EVALUATION OF THE RADM GAS-PHASE CHEMICAL MECHANISM
by
William P. L. Carter
Statewide Air Pollution Research Center
University of California
Riverside, California 92521
and
Fredrick W. Lurmann
Lurmann Associates
1819 State St., Suite D
Santa Barbara, California 93103
Cooperative Agreement No. CR-81U558-01-0
Project Officer
Marcia C. Dodge
Chemical Processes and Characterization Division
Atmospheric Research and Exposure Assessment Laboratory
ARCHIVE Research Triangle Park, North Carolina 27711
EPA
600-
3-
ATMOSPHERIC RESEARCH AND EXPOSURE ASSESSMENT LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U. S. ENVIRONMENTAL PROTECTION AGENCY
RESEARCH TRIANGLE PARK, NC 27711
iPA-89:radmreport
-------
EVALUATION OF THE RADM GAS-PHASE CHEMICAL MECHANISM
by
William P. L. Carter
Statewide Air Pollution Research Center
University of California
Riverside, California 92521
and
Fredrick W. Lurmann
Lurmann Associates
1819 State St., Suite D
Santa Barbara, California 93103
IP
L, Cooperative Agreement No. CR-814558-01-0
Project Officer
Marcia C. Dodge
Chemical Processes and Characterization Divi-sion
Atmospheric Research and Exposure Assessment Laboratory
Research Triangle Park, North Carolina 27711
ATMOSPHERIC RESEARCH AND EXPOSURE ASSESSMENT LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U. S. ENVIRONMENTAL PROTECTION AGENCY
RESEARCH TRIANGLE PARK, NC 27711
CPA-89:radmreport
-------
NOTICE
The information in this document has been funded by the United States
Environmental Protection Agency under Cooperative Agreement No. CR-81M558-
01-0 to the University of California at Riverside. It has been subject to
the Agency's peer and administrative review, and it has been approved for
publication as an EPA document. Mention of trade names or commercial
products does not constitute endorsement or recommendation for use.
ii
-------
PREFACE
This final report describes an independent evaluation of the gas-
phase mechanism developed by Dr. William R. Stockwell of the State Univer-
sity of New York at Albany for use in the Regional Acid Deposition Model
(RADM). Although the authors are solely responsible for the preparation
of this report and the development of its conclusions, the work described
herein includes significant contributions by Dr. Stockwell and several
other scientists. Dr. Stockwell participated in this effort by providing
data files necessary to implement the several versions of the mechanism
that were evaluated. He also evaluated the reviews and recommendations
made by the authors throughout the program, incorporating many of these
recommendations into the mechanism and generally keeping the authors
informed on the continuing evolution of his mechanism and its implementa-
tion into RADM. In addition, Stockwell assisted our evaluation of the
effects of peroxy radical approximations by providing us with a detailed
mechanism to use as the standard. Without his cooperation and participa-
tion, the utility of this program to the overall assessment of the RADM
model would have been highly limited.
In addition to Dr. Stockwell, several other scientists provided
valuable assistance in this program. Dr. Paulette Middleton of the
National Center of Atmospheric Research solicited our advice and partici-
pation in the development of the RADM emissions processing system and is
responsible for many of the ideas discussed in Section 3.2 of this
report. Dr. Harvey E. Jeffries and co-workers at the University of North
Carolina, as subcontractors to this program, prepared the data files used
to implement their new light characterization model for the UNC chamber,
reviewed the input files we used when modeling the UNC chamber runs, and
provided us with data for UNC runs we had not modeled previously. Dr.
Roger Atkinson of the Statewide Air Pollution Research Center (SAPRC) at
U. C. Riverside provided valuable assistance in the initial review of the
RADM gas-phase mechanism for consistency with current chemical data and
the results of the latest data evaluations. Finally, the authors wish to
acknowledge the role played by Dr. Marcia C. Dodge of EPA/AREAL in
conceiving of the need for an independent evaluation of the RADM gas-phase
mechanism, and providing leadership in its overall direction, and her
active interest in its progress.
The authors also wish to acknowledge the SAPRC staff members who
provided major assistance in the preparation of this report. Ms. Minn P.
Poe is responsible for the preparation of many of the figures, and also
provided computer programming assistance throughout this program. Ms.
Christy J. LaClaire is responsible for most of the word processing for
this report.
iii
-------
ABSTRACT
Results are presented of a program to independently evaluate the gas
phase mechanism developed for use in the second version of the Regional
Acid Deposition Model (RADM-II). This work consisted of the following:
(1) An initial review of the 1987 version of the RADM mechanism was
carried out and several modifications were made to the mechanism prior to
the major evaluation effort described in this report. (2) Assistance was
provided to the RADM team in developing methods for processing emissions
input for the RADM model and for representing aggregated VOCs in the
model. (3) The RADM mechanism was tested by comparing predictions against
results of over 550 environmental chamber experiments carried out at the
University of California at Riverside and at the University of North
Carolina (UNC). The testing using the UNC experiments incorporated an
improved chamber light characterization model developed at UNC. Statis-
tical summaries of the overall results and results for individual experi-
ments are presented and discussed. (4) A series of 90 test problems for
use in sensitivity calculations to assess effects of alternative assump-
tions for gas-phase mechanisms used in regional models were developed, and
these were used to test several condensation approaches for the RADM
mechanism. (5) As a result of this evaluation against the chamber data
and the sensitivity test calculations, recommendations were made for
modifications to the RADM mechanism concerning the representation of the
gas-phase chemistry of the aromatics and the condensation of the represen-
tation of the higher alkanes. Two modified RADM mechanisms, one involving
primarily changes in parameter values, and one where more extensive
changes and use of fewer model species are involved, were developed.
These modified mechanisms and their performance in simulating the chamber
data and the sensitivity test problems are discussed. The version of the
mechanism involving primarily parameter changes has been adopted by the
RADM team for implementation into the model.
It is concluded that the RADM mechanism, as modified as a result of
this program, represents the state-of-the-art in gas-phase atmospheric
reaction mechanisms and is suitable for use in the regional acid deposi-
tion model. However, there are still major uncertainties in our under-
standing of atmospheric chemistry. Also, available chamber data are not
sufficient for evaluating the ability of mechanisms to represent the low
NOX conditions characteristic of much of the regional modeling domain, nor
for evaluating their prediction of peroxide species or organic acids. The
major areas of uncertainty and needs for future research are discussed.
iv
-------
CONTENTS
Page
Tables vii
Figures xi
1. Introduction 1
2. The March 1988 RADM and Recommended Modified RADM Gas-Phase
Chemical Mechanisms 5
2.1 The March 1988 RADM Mechanism 6
2.2 Recommended Modifications to the RADM Mechanism 29
2.2.1 Recommended Modifications to the Aromatic Mechanism 29
2.2.2 Condensed Higher Alkane Representation 37
2.2.3 Modifications to the XN02 Reactions 38
2.2.4 Omitted OLN Reactions 40
2.2.5 Parameter Modifications Resulting from Updated
Emissions Assignments - 40
2.3 The Recommended Modified Mechanism (RADM-M) 41
2.4 The Modified Parameter Mechanism (RADM-P) 42
3. Aggregation and Representation of Organic Emissions in RADM 57
3.1 Introduction 57
3.2 Aggregation of VOC Emissions Input for RADM 58
3.2.1 Development of a Mechanism-Independent VOC
Emissions Classification System 59
3.2.2 Recommended Extensions to the VOC Emissions
Classification System 73
3.2.3 Aggregation of Emissions into the RADM Mechanism 76
3.3 Derivation of Kinetic and Mechanistic Parameters for
Lumped Model Species in RADM 87
3.3.1 Mechanistic Parameters for Lumped Alkanes 93
3-3-2 Mechanistic Parameters for Lumped Aromatics 97
3.3.3 Mechanistic Parameters for Lumped Alkenes 99
4. Evaluation Against Environmental Chamber Data:
Characterization and Lumping Procedures 100
4.1 Introduction 100
4.2 Characterization of Chamber Conditions 102
4.2.1 Light Characterization 103
4.2.2 Representation of Other Chamber-Dependent
Effects 112
4.3 Representation of Organics in Chamber Experiments 138
-------
CONTENTS
(continued)
5. Evaluation Against Environmental Chamber Data: Results 142
5.1 Organization of the Evaluation 143
5.2 Measures of Model Performance 146
5.3 Performance on Characterization Runs 149
5.4 Performance of the RADM Mechanism on Single Organic Runs 155
5.4.1 Carbonyl-NO Runs 155
5.4.2 Alkene-N0x Runs 162
5.4.3 Alkane-NO Runs 169
5.4.4 Aromatic-N0x Runs 175
5.4.5 Natural Hydrocarbon Runs 184
5.4.6 Performance in Simulations of Formaldehyde Yields 190
5.4.7 Performance in Simulations of PAN Yields 192
5.5 Performance of the RADM Mechanism on Runs with
Organic Mixtures 195
5.6 Tests of the RADM VOC Aggregation Approach 218
5.7 Performance of the Recommended Modified Mechanisms 219
6. Evaluation of Alternative Chemical Assumptions 242
6.1 Test Problems for Sensitivity Analysis 242
6.2 Presentation of Sensitivity Test Results 256
6.3 Treatment of Peroxy Radical Reactions 258
6.4 Lumping of Higher Alkanes 290
6.5 The Recommended vs. the March 1988 RADM Mechanisms 292
7. Summary and Conclusions 310
7.1 Evaluation of Kinetic and Mechanistic Parameters 312
7.2 Development of the Linkage Between the VOC Emissions
Inventory and the RADM Mechanism 314
7.3 Mechanism Evaluation Against Chamber Data 315
7.4 Sensitivity Testing of Alternate Mechanistic Assumption 320
7.5 Recommendations 322
7.5.1 Longer Term Research Needs 323
7.5.2 Shorter Term Needs 325
References 328
APPENDICES
A Selected Results of Simulations of All Experiments
Modeled Using the RADM Mechanism 333
B Selected Results of Simulations of All Organic-N0x
Experiments Modeled Using the RADM-M Mechanism 356
C Selected Results of Simulations of All Organic-N0x
Experiments Modeled Using the RADM-P Mechanism 373
vi
-------
TABLES
Table
Number
1. Model Species Employed in the March 1988 Version of the RADM
Gas-Phase Chemical Mechanism 8
2. Listing of Reactions in the March 1988 Version of the RADM
Gas-Phase Chemical Mechanism 11
3. Absorption Coefficients and Quantum Yields Used in Versions
of the RADM Gas-Phase Chemical Mechanism Evaluated in this
Study 19
4. Organic Model Species Employed in the Recommended Modified
Version of the RADM Gas-Phase Chemical Mechanism (RADM-M) 43
5. Listing of the Organic Reactions in the Recommended Modified
Version of the RADM Gas-Phase Chemical Mechanism (RADM-M) 45
6. Listing of the Organic Reactions in the Modified Parameter
Version of the RADM Gas-Phase Chemical Mechanism (RADM-P) 51
7. Summary of the Emissions Groupings Using the 32-Class Scheme,
Amounts of Each Emitted in the NAPAP Total U.S. Inventory, and
Estimated Amounts Reacted in Conditions Representative of RADM
Appl ications 71
8. Assignments of Emissions Groups to RADM Model Species 77
9. Averages of Parameters for Emissions Groups Lumped as C^
Alkanes and for RADM Lumped Alkane Model Species, Derived Using
the NAPAP Total U.S. Emissions Inventory 89
10. Averages of Parameters for Emissions Groups Lumped as Aromatics
and for RADM Lumped Aromatic Model Species, Derived Using the
NAPAP Total U.S. Emissions Inventory 90
11. Averages of Parameters for Emissions Groups Lumped as Non-Ethene
Alkenes and for RADM Lumped Non-Ethene Alkene Model Species,
Derived Using the NAPAP Total U.S. Emissions Inventory 91
12. Averages of Parameters for Emissions Represented by the Combined
Higher Alkane (HC5 + HC8) Model Species Recommended for the
RADM-M Mechanism 98
13. ^ummary of Chamber-Dependent Parameters Used in the Model
Simulat ions of the SAPRC EC Chamber Runs 117
14. Summary of Chamber-Dependent Parameters Used in the Model
Simulations of the SAPRC ITC Chamber Runs 119
vii
-------
TABLES
(continued)
Table
Number Page
15. Summary of Chamber-Dependent Parameters Used in the Model
Simulations of the SAPRC OTC Chamber Runs .......................... 121
16. Values of the Continuous Chamber Radical Input Parameter, kRg,
used in Modeling SAPRC ITC and OTC Chamber Runs, and Averages
of Measured Values Used to Derive Them ............................. 124
17. Summary of Chamber-Dependent Parameters Used in the Model
Simulations of the UNC Chamber Runs ................................ 125
18. Values of Adjustable Chamber-Dependent Parameters Employed
in the Simulations of UNC Chamber Runs ............................. 127
19. Selected Performance Statistics in the RADM Simulations of
the N0x-Air, C0-N0x-Air, Pure Air, and Carbonyl-Air Runs ........... 150
20. Performance Statistics on Maximum Ozone and on NO Oxidation
and Ozone Formation Rates in the Simulations of the
Organic-N0x-Air Runs Using the RADM Mechanism ...................... 156
21. Performance Statistics on Maximum Formaldehyde in the
Simulations of the Organic-N0x-Air Runs Using the RADM
Mechanism [[[ 191
22. Performance Statistics on Maximum PAN in the Simulations of
the Organic-N0x-Air Runs Using the RADM Mechanism .................. 193
23. Comparison of Average Normalized Biases and Errors in
Simulations of Chamber Experiments using the RADM Mechanism
with Run-to-Run Adjustment of Rate Constants with Simulations
using the Standard RADM VOC Aggregation Procedure .................. 220
21. Performance Statistics on Maximum Ozone and NO Oxidation
and Ozone Formation Rates in the Simulations of
Organic-N0x-Air Runs Using the RADM-M Mechanism
25. Performance Statistics on Maximum Ozone and NO Oxidation
and Ozone Formation Rates in the Simulations of
Organic-N0x-Air Runs Using the RADM-P Mechanism .................... 226
26. Performance Statistics on Maximum Formaldehyde in the
Simulations of Organic-N0x-Air Runs Using the RADM-M
Mechanism [[[ 228
27. Performance Statistics on Maximum Formaldehyde in the
Simulations of Organic-N0x-Air Runs Using the RADM-P
-------
TABLES
(continued)
Table
Number Page
28. Performance Statistics on Maximum PAN in the Simulations of
Organic-N0x-Air Runs Using the RADM-M Mechanism 230
29. Performance Statistics on Maximum PAN in the Simulations of
Organic-N0x-Air Runs Using the RADM-P Mechanism 231
30. Comparisons of Average Normalized Biases and Errors in Maximum
Ozone and NOX Oxidation Rates in the Simulations of Organic-
NO -Air Runs Using the RADM, RADM-M, and RADM-P Mechanisms 232
X
31. Environmental Conditions Used in Test Problems for Sensitivity
Calculations 244
32. Compositions of ROG Surrogate Mixtures Used in Test
Calculations ' 249
33. Compositions of ROG Surrogates Used in the Test Calculations
for the RADM Mechanism 252
34. Summary of Conditions of Test Calculations Used for
Sensitivity Studies 253
35. Organic Peroxy + Organic Peroxy Reactions Used in Stockwell's
"Detailed Radical" Version of the RADM Mechanism 262
36. Listing of the Organic Reactions in the Modified RADM
Mechanism Incorporating the 1986 SAPRC Approach for
Representing Reactions of Organic Peroxy Radicals 270
37. Results of Sensitivity Test Calculations Using the Evaluated
RADM Mechanism as the Test Mechanism, Relative to the
"Detailed Radical" Mechanism as the Standard 277
38. Results of Sensitivity Test Calculations for the Version of
the RADM Mechanism Using the 1986 SAPRC Peroxy Radical
Representation as the Test Mechanism, Relative to the
"Detailed Radical" Mechanism as the Standard 282
39. Results of Sensitivity Test Calculations Using the Version of
the RADM Mechanism with the Higher Alkane Species HC5 and HC8
Lumped Together, Relative to the Standard RADM Mechanism 293
40. Summary of Results of Sensitivity Test Calculations Using the
Recommended Modified (RADM-M) and the Recommended Modified
Parameter (RADM-P) Mechanisms, Relative to the March 1988
RADM Mechanism as the Standard 298
IX
-------
TABLES
(continued)
Table
Number
41. Results of Sensitivity Test Calculations of Urban Ozone
Formation Using the Recommended Modified (RADM-M) and the
Recommended Modified Parameter (RADM-P) Mechanisms, Relative
to the March 1988 RADM Mechanism as the Standard 301
42. Results of Sensitivity Test Calculations of Urban H202
Formation Using the Recommended Modified {RADM-M) and the
Recommended Modified Parameter (RADM-P) Mechanisms, Relative
to the March 1988 RADM Mechanism as the Standard 304
-------
FIGURES
Figure
Number Page
1. Comparison of Fits of RADM and Modified RADM Simulations
to Results of Toluene and Xylene-N0x-Air Chamber Experiments 36
2. Experimental and Calculated Concentration-Time Plots for
Selected Species in the UNC Formaldehyde-N0v-Air Run AU0279B 113
A
3. Experimental and Calculated Concentration-Time Plots for
Selected Species in the UNC Complex Mixture-NO -Air Run
ST1682B 114
4. Experimental and Calculated Concentration-Time Plots for
Selected Species in the UNC Complex Mixture-NO -Air Run
ST2981R 115
5. Comparison of Results of UNC Characterization Simulations
Using Various Mechanisms and Chamber and Light-Character-
ization Models 129
6. Hierarchy of Species for Mechanism Testing 144
7. Calculated vs Experimental Changes in [NO] in NO -Air and
C0-N0v-Air Runs 151
A
8. Calculated vs Experimental Changes in [N02] in NO -Air and
C0-N0v-Air Runs 151
A
9. Calculated vs Experimental Maximum Ozone in the Pure Air and
Carbonyl-Air Runs 153
10. Calculated vs Experimental Rate of Ozone Formation in the
Pure Air and Carbonyl-Air Runs 153
11. Histogram of Normalized Biases in Maximum Ozone in Pure Air and
Carbonyl-Air Runs .- 154
12. Normalized Biases in Maximum Ozone vs HC/NOX Ratios in
Carbonyl-Air Runs 154
13. Calculated vs Experimental Maximum Ozone in Carbonyl-N0x
Runs 159
1U. Calculated vs Experimental Rates of NO Oxidation and Ozone
Formation in the Carbonyl-N0v Runs 159
X
15. Histogram of Normalized Biases in Maximum Ozone in
Carbonyl-N0x Runs 160
16. Normalized Bias in Maximum Ozone vs HC/NOX Ratio in
Carbonyl-N0x-Runs 160
xi
-------
FIGURES
(continued)
Figure
Number Page
17. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of UNC Acetaldehyde-N0x Run
AU2482B 161
18. Calculated vs Experimental Maximum Ozone in the Alkene-N0x
Runs for the RADM Mechanism 163
19. Calculated vs Experimental Rates of NO Oxidation and Ozone
Formation in the Alkene-N0v Runs for the RADM Mechanism 163
A
20. Histogram of Normalized Biases in Maximum Ozone in Alkene
Runs for the RADM Mechanism 164
21. Normalized Bias in Maximum Ozone vs HC/NOX in Alkene Runs
for the RADM Mechanism 164
22. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of UNC Ethene-N0x Run
OC058MB 165
23. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of the Propene-N0x Run
EC-216 166
2M. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of UNC Propene-N0x Run
JL2983B 167
25. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of UNC 1-Butene-NOx Run
ST2583B 168
26. Calculated vs Experimental Maximum Ozone in the Alkane-N0x
Runs for the RADM Mechanism 170
27. Calculated vs Experimental Rates of NO Oxidation and Ozone
Formation in the Alkane-N0x Runs for the RADM Mechanism 170
28. Histogram of Normalized Biases in Maximum Ozone in Alkane
Runs for the RADM Mechanism 171
29. Normalized Bias in Maximum Ozone vs HC/NOX in Alkane Runs
for the RADM Mechanism 171
30. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of n-Butane-NOx Run EC-130 172
xii
-------
FIGURES
(continued)
Figure
Number Page
31. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of n-Octane-NOx Run ITC-552 173
32. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of UNC 2,3-Dimethylbutane
-NOX Run OC1789R 174
33. Calculated vs Experimental Maximum Ozone in the Aromatic-
NOX Runs for the RADM Mechanism 176
3^. Calculated ys Experimental Rates of NO Oxidation and Ozone
Formation in the Aromatic-N0x Runs for the RADM Mechanism 176
35. Histogram of Normalized Biases in Maximum Ozone in Aromatic
Runs for the RADM Mechanism 177
36. Normalized Bias in Maximum Ozone ys HC/NO in Aromatic Runs
for the RADM Mechanism 177
37. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of the Benzene-NO Run
ITC-562 178
38. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of the Toluene-NO Run
EC-327 179
39. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of the UNC Toluene-NO Run
JL3080R 180
40. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of the m-Xylene-NO Run
ITC-702 181
41. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of the UNC o-Xylene-NO Run
JL3080B 182
42. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of the Mesitylene-NO Run
EC-901 V 183
43. Calculated ys Experimental Maximum Ozone in the Natural
Hydrocarbon Runs for the RADM Mechanism 185
xiii
-------
FIGURES
(continued)
Figure
Number Page
44. Calculated vs Experimental Rates of NO Oxidation and Ozone
Formation in the Natural Hydrocarbon Runs for the RADM
Mechanism 185
45. Histogram of Normalized Bias in Maximum Ozone in the Natural
Hydrocarbon Runs for the RADM Mechanism 186
46. Normalized Bias in Maximum Ozone vs HC/NOX in the Natural
Hydrocarbon for the RADM Mechanism 186
47. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of the Isoprene-N0x Run
ITC-811 187
48. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of the UNC Isoprene-NO Run
JL1680B 188
49. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of the UNC a-Pinene-NO Run
JL1580B 189
50. Calculated vs Experimental Maximum Formaldehyde in the
Single Organic-NOK Runs for the RADM Mechanism 192
51. Calculated vs Experimental Maximum PAN in the Single
Organic-N0x Runs for the RADM Mechanism 194
52. Calculated vs Experimental Maximum Ozone in the Simple
Mixture Runs for the RADM Mechanism 196
53. Calculated vs Experimental Rates of NO Oxidation and Ozone
Formation in the Simple Mixture Runs for the RADM Mechanism 196
54. Histogram of Normalized Biases in Maximum Ozone in the Simple
Mixture Runs for the RADM Mechanism 197
55. Normalized Bias in Maximum Ozone vs HC/NOX in the Simple
Mixture Runs for the RADM Mechanism 197
56. Calculated vs Experimental Maximum Ozone in the Surrogate
Mixture Runs for the RADM Mechanism 198
57. Calculated vs Experimental Rates of NO Oxidation and Ozone
Formation in the Surrogate Mixture Runs for the RADM
Mechanism 198
xiv
-------
FIGURES
(continued)
Figure
Number
58. Histogram of Normalized Biases in Maximum Ozone in the
Surrogate Mixture Runs for the RADM Mechanism 199
59. Normalized Bias in Maximum Ozone vs HC/NOX in the Surrogate
Mixture Runs for the RADM Mechanism 199
60. Normalized Bias in Maximum Ozone vs Initial NOX in the
Surrogate Mixture Runs for the RADM Mechanism 200
61. Normalized Bias in Maximum Ozone vs Initial HC in the
Surrogate Mixture Runs for the RADM Mechanism 200
62. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of the UNC Mixed Alkene Run
OC1278R , 201
63. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of the 7 Hydrocarbon
Surrogate Run EC-237 202
64. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of the 8 Hydrocarbon
Surrogate, Multi-Day Run ITC-630 203
65. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of the 8 Hydrocarbon
Surrogate, Multi-Day Run ITC-633 204
66. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of the 8 Hydrocarbon
Surrogate, Multi-Day Run ITC-637 205
67. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of the 8 Hydrocarbon
Surrogate, Multi-Day Run OTC-192B 206
68. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of the 8 Hydrocarbon
Surrogate, Multi-Day Run OTC-195A 207
69. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of the 8 Hydrocarbon
Surrogate, Multi-Day Run OTC-202B 208
70. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of the 8 Hydrocarbon
Surrogate, Multi-Day Run OTC-224B 209
xv
-------
FIGURES
(continued)
Figure
Number Page
71. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of the UNC 3 Hydrocarbon
Surrogate Run JN1483R 210
72. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of the UNC Miscellaneous
Surrogate Run AU2681B 211
73. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of the UNC Synurban
Surrogate Run JN2685R 212
7M. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of the UNC Synauto
Surrogate Run AU0684R 213
75. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of the UNC Auto Exhaust
Run AU1183R 214
76. Calculated vs Experimental Maximum Formaldehyde in the
Simple Mixture Runs for the RADM Mechanism 216
77. Calculated vs Experimental Maximum Formaldehyde in the
Surrogate Mixture Runs for the RADM Mechanism 216
78. Calculated vs Experimental Maximum PAN in the Simple
Mixture Runs for the RADM Mechanism 217
79. Calculated vs Experimental Maximum PAN in the Surrogate
Mixture Runs for the RADM Mechanism 217
80. Plots of Maximum Ozone Yields in Simulations of Organic
Mixture Experiments using the RADM Mechanism with Run-to-Run
Adjustment of Rate Constants Against Yields Calculated Using
the Standard RADM VOC Aggregation Procedure 222
81. Calculated vs Experimental Maximum Ozone in the Surrogate
Mixture Runs for the RADM-M Mechanism 235
82. Calculated vs Experimental Maximum Ozone in the Surrogate
Mixture Runs for the RADM-P Mechanism 235
83. Calculated vs Experimental Maximum Ozone in the Alkane-NO
Runs for the RADM-M Mechanism 236
8M. Calculated vs Experimental Maximum Ozone in the Alkane-NO
Runs for the RADM-P Mechanism 236
xv i
-------
FIGURES
(continued)
Figure
Number Page
85. Calculated vs Experimental Maximum Ozone in the Aromatic-
NOY Runs for the RADM-M Mechanism .................................. 237
A
86. Calculated vs Experimental Maximum Ozone in the Aromatic-
NOX Runs for the RADM-P Mechanism .................................. 237
87. Comparisons of Maximum Ozone Calculated using the RADM-M
and the Unmodified RADM Mechanisms for Simple and Surrogate
Mixture Runs [[[ 240
88. Comparisons of Maximum Ozone Calculated using the RADM-P
and the Unmodified RADM Mechanisms for Simple and Surrogate
Mixture Runs [[[ 240
89. Concentration-Time Plots for Selected Species Calculated
Using the RADM, RADM-M, and RADM-P Mechanisms for the Summer
Surface Static Urban Test Problem with Initial NOX of 67 ppb
and a ROG/NOV Ratio of 10 .......................................... 307
A
90. Concentration-Time Plots for Selected Species Calculated
Using the RADM, RADM-M, and RADM-P Mechanisms for the Winter
-------
1. INTRODUCTION
The Regional Acid Deposition Model (RADM) plays a central role in the
National Acid Precipitation Assessment Program's (NAPAP) plan to develop
an understanding of source-receptor relationships for acidifying pollut-
ants in the United States. This model is needed to formulate appropriate
and cost-effective emission control policies to mitigate the potentially
damaging effects of acid deposition. While transport, deposition, and
chemistry all play important roles in the acid deposition phenomenon, the
chemistry plays a particularly important role in determining the response
of the acid deposition system to major changes in NO and SOV emissions.
