range to heat a Irving space) and
improperly operating gas appliances or vented heating systems (e g , out of repair gas range
or improper operation of a gas wall or floor furnace) can be important contributors to indoor
NO2 concentrations, but little data are available to assess the extent of that contribution
(Deals et al, 1987) Data have not been reported that would allow for an assessment of the
contributions of wood- or coal-burning stoves or fireplaces to indoor NO2 concentrations, but
such a contribution would be expected to be small Cigarette smoking is expected to add
little NO2 to homes (Wilson et al, 1986, Leaderer et al, 1986a, Good et al, 1982)
Efforts to model indoor NO2 levels have employed both (1) physical/chemical and
(2) empirical/statistical models Physical/chemical models have largely been applied to test-
house data or to small samples of homes where detailed data on the factors impacting the
concentrations have been measured These models have been used, with varying success in
explaining measured indoor levels of NO2, in user-friendly computer models (Weir et al,
1990) for predicting NO2 for specific indoor settings, and to estimate indoor concentration
7-93
-------
distributions (Traynor et al, 1987, Hemphill et al, 1987, Billick et al, 1989) Various
empirical/statistical models have also been developed from large field-study data bases
These employ questionnaire responses and measured physical data (house volume, etc ) as
key independent variables and have met with moderate success (Butler et al, 1990, Drye
et al., 1989; Wilson et al, 1986).
The removal of NO2 indoors by surfaces (reactive decay) is often equal to or greater
than infiltration in removing NO2. Nitrogen dioxide reactive decay rates vary as a function
of the type of material and surface area of the material The degree of mixing or turbulence
in a space is also important, as is relative humidity A by-product of NO2 removal by
materials is NO production and a saturation effect may occur for some materials Reactive
decay of NO2 in residences is highly variable between residences, within rooms in a
residence, and on a temporal basis within a residence (Spicer et al, 1986, Leaderer et al,
1986b; Borrazzo et al, 1987b).
Indoor concentrations of HONO appear to be higher than outdoors, even when indoor
NO2 concentrations do not exceed outdoor levels A possible mechanism for this is the
heterogeneous reaction of NO2 with water In homes where unvented combustion sources
are used, elevated HONO levels may be associated with direct emissions of HONO from the
flame as well as with heterogeneous reactions of the produced NO2 with water Nitric acid
has been measured indoors during a summer period at concentrations lower than ambient
levels Indoor production of HNO3 has been postulated (Brauer et al, 1991)
7-94
-------
REFERENCES
AIRS, Aerometnc Information Retrieval System [database] (1991) [Data on NOX] Research Triangle Park, NC
U S Environmental Protection Agency, Office of Air Quality Planning and Standards Disc, IBM 3090
AIRS, Aerometnc Information Retrieval System [database] (199 2) [Data on NOX] Research Triangle Park, NC
U S Environmental Protection Agency, Office of Air Quality Planning and Standards
Allegnni, I, De Santis, F (1989) Measurement of atmospheric pollutants relevant to dry acid deposition Grit
Rev Anal Chem 21 237-255
Atkins, D H F , Law, D V (1987) Indoor-outdoor nitrogen dioxide concentration ratios for homes with gas
and electric cooking In Seifert, B , Esdorn, H , Fischer, M , Rueden, H , Wegner, J , eds Indoor air
'87 proceedings of the 4th international conference on indoor air quality and climate, v 1, volatile
organic compounds, combustion gases, particles and fibres, microbiological agents, August, Berlin,
Federal Republic of Germany Berlin, Federal Republic of Germany Institute for Water, Soil and Air
Hygiene, pp 383-389
Austin, B S , Rosenbaum, A S , Hayes, S R (1988) User's guide to the NEM/SAI exposure model
San Rafael, CA Systems Applications, Inc , SYSAPP-8S/051
Beals, S A , Holiman, J C , Kubo, R , Rubio, S A , Stanford, R , Colome, S D , Wilson, A L (1987)
Residential indoor air quality characterization study of nitrogen dioxide Phase n final report the wall
and floor furnace inspection study Los Angeles, CA Southern California Gas Company
Billick, I H (1991) An update of the natural gas industry's research related to indoor air quality Presented at
American Gas Association 1991 distnbution/transmission conference, April-May, Nashville, TN
Billick, I H , Baker, P E , Colome, S D (1989) Simulation oi indoor nitrogen dioxide concentrations
In Harper, J P , ed Combustion processes and the quality of the indoor environment transactions, of an
international specialty conference, September 1988, Niagara Falls, NY Pittsburgh, PA Air & Waste
Management Association, pp 151-172 (A&WMA transactions^ series TR-15)
Bellinger, M J , Hahn, C J , Parrish, D D , Murphy, P C , Albntton, D L , Fehsenfeld, F C (1984) NOX
measurements in clean continental air and analysis of the contributing meteorology J Geophys Res
[Atmos ] 89 9623-9631
Borrazzo, J E , Osborn, J F , Fortmann, R C , Keefer, R. L , Davidson, C I (1987a) Modeling and
monitoring of CO, NO and NO2 in a modern townhouse Atmos Environ 21 299-311
Borrazzo, J E , Peters, C , Peck, S , Davidson, C I (1987b) Determination of NO2 loss rates from
concentration measurements in an occupied urban residence In Seifert, B , Esdorn, H , Fischer, M ,
Rueden, H , Wegner, J , eds Indoor air '87 proceedings of the 4th international conference on indoor
air quality and climate, v 1, volatile organic compounds, combustion gases, particles and fibres,
microbiological agents, August, Berlin, Federal Republic of Germany Berlin, Federal Republic of
Germany Institute for Water, Soil and Air Hygiene, pp 321-325
Brauer, M , Ryan, P B , Suh, H H , Koutratas, P , Spengler, I D , Leslie, N P , Billick, I H (1990)
Measurements of nitrous acid inside two research houses Environ Sci Technol 24 1521-1527
Brauer, M , Koutrakis, P , Keeler, G J , Spengler, J D (199 Tl Indoor and outdoor concentrations of inorganic
acidic aerosols and gases J Air Waste Manage Assoc 41 171-181
7-95
-------
Butler, D A , Ozkaynak, H , Billick, I H , Spengler, J D (1990) Predicting indoor NC^ concentrations as a
function of home characteristics and ambient NO2 levels In Indoor air '90 precedings of the 5th
international conference on indoor air quality and climate, volume 2, characteristics of indoor air,
July-August; Toronto, ON, Canada Ottawa, ON, Canada International Conference on Indoor Air
Quality and Climate, Ihc , pp 519-524
Cadle, S H (1985) Seasonal variations in mtnc acid, nitrate, strong aerosol acidity, and ammonia in an urban
area. Atmos. Environ 19 181-188
Cadle, S H , Mulawa, P A (1988) Atmospheric summertime concentrations and estimated dry deposition rates
of nitrogen and sulfur species at a Smoky Mountain site in North Carolina Presented at 81st annual
meeting of the Air Pollution Control Association, June, Dallas, TX Pittsburgh, PA Air Pollution
Control Association, paper no 88-119 3
Cadle, S H.; Countess, R J , Kelly, N A (1982) Nitric acid and ammonia in urban and rural locations
Atmos Environ 16 2501-2506
Clausing, P , Mak, J. K., Spengler, J D , Letz, R (1984) Personal NO2 exposures of high school students
In Berglund, B , Lindvall, T , Sundell, J , eds Indoor air proceedings of the 3rd international
conference on indoor air quality and climate, v 4, chemical characterization and personal exposure,
August, Stockholm, Sweden Stockholm, Sweden Swedish Council for Building Research, pp 135-139
Available from NTIS, Springfield, VA, PB85-104214
Colome, S D., Wilson, A L , Spengler, J D (1992) Nitrogen dioxide exposure studies volume V, personal
exposure to nitrogen dioxide in the Los Angeles basin a microenvironmental approach Chicago, IL
Gas Research Institute, report no GRI-92/0427
Dasch, J. M , Cadle, S H (1990) The removal of nitric acid to atmospheric particles during a wintertime field
study Atmos Environ Part A 24 2557-2562
Dasch, J M , Cadle, S H , Kennedy, K G , Mulawa, P A (1989) Comparison of annular denuders and filter
packs for atmospheric sampling Atmos Environ 23 2775-2782
Decker, C E , Ripperton, L A , Worth, J J B , Vukovich, F M , Bach, W D , Tommerdahl, J B ,
Smith, F , Wagoner, D E (1976) Formation and transport of oxidants along Gulf Coast and in northern
U S. Research Triangle Park, NC U S Environmental Protection Agency, Office of Air Quality
Planning and Standards, EPA report no EPA-450/3-76-033 Available from NTIS, Springfield, VA,
PB-260384
Drye, E. E., O2kaynak, H , Burbank, B , Billick, I H , Baker, P E , Spengler, J D , Ryan, P B , Colome,
S D (1989) Development of models for predicting the distribution of indoor nitrogen dioxide
concentrations JAPCA39 1169-1177
Federal Register (1979) Air programs, ambient air quality monitoring, data reporting, and surveillance
provisions. F R (May 10) 44 27558-27604
Fehsenfeld, F C ; Dickerson, R R , Hubler, G , Luke, W T , Nunnermacker, L J , Williams, E J , Roberts,
J M ; Calvert, J G , Curran, C M , Delany, A C , Eubank, C S , Fahey, D W , Fried, A ,
Gandrud, B W , Langford, A O , Murphy, P C , Norton, R B , Pickering, K E , Ridley, B A
(1987) A ground-based intercompanson of NO, NOX, and NOy measurement techniques J Geophys
Res [Atmos ] 92 14710-14722
7-96
-------
Ferns, B G , Jr , Speizer, F E , spengler, J D , Dockery, D , Bishop, Y M M , Wolfson, M , Humble, C
(1979) Effects of sulfur oxides and respirable particles on human health Methodology and demography
of populations in study Am Rev Respir Dis 120 767-779
Fortmann, R C , Nagda, N L , Harper, J P (1987) Radon mitigation through residential pressunzation control
strategy In Seifert, B , Esdorn, H , Fischer, M , Rueden, H , Wegner, J , eds Indoor air '87
proceedings of the 4th international conference on indoor air quality and climate, v 2, environmental
tobacco smoke, multicomponent studies, radon, sick buildings, odours and irritants, hyperreactivities and
allergies, August, Berlin, Federal Republic of Germany Berlin, Federal Republic of Germany Institute
for Water, Soil and Air Hygiene, pp 300-304
Fugas, M (1976) Assessment of total exposure to an air pollutant In International conference on environmental
sensing and assessment, v 2, September, Las Vegas, NV (SFew York, NY Institute of Electrical and
Electronic Engineers, paper no 38-5
Goldstein, B D , Melia, R J W , Chum, S , Florey, C du V , Clark, D , John, H H (1979) The relation
between respiratory illness in primary schoolchildren and the use of gas for cooking n Factors affecting
nitrogen dioxide levels in the home Lit J Epidemiol 8 339-345
Goldstein, I F , Hartel, D , Andrews, L R (1985) Monitoring personal exposure to nitrogen dioxide Presented
at 78th annual meeting of the Air Pollution Control Association, June, Detroit, MI Pittsburgh, PA Air
Pollution Control Association, paper no 85-85 7
Good, B W , Vilcins, G , Harvey, W R , Clabo, D A , Jr , Lewis, A L (1982) Effect of cigarette smoking
on residential NO2 levels Environ Lat 8 167-175
Gnmsrud, D T , Sherman, M H , Sonderegger, R C (1982) Calculating infiltration implications for a
construction quality standard In Proceedings of the ASHRAE-EOE conference on the thermal
performance of the exterior envelope of buildings U, December, Las Vegas, NV New York, NY
American Society of Heating, Refrigerating and Air-Conditiomng Engineers, pp 422-452
Grot, R A , Clark, R E (1979) Air leakage characteristics and weathenzation techniques for low-income
housing In Proceedings of the ASHRAE-DOE conference on the thermal performance of the exterior
envelope of buildings, December, Orlando, FL New York, NY American Society of Heating,
Refrigerating and Air-Conditiomng Engineers, pp 178-194
Hanson, P J , Taylor, G E , Jr , Vose, J (1992) Experimental laboratory measurements of reactive N gas
deposition to forest landscape surfaces biological and environmental controls In Johnson, D W ,
Lindberg, S E , eds Atmosphenc deposition and forest nutnent cycling a synthesis of the integrated
forest study New York, NY Sprmger-Verlag, pp 166-177 (Ecological studies 91)
Harlos, D P , Spengler, J D , Bilhck, I (1987) Continuous nitrogen dioxide monitoring during cooking and
commuting personal and stationary exposures In Seifert, B , Esdorn, H , Fischer, M , Rueden, H ,
Wegner, J , eds Indoor air '87 proceedings of the 4th international conference on indoor air quality and
climate, v 1, volatile organic compounds, combustion gases, particles and fibres, microbiological agents,
August, Berlin, Federal Republic of Germany Berlin, Federal Republic of Germany Institute for Water,
Soil and Air Hygiene, pp 278-282
Hayes, S R (1989) Estimating the effect of being indoors on total personal exposure to outdoor air pollution
JAPCA39 1453-1461
7-97
-------
Hemphill, C P , Ryan, P B , Bilhck, I H , Nagda, N L , Koontz, M D , Fortmann, R C (1987) Estimation
of nitrogen dioxide concentrations in homes equipped with unvented gas space heaters In Seifert, B ,
Esdorn, H , Fischer, M , Rueden, H , Wegner, J , eds Indoor air '87 proceedings of the 4th
international conference on indoor air quality and climate, v 1, volatile organic compounds, combustion
gases, particles and fibres, microbiological agents, August, Berlin, Federal Republic of Germany Berlin,
Federal Republic of Germany Institute for Water, Soil and Air Hygiene, pp 420-424
Hosein, H R , Bouhuys, A (1979) Possible environmental hazards of gas cooking [letter] Br Med J (6156)
125.