A A
The chemistry submodel is the only component of the overall modeling
system that is capable of predicting nonlinearly in the relationship
between source emissions and acid deposition. Thus, it is critically
important that any chemical mechanism incorporated into the RADM model be
fully evaluated using the best available data and that it be recognized by
the scientific community as representing the current state-of-the-art in
our understanding of atmospheric chemistry.
Although the full RADM model must incorporate both gas phase and
aqueous phase chemistry, the study described in this report addresses only
the gas phase portion of the chemistry module. As presently implemented,
only the gas phase species are transported within the model, and their
concentrations at each time step are used to drive the aqueous phase
chemistry, based on stationary state approximations. Therefore, the gas
phase chemistry module contributes substantially to any non-linearities
predicted by the model. The gas phase chemistry module is also important
because, as implemented in the first version of the RADM model (RADM-I),
approximately 75J of the computer time is consumed in calculating the gas
phase transformations. This is with a smaller version of the gas phase
chemical mechanism (Stockwell, 1986) than the one proposed for use in the
second version of this model. Thus, clearly the efficiency and appro-
priateness of the level of detail in the gas phase transformation mechan-
ism, as well as its accuracy and scientific validity, is an important
issue which needs to be addressed.
Because of computer constraints, the gas-phase chemistry module in
the first version of the RADM model (Stockwell, 1986) was highly condensed
-------
and contained a number of approximations that were chemically unreal-
istic. With the expanded memory capacities of the newer computers, a new,
more detailed gas-phase chemical mechanism was proposed for use in the
second version of the RADM model (designated RADM-II) which is now under
development (NCAR, 1987). Although in terms of chemical detail and
consistency with current knowledge, this new RADM gas-phase mechanism is
an improvement over the previous version, it had not undergone the
external review and comprehensive evaluation that is necessary for such an
important component of a model which will be used in making policy
decisions. For example, this new mechanism had undergone only limited
testing against the results of a small number of environmental chamber
experiments. In contrast, the mechanisms most recently developed for use
in oxidant and EKMA models have been tested against results of hundreds of
experiments carried out in several different chambers (Carter et al.,
1986; Gery et al., 1988). The December 1986 EPA workshop on evaluation
and documentation of chemical mechanisms for use in air quality simulation
models concluded that an extensive testing program against chamber data is
an important part of the overall evaluation process for such mechanisms
(EPA, 1987). In addition, the new RADM mechanism had not undergone
external review for consistency with available laboratory data and current
chemical knowledge. Such a review and evaluation would be particularly
useful prior to the implementation of the mechanism in RADM, when any
problems can be more readily corrected. To address this need, the EPA
contracted the Statewide Air Pollution Center (SAPRC) at the University of
California at Riverside to carry out an independent review and evaluation
of this new proposed gas-phase chemical mechanism for RADM, and to make
suggestions and recommendations concerning areas where the mechanism might
be improved.
This evaluation program was a cooperative effort involving the
authors of this report at SAPRC and Lurmann and Associates, Dr. Harvey
Jeffries and his group at the University of North Carolina (UNC), Dr.
William Stockwell at the State University of New York (SUNY), and Dr.
Paulette Middleton at the National Center for Atmospheric Research
(NCAR). The evaluation itself was carried out at SAPRC with the assist-
ance of Lurmann and Associates. Dr Jeffries and his group assisted in
this effort by reviewing the input data used when testing the mechanism
-------
against results of UNC chamber experiments and preparing input data to
implement a new model for light characterization in that chamber (Jeffries
et al.f 1989a). Dr. Stockwell, the principal developer of the RADM-II
mechanism and the member of the RADM team ultimately responsible for the
gas-phase mechanism which will be used in RADM-II, provided the initial
and modified versions of the mechanism which were evaluated in this
program, and served as an overall consultant on this project. Dr.
Middleton is responsible for preparing the emissions input for the RADM
model and as part of this program, SAPRC worked with her in developing
appropriate procedures for processing organic emissions data for use with
this mechanism. This report describes primarily the work on this
evaluation effort carried out by SAPRC and Lurmann and Associates.
This evaluation effort involved several separate, though related,
tasks: (1) SAPRC carried out an initial review of a preliminary version
of the RADM-II mechanism (NCAR, 1987) to determine its consistency with
current data and data evaluations and made recommendations for modifica-
tions. As a result of these recommendations, Stockwell made some modifi-
cations in the RADM-II mechanism prior to its implementation on SAPRC
computers for further evaluation. (2) SAPRC also made recommendations
regarding organic emissions processing procedures and methods used in the
mechanism to represent organic emissions. Most of these recommendations
were subsequently implemented in the RADM-II mechanism and in emissions
processing procedures developed for it. (3) With the assistance of UNC,
modifications were made in the light characterization and chamber effects
models used when testing mechanisms against UNC chamber data. (4) The
predictions of the RADM-II mechanism, as modified by Stockwell as a result
of the initial review, was tested against the results of approximately 550
UNC and SAPRC chamber runs, and a statistical analysis of the results was
carried out. Other modified versions of the RADM mechanism were tested
against these chamber data as well. (5) A number of alternative mechanis-
tic assumptions and condensation approaches for the RADM mechanism were
tested using a series of 90 test problems representing a range of condi-
tions expected in regional model applications. (6) Additional recommenda-
tions for modifications of the RADM mechanism were made as a result of the
simulations of the chamber experiments and the test calculations.
Stockwell incorporated some of these recommendations in a "final" version
-------
of the RADM mechanism, which, as of this writing, is expected to be the
version which will be implemented into the RADM-II model. This mechanism
is included among those which were simulated in the chamber experiments
and test calculations.
This report is organized into seven sections. Following this intro-
ductory section, Section 2 describes and lists the version of the RADM-II
mechanism which was evaluated in the model simulations carried out in this
study, and two modified versions which were recommended as a result of
this program. Section 3 describes the work carried out in this program
concerning aggregation and representation of organic emissions data for
use with the RADM-II mechanism. Section ^ describes the procedures used
when evaluating the mechanisms against the chamber data, with emphasis
particularly on the changes in the procedures and input data used,
compared with previous studies. Section 5 presents the results of the
simulations of the evaluated and recommended modified RADM mechanisms
against the chamber data, including statistical summaries of the perform-
ance of the mechanisms and representative concentration-time plots.
Section 6 describes the test calculations carried out to test alternative
mechanistic assumptions and degrees of condensation, including a
discussion of the 90 simulations employed, the mechanistic alternatives
tested, and the results obtained. Finally, Section 7 gives a summary of
the results from this program and the conclusions.
-------
2. THE MARCH 1988 RADM AND RECOMMENDED MODIFIED RADM
GAS-PHASE CHEMICAL MECHANISMS
In this section, we summarize and list the versions of the RADM-II
chemical mechanisms which were evaluated during this study, and describe
the modifications to the mechanism which are recommended as a result of
this program. At the start of this program, the current RADM-II mechanism
was the version documented in the NCAR (1987) report. However, as
indicated earlier, Stockwell modified this mechanism shortly after this
program began as a result of an initial review we carried out on this
mechanism. This modified mechanism, which was dated March 30, 1988, is
the version that was implemented at SAPRC for evaluation purposes. There-
fore, the March 1988 version, and not the 1987 mechanism, is the version
of the mechanism whose evaluation is discussed in this report. A listing
of this mechanism is presented in Section 2.2. In the subsequent
discussion, this version of the mechanism will be designated simply as the
"RADM" or the "March 1988 RADM" mechanism.
The results of our evaluation efforts, described in more detail in
Sections 3, 5 and 6 of this report, indicated several areas where improve-
ments to the RADM mechanism were warranted. These included changes in the
representation of the unknown aspects of the aromatic photooxidation
mechanisms to improve its performance in simulations of the chamber data
while reducing the number of species involved, a condensation in the
representation of the alkanes, some changes in rate constant and product
yield parameters for lumped species to better correspond to our most
recent analysis of emissions data, correction of an error in the methyl-
glyoxal quantum yields, addition of a few reactions omitted from the NO^ *
alkene mechanism, and minor improvements in the chemistry of the condensed
representations of the reactions of nitrophenols. The modifications we
recommended to the RADM mechanism are described in Section 2.2, and the
resulting modified mechanism is listed in Section 2.3- In the subsequent
discussion, this modified version of the RADM-II mechanism is designated
as the "recommended modified," or simply as the "RADM-M," mechanism.
By the time we completed the evaluation effort and formulated our
recommendations for modifying the mechanism, much of the work in imple-
menting the RADM mechanism in the overall RADM-II model software had
-------
already been carried out. Although Stockwell and the RADM team were
willing to implement our recommendations, we were informed that
modifications involving more than Just changes in kinetic or mechanistic
parameters (i.e., rate constant values, product yield coefficients, or
absorption coefficient and quantum yield data) would be difficult to
implement without adversely impacting the RADM-II development schedule.
Therefore, we developed a set of recommendations for modifications of the
RADM mechanism that involved only changes in parameter values. These
changes included revising the aroraatics mechanism, updating the rate
constants and product yield parameters for lumped species based on the
most recent analysis of emissions data, and correcting the methylglyoxal
quantum yields. In addition, the RADM team was willing to add the few
reactions which were omitted from the NO? + alkene mechanism, so this
change was also incorporated into this version of the mechanism. This
second modified version of the RADM mechanism is designated as the
"modified parameter," or simply the "RADM-P," mechanism. It is described
and listed in Section 2.4.
The RADM team has accepted the recommendations for modifications
which were incorporated in the RADM-P mechanism, and this version of the
mechanism is now being implemented into the RADM-II model (Stockwell,
private communication, 1989).
2.1 The March 1988 RADM Mechanism
The RADM gas-phase chemical mechanism that was used as the starting
point in this evaluation effort is the March 1988 update of the RADM-II
mechanism described in the NCAR (1987) report. This is a condensed mech-
anism which updated and extends the earlier mechanism of Stockwell (1986),
with the representation of the reactions of the aroraatics based on that of
Lurmann et al. (1986). However, this mechanism has a more detailed
representation of the reactions of the alkanes, organic acids, organic
hydroperoxides, and the radical-radical reactions occurring in the absence
of NOX. The increased level of detail for the acids, organic peroxides,
and radical-radical reactions is appropriate for a mechanism to be used
for acid deposition modeling.
The modifications made to the NCAR (1987) mechanism included some
relatively minor changes in some of the rate constants to be consistent
-------
with the latest evaluations of NASA (1987) and IUPAC (1988). Changes were
also made in the representation of glycolaldehyde, an acetaldehyde oxida-
tion product, and in the representation of lumped higher alkanes to be
more consistent with observed organic nitrate yields from such compounds
(Carter and Atkinson, 1985). Some changes in rate constants and product
yields for lumped species based on results of a preliminary analysis of
emissions data (see Section 3) were also made. The absorption coeffi-
cients and quantum yields for the photolysis reactions were not changed.
Most of the changes made to the mechanism were relatively minor; the basic
characteristics, approaches, and types of chemical approximations employed
in this version are the same as the version documented in the NCAR (1987)
report.
The model species employed in the RADM mechanism are listed in Table
1. There are a total of 18 inorganic species, 14 species representing the
various types of organic oxidation products, 11 species representing
emitted hydrocarbons, and 15 species representing reactive organic inter-
mediates. The set of inorganic species is the same as that employed in
most other state-of-the-art mechanisms. Like the Lurmann et al. (1986)
mechanism and the more recent mechanisms developed at SAPRC (Carter et
al., 1986, 1987; Lurmann et al., 1987; Carter, 1988), but unlike the
Carbon Bond mechanisms (Gery et al., 1988), the reactions of the organics
are represented on the molecular level, with some species being represent-
ed explicitly, and others being represented using either the "surrogate
species" or the "lumped molecule" lumping approaches. The organic species
represented explicitly in RADM are methane, ethane, ethene, formaldehyde,
formic acid, methyl hydroperoxide, and isoprene. The surrogate species
approach involves using a single representative compound to represent
reactions of other compounds assumed to have similar mechanisms or reac-
tivity. For example, the surrogate acetaldehyde (denoted as ALD) is used
to represent all C2+ aldehydes. The lumped molecule approach involves
representing mixtures of compounds by pseudo-molecules whose rate
constants and perhaps other mechanistic parameters are determined based on
the mixture they are used to represent. This lumping technique is used to
represent Co+ alkanes and alkenes and (to some extent) the aromatic
hydrocarbons. No attempt is made to conserve carbon in this mechanism, as
is the case for most other current molecularly-based mechanisms.
-------
Table 1. Model Species Employed in the March 1988 Version of the RADM
Gas-Phase Chemical Mechanism
Name Description
Transported Inorganic Species
S02 Sulfur Dioxide
03 Ozone
H202 Hydrogen Peroxide
NO Nitric Oxide
N02 Nitrogen Dioxide
N205 Nitrogen Pentoxide
HONO Nitrous Acid
HN03 Nitric Acid
HN04 Peroxy Nitric Acid
CO Carbon Monoxide
Inorganic Radical and Atomic Intermediates
HO Hydroxyl (OH) Radicals
H02 Hydroperoxyl Radicals
N03 Nitrate Radicals
03P 0(3P) Atoms
01D 0('D) Atoms
Constant Inorganic Species
02 Oxygen
N2 Nitrogen
M Air
H20 Water
Non-Reacting Inorganic Species
H2 Hydrogen
C02 Carbon Dioxide
SULF Sulfuric Acid (or sulfates)
Organic Hydrocarbons
CHI Methane
ETH Ethane
HC3 C3+ Alkanes with kOH < 5 x 103 (a)
HC5 Alkanes with kOH Between 0.5 and 1.0 x 101*
HC8 Alkanes with kOH > 1.0 x 1CT
OL2 Ethene
OLT Terminal Monoalkenes
OLI Internal Alkenes, Terpenes, Conjugated Dialkenes
ISO Isoprene
TOL Aromatic Hydrocarbons with kOH < 2 x 1o!j
XYL Aromatic Hydrocarbons with kOH > 2 x 10^
(continued)
8
-------
Table 1 (continued) - 2
Name Description
Organic Oxidation Products
HCHO Formaldehyde
ALD Lumped Higher Aldehydes
KET Ketones
GLY Glyoxal
MGLY Methyl Glyoxal
DCB Uncharacterized Reactive Aromatic Fragmentation Products
CSL Phenols and Cresols
PAN Acyl Peroxy Nitrates (other than those formed from DCB)
TPAN Acyl Peroxy Nitrates Formed from DCB
ONIT Other Organic Nitrates (primarily alkyl nitrates)
OP1 Methyl Hydroperoxide
OP2 Higher Organic Hydroperoxides
ORA1 Formic Acid
ORA2 Higher Organic Acids (other than peroxy acetic acid)
PAA Peroxy Acetic Acid
Organic Radicals
M02 Methyl Peroxy Radicals
ETHP Ethyl Peroxy Radicals
HC3P Peroxy Radicals from OH + HC3
HC5P Peroxy Radicals from OH + HC5
HC8P Peroxy Radicals from OH + HC8
OL2P Peroxy Radicals from OH + Ethene
OLTP Peroxy Radicals from OH + Isoprene and Terminal Alkenes
OLIP Peroxy Radicals from OH + Internal Alkenes
KETP Peroxy Radicals from OH + Ketones
TOLP Peroxy Radicals from OH + Benzene and Monoalkyl Benzenes
XYLP Peroxy Radicals from OH + Xylenes and More Reactive Aromatics
OLN Peroxy Radicals from NO? + Alkenes
X02 Chemical "Operator" Used to Account for NO-to-N02 Conversions
due to Secondarily Formed Organic Peroxy Radicals
AC03 Acyl Peroxy Radicals (other than those formed from DCB)
TC03 Acyl Peroxy Radicals Formed from DCB
XN02 Chemical Operator Used to Account for NOX Loss Due to
Reactions of NO? with Phenols and Cresols.
(a) kOH is the OH radical rate constant at 300 K in ppm~1 min .
-------
The RADM mechanism includes a relatively detailed, though still
approximate, representation of the reactions of peroxy radical inter-
mediates. This representation involves significantly more reactions than
used for these processes in most other current lumped mechanisms. It also
uses more model species to represent the products of these reactions in
the absence of NOX, with separate representations of methyl and higher
hydroperoxides, formic and higher acids, and separate representation of
peroxyacetic acid. Most of the reacting primary organics and organic
products have separate representations of their peroxy radical products,
and the mechanism includes explicit representation of the reactions of
these radicals with NO, with NC^ (for acyl peroxy radicals), with HC^,
with methyl peroxy (M02), and with acetylperoxy (AC03) radicals. In
representing explicitly almost all of the possible peroxy + M02 and peroxy
+ AC03 reactions, this mechanism is more detailed in this regard than most
other current mechanisms, which either ignore most of these processes
(Gery et al.f 1988), or uses approximations involving chemical "operators"
to represent them (Carter et al., 1986; Lurmann et al., 1987; Carter,
1988). However, this mechanism omits the reactions of the peroxy species
OLN (formed in the NO? + alkene reactions) with M02 and AC03, and ignores
peroxy + peroxy reactions not involving M02 and AC03. The effects of
alternative representations of peroxy radical reactions in this mechanism
are discussed in Section 6.3.
Table 2 gives a listing of the reactions and the thermal rate
constants used in the March 1988 RADM mechanism, and Table 3 gives the
absorption coefficients and quantum yields used for its photolysis reac-
tions. The reactions and thermal rate constants were obtained from
Stockwell in computer readable format, along with input data and results
of a test calculation employing constant photolysis rates. This mechanism
was implemented on the computers at SAPRC and Lurmann and Associates
(which each use different model calculation software), and the results of
test problem were duplicated. The absorption coefficients and quantum
yields used in this mechanism for atmospheric simulations were transmitted
to us separately, and we used these to calculate photolysis rates as a
function of zenith angle for the solar spectrum used in the test problems
described in Section 6.1 of this report. Computer-readable data files
implementing this mechanism are available from the authors of this report
upon request (with update information available from Stockwell at SUNY).
10
-------
Table 2. Listing of Reactions in the March 1988 Version of the RADM
Gas-Phase Chemical Mechanism
Rxn.
Label
(a) k(300)
Kinetic Parameters (b)
A Ea B
Reactions (c)
Inorganic Reactions (d)
001
002
003
004
005
006
007
008
009
022
023
024
025
026
027
028
029
030
031
032
033A
033B
034A
034B
035
036
037
038
039
040
041
042
(Phot. Set = N02R )
(Phot. Set = 030 1DR )
(Phot. Set = 0303PR )
(Phot. Set = HONOR )
(Phot. Set = HN03R )
(Phot. Set = HN04R )
(Phot. Set = N03NOR )
(Phot. Set = N03N02R )
(Phot. Set = H202R )
2
1
3
5
3
2
1
3
1
2
. 16E-05
.42E+04
.82E+04
.93E+04
.23E+05
.76E+01
. 02E+02
. 05E+00
.21E+04
. OOE+03
2
9
2
4
3
2
2
1
5
6
6
. 16E-05
.54E+03
. 64E+04
.70E+04
.23E-05
.94E+03
35E+03
.61E+01
.43E+03
(Falloff
.46E-03
. 90E+03
F= 0.60
0
-0
-0
-0
0
2
1
0
-0
.00
.24
.22
.14
.00
.78
.87
.99
.48
-4.30
-1.00
-1.00
-1.00
-1.00
-1.00
-1.00
-1.00
-1.00
Kinetics)
0
0
i
(Equilibrium
3
2
1
1
9
2
7
6
4
4
6
3
1
.24E-03
.55E+03
.79E-03
.34E-01
. 38E-02
.49E+03
.05E+03
.93E-10
.94E-02
. 11E+04
.08E-01
.67E+03
.85E+03
1
3
6
1
2
4
2
2
1
2
2
3
3
.95E+13
.23E+02
.82E-05
.11E-05
. 34E-06
.84E+03
(Falloff
.51E-02
.20E+04
F= 0.60
. 19E-10
.06E+02
.50E+04
.67E+01
. 67E+03
21
-1
-1
-5
-6
0
.00
.00
n =
-5.20
-2.40
1.00
with 031)
.66
.23
.95
.60
.32
.40
1.00
-1.00
-2.00
-2.00
-2.00
-1.00
Kinetics)
0
0
.00
.00
n=
-1
4
-0
2
0
.05
.97
.30
.44
.00
-4.60
-1.50
1.00
-2.00
-1.00
-1.00
-1.00
-1.00
(Falloff Kinetics)
7
2
.90E-02
.20E-1-03
0
0
F= 0.60
.00
.00
n=
-6.30
-1.50
1.00
N02 *
03 +
03 +
HONO
HN03
HN04
N03 +
N03 +
H202
03P +
03P +
01D +
01D +
01D +
03 +
03 +
03 +
H02 +
H02 +
(kO)
(kinf)
HN04
(Keq)
H02 +
H02 +
H02 +
H02 +
H202
NO +
(kO)
(kinf)
NO +
03 +
N03 +
N03 +
N03 +
N03 +
(kO)
(kinf)
HV =
HV =
HV =
+ HV
+ HV
+ HV
HV =
HV =
+ HV
02 +
N02
N2 =
02 =
H20
NO =
HO =
H02 =
NO =
N02
= H02
H02
H02
H02
H02
+ HO
HO =
NO +
N02 =
NO =
N02
H02
N02
03P +
01D +
03P f
= HO *
= HO +
= H02
NO +
N02 +
= HO +
NO
02
02
NO
N02
+ N02
02
03P
HO
M = 03
= NO t-
03P +
03P +
= HO +
N02 +
H02 +
HO +
N02 +
= HN04
+ N02
= H202
+ M =
+ H20
+ H20
= H02
HONO
02
N2
02
HO
02
02
12 02
HO
H202 +
= H202
= H202
* H20
02 = *2 N02
N03
N02 +
= NO +
= HN03
= N205
N02
N02 +
+ 02
M
02
(continued)
11
-------
Table 2 (continued) - 2
Rxn.
1 aha 1
Ltd Dei.
(a)
043
044
045
046
047
048
049
050A
050B
010
011
012
013
014
015
016
017
018
019
020
021
Kinetic Parameters (b)
k(300) A Ea B
(Equilibrium with 042)
2.27E-03 3.72E+13 22.26 1.00
2.94E-06 2.94E-06 0.00 -1.00
1.66E+04 (Falloff Kinetics)
9.34E-02 0.00 -5.20
3.52E+04 0.00 -2.30
F= 0.60 n= 1.00
2.16E+02 2.16E+02 0.00 -1.00
6.77E+03 1.91E+03 -0.75 -1.00
1.45E+05 6.75E+04 -0.46 -1.00
1.29E+03 (Falloff Kinetics)
1.08E-02 0.00 -5.30
2.20E+03 0.00 -1.00
F= 0.60 n= 1.00
2.20E+02 2.20E+02 0.00 -1.00
1.31E-04 1.31E-04 0.00 -2.00
(Phot. Set = HCHOMR )
(Phot. Set = HCHORR )
(Phot. Set = ALDR )
(Phot. Set = OPR )
(Phot. Set = OPR )
(Phot. Set = PAAR )
(Phot. Set = KETR )
(Phot. Set = GLY2R )
(Phot. Set = GLY1R )
(Phot. Set = MGLYR )
(Phot. Set = DCBR )
(Phot. Set = ONJTR )
Rossfif1 1 finQ ( r* j
I\ wu w WXwllv \ w /
N205 = N02 + N03
(Keq)
N205 + H20 = #2 HN03
HO + N02 = HN03
(kO)
(kinf)
HO + HN03 = N03 + H20
HO + HN04 = N02 + H20
HO * H02 = H20 + 02
HO + S02 = SULF + H02
(kO)
(kinf)
CO + HO = H02 + C02
CO + HO + M = H02 + C02
Organic Photolysis Reactions
HCHO + HV = H2 + CO
HCHO + HV = H02 + H02 + CO
ALD * HV = M02 + H02 + CO
OP1 f HV = HCHO + H02 + HO
OP2 + HV = ALD + H02 + HO
PAA + HV = M02 * C02 * HO
KET + HV = AC03 + ETHP
GLY + HV = #.13 HCHO + #1.87 CO
GLY * HV = #.45 HCHO + 11.55 CO +
#.80 H02
MGLY + HV = AC03 + H02 + CO
DCB +'HV = #.98 H02 + #.02 AC03 +
TC03
ONIT + HV = #.20 ALD + #.80 KET *
H02 + N02
051
052
053
054
1
4
3
7
.29Ef
.11E+
.76E-h
.86E+
01
02
03
03
9
1
2
2
.23E+
.81E+
.28Ef
.79E+
02
03
04
04
2
0
1
0
.54
.88
.07
.75
1
1
-1
-1
.00
.00
.00
.00
CH4
ETH
HC3
#
#
HC5
Organic + OH Reactions
+ HO = M02 + H20
+ HO = ETHP + H20
+ HO = #.90 HC3P + #.10 H02 +
,014 HCHO + 1.069 ALD +
.026 KET + H20
+ HO = HC5P + #.27 X02 + H20
(continued)
12
-------
Table 2 (continued) - 3
Rxn.
Label
(a)
055
056
057
058
074
059
060
063
064
065
066
067
061
062
068
069
070
071
072
073
Kinetic Parameters (b)
k(300)
1.
1.
3.
9.
1.
9.
66E+04
24E+04
82E+04
24E+04
46E+05
02E+03
4.54E+04
1 . 32E+04
2,
1,
1
2
5
5
4
1
1
1
2
3
.37E+04
.47E+03
. 69E+04
.50E+04
.87E+04
.28E+04
. 11E+04
.47E+04
.47E+04
.47E+04
.06E+02
.76E+03
5.
3.
7.
1.
3.
3.
3.
1,
1,
1,
1,
2
5
5
4
1
1
1
9
2
A
87E+04
16E+03
12E+03
48E+04
74E+04
08E+03
08E-.-04
.32E+04
.01E+04
.76E+04
.69E+04
.50E+04
.87E+04
.28E+04
. 11E+04
. 47E+04
.47E+04
.47E+04
. 06E+02
.28E+04
Ea
0.
-0.
-1.
-1.
-0.
-0.
-0.
0.
-0.
1.
0.
0.
0.
0.
75
82
00
09
81
64
23
00
51
,48
,00
,00
,00
.00
0.00
0.00
0,
.00
0.00
0
1
.88
.07
B
-1.
-1.
-1.
-1.
-1.
-1.
-1.
-1.
-1.
-1.
-1.
-1.
-1.
-1.
-1.
-1.
-1.
-1.