Jacobson, J S , McManus, J M (1985) Pattern of atmospheric sulphur dioxide occurrence an important
criterion in vegetation effects assessment Atmos Environ 19 501-506
Jcnlon, M E , Cox, R A , Williams, D J (1988) Laboratory studies of the kinetics of formation of nitrous acid
from the thermal reaction of nitrogen dioxide and water vapour Atmos Environ 22 487-498
Kaplan, W A., Wofsy, S C , Keller, M , Da Costa, J M (1988) Emission of NO and deposition of O3 in a
tropical forest system J Geophys Res 93 1389-1395
Keller, M D , Lanese, R R , Mitchell, R I, Cote, R W (1979) Respiratory illness in households using gas
and electricity for cooking I survey of incidence Environ Res 19 495-503
Kelly, T. J ; Stedman, D H , Ritter, J A , Harvey, R B (1980) Measurements of oxides of nitrogen and nitric
acid in clean air J Geophys Res 85 7417-7425
Kelly, N. A , Wolff, G T , Ferman, M A (1982) Background pollutant measurements in air masses affecting
the eastern half of the United States—I air masses arriving from the northwest Atmos Environ
16- 1077-1088
Kelly, T J ; Tanner, R L , Newman, L , Galvin, P J , Kadlecek, J A (1984) Trace gas and aerosol
measurements at a remote site in the northeast U S Atmos Environ 18 2565-2576
Kelly, N A ; Ferman, M A , Wolff, G T (1986) The chemical and meteorological conditions associated with
high and low ozone concentrations in southeastern Michigan and nearby areas of Ontario J Air Pollut
Control Assoc 36 150-158
Koontz, M D , Fortmann, R C , Nagda, N L , Bilhck, I H (1986) Protocol for an indoor air quality
monitoring survey conducted in Texas Presented at 79th annual meeting of the Air Pollution Control
Association, June, Minneapolis, MN Pittsburgh, PA Air Pollution Control Association, paper no
86-63
Koontz, M D , Fortmann, R C , Nagda, N L (1988) A topical report on a field monitoring study of homes
with unvented gas space heaters v IV—quality assurance/control procedures and results [final report
(January 1985-June 1987)] Chicago, IL Gas Research Institute, report no GRI-87/0044 3
Koontz, M D , Mehegan, L L , Nagda, N L (1992) Distribution and use of cooking appliances that can affect
indoor air quality topical report (November 1985—December 1991) Chicago, IL Gas Research
Institute, report no GRI-93/0013 Available from NTIS, Springfield, VA, PB93-185452/XAB
Koutrakis, P , Bnggs, S L K , Leaderer, B P (1992) Source apportionment of indoor aerosols in Suffolk and
Onandaga Counties, New York Environ Sci Technol 26 521-527
7-98
-------
Lambert, W E (1991) [Letter to Dr Dennis Kotchmar concerning nitrogen dioxide measurements for a sample
of 100 homes participating in the UNM study of infant respiratory illness] Albuquerque, NM The
University of New Mexico, New Mexico Tumor Registry Medical Center, July 10
Lane, P I, Bell, J N B (1984) The effects of simulated urban air pollution on grass yield part I— description
and simulation of ambient pollution Environ Pollut Ser B 8 245-263
Leaderer, B P (1982) Air pollutant emissions from kerosene space heaters Science (Washington, DC)
218 1113-1115
Leaderer, B P , Stolwijk, J A J , Zagramski, R T , Qing-Shan, M (1984) A field study of indoor air
contaminant levels associated with unvented combustion souices Presented at 77th annual meeting of the
Air Pollution Control Association, June, San Francisco, CA Pittsburgh, PA Air Pollution Control
Association, paper no 84-33 3
Leaderer, B P , Zagramski, R T , Berwick, M , Stolwijk, J A J (1986a) Assessment of exposure to indoor
air contaminants from combustion sources methodology and application Am J Epidemiol
124 275-289
Leaderer, B P , Renes, S , Bluyssen, P , Van De Loo, H (1986b) Chamber studies of NO2, SO2 and RSP
deposition rates indoors Presented at 79th annual meeting of the Air Pollution Control Assbciation,
June, Minneapolis, MN Pittsburgh, PA Air Pollution Control Association, paper no 86-38 3
Leaderer, B P , Zagramski, R T , Berwick, M , Stolwijk, J A J (1987) Predicting NO2 levels in residences
based upon sources and source use a multivanate model Almos Environ 21 361-368
Lebret, E , Noy, D , Boley, J , Brunekreef, B (1987) Real-time concentration measurements of CO and NO2 in
twelve homes In Seifert, B , Esdorn, H , Fischer, M , Rueden, H , Wegner, J , eds Indoor air '87
proceedings of the 4th international conference on indoor ail quality and climate, v 1, volatile organic
compounds, combustion gases, particles and fibres, microbiological agents, August, Berlin, Federal
Republic of Germany Berlin, Federal Republic of Germany Institute for Water, Soil and Air Hygiene,
pp 435-439
Lee, K , Yanagisaswa, Y , Spengler, J D , Ozkaynak, H (1992) Nitrogen dioxide exposure studies volume VI,
the NO2 badge Chicago, IL Gas Research Institute, report no GRI-92/0428
Lefohn, A S (1989) The characterization of sulfur dioxide, nitrogen dioxide, and ozone gaseous concentrations
at Mountain Cloud Chemistry Program monitoring sites in the United States Research Triangle Park,
NC U S Environmental Protection Agency, Atmospheric Research and Exposure Assessment
Laboratory
Lefohn, A S , Tingey, D T (1984) The co-occurrence of potentially phytotoxic concentrations of various
gaseous air pollutants Atmos Environ 18 2521-2526
Lefohn, A S , Davis, C E , Jones, C K , Tingey, D T , Hogsett, W E (1987) Co-occurrence patterns of
gaseous air pollutant pairs at different minimum concentrations in the United States Atmos Environ
21 2435-2444
Lefohn, A S , Benkovitz, C M , Tanner, R L , Smith, L A , Shaidwick, D S (1991) Air quality
measurements and characterizations for terrestrial effects research In Irving, P M , ed Acidic
deposition state of science and technology, volume I, emissions, atmospheric processes, and deposition
Washington, DC The U S National Acid Precipitation Assessment Program (State of science and
technology report no 7)
7-99
-------
Legge, A H , Peake, E , Strosher, M , Nosal, M , McVehil, G E , Hansen, M (1990) Characteristics of the
background air quality In Legge, A H , Krupa, S V , eds Acidic deposition sulphur and nitrogen
oxides Chelsea, MI Lewis Publishers, pp 129-248
Lindberg, S E ; Bredemeier, M , Schaefer, D A , Qi, L (1990) Atmospheric concentrations and deposition of
nitrogen and major ions in conifer forests in the United States and Federal Republic of Germany Atmos
Environ Part A 24 2207-2220
Macriss, R A , Elkins, R H (1977) Control of the level of NO^ in the indoor environment In Kasuga, S ,
Suzuki, N , Yamada, T , Kimura, G , Ihagaki, K , Onoe, K , eds Proceedings of the fourth
international clean air congress, May, Tokyo, Japan Tokyo, Japan Japanese Union of Air Pollution
Prevention Associations, pp 510-514
Magee, P N , ed (1982) Nitrosamines and human cancer Cold Spring Harbor, NY Cold Spring Harbor
Laboratory (Banbury report 12)
Marbury, M C , Harlos, D P , Samet, J M , Spengler, J D (1988) Indoor residential NO2 concentrations in
Albuquerque, New Mexico JAPCA 38 392-398
Martinez, J. R , Singh, H B (1979) Survey of the role of NOX in nonurban ozone formation Research Triangle
Park, NC U.S Environmental Protection Agency, Office of Air Quality Planning and Standards, EPA
report no EPA-450/4-79-035 Available from NTIS, Springfield, VA, PB80-122815
McCarthy, S M , Yarmac, R F , Yocom, J E (1987) Indoor nitrogen dioxide exposure the contribution from
unvented gas space heaters In Seifert, B , Esdorn, H , Fischer, M , Rueden, H , Wegner, J , eds
Indoor air '87 proceedings of the 4th international conference on indoor air quality and climate, v 1,
volatile organic compounds, combustion gases, particles and fibres, microbiological agents, August,
Berlin, Federal Republic of Germany Berlin, Federal Republic of Germany Institute for Water, Soil and
Air Hygiene, pp 478-482
Melia, R J. W ; Florey, C du V , Morris, R W , Goldstein, B D , Clark, D , John, H H (1982) Childhood
respiratory illness and the home environment I Relations between nitrogen dioxide, temperature and
relative humidity Int J Epidemiol 11 155-163
Miyazaki, T (1984) Adsorption characteristics of NOX by several lands of interior materials In Berglund, B ,
Lindvall, T , Sundell, J , eds Indoor air proceedings of the 3rd international conference on indoor air
quality and climate, v 4, chemical characterization and personal exposure, August, Stockholm, Sweden
Stockholm, Sweden Swedish Council for Building Research, pp 103-110 Available from NTIS,
Springfield, VA, PB85-104214
Moschandreas, D J , Relwam, S M , Macriss, R A , Cole, J T (1984) Differences and similarities of two
techniques used to measure emission rates from unvented gas appliances In Berglund, B , Lindvall, T ,
Sundell, J , eds Indoor air proceedings of the 3id international conference on mdoor air quality and
climate, v 4, chemical characterization and personal exposure, August, Stockholm, Sweden Stockholm,
Sweden Swedish Council for Building Research, pp 375-379 Available from NTIS, Springfield, VA,
PB85-104214.