1,
-1,
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
,00
,00
.00
.00
.00
Reactions (
HC8 +
OL2 +
OLT H-
OLI +
ISO +
TOL +
# . 16
XYL +
#.17
HCHO +
ALD -
KET *
GLY *
MGLY -»
CSL +
#.9
CSL *
DCB +
OP1 +
#.5
OP2 +
#.5
PAA +
PAN +
ONIT <
HO =
HO =
HO =
HO =
HO =
HO =
H02
HO =
H02
0
HC8P * #.78 X02 + H20
OL2P
OLTP
OLIP
OLTP
#.84 TOLP + #.16 CSL +
#.83 XYLP -- #.17 CSL +
HO = H02 -i- CO + H20
HO =
HO =
HO =
AC03 + H20
KETP * H20
H02 + #2 CO * H20
- HO = AC03 + CO 4- H20
HO =
TC03
HO =
HO =
HO =
HO
HO =
HO
HO =
HO =
* HO
Organic +
093
094
095
096
097
098
099
100
101
102
9
3
9
3
3
3
1
2
1
8
.23E-01
. 65E+00
.23E-01
.65E+00
. 65E+00
.23E-I-04
.72E-01
.33E+01
.84E+03
.53E+02
8
2
8
2
2
3
2
1
4
8
.81E+02
.06E+03
.81E+02
.06E+03
.06E+03
.23E+04
.94E+03
.47E-.-04
.74E+04
.53E+02
4
3
4
3
3
0
5
3
1
0
.09
.78
.09
.78
.78
.00
.81
.84
.94
.00
-1
-1
-1
-1
-1
-1
-1
-1
-1
-1
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
HCHO
ALD +
GLY +
MGLY -
DCB +
CSL f
OL2 +
OLT >
OLI +
ISO +
« N03
N03
N03
« N03
N03
N03
N03
N03
N03
N03
#.1 H02 * #.9 X02 +
CSL
TC03 * H20
#.5 M02 + #.5 HCHO +
#.5 HC3P * #.5 ALD +
AC03 -* H20
HCHO + N03 + X02
= HC3P + N02
N03 Reactions
= H02 + HN03 + CO
= AC03 f HN03
= HN03 + H02 + #2 CO
= HN03 * AC03 + CO
= HN03 -t- TC03
= HN03 + XN02 -t- #.5 CSL
= OLN
= OLN
= OLN
= OLN
103 2.72E-03 1.76E+01 5.23 -1.00
Organic + 03 Reactions
OL2 + 03 = HCHO -H #.42 CO +
#.4 ORA1 * #.12 H02
(continued)
13
-------
Table 2 (continued) - 4
Rxn. Kinetic Parameters (b)
(a) k(300) A Ea B
104 1.74E-02 1.94E+01 4.18-1.00 OLT
106 3.02E-01 1.33E+01 2.26-1.00 OL
r «
.33
.20
.22
+
.10
.29
#.26
106 2.20E-02 1.81E+01 4.00 -1.00 ISO +
#.33
03 =
CO +
ORA2
M02
03 =
KET
ORA2
H02
03 =
CO +
#
+
#
+
+
#
^
/
.53 HCHO
#.20 ORA1
+
*
#
.
.18
#
+
#
.
#
m
.53
#.
#.20 ORA2 +
#.22 M02
+
1
+
#
.50
ALD +
.23 H02 +
10 HO
HCHO
23 CO
+
*-
-t-
#
#
#
.06
.72
.06
CH4
ALD +
ORA1 +
.09 CH4 +
14 HO
HCHO
+
+
#
#
.31
.50
M02
ALD *
20 ORA1 +
#
.
.23
10 HO
H02 +
075 7.52E+03 4.11E+03 -0.36 -1.00
076 2.90E-02 1.17E+18 26.91 0.00
077 6.90E+03 6.90E+03 0.00 -1.00
078 2.90E-02 1.17E+18 26.91 0.00
086 1.12E+04 6.17E+03 -0.36 -1.00
087 1.12E+04 6.17E+03 -0.36 -1.00
079 1.12E+04 6.17E+03 -0.36 -1.00
080 1.12E+04 6.17E+03 -0.36 -1.00
081 1.12E+04 6.17E+03 -0.36 -1.00
082 1.12E+OM 6.17E+03 -0.36 -1.00
083 1.12E+04 6.17E+03 -0.36 -1.00
084 1.12E+04 6.17E+03 -0.36 -1.00
085 1.12E+04 6.17E-I-03 -0.36 -1.00
088 1.12E+04 6.17E+03 -0.36 -1.00
089 1.12E+04 6.17E+03 -0.36 -1.00
Organic Peroxy + N02 Reactions
AC03 + N02 = PAN
PAN = AC03 f N02
TC03 f N02 = TPAN
TPAN = TC03 + N02
Organic Peroxy + NO Reactions
AC03 * NO = M02 + N02
TC03 * NO = N02 i- #.92 H02 +
#.89 GLY + #.11 MGLY +
#.05 AC03 + #.95 CO + #2 X02
M02 + NO = HCHO + H02 * N02
HC3P + NO = #.69 ALD + #.26 KET +
#.14 HCHO -t- #.03 ONIT +
#.97 N02 + #.97 H02
HC5P + NO = #.36 ALD * #.76 KET +
#.10 ONIT + #.90 N02 + #.90 H02
HC8P + NO = #.44 ALD * #1.05 KET +
#.05 HCHO + #.24 ONIT +
#.76 N02 + #.76 H02
OL2P + NO = #1.6 HCHO + H02 +
N02 + #.2 ALD
OLTP * NO = ALD + HCHO i- H02 + N02
OLIP + NO = H02
#.28 HCHO + #
TOLP + NO = N02
#.724 MGLY + #
#.905 DCB
XYLP * NO = N02
HCHO i- H02
#1.45 ALD +
10 KET + N02
H02 +
181 GLY +
H02 + MGLY + DCB
(continued)
14
-------
Table 2 (continued) - 5
Rxn.
Label
(a) k(300)
Kinetic Parameters (b)
A Ea B
Reactions (c)
090
091
092
150
107
108
109
110
111
112
113
111*
115
117
118
120
146
116
119
142
145
130
134
135
136
137
138
139
1.12E+04 6.17E+03 -0.36 -1.00
1.12E+04 6.17E+03 -0.36 -1.00
1.12E+04 6.17E+03 -0.36 -1.00
1.12E+04 6.17E+03 -0.36 -1.00
8.61E+03
ETHP + NO = ALD +
KETP + NO = MGLY
OLN + NO = HCHO +
X02 + NO = N02
H02 + N02
+ N02 + H02
ALD + #2 N02
Organic Peroxy + H02 Reactions
1.13E+02 -2.58 -1.00 H02 +
(Same
(Same
(Same
(Same
(Same
(Same
(Same
(Same
(Same
(Same
(Same
(Same
(Same
(Same
k
k
k
k
k
k
k
k
k
k
k
k
k
k
as
as
as
as
as
as
as
as
as
as
as
as
as
as
107 )
107 )
107 )
107 )
107 )
107 )
107 )
107 )
107 )
107 )
107 )
107 )
107 )
107 )
H02 +
H02 +
H02 +
H02 +
H02 +
H02 +
H02 +
H02 +
H02 +
H02 +
H02 +
H02 +
H02 +
H02 +
M02 :
ETHP
HC3P
HC5P
HC8P
OL2P
OLTP
OL1P
KETP
TOLP
XYLP
OLN
X02
AC03
TC03
: OP1
= OP2
= OP2
= OP2
= OP2
= OP2
= OP2
= OP2
= OP2
= OP2
= OP2
= ON IT
= OP2
= PAA
= OP2
Organic Peroxy + AC03 Reactions
3.64E+03 1.75Ef03 -0.44 -1.00
3.64E+03 1.75E+03 -0.44 -1.00
2.93E+03 1.41E+03 -0.44 -1.00
1.04E+03 4.99E+02 -0.44 -1.00
3.06E+02 1.47E+02 -0.44 -1.00
2.57Ef02 1.23E+02 -0.44 -1.00
2.20E+02 1.06E+02 -0.44 -1.00
1.04E+03 4.99E+02 -0.44 -1.00
1.04E+03 4.99E+02 -0.44 -1.00
AC03 * AC03
AC03 + TC03
*.890 GLY
*.05 AC03
M02 + AC03 =
#.5 M02 +
ETHP + AC03
1.5 M02 -K
HC3P t- AC03
1.5 H02 +
HC5P + AC03
*.86 KET +
1.5 ORA2
HC8P + AC03
#.5 H02 +
OL2P + AC03
1.5 H02 +
OLTP + AC03
#.5 H02 +
= #2 M02
= M02 + #.92 H02 -H
+ 1.11 MGLY +
* 1.950 CO + *2 X02
HCHO + #.5 H02 +
1.5 ORA2
= ALD + #.5 H02 *
#.5 ORA2
= 1.2 ALD + 1.8 KET +
1.5 M02 - 1.5 ORA2
= 1.11 ALD +
1.5 H02 +1.5 M02 +
= 1.10 ALD + #.9 KET +
#.5 M02 +1.5 ORA2
=1.8 HCHO + #.6 ALD +
1.5 M02 + 1.5 ORA2
= ALD + 1.5 HCHO +
1.5 M02 +1.5 ORA2
(continued)
15
-------
Table 2 (continued) - 6
Rxn.
Label
(a) k(300)
Kinetic Parameters (b)
A Ea B
Reactions (c)
140 1.28E+02 6.17E+01 -0.44 -1.00
141
143
144
148
121
122
123
124
125
126
127
128
129
131
132
133
1.28E+02
1.28E+02
1 . 28E+02
1.28E+02
5.81E+02
4.28E+02
1.28E+02
1.04E+02
8.86E+01
4.28E+02
4.28E+02
5.19E+01
5.19Ei-01
5.19E+01
5.19E+01
2.93E+03
2.79E+02 -0.44 -1.00
2.06E+02 -0.44 -1.00
6.17E+01 -0.44 -1.00
4.99E+01 -0.44 -1.00
4.26E+01 -0.44 -1.00
2.06E+02 -0.44 -1.00
2.06E+02 -0.44 -1.00
2.50E+01 -0.44 -1.00
2.50E+01 -0.44 -1.00
2.50Ef01 -0.44 -1.00
2.93E+03 1.41E+03 -0.44 -1.00
147 5.19E+01 2.50Ef01 -0.44 -1.00
OLIP + AC03 = #.725 ALD +
#.55 KET + #.14 HCHO + #.5 H02 +
#.5 M02 + t.5 ORA2
KETP + AC03 = MGLY + #.5 H02 +
#.5 M02 + #.5 ORA2
AC03 * TOLP = M02 + #.8 MGLY *
8.2 GLY + DCB + H02
AC03 * XYLP = M02 + MGLY + DCB +
H02
X02 + AC03 = M02
Organic Peroxy + M02 Reactions
M02 + M02 = #1.5 HCHO + H02
M02 -t- ETHP = #.75 HCHO + H02 +
#.75 ALD
M02 + HC3P = #.75 HCHO * H02 *
#.15 ALD * #.60 KET
M02 + HC5P = #.75 HCHO * H02 *
#.105 ALD + #.645 KET
M02 + HC8P = #.75 HCHO + H02 *
#.075 ALD * #.675 KET
M02 + OL2P = #1.55 HCHO * H02 +
#.35 ALD
M02 + OLTP = #1.25 HCHO i- H02 +
#.75 ALD
M02
H02
H02
OLIP = 1.89 HCHO
#.725 ALD + #.55 KET
M02 + KETP = #.75 HCHO
#.75 MGLY
M02 + TOLP = #.976 HCHO +
#1.9 H02 + #.724 MGLY +
#.181 GLY + #.905 DCB
M02 1- XYLP = HCHO + #2 H02 +
MGLY + DCB
M02 + TC03 = #.5 HCHO + #.5 ORA2 +
#.46 H02 * #.445 GLY +
#.055 MGLY + #.025 AC03 +
#.475 CO + X02
M02 + X02 = HCHO + H02
(continued)
16
-------
Table 2 (continued) - 7
Kinetic Parameters (b)
Rxn.
Label
(a) k(300) A Ea B
Reactions (c)
Other Organic Peroxy + Peroxy
Reactions
149 1.10E+00 5.28E-01 -0.44 -1.00 X02 + X02 =
151 1.12E+04 6.17E+03 -0.36 -1.00
152 (Same k as 107 )
153 5.19E+01 2.50E+01 -0.44 -1.00
154 1.28E+02 6.17E+01 -0.44 -1.00
155 1.10E+00 5.28E-01 -0.44 -1.00
XN02 Reactions
XN02 + N02 = ONIT
XN02 + H02 = OP2
XN02 + M02 = HCHO
XN02 + AC03 = M02
XN02 + XN02 =
H02
Notes:
(a) Reaction labels in this listing are the reaction sequence numbers
used by Stockwell to implement this mechanism. Some reactions were
re-arranged for the purpose of this listing. Some reactions were
split into two reactions for implementation on SAPRC modeling
software.
(b) Except as indicated otherwise, rate constants for reactions in this
table are given by the expression
k = A (T/300)8 e~Ea/RT
where k and A are in ppm, minute units, T is the temperature in
degrees K, and R is 0.0019872 kcal deg mole . If the
notation "Phot. Set" is used, then this is a photolysis reaction
whose rate constant is calculated using the indicated set of
absorption coefficients and quantum yields given in Table 3. If
the notation "Falloff" is used, then this reaction is both
temperature and pressure dependent, and its rate constant is given
by
k = t(k0 M) / (1 + [k0 M / kinf] ) ]
where
g = 1 / [1 + (Iog10 [k0 M / kinf] / n)2 ],
M is the pressure in ppm, k« is the low pressure limiting rate
constant calculated as a function of temperature from its
(continued)
17
-------
Table 2 (continued) - 8
corresponding A, Ea and B parameters given on the next line,
is the high pressure limiting rate constant calculated from its
parameters given on the line below that, and the parameters F
and n are given as indicated in the line below that. If the
notation "Equilibrium with" is used, the rate constant is calculated
by multiplying the rate constant for the indicated reaction by the
equilibrium constant, whose T=300 K value and temperature dependence
parameters, analogous to A, Ea and B, are given on the following
line.
(c) If a number preceded by a "K" appears on the list of reaction
products, then the number is a product yield coefficient for the
species whose name follows it. Otherwise the product yield is 1.0.
(d) These reactions are also used in the modified (RADM-M and RADM-P)
versions of this mechanism.
18
-------
Table 3. Absorption Coefficients and Quantum Yields Used in Versions
of the RADM Gas-Phase Chemical Mechanism Evaluated in this
Study
WL
(nm)
Abs
(cm2)
QY
WL
(nm)
Abs
(cm2)
QY
WL
(nm)
Abs
(cm2)
QY
Data Used in the March 1988 and the Recommended Versions of the
RADM Mechanism
Photolysis Set = N02R
186.0 2.59E-19 1.000
191.4 2.73E-19 1.000
197.0 2.46E-19 1.000
203.0 2.92E-19 1.000
209.4 4.10E-19 1-000
216.2 4.60E-19 1.000
223.5 3.70E-19 1.000
231.2 2.76E-19 1.000
239.5 7.85E-20 1.000
248.4 2.13E-20 1.000
258.1 1.60E-20 1.000
268.5 2.72E-20 1.000
279.7 5.26E-20 1.000
292.0 9.39E-20 1.000
303.0 1.51E-19 1-000
306.0 1.53E-19 1.000
309.0 1.81E-19 ^.000
312.0 2.03E-19 1.000
316.0 2.20E-19 1.000
330.0 3.04E-19 1.000
345.0 4.05E-19 1.000
360.0 4.94E-19 1.000
375.0 5.63E-19 1.000
390.0 6.00E-19 0.981
405.0 5.49E-19 0.356
420.0 O.OOE-01 0.018
187.8 2.72E-19 1.000
193.2 2.51E-19 1.000
199.0 2.46E-19 1.000
205.1 3.53E-19 1.000
211.6 4.36E-19 1.000
218.6 4.17E-19 1.000
226.0 3.89E-19 1.000
233-9 1.71E-19 1.000
242.4 6.12E-20 1.000
251.6 1.40E-20 1.000
261.4 1.80E-20 1.000
272.1 3-27E-20 1.000
283.7 6.27E-20 1.000
296.3 1.09E-19 1.000
304.0 1.56E-19 1.000
307.0 1.61E-19 1.000
310.0 1.83E-19 1.000
313.0 2.01E-19 1.000
320.0 2.50E-19 1.000
335.0 3.40E-19 1.000
350.0 4.46E-19 1.000
365.0 5.29E-19 1.000
380.0 5.61E-19 1.000
395.0 5.69E-19 0.922
410.0 5.98E-19 0.135
189.6 2.85E-19 1.000
195.1 2.44E-19 1.000
201.0 2.82E-19 1.000
207.3 3-74E-19 1.000
213.9 4.45E-19 1.000
221.0 4.54E-19 1.000
228.6 2.69E-19 1.000
236.7 1.62E-19 1.000
245.4 2.88E-20 1.000
254.8 1.49E-20 1.000
264.9 2.18E-20 1.000
275.9 4.10E-20 1.000
287.8 7.62E-20 1.000
300.5 1.26E-19 1.000
305.0 1.55E-19 1.000
308.0 1.65E-19 1.000
311.0 1.91E-19 1.000
314.0 1.99E-19 1-000
325-0 2.81E-19 1.000
340.0 3.84E-19 1.000
355.0 4.88E-19 1-000
370.0 5.40E-19 1.000
385.0 5.72E-19 0.997
400.0 6.18E-19 0.693
415.0 5.77E-19 0.060
Photolysis Set = 0301DR
186.0 6.23E-19 0.900
191.4 4.75E-19 0.900
197.0 3.46E-19 0.900
203.0 3.27E-19 0.900
209.4 5.46E-19 0.900
216.2 1.20E-18 0.900
223.5 2.57E-18 0.900
231.2 4.94E-18 0.900
239.5 8.11E-18 0.900
187.8 5.76E-19 0.900
193.2 4.27E-19 0.900
199.0 3-22E-19 0.900
205.1 3.65E-19 0.900
211.6 7.05E-19 0.900
218.6 1.56E-18 0.900
226.0 3-24E-18 0.900
233.9 5.93E-18 0.900
242.4 9.20E-18 0.900
189.6 5.26E-19 0.900
195.1 3.82E-19 0.900
201.0 3.14E-19 0.900
207.3 4.37E-19 0.900
213.9 9.20E-19 0.900
221.0 2.00E-18 0.900
228.6 4.04E-18 0.900
236.7 7.01E-18 0.900
245.4 1.02E-17 0.900
(continued)
19
-------
Table 3 (continued) - 2
WL
(nm)
Abs
(cm )
QY
WL
(nm)
Abs
(cm2)
QY
WL
(nm)
Abs
(cm2)
QY
248.4 1.10E-17 0.900
258.1 1.14E-17 0.900
268.5 8.35E-18 0.900
279.7 4.03E-18 0.900
292.0 1.11E-18 0.900
303-0 2.64E-19 0.900
306.0 1.79E-19 0.837
309.0 1.19E-19 0.639
312.0 8.01E-20 0.304
316.0 4.69E-20 0.065
Photolysis Set = 0303PR
186.0 6.23E-19 0.100
191.4 4.75E-19 0.100
197.0 3.46E-19 0.100
203.0 3.27E-19 0.100
209.4 5.46E-19 0.100
216.2 1.20E-18 0.100
223.5 2.57E-18 0.100
231.2 4.94E-18 0.100
239.5 8.11E-18 0.100
248.4 1.10E-17 0.100
258.1 1.14E-17 0.100
268.5 8.35E-18 0.100
279.7 4.03E-18 0.100
292.0 1.11E-18 0.100
303.0 2.64E-19 0.100
306.0 1.79E-19 0.163
309.0 1.19E-19 0.361
312.0 8.01E-20 0.696
316.0 4.69E-20 0.935
330.0 6.69E-21 1.000
345.0 7.21E-22 1.000
360.0 5.49E-23 1.000
415.0 3.14E-23 1.000
430.0 6.83E-23 1.000
445.0 1.49E-22 1.000
460.0 3-57E-22 1.000
475.0 4.89E-22 1.000
490.0 8.28E-22 1.000
505.0 1.62E-21 1.000
520.0 1.78E-21 1.000
535.0 2.74E-21 1.000
550.0 3.17E-21 1.000
251.6 1.14E-17 0.900
261.4 1.08E-17 0.900
272.1 6.93E-18 0.900
283.7 2.79E-18 0.900
296.3 6.37E-19 0.900
304.0 2.33E-19 0.899
307.0 1.57E-19 0.789
310.0 1.03E-19 0.531
313.0 6.90E-20 0.216
320.0 2.65E-20 0.000
187.8 5.76E-19 0.100
193.2 4.27E-19 0.100
199.0 3.22E-19 0.100
205.1 3.65E-19 0.100
211.6 7.05E-19 0.100
218.6 1.56E-18 0.100
226.0 3.24E-18 0.100
233.9 5.93E-18 0.100
242.4 9.20E-18 0.100
251.6 1.14E-17 0.100
261.4 1.08E-17 0.100
272.1 6.93E-18 0.100
283.7 2.79E-18 0.100
296.3 6.37E-19 0.100
304.0 2.33E-19 0.101
307.0 1.57E-19 0.211
310.0 1.03E-19 0.469
313.0 6.90E-20 0.784
320.0 2.65E-20 1.000
335.0 3-06E-21 1.000
350.0 2.66E-22 1.000
365.0 O.OOE-01 1.000
420.0 3.99E-23 1.000
435.0 8.66E-23 1.000
450.0 1.71E-22 1.000
465.0 3.68E-22 1.000
480.0 7.11E-22 1.000
495.0 9.09E-22 1.000
510.0 1.58E-21 1.000
525.0 2.07E-21 1.000
540.0 2.88E-21 1.000
555.0 3.36E-21 1.000
254.8 1.16E-17 0.900
264.9 9.64E-18 0.900
275.9 5-43E-18 0.900
287.8 1.81E-18 0.900
300.5 3.63E-19 0.900
305.0 1.99E-19 0.872
308.0 1.36E-19 0.725
311.0 9.13E-20 0.413
314.0 6.33E-20 0.150
189.6 5.26E-19 0.100
195.1 3.82E-19 0.100
201.0 3.14E-19 0.100
207.3 4.37E-19 0.100
213.9 9.20E-19 0.100
221.0 2.00E-18 0.100
228.6 4.04E-18 0.100
236.7 7.01E-18 0.100
245.4 1.02E-17 0.100
254.8 1.16E-17 0.100
264.9 9.64E-18 0.100
275.9 5.43E-18 0.100
287.8 1.81E-18 0.100
300.5 3.63E-19 0.100
305.0 1.99E-19 0.128
308.0 1.36E-19 0.275
311.0 9.13E-20 0.587
314.0 6.33E-20 0.850
325.0 1.30E-20 1.000
340.0 1.37E-21 1.000
355.0 1.09E-22 1.000
410.0 2.91E-23 1.000
425.0 6.54E-23 1.000
440.0 1.25E-22 1.000
455.0 2.12E-22 1.000
470.0 4.06E-22 1.000
485.0 8.43E-22 1.000
500.0 1.22E-21 1.000
515.0 1.60E-21 1.000
530.0 2.55E-21 1.000
545.0 3.07E-21 1.000
560.0 3.88E-21 1.000
(continued)
20
-------
Table 3 (continued) - 3
WL Abs QY
( nm ) ( cm )
565.0 4.31E-21 1.000
580.0 4.55E-21 1.000
595.0 4.61E-21 1.000
610.0 4.54E-21 1.000
625.0 3.60E-21 1.000
640.0 2.74E-21 1.000
660.0 2.07E-21 1.000
690.0 1.11E-21 1.000
720.0 6.40E-22 1.000
Photolysis Set = HONOR
311.0 2.50E-22 1.000
314.0 4.50E-21 1.000
325.0 3-93E-20 1.000
340.0 1.69E-19 1.000
355.0 2.58E-19 1.000
370.0 2.06E-19 1.000
385-0 1.46E-19 1.000
Photolysis Set = HN03R
191.4 1.37E-17 1.000
197.0 8.54E-18 1.000
203.0 3.60E-18 1.000
209.4 9.93E-19 1.000
216.2 2.57E-19 1.000
223.5 9.10E-20 1.000
231.2 4.66E-20 1.000
239.5 2.48E-20 1.000
248.4 1.92E-20 1.000
258.1 1.87E-20 1.000
268.5 1.57E-20 1.000
279.7 1.01E-20 1.000
292.0 4.44E-21 1.000
303.0 1.74E-21 1.000
306.0 1.24E-21 1.000
309.0 7.85E-22 1.000
312.0 5.15E-22 1.000
316.0 2.00E-22 1.000
330.0 O.OOE-01 1.000
WL Abs QY
( nm ) ( cm )
570.0 4.67E-21 1.000
585.0 4.35E-21 1.000
600.0 4.89E-21 1.000
615.0 4.24E-21 1.000
630.0 3-43E-21 1.000
644.8 2.61E-21 1.000
670.0 1.72E-21 1.000
700.0 9-13E-22 1.000
730.0 5.14E-22 1.000
312.0 2.02E-21 1.000
316.0 5.26E-21 1.000
330.0 8.51E-20 1.000
345.0 1.14E-19 1.000
360.0 7.31E-20 1.000
375.0 3.71E-20 1.000
390.0 1.42E-20 1.000
193.2 1.22E-17 1.000
199.0 6.66E-18 1.000
205.1 2.48E-18 1.000
211.6 6.64E-19 1.000
218.6 1.65E-19 1.000
226.0 7.26E-20 1.000
233.9 3.68E-20 1.000
242.4 2.20E-20 1.000
251.6 1.90E-20 1.000
261.4 1.77E-20 1.000
272.1 1.40E-20 1.000
283.7 8.13E-21 1.000
296.3 2.95E-21 1.000
304.0 1.55E-21 1.000
307.0 1.09E-21 1.000
310.0 6.71E-22 1.000
313.0 4.37E-22 1.000
320.0 8.50E-23 1.000
WL Abs QY
( nm ) ( cm )
575.0 4.75E-21 1.000
590.0 4.42E-21 1.000
605.0 4.84E-21 1.000
620.0 3.90E-21 1.000
635.0 3.17E-21 1.000
651.1 2.40E-21 1.000
680.0 1.37E-21 1.000
710.0 7.93E-22 1.000
313.0 3.97E-21 1.000
320.0 3-93E-20 1.000
335.0 6.31E-20 1.000
350.0 1.07E-19 1.000
365.0 1.84E-19 1.000
380.0 8.25E-20 1.000
395.0 O.OOE-01 1.000
195.1 1.04E-17 1.000
201.0 5.12E-18 1.000
207.3 1.67E-18 1.000
213-9 3.78E-19 1.000
221.0 1.22E-19 1.000
228.6 5.77E-20 1.000
236.7 3.00E-20 1.000
245.4 2.03E-20 1.000
254.8 1.89E-20 1.000
264.9 1.67E-20 1.000
275.9 1.21E-20 1.000
287.8 6.17E-21 1.000
300.5 2.08E-21 1.000
305.0 1.39E-21 1.000
308.0 9.35E-22 1.000
311.0 5.93E-22 1.000
314.0 3.30E-22 1.000
325.0 3.50E-23 1.000
(continued)
21
-------
Table 3 (continued) - 4
WL Abs QY
(nm) (cm )
Photolysis Set = HN04R
189.6 1.00E-17 1.000
195.1 6.60E-18 1.000
201.0 4.00E-18 1.000
207.3 2.40E-18 1.000
213.9 1.50E-18 1.000
221.0 1.00E-18 1.000
228.6 7.50E-19 1.000
236.7 6.00E-19 1.000
245.4 4.40E-19 1.000
254.8 3.00E-19 1.000
264.9 2.00E-19 1.000
275.9 1.25E-19 1.000
287.8 4.50E-20 1.000
300.5 1.35E-20 1.000
305.0 9.00E-21 1.000
308.0 7.00E-21 1.000
311.0 5.50E-21 1.000
314.0 4.20E-21 1.000
325.0 1.50E-21 1.000
Photolysis Set = N03NOR
585.0 2.66E-18 0.032
600.0 2.69E-18 0.311
615.0 1.84E-18 0.194
630.0 4.89E-18 0.078
WL Abs QY
(nm) (cm2)
191.4 9.00E-18 1.000
197.0 5.60E-18 1.000
203-0 3.40E-18 1.000
209.4 2.10E-18 1.000
216.2 1.30E-18 1.000
223.5 9.00E-19 1.000
231.2 6.80E-19 1.000
239.5 5.50E-19 1.000
248.4 4.00E-19 1.000
258.1 2.70E-19 1.000
268.5 1.80E-19 1.000
279.7 9.00E-20 1.000
292.0 3.10E-20 1.000
303.0 1.10E-20 1.000
306.0 8.50E-21 1.000
309.0 6.60E-21 1.000
312.0 5.10E-21 1.000
316.0 3.30E-21 1.000
330.0 1.00E-21 1.000
590.0 4.74E-18 0.191
605.0 3.05E-18 0.272
620.0 4.61E-18 0.156
635.0 1.43E-18 0.039
WL Abs QY
(nra) (cm )
193-2 7.60E-18 1.000
199.0 4.80E-18 1.000
205-1 2.90E-18 1.000
211.6 1.70E-18 1.000
218.6 1.10E-18 1.000
226.0 8.20E-19 1.000
233.9 6.50E-19 1.000
242.4 5.00E-19 1.000
251.6 3.50E-19 1.000
261.4 2.35E-19 1.000
272.1 1.55E-19 1.000
283.7 6.50E-20 1.000
296.3 2.10E-20 1.000
304.0 1.00E-20 1.000
307.0 7.70E-21 1.000
310.0 6.00E-21 1.000
313.0 4.50E-21 1.000
320.0 2.40E-21 1.000
335.0 O.OOE-01 1.000
595.0 3.74E-18 0.350
610.0 1.51E-18 0.233
625.0 8.17E-18 0.117
640.0 1.08E-18 0.000
Photolysis Set = N03N02R
405.0 2.80E-20 1.000
420.0 8.20E-20 1.000
435.0 1.84E-19 1.000
450.0 2.83E-19 1.000
465.0 4.34E-19 1.000
480.0 6.44E-19 1.000
495.0 9.67E-19 1.000
510.0 1.32E-18 1.000
525.0 1.49E-18 1.000
540.0 1.83E-18 1.000
555.0 2.68E-18 1.000
570.0 2.49E-18 1.000
585.0 2.66E-18 0.968
600.0 2.69E-18 0.578
410.0 4.45E-20 1.000
425.0 1.05E-19 1.000
440.0 1.94E-19 1.000
455.0 3.35E-19 1.000
470.0 5.10E-19 1.000
485.0 6.86E-19 1.000
500.0 9.90E-19
515.0 1.40E-18
530.0 1.93E-18
545.0 1.82E-18
560.0 3.07E-18
575.0 2.61E-18
.000
.000
.000
.000
.000
.000
590.0 4.74E-18 0.809
605.0 3.05E-18 0.506
415.0 5.35E-20 1.000
430.0 1.29E-19 1.000
445.0 2.23E-19 1-000
460.0 3.72E-19 1.000
475.0 6.03E-19 1.000
490.0 8.80E-19 1.000
505.0 1.09E-18 1.000
520.0 1.45E-18 1.000
535.0 2.04E-18 1.000
550.0 2.35E-18 1.000
565.0 2.54E-18 1.000
580.0 2.91E-18 1.000
595.0 3.74E-18 0.650
610.0 1.51E-18 0.433
(continued)
22
-------
Table 3 (continued) - 5
WL
(nm)
Abs
(cm2)
QY
WL
(nm)
Abs
(cm )
QY
WL
(nm)
Abs
(cm2)
QY
615.0 1.84E-18 0.361
630.0 4.89E-18 0.114
Photolysis Set = H202R
620.0 4.61E-18 0.289
635.0 1.43E-18 0.072
625.0 8.17E-18 0.217
640.0 1.08E-18 0.000
191.4
197.0
203.0
209.4
216.2
223.5
231.2
239.5
248.4
258.1
268.5
279.7
292.0
303.0
306.0
309.0
312.0
316.0
330.0
345.0
3.21E-19
5.27E-19
4.34E-19
3.56E-19
2.88E-19
2.26E-19
1.71E-19
1.25E-19
8.81E-20
5.80E-20
3.54E-20
2.02E-20
1.03E-20
5.59E-21
4.69E-21
4.00E-21
3.38E-21
2.65E-21
1.19E-21
5.03E-22
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
.000
193.
199-
205.
211.
218.
226.
233.
242.
251-
261.
272.
283.
296.
304.
307.
310.
313.
320.
335.
350.
2 3
0 4
1 4
6 3
6 2
0 2
9 1
4 1
6 7
4 4
1 2
7 1
3 8
0 5
0 4
0 3
0 3
0 2
0 9
0 3
.OOE-19
.92E-19
.07E-19
.33E-19
.67E-19
.07E-19
.55E-19
. 12E-19
.76E-20
.92E-20
.96E-20
.61E-20
. 16E-21
.26E-21
.46E-21
.77E-21
. 19E-21
.07E-21
.36E-22
.25E-22
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
195
201
207
213
221
228
236
245
254
264
275
287
300
305
308
311
314
325
340
355
.1 2.
.0 4.
.3 3.
.9 3.
.0 2.
.6 1.
.7 1.
.4 9-
.8 6.
.9 4.
.9 2.
.8 1.
.5 6.
.0 H.
.0 4.
.0 3.
.0 2.
.0 1.
.0 6.
.0 0.
80E-19
61E-19
82E-19
10E-19
46E-19
89E-19
39E-19
93E-20
78E-20
21E-20
46E-20
29E-20
47E-21
93E-21
23E-21
58E-21
99E-21
53E-21
86E-22
OOE-01
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
Photolysis Set = HCHOMR
251.6 3-91E-22 0.480
261.4 5.06E-21 0.498
272.1 1.16E-20 0.457
283.7 2.27E-20 0.325
296.3 2.99E-20 0.270
304.0 6.33E-20 0.247
307.0 2.04E-20 0.251
310.0 1.75E-20 0.260
313.0 1.25E-20 0.273
320.0 1.53E-20 0.394
335.0 2.13E-21 0.754
350.0 1.52E-21 0.338
365.0 0. OOE-01 0.005
254.8 1.18E-21 0.489
264.9 5.55E-21 0.495
275.9 1.60E-20 0.386
287.8 2.13E-20 0.306
300.5 1.52E-20 0.255
305.0 4.67E-20 0.246
308.0 1.41E-20 0.254
311.0 7.28E-21 0.264
314.0 3.92E-20 0.279
325.0 2.11E-20 0.508
340.0 1.63E-20 0.616
355.0 7.17E-21 0.198
258.1 2.81E-21 0.494
268.5 9.35E-21 0.484
279.7 1.58E-20 0.342
292.0 2.26E-20 0.287
303.0 2.28E-20 0.249
306.0 4.50E-20 0.248
309.0 2.96E-20 0.257
312.0 1.34E-20 0.269
316.0 3-95E-20 0.310
330.0 1.92E-20 0.676
345.0 6.67E-21 0.471
360.0 9.07E-23 0.085
(continued)
23
-------
Table 3 (continued) - 6
WL
(nm)
Abs
(cm2)
QY
WL
(nm)
Abs
(cm2)
QY
WL
(nm)
Abs
(cm2)
QY
Photolysis Set = HCHORR
251.6 3.91E-22 0.340
261.1 5.06E-21 0.319
272.1 1.16E-20 0.405
283.7 2.27E-20 0.606
296.3 2.99E-20 0.742
304.0 6.33E-20 0.754
307.0 2.04E-20 0.753
310.0 1.75E-20 0.749
313.0 1.25E-20 0.732
320.0 1.53E-20 0.593
335.0 2.13E-21 0.122
Photolysis Set = ALDR
264.9 2.82E-20 0.340
275.9 4.01E-20 0.492
287.8 4.55E-20 0.531
300.5 3.90E-20 0.420
305.0 3.34E-20 0.351
308.0 3-11E-20 0.304
311.0 2.66E-20 0.255
314.0 2.25E-20 0.204
325.0 1.09E-20 0.050
Photolysis Set = OPR
211.6 3.49E-19 1.000
218.6 2.42E-19 1.000
226.0 1.71E-19 1.000
233.9 1.19E-19 1.000
242.4 8.07E-20 1.000
251.6 5.48E-20 1.000
261.4 3.61E-20 1.000
272.1 2.29E-20 1.000
283.7 1.28E-20 1.000
296.3 6.98E-21 1.000
304.0 4.84E-21 1.000
307.0 4.12E-21 1.000
310.0 3.41E-21 1.000
313.0 2.95E-21 1.000
320.0 1.94E-21 1.000
335.0 8.50E-22 1.000
350.0 3.87E-22 1.000
254.8 1.18E-21 0.323
264.9 5.55E-21 0.333
275.9 1.60E-20 0.456
287.8 2.13E-20 0.680
300.5 1.52E-20 0.750
305.0 4.67E-20 0.755
308.0 1.41E-20 0.752
311.0 7.28E-21 0.745
314.0 3.92E-20 0.723
325.0 2.11E-20 0.458
340.0 1.63E-20 0.003
268.5 3.25E-20 0.369
279.7 4.34E-20 0.557
292.0 4.43E-20 0.502
303.0 3.54E-20 0.383
306.0 3.27E-20 0.335
309.0 2.97E-20 0.288
312.0 2.50E-20 0.238
316.0 2.05E-20 0.169
330.0 6.43E-21 0.006
213.9 3.14E-19 1.000
221.0 2.12E-19 1.000
228.6 1.50E-19 1.000
236.7 1.05E-19 1.000
245.4 7.18E-20 1.000
254.8 4.84E-20 1.000
264.9 3.16E-20 1.000
275.9 1.91E-20 1.000
287.8 1.03E-20 1.000
300.5 5.70E-21 1.000
305.0 4.60E-21 1.000
308.0 3.88E-21 1.000
311.0 3.25E-21 1.000
314.0 2.80E-21 1.000
325.0 1.50E-21 1.000
340.0 6.19E-22 1.000
355.0 2.00E-22 1.000
258.1 2.81E-21 0.317
268.5 9.35E-21 0.365
279.7 1.58E-20 0.522
292.0 2.26E-20 0.724
303.0 2.28E-20 0.753
306.0 4.50E-20 0.754
309.0 2.96E-20 0.751
312.0 1.34E-20 0.740
316.0 3.95E-20 0.690
330.0 1.92E-20 0.305
345.0 6.67E-21 0.000
272.1 3.68E-20 0.421
283.7 4.50E-20 0.552
296.3 4.22E-20 0.463
304.0 3.41E-20 0.367
307.0 3.19E-20 0.319
310.0 2.81E-20 0.272
313.0 2.36E-20 0.221
320.0 1.57E-20 0.106
335.0 2.79E-21 0.000
216.2 2.79E-19 1.000
223.5 1.92E-19 1.000
231.2 1.33E-19 1.000
239-5 9.07E-20 1.000
248.4 6.26E-20 1.000
258.1 4.19E-20 1.000
268.5 2.70E-20 1.000
279.7 1.54E-20 1.000
292.0 8.36E-21 1.000
303.0 5.08E-21 1.000
306.0 4.36E-21 1.000
309.0 3.64E-21 1.000
312.0 3.10E-21 1.000
316.0 2.50E-21 1.000
330.0 1.12E-21 1.000
345.0 5.00E-22 1.000
360.0 O.OOE-01 1.000
(continued)
24
-------
Table 3 (continued) - 7
WL Abs QY
(nm) (cm)
Photolysis Set = PAAR
191.4 8.98E-20 1.000
197.0 1.48E-19 1-000
203.0 1.21E-19 1.000
209-4 9.98E-20 1.000
216.2 8.07E-20 1.000
223.5 6.33E-20 1.000
231.2 4.80E-20 1.000
239.5 3.49E-20 1.000
248.4 2.47E-20 1.000
258.1 1.62E-20 1.000
268.5 9.91E-21 1.000
279-7 5.66E-21 1.000
292.0 2.88E-21 1.000
303.0 1.57E-21 1.000
306.0 1.31E-21 1.000
309.0 1.12E-21 1.000
312.0 9.47E-22 1.000
316.0 7.43E-22 1.000
330.0 3.32E-22 1.000
345.0 1.U1E-22 1.000
Photolysis Set = KETR
279.7 6.00E-20 0.077
292.0 5.10E-20 0.077
303.0 3.00E-20 0.077
306.0 2.30E-20 0.077
309.0 1.70E-20 0.077
312.0 1.20E-20 0.077
316.0 7.80E-21 0.077
330.0 O.OOE-01 0.077
Photolysis Set = GLY1R
233.9 3.40E-21 1.000
242.4 5.73E-21 1.000
251.6 9.05E-21 1.000
261.4 1.62E-20 1.000
272.1 2.48E-20 1.000
283.7 3.18E-20 1.000
296.3 3.37E-20 1.000
304.0 2.90E-20 1.000
307.0 2.72E-20 1.000
WL Abs QY
( nm ) ( cm )
193.2 8.41E-20 1.000
199.0 1.38E-19 1.000
205.1 1.14E-19 1.000
211.6 9.32E-20 1.000
218.6 7.48E-20 1.000
226.0 5.79E-20 1.000
233.9 4.33E-20 1.000
242.4 3.13E-20 1.000
251.6 2.17E-20 1.000
261.4 1.38E-20 1.000
272.1 8.29E-21 1.000
283.7 4.52E-21 1.000
296.3 2.29E-21 1.000
304.0 1.47E-21 1.000
307.0 1.25E-21 1.000
310.0 1.06E-21 1.000
313.0 8.93E-22 1.000
320.0 5.81E-22 1.000
335.0 2.62E-22 1.000
350.0 9.10E-23 1-000
283.7 5.80E-20 0.077
296.3 4.10E-20 0.077
304.0 2.80E-20 0.077
307.0 2.10E-20 0.077
310.0 1.50E-20 0.077
313.0 1.10E-20 0.077
320.0 3.90E-21 0.077
236.7 4.01E-21 1.000
245.4 6.62E-21 1.000
254.8 1.12E-20 1.000
264.9 2.02E-20 1.000
275.9 2.76E-20 1.000
287.8 3.22E-20 1.000
300.5 3.43E-20 1.000
305.0 2.75E-20 1.000
308.0 2.72E-20 1.000
WL Abs QY
(nm) (cm2)
195.1 7.84E-20 1.000
201.0 1.29E-19 1-000
207.3 1.07E-19 1-000
213.9 8.68E-20 1.000
221.0 6.90E-20 1.000
228.6 5.29E-20 1.000
236.7 3.90E-20 1.000
245.4 2.78E-20 1.000
254.8 1.90E-20 1.000
264.9 1.18E-20 1.000
275.9 6.90E-21 1.000
287.8 3.61E-21 1.000
300.5 1.81E-21 1.000
305.0 1.38E-21 1.000
308.0 1.18E-21 1.000
311.0 1.00E-21 1.000
314.0 8.38E-22 1.000
325.0 4.28E-22 1.000
340.0 1.92E-22 1.000
355.0 O.OOE-01 1.000
287.8 5.50E-20 0.077
300.5 3.20E-20 0.077
305.0 2.50E-20 0.077
308.0 1.90E-20 0.077
311.0 1.40E-20 0.077
314.0 1.00E-20 0.077
325.0 1.80E-21 0.077
239.5 4.86E-21 1.000
248.4 7.54E-21 1.000
258.1 1.33E-20 1.000
268.5 2.24E-20 1.000
279.7 2.89E-20 1.000
292.0 3.21E-20 1.000
303.0 3.07E-20 1.000
306.0 2.72E-20 1.000
309.0 2.72E-20 1.000
(continued)
25
-------
Table 3 (continued) - 8
WL Abs QY
(nm) (cnr)
310.0 2.73E-20 1.000
313.0 2.68E-20 1.000
320.0 1.51E-20 1.000
335.0 2.29E-20 1.000
Photolysis Set = GLY2R
355.0 2.86E-22 0.024
370.0 7.64E-21 0.024
385.0 1.66E-20 0.024
400.0 3-36E-20 0.024
415.0 6.43E-20 0.024
430.0 6.77E-20 0.024
445.0 7.15E-20 0.024
460.0 O.OOE-01 0.024
Photolysis Set = ONITR
264.9 2.91E-20 1.000
275.9 2.01E-20 1.000
287.8 9-92E-21 1.000
300.5 4.07E-21 1.000
305.0 2.68E-21 1.000
308.0 1.99E-21 1.000
311.0 1.38E-21 1.000
314.0 9.14E-22 1.000
325.0 1.90E-22 1.000
WL Abs QY
( nm ) ( cm )
311.0 2.79E-20 1.000
314.0 2.49E-20 1.000
325.0 1.27E-20 1.000
340.0 3.58E-21 1.000
360.0 2.08E-21 0.024
375.0 1.07E-20 0.024
390.0 3.03E-20 0.024
405.0 3.66E-20 0.024
420.0 5.46E-20 0.024
435.0 5.99E-20 0.024
450.0 7.30E-20 0.024
268.5 2.66E-20 1.000
279.7 1.64E-20 1.000
292.0 7.49E-21 1.000
303.0 3.28E-21 1.000
306.0 2.44E-21 1.000
309.0 1.76E-21 1.000
312.0 1.22E-21 1.000
316.0 6.87E-22 1.000
330.0 O.OOE-01 1.000
WL Abs QY
(nm) (cm )
312.0 2.83E-20 1.000
316.0 2.12E-20 1.000
330.0 1.42E-20 1.000
345.0 O.OOE-01 1.000
365.0 3.44E-21 0.024
380.0 1.59E-20 0.024
395.0 2.63E-20 0.024
410.0 4.56E-20 0.024
425.0 9.22E-20 0.024
440.0 1.17E-19 0.024
455.0 2.01E-19 0.024
272.1 2.37E-20 1.000
283.7 1.29E-20 1.000
296.3 5.62E-21 1.000
304.0 2.98E-21 1.000
307.0 2.21E-21 1.000
310.0 1.54E-21 1.000
313.0 1.07E-21 1.000
320.0 4.04E-22 1.000
Data Used in the March 1988 Version of the RADM Mechanism, but Not
in the Recommended Versions
Photolysis Set = MGLYR
233.9 1.23E-20 1.000
242.4 1.70E-20 1.000
251.6 1.80E-20 1.000
261.4 1.99E-20 1.000
272.1 2.61E-20 1.000
283.7 2.93E-20 1.000
296.3 2.32E-20 1.000
304.0 1.84E-20 1.000
307.0 1.55E-20 1.000
310.0 1.26E-20 1.000
313.0 1.19E-20 1.000
320.0 6.44E-21 1.000
236.7 1.47E-20 1.000
245.4 1.67E-20 1.000
254.8 1.81E-20 1.000
264.9 2.12E-20 1.000
275.9 2.79E-20 1.000
287.8 2.93E-20 1.000
300.5 2.14E-20 1.000
305.0 1.75E-20 1.000
308.0 1.45E-20 1.000
311.0 1.23E-20 1.000
314.0 1.17E-20 1.000
325.0 4.00E-21 1.000
239.5 1.69E-20
248.4 1.75E-20
258.1 1.89E-20
268.5 2.38E-20
279.7 2.96E-20
292.0 2.71E-20
303.0 1.93E-20
306.0 1.65E-20
309.0 1.35E-20
312.0 1.21E-20
316.0 1.04E-20
330.0 2.00E-21
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
(continued)
26
-------
Table 3 (continued) - 9
WL
(nm)
Abs
(cm2)
QY
WL
(nm)
Abs
(cm2)
QY
WL
(nm)
Abs
(cm2)
QY
335.0 2.50E-22 1.000 340.0 O.OOE-01 1.000 355.0 7.23E-22 0.153
360.0 2.46E-21 0.153 365.0 4.00E-21 0.153 370.0 5.55E-21 0.153
375.0 8.00E-21 0.153 380.0 1.08E-20 0.153 385.0 1.51E-20 0.153
390.0 1.99E-20 0.153 395.0 2.40E-20 0.153 400.0 2.87E-20 0.153
405.0 3.75E-20 0.153 410.0 4.42E-20 0.153 415.0 4.82E-20 0.153
420.0 4.96E-20 0.153 425.0 4.97E-20 0.153 430.0 5.18E-20 0.153
435.0 5.87E-20 0.153 440.0 6.01E-20 0.153 445.0 6.06E-20 0.153
450.0 5.40E-20 0.153 455.0 2.97E-20 0.153 460.0 6.96E-21 0.153
465.0 2.00E-21 0.153 470.0 O.OOE-01 0.153
Photolysis Set = DCBR
186.0 2.59E-21 1.000 187.8 2.72E-21 1.000 189.6 2.85E-21 1.000
191.4 2.73E-21 1.000 193-2 2.51E-21 1.000 195.1 2.44E-21 1.000
197.0 2.46E-21 1.000 199.0 2.46E-21 1.000 201.0 2.82E-21 1.000
203.0 2.92E-21 1.000 205.1 3.53E-21 1.000 207.3 3.74E-21 1.000
209.4 4.10E-21 1.000 211.6 4.36E-21 1.000 213.9 4.45E-21 1.000
216.2 4.60E-21 1.000 218.6 4.17E-21 1.000 221.0 4.54E-21 1.000
223.5 3.70E-21 1.000 226.0 3.89E-21 1.000 228.6 2.69E-21 1.000
231.2 2.76E-21 1.000 233-9 1.71E-21 1.000 236.7 1.62E-21 1.000
239-5 7.85E-22 1.000 242.4 6.12E-22 1.000 245.4 2.88E-22 1.000
248.4 2.13E-22 1.000 251.6 1.40E-22 1.000 254.8 1.49E-22 1.000
258.1 1.60E-22 1.000 261.4 1.79E-22 1.000 264.9 2.18E-22 1.000
268.5 2.72E-22 1.000 272.1 3.27E-22 1.000 275.9 4.10E-22 1.000
279.7 5.26E-22 1.000 283.7 6.27E-22 1.000 287.8 7.62E-22 1.000
292.0 9.39E-22 1.000 296.3 1.09E-21 1.000 300.5 1.26E-21 1.000
303.0 1.51E-21 1.000 304.0 1.56E-21 1.000 305.0 1.55E-21 1.000
306.0 1.53E-21 1.000 307.0 1.61E-21 1.000 308.0 1.65E-21 1.000
309.0 1.81E-21 1.000 310.0 1.83E-21 1.000 311.0 1.91E-21 1.000
312.0 2.03E-21 1.000 313.0 2.01E-21 1.000 314.0 1.99E-21 1.000
316.0 2.20E-21 1.000 320.0 2.50E-21 1.000 325.0 2.81E-21 1.000
330.0 3.04E-21 1.000 335.0 3.40E-21 1.000 340.0 3.84E-21 1.000
345.0 4.05E-21 1.000 350.0 4.47E-21 1.000 355.0 4.88E-21 1.000
360.0 4.94E-21 1.000 365.0 5.29E-21 1.000 370.0 5.40E-21 1.000
375.0 5.63E-21 1.000 380.0 5.61E-21 1.000 385.0 5.72E-21 0.997
390.0 6.00E-21 0.981 395.0 5.69E-21 0.922 400.0 6.18E-21 0.693
405.0 5.49E-21 0.356 410.0 5.98E-21 0.135 415.0 5.77E-21 0.060
420.0 O.OOE-01 0.018
Data Used in the Recommended Versions of the RADM Mechanism, but Not
in the March 1988 Version
Photolysis Set = MGLYRM
231.2 O.OOE-01 1.000 233-9 1.23E-20 1.000 236.7 1.47E-20 1.000
239.5 1.69E-20 1.000 242.4 1.70E-20 1.000 245.4 1.67E-20 1.000
(continued)
27
-------
Table 3 (continued) - 10
WL
(nm)
Abs
(cm2)
QY
WL
(nm)
Abs
(cm2)
QY
WL
(nm)
Abs
(cm2)
QY
248.4 1.75E-20 1.000 251.6 1.80E-20 1.000 254.8 1.81E-20 1.000
258.1 1.89E-20 1.000 261.4 1.99E-20 1.000 264.9 2.12E-20 1.000
268.5 2.38E-20 1.000 272.1 2.61E-20 1.000 275.9 2.79E-20 1.000
279.7 2.96E-20 1.000 283-7 2.93E-20 1.000 287.8 2.93E-20 1.000
292.0 2.71E-20 1.000 296.3 2.32E-20 1.000 300.5 2.14E-20 1.000
303.0 1.93E-20 1.000 304.0 1.84E-20 1.000 305.0 1.75E-20 1.000
306.0 1.65E-20 1.000 307.0 1.55E-20 1.000 308.0 1.45E-20 1.000
309.0 1.35E-20 1.000 310.0 1.26E-20 1.000 311.0 1.23E-20 1.000
312.0 1.21E-20 1.000 313.0 1.19E-20 1.000 314.0 1.17E-20 1.000
316.0 1.04E-20 1.000 320.0 6.44E-21 1.000 325.0 4.00E-21 1.000
330.0 2.00E-21 1.000 335.0 2.50E-22 1.000 340.0 O.OOE-01 1.000
345.0 O.OOE-01 0.103 350.0 O.OOE-01 0.103 355.0 7.23E-22 0.103
360.0 2.46E-21 0.103 365.0 4.00E-21 0.103 370.0 5.55E-21 0.103
375.0 8.00E-21 0.103 380.0 1.08E-20 0.103 385.0 1.51E-20 0.103
390.0 1.99E-20 0.103 395.0 2.40E-20 0.103 400.0 2.87E-20 0.103
405.0 3.75E-20 0.103 410.0 4.42E-20 0.103 415.0 4.82E-20 0.103
420.0 4.96E-20 0.103 425.0 4.97E-20 0.103 430.0 5.18E-20 0.103
435.0 5.87E-20 0.103 440.0 6.01E-20 0.103 445.0 6.06E-20 0.103
450.0 5.40E-20 0.103 455.0 2.97E-20 0.103 460.0 6.96E-21 0.103
465.0 2.00E-21 0.103 470.0 O.OOE-01 0.103
Photolysis Set = DCBRM
186.0 7.90E-20 1.000 187.8 7.90E-20 1.000 189.6 7.90E-20 1.000
191.4 7.90E-20 1.000 193.2 7.90E-20 1.000 195.1 7.90E-20 1.000
197.0 7.90E-20 1.000 199-0 7.90E-20 1.000 201.0 7.90E-20 1.000
203.0 7.90E-20 1.000 205.1 7.90E-20 1.000 207.3 7.90E-20 1.000
209.4 7.90E-20 1.000 211.6 7.90E-20 1.000 213-9 7.90E-20 1.000
216.2 7.90E-20 1.000 218.6 7.90E-20 1.000 221.0 7.90E-20 1.000
223-5 7.90E-20 1.000 226.0 7.90E-20 1.000 228.6 7.90E-20 1.000
231.2 7.90E-20 1.000 233.9 7.90E-20 1.000 236.7 7.90E-20 1.000
239.5 7.90E-20 1.000 242.4 7.90E-20 1.000 245.4 7.90E-20 1.000
248.4 7.90E-20 1.000 251.6 7.90E-20 1.000 254.8 7.90E-20 1.000
258.1 7.90E-20 1.000 261.4 7.90E-20 1.000 264.9 7.90E-20 1.000
268.5 7.90E-20 1.000 272.1 7.90E-20 1.000 275.9 7.90E-20 1.000
279.7 7.90E-20 1.000 283-7 7.90E-20 1.000 287.8 7.90E-20 1.000
292.0 7-90E-20 1.000 296.3 7.90E-20 1.000 300.5 7-90E-20 1.000
303.0 7.90E-20 1.000 304.0 7.90E-20 1.000 305.0 7.90E-20 1.000
306.0 7.90E-20 1.000 307.0 7.90E-20 1.000 308.0 7.90E-20 1.000
309-0 7.90E-20 1.000 310.0 7.90E-20 1.000 311-0 7.90E-20 1.000
312.0 7.90E-20 1.000 313.0 7.90E-20 1.000 314.0 7.90E-20 1.000
316.0 7.90E-20 1.000 320.0 7.90E-20 1.000 325.0 7.90E-20 1.000
330.0 7.90E-20 1.000 335.0 7.90E-20 1.000 340.0 7.90E-20 1.000
345.0 7.90E-20 1.000 350.0 7.90E-20 1.000 355.0 7.90E-20 0.500
360.0 7.90E-20 0.000 365.0 O.OOE-01 0.000
28
-------
2.2 Recommended Modifications to the RADM Mechanism
The results of this evaluation study indicated that for the most part
the March 1988 version of the RADM gas-phase chemical mechanism is
chemically reasonable and performs as well as can reasonably be expected
in simulating most of the available chamber data. However, there were
aspects of the mechanism which we recommended be modified prior to its use
in the RADM-II model. The results of the chamber simulations, discussed
in Section 5, show that the performance of the RADM gas-phase mechanism in
fitting results of aromatic-NOx-air chamber experiments was not quite as
good as should be expected given the performance of other current gas-
phase chemical mechanisms. The test calculations described in Section 6.3
show that the representation of the higher alkanes could be condensed
without significantly affecting model predictions. In addition, there
were several other areas where it could be argued that minor modifications
to the current mechanism might improve its chemical reasonableness. The
recommended modifications to the RADM gas-phase chemical mechanism are
described and justified below.