Moschandreas, D J ; Relwam, S M , O'Neill, H J , Cole, J T , Elkins, R H , Macriss, R A (1985)
Characterization of emission rates from mdoor combustion sources Chicago, IL Gas Research Institute,
report no GRI 85/0075 Available from NTIS, Springfield, VA, PB86-103900
National Research Council (1981) Indoor pollutants Washington, DC National Academy Press
7-100
-------
National Research Council (1991) Rethinking the ozone problem in urban and regional air pollution
Washington, DC National Academy Press
Nazaroff, W W , Cass, G R (1986) Mathematical modeling of chemically reactive pollutants in indoor air
Environ Sci Technol 20 924-934
Neas, L M , Dockery, D W , Ware, J H , Spengler, J D , Speizer, F E , Ferns, B G , Jr (1991)
Association of indoor nitrogen dioxide with respiratory symptoms and pulmonary function in children
Am J Epidemiol 134 204-219
New York State Department of Health (1989) Influence of surficial soil and bedrock on indoor radon in New
York state homes task n, subtask of an investigation of infiltration and indoor air quality in New York
state homes Albany, NY New York State Energy Research and Development Authority, energy
authority report no 89-14
Noy, D , Lebret, E , Boleij, J , Brunekreef, B (1984) Integrated NO2 exposure estimates In Berglund, B ,
Lindvall, T , Sundell, J , eds Indoor air proceedings of the 3rd international conference on indoor air
quality and climate, v 4, chemical characterization and personal exposure, August, Stockholm, Sweden
Stockholm, Sweden Swedish Council for Building Research, pp 37-42 Available from NTIS,
Springfield, VA, PB85-104214
Ott, W (1989) Human activity patterns a review of the literature for estimating time spent indoors, outdoors,
and in-transit In Starks, T H , ed Proceedings of the research planning conference on human activity
patterns Las Vegas, NV U S Environmental Protection Agency, Environmental Monitoring Systems
Laboratory, chapter 3, EPA report no EPA-600/4-89-004 Available from NTIS, Springfield, VA,
PB89-166599
Ozkaynak, H , Ryan, P B , Allen, G A , Turner, W A (1982) Ikdoor air quality modeling compartmental
approach, with reactive chemistry Environ Eat 8 461-471
Palmes, E D , Gunmson, A F , DiMattio, J , Tomczyk, C (1976) Personal sampler for nitrogen dioxide
Am lad Hyg Assoc J 37 570-577
Palmes, E D , Tomczyk, C , DiMattio, J (1977) Average NO2 concentrations in dwellings with gas or electric
stoves Atmos Environ 11 869-872
Parkhurst, W J , Humphreys, M P , Harper, J P , Spengler, J D (1988) Influence of indoor combustion
sources on indoor air quality Environ Prog 7 257-261
Parrish, D D , Williams, E , Norton, R B , Fehsenfeld, F C (1985) Measurement of odd-nitrogen species and
O3 at Point Arena, California EOS Trans Am Geophys Union 66 820
Parrish, D D , Trainer, M , Williams, E J , Fahey, D W , Huebler, G , Eubank, C S , Liu, S C , Murphy,
P C , Albntton, D L , Fehsenfeld, F C (1986) Measurements of the NOX-O3 photostationary state at
Niwot Ridge, Colorado J Geophys Res [Atmos ] 91 5361-5370
Parrish, D D , Buhr, M , Norton, R , Fehsenfeld, F (1988) Study of atmospheric oxidation processes at a
rural, eastern U S site EOS Trans Am Geophys Union 69 1056
Petreas, M , Liu, K -S , Chang, B -H , Hayward, S B , Sexton, K (1988) A survey of nitrogen dioxide levels
measured inside mobile homes JAPCA 38 647-651
7-101
-------
Puts, J N., Jr , Grosjean, D , Van Cauwenberghe, K , Schmid, J P , Fitz, D R (1978) Photooxidation of
aliphatic amines under simulated atmospheric conditions formation of nitrosamines, mtramines, amides,
and photochemical oxidant Environ Sci Technol 12 946-953
Pitts, J. N , Jr , Sanhueza, E , Atkinson, R , Carter, W P L , Winer, A M , Haras, G W , Plum, C N
(1984) An investigation of the dark formation of nitrous acid in environmental chambers Int J Chem
Kinct. 16 919-939
Pitts, J. N , Jr , Biermann, H W , Tuazon, E C , Green, M , Long, W D , Winer, A M (1989)
Time-resolved identification and measurement of indoor air pollutants by spectroscopic techniques
gaseous nitrous acid, methanol, formaldehyde and formic acid J Air Waste Manage Assoc
39. 1344-1347
Pratt, G C , Hendnckson, R C , Chevone, B I , Chnstopherson, D A , O'Brien, M V , Krupa, S V (1983)
Ozone and oxides of nitrogen in the rural upper-midwestern USA Atmos Environ 17 2013-2023
Quackenboss, J J , Spengler, J D , Kanarek, M S , Letz, R , Duffy, C P (1986) Personal exposure to
nitrogen dioxide relationship to indoor/outdoor air quality and activity patterns Environ Sci Technol
20 775-783
Quackenboss, J J , Lebowitz, M D , Hayes, C (1987) Epidemiological study of respiratory responses to
indoor-outdoor air quality In Seifert, B , Esdorn, H , Fischer, M , Rueden, H , Wegner, J , eds
Indoor air '87. proceedings of the 4th international conference on indoor air quality and climate, v 2,
environmental tobacco smoke, multicomponent studies, radon, sick buildings, odours and irritants,
hyperreactivities and allergies, August, Berlin, Federal Republic of Germany Berlin, Federal Republic of
Germany Institute for Water, Soil and Air Hygiene, pp 198-202
Quackenboss, J J , Bronmmann, D , Camilli, A E , Lebowitz, M D (1988) Bronchial responsiveness in
children and adults in association with formaldehyde, particulate matter, and environmental tobacco
smoke exposures Am Rev Respir Dis 137(suppl) 253
Research Triangle Institute (1975) Investigation of rural oxidant levels as related to urban hydrocarbon control
strategies Research Triangle Park, NC U S Environmental Protection Agency, Office of Air Quality
Planning and Standards, EPA report no EPA-450/3-75-036 Available from NTIS, Springfield, VA,
PB-242299.
Research Triangle Institute (1990) An investigation of infiltration and indoor air quality Albany, NY New York
State Energy Research and Development Authority, report no NYERDA 90-11 Available from NTIS,
Springfield, VA, PB91-119156
Robinson, J. P (1977) How Americans use their time a social psychological analysis of everyday behavior New
York, NY Praeger Publishers
Ryan, P. B , Spengler, J D (1992) Nitrogen dioxide exposure studies volume n, the Boston residential home
characterization study Chicago, JJL Gas Research Institute, report no GRI-92/0424
Ryan, P. B ; Spengler, J D., Letz, R (1983) The effects of kerosene heaters on indoor pollutant concentrations
& monitoring and modeling study Atmos Environ 17 1339-1345
Ryan, P. B ; Soczek, M L , Treitman, R D , Spengler, J D , Bilhck, I H (1988) The Boston residential
N©2 characterization study—n survey methodology and population concentration estimates Atmos
Environ 22 2115-2125
7-102
-------
Ryan, P B , Schwab, M , Spengler, J D , Colome, S D , Wilson, A L (1992) Nitrogen dioxide exposure
studies, volume IV human exposure to nitrogen dioxide in the Los Angeles basin Chicago, IL Gas
Research Institute and Los Angeles, CA Southern California Gas Company
Sakamaki, F , Hatakeyama, S , Akimoto, H (1983) Formation of nitrous acid and nitric oxide in the
heterogeneous dark reaction of nitrogen dioxide and water vapor in a smog chamber Int J Chem
Kinet 15 1013-1029
Samet, J M , Lambert, W E , Skipper, B J , Gushing, A H , McLaren, L C , Schwab, M , Spengler, J D
(1992) A study of respiratory illnesses in infants and nitrogen dioxide exposure Arch Fjiviron Health
47 57-63
Sheldon, L S , Hartwell, T D , Cox, B G , Sickles, J E , H, Pdhzzan, E D , Smith, M L , Perntt, R L ,
Jones, S M (1989) An investigation of infiltration and indoor air quality final report Albany, NY
New York State Energy Research and Development Authority, New York State ERDA contract no
736-CON-BCS-85
Sparks, L E (1988) Indoor air—sources indoor air quality model version 1 0 Research Triangle Park, NC
U S Environmental Protection Agency, Air and Energy Engineering Research Laboratory, EPA report
no EPA-600/8-88-097
Speizer, F E , Ferns, B , Jr , Bishop, Y M M , Spengler, J (1980) Respiratory disease rates and pulmonary
function in children associated with NO2 exposure Am Rev Respir Dis 121 3-10
Spengler, J , Samet, J (1992) NO2 concentrations in a subsample of Albuquerque NM homes participating in the
infant respiratory health study conducted by J Samet, University of New Mexico, New Mexico Tumor
Registry Medical Center Personal communication February
Spengler, J D , Duffy, C P , Letz, R , Tibbitts, T W , Ferns, B G , Jr (1983) Nitrogen dioxide inside and
outside 137 homes and implications for ambient air quality standards and health effects research Environ
Sci Technol 17 164-168
Spengler, J D , O2kaynak, H , McDermott, A , Brauer, M (1992a) Review of the draft air quality cntena for
oxides of nitrogen (EPA/600/8-91/049) Boston, MA Harvard School of Public Health, Department of
Environmental Health
Spengler, J D , Ryan, P B , Schwab, M , Colome, S D , Wilson, A L (1992b) Nitrogen dioxide exposure
studies volume IV, personal exposure to nitrogen dioxide in the Los Angeles basin Chicago, IL Gas
Research Institute, report no GRI-92/0426
Spengler, J D , Brauer, M , Samet, J M , Lambert, W E (1993) Nitrous acid in Albuquerque, New Mexico,
homes Environ Sci Technol 27 841-845
Spicer, C W (1977) The fate of nitrogen oxides in the atmosphere In Pitts, J N , Jr , Metcalf, R L , Lloyd,
A C , eds Advances in environmental science and technology v 4 New York, NY John Wiley and
Sons, Inc , pp 163-261
Spicer, C W , Coutant, R W , Ward, G F (1986) Investigation of nitrogen dioxide removal from indoor air
[final report (December 1984—September 1986)] Chicago, IL Gas Research Institute, report
no GRI-86/0303 Available from NTIS, Springfield, VA, PB87-185971
Spicer, C W , Coutant, R W , Ward, G F , Joseph, D W , Gaynor, A J , Bilhck, I H (1989) Rates and
mechanisms of NO2 removal from indoor air by residential materials Environ Int 15 643-654
7-103
-------
Svensson, R , Ljungstroem, E , Lmdqyist, O (1987) Kinetics of the reaction between nitrogen dioxide and water
vapour Atmos Environ 21 1529-1539
Szalai, A , ed (1972) The use of time daily activities of urban and suburban populations in 12 countries The
Hague, The Netherlands Mouton and Co
Taylor, G E , Ross-Todd, B M , Allen, E , Conkkn, P , Edmonds, R , Joranger, E , Miller, E , Ragsdale, L,
Shepard, J.; Silsbee, D , Swank, W (1992) Patterns of tropospehnc ozone in forested landscapes of the
Integrated Forest Study In Johnson, D W , Lindberg, S E , eds Atmospheric deposition and forest
nutrient cycling—a synthesis of the Integrated Forest Study New York, NY Springer-Verlag, pp 50-71
Tikalsky, S , Reisdorf, K , Flickinger, J , Tofczke, D , Haywood, J , Annen, L , Kanarek, M , Kaarakka, P ,
Prrns, E (1987) Gas range/oven emissions impact analysis [final report (July 1985-December 1987)]
Chicago, IL Gas Research Institute, report no GRI-87/0119 Available from NTTS, Springfield, VA,
PB88-232756
Torres, A. L , Buchan, H (1988) Troposphenc nitric oxide measurements over the Amazon Basin J Geophys
Res 93 1396-1406
Traynor, G W , Anthon, D W , Hollowell, C D (1982) Technique for determining pollutant emissions from a
gas-fired range Atmos Environ 16 2979-2987
Traynor, G W , Apte, M G , Carruthers, A R , Dillworth, J F , Grimsrud, D T , Thompson, W T (1984)
Indoor air pollution and interroom pollutant transport due to unvented kerosene-fired space heaters
Berkeley, CA Lawrence Berkeley Laboratory, report no LBL-17600 Available from NTIS,
Springfield, VA, DE84015949
Traynor, G W ; Aceti, J C , Apte, M G , Smith, B V , Green, L L , Smith-Reiser, A , Novak, K M ,
Moses, D O (1987) Macromodel for assessing indoor exposures to combustion-generated pollutants
In Seifert, B , Esdorn, H , Fischer, M , Rueden, H , Wegner, J , eds Indooi air '87 proceedings of
the 4th international conference on indoor air quality and climate, v 1, volatile organic compounds,
combustion gases, particles and fibres, microbiological agents, August, Berlin, Federal Republic of
Germany. Berlin, Federal Republic of Germany Institute for Water, Soil and Air Hygiene, pp 273-277
U S Bureau of the Census (1982) 1980 census of population and housing supplementary report provisional
estimates of social, economic, and housing characteristics states and selected standard metropolitan
statistical areas Washington, DC U S Department of Commerce, report no PHC 80-S1-1 Available
from GPO, Washington, DC
U S. Bureau of the Census (1989) 1990 Census of population and housing—content determination reports
Housing plumbing, equipment, and fuels Washington, DC U S Department of Commerce, report no
1990 CDR-12 Available from GPO, Washington, DC
U.S Bureau of the Census (1992) 1990 Census of population and housing summary social, economic, and
housing characteristics—United States Washington, DC U S Department of Commerce, report no 1990
CPH-5-1 Available from GPO, Washington, DC
U.S. Environmental Protection Agency (1990) National air quality and emissions trends report, 1988 Research
Triangle Park, NC Office of Air Quality Planning and Standards, EPA report no EPA/450/4-90/002
Available from NTIS, Springfield, VA, PB90-200114/XAB
U.S. Environmental Protection Agency (1991a) National air quality and emissions trends report, 1989 Research
Triangle Park, NC Office of Air Quality Planning and Standards, EPA report no EPA/450/4-91/003
Available from NTIS, Springfield, VA, PB91-172247/XAB
7-104
-------
U S Environmental Protection Agency (1991b) National air quality and emissions trends report, 1990 Research
Triangle Park, NC Office of Air Quality Planning and Standards, EPA report no EPA-450/4-91-023
Available from NTIS, Springfield, VA, PB92-141555/XAB
Wade, W A , JH, Cote, W A , Yocom, J E (1975) A study of indoor air quality J Air Pollut Control
Assoc 25 933-939
Weir, B R , Anderson, G E , Hayes, S R , Greenfield, S M (1990) Specification of indoor air model
characteristics In Indoor air '90 precedings of the 5th international conference on indoor air quality and
climate, volume 4, building and system assessments and solutions, July-August, Toronto, ON, Canada
Ottawa, ON, Canada International Conference on Indoor Air Quality and Climate, Inc , pp 231-236
Weschler, C J , Brauer, M , Koutrakis, P (1992) Indoor ozone and nitrogen dioxide a potential pathway to the
generation of nitrate radicals, dimtrogen pentaoxide, and mtnc acid indoors Environ Sci Technol
26 179-184
Wiley, J A , Robinson, J P , Piazza, T , Garrett, K , Cirksena, R , Cheng, Y -T , Martin, G (1991a) Activity
patterns of California residents Sacramento, CA California Air Resources Board, contract no
A6-177-33
Wiley, J A , Robinson, J P , Piazza, T , Garrett, T , Cirksena, K , Cheng, Y -T , Martin, G (1991b) Study of
children's activity patterns final report Sacramento, CA California Air Resources Board, contract
no A-733-149
Wilson, A L , Colome, S D , Baker, P E , Becker, E W (1986) Residential indoor air quality
characterization study of nitrogen dioxide Phase I Volumes 1, 2 and 3 Southern California Gas
Company, October
Winer, A M , Biermann, H W (1991) Measurements of nitrous acid, nitrate radicals, formaldehyde and
nitrogen dioxide for the Southern California Air Quality Study by differential optical absorption
spectroscopy In Conference on chemical sensing of the environment measurement of atmospheric
gases, January, Los Angeles, CA Proc SPJJE-Int Soc Opt Eng (1433) 44-55
Wolff, G T (1984) On the nature of nitrate in coarse continental aerosols Atmos Environ 18 977-981
Wolff, G T , Korsog, P E (1989) Atmospheric concentrations and regional source apportionments of sulfate,
nitrate and sulfur dioxide in the Berkshire Mountains in western Massachusetts Atmos Environ
23 55-65
Wolff, G T , Ruthkosky, M S , Stroup, D P , Korsog, P E , Ferman, M A , Wendel, G J , Stedman,
D H (1986a) Measurements of SOX, NOX and aerosol species on Bermuda Atmos Environ
20 1229-1239
Wolff, G T , Kelly, N A , Ferman, M A , Ruthkosky, M S , Stioup, D P , Korsog, P E (1986b)
Measurements of sulfur oxides, nitrogen oxides, haze and fine particles at a rural site on the Atlantic
coast J Air Pollut Control Assoc 36 585-591
Wolff, G T , Ruthkosky, M S , Stroup, D P , Korsog, P E (1991) A characterization of the principal PM-10
species in Claremont (summer) and Long Beach (fall) during SCAQS Atmos Environ Part A
25 2173-2186
Yamanaka, S (1984) Decay rates of nitrogen oxides in a typical Japanese living room Environ Sci Technol
18 566-570
7-105
-------
Yanagisawa, Y.; Nishimura, H (1982) A badge-type personal sampler for measurement of personal exposure to
NO2 and NO in ambient air Environ Int 8 235-242
7-106
-------
8. ASSESSING TOTAL HUMAN EXPOSURE
TO NITROGEN DIOXIDE
8.1 INTRODUCTION
In the course of their daily activities, humans are exposed to nitrogen dioxide (NQ^) m
a number of settings or environments (residential, industrial, nomndustnal occupational,
transportation, outdoors, etc ) Human exposure to NO2 consists of contact at the air
boundary layer between the human and the environment at a specific concentration for a
specified period of tune The integrated NO2 exposure is the sum of the individual NO2
exposures over all possible time intervals for all settings or environments Thus, the units of
exposure are concentration multiplied by time
The assessment of human exposures to NO2 can be represented by the following
simplified basic model that was first proposed by Fugas (1976) and used by a number of
other researchers (e g , Duan, 1981, Sexton and Ryan, 1988)
(8-1)
where E is the total or integrated exposure to NO2 for an individual expressed as an average
concentration over a specified period of time across all microenvironments, ET is the average
NO2 exposure in the il microenvironment, ft is the fraction (percentage) of time spent in the
ii^ A|^
i microenvironment, and cz is the NO2 concentration in the i microenvironment This
model essentially represents exposures as a linear combination of the average concentration
in each microenvironment, weighted by an individual's time in a microenvironment type
A microenvironment is defined as a three-dimensional space having a volume in which the
pollutant concentration is considered or assumed to be spatially uniform In theory, there
can be a large number of microenvironments, in practice, however, microenvironments are
aggregated into a reduced number for air monitoring and modeling applications (outdoors,
public transportation, indoor residential, nomndustnal occupational, industrial, public access
buildings, etc ) The NO2 concentration in each microenvironment can show substantial
spatial and temporal variability and is a complex function of the sources, source use, and
8-1
-------
dispersal and removal mechanisms A more detailed discussion of the general theoretical
concepts in assessing exposure to air contaminants can be found in a recent report prepared
by the National Research Council (1991)
Until recently, efforts to assess adverse health and comfort effects associated with NO2
and subsequent efforts to mitigate or reduce exposures have focused on measurements of
ambient air quality obtained from fixed-location monitoring sites Numerous studies on
indoor concentrations of NO2 (see Chapter 7) combined with studies on human activity
patterns (e.g., Szalai, 1972, Chapin, 1974, Robinson, 1977, Ott, 1989, Schwab et al, 1990,
Wiley et al., 1991a,b) clearly underscore the need to consider contributions of individual
microenvironments in assessing NO2 exposures Outdoor NO2 levels also impact indoor
concentrations, and they can dominate indoor concentrations when there are no indoor
sources or when outdoor concentrations are elevated (Chapter 7) The magnitude of that
impact is determined by outdoor levels, temporal variability in outdooi levels, and building
characteristics (mfiltration/ventilation and indoor decay rates) In contrast, outdoor levels do
not well represent indoor concentrations where there are indoor NO2 sources, particularly
when outdoor levels are low and show little variability Direct individual exposure to
outdoor concentrations accounts on average for a small portion of a person's total exposure
because of the small amount of time typically spent outdoors It is important to note,
however, that for some significant segments of the population (e g , police, other outdoor
workers, etc.), outdoor exposures can account for a high fraction of their total exposure
Total personal exposure to NO2 and a consideration of the microenvironments in which those
exposures take place is essential in assessing the potential risk for associated health effects
and in developing effective mitigation measures
Central to the design of a human exposure assessment effort is the identification of the
health or comfort effect under study, ascertainment of the individual contaminant thought to
be associated with that effect, and specification of the contaminant exposure on a time scale
corresponding to the effect It is in fact the biological response time that is central to the
development of the exposure assessment methodology The principal biological response of
human exposures to NO2 known at this tune is respiratory disease Given that the current an
quality health standard (primary NAAQS) for NO2 reflects a long-term average
concentration, emphasis is placed here on assessing long-term exposures to NO2 on the order
8-2
-------
of days or weeks It is important to note, however, that short-term high episodes can elevate
total exposures (concentration x tune) and may be of importance in impacting health
(Chapter 13) Also, other exposure measures may be appropriate
In developing and conducting an exposure assessment protocol, the application of the
acquired data must be considered The three principal applications for NO2 exposure
assessment efforts are environmental epidemiology, risk assessment, and risk management
In environmental epidemiology, misclassifications and the influence of confounders can be
minimized when NO2 exposures of the study population are adequately assessed
In addition, the application of an appropriate NO2 exposure assessment methodology will
reduce the error or uncertainty in calculating risks associated with NO2 exposure, aid in
formulating cost effective mitigation efforts to minimize risk, and permit the monitoring of
progress in reducing the risk
Adequate exposure assessment for NO2 is particularly critical in conducting and
evaluating epidemiological studies As discussed in more detail in Chapter 13, failure to
measure or estimate exposures adequately and to address the uncertainty in the measured
exposures can lead to misclassification errors (Shy et al, 1978, Gladen and Rogan, 1979,
Ozkaynak et al , 1986, Willet, 1989, Dosemeci et al , 1990, Lebret, 1990) Early
epidemiological studies comparing incidence of respiratory illness in children living in homes
with and without gas stoves used the presence of a gas cooker as a simple categorical
variable of NO2 exposure Such a dichotomous grouping can result in a severe
nondifferential misclassification error in assigning exposure categories This type of error is
likely to underestimate the true relationship and could possibly result in a null finding
Where categories of exposure are assigned based upon measured or estimated concentrations,
the intermediate categories of exposure may be biased either away from or towards the null
In support of epidemiologic studies of NO2, it is impoitant to employ an exposure assessment
protocol that will minimize the misclassification and characterize the uncertainty associated
with assessed exposures (Lambert et al, 1990)
Exposures to NO2 can be assessed by either direct or indirect methods Direct methods
include biomarkers and personal monitoring (breathing zone measurements) Indirect
methods refer to the coupling of measurements of NO2 concentrations in various settings or
environments with models to estimate exposure In this chapter, the available data on human
8-3
-------
exposure to NO2 will be reviewed within the context of the direct and indirect methods of
exposure assessment
8.2 DIRECT METHODS
8.2.1 Biomarkers
Biological markers are cellular, biochemical, or molecular measuies that are obtained in
biological media such as human tissues, cells, or fluids and are indicative of exposure to
environmental chemicals (National Research Council, 1989) Biomarkers of exposure can
theoretically integrate total intake to the body from multiple sources of exposure to
environmental contaminants If they are stable over time, they can be used to integrate
exposure over time. They can be useful tools in elucidating mechanisms of disease or in
extrapolations between internal doses, routes of exposure, and species or tissues, but do not
necessarily provide the direct link between environmental exposure and disease Biomarkers
may be measures of the contaminant or its metabolites that are directly related to the specific
contaminant associated with the effect outcome (e g , lead) or may only be a surrogate for
exposure to a complex source of environmental contaminants (e g , cotinine)
The relationship between the biological marker and the air contaminant concentration
and length of exposure is typically poorly understood Biomarkers are indicators that an
exposure has taken place, but are not necessarily a measure of exposure Biomarkers by
themselves do not provide information on the environment or setting in which the exposure
takes place and hence on the factors that control the exposure
Hydroxyproline excretion, an indicator of increased collagen cataboksm or connective
tissue injury, has been proposed as a biomarker for exposure to NO2 Yanagisawa et al
(1986, 1988) found an association between the hydroxyproline to creatinine ratio and daily
average personal NO2 levels in a sample of 800 women in two communities near Tokyo
Hydroxyproline, however, was found to be significantly correlated with passive and active
smoking. In chamber studies of normal males exposed to 0 6 ppm NO2 for 3 consecutive
days for 4 h/day, Muelenaer et al (1988) saw no significant changes in hydroxyprohne
excretion More recently, Maples et al (1991) have studied the potential for the
8-4
-------
nitric oxide/heme protein complex to serve as a useful biologic marker for NO2 exposure
The impact on hydroxyproline excretion of diet is not well characterized
8.2.2 Personal Monitoring
Personal air monitoring provides a direct measurement of the concentrations of air
contaminants in the immediate breathing zone of an individual as a measure of personal
exposure It provides a measure of exposure across the microenvironments or settings in
which individuals spend their tune Personal monitoring employs samplers (worn by
subjects) that record pollutant concentrations that individuals are exposed to in the course of
their normal activity for periods of seconds to several days These monitors can be
(a) passive dosimeters (e g , Palmes tubes or Yanagisawa badges), that provide an integrated
measure of exposure, or (b) portable monitors with data loggers, which can provide a nearly
continuous measure of exposure Little personal NO2 exposure data have been reported
using portable monitors The vast majority of personal NO2 exposure data have been
gathered using passive dosimeters Measures of integrated personal exposure (passive
dosimetry) by themselves are not adequate to determine the major sources of exposure and
the settings in which the exposures take place Personal integrated measures of exposure
need to be supplemented by personal activity diaries and measures of the factors affecting
those exposures if effective mitigation measures are to be developed and instituted A direct
measure of personal exposure (integrated or continuous) in epidemiologic studies of NO2
health effects, however, is highly desirable to minimize misclassification and to uncover
exposure-response relationships Personal exposures on whole study populations or selected
subsamples are not easy to obtain and can be expensive
The ability to obtain measures of personal exposure is in large part controlled by the
availability of accurate, small, easy to use, and inexpensive personal monitors The Palmes
tube (Palmes et al , 1976) and the NO2 badge (Yanagisawa and Nishunura, 1982) have
provided the sensitivity and specificity necessary to conduct personal NO2 air monitoring at a
reasonable cost These monitors are passive samplers that utilize diffusion to control
delivery of gases to the collection medium The samples are then returned to a laboratory
for subsequent laboratory analysis These passive samplers have been used extensively in the
8-5
-------
monitoring of various microenvironments (Chapter 7) and to a lesser extent in personal
exposure monitoring.