2.2.1 Recommended Modifications to the Aromatic Mechanism
There are major gaps in our understanding of the atmospheric
reactions of aromatics, and the available laboratory data are not suffi-
cient to serve as a basis for deriving models for the reactions of these
compounds. Thus, we have no choice but to use simulations of environ-
mental chamber experiments as essentially the only means to Judge other-
wise chemically reasonable alternative assumptions concerning the unknown
portions of the aromatics mechanisms. The performance of the RADM mech-
anism in simulating results of toluene-NOx-air and xylene-NOx-air environ-
mental chamber runs, discussed in Section 5, indicated that there was room
for improvement in this mechanism. The RADM mechanism was found to have
considerably more scatter in the simulations of maximum ozone yields and
rates of NO oxidation and 0^ formation than the SAPRC mechanism. By
comparison, Jeffries (1989b) found that the Carbon Bond-IV mechanism a
more condensed mechanism than RADM performed comparably to the SAPRC
mechanism in simulating the aromatic runs, being somewhat better in
simulating the UNC toluene runs, and somewhat worse in simulating UNC
xylene runs. Thus by the criterion of performance in simulating the
chamber data, the model used in the RADM mechanism to represent the
29
-------
unknown portions of the toluene and xylene mechanisms which is based
largely on an earlier published aromatics mechanism (Lurmann et al., 1986)
must be judged to be less reliable than the models used in the current
SAPRC or Carbon Bond mechanisms. Since the mechanisms for toluene and
xylene serve as the basis for the representation of all the aromatics
hydrocarbons, it is clearly important that the RADM-II model incorporate
mechanisms for these compounds that are as reliable as possible. For this
reason, a major recommendation resulting from this study was that the
aromatics photooxidation mechanism be modified to improve its performance
in simulations of the toluene and xylene chamber data.
To address this concern, we investigated how the RADM mechanism might
be modified to improve its performance in simulating the results of these
chamber experiments. The modifications examined involved making the RADM
aromatics mechanism more like the aromatics mechanism used in the current
SAPRC mechanism (Carter et al., 1988). This approach was employed because
(1) independent development of an entirely new aromatics mechanism was
beyond the scope of this project; (2) we are familiar with the performance
and implementation of the SAPRC aromatics model and know how to adjust it
to optimize its ability to fit the chamber data; and (3) although the
aromatics model in Carbon Bond IV also performs as satisfactorily in
simulating chamber data, its use of lumped structure groups makes it less
straightforward to implement in a lumped molecule mechanism such as
RADM. In addition, we are less familiar with how to optimize the ability
of the Carbon Bond mechanism to fit chamber data. However, it must be
recognized that the SAPRC aromatics mechanism uses a highly idealized and
simplified representation of the unknown aspects of aromatic reactions,
and no claim is made that it is necessarily any more chemically reasonable
than the aromatics representation used in any other current mechanism,
including that in the RADM mechanism itself.
Two levels of modifications of the aromatic mechanism were examined
in the course of this study: one in which the only changes involved
parameter values, and the other where more extensive changes were made.
The latter is the version of the modified aromatics mechanism which is
designated as RADM-M and is listed in Section 2.3- The former is referred
to as the "modified parameter" aromatics mechanism, which was recommended
for implementation if more extensive changes in the mechanism were not
30
-------
feasible. It is designated as the RADM-P mechanism and is listed in
Section 2.4. In the subsequent discussion, the modifications discussed
for toluene refer to the reactions of the RADM model species TOL, and the
modifications discussed for xylenes refer to reactions of the model
species XYL. In both cases, the following parameter changes were
recommended:
(1) It was recommended that the yields of methyl glyoxal (MGLY) and
(for the modified parameter mechanism) glyoxal (GLY) be reduced to be more
consistent with experimental yields of these products from toluene and m-
xylene (Atkinson, 1989, and references therein). In addition, the yield
of cresols (CSL) from toluene was increased slightly to be more consistent
with the results of Gery et al. (1985) and Atkinson et al. (1989); this
also resulted in somewhat improved fits of model simulations of maximum
ozone yields in toluene-NOx-air experiments. The specific changes in
yields of these products are as follows:
RADM RADM-M and RADM-P
MGLY yield from TOLP 0.72*4 0.17
MGLY yield from XYLP 1.0 0.45
GLY yield form TOLP 0.181 0.16
CSL yield form TOL 0.16 0.25
(2) Although the available documentation of the derivation of
photolysis rates to be used in the RADM mechanism (NCAR, 1987) states that
the methylglyoxal quantum yields in the higher wavelength band are based
on the data of Plum et al. (1983), the RADM mechanism actually uses higher
quantum yield than indicated by those data. In particular, as shown in
Table 3 (see photolysis file "MGLYR"), the RADM mechanism uses a wave-
length-independent quantum yield of 0.153 for wavelengths above 350 run,
while the data of Plum et al. (1983) indicate that the quantum yield
should be 0.107. We therefore recommended that the methylglyoxal quantum
yields for these wavelengths be decreased to the lower value which is more
consistent with the data of Plum et al. (1983). The modified sets of
quantum yields recommended for methylglyoxal are listed in Table 3
(photolysis file "MGLYRM").
(3) The RADM mechanism uses the model species DCB to represent the
uncharacterized aromatic ring fragmentation products. The photolysis
31
-------
rates of these compounds, and the wavelengths where they photolyze, are
unknown. In the RADM mechanism, the species DCB photolyzes at the same
wavelengths as NC^, but at a rate which is 10 times slower. The ratio to
NCs is presumably based on results of simulations of chamber runs.
However, in the development of the SAPRC mechanism, it was found that
better performance in simulating the chamber data, especially runs carried
out using a blacklight light source, could be obtained if it was assumed
that this species photolyzes more rapidly, and photolyzes primarily at
much lower wavelengths than those responsible for the photolysis of NO^.
Therefore, we recommended that the absorption coefficient, quantum yield
products used in the SAPRC mechanism for the unknown aromatic fragmenta-
tion species be used in the RADM mechanism for DCB. This consists of
assuming the absorption coefficient x quantum yield product is constant at
?o ? i
7.9x10 cm molecule , base e, at wavelengths of 350 nm or below, and
decreases linearly to zero at 360 nm. This is essentially arbitrary, but
allows good fits to aromatic-NO -air experiments carried out using a
A
variety of light sources. The modified absorption coefficient x quantum
yield products are listed in Table 3 (photolysis set "DCBRM").
(H) The yields of the DCB species from toluene and xylene were
adjusted to optimize the fits of model simulations to results of selected
SAPRC toluene-NOx-air and m-xylene-NOx-air experiments The approach
employed, and the specific experiments whose results were utilized in the
optimizations is the same as those used to derive the uncharacterized
aromatic fragmentation product yields in the current SAPRC mechanism
(Carter, 1988). The best fit DCB yields depended on the other changes
made to the aromatic mechanism, which differed depending on whether
changes other than just modifying parameter values were made. Thus, the
optimum DCB yields for the modified parameter (RADM-P) mechanism are
slightly different from those for the recommended (RADM-M) mechanism where
other changes were made. These are indicated below:
RADM RADM-P RADM-M
DCB yield from TOLP 0.905 0.70 0.68
DCB yield from XYLP 1.0 0.806 0.79
32
-------
(5) The RADM mechanism uses an OH + XYL rate constant which was
derived based on an analysis of total U. S. emissions data which we
carried out during the initial review of the 1987 version of the RADM
mechanism. However, since the time this initial analyses was carried out,
the emissions assignments and rate constant estimates have been updated as
a result of the work carried out for the California ARB (Carter, 1988),
and there have also been some changes to methods used for aggregating
emissions input into RADM. (See Section 3 of this report for a discussion
of our analyses for aggregating emissions into RADM.) Therefore, an up-
dated analysis of the emissions data was carried out for the purpose of
developing our final recommendations in this regard, and the results are
documented in Section 3.3. This updated analysis resulted in slight
changes in the recommended OH radical rate constants for XYL and some of
the other lumped species. In the recommended mechanisms, the A factor
for the OH + XYL reaction was changed from 2.1x10~11 to 1.89x10"11 cm^
molecule" sec, to correspond better to the revised emissions analysis.
This value was used in calculating the optimum DCB yields from XYLP,
discussed above.
The modified product yields recommended for toluene and xylene given
above actually refer to the yields of these products formed in the reac-
tion of their corresponding peroxy radicals (TOLP and XYLP, respectively)
with NO. The RADM mechanism also has these radical species reacting with
H02, methyl peroxy (M02), and acetyl peroxy (AC03) radicals. The
recommendations do not affect the species representing the products formed
in the HOp reactions (which is the lumped higher hydroperoxide species
OP2), but do affect the species formed in the other two reactions. For
these reactions, the recommended product yields were derived based on
assuming that the set of organic products formed from TOLP and XYLP are
the same as those formed in the NO reaction, with the radical and M02 or
AC03 product yields being determined based on assuming these reactions
occur 50^ via RO + RO + 02 formation, and 50% via ROH formation. This may
be somewhat arbitrary, but represents a systematic and self-consistent
method for deriving product yields for the many peroxy radical species in
this mechanism and, in view of our lack of knowledge of what the actual
reactions and species are, appears to be as good as any other set of
assumptions.
33
-------
The above are the modifications recommended for the aromatic mechan-
ism that involve only changes of parameter values, and thus are the
modifications implemented in the RADM-P version of the mechanism.
However, if more extensive modifications could be made, then it was
recommended that the representation of the aromatic fragmentation products
be simplified. Chamber simulations carried out in this study indicate
that a simplified DCB mechanism, where the species TPAN and TC03 are
replaced by PAN and AC03, respectively, and where GLY is eliminated,
performs equally well, or even slightly better, in simulating the results
of the chamber experiments, yet requires three fewer species. In view of
the fact that the more complex "DCB" mechanism used in the evaluated RADM
mechanism is based on theories of aromatic reactions (e.g., Atkinson et
al. 1980) which are now known to be incorrect or at best incomplete (e.g.,
Atkinson, 1988 and references therein), that mechanism has no claim to
greater chemical reasonableness than the more simplified representation
that is recommended. The recommended simplifications are implemented in
the RADM-M mechanism, and the specific changes involved are discussed
below. (All these simplifications were implemented in the RADM-M mechan-
ism before the optimum DCB yields were calculated.)
(1) Glyoxal (GLY) makes a relatively small contribution to the reac-
tivity of aromatics compared to the much more reactive products methyl
glyoxal (MGLY) and DCB. Thus, it is recommended that GLY be removed from
the RADM mechanism, where it is formed only from TOL, and its contribution
to reactivity be represented by having it lumped with the other aromatic
fragmentation products which are represented by DCB. (The optimization of
the yield of DCB from TOL in a mechanism where GLY is removed will auto-
matically take its contribution to reactivity into account). Since
glyoxal in itself is not (to our knowledge) of special significance to
acid deposition, there is no reason to represent it explicitly in regional
acid deposition models if its contribution to reactivity can be satisfac-
torily represented in other ways. If it is important, its formation from
xylenes and other aromatics should also be included. We assume that this
is not the case.
(2) The products of the reactions of DCB are assumed to be similar
to those for the reactions of methyl glyoxal, and are represented as
follows:
-------
DCBX * HV = H02 + CO * AC03
DCBX + HO = AC03
(DCB is renamed DCBX in the RADM-M mechanism to indicate that it reacts
differently.) The photolysis reaction occurs with the same rate and spec-
tral response as used in the modified parameter mechanism, which in turn
is the same as that used in the SAPRC mechanism for unknown aromatic
fragmentation products (as discussed above). The OH radical reaction is
assumed to have the same rate constant as that for methyl glyoxal. The
NO? radical reaction is unlikely to be a significant loss route for this
species (or for NOo), and is ignored. Although this simpler representa-
tion of DCB reactions is not necessarily any more chemically reasonable
than that used in RADM or RADM-P, it has the significant advantage that it
results in the removal of TC03 and TPAN from the mechanism since the DCB
reactions are the only source for these species in the evaluated RADM
mechanism.
The removal of TPAN from the mechanism means that PAN is used to
represent all types of PAN analogues that might be formed in the reactions
of unidentified aromatic fragmentation products. This is not inconsistent
with the general concept of the level of detail intended for this mech-
anism, since PAN is not represented explicitly in any case. In the RADM
mechanism, the model species PAN represnts not only PAN, but PPN and other
higher PAN analogues formed in non-aromatic systems. Thus, in the
recommended modification for RADM-M, the higher PAN analogues (if any)
formed from unknown aromatic fragmentation products are in effect lumped
with the other higher PAN analogues. We see no a-priori reason why it is
necessary to have separate representations of the higher PAN analogues
formed from aroraatics when this is not the case for those formed from
alkanes, alkenes, and other species.
Comparisons of the performance of the RADM, the modified parameter
RADM-P and the recommended RADM-M mechanisms in simulating the maximum
ozone yields and rates of NO oxidation and ozone formation in the
aromatic-NO-air experiments is shown in Figure 1. (See Section 4 for a
A
discussion of how the simulations of the environmental chamber experiments
were carried out, and see Section 5 for a more extensive discussion of the
performance of the RADM and the RADM-M mechanisms in simulating these
35
-------
RADM
RADM-P
RADM-M
Fits to Maximum Ozone
[Calculated - Experimental
(ppm)]
Model
<
0.33 -
0.27 -
0.21 -
0.15 -
0.09 -
0.03 -
0.03 -
0.09 -
0.15 -
0.21 -
0.27 -
>
Model
Low
-0.33
-0.27
-0.21
-0.15
-0.09
-0.03
0.03
0.09
0.15
0.21
0.27
0.33
0.33
High
:5
:2
: 14446
:333
:3
:3
:24
: 111111111446
:116
: 133466
:33
:3
:1246
:11111
:11366
:13334
Fits to Average d([Oo]-[NO])/dt
[(Calculated - Experimental)/Calculated]
Model Slow
<
-1.00 -
-0.82 -
-0.64 -
-0.45 -
-0.27 -
-0.09 -
0.09 -
0.27 -
0.45 -
0.64 -
0.82 -
>
Model
-1.00
-0.82
-0.64
-0.45
-0.27
-0.09
0.09
0.27
0.45
0.64
0.82
1.00
1.00
Fast
; ;
J J
* J
:125 :
* 1 1
* I I
:1144 :133
--: 13336 -: 112334456
:1 1111 13446 :1 11 11366
:1366 :1446
: :1
:1 :
1
* 1 *
11
133
111112334456
11366
1446
1
1
Figure 1. Comparison of Fits of RADM and Modified RADM Simulations
to Results of Toluene and Xylene-N0v-Air Chamber Experiments.
A
Symbols:
1: EC Toluene
2: ITC Toluene
3: UNC Toluene
4: EC m-Xylene
5: ITC m-Xylene
6: UNC o-Xylene
Mechanisms:
RADM
RADM-P
RADM-M
March 1988 Mechanism
Recommended Parameter-only
Modifications
Recommended Modifications
36
-------
experiments). Both modified versions of RADM perform better than the
current version in simulating these results, with fewer cases of large
discrepancies and more cases of good fits being obtained. Despite its
more condensed DCB mechanism, the more extensively modified RADM-M
mechanism seems to be slightly better in its overall distribution of
discrepancies than the version where only the parameters are modified,
though the differences are probably not significant. It is significant,
however, that the version of the mechanism with the more detailed (though
not necessarily more accurate) representation of DCB mechanism performs no
better in simulating the results of these experiments than does the
version with three fewer species.
2.2.2 Condensed Higher Alkane Representation
One area where the RADM mechanism is more detailed than the
current ADOM (Lurmann and Karamchandani, 1987), Carbon Bond (Gery et al.,
1988), and other lumped . mechanisms used in regional models is its
representation of the reactions of the higher (C^) alkanes. RADM uses
three lumped species to represent the C?+ alkanes (plus one additional
species each to represent their peroxy radical intermediates), while both
ADOM and Carbon Bond use only one lumped species for these compounds.
Test calculations were carried out as part of this study to determine
whether a moderate level of condensation of the RADM alkane mechanism,
where the two higher alkane model species HC5 and HC8 were lumped together
and represented by a single species, would have a significant effect on
model predictions relative to the evaluated version. The results, given
in detail in Section 6.4, indicate that using this condensation has
insignificant effects on simulations of species of interest in regional
modeling simulations for the wide range of chemical conditions examined.
Because of this, we implemented this condensation in the RADM-M mechanism.
The OH radical rate constant and product yield parameters recommended
for the new lumped higher alkane species were derived in a manner
analogous to the derivation of the recommended parameters for the lumped
higher alkane species in the current RADM mechanism, as discussed in
Section 3.3.1 of this report. Specifically, values of average OH radical
rate constants and product coefficients calculated for the aggregated
emissions groups represented by HC5 and HC8 in the evaluated mechanism
were further averaged, using reactivity weighing and relative molar
37
-------
emissions in the NAPAP total U.S. anthropogenic emissions inventory, to
determine values for these parameters which are appropriate for this
lumped HC5+HC8 alkane model species. The averages of the parameters, and
the contributions of the various emissions groups to the values derived,
are included with the tabulations given for the other RADM model species
in Section 3.3.1- Since the average number of carbons calculated for this
lumped species based on the NAPAP inventory is approximately 6.3, this new
lumped model species is designated "HC6," to be consistent with the
nomenclature used in the RADM mechanism. The listing of the RADM-M
mechanism shows the reactions recommended for this model species, which
replaces the model species HC5 and HC8 in the RADM and RADM-P mechanisms.
(Note that the specific kinetic and mechanistic parameters recommend-
ed for the new lumped "HC6" species based in the analysis in Section 3.3.1
are slightly different from those used in the sensitivity tests discussed
in Section 6.4. This is because the parameters used in the sensitivity
tests were derived based on those used for HC5 and HC8 in the RADM mechan-
ism, for more straightforward comparison with that mechanism. As discussed
below in Section 2.2.5, the mechanistic parameters for HC5 and HC8 in the
evaluated mechanism are slightly different from those which are currently
recommended and implemented in RADM-P.)
It is possible that it may not be much worse an approximation to lump
HC3 with the higher alkane classes, as is done in the ADOM and (in effect)
the Carbon Bond mechanisms. Test calculations carried out using the most
recent SAPRC mechanism indicate that much more severe condensations of the
alkane mechanisms can be employed without significantly affecting model
predictions, provided the parameters for the lumped species are adjusted
appropriately based on the mixture being represented. However, more
severe approximations for alkane representation in RADM were not examined,
and are not currently recommended.
2.2.3 Modifications to the XN02 Reactions
The species "XN02" is used in the RADM mechanism to represent
the additional NOX consumption caused by reactions of NO^ radicals with
dinitrophenols. As such, it actually represents a phenoxy-radical-like
species. However, in the RADM mechanism, this species is represented as
forming organic nitrates (ONIT) when it reacts with NOg, forming organic
hydroperoxides (OP2) when it reacts with H02, and reacting with M02 and
38
-------
AC03 as if it were a peroxy radical. We believe that it is more chemical-
ly reasonable not to have it form ONIT or OP2 in its respective reactions
with NC>2 and H02, and that it is equally chemically reasonable to use a
simpler representation of its loss in the absence of N02 or HOp. There-
fore, modifications in this aspect of the mechanism are recommended for
both improved chemical reasonableness and to simplify the mechanism.
These modifications are implemented in the RADM-M mechanism, but could not
be implemented into RADM-P.
The reaction of XN02 with N02 actually represents the reaction of
nitro-substituted phenoxy radicals with N02 to form dinitrophenols. These
species probably condense into the aerosol phase and do not react; but if
they react at all, it is estimated that the most likely process would be
reaction with NO^ to consume even more NOX. However, the RADM mechanism
in effect represents dinitrophenols as alkyl nitrates (ONIT), whose subse-
quent reactions tend to re-introduce NOV into the system. For this
A
reason, it is considered to be more chemically reasonable for the nitrogen
in the XN02 + N02 reaction to be "lost", rather than for it to re-appear
as ONIT. Therefore, we recommend that the XN02 + N02 reaction have no
products, or, if there is a need to keep track of N-balance in the RADM
simulations, its products be represented by some unreactive N-containing
"counter" species.
The reaction of XN02 with H02 represents a reaction of H02 with some
type of phenoxy radical. The most likely organic product of such a reac-
tion would be a phenol or nitrophenol, and not a hydroperoxide. There-
fore, the use of OP2 as a product in this reaction is probably inappro-
priate. Since hydroperoxides are assumed to be involved in formation of
acid deposition, we recommend that this source of OP2 be removed, and that
the XN02 + H02 reaction also have no reactive products.
The reactions used to represent the removal of XN02 in those cases
where it is formed in the absence of significant amounts of H02 and N02
are usually not important, since usually removal by N02 dominates, and the
H02 reaction is probably adequate to represent its removal in most other
cases. The use of peroxy-like reactions for this purpose is not
necessarily inappropriate, but is essentially arbitrary. However, this
representation introduces unnecessary complexity into the mechanism,
probably making it more difficult to implement in "solvers" where the
39
-------
steady state approximation is used for XN02. To simplify the mechanism
and make it easier to implement, we recommend using a simple first-order
loss process to represent the consumption of XN02 in cases where levels of
both H02 and NOp are low. The rate constant we chose for this process in
RADM-M (2.0x10"^ sec"1) is essentially arbitrary, and corresponds to the
loss rate by reaction with 1 ppt of NC^, or approximately 2 ppt of HC^.
Test calculations show that this simpler representation has essentially no
effects on model predictions representative of conditions likely to occur
in regional model simulations.
2.2,4 Omitted OLN Reactions
The RADM mechanism includes the reactions of M02 and AC03 with
all the organic peroxy radical species in the mechanism except for OLN,
the peroxy radical which is formed when NO^ reacts with alkenes. This
probably was an oversight, since there is no reason why OLN should be
treated differently than the other organic peroxy radical species in this
regard. Because under certain conditions at nighttime OLN can be the
dominant peroxy radical, it is advisable that the OLN + OLN reaction be
included in the mechanism, since that reaction might be its major sink
under those conditions. Since the RADM development team were able to
readily add these reactions to the mechanism being implemented into RADM
(Stockwell, private communication, 1989), these changes are incorporated
in the RADM-P as well as the RADM-M recommended modified mechanisms. The
rate constants and mechanisms used for these reactions are those of
Stockwell (private communication, 1988) derived for the "detailed radical"
version of the RADM mechanism, given in Section 6.2.
2.2.5 Parameter Modifications Resulting from Updated Emissions
Assignments
As indicated above and discussed in more detail in Section
3.3, kinetic, and in some cases product, yield parameters for a number of
lumped species in the RADM mechanism art derived based on an analysis of
the species they represent in the NAPAP total U.S. emissions inventory.
The parameters used in the evaluated RADM mechanism for a number of these
species are based on recommendations we made in this regard as part of our
initial review of the 1987 version of the mechanism. However, an updated
analysis of these data, discussed in Section 3.3 of this report, resulted
in slight changes being recommended for several of these parameters. The
40
-------
specific changes which were recommended are given in the appropriate
tabulations in Section 3.3, and are incorporated into both the recommended
modified RADM-M and RADM-P mechanisms.
The recommended product yields given in the tables in Section 3.3
refer to yields under conditions where the reactions of peroxy radicals
with NO dominate. The RADM mechanism also has peroxy radicals reacting
with H02> M02, and AC03. The modified recommended product yields do not
affect the products of the HC^ reactions, since these are represented by
separate hydroperoxide species. However, they do affect the products
formed in the M02 or AC03 reactions. The products in those reactions were
derived based on assuming that the products from M02 and AC03 are the same
as in the RADM mechanism, and that the oxygenated products formed from the
lumped peroxy radicals (e.g., HC3P, etc.) are the same as those formed in
the NO reaction, excluding ONIT. In all cases, the radical yields were
derived based on assuming the reactions occur 50% via RO + RO + Oo
formation, and 50% via ROH formation. This is the same approach as
discussed above for TOLP and XYLP in the modified aromatics mechanisms and
is in line with the representation for most of these reactions in other
current mechanisms.
2.3 The Recommended Modified Mechanism (RADM-M)
The RADM-M mechanism is the version of the RADM mechanism which
incorporates all the modifications recommended as a result of this study.
It has the same inorganic reaction set and (except for the aromatic
product species DCB and its PAN analogue and the rate of methyl glyoxal
photolysis) the same mechanisms for reactive product species as employed
in the RADM mechanism. Except for the lumping together of the two higher
alkane classes and slight simplifications in mechanisms for some aromatic
products, this mechanism represents the same level of condensation and
chemical detail as employed in the RADM mechanism. However, results of
test calculations representing the range of chemical conditions likely to
be encountered in regional model simulations, given in Section 6.5 of this
report, indicate that there are cases where these mechanisms can give
significantly different predictions. These differences are attributed
primarily to changes in the aromatics mechanism that were made to improve
its performance in simulating results of chamber data. Other test
-------
calculations suggest that the other modifications, which are recommended
primarily to yield a mechanism that is somewhat more compact and (in the
case of reactions involving XN02) somewhat more chemically reasonable,
should have relatively little effect on predictions under most conditions.