There are relatively few studies reported where subjects wore passive NO2 monitors in
the course of their daily activities to assess personal NO2 exposure la such studies, personal
exposures were typically compared to NO2 measurements made in the various
microenvironments in which they spent their tune Frequently, these studies obtained
supplemental information, through tune activity diaries, on the time subjects spent in various
microenvironments in which NO2 measurements were made. When rnicroenvironmental
measurements and time activity diaries are obtained, the personal exposure measurements are
compared to the rnicroenvironmental data, and personal exposure models are developed using
both the microenvironmental measurements and tune activity diaries Many of the personal
monitoring studies have in fact resulted in the development and testing of indirect methods of
assessing personal NO2 exposures
An extensive study on personal NO2 exposures as a function of outdoor and indoor
concentrations, indoor sources and time/activity patterns was reported by Quackenboss et al
(1986). In this study, 1-week NO2 samples were obtained during both a winter and summer
period using Palmes tubes for 350 volunteers residing in 82 homes in the rural Portage, WI,
area. The personal samples were supplemented with household measurements made outside
and in the kitchen and bedroom of each house, time spent in various microenvironments
(inside of home, outside, inside a motor vehicle, inside at work or school, and at other
indoor locations) and information on household characteristics (i e , existence of gas or
electric stove) Average NO2 personal exposures were weakly correlated to outdoor NO2
concentrations in the winter and moderately correlated in the summer (Figure 8-1) Personal
exposures were strongly correlated to home average (average of the bedroom and kitchen)
concentrations for homes with and without gas cooking ranges (Figure 8-2) Outdoor
concentrations were considerably lower than personal exposures for individuals in homes
with gas cooking ranges and higher than personal exposures for individuals in homes with
electric stoves A comparison of the indoor and outdoor NO2 concentrations for this study as
a function of season and use of a gas or electee cooking range is discussed in Chapter 7
The low explained variability in personal exposures by outdoor levels in this study may be
8-6
-------
NITROGEN DIOXIDE LEVELS IN SUMMER
t
1
1
fe
Ul
5;
2
o
i
0.
Ul
C5
£
<
ou
40
30
20
10
9
8
7
6
5
4
3
2
: *
= * *
: ... ** * H# * o
* * * ° *
'=• ****i*°^i0J>°°° o
r * o ° c&°< ° ° °
O v
- * 0
—
—
'- LEGEND.
_
r o Electric
E_ * Gas
~l I I I I I I I I Illlli I I I I I I I I ll I I I I I II I I I I I I I I I I I I Illll
2 S 10 20
OUTDOOR CONCENTRATION
SO
NITROGEN DIOXIDE LEVELS IN WINTER
t
S
^ 60
Ul
oc
I"
_J
^f
^3
i£*
S 10
cc
Ul
0.
Ul 6
<3
m
< 2
~ *
^_
- o
—
—
1
L
Z.
=_
I
"II 1 1 1 1 1 1 1 III III t
** *
* * *
>*** *
*** * *****,
* *Vsooo5
0 OD 0 ° 0 O°
o w-1 o ^i Q
CP oy
0 0 %
0 ° ° 0
0 LEGEND
o Electric
* Gas
1 H 1 1 1 1 ll 1 1 M 1 M II 1 1 1 1 1 1 1 M 1 Illtll t 1 1 1 1 1 1
5 10 2D
OUTDOOR CONCENTRATION
50
Figure 8-1. Average personal nitrogen dioxide exposure for each household compared
with outdoor concentrations for summer and whiter.
Source Quackenboss et al (1986)
8-7
-------
NITROGEN DIOXIDE LEVELS IN SUMMER
t5""
E
i
^?
LU
CC
u
CO
2
1
o
CC
LU
Q.
LU
CD
2
LU
$
/u
60
50
40
30
20
10
9
8
7
6
5
„
—
_
: *
:L
: **
O $ * J-
r ° ° * * **
— .ojjc * ;(: >k
=• ° Q0° 0o^°a>0°Xs^
- 0*00«&%>^>J?' *
:o ° ° ° LEGEND
V
O ^t
I 0 ° ° Electric
r * Gas
_
' i i 1 1 1 1 1 1 1 1 i i i 1 1 1 1 1 1 1 i i i 1 1 1 1 1 1 1 1 ii ii MI
5 10 20 50
HOME AVERAGE CONCENTRATION
NITROGEN DIOXIDE LEVELS IN WINTER
fT 100
E
•S
LU 50
CC
1
§ 20
tr 1
O 10
V)
CC
Q.
UJ 5
CD
jjg
LU
2j -
^
f-
r~
£
I
=
z.
-
\ °
; o
z.
~ o o°
ro°
~~
~^-
:.
E_
"
-i ii i In 11 hnil
%
*
* * * *
* ****
* * .»(*** *
0 ° Q, **
ik ^
0 ° **
oto00^0
(^>°o^ ^o
% "
LEGEND
0 Electric
* Gas
1 i 1 illlll i I nhllll i i II Illlllillil i 1 i hhl ih
2 5 10 20 50
HOME AVERAGE CONCENTRATION
100
Figure 8-2. Average personal nitrogen dioxide exposure for each home compared with
average indoor concentrations for summer and winter.
Source Quackenboss et al (1986)
8-8
-------
due in part to the low outdoor levels measured in Portage, WI, during the study (typically
less than 20 /*g/m3 [see Table 7-4])
The tune/activity diary results reported by Quaickenboss et al (1986) showed that the
subjects spent more than 65% of their tune at home, whereas about 15% of their tune was
spent outdoors in the summer and 5% in the winter Approximately 4% of the time was
spent in motor vehicles, up to 15 % at work or school, and approximately 8 % in other indoor
environments The study estimated personal exposures from home and average indoor and
outdoor levels weighted by the proportion of tune spent there for three categories of people
(student, worker, and other) by season and stove type The estimated exposures were then
compared to the measured personal exposures The explained variance ranged from 3 to
71 % The measured exposures were generally not well predicted by the estimated
exposures
In an earlier study (a pilot for the one discussed above), Quackenboss et al (1982)
measured personal NO2 exposures (from 5 to 7 days) of 66 family members in 19 homes in
Portage, WI, and NO2 concentrations in a bedroom, kitchen, and outdoors for each
residence Time budgets were obtained for each subject and information on cooking fuel was
obtained for each house Personal exposures were found to be strongly associated with
bedroom (r = 0 84, gas, r = 0 63, electric) and kitchen (r = 0 71, gas, r = 0 6, electric)
concentrations for both the gas and electric cooking homes, but poorly associated with
outdoor concentrations (r = 0 4) Several models using microenvironment monitoring, time
budgets, and household characteristics were applied to the data to relate average personal
NO2 exposures to indoor concentrations The models explained as much as 90% of the
variation in personal exposures
As part of an epidemiological study of the impact of kerosene heaters on health,
Leaderer et al (1986) compared personal 1-week NO2 measurements on 23 adult subjects
with 1-week NO2 measurements made in three locations in the subjects' homes (kitchen,
living room, and bedroom) and outdoors The homes monitored had a mix of NO2 sources
(gas stove, kerosene heater, gas stove and kerosene heater, and no known source) Eighty
percent of the variation in total personal NO2 exposures were accounted for by variations in
indoor average NO2 concentrations for homes with the mix of sources (Figure 8-3) Total
personal NO2 exposures were found to be 90% of the house average concentration When
8-9
-------
140-
120-
100-
i
60-
1
I 40-
20-
• Kerosene Heaters and Gas Stoves
• No Source
A Gas Stoves
V Kerosene Heaters
V T
A
I
20
40 60 80
Average NO2/House (wj/rr?)
100
120
140
Figure 8-3. Comparison of the house average 2-week nitrogen dioxide concentrations
with the total personal nitrogen dioxide levels measured over the same tune
period for one adult resident in each house, New Haven, CT, area, whiter
1983. Each house is identified by the type of sources hi the house and the
fitted regression line is presented.
Source: Leaderer et al (1986)
compared to concentrations measured in different locations in the house, the bedroom
concentration was the best predictor (R2 = 0 88) Total personal exposures to NO2 were
not found to be related to outdoor concentrations of NO2 measured at each residence
In a recent study, Harlos et al (1987) obtained 1-day personal NO2 measurements for
15 infants in homes with gas ranges Corresponding NO2 measurements were made in the
infants' bedrooms and in the living room and kitchen of each house As might be expected,
the measured personal exposure correlated well with the bedroom concentrations (r = 0 78)
An effort to model the infants' personal exposure based upon the infants' activity patterns
8-10
-------
and room concentrations resulted in a slightly stronger relationship with measured personal
exposure (r = 0 82)
In a study of 500 junior high students in Waterlown, MA (Clausmg et al, 1986),
personal NO2 concentrations were measured over a 3- to 4-day period between November
and December of 1982 Nitrogen dioxide measurements were made in the bedroom, living
room, kitchen, and outside for 200 of the homes of the students on a tune scale that
corresponded to the personal monitoring Tune/activity diaries were kept by the subjects and
a questionnaire was utilized to obtain information on home characteristics, particularly
sources of NO2 and their use A variety of models utilizing indoor and outdoor measured
NO2 concentrations, tune activity information, and home characteristics were explored to
explain variations in measured personal NO2 exposure The correlation between personal
exposure and outdoor concentrations was not significant at the 0 05 level Models that
included indoor NO2 concentrations explained from 60 to 90% of the variation in personal
NO2 exposures Excluding indoor concentrations and using cooking range characteristics
(gas range, presence of pilot lights, etc ) explained only 40% of the variation in personal
NO2 exposures
As part of a large study on indoor air pollution in the Netherlands (a sample of over
900 homes), a small study was conducted on a subsample of 14 families (11 mothers,
11 school children, and 8 preschool children) to determine relationships between measured
indoor NO2 concentrations and personal exposures (Hoek et al, 1984) Weekly average
personal exposures were measured for mothers and for primary (mean age = 78 years) and
preschool (mean age = 34 years) children, and weekly NO2 concentrations were monitored
in several locations A self-administered questionnaiie was used to obtain tune/activity
patterns Measured and calculated (from microenvircnment measurements and tune/activity
iy
data) personal exposures were highly correlated (R from 82 % for the whole group to 92 %
for the preschool children), but the calculated levels were on average 20% lower than the
measured personal exposures The mothers' exposure was found to be a good predictor of
o
the primary and preschool children (R of 91% and 93%, respectively), but overestimated
the children's exposure by approximately 20%
One-day personal exposures and corresponding microenvironment NO2 concentrations
(home, workplace, and outdoors) were measured for 20 housewives and 44 office workers
8-11
-------
(same office building) in Tokyo (Nitta and Maeda, 1982) Activity diaries were obtained
during the sampling period. Repeat sampling during a different season was conducted on a
subsample of the population Outdoor concentrations were found to be weak predictors of
personal exposures, with correlation coefficients no higher than 0 33 for all comparisons by
group (housewife or worker) and season Housewife personal exposures compared well with
both modeled personal exposures (from microenvironmental measurements and activity
diaries) for winter (r = 0 88 and 0 89) and moderately for the summer (r = 0 62 and 0 57)
Results for the office workers were similar
In another study, daily averages of personal NO2 exposures and indoor and outdoor
NO2 concentrations were measured for 40 housewives and their preschool children
(subsample) living in and near Tokyo for a winter, spring, and summer penod (Yanagisawa
et al., 1984) An additional nine housewives were monitored for seven 1-day periods each
month for a year. Gas ranges were used for cooking in almost all houses Activity diaries
and information on home characteristics were obtained Outdoor concentrations were not
significantly related to personal exposures Indoor levels were above outdoor levels, with the
living room and bedroom concentration averages close to the personal exposures of the
housewives and preschool children for the different locations and seasons Kitchen
concentrations were on average higher than personal exposures Measured and calculated
personal exposures for the nine housewives agreed well
Hoek et al (1984) monitored 12 primary school children for a 1-week penod for NO2
exposures as part of a study of indoor air quality and respiratory symptoms Time/activity
patterns and NO2 concentrations in the bedroom, living room, and kitchen of the homes of
the subjects were also obtained Outdoor NO2 concentrations were obtained from a central
monitoring site The homes contained geysers (unvented, gas-fired, hot water sources at the
water tap), an important source of NO2 Living room and bedroom concentrations best
predicted personal exposures (R2 of 0 64 and 0 63), whereas the kitchen concentrations
explained 35% of the variation of personal exposures Overall, 84% of the variation in
personal NO2 exposure was explained by variations in the modeled exposure, using
time/activity and microenvironment measurements
Dockery et al. (1981) measured 1-week personal NO2 exposures along with bedroom,
kitchen, and outdoor concentrations for nine families in Topeka, KS, during a summer
8-12
-------
period The homes had a mix of electric and gas cooking ranges, and activity diaries were
obtained Mean personal exposures were similar to the bedroom means Measured and
calculated personal exposures agreed well, with 86% of the variation being explained
As part of an epidemiologic study of an association between personal exposure to NO2
and respiratory illness in Hong Kong (Koo et al, 1990), 24-h personal measurements of NO2
using passive badges (Yanagisawa and Nishimura, 1982) were acquired on 362 children
age 13 and under and attending the same school, and 319 of their mothers The passive
badges were also hung in the school classrooms and school playground A questionnaire was
used to obtain data on indoor sources of air pollution (e g , smoking habits of family
members, types of heating and cooking fuels, frequency of cooking, ventilation patterns,
burning of incense and mosquito coils, and mother's exposure to dust or fumes in the
workplace) The sampling was conducted during warm weather (average ambient
temperature of 27 °C) Ventilation was supplied to the school through open windows
Variations in classroom NO2 concentrations explained 56% of the variation in personal
exposure levels of the children Personal NO2 exposures of the mothers was not significantly
correlated with the personal exposures of their children (p > 0 05) Neither the children's
nor the mothers' personal NO2 exposure related to number of cigarettes smoked in the home
or the number of hours of exposure to cigarette smoke The children's exposure was not
related to cooking and heating habits, whereas the mothers' level was highest for liquified
petroleum gas and kerosene users and lowest for piped gas users Increases in mothers'
personal NO2 levels were also seen when kitchens did not have ventilating fans (11%), when
incense was burned in the home (10%), and when the mother reported dust exposure in the
workplace (21 %)
The largest field study of personal exposures reported to date was conducted in the
Los Angles Basin (Schwab et al, 1990, Spengler et al, 1992, Colome et al, 1992) In this
study, 700 individuals from 500 households were moinitored for two consecutive 24-h periods
for their integrated NO2 exposure using Yanagisawa filter-badge monitors In addition, NO2
concentrations were monitored in both the bedroom and outside the home of each participant
over the 48-h period that they were monitored for peisonal exposures Data were collected
on housing units (including indoor sources), personal characteristics, and tune/activity
patterns The objective of the study was to investigate the seasonal, spatial, and
8-13
-------
demographic trends in personal NO2 exposure and the influence of indoor and outdoor NO2
concentrations and activity patterns on NO2 exposure Seasonal variations in outdoor,
indoor, and personal exposures were apparent Concentrations away from home averaged
0.007 to 0 030 ppm, depending on the cycle, higher than the at-home concentrations
Personal exposures of participants who cooked with gas averaged 0 010 ppm to 0 028 ppm
higher than those without gas
Several analysis techniques (stepwise regression, analysis of variance, and path
analysis) were applied to the collected data to determine the relative contribution to personal
NC>2 exposures of factors hypothesized to influence those exposures The factors considered
in the analysis included indoor and outdoor concentrations of NO2, activity patterns of the
subjects; type of cooking range (electric, gas without pilot lights, and gas with pilot lights),
season (winter versus summer), location (potential high, medium, and low ambient NO2
levels); and six population subgroups defined by age, sex, and employment status In the
analysis, the two consecutive 24-h personal NO2 measurements were averaged, with the
resultant variable used as the dependent variable in the analysis
The analysis indicated that the tune spent in each of the microenvironments considered
was a poor predictor of personal exposure (R typically less than 0 10) The explained
variation in personal exposures increased to 55 % when the bedroom concentration was
included in the model and decreased to 42 % when only outdoor concentrations were
considered. Only small differences in the time/microenvironment relationships were
observed between groups defined by range-type or population subgroups (explained variation
changes of 1 to 10% were observed) Analysis of variance showed that cooking range type,
season, and geographic location explained 30% of the personal NO2 exposure and the
addition of outdoor NO2 levels increased the explained variation to 54% Inclusion of
bedroom concentration increased the explained variation to 62% The model with outdoor
ty
levels alone produced an R of 0.48, whereas using bedroom concentrations alone resulted in
an R2 of 0 59.
Colome et al (1992) indicate that in the higher ambient NO2 levels, observed in
Los Angeles, personal concentrations measured outside the home were generally higher than
personal concentrations encountered while at home, reinforcing the importance of outdoor
concentrations of NO2 Models developed for each of the population subgroups using
8-14
-------
bedroom and outdoor NO2 concentrations and range lype resulted in explained variations in
personal NO2 levels ranging from 60 to 78% The study estimates that in the Los Angeles
Basin, 68 % of the variation in personal NO2 exposures is explained by measured outdoor and
indoor concentrations and that 32% of the variation is unexplained by the parameters
measured The authors noted that tune/activity measures by themselves are weak predictors
of exposure, whereas surrogates of exposure such as cooking range type and location are
relatively good predictors Outdoor NO2 concentrations in the Los Angeles Basin are
considerably higher than in most other locations in the United States and show more
variability in tune than most locations in the United States, and as such, aspects of the results
of this study may not be applicable to other regions, especially those with lower ambient
NO2 levels Following the discussion of a similar study in Boston, discussed next, is a
presentation of comparisons between the Boston and Los Angeles results
One additional large field study of personal exposure to NO2 was conducted in the
Boston area (Ryan et al , 1988, Ryan and Spengler, 1992, Ryan et al , 1992) This study of
approximately 400 homes was conducted during 1985 and 1986 by the same research group
using the same protocols as discussed above in the Los Angeles study Ryan and Spengler
(1992) state that the response rate of approximately 60 % was comparable to other exposure
studies In Boston, outdoor NO2 contributes to indoor NO2 levels, but the outdoor levels
explain less than 10% of the variation in indoor levels in the gas range homes The
important sources of indoor NO2 are gas range use and the presence of a continuously lit
pilot light Levels in the kitchen are about twice those experienced in other rooms of the
house Other sources and home factors are relatively minor modifiers of indoor
concentrations
Ryan and Spengler (1992) report that, in Boston, the homes with gas ranges that have
pilot lights show consistently 0 010 to 0 015 ppm higher NO2 concentrations across the
distribution than for homes with a gas range without pilot lights For homes with gas
cooking, the indoor concentrations always exceeded outdoor concentrations for those with
pilot lights The highest NO2 concentrations for gas cooking homes were measured in the
kitchen, followed by the living room, then the bedroom Kitchen concentrations exceeded
ambient levels by 0 014 to 0 015 ppm Exploring other nongas combustion variables,
including wood burning stove, smoking, air-conditiomng, and fireplace, adds little intuitive
8-15
-------
or quantitative understanding The most consistent influence is seen in lower indoor NO2
levels with the presence of a wood stove It is believed that this variable is a surrogate for
other factors not immediately apparent The ratio of indoor to outdoor NO2 concentrations
reveals the presence or absence of sources For electric range homes, the geometric mean
indoor/outdoor (I/O) ratio is approximately 0 4 in the fall, 0 75 in the summer, and 0 5 in
the winter sampling periods The I/O ratio is clearly higher for the homes with gas pilot
lights. But the overall geometric mean is approximately 1 25, 1 20, and 1 50 for the fall,
summer, and winter sampling periods, respectively For the Boston Standard Metropolitan
Statistical Area, presence of gas range results in an effective increase in NO2 concentration
of 0.010 to 0 020 ppm depending on season The variability in this effective source is large
with a coefficient of variation of approximately 100% The source of this variability
includes home characteristics such as presence of pilot lights or other unvented appliances,
and use patterns
Ryan et al (1992) indicate that, in Boston, bedroom NO2 concentrations were
significantly related to personal NO2 exposures Personal exposures of individuals living in
homes with gas ranges with pilot lights were 0.015 ppm greater, on average, than the NO2
exposures of those with electric ranges The difference between NO2 exposures of
participants living in homes with electric appliances versus those living in homes with ranges
with no pilot lights was less, about 0 Oil ppm
Ryan et al. (1992) provide comparisons between the Boston and Los Angeles results
There are several similarities between the results of the analysis of the Boston Personal
Monitoring Study and the results of the Los Angeles Personal Exposure Study Specifically,
no strong correlations between personal exposure and the amount of time individuals spent in
each activity category were found in either study Also, bedroom concentrations have a
strong influence on personal exposure in both cities In Boston, 48 % of the variation in
personal NO2 exposure is explained by variations in bedroom concentrations, whereas the
figure is 53 % in Los Angeles
There are also some salient differences in the multivanate analysis results between
Boston and Los Angeles Specifically, in Los Angeles, the outdoor concentration appeared
to play a dominant role in influencing indoor concentrations, 40% of the variation in
bedroom concentrations is explained by variations in outdoor concentrations The outdoor
8-16
-------
concentrations also played a strong influence on personal exposure (R is 41) In Boston,
however, this is not the case, only 9 % of the variation in bedroom concentrations is
explained by outdoor concentrations The correlation between personal exposure and outdoor
concentrations is even weaker (R2 = 0 05) A portion of this reduced explanatory power
may be due to the use of previous-year data, rather than concurrent ambient data
Additionally, although in both cities those who have a gas range have personal exposure
levels that are 0 010 ppm higher than the personal exposure levels of individuals with electric
ranges, the entire scale is 0 010 ppm higher in Los Angeles
The measured data on personal NO2 exposures indicate that outdoor measurements of
NO2 (measured in the vicinity of residences) by themselves only explain typically less than
50% of the variations in personal exposure, primarily because of the small amount of tune
individuals spend outdoors, the impact of buildings on removing outdoor NO2, and the
important contnbution from indoor sources Outdoor NO2 concentrations, however,
dominate indoor levels when there are no indoor sources and when outdoor concentrations
are high (e g , in Los Angeles) No studies have been reported that examine the relationship
between NO2 concentrations measured at central outdoor sites and personal NO2 exposures
It is likely, however, given the outdoor spatial variability of NO2, that in general NO2 levels
measured at central sites may be poor predictors of personal NO2 exposures In some
specific outdoor locations, central sites may be bettei than others Nitrogen dioxide
concentrations measured in the living room or bedroom of a house or whole-house average
measurements are better predictors of personal exposure than outdoor measurements for the
population as a whole It is important to note that there may be significant segments of the
population (e g , infants, police, etc ) for which indoor NO2 levels may not be good
predictors of their exposure Calculated exposures from microenvironmental measurements
and activity diaries are good indicators of personal exposure
8.3 INDIRECT METHODS
Indirect methods employ various degrees of microenvironmental monitoring and
questionnaires to estimate an individual's or population's total NO2 exposure They attempt
to measure and understand the underlying relationships between personal exposure and the
8-17
-------
variables causing exposure so that NO2 exposures in other populations in other locations can
be estimated Such models can provide exposure frequency distributions for the entire
population or segments of the population
Estimates of NO2 concentrations in various microenvironments (microenvironmental
monitoring) can provide information on the spatial and temporal distribution of such
concentrations in those environments and the factors (sources, emission rates, removal and
dispersal mechanisms, etc ) affecting them Estimates of microenvironmental NO2
concentrations combined with time/activity patterns can be used to estimate or model total
NO2 exposure (i.e , Equation 8-1) Questionnaires are used to gather information on
microenvironmental factors (sources, source use, volume, etc ), human activities, or time
budgets or for the simple categorization of an individual's exposure status
Physical/chemical, empirical/statistical, or hybrid models have been used to estimate
NO2 concentrations in various microenvironments These models utilize inputs on sources
and emissions, air contaminant dispersal, contaminant reactions, removal mechanisms, and
responses to questionnaires Several attempts to model indoor concentrations of NO2 were
reviewed in Chapter 7 Section 7 3 should be referred to for a detailed discussion of efforts
to model indoor NO2 concentrations The indoor modeling approaches have been very
limited in applications because of the lack of data on the variability of the input parameters
(e.g , emission rates, source use, mixing, reactive decay rates, etc ) in actual indoor
environments and the lack of adequate questionnaires for obtaining information on building
or household characteristics Efforts are under way to attempt to standardize questionnaires
for use in indoor air quality studies (i e , Lebowitz et al, 1989a,b) Such efforts will
provide better input data to models for predicting indoor NO2 concentrations Ambient
models do not provide spatial and temporal NO2 concentrations on a scale needed to assess
concentrations outside of various microenvironments or at the interface between the outdoor
air and an individual No models exist for transit microenvironments
The primary method for determining exposures in epidemiologic studies of the effects
of NO2 has been the use of questionnaires (e g , Melia et al, 1977, 1979) Typically, the
questionnaires determined such information as the geographic location of a respondent's
residence, type of occupation, and whether a source exists in a house (gas cooking stove,
kerosene heater, etc ) Reliance on this type of data for exposure, as discussed earlier
8-18
-------
(Section 8 2), can result in serious misclassification errors due to faulty exposure
specifications and failure to adequately account for confounding factors These errors result
in the increased probability of masking or misrepresenting any exposure-effect relationships
that may exist It is not enough, for example, to determine a respondent's exposure status
based upon their response to the question of existence of a gas cooking range in their home
Chapter 7 indicates that there is a wide distribution of indoor NO2 levels in homes with gas
cooking ranges related to outdoor levels, indoor sources, indoor source use, condition of the
source, season of the year, indoor mixing, infiltration, and indoor building materials and
furnishings
Total personal exposures to NO2 are assessed indirectly by using a surrogate
microenvironmental measurement or by combining modeled microenvironmental
concentrations with tune/activity patterns Section 822 indicates that a whole-house NO2
measurement or a bedroom NO2 measurement is highly correlated with total personal NO2
exposure The good agreement is in large part due to the tune spent in the residential
environment by individuals Presumably, a model that predicts whole-house NO2 levels
could be used to represent personal exposures Combining microenvironmental monitoring
data or modeled concentrations with time/activity patterns (Equation 8-1) for an individual
has been used in a number of the studies discussed in Section 822 The results indicated
good agreement between measured personal exposures and those calculated from time/activity
patterns and microenvironmental monitoring Due to the potential for higher levels indoors
and the high level of tune spent indoors, outdoor levels by themselves are typically only
weak predictors of total exposure, although they can contribute substantiaEy to indoor levels
where people spend their tune In areas with high ambient NO2 levels, such as the
Los Angeles Basin, the influence of outdoor levels is greater than in areas where outdoor
levels are low and show little variability The tune/activity portions of these studies and the
extensive review of tune/activity studies conducted by Ott (1989) clearly demonstrate the
importance of the home environment in terms of a tune budget Figures 8-4a and 8-4b (Ott,
1989) highlight the portion of tune spent in the home by individuals employed outside the
home and by full-tune homemakers It should be emphasized again, however, that Figures
8-4a and 8-4b represent average tune budgets and that segments of the population can deviate
significantly from those averages
8-19
-------
INDOORS OTHER
5.4*
IN-TRANSIT
42*.