The organic model species used in the RADM-M mechanism are listed in
Table 4, and their reactions and thermal rate parameters are listed in
Table 5. The inorganic species and their reactions and rate parameters
are not included in these tables because they are the same as those listed
for the RADM mechanism in Tables 1 and 2, respectively. The absorption
coefficients and quantum yields used for the photolysis reactions in this
mechanism are given in Table 3-
2.4 The Modified Parameter Mechanism (RADM-P)
The RADM-P mechanism is the version of the RADM mechanism which
incorporates all the modifications recommended as a result of this study
that can be implemented by changing only kinetic or product yield
values. Because of practical difficulties involved in making the more
extensive changes required to implement RADM-M, this is the version of the
mechanism which is being implemented into the RADM-II model (Stockwell,
private communication, 1989). This mechanism has the same number of
species and (except for the three OLN reactions which were added) the same
number of reactions as the March 1988 RADM mechanism, and thus represents
almost exactly the same level of chemical detail and condensation. Its
major modifications are the changes in parameters involved in the aromatic
mechanisms which were made to improve its performance in simulating the
chamber data. As with the RADM-M mechanism, test calculations discussed
in Section 6.5 indicate that these changes result in cases where this
mechanism can yield significantly different predictions in simulations
representing conditions of regional model simulations. The other modifi-
cations concern the addition of the usually relatively unimportant OLN
reactions and the minor changes in parameters for lumped species based on
the updated analysis of the emissions data.
-------
Table M. Organic Model Species Employed in the Recommended Modified
Version of the RADM Gas-Phase Chemical Mechanism (RADM-M)
Name Description
Organic Hydrocarbons
CH4 Methane
ETH Ethane
HC3 C3+ Alkanes with kOH < 5 x 103 (a)
HC6 Alkanes with kOH > 0.5 x 1 2 x 10
Organic Oxidation Products
HCHO Formaldehyde
ALD Lumped Higher Aldehydes
KET Ketones
MGLY Methyl Glyoxal
DCBX Uncharacterized Reactive Aromatic Fragmentation Products
CSL Phenols and Cresols
PAN Acyl Peroxy Nitrates
ONIT Other Organic Nitrates (primarily alkyl nitrates)
OP1 Methyl Hydroperoxide
OP2 Higher Organic Hydroperoxides
ORA1 Formic Acid
ORA2 Higher Organic Acids (other than peroxy acetic acid)
PAA Peroxy Acetic Acid
Organic Radicals
M02 Methyl Peroxy Radicals
ETHP Ethyl Peroxy Radicals
HC3P Peroxy Radicals from OH + HC3
HC6P Peroxy Radicals from OH + HC6
OL2P Peroxy Radicals from OH + Ethene
OLTP Peroxy Radicals from OH + Isoprene and Terminal Alkenes
OLIP Peroxy Radicals from OH * Internal Alkenes
KETP Peroxy Radicals from OH + Ketones
TOLP Peroxy Radicals from OH + Benzene and Monoalkyl Benzenes
XYLP Peroxy Radicals from OH -»- Xylenes and More Reactive Aromatics
OLN Peroxy Radicals from NOo + Alkenes
(continued)
-------
Table l» (continued) - 2
Name Description
X02 Chemical Operator Used to Account for NO-to-NC^ conversions
due to Secondarily Formed Organic Peroxy Radicals
AC03 Acyl Peroxy Radicals
XN02 Chemical Operator Used to Account for NOX Loss Due to
Reactions of NO? with Phenols and Cresols.
(a) kOH is the OH radical rate constant at 300 K in ppm min .
-------
Table 5. Listing of the Organic Reactions in the Recommended Modified
Version of the RADM Gas-Phase Chemical Mechanism (RADM-M)
Rxn.
Label
(a)
010
011
012
013
014
015
016
019
M01
021
Kinetic
k(300)
(Phot.
(Phot.
(Phot.
(Phot.
( Phot ,
(Phot,
(Phot,
(Phot,
(Phot
(Phot
Parameters (b)
A
Ea
B
, Set = HCHOMR )
. Set = HCHORR )
. Set = ALDR )
. Set = OPR )
. Set = OPR )
. Set = PAAR )
. Set = KETR )
. Set = MGLYRM )
. Set = DCBRM )
. Set = ONITR )
Reactions (b)
Organic Photolysis Reactions
HCHO + HV = H2 * CO
HCHO + HV = H02 * H02 + CO
ALD + HV = M02 + H02 * CO
OP1 * HV = HCHO + H02 + HO
OP2 + HV = ALD t- H02 * HO
PAA * HV = M02 + C02 + HO
KET -i- HV = AC03 f ETHP
MGLY + HV = AC03 + H02 + CO
DCBX + HV = H02 + CO + AC03
ONIT + HV = #.20 ALD * #.80 KET +
H02 + N02
Organic + OH Reactions
051
052
053
M02
056
057
058
074
059
060
063
064
065
067
M03
061
062
069A
069B
070A
070B
1
4
3
1
1
4
9
1
9
4
1
2
1
2
2
5
5
7
7
7
7
.29E+01
.11E+02
.86Ef03
.09E+04
.24E+04
. 19E+04
.79E+04
.46E+05
.02E+03
.09E+04
. 32E+04
.37E+04
.47E+03
. 50E+04
.50E+04
.87E+04
.28E+04
.34E+03
.34E+03
.34E+03
.34E+03
9.
1.
2.
3.
3.
7.
1.
3.
23E+02
81E+03
33E+04
86E+04
16E+03
81E+03
57E+04
74E+04
3.08E+03
2.77E+04
1.
.32E+04
1.01E+04
1.76E+04
2.50E+04
2.50E+04
5.87E+04
5
7
7
7
7
.28E+04
.34E+03
.34E+03
.34E+03
.34E+03
2.54
0.88
1.07
0.75
-0.82
-1.00
-1.09
-0.81
-0.64
-0.23
0.00
-0.51
1.48
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
1
1
-1
-1
-1
-1
-1
-1
-1
-1
-1
-1
-1
-1
-1
-1
-1
-1
-1
-1
-1
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
CH4
ETH
HC3
#,
#
HC6
OL2
OLT
OLI
ISO
TOL
#
XYL
#
+ HO =
+ HO =
+ HO =
M02 H
ETHP
#.83
i- H20
+ H20
HC3P -i- #.17 H02 +
.009 HCHO + #.075 ALD +
.025 KET + H20
+ HO =
+ HO =
+ HO =
+ HO =
+ HO =
+ HO =
.24 H02
+ HO =
.17 H02
HCHO + HO
ALD
KET
+ HO =
+ HO =
MGLY + HO
DCBX + HO
CSL
#
CSL
OP1
HO
OP2
HO
+ HO =
.9 AC03
+ HO =
+ HO =
+ OP1 =
+ HO =
+ OP2 =
HC6P
OL2P
OLTP
OLIP
OLTP
#.76
#.83
= H02
AC03
KETP
+ #.51 X02 + H20
TOLP + #.24 CSL +
XYLP + #.17 CSL +
+ CO + H20
+ H20
+ H20
= AC03 + CO + H20
= AC03
#.1
CSL
M02
HO +
HC3P
HO *
H02 +1.9 X02 +
HCHO
ALD
(continued)
45
-------
Table 5 (continued) - 2
Rxn.
Label
(a)
071
072
073
Kinetic Parameters (b)
1.
2.
3.
k(300)
47E+04
06E-t-02
76E+03
1
9
2
A
.47E+04
.06E+02
.28E+04
0
0
1
Ea
.00
.88
.07
-1
1
-1
B
.00
.00
.00
Reactions (b)
PAA
PAN
ONIT
+ HO = AC03 + H20
+ HO = HCHO + N03 + X02
* HO = HC3P + N02
Organic + N03 Reactions
093
094
096
098
099
100
101
102
9.
3.
3.
3.
1.
2.
1.
8.
23E-01
65E+00
65E+00
23E+04
72E-01
65E+01
84E+03
53E+02
8
2
2
3
2
1
4
8
.81E+02
.06E+03
.06E+03
.23E+04
.94E+03
.47E+04
.74E+04
.53E+02
4
3
3
0
5
3
1
0
.09
.78
.78
.00
.81
.77
.94
.00
-1
-1
-1
-1
-1
-1
-1
-1
.00
.00
.00
.00
.00
.00
.00
.00
HCHO
ALD
MGLY
CSL
OL2
OLT
OLI
ISO
+ N03 = H02 + HN03 * CO
+ N03 = AC03 + HN03
+ N03 = HN03 + AC03 * CO
+ N03 = HN03 + XN02 * 1.
+ N03 = OLN
-»- N03 = OLN
+ N03 = OLN
* N03 = OLN
5 CSL
Organic 03 Reactions
103
104
2.
1.
72E-03
74E-02
1
1
.76E+01
.94E+01
5
4
.23
.18
-1
-1
.00
.00
OL2
#.
OLT
#.
#.
+ 03 = HCHO + #.42 CO +
4 ORA1 +1.12 H02
+ 03 = #.53 HCHO + #.50
33 CO + #.20 ORA1 +
20 ORA2 + #.23 H02 +
#.22 M02 + #.10 HO + #.06
105
106
2.43E-01
2.20E-02
1
1
.07E+01
.81E+01
2
4
.26
.00
-1
-1
.00
.00
OLI
.
.
^
ISO
.
m
+ 03 = 1.18 HCHO + #.72
10 KET + #.23 CO + #.06
29 ORA2 + #.09 CH4 +
26 H02 + #.14 HO + #.31
+ 03 = #.53 HCHO + #.50
33 CO + #.20 ORA1 +
20 ORA2 + #.23 H02 +
22 M02 + #.10 HO
ALD +
CH4
ALD +
ORA1
M02
ALD +
075 7.52E+03 4.11E+03 -0.36 -1.00
076 2.90E-02 1.17E+18 26.91 0.00
086 1.12E+04 6.17E+03 -0.36 -1.00
079 1.12E+04 6.17E+03 -0.36 -1.00
080 1.12E+04 6.17E+03 -0.36 -1.00
Organic Peroxy + N02 Reactions
AC03 + N02 = PAN
PAN = AC03 * N02
Organic Peroxy + NO Reactions
AC03 + NO = M02 + N02
M02 + NO = HCHO + H02 + N02
HC3P + NO = #.75 ALD + #.25 KET +
#.09 HCHO + #.036 ONIT +
#.964 N02 + #.964 H02
(continued)
46
-------
Table 5 (continued) - 3
Rxn.
Label
(a)
M04
083
084
085
088
089
090
091
092
150
Kinetic
k(300)
1.12E+04
1.12E+04
1.12E+04
1.12E+04
1.12E+04
1.12E+04
1.12E+04
1.12E+04
1.12E+04
1.12E+04
6.
6.
6.
6.
6.
6.
6.
6.
6.
6.
Parameters (b)
A
17E+03
17E+03
17E+03
17E+03
17E+03
17E+03
17E+03
17E+03
17E+03
17E+03
Ea
-0.36 -1,
B
,00
-0.36 -1.00
-0.36 -1.00
-0.36 -1.00
-0.36 -1
-0.36 -1
-0.36 -1
-0.36 -1
-0.36 -1
-0.36 -1
.00
.00
.00
.00
.00
.00
Reactions (b)
HC6P +
#.15
#.85
OL2P +
N02
OLTP +
OLIP +
#.28
TOLP +
#.17
XYLP +
#.45
ETHP +
KETP +
OLN +
X02 +
NO = #.38 ALD + #.87 KET +
ONIT + #.03 HCHO +
N02 + #.85 H02
NO = #1.6 HCHO + H02 +
+ #.2 ALD
NO = ALD + HCHO + H02 + N02
NO = H02 + #1.45 ALD +
HCHO + #.10 KET + N02
NO = N02 + H02 +
MGLY + #.68 DCBX
NO = N02 + H02 +
MGLY + #.79 DCBX
NO = ALD + H02 -t- N02
NO = MGLY + N02 + H02
NO = HCHO + ALD + #2 N02
NO = N02
Organic Peroxy + H02 Reactions
107
108
109
M05
112
113
114
115
117
118
120
146
116
8.61E+03
1.
13E+02
(Same
(Same
(Same
(Same
(Same
(Same
(Same
(Same
(Same
(Same
(Same
(Same
-2.58 -1
k as 107
k as 107
k as 107
k as 107
k as 107
k as 107
k as 107
k as 107
k as 107
k as 107
k as 107
k as 107
.00
)
)
)
)
)
)
)
)
)
)
)
)
H02 +
H02 +
H02 +
H02 +
H02 +
H02 +
H02 +
H02 +
H02 +
H02 +
H02 +
H02 +
H02 +
M02 = OP1
ETHP = OP2
HC3P = OP2
HC6P = OP2
OL2P = OP2
OLTP = OP2
OLIP = OP2
KETP = OP2
TOLP = OP2
XYLP = OP2
OLN = ONIT
X02 = OP2
AC03 = PAA
Organic Peroxy + AC03 Reactions
142
130
134
135
3.64E-t-03
2.93E+03
1.04E+03
3.06E+02
1
1
4
1
.75E+03 -0.44 -1
.41E+03 -0.44 -1
.99E+02 -0.44 -1
.47E+02 -0.44 -1
.00
.00
.00
.00
AC03 + AC03 = #2 M02
M02 +
#.5
AC03 = HCHO + #.5 H02 +
M02 + #.5 ORA2
ETHP + AC03 = ALD + #.5 H02 +
#.5
HC3P H
M02 + #.5 ORA2
* AC03 = #.77 ALD +
#.26 KET + #.09 HCHO + #.5 H02 -t
#.5
M02 + #.5 ORA2
(continued)
47
-------
Table 5 (continued) - 4
Rxn.
Label
(a)
M06
138
139
140
141
143
144
M07
148
Kinetic Parameters (b)
k(300)
2.57E+02
1
1
1
1
1
1
1
1
.04E+03
.04E+03
.28E+02
.28E+02
.28E+02
.28E+02
.28E+02
.28E+02
1.
4.
4.
6,
6.
A
23E+02
99E+02
99E+02
17E+01
17E+01
6.17E+01
6
6
6
. 17E+01
.17E+01
.17E+01
-0.
-0.
-0.
-0.
-0.
-0.
Ea
44
44
44
44
44
44
-0.44
-0.44
-0.44
-1.
-1,
-1
-1
-1
-1
-1
-1
-1
B
00
00
00
.00
.00
.00
.00
.00
.00
Reactions (b)
HC6P
11
1.
OL2P
1.
OLTP
1.
OLIP
#.
1.
KETP
1.
+ AC03 =
.02 KET +
5 H02 + 1
+ AC03 =
5 H02 + 1
+ AC03 =
5 H02 + 1
+ AC03 =
#.45 ALD +
#.04 HCHO +
.5 M02 + #.5 ORA2
#.8 HCHO + #.6 ALD -
.5 M02 + #.5 ORA2
ALD + #.5 HCHO +
.5 M02 + #.5 ORA2
#.725 ALD +
55 KET + #.14 HCHO +1.5 H02
5 M02 + #
+ AC03 =
5 M02 + #
AC03 + TOLP =
1.
*.
68 DCBX +
5 ORA2
AC03 + XYLP =
i.
i.
OLN
1.
X02
79 DCBX +
5 ORA2
+ AC03 =
5 ORA2 +
+ AC03 =
.5 ORA2
MGLY + #.5 H02 +
.5 ORA2
#.17 MGLY +
#.5 H02 + #.5 M02 +
#.45 MGLY +
#.5 H02 + #.5 M02 +
HCHO + ALD +
N02 + #.5 M02
M02
Organic Peroxy + M02 Reactions
121
122
123
M08
126
127
128
129
131
132
5
4
1
1
4
4
5
5
5
5
.81E+02
.28E+02
.28E+02
.04E+02
.28E+02
.28E+02
.19E+01
.19E+01
.19E+01
.19E+01
2
2
6
4
2
2
2
2
2
2
.79E+02
.06E+02
.17E+01
.99E+01
.06E+02
.06E+02
.50E+01
.50E+01
.50E+01
.50E+01
-0.44
-0.44
-0.44
-0.44
-0.44
-0.44
-0
-0
-0
-0
.44
.44
.44
.44
-1
-1
-1
-1
-1
-1
-1
-1
-1
-1
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
M02
M02
1.
M02
t.
M02
#.
M02
1.
M02
*
M02
1
M02
1
M02
t
M02
*
+ M02 = #
+ ETHP =
75 ALD
+ HC3P =
77 ALD +
+ HC6P =
45 ALD +
+ OL2P =
35 ALD
+ OLTP =
75 ALD
+ OLIP =
725 ALD
+ KETP =
.75 MGLY
+ TOLP =
.17 MGLY H
+ XYLP =
1.5 HCHO + H02
#.75 HCHO + H02 +
.84 HCHO + H02 +
.26 KET
.79 HCHO + H02 +
1.02 KET
1.55 HCHO + H02 +
#1.25 HCHO + H02 +
#.89 HCHO + H02 +
#.55 KET
#.75 HCHO + H02 +
#.75 HCHO + H02 +
- #.68 DCBX
#.75 HCHO + H02 +
.45 MGLY + #.79 DCBX
(continued)
48
-------
Table 5 (continued) - 5
Rxn.
Label
(a)
Kinetic Parameters (b)
k(300) A Ea B
Reactions (b)
M09 5.19E+01 2.50E+01 -0.44 -1.00 M02 + OLN = 11.75 HCHO
ALD + N02
147 5.19E+01 2.50E+01 -0.44 -1.00 M02 + X02 = HCHO + H02
#.5 H02
149
M10
M11
M12
M13
Other Organic Peroxy + Peroxy
Reactions
1.10E+00 5.28E-01 -0.44 -1.00 X02 + X02 =
1.10E+00 5.28E-01 -0.44 -1.00 OLN + OLN = 12 HCHO + #2 ALD
12 N02
XN02 Reactions
1.12E+04 6.17EH-03 -0.36 -1.00
(Same k as 107 )
1.20E-02 (No T Dependence)
XN02 + N02 = (Loss-of-NOx)
XN02 + H02 =
XN02 =
Notes:
(a) Reactions which correspond directly to reactions in the March 1988
RADM mechanism in Table 2 are given the same reaction labels as used
for them in Table 2. Reactions that were added or significantly
modified relative to the mechanism in Table 2 are indicated by the
label "Mxx".
(b) See Footnotes (b) and (c) in Table 2 for a description of the
formats used in the kinetic parameter and the reaction listings.
49
-------
The model species used in the RADM-P mechanism are the same as those
listed for the RADM mechanism in Table 1. The organic reactions and their
thermal rate parameters for the RADM-P mechanism are listed in Table 6.
The inorganic reactions and rate parameters are also the same as those of
the mechanism listed in Table 2 and are thus not included in Table 6. The
absorption coefficients and quantum yields used for the photolysis reac-
tions in this mechanism are the same as used for the RADM-M mechanism and
are given in Table 3. This mechanism was implemented by Stockwell on his
modeling software at SUNY, and test calculations indicate that the mech-
anism implemented at SUNY is consistent with its implementation at SAPRC.
50
-------
Table 6. Listing of the Organic Reactions in the Modified Parameter
Version of the RADM Gas-Phase Chemical Mechanism (RADM-P)
Rxn.
Label
(a) k(300)
Kinetic Parameters (b)
A Ea B
Reactions (b)
010
011
012
013
014
015
016
017
018
019
020
( Phot .
(Phot.
( Phot .
(Phot.
(Phot.
(Phot.
(Phot.
(Phot.
(Phot.
(Phot.
(Phot.
Set
Set
Set
Set
Set
Set
Set
Set
Set
Set
Set
= HCHOMR )
= HCHORR )
= ALDR )
= OPR )
= OPR )
= PAAR )
= KETR )
= GLY2R )
= GLY1R )
= MGLYRM )
= DCBRM )
021
Organic Photolysis Reactions
HCHO + HV = H2
HCHO + HV = H02
CO
t- H02
ALD + HV = M02 + H02
OP1 + HV = HCHO + H02
OP2 * HV = ALD + H02
PAA
KET
(Phot. Set = ONITR )
CO
CO
HO
HO
+ HV = M02 + C02 -i- HO
+ HV = AC03 + ETHP
GLY f HV = #.13 HCHO + #1.87 CO
GLY 1- HV = #.45 HCHO + #1.55 CO
#.80 H02
MGLY + HV = AC03 + H02 + CO
DCB + HV = #.98 H02 + #.02 AC03
TC03
ONIT + HV = #.20 ALD + #.80 KET
H02 * N02
Organic + OH Reactions
051
052
053
054
055
056
057
058
074
059
060
063
064
065
066
067
061
1,
4
.29E+01
.11E+02
3.86E+03
7
1
1
4
9
1
9
4
1
2
1
1
2
5
.16E+03
.51E+04
.24E+04
.19E+04
.79E+04
.46E+05
.02E+03
.09E+04
.32E+04
.37E+04
.47E+03
.69E+04
.50E+04
.87E+04
9.23E+02
1.81E+03
2.33E+04
2.54E+04
5.34E+04
3.16E+03
7.81E+03
1.57E+04
3.74E+04
3.08E-1-03
2 . 77E+04
1.32E+04
1.01E+04
1.76E+04
1.69E+04
2.50E+04
5.87E+04
2.54
0.88
1.07
0.75
0.75
-0.82
-1.00
-1.09
-0.81
-0.64
-0.23
0.00
-0.51
1.48
0.00
0.00
0.00
1
1
-1
-1
-1
-1
-1
-1
-1
-1
-1
-1
-1
-1
-1
-1
-1
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
CH4
ETH
HC3
#.
#.
HC5
HC8
OL2
OLT
OLI
ISO
TOL
#
XYL
*
+ HO =
+ HO =
+ HO =
M02 ,
ETHP
#.83
i- H20
+ H20
HC3P *
#.17 H02 +
009 HCHO + #.075 ALD +
025 KET + H20
+ HO =
+ HO =
+ HO =
+ HO =
+ HO =
+ HO =
+ HO =
.25 H02
+ HO =
.17 H02
HCHO + HO
ALD
KET
GLY
+ HO =
+ HO =
+ HO =
MGLY + HO
CSL
#
+ HO =
.9 TC03
HC5P
HC8P
OL2P
OLTP
OLIP
OLTP
#.75
#.83
= H02
AC03
KETP
H02
+ #.25
+ #.75
TOLP t-
XYLP +
+ CO +
+ H20
+ H20
+ #2 CO
= AC03 + CO
#.1
H02 + #
X02 + H20
X02 + H20
#.25 CSL *
#.17 CSL +
H20
I- H20
+ H20
.9 X02 +
(continued)
51
-------
Table 6 (continued) - 2
Rxn.
Label
(a)
062
068
069A
069B
070A
070B
071
072
073
Kinetic Parameters (b)
k(300)
5
4
7
7
7
7
1
2
3
.28E+04
. 11E+04
. 34E+03
. 34E+03
. 34E+03
.34E+03
.47E+04
.06E+02
. 76E+03
5
4
7
7
7
7
1
9
2
A
.28E+04
.11E+04
.34E-I-03
.34E+03
.34E+03
.34E+03
.47E+04
.06E+02
.28E-I-04
0
0
0
0
0
0
0
0
1
Ea
.00
.00
.00
.00
.00
.00
.00
.88
.07
B
-1.00
-1.00
-1.00
-1.00
-1.00
-1.00
-1.00
1.00
-1.00
Reactions (
CSL
DCB
OP1
HO +
OP2
HO +
PAA
PAN
ONIT
+ HO =
+ HO =
+ HO =
OP1 =
- HO =
OP2 =
-i- HO =
+ HO =
» HO =
b)
CSL
TC03 + H20
M02
HO + HCHO
HC3P
HO + ALD
AC03 + H20
HCHO + N03 + X02
HC3P + N02
Organic + N03 Reactions
093
094
095
096
097
098
099
100
101
102
9
3
9
3
3
3
1
2
1
8
.23E-01
.65E+00
.23E-01
. 65E+00
.65E+00
. 23E+04
.72E-01
.65E+01
.84E+03
.53E-I-02
8
2
8
2
2
3
2
1
14
8
.81E+02
.06E+03
.81E+02
.06E+03
.06E+03
.23E+04
.94E+03
.47E+04
.74E+04
.53E+02
4
3
4
3
3
0
5
3
1
0
.09
.78
.09
.78
.78
.00
.81
.77
.94
.00
-1.00
-1.00
-1.00
-1.00
-1.00
-1.00
-1.00
-1.00
-1.00
-1.00
HCHO
ALD
GLY
MGLY
DCB
CSL
OL2
OLT
OLI
ISO
+ N03
+ N03 =
+ N03 =
* N03
+ N03 =
+ N03 =
* N03 =
+ N03 =
+ N03 =
+ N03 =
= H02 + HN03 > CO
AC03 + HN03
HN03 - H02 > #2 CO
= HN03 -i- AC03 + CO
HN03 + TC03
HN03 -* XN02 + #.5 CSL
OLN
OLN
OLN
OLN
Organic + 03 Reactions
103
104
105
106
2
1
2
2
. 72E-03
.74E-02
.43E-01
.20E-02
1
1
1
1
.76E+01
.94E+01
.07E+01
.81E+01
5
4
2
4
.23
.18
.26
.00
-1.00
-1.00
-1.00
-1.00
OL2
#.
OLT
.
^
OL
.
ISO
.
.
+ 03 =
4 ORA1
+ 03 =
33 CO +
20 ORA2
22 M02
+ 03 =
10 KET
29 ORA2
26 H02
+ 03 =
33 CO +
20 ORA2
22 M02
HCHO + #.42 CO t-
f #.12 H02
#.53 HCHO + #.50 ALD +
#.20 ORA1 +
+ #.23 H02 +
+ #.10 HO -- .06 CH4
#.18 HCHO + .72 ALD +
* #.23 CO + .06 ORA1
+ #.09 CH4
+ #.14 HO + .31 M02
#.53 HCHO + .50 ALD +
#.20 ORA1 +
+ #.23 H02 +
-t- #.10 HO
Organic Peroxy + N02 Reactions
075 7.52E+03 4.11E+03 -0.36 -1.00
076 2.90E-02 1.17E+18 26.91 0.00
AC03 + N02 = PAN
PAN = AC03 + N02
(continued)
52
-------
Table 6 (continued) - 3
Rxn.
Label
(a)
077
078
Kinetic
k(300)
6.90E+03
2.90E-02
6.
1.
Parameters (b)
A
90E+03
17E+18
Ea B
0.00 -1.
26.91 0.
00
00
Reactions (b)
TC03 +
TPAN =
N02 =
TC03
TPAM
+ N02
Organic Peroxy + NO Reactions
086
087
079
080
1.12E+OM
1.12E+04
1.12E+04
1.12E+04
6.
6.
6.
6.
17E+03
17E+03
.17E+03
, 17E+03
-0.36 -1.
-0.36 -1.
-0.36 -1,
-0.36 -1,
.00
,00
.00
.00
AC03 +
TC03 +
1.89
#.05
NO =
NO =
GLY
AC03
M02 + N02
N02 + #.92 H02 +
+ #.11 MGLY
+ #.95 CO +
M02 + NO = HCHO + H02 +
HC3P +
1.09
NO =
HCHO
1.964 M02
081
082
083
084
085
088
089
090
091
092
153
1.12E+04
1.12E+04
1.12E+04
1.12E+04
1.12E+04
1.12E+04
1.12E+04
1.12E+OM
1.12E+04
1 . 12E+04
1.12E+04
6.