OUTDOORS
Figure 8-4a. Proportion of time spent by women who are full-time homemakers in
indoor, outdoor, and in-transit microenvironments.
Source' Data from time budget studies in 44 U S cities (Szalai, 1972, Robinson, 1977), additional
interpretation and analyses appear in Ott (1989)
IN TRANSIT
6%
OUTDOORS
2%
INDOORS OTHER
1%
Figure 8-4b. Proportion of time spent by employed persons in indoor, outdoor, and
in-transit microenvironments.
Source: Data from time budget studies in 44 U S cities (Szalai, 1972, Robinson, 1977), additional
interpretation and analyses appear in Ott (1989)
8-20
-------
8.3.1 Personal Exposure Models
Billick et al (1991) report an effort to develop a model predictive of personal NO2
exposure based upon microenvironmental measurements and time/activity patterns Data
used to develop the model come from eight different sites in the United States and were
obtained from three different studies (Wilson et al, 1986, Ryan et al, 1988, Spengler et al,
1987) The data base, reviewed all or in part in earlier publications (e g , Butler et al ,
1990, Drye et al, 1989, Spengler et al, 1989), contains over 6,400 measurements of NO2
(Palmes tubes) in over 1,700 households and represents a range of exposure levels over a
diverse set of geographic and climatic conditions in the U S The model was developed in
two steps The first step was to develop a model predictive of indoor/outdoor and of
in-vehicle/outdoor levels of NO2 The second step involved the integration of time/activity
data with the models developed in the first step
The indoor/outdoor residential model was developed from a simplified version of the
general mass-balance equation (see Equations 7-1, 7-2, and 7-3) The model (Sexton et al,
1983, Drye et al , 1989) is given below
Cm = ™Cout 4 b, (8-2)
where Cm is the indoor concentration (micrograms per cubic meter), m is the penetration
coefficient for outdoor NO2, Cout is the outdoor concentration (micrograms per cubic meter),
and b is the concentration contribution by indoor sources (micrograms per cubic meter)
In the above model, m and b are estimated from a multivartate analysis of data collected in
the eight studies The model was estimated for NO2 concentrations separately for indoor
location (e g , bedroom, kitchen, in-vehicle), season (winter, summer), and for cooking
range type (gas versus electric) All the models used measured outdoor NO2 levels and the
models for gas versus electric ranges used the inverse volume of the dwelling The model
for electric range contained a dichotomous independent variable for the presence of a gas
furnace, whereas the model for gas cooking range dwellings was characterized by whether
the appliance employs a continuously burning pilot light and whether a microwave oven is
present An earlier version of this model (Drye et al, 1989), based on a more limited data
set, is discussed in Chapter 7
8-21
-------
The model for in-vehicle/outdoor relationships uses data obtained in a study of NO2
concentration measured inside and outside vehicles (Chan et al, 1990) The predictive
models fitted from the data for dwellings are shown in Tables 8-1 and 8-2 The
in-vehicle/outdoor model uses a range of indoor/outdoor ratios from 0 2 to 3 32 The fitted
o
models resulted in R values ranging from 0 12 to 0 64, indicating that from 36 to 82% of
the variation in indoor variations in NO2 levels remained unexplained
TABLE 8-1. ELECTRIC-RANGE HOME LEAST SQUARES REGRESSION
COEFFICIENTS AND T-STATISTICS (IN PARENTHESES)3
Kitchen
Variable
Ambient NO2 (ppb)
l/(Home Volume)b
Furnace Fuel
(gas = 1)
R2
Observations
Summer
0 61 (28 44)
-2 52 (0 59)
4 50 (8 16)
058
646
Winter
0 52 (15 27)
42 22 (5 39)
-0 43 (0.44)
021
754
Bedroom
Summer
0 58 (32 15)
-2 87 (0 81)
3 64 (7 87)
064
641
Winter
0 43 (11 23)
41 81 (4 65)
-0 01 (0 01)
0 12
736
Regression Equation- Indoor NO2 = /Jl (ambient NO2 level) + /J2 (I/home volume) + j83 (gas furnace
present)
8Shaded t-statistics values are nonsignificant at the 95% confidence level NO2 = Nitrogen dioxide
Units. I/thousands of cubic feet
Source BiUicketal (1991)
The model used to estimate the tune-weighted average NO2 exposure was presented
simply as the sum of the products of the tune spent in each of the three environments
(in dwellings, outdoors, and in vehicles) multiplied by the predicted concentration in each
environment, divided by the sum of the tune spent in each environment (Equation 8-1) The
concentrations in each environment were determined from the above model (Equation 8-2),
whereas the time in each environment is developed from simulations formed by taking
random draws from distributions of observed amounts of time spent outdoors and in vehicles
(e.g., Schwab et al., 1990) The utility of the model stems from its capacity to estimate a
8-22
-------
TABLE 8-2. GAS-RANGE HOME LEAST SQUARES REGRESSION COEFFICIENTS
AND T-STATISTICS (IN PARENTHESES)3
Kitchen
Variable
Ambient NO2 (ppb)
No Pilot/Microwave
No Pilot/No Microwave
Pilot/Microwave
Pilot/No Microwave
l/(Home Volume)b
R2
Observations
Summer
085
251
405
1430
1666
2821
(29
08)
(1.69)
(1
(11
(10
(3
058
876
96)
10)
17)
31)
Winter
070
748
739
26 11
2896
3785
(20
(3
50)
06)
(2,136)
(12
(10
(2
041
952
69)
47)
59)
Bedroom
Summer
070
122
256
685
760
2804
(34
a>
(1,
(7
(6
(4
063
868
42)
17)
77)
55)
62)
55)
Winter
053
2 14
-034
11 92
1306
5620
(16 60)
-------
8.4 SUMMARY
Exposure to NO2 occurs across a number of microenvironments or settings
An individual's integrated exposure is the sum of all of the individual NO2 exposures over all
time intervals for all microenvironments, weighted by the tune in each microenvironment
(Fugas, 1976, Sexton and Ryan, 1988, National Research Council, 1991) The assessment
of human exposures to NO2 can be represented by the following simplified basic model
(8-1)
Accurate assessments of total NO2 exposure and the environments in which exposures take
place are essential to minimize misclassification errors in epidemiologic studies (Shy et al ,
1978), in defining population exposure distributions in risk assessment, and in developing
effective mitigation measures in risk management
Personal NO2 exposures can be assessed by direct and indirect measures Direct
measures include biomarkers and use of personal monitors (Yanagisawa et al , 1986, 1988,
Maples et al., 1991). No validated biomarkers for exposure are presently available for NO2
A limited number of studies have been conducted in which personal exposures to NO2 were
measured using passive monitors (Quackenboss et al , 1986, Leaderer et al , 1986, Harlos
et al., 1987). These studies generally indicate that outdoor levels of NO2, although related
to and contributing substantially to both personal levels and indoor concentrations, are by
themselves poor predictors of personal exposures for most populations Average indoor
residential concentrations (e g , whole-house average or bedroom level) tend to be the best
predictor of personal exposure, typically explaining 50 to 60% of the variation in personal
exposures. In selected populations, the indoor residential environment may not be a good
predictor of total exposure because of the higher percentages of tune spent in different
environments and/or the potential for unusual NO2 concentrations
Indirect estimations of personal exposure to NO2 employ various degrees of both
microenvironmental monitoring and questionnaires to estimate an individual's or population's
total exposure. One such model, developed from an extensive monitoring and questionnaire
data base, can estimate population exposure distributions from easily obtained data on
outdoor NO2 concentrations, housing characteristics, and tune/activity patterns (Billick et al ,
8-24
-------
1991) This indoor/outdoor residential model (Sexton et al, 1983; Drye et al, 1989),
developed from a simplified version of the general mass-balance equation (see Equations 7-1,
7-2, and 7-3), is given below
Cin = mCout + b, (8-2)
where Cm is the indoor concentration (micrograms per cubic meter), m is the penetration
coefficient for outdoor NO2, Cout is the outdoor concentration (micrograms per cubic meter),
and b is the concentration contribution by indoor sources (micrograms per cubic meter) The
model used to estimate the tune-weighted average NO2 exposure was presented simply as the
sum of the products of the tone spent in each of the three environments (in dwellings,
outdoors, and in vehicles) multiplied by the predicted concentration m each environment,
divided by the sum of the time spent m each environment (Equation 8-1) This model is
proposed for use m evaluating the impact of various NO2 mitigation measures The model is
promising, but it has not yet been validated and the uncertainty associated with it has not
been characterized
8-25
-------
REFERENCES
Bilhck, I. H (1990) Estimation of population exposure to nitrogen dioxide Toxicol Ind Health. 6 325-335
Bilhck, I H , Ozkaynak, H , Butler, D A , Spengler, J D (1991) Predicting personal exposures to NO2 for
population-based exposure and risk evaluations Presented at 84th annual meeting of the Air and Waste
Management Association, June, Vancouver, BC, Canada Pittsburgh, PA Air and Waste Management
Association, paper no 91-172 9
Butler, D A , Ozkaynak, H , Bilhck, I H , Spengler, J D (1990) Predicting indoor NO2 concentrations as a
function of home characteristics and ambient NO2 levels In Indoor air '90 precedings of the 5th
international conference on indoor air quality and climate, volume 2, characteristics of indoor air,
July-August, Toronto, ON, Canada Ottawa, ON, Canada International Conference on Indoor Air
Quality and Climate, Inc , pp 519-524
Chan, C (1990) Commuter exposures to volatile organic compounds [Ph D dissertation] Boston, MA Harvard
School of Public Health
Chapin, F. S , Jr (1974) Human activity patterns in the city New York, NY John Wiley & Sons (Clark,
T. N , ed Wiley series in urban research)
Clausing, P , Mak, J K , Spengler, J D , Letz, R (1986) Personal NO2 exposures of high school students
Environ Int 12 413-417
Colome, S D , Wilson, A L , Spengler, J D (1992) Nitrogen dioxide exposure studies—volume V, personal
exposure to nitrogen dioxide in the Los Angeles basin a microenvironmental approach Chicago, IL Gas
Research Institute, report no GRI-92/0427
Dockery, D W.; Spengler, J D , Reed, M P , Ware, J (1981) Relationships among personal, indoor and
outdoor NO2 measurements Environ Int 5 101-107
Doscmeci, M , Wacholder, S , Lubin, J H (1990) Does nondifferential misclassification of exposure always
bias a true effect toward the null value'' Am J Epidenuol 132 746-748
Drye, E E., Ozkaynak, H , Burbank, B , Bilhck, I H , Baker, P E , Spengler, J D , Ryan, P B , Colome,
S D (1989) Development of models for predicting the distribution of indoor nitrogen dioxide
concentrations JAPCA 39 1169-1177
Duan, N (1981) Microenvironment types a model for human exposure to air pollution Stanford, CA Stanford
University, Dept of Statistics, SIMS technical report no 47
Fugas, M (1976) Assessment of total exposure to an air pollutant In International conference on environmental
sensing and assessment, v 2, September, Las Vegas, NV New York, NY Institute of Electrical and
Electronic Engineers, paper no 38-5
Gladcn, B , Rogan, W J (1979) Misclassification and the design of environmental studies Am J Epidenuol
109 607-616
Harlos, D P , Marbury, M , Samet, J , Spengler, J D (1987) Relating indoor NC^ levels to infant personal
exposures Atmos Environ 21 369-376
Hoek, G., Brunekreef, B , Meyer, R , Scholten, A , Boley, J (1984) Indoor nitrogen dioxide pollution and
respiratory symptoms of schoolchildren Int Arch Occup Environ Health 55 79-86
8-26
-------
Koo, L C , Ho, J H-C , Ho, C -Y , Matsuki, H , Shimizu, H , Mori, T , Tominaga, S (1990) Personal
exposure to nitrogen dioxide and its association with respiratory illness in Hong Kong Am Rev Respir
Dis 141 1119-1126
Lambert, W E , Samet, J M , Stidley, C A , Spengler, J D (1990) Classification of chrome exposure to
nitrogen dioxide In Indoor air '90 precedings of the 5th international conference on indoor air quality
and climate, volume 2, characteristics of indoor air, July-August, Toronto, ON, Canada Ottawa, ON,
Canada International Conference on Indoor Air Quality and Climate, Inc , pp 537-542
Leaderer, B P , Zagramski, R T , Berwick, M , Stolwyk, J A J (1986) Assessment of exposure to indoor air
contaminants from combustion sources methodology and application Am J Epidemiol 124 275-289
Lebowitz, M D , Quackenboss, J J , Soczek, M L , Kollander, M , Colome, S (1989a) The new standard
environmental inventory questionnaire for estimation of indoor concentrations JAPCA 39 1411-1419
Lebowitz, M D , Quackenboss, J J , Soczek, M L , Colome, S D , Lioy, P J (1989b) Workshop
development of questionnaires and survey instruments In Nagda, N L , Harper, J P , eds Design and
protocol for monitoring indoor air quality Philadelphia, PA American Society for Testing and Materials,
pp 203-216 (ASTM special technical publication 1002
Lebret, E (1990) Errors in exposure measures Toxicol Ind Health 6 147-156
Maples, K R , Sandstrom, T , Su, Y -F , Henderson, R F (1991) The nitric oxide/heme protem complex as a
biologic marker of exposure to nitrogen dioxide in humans, rats, and in vitro models Am J Respir
CellMol Biol 4 538-543
Meha, R J W , Florey, C du V , Altman, D G , Swan, A V (1977) Association between gas cooking and
respiratory disease in children Br Med J 2 149-152
Meha, R J W , Florey, C du V , Chinn, S (1979) The relation between respiratory illness in primary
schoolchildren and the use of gas for cooking I—results from a national survey Int J Epidemiol
8 333-338
Muelenaer, P , Reid, H , Morns, R , Saltzman, L , Horstman, D , Collier, A , Henderson, F (1987) Urinary
hydroxyprohne excretion in young males exposed experimentally to nitrogen dioxide In Seifert, B ,
Esdorn, H , Fischer, M , Rueden, H , Wegner, J , eds Indoor air '87 proceedings of the 4th
international conference on indoor air quality and climate, v 2, environmental tobacco smoke,
multicomponent studies, radon, sick buildings, odours and irritants, hyperreactivities and allergies,
August, Berlin, Federal Republic of Germany Berlin, Fede ral Republic of Germany Institute for Water,
Soil and Air Hygiene, pp 97-103
National Research Council (1989) Biologic markers of pulmonary toxicology Washington, DC National
Academy Press
National Research Council (1991) Human exposure assessment for airborne pollutants advances and
opportunities Washington, DC National Academy of Sciences
Nitta, H , Maeda, K (1982) Personal exposure monitoring to nitrogen dioxide Environ Int 8 243-248
Ott, W (1989) Human activity patterns a review of the literature for estimating time spent indoors, outdoors,
and in-transit In Starks, T H , ed Proceedings of the research planning conference on human activity
patterns Las Vegas, NV US Environmental Protection Agency, Environmental Monitoring Systems
Laboratory, chapter 3, EPA report no EPA-600/4-89-004 Available from NITS, Springfield, VA,
PB89-166599
8-27
-------
Ozkaynak, H , Ryan, P. B , Spengler, J D , Laird, N M (1986) Bias due to misclassification of personal
exposures in epidermologic studies of indoor and outdoor air pollution In Berglund, B , Berglund,
U., Lindvall, T , Spengler, J , Sundell, J , eds Indoor air quality papers from the third international
conference on indoor air quality and climate, August 1984, Stockholm, Sweden Environ Int
12- 389-393
Palmes, E D., Gunmson, A F , DiMattio, J , Tomczyk, C (1976) Personal samplei for nitrogen dioxide
Am Ind Hyg Assoc J 37 570-577
Quackenboss, J J , Kanarek, M S , Spengler, J D , Letz, R (1982) Personal monitoring for nitrogen dioxide
exposure, methodological considerations for a community study Environ Lit 8 249-258
Quackenboss, J J , Spengler, J D , Kanarek, M S , Letz, R , Duffy, C P (1986) Personal exposure to
nitrogen dioxide relationship to indoor/outdoor air quality and activity patterns Environ Sci Technol
20 775-783.
Robinson, J. P. (1977) How Americans use their time a social psychological analysis of everyday behavior
New York, NY Praeger Publishers
Ryan, P. B , Koutrakis, P (1990) Indoor air chemistry an emerging field In Indoor air '90 precedings of the
5th international conference on indoor air quality and climate, volume 2, characteristics of indoor air,
July-August, Toronto, ON, Canada Ottawa, ON, Canada International Conference on Indoor Air
Quality and Climate, Inc , pp 489-494 »
Ryan, P B ; Spengler, J D (1992) Nitrogen dioxide exposure studies—volume n, the Boston residential home
characterization study Chicago, IL Gas Research Institute, report no GRI-92/0424
Ryan, P B , Soczek, M L , Treitman, R D , Spengler, J D , Bilhck, I H (1988) The Boston residential NO2
characterization study—H survey methodology and population concentration estimates Atmos Environ
22 2115-2125
Ryan, P B , Schwab, M , Spengler, J D (1992) Nitrogen dioxide exposure studies—volume m, personal
exposure to nitrogen dioxide in Boston a microenvironmental approach Chicago, IL Gas Research
Institute, report no GRI-92/0425 2
Schwab, M , Colome, S D , Spengler, J D , Ryan, P B , Bilhck, I H (1990) Activity patterns applied to
pollutant exposure assessment data from a personal monitoring study in Los Angeles Toxicol Ind
Health 6 517-532
Sexton, K ; Ryan, P B (1988) Assessment of human exposure to air pollution methods, measurements, and
models. In Watson, A Y , Bates, R R , Kennedy, D , eds Air pollution, the automobile, and public
health Washington, DC National Academy Press, pp 207-238
Sexton, K., Letz, R , Spengler, J D (1983) Estimating human exposure to nitrogen dioxide an indoor/outdoor
modeling approach Environ Res 32 151-166
Shy, C. M , Kleinbaum, D G , Morgenstern, H (1978) The effect of misclassification of exposure status m
epidemiological studies of air pollution health effects Bull N Y Acad Med 54 1155-1165
8-28
-------
Spengler, J D , Ware, J , Speizer, F , Ferns, B , Dockery, D , Lebret, E , Brunnekreef, B (1987) Harvard's
indoor air quality respiratory health study In Seifert, B , Esdorn, H , Fischer, M , Rueden, H ,
Wegner, J , eds Indoor air '87 proceedings of the 4th international conference on indoor air quality and
climate, v 2, environmental tobacco smoke, multicomponent studies, radon, sick buildings, odours and
irritants, hyperreactivities and allergies, August, Berlin, Federal Republic of Germany Berlin, Federal
Republic of Germany Institute for Water, Soil and Air Hygiene, pp 218-223
Spengler, J , Ryan, P B , Schwab, M , Colome, S , Wilson, A L , Billick, I, Becker, E (1989) An overview
of the Los Angeles personal monitoring study Presented alt First international symposium on total
exposure assessment methodology a new horizon Las Vegas, NV, pp 66-85
Spengler, J D , Ryan, P B , Schwab, M , Colome, S D , Wilson, A L (1992) Nitrogen dioxide exposure
studies—volume IV, personal exposure to nitrogen dioxide in the Los Angeles basin Chicago, IL Gas
Research Institute, report no GRI-92/0426
Szalai, A , ed (1972) The use of tune daily activities of urban and suburban populations in 12 countries The
Hague, The Netherlands Mouton and Co
Wiley, J A , Robinson, J P , Piazza, T , Garrett, K , Cirksena, K , Cheng, Y -T , Martin, G (1991a) Activity
patterns of California residents Sacramento, CA California Air Resources Board, contract
no A6-177-33
Wiley, J A , Robinson, J P , Piazza, T , Garrett, T , Cirksena, K , Cheng, Y -T , Martin, G (1991b) Study of
children's activity patterns final report Sacramento, CA California Air Resources Board, contract
no A-733-149
Willett, W. (1989) An overview of issues related to the correction of non-differential exposure measurement error
in epidenuologic studies Stat Med 8 1031-1040,1071-1073
Wilson, A L , Colome, S D , Baker, P E , Becker, E W (1986) Residential indoor air quality
characterization study of nitrogen dioxide Phase I Volumes 1, 2 and 3 Southern California Gas
Company, October
Yanagisawa, Y , Nishimura, H (1982) A badge-type personal sampler for measurement of personal exposure to
NO2 and NO in ambient air Environ Iht 8 235-242
Yanagisawa, Y , Matsuki, H , Osaka, F , Kasuga, H , Nishimura, H (1984) Annual variation of personal
exposure to nitrogen dioxide In Berglund, B , Lindvall, T , Sundell, J , eds Indoor air proceedings of
the 3rd international conference on indoor air quality and climate, v 4, chemical characterization and
personal exposure, August, Stockholm, Sweden Stockholm, Sweden Swedish Council for Building
Research, pp 33-36 Available from NITS, Springfield, VA, PB85-104214
Yanagisawa, Y , Nishimura, H , Matsuki, H , Osaka, F , Kasuga, H (1986) Personal exposure and health effect
relationship for NO2 with urinary hydroxyproline to creatinine ratio as indicator Arch Environ Health
41- 41-48
Yanagisawa, Y , Nishimura, H , Matsuki, H , Osaka, F , Kasuga, H (1988) Urinary hydroxyprohne to
creatinine ratio as a biological effect marker for exposure to NO2 and tobacco smoke Atmos Environ
22 2195-2203
«U S GOVERNMENT PRINTING OFFICE 1993-550-001/80323
8-29
-------
-------
-------
rn
•o
o
o
00
cb
CD
0]
°§ S
7;
O
o£
P
< CO
S* m
CD
c
CO
(D
oomc
O
3
» s
CD
Q.
'CD
o^s
~T" ^ JD
^1^
*» ^ 2.
g3 CD
CD O
3 Ef
CD
DO >
CD
-------
|