6
6
6
6
6
6
6
6
6
6
. 17E+03
. 17E+03
. 17E+03
. 17E+03
. 17E+03
. 17E+03
. 17E+03
.17E+03
.17E+03
.17E+03
.17E+03
-0.36 -1.
-0.36 -1
-0.36 -1
-0.36 -1
-0.36 -1
-0.36 -1
-0.36 -1
-0.36 -1
-0.36 -1
-0.36 -1
-0.36 -1
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
HC5P +
1.08
HC8P +
1.04
#.76
OL2P *
N02 -
OLTP +
OLIP +
#.28
TOLP +
#.17
XYLP +
#.45
ETHP +
KETP +
OLN +
X02 +
NO =
ONIT
NO =
HCHO
#.75 ALD + #
+ #.036 ONIT
+ #.964 H02
#.38 ALD + #
+ #.92 N02 +
#.35 ALD + #
+ #.24 ONIT
+
#2 X02
N02
.25 KET +
+
.69 KET +
#.92 H02
1.06 KET +
-»-
N02 + #.76 H02
NO =
« #.2
NO =
NO =
#1.6 HCHO +
ALD
H02 +
ALD + HCHO + H02 + N02
H02 + #1.45
HCHO +1.10 KET
NO =
MGLY
NO =
MGLY
NO =
NO =
NO =
NO =
N02 + H02 +
+ #.16 GLY -
N02 + H02 +
+ #.806 DCB
ALD + H02 +
MGLY + N02 H
HCHO + ALD +
N02
ALD +
+ N02
- #.70 DCB
N02
K H02
#2 N02
Organic Peroxy + H02 Reactions
107
108
109
110
111
112
113
114
115
117
8.61E+03
1
.13E+02
(Same
(Same
(Same
(Same
(Same
(Same
(Same
(Same
(Same
-2.58 -1
k as 107
k as 107
k as 107
k as 107
k as 107
k as 107
k as 107
k as 107
k as 107
.00
)
)
)
)
)
)
)
)
)
H02 +
H02 --
H02 +
H02 «
H02 +
H02 +
H02 +
H02 +
H02 +
H02 *
M02 =
ETHP
HC3P
HC5P
HC8P
OL2P
OLTP
OLIP
KETP
TOLP
OP1
= OP2
= OP2
= OP2
= OP2
= OP2
= OP2
= OP2
= OP2
= OP2
(continued)
53
-------
Table 6 (continued) - 4
Rxn.
Label
(a) k(300)
Kinetic Parameters (b)
A Ea B
Reactions (b)
118
120
149
116
119
(Same k as 107 )
(Same k as 107 )
(Same k as 107 )
(Same k as 107 )
(Same k as 107 )
H02
H02
H02
H02
H02
+ XYLP = OP2
+ OLN = ONIT
+ X02 = OP2
+ AC03 = PAA
+ TC03 = OP2
143 3.64E+03 1.75E+03 -0.44 -1.00
146 3.64E+03 1.75E+03 -0.44 -1.00
130 2.93E+03 1.41E+03 -0.44 -1.00
133 2.93E+03 1.41E+03 -0.44 -1.00
135 1.04E+03 4.99E-I-02 -0.44 -1.00
136 3.06E+02 1.47E+02 -0.44 -1.00
137 2.57E+02 1.23E+02 -0.44 -1.00
138 2.20E+02 1.06E+02 -0.44 -1.00
139 1.04E+03 4.99E+02 -0.44 -1.00
140 1.04E+03 4.99E+02 -0.44 -1.00
141 1.28E+02 6.17E+01 -0.44 -1.00
142 1.28E+02 6.17E+01 -0.44 -1.00
144 1.28E+02 6.17E+01 -0.44 -1.00
145 1.28E+02 6.17E+01 -0.44 -1.00
147 1.28E+02 6.17E+01 -0.44 -1.00
Organic Peroxy + AC03 Reactions
AC03 + AC03 = 12 M02
AC03 + TC03 = M02 + #.92 H02 +
1.890 GLY + 1.11 MGLY +
1.05 AC03 f 1.950 CO + 12 X02
M02 + AC03 = HCHO + #.5 H02 +
4.5 M02 + 4.5 ORA2
M02 * TC03 =1.5 HCHO + 4.5 ORA2 +
1.46 H02 + 1.445 GLY +
1.055 MGLY + 1.025 AC03 +
#.475 CO + X02
ETHP + AC03 = ALD + *.5 H02 +
1.5 M02 +1.5 ORA2
HC3P + AC03 = #.77 ALD +
1.26 KET -K 1.5 H02 * 1.5 M02 +
1.5 ORA2
HC5P + AC03 = 1.41 ALD -»-
1.75 KET + 4.5 H02 + #.5 M02 +
1.5 ORA2
HC8P -K AC03 = #.46 ALD +
#1.39 KET + #.5 H02 + #.5 M02 +
#.5 ORA2
OL2P + AC03 = #.8 HCHO + #.6 ALD +
#.5 H02 + #.5 M02 + #.5 ORA2
OLTP + AC03 = ALD + #.5 HCHO +
#.5 H02 + #.5 M02 + #.5 ORA2
OLIP + AC03 = #.725 ALD +
#.55 KET * #.14 HCHO * #.5 H02 +
#.5 M02 + #.5 ORA2
KETP + AC03 = MGLY + #.5 H02 +
#.5 M02 + #.5 ORA2
AC03 + TOLP = M02 + #.17 MGLY +
#,16 GLY * #.70 DCB + H02
AC03 + XYLP = M02 + #.45 MGLY +
#.806 DCB + H02
AC03 + OLN = HCHO + ALD +
#.5 ORA2 + N02 + 4.5 M02
(continued)
54
-------
Table 6 (continued) - 5
Rxn,
Label
(a)
151
Kinetic
k(300)
1.28E+02
6.
Parameters (b)
A
17E+01
Ea
-0.44
B
-1.00
Reactions (b)
X02 + AC03 =
M02
Organic Peroxy + M02 Reactions
121
122
123
124
125
126
127
128
129
131
132
134
150
5.81E+02
4.28E+02
1.28E+02
1.04E+02
8.86E+01
4.28E+02
4.28E+02
5.19E+01
5.19E+01
5.19E+01
5.19E+01
5.19E+01
5.19E+01
2.
2.
6.
4.
4.
2.
2.
2.
2.
79E+02
06E+02
17E+01
99E+01
26E+01
06E+02
06E+02
50E+01
50E+01
2.50E-I-01
2.50E+01
2.50E+01
2.50E+01
-0.
-0.
-0.
-0.
-0.
-0.
-0.
-0.
-0.
44
44
44
44
44
44
44
44
44
-0.44
-0.44
-0.44
-0.44
-1.00
-1.00
-1.00
-1.00
-1.00
-1.00
-1.00
-1.00
-1.00
-1.00
-1.00
-1.00
-1.00
M02 + M02 = #1.5 HCHO »
M02 » ETHP =
#.75 ALD
M02 -- HC3P =
#.77 ALD +
M02 -t- HC5P =
#.41 ALD -t-
M02 1- HC8P =
#.46 ALD +
M02 + OL2P =
#.35 ALD
M02 + OLTP =
#.75 ALD
M02 -H OLIP =
#.725 ALD
M02 + KETP =
#.75 MGLY
M02 + TOLP =
#.17 MGLY
M02 + XYLP =
#.45 MGLY
M02 + OLN =
ALD + N02
X02 + M02 =
#.75 HCHO
.84 HCHO
.26 KET
.77 HCHO
.75 KET
.80 HCHO
#1.39 KET
- H02
+ H02 +
+ H02 +
+ H02 +
+ H02 +
#1.55 HCHO -i- H02 +
#1.25 HCHO * H02 +
#.89 HCHO
+ #.55 KET
#.75 HCHO
HCHO + #2
+ #.16 GLY
HCHO + #2
-.- H02 +
+ H02 -.-
H02 +
+ #.70 DCB
H02 +
+ #.806 DCB
#1.75 HCHO
HCHO + H02
Other Organic Peroxy +
148
152
1.10E+00
1.10E+00
5
5
.28E-01
.28E-01
-0
-0
.44
.44
-1.00
-1.00
Reactions
OLN -f OLN =
#2 N02
X02 + X02 =
#2 HCHO +
I- #.5 H02
Peroxy
#2 ALD +
XN02 Reactions
154
155
156
157
158
1.12E+04
5.19E+01
1.28E+02
1.10E+00
6
2
6
5
.17E-I-03
(Same
.50E+01
.17E+01
.28E-01
-0
.36
-1.00
k as 107 )
-0
-0
-0
.44
.44
.44
-1.00
-1.00
-1.00
XN02 + N02 =
XN02 + H02 =
XN02 * M02 :
XN02 + AC03
XN02 * XN02
: ONIT
: OP2
: HCHO + H02
= M02
=
(continued)
55
-------
Table 6 (continued) - 6
Notes:
(a) Reaction labels shown in this table are the reaction sequence
number used by Stockwell to implement this mechanism at SUNY.
Some reactions were re-arranged for the purpose of this listing.
Some reactions were split into two reactions for implementation
on SAPRC modeling software.
(b) See Footnotes (b) and (c) in Table 2 for a description of the
formats used in the kinetic parameter and the reaction listings,
56
-------
3. AGGREGATION AND REPRESENTATION OF ORGANIC EMISSIONS IN RADM
3.1 Introduction
The gas-phase reactions of volatile organic compounds (VOCs) are
important in regional oxidant and acid deposition models for a number of
reasons. They play a major role in the formation of oxidants on the local
and regional scale, contribute to the generation (or inhibition) of the
radicals responsible for converting sulphur oxides to sulphuric acid and
nitrogen oxides to nitric acid, and are involved in the formation of
peroxides. VOCs differ significantly in their effects on these processes,
and the model must be able to represent these differences in reactivity
with a reasonable degree of accuracy. This is a difficult problem, not
only because there remain significant uncertainties in our knowledge of
the atmospheric reactions of most types of VOCs, but also because the many
hundreds (if not thousands) of different VOCs emitted into the atmosphere
must be represented by only a limited number of model species. For
example, the 1985 NAPAP inventory for anthropogenic emissions of VOCs in
the United States (Wagner et al.f 1986; Shareef et al., 1988) include
emissions for over 600 different chemical compounds, while the gas-phase
mechanism proposed for RADM-II has only 17 model species which are used to
represent them. The methods used to represent these hundreds of compounds
with the limited number of model species is therefore an important
component of the gas-phase mechanism in RADM.
As part of the most recent update of the SAPRC gas-phase mechanism
for oxidant models, we had previously, under funding from the California
Air Resources Board, developed new VOC emissions processing procedures for
representing in such models the large numbers of chemical species in
detailed emissions inventories (Carter et al., 1988a,b). These procedures
were designed to utilize, as much as possible, the chemical information
available in the detailed VOC emissions inventories to derive the kinetic
and mechanistic parameters for the model species. As part of this effort,
kinetic and mechanistic parameters were assigned to the major categories
of compounds used in current detailed emissions inventories. In addition,
emissions processing software was developed to utilize these assignments
to derive the parameters for the lumped model species which best corres-
pond to those of the emissions profile being represented. Although the
57
-------
complexity of this new emissions processing system is such that it is not
practical to utilize it in its current form in the RADM model, the
procedures, data bases, and software developed in this effort could be
utilized to recommend appropriate methods for representing VOC emissions
in RADM.
Therefore, as part of this gas-phase mechanism evaluation program for
RADM, we assisted the RADM team in the development of procedures for
processing and representing VOC emissions for RADM-II. This effort has
involved: (1) working with Dr. Paulette Middleton of the RADM team in
the development of a new emissions processing and aggregation system for
regional models which can be used with a variety of chemical mechanisms;
(2) working with Stockwell and Middleton to determine how best to
distribute the various VOC compounds in emissions inventories among the
species used in the RADM mechanism; and (3) working with Stockwell in
deriving the kinetic and mechanistic parameters for the lumped VOC species
in the RADM mechanism which best represent those of the sets of emitted
species they are intended to represent. The results of the first two
efforts are discussed below in Section 3.2, and the results of the third
are discussed in Section 3-3-
3.2 Aggregation of VOC Emissions Input for RADM
The current NAPAP inventory of VOC emissions in the RADM modeling
region gives mass emissions of over 550 types of chemicals from several
thousand different source categories throughout the United States. These
detailed emissions data are not input into the RADM model directly, but
instead must be aggregated into the much smaller number of VOC categories
which are utilized in the model. Because of the extremely large size of
the data bases involved, this is a very expensive procedure. In the past,
the procedure has been to aggregate the detailed VOC emissions data into
the species used by the model in a single step. Since the number and
classification of VOC species in a model in general depends on the
chemical mechanism used, the aggregation procedure must be carried out
separately for each chemical mechanism, and once this procedure has been
carried out, the representation of VOCs used in the mechanism cannot be
readily modified.
58
-------
Because there are currently available a number of chemical mechanisms
which can be utilized in regional models, each in general using differing
methods for aggregating VOC emissions, it would be advantageous to develop
a procedure to aggregate the detailed emissions data into a more manage-
able data base which still contains all the information required by the
different mechanisms, including mechanisms which may be developed in the
future. To address this, Middleton proposed developing a two-step
emissions aggregation process for the RADM model, where the detailed data
are first aggregated into a manageable number of emissions categories
which can be utilized by a variety of chemical mechanisms, and then these
emission categories are further aggregated into the smaller set of VOC
species which are directly input into the RADM mechanism. Since, in
principle, only the second, much less expensive, aggregation step is
necessarily mechanism-dependent, this means that this data base can be
more readily utilized by alternative chemical mechanisms either for
different types of applications, or for mechanism intercomparison
studies. The development of the mechanism-independent VOC emissions
classification scheme as it will be used for RADM, and recommended
extensions to this scheme which are not yet implemented, are discussed
below. Following that, the procedures developed for further aggregating
these emissions into RADM model species are discussed.
3.2.1 Development of a Mechanism-Independent VOC Emissions
Classification System
The primary objective in the development of a mechanism-
independent VOC emissions aggregation scheme is to utilize a minimum
number of chemical groupings while retaining, as much as possible, the
types of chemical information needed to represent the various types of
relevant impacts that VOCs might have on model simulations. These types
of impacts, and their magnitudes on model outputs of interest, are
referred to under the general term of "reactivity," where relevant
measures of reactivity depend on the application of the model. Therefore,
the objective in developing a general emissions classification scheme
would be to organize the VOCs according to their reactivity where each
class has compounds which are similar in all aspects of reactivity
important for the model application. Since the aspects of reactivity
which are of greatest interest depend on the application of the model, the
59
-------
most appropriate aggregation scheme will also be dependent on the intended
application of the data.
In practice, if the number of categories which can be used is
limited, compounds with different reactivity characteristics would have to
be lumped together in the same category. Given a set number of cate-
gories, the extent to which reactivities should differ within a given
class would depend on how important the compounds are expected to be in
affecting the model, which in turn depends both on how much is emitted and
on how reactive they are (i.e., on the magnitude of the effects of a given
amount of the VOC on the relevant predictions of the model). For example,
reactive compounds which are emitted in very large amounts (such as, for
example, ethylene) should be in classes by themselves, and categories
representing groups of relatively reactive types of compounds with signif-
icant amounts of emissions should represent relatively narrow ranges of
reactivity. On the other hand, it is not worthwhile to have separate
classes for reactive compounds of very low emissions; they can be grouped
into classes which are closest to them in reactivity, even if the classes
are not particularly close. Furthermore, compounds which are essentially
unreactive for the purposes of the particular model application can be
lumped together into a single "inert" category, regardless of how much may
be emitted. However, a compound which is "inert" in one application may
be reactive in another, so if the categorization scheme is to be used for
several different applications, there may be more than one category which
is treated as "inert" in a given type of application.
In the case of aggregation of emissions for models such as RADM, the
relevant reactivity criteria concern reactivity with respect to oxidant
formation and reactivity with respect to formation of acidic species.
Reactivity with respect to oxidant formation is influenced by how rapidly
the VOC reacts in the atmosphere, and, if it reacts sufficiently rapidly,
by the nature of its reaction mechanism. Therefore, compounds which react
sufficiently rapidly that they participate in the oxidant formation
process, yet react at significantly different rates, should be placed in
different reactivity categories, while compounds which do not react to a
significant extent in the time frame of the model application can all be
lumped together and be classified as "inert." If a VOC reacts, the nature
of its reaction mechanism is also important, since different aspects of a
60
-------
VOCs reaction can affect the system in different ways. For example, the
reactions of a VOC can have a relatively direct effect on ozone formation
by forming radicals which convert NO to NOo (which causes ozone formation
in the presence of sunlight), and more indirectly (but often more signifi-
cantly) by enhancing or inhibiting levels of the radicals which control
the overall rate of the oxidant formation process, and by the formation of
reactive organic products. For this reason, the emissions categorization
scheme must not group together types of compounds with significantly
different types of reaction mechanisms, even if they may react at similar
rates. At a minimum, the classification scheme should include separate
categories of compounds which are represented with different sets of
reactions in current oxidant formation mechanisms. Finally, since the
model is concerned with formation of acidic species, VOCs which are acids
themselves should not be lumped into the same categories as those which
are not.
Ideally, it would be desirable that the emissions aggregation scheme
used for RADM and other oxidant and acid deposition models also be suit-
able for use with models concerned with global effects and the strato-
sphere. This would require the use of additional VOC categories. For
global models, reactions occur on a much longer time scale, and thus many
compounds classified as "inert" for regional model applications would have
to be placed in separate "reactive" categories for global models. In
addition, for models where the stratosphere is of interest, separate
categories would have to be created for halogen-containing compounds which
are transported to the stratosphere.
Unfortunately, current limitations in the software utilized by the
EPA and its contractors for processing emissions input for RADM restrict
the number of categories to 32. Our analysis indicates that this may be
sufficient to convey the necessary reactivity information required for
local and regional oxidant and acid deposition models using currently
available chemical mechanisms, but is not sufficient to convey the necess-
ary additional information required by global and stratospheric models.
Therefore, it was determined that the initial emissions aggregation scheme
to be developed for RADM would not contain the additional VOC categories
required by such models.
61
-------
An analysis of the total emissions into the contiguous United States
of the approximately -550 different types of VOCs in the NAPAP inventory
was carried out to determine the most appropriate aggregation scheme,
involving no more than 32 categories, to use for regional oxidant and acid
rain models. This was carried out jointly by SAPRC as part of this
program and by Middleton and co-workers at NCAR as part of the RADM
development effort. NCAR determined the major overall characteristics of
the classification scheme and the types of organics that had to be
represented in separate classes because of their importance in the
emissions inventory; they also had the final responsibility for the system
which was developed. SAPRC provided guidance to NCAR in this task by
analyzing the reactivity characteristics of the various types of VOCs in
the inventory, and making recommendations as to how species should be
classified based on their reactivity. This analysis utilized the kinetic
and mechanistic assignments which were made for emitted VOC categories as
part of its development of the new emissions processing system for the new
California ARB model (Carter, 1988), together with new assignments which
had to be made as part of this program for a number of major VOC categor-
ies which are on the NAPAP, but not the California inventories.
A number of emissions categories were assigned without direct
consideration of reactivity, but based on considerations of importance of
individual compounds in the total mass emissions, or on considerations of
special chemical characteristics. In addition, special categories were
created for emissions where there is information missing concerning their
chemical characteristics. These are as follows:
(1) On the basis of total amounts of mass emitted into the United
States, combined with considerations of their special chemical character-
istics, eight individual compounds were determined to be sufficiently
important that they should not be aggregated with other species, but
should be in categories by themselves. These were methane, ethane,
propane, ethene, propene, acetylene, formaldehyde, and acetone. Initial-
ly, it was also proposed that benzene, methanol, ethanol, isopropyl
alcohol, perchloroethylene, and methyl chloroform be in their own cate-
gories because of large current or (in the case of methanol) possible
future emissions; but it was decided that given the 32-class limitation
there were an insufficient number of classes available for this purpose.
62
-------
These compounds were lumped with other species either because they are
relatively unreactive, or because their reactivities were judged to be
fairly similar to those of other species. Note, however, that many of
these compounds dominate the categories in which they were placed.
(2) Organic acids were lumped into their own separate category
because of their obvious role in acidification. Their contribution to
reactivity with respect to oxidant formation is minor, so it was not
necessary to differentiate them further on this basis. Although the RADM
mechanism represents formic acid separately from the higher organic acids,
the emissions of formic acid in the NAPAP inventory are insufficient to
justify utilization of a separate emissions category grouping just for
this compound.
(3) Two aggregation categories were created for "chemical compounds"
listed in the current detailed NAPAP (and California) emissions inventor-
ies which are not true compounds at all, but represent unspecified
mixtures of compounds of ambiguous reactivity. The category "Alkane/
Aromatic Mix" was used to represent emissions which are designated as
"Naphtha," "Mineral Spirits," or "Lactol Spirits." These are solvents
believed to consist primarily of Cg^ alkanes and aromatics, but whose
exact chemical compositions are unspecified. The category "Alkenes
(Primary/Internal Mix)" was used to represent emissions designated as "Cn
alkenes" (where n>^ 4), without reference to the location of the double
bond. The location of the double bond significantly affects the reac-
tivity of the alkene, and is a necessary basis for the reactivity classif-
ications of these compounds.
(4) An "unreactive" category was included for compounds which are
estimated to react too slowly to be important in affecting chemical
transformations on the regional scale. Note, however, that many compounds
in this category are not "unreactive" on a global scale or in the
stratosphere.
(5) An "unidentified" category was included for unidentified VOCs.
These constitute somewhat less than 6J of the mass in the NAPAP total U.S.
anthropogenic emissions inventory.
(6) An "unassigned" category was included for those VOCs which were
identified, but for which no reactivity assignments or classifications
were made. As indicated above, as part of this and our previous ARB
63
-------
program (Carter, 1988) we made mechanistic classifications and (where
applicable) reactivity assignments for most of the detailed emissions
categories in the California and NAPAP inventories. However, a number of
compounds in the NAPAP inventory were not categorized because they both
were very difficult to classify and had small total U.S. emissions. These
"unassigned" compounds constitute less than 1/t of the mass in the NAPAP
total U.S. anthropogenic emissions inventory.
The above amounts to a total of 1U categories, leaving 18 categories
available for all other types of compounds in a 32-category system. Based
on considerations of differences in chemical reaction mechanisms, together
with considerations of total amounts of different types of chemicals
emitted, it was determined that the remaining categories should, at a
minimum, differentiate between the following general types of chemicals
(listed in order of mass emitted):
Alkanes (other than methane, ethane, and propane)
Aromatics
"Others" (alcohols, ethers, alcohol ethers, esters, and other
types of compounds which are generally represented as reacting
analogously to alkanes in current mechanisms)
Alkenes (other than ethene and groups of chemicals classified as
"primary/internal mix")
Haloalkenes
Aldehydes (other than formaldehyde)
Styrenes
Phenols and Cresols
There are a number of other chemical classes emitted, but it was judged
not to be worthwhile to assign them to separate groups because either the
amounts emitted were too small or they were judged to be of minor
reactivity with respect to oxidant or acid formation, and thus could be
classified as "inert."
In the case of the first four groups listed above (alkanes,
aromatics, "others," and alkenes), the variety of rates of reaction of the
-------
individual species are such that these groups needed to be further broken
down on the basis of differences in reaction rates. At a minimum, the
numbers of reactivity categories for each type of chemical should be such
that there are at least as many as the number of lumped species which
might be used for them in the models, and the criteria used to make the
reactivity differentiations should also be compatible with those used in
the models. Based on these considerations, the reactivity classifications
were derived as follows:
Alkenes: Most current mechanisms differentiate between ethene, other
terminal alkenes, and internal alkenes. This differentiation is therefore
also used in the emissions categorization. As indicated above, ethene and
propene are given emissions categories by themselves, so the differentia-
tion of the mechanisms between ethene and other alkenes is already taken
care of. The terminal alkenes other than propene are lumped into the
group designated "alkenes (primary)," and the more reactive alkenes with
internal double bonds are lumped into "alkenes (internal)."
However, certain "terminal" alkenes, such as 1,3-butadiene and beta-
pinene have atmospheric reaction rates which are more typical of the
internal alkenes, and lumping these into the "primary alkenes" group
seemed to us to be inappropriate. To provide an objective criterion to
determine the classification of these more reactive alkenes, all alkenes
(other than ethene or propene) which react with OH radicals with rate
it 1 1
constants less than 7.5 x 10 ppm min were assigned to the "primary"
group, and alkenes with higher rate constants were assigned to the
"internal" group. This rate constant lies well between the typical ranges
for "ordinary" primary and internal alkenes, but results in the unusually
reactive "primary" alkenes such as beta-pinene being grouped with the more
reactive internal alkenes, which have more similar atmospheric reaction
rates.
Note that this classification for the alkenes is appropriate for
inventories which contain anthropogenic emissions only. If biogenic
emissions are to be merged into the same scheme, then separate categories
should be reserved for them. This is particularly important for regional
modeling applications, since biogenics tend to dominate the.mass emissions
for large regions of the modeling domain. The biogenic emissions inven-
tory includes species such as isoprene, alpha-pinene, other terpenes, and
65
-------
unidentified biogenics (Middleton, private communication, 1988), and, at a
minimum, separate categories for each of these should be included in any
merged emissions grouping system. However, in this discussion it is
assumed that biogenic emissions are processed separately from anthropo-
genic emissions, and thus separate categories for biogenic emissions are
not included in this scheme.
Cnt Alkanes: The RADM mechanism uses three different species to
represent the C^+ alkanes. Other mechanisms use fewer species for this
purpose, and unlike the case for the alkenes, the- mechanisms generally
differ on how the lumped groups of alkanes are defined in the model. To
allow for maximum flexibility in use with mechanisms with differing
aggregation schemes for the alkanes, four different reactivity groups are
used for the emitted C^+ alkanes, with (as indicated above) methane,
ethane, and propane being in groups by themselves.
The criterion used in this aggregation scheme to determine the reac-
tivity classifications for the alkanes is based solely on how rapidly they
react in the atmosphere. This is an important factor in measuring reac-
tivity, since it determines how much of the alkane which is emitted under-
goes reaction and thus participates in chemical transformations. However,
the rate of reaction is not the only factor affecting the reactivity of
alkanes; aspects such as the alkyl nitrate yield in the reaction is also
important (Carter and Atkinson, 1985, 1989). Nevertheless, at least to
some extent the nitrate yields and other aspects are correlated with the
size of the molecule, which in turn is correlated with how rapidly it
reacts.
Alkanes are consumed in the atmosphere primarily by reaction with OH
radicals, but the rate at which they react can vary significantly depend-
ing on the size of the alkane, and to a lesser extent on its structure.
For example, the rate constants for n-butane and n-pentadecane, both
important C4+ alkanes in the NAPAP inventory, have OH radical rate
constants which differ by approximately an order of magnitude. 2,2-
Dimethylbutane and cyclohexane have the same number of carbons, yet the
latter reacts more than twice as rapidly as the former. Most other C^
alkanes with significant emissions in the NAPAP inventory have rate
constants between those of n-butane and n-pentadecane.
66
-------
Based on an analysis of the amounts of alkanes with various ranges of
rate constants in the available NAPAP total U.S. inventory, the four C^
alkane reactivity classes are defined as follows (where kOH is the OH
radical rate constant at T=300 K in units of ppm min"1):
(1) The group designated "alkanes (<0.5 react)," consists of all
alkanes where kOH is less than 5 x 1CT ppm" min . This group is
dominated by n-butane and isobutane, but also contains 2,2-dimethylbutane.
(2) The group designated "alkanes (0.5-1 react)," contains alkanes
with kOH between 0.5 and 1 x 10 ppm min . It is dominated by pentanes
and hexanes. This is the second largest group of alkanes in the NAPAP
inventory in terms of estimated amount reacted in regional model simula-
tions (see below).
(3) The group designated "alkanes (1-2 react)" consists of alkanes
U 1 1
with kOH between 1 and 2 x 10 ppm min"1. It is dominated by n-heptane
and other C^-C^Q isomers. This is the largest group of alkanes in terms
of estimated amount reacted in regional model simulations.
(4) The group designated "alkanes (>2 react)" consists of alkanes
with kOH greater than 2 x 10 ppm min. Almost HQ% of the emissions of
this group is accounted for by n-pentadecane, with the remainder being
various C11+ acyclic or CQ+ cyclic alkanes. Although this is the smallest
group of Cjj+ alkanes in terms of estimated amount reacted, it still makes
a non-negligible contribution to the emissions, consisting of over 3£ of
the mass of VOC emitted.
Aromatics: The aromatics, like the alkanes, are consumed in the
atmosphere primarily by reaction with OH radicals, and it is the rate of
this reaction which was used to define their reactivity classes. However,
in terms of distribution of reaction rates and fractions of emitted
aromatics estimated to react in regional model simulations, the aromatics
tend to fall into three distinct groups. These groups also correspond
reasonably well with how they are aggregated in current lumped kinetic
mechanisms. These reactivity classifications are as follows:
(1) The lowest reactivity group is dominated by benzene, but also
includes several halobenzenes which are also in the NAPAP inventory. The
halobenzenes react somewhat slower than benzene, but not so slowly that
they should be considered unreactive. Since there are not enough halo-
benzenes emitted to Justify placing them in their own aggregation group,
they are lumped with benzene.
67
-------
(2) The middle reactivity aromatics group is designated "aromatics
(<2 react)," and is defined as those aromatics which react with OH
radicals more rapidly than benzene, but with rate constants of less than 2
x 10 ppm min . The compounds in this group have a relatively narrow
range of kOH values, being around 0.8-1.2 x 10 ppm min , and consist
entirely of toluene and other monoalkyl benzenes. Toluene is the dominant
compound in this group, accounting for over 80% of the mass of such
compounds emitted in the NAPAP inventory.
(3) The most reactive group of aromatics is designated "aromatics
(>2 react)," and is defined as those aromatics which have kOH greater than
2 x 10 ppm" min . This group consists primarily of di- and poly-alkyl
benzenes, but it also includes naphthalenes and indans, which account for
approximately 5J of the mass emissions into this group. This is the
largest group of aromatics, both in terms of mass emissions and in terms
of estimated amounts reacting in a model simulation. Almost 10/t of the
estimated number of moles of VOC reacted in a model simulation corresponds
to compounds in this group.
It should be noted that had it been practical to include more than 32
groups in this total aggregation scheme, the most reactive aromatics group
would have been broken down further with dialkylbenzenes, tri- and poly-
alky Iben2enes, and naphthalenes and indans each being in their own groups.
This would permit the scheme to be used with the most detailed version of
the SAPRC/ERT mechanism described by Lurmann et al. (1987), which has
separate model species for xylenes and trialkylbenzenes. In addition,
smog chamber studies (Carter et al., 1987) indicate that the naphthalenes
are much less reactive in N0v-air irradiations than the di- and poly-
A
alkylbenzenes, despite their high OH radical rate constants. However,
this grouping corresponds to the lumping schemes used in the RADM and
other chemical mechanisms in current regional models, which generally lump
all polyalkylbenzenes with the xylenes. The emissions of the naphthalenes
and the indans are not sufficient to justify creating a separate group for
them in this 32-class scheme.
"Others": Compounds such as alcohols, ethers, alcohol ethers, and
other compounds (primarily 0-containing species which are not aldehydes,
ketones, organic acids, and which do not contain aromatic rings or carbon-
carbon double bonds) are lumped into the miscellaneous group designated
68
-------
"others." All these compounds are believed to react in the atmosphere
primarily with OH radicals, and current kinetic mechanisms (including
RADM) generally lump them with the alkanes. There is an insufficient
number of available groups in the 32-class scheme to represent the variety
of chemical types of this group separately, so they are aggregated into
four reactivity classes based on their OH radical rate constants, using a
method similar to that used for the alkanes. These categories are as
follows:
(1) Compounds with OH radical rate constants less than 2.5 x 10^
ppm rain"' are lumped into the group designated "others (<0.25 react)."
The major compounds in this group are methyl and ethyl acetates (-51J of
the emitted mass in this group) and methanol (-40% of the emitted mass).
Emissions of this class are relatively minor in the current NAPAP total
U.S. inventory (consisting of less than 0.4/1 of the mass emitted and less
than Q.2% of the estimated moles reacted), but these compounds are too
reactive to be treated as "inert," and too unreactive to be lumped into
the next highest reactivity "others" group. In addition, in view of the
interest in using methanol as an alternative fuel, emissions into this
class may become much more important in the future.
(2) Compounds with kOH values between 2.5 and approximately 5.0 x
10^ ppm"1 min"1 are lumped into the group "others (0.25-0.5 react)." This
group is dominated by ethanol, which accounts for over 85? of the mass of
this group in the NAPAP total U.S. inventory. It is also non-negligible
in terms of total emissions, accounting for almost 4% of the estimated
moles of VOC reacted in the model simulations. [Subsequent to the
development of this classification system, we revised our assignment of
the OH radical rate parameters for ethanol to be consistent with the most
recent recommendation of Atkinson (1989). This yields a 300 K rate
constant which is slightly higher (by less than }%) than the nominal upper
limit rate constant for this group. However, ethanol was not re-assigned,
since otherwise there would be no significant emissions for this group.]
(3) Compounds with kOH values between approximately 0.5 and 1.0 x
10 ppm min are lumped into the group "others (0.5-1 react)." The
major compounds in this group are isopropyl alcohol (-60J of the mass) and
propyl and higher acetates (-25%). This group accounts for an estimated
4-5/t of the VOC mass reacted in model simulations.
69
-------
(4) Compounds with kOH values higher than 1.0 x 1CT ppm'1 min~ are
lumped into the group "others (>1 react)." The major compounds in this
group consist of glycols and cellosolves and other alcohol ethers. [It
also has very minor contributions by alkynes, allenes, and amines, which
react within this rate constant range. In our opinion, these are not well
represented by any of the other chemical classes in this aggregation
scheme (for example, the alkynes and allenes are not lumped with the
alkenes since they do not react significantly with ozone), and are not
emitted in sufficient quantities to justify being in their own group. ]
The emissions of compounds in this group are small but non-negligible,
accounting for approximately 1* of the mass emitted and 1-2* of the
estimated moles reacted in the total U.S. inventory.
Summary of Emissions Groups: Table 7 gives a summary of all the
groups in this 32-class aggregation scheme, and the total mass and molar
emissions of compounds in each group of the current NAPAP inventory for
total anthropogenic emissions in the contiguous United States. Note that
the detailed emissions data bases give mass emissions, but these are
aggregated into the 32-class scheme on a molar basis, with the number of
moles of each type of emitted compound being calculated based on its
molecular weight to determine its contribution to the aggregated emissions
group. The table therefore also shows the percent contribution of each
group to the total mass and molar emissions. Finally, to give an indica-
tion of the relative importance of the emissions of each category on model
simulations, Table 7 also gives estimates of the number of moles of
compounds in each group that is estimated to undergo chemical reaction in
a 2-3 day RADM simulation. These latter estimates were made as indicated
in footnotes to the table, and as discussed in the following section.
It can be seen that the most important group in terms of moles
emitted is methane, accounting for almost kQl of the moles emitted in the
inventory, but it accounts for less than 0.2% of the moles reacted. In
terms of moles reacted, the most important single group is ethene,
accounting for almost 21/1 of the moles reacted, followed by the middle
reactivity C^+ alkane group and the high reactivity group of aromatics.
Alkanes, aromatics, and alkenes account for -80J, aldehydes and ketones
for -1%, and all other types of compounds for -11J of the total estimated
number of moles of VOC reacted. As indicated in Table 7, approximately 1%
70
-------
Table 7. Summary of the Emissions Groupings Using the 32-Class Scheme,
Amounts of Each Emitted in the NAPAP Total U.S. Inventory, and
Estimated Amounts Reacted in Conditions Representative of RADM
Applications
C1-C2 Alkanes
Methane
Ethane
Total
Higher Alkanes
Propane
Alkanes (<0.5 react)
Alkanes (0.5-1 react)
Alkanes (1-2 react)
Alkanes ( >2 react)
Alkane/ Aromatic Mix (b)
Total
Ethene
Higher Alkenes
Propene
Alkenes (primary)
Alkenes (internal)
Alkenes (mix)
Total
Aromatics
Benzene, Halobenzenes
Aromatics (<2 react)
Aromatics (>2 react)
Phenols and Cresols
Styrenes
Total
Aldehydes and Ketones
Formaldehyde
Higher Aldehydes
Acetone
Higher Ketones
Total
Miscellaneous
Organic Acids (c)
Acetylene
Haloalkenes
Total
M.Kg
2.83
0.46
3.29
0.12
1.40
2.03
2.49
0.60
0.25
6.89
1.27
0.44
0.33
0.76
0.20
1.73
0.60
1.56
2.03
0.05
0.08
4.31
0.32
0.17
0.21
0.20
0.89
0.04
0.36
0.52
0.92
A rn/*"M i n V c
rulivJlln Ui
12.18
1.98
14.16
0.53
6.01
8.72
10.71
2.60
1.06
29.63
5.45
1.91
1.42
3.28
0.85
7.46
2.59
6.69
8.72
0.19
0.36
18.55
1.37
0.71
0.90
0.85
3.83
0.17
1.56
2.24
3.97
iPm i f ^ ^H
dlu A. w v\?U
K.Mole
176.7
15.3
192.0
2.8
23.9
25.5
21.9
3.1
1.8
78.9
45.2
10.6
4.6
11.0
1.9
28.0
7.3
16.1
16.3
0.3
1.5
41.5
10.7
2.5
3.6
2.2
19.0
0.3
14.0
2.1
16.4
*
38.42
3.33
41.75
0.61
5.20
5.53
4.76
0.67
0.38
17.16
9.82
2.29
1.00
2.38
0.41
6.09
1.58
3.51
3.54
0.06
0.33
9.02
2.32
0.54
0.79
0.48
4.12
0.07
3.04
0.46
3.57
PAO /%f- Arl
rvcaC wCU
K.Mole
0.2
0.7
0.9
0.5
8.0
13.2
16.7
2.8
1.6
42.9
33.9
10.5
4.6
11.0
1.9
27.8
1.4
10.2
16.1
0.3
1.0
29.0
8.2
2.3
0.1
0.3
11.0
0.0
1.7
0.1
1.8
( a ">
V a )
n
0.15
0.41
0.55
0.31
4.86
8.06
10.19
1.73
0.98
26.13
20.65
6.37
2.78
6.68
1.13
16.96
0.84
6.23
9.81
0.17
0.58
17.64
4.98
1.41
0.08
0.20
6.68
0.00
1.01
0.04
1.05
(continued)
71
-------
Table 7 (continued) - 2
Emissions Group
Amounts Emitted
M.Kg % K.Mole
Reacted (a)
K.Mole %
"Others"
Others (<0.25 react)
Others (0.25-0.5 react)
Others (0.5-1 react)
Others (>1 react)
Total
Unreactive
Unknown/Unassigned
Unidentified
Unassigned
Total
0.10
0.80
0.88
0.30
2.08
0.35
0.42
3.44
3.76
1.28
8.90
1.50
2.0
16.7
12.5
3.3
34.7
4.3
0.43
3.63
2.73
0.72
7.53
0.94
0.3
6.3
7.7
2.7
16.9
0.0
0. 17
3.85
4.67
1.66
10.34
0.00
1.36
0.16
1.52
5.83
0.72
6.55
(a) Estimated assuming reaction with OH radicals is the main
consumption process for compounds in the groups, and assuming that
the ratio of reacted to emitted compounds can be estimated by
1 - exp (-kOH x INTOH), where kOH is the average OH radical rate
constant for compounds in the group, and INTOH is a parameter
reflecting the effective integrated OH radical levels in the model
simulation (Carter and Atkinson, 1989). The tabulated values are
calculated assuming INTOH =110 ppt-min, based on analyses of
results of previous RADM model simulations (Stockwell, private
communication, 1988), and the kOH values are weighed averages of
those assigned for compounds in the groups (Carter, 1988). See
Section 3.2.3.
(b) These are estimated to consist of approximately 90? alkanes and 10J
aromatics (on a molar basis), so for the purpose of this summary
they are counted as alkanes.
(c) Amount reacted using kOH of acetic acid, the major emitted organic
acid in the NAPAP inventory.
72
-------
of the mass in the current NAPAP total U.S. inventory is either unknown or
unassigned.
3.2.2 Recommended Extensions to the VQC Emissions Classification
System
The 32-class VOC classification system described in the
previous section is the version which is being implemented for use with
the RADM-II model. Although we consider this system to be satisfactory
for this purpose, we have several recommendations, discussed below, for
extending its utility for use with other mechanisms besides that which
will be used in RADM-II, and for other types of models. All of these
recommendations require updating the software currently used for aggre-
gating detailed emissions data.
Use with the Carbon Bond Mechanism. The emissions aggregation system
discussed above was designed primarily for use with mechanisms, such as
that used in RADM, which represent reactions of emitted VOCs on a molecu-
lar level, where, in general, only one model species is used to represent
the reaction of any type of emitted molecule. However, the Carbon Bond
mechanism (Whitten et al., I960; Gery et al., 1988), which is widely used
in urban airshed modeling applications, represents portions of some types
of molecules reacting independently, and thus requires at least some
portions of the emissions to be aggregated in terms of reactive portions
of molecules, rather than molecules as a whole. For that reason, this 32-
class system, as described above, cannot be used in aggregating emissions
input for models using the Carbon Bond mechanism. However, as indicated
below, this limitation can be readily removed by adding two new emissions
categories.
Although as originally conceived the Carbon Bond mechanism was
radically different from molecularly-based mechanisms in its treatment of
parts of molecules as reacting independently, in recent years the need for
improved performance in simulating results of chamber experiments has
required it to evolve more towards the molecularly-based mechanisms, with
the latest version (Gery et al., 1988) using almost the same numbers and
types of species to represent alkenes, aroraatics, and oxygenates as other
state-of-the art mechanisms. The principal remaining difference between
the current version of Carbon Bond and the molecular-based mechanisms, is
the use of a single-carbon model species called "paraffin bond" (PAR) to
73
-------
represent both the alkanes and the "extra" carbons on molecules which have
more carbons than the species in the model representing them. In addi-
tion, the double bonds in some diolefins and other bifunctional compounds
are represented as reacting independently. The present version of the
emissions aggregation system does not contain the information necessary to
determine the levels of the PAR model species or the contributions of
bifunctional compounds to levels of alkene model species.
The deficiency with respect to aggregation of the "paraffin bond"
species can be addressed by adding a separate category designated "reac-
tive carbon," and using it to tabulate the total number of carbons in
reactive emitted species. (The specific criterion used for "reactive" in
this context would be determined by the criterion presently used when
assigning contributions to PAR when aggregating emissions for Carbon
Bond.) The levels of PAR emissions to use in the model could then be
determined by subtracting from this number the number of carbons in the
other model species used to represent the emissions. For example, if a
simplified emissions inventory contained 1 mole each of n-butane, ethanol,
and butyl benzene, then the corresponding emissions aggregation would
include 1 mole each of "alkanes (0.25-0.5)," "others (0.25-0.5 react),"
"Aromatic (<2 react)," and 16 moles of the special "reactive carbon" group
(4 moles from n-butane, 2 from ethanol, and 10 from butyl benzene). In
models using the Carbon Bond mechanism, the emissions group "Aromatic (<2
react)" would be represented by the seven-carbon species "TOL," and the
"PAR" levels would be calculated by subtracting these 7 moles of reactive
carbons already represented in TOL from the total of 16 moles of reactive
carbon, yielding 9 moles of "PAR." Since the "alkanes" and "others"
emissions in this inventory are already represented by "PAR," the aggrega-
tions in these groups are not used in determining input for models using
this mechanism.
An analogous method could be used to account for the extra double
bonds in poly functional alkenes. In this case, it is recommended that
total terminal double bonds and total internal double bonds be aggregated
into separate groups, since these are represented differently in the
Carbon Bond mechanism.
In view of the extensive use of the Carbon Bond mechanism, it is
perhaps surprising that this emissions aggregation system was not extended
-------
so it could be used for this mechanism. The principle reason for this
appears to be that the software presently used to process the NAPAP
emissions input data cannot be readily adapted to be able to distribute
more than one detailed emissions class into more than one aggregated
emissions category. Aggregation for Carbon Bond obviously requires this
capability. This artificial limitation would presumably be addressed when
the emissions processing software is upgraded.
Aggregation of Aromatics. The present 32-class scheme lumps the
xylenes and other dialkylbenzenes, tri- and polyalkyl benzenes, and
naphthalenes into a single group ("Aromatics >2 react"). This is satis-
factory for use with the RADM and most other lumped mechanisms, which only
uses a single model species (generally designated XYL) for these com-
pounds. However, the most detailed version of the SAPRC/ERT mechanism
(Lurmann et al., 1987) has a separate model species for the tri- and poly-
alkylbenzenes, and thus this grouping is not adequate for use with this
mechanism. Splitting the "Aromatics (>2 react)" group into "Aromatics (2-
4 react)" and "Aromatics (>4 react)," with the trialkylbenzenes and other
aromatics which react faster than 4 x 10 ppm min being lumped into
the.latter group, would address this problem. Both of these groups would
be significant in terms of amounts of emitted species, and thus aggre-
gating them separately would not be inappropriate on that basis.
It would also be desirable to aggregate the naphthalenes and other
tri- and poly-cyclic aromatics in separate groups from the alkylbenzenes.
Environmental chamber and modeling studies carried out at SAPRC indicate
that these species are much less reactive than alkylbenzenes, despite
similar or higher OH radical rate constants (Carter et al., 1987). There-
fore, lumping them with alkylbenzenes will result in models significantly
overpredicting their reactivity in model simulations. On the other hand,
the naphthalenes are relatively unimportant in terms of amounts emitted,
contributing only approximately 0.4jt of the mass emissions in the NAPAP
total U.S. anthropogenic inventory. However, there may be special urban
pollution cases where their contributions may be more important.
Use with Global and Stratospheric Models. As indicated above, this
aggregation scheme was developed for use with local and regional oxidant
and acid deposition models, and additional classes would have to be added
for it to be used for other types of modeling applications. For example,
75
-------
global models may require more distinction among less reactive compounds,
and stratospheric models would require halogenated compounds to be split
out from the "unreactive" category, or perhaps separate categories tabu-
lating the total numbers of Cl and Br atoms which reach the stratos-
phere. A discussion of the input requirements for such models is beyond
the scope of this report.
3.2.3 Aggregation of Emissions into the RADM Mechanism
The gas-phase RADM mechanism contains 17 different model
species which can be used to represent the reactions of various types of
emitted reactive VOCs. These include five species for alkanes, four for
alkenes, two for aromatics, two for aldehydes, one for ketones, one for
phenols, and two for organic acids. As indicated above, the system
developed for processing anthropogenic emissions input for the RADM model
involves a two-step procedure where first the detailed emissions data,
given in terms of mass emissions of over 550 compounds, are aggregated
into molar emissions of 32 compounds judged to be similar in reactivity,
and then these are further aggregated into molar emissions of the 16
species used in RADM. As part of this program, SAPRC was asked to make
recommendations for how to aggregate emissions organized into the 32-class
system into the species in the RADM mechanism. The system that was
recommended, which we anticipate will be used in RADM, is summarized in
this section.
The allocation of the 32 emissions groups to species in the RADM
mechanism is summarized in Table 8. In many cases these allocations are
reasonably obvious; cases that are less obvious are discussed below for
the various types of chemicals. As indicated in the table, most of the
emissions groups are represented by the RADM species on a simple mole-per-
mole basis, with one mole of RADM species being used for each mole of
compound emitted. However, in some cases, particularly those where
emissions groups of significantly different reactivities have to be lumped
together, "reactivity weighing" is employed. This involves use of a reac-
tivity weighing factor in calculating the number of moles of model species
from the moles of emitted species. The general principle of reactivity
weighing will be discussed first, followed by a discussion of specific
considerations involved for the various types of chemicals.
76
-------
Table 8. Assignments of Emissions Groups to RADM Model Species
Emissions Class Mix
____ Fao
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
Description (a)
Methane
Ethane
Propane
Alkanes (0.25-0.5 react)
Alkanes (0.5-1 react)
Alkanes (1-2 react)
Alkanes (>2 react
Alkane/Aromatic Mix 0.91
0.09
Ethene
Propene
Alkenes (Primary)
Alkenes (Internal)
Alkenes (Prim/Inter Mix) 0.50
0.50
Benzene, Halobenzenes
Aromatics (<2 react)
Aromatics ( >2 react)
Phenols and Cresols
Styrenes 1.0
1.0
Formaldehyde
Higher Aldehydes
Acetone
Higher Ketones
Organic Acids
Acetylene
Haloalkenes
Unreactive
Others (<0.25 react)
Others (0.25-0.5 react)
Others (0.5-1. react)
Others (>1 react)
RADM
Class
CH4
ETH
HC3
HC3
HC5
HC8
HC8
HC8
XYL
OL2
OLT
OLT
OLI
OLT
OLI
TOL
TOL
XYL
CSL
TOL
OLT
HCHO
ALD
KET
KET
ORA2
HC3
HC3
_
HC3
HC3
HC5
HC8
Weighing Factor (b,c)
Factor
1.0
1.0
0.519
0.964
0.956
0.945
1.141
1.002
0.090
1.0
1.0
1.0
1.0
0.500
0.500
0.293
1.0
1.0
1.0
1.0
1.0
1.0
1.0
0.253
1.0
1.0
0.343
0.078
1.0
0.404
1.215
1.075
1.011
k(emit)
0.180
0.369
0.668
1.315
2.339
2.018
0.185
0.035
0.115
0.025
0.137
0.496
0.801 '
1.552
k(RADM)
0.386
0.386
0.715
1.507
1.507
1.507
0.902
0.147
0.386
0.386
0.386
0.386
0.715
1.507
(a) Molar distribution of RADM species used for groups represented by
more than one RADM species.
(b) "Factor" is number of moles of model species used to represent
the emissions group. If reactivity weighing is not employed, the
factor is 1.0 or "Mix Fac.," if applicable. If reactivity weighing
is employed, these are given by
(continued)
77
-------
Table 8 (continued) - 2
(c)
Factor = Mix Fac. (if applicable)
x (1-exp[-k(emit)xINTOH])/(1-exp[-k(RADM)xINTOH])
where k(emit) and k(RADM) are respectively the 300 K OH rate
constants for the emissions group and the RADM species (given in
the table in units of 10 ppm"1 rain"1), and INTOH is assigned a
value of 110 ppt-min. See text.
The tabulated k(RADM) values for the lumped alkane model species
are for the RADM-P mechanism. The March 1988 RADM mechanism used
k(RADM) values for HC3, HC5, and HC8 of 0.376, 0.786, and 1.655,
respectively.
78
-------
Reactivity Weighing. Reactivity weighing is used as a means to
represent groups of compounds with similar reaction mechanisms, but with
significantly differing reaction rates, with a single model species. It
is based on the assumption that, at least for compounds with similar reac-
tion mechanisms, the effects of emissions of different compounds on
results of a model simulation is approximately proportional to the amount
of compound which undergoes chemical reaction in the simulation. For
example, if compound "A" reacts rapidly and essentially is all consumed in
a model simulation, while compound "B" reacts more slowly and only half is
consumed, then, under this approximation, emissions of 1 mole of "B" has
the same effect as emissions of 0.5 moles of "A." If this is assumed,
then the appropriate amount of model species to use to represent a given
amount of emitted species would be such that the estimated amounts of each
which react in the simulation are the same. Thus, for example, if an
emissions group E is to be represented in a model simulation by model
species M, then the reactivity weighing factor for E would be derived such
that:
Moles of M Reacting
in Simulation
Moles of E Reacting
in Simulation
or
Moles of M
used to
Represent E
Fraction
x of M which =
Reacts
Moles of
E Emitted
X
Fraction
of E which
Reacts
Therefore,
Reactivity
Weighing
Factor for E
Moles of M
used to
Represent E
Fraction
of E which
Reacts
(I)
Moles of E
Emitted
Fraction
of M which
Reacts
Thus, for example, if half of the emitted species is estimated to react
under the conditions of a model simulation, and these species are repre-
sented by a more rapidly reacting model species where 75? is estimated to
react under the same conditions, then the appropriate reactivity weighing
factor would be 0.5/0.75 = 0.667.
79
-------
The use of this reactivity weighing approach requires an ability to
estimate fractions of emitted species which undergo reaction in represen-
tative model simulations. Such estimates are relatively straightforward
for compounds which react in the atmosphere primarily with hydroxyl
radicals, as is the case for most of the types of compounds where reac-
tivity weighing is employed. For such compounds, we have shown that the
fraction of compound which reacts in an airshed model simulation can be
fit reasonably well by the following (Carter and Atkinson, 1989):
Moles Compound
Reacted - kOH x INTOH
= ( 1 - e ) (II)
Moles Compound
Emitted
where kOH is the OH radical rate constant and INTOH is a measure of the
effective integrated OH radical levels under the conditions of the model
simulation. (The above equation is exact if all the VOC is present at the
beginning of the simulation, but is approximate otherwise.) If this
approximation is employed, and if an emissions group E is to be represent-
ed by the model species M, then combining Equations (I) and (II) yields:
- kOH(e) x INTOH
Reactivity Moles of M ( 1 - e )
Weighing = Representing = (III)
Factor one mole of E - kOH(m) x INTOH
for E ( 1 - e )
where kOH(e) is the average OH radical rate constant for species in
emissions group E, and kOH(m) is the OH radical rate constant used in the
gas-phase mechanism for model species M. Note that if E and M are both
rapidly reacting, then the reactivity weighing factor is always unity (and
thus in effect reactivity weighing need not be employed); while if they
both react slowly, then the reactivity factor is proportional to the ratio
of the rate constants, and independent of the estimated INTOH parameter.
The appropriate value to use for INTOH in derivation of the reac-
tivity weighing factors will depend on the conditions of the model simula-
tion, since it is dependent on both the overall OH radical levels in the
simulation, and the length of time the simulation is carried out. For
80
-------
RADM simulations lasting 2-3 days, Stockwell and co-workers (Stockwell,
private communication, 1988) estimate that an INTOH value of 110 ppt-min
is appropriate. Therefore, that value is used in deriving our recommended
weighing factors. Note that as indicated above the estimates of INTOH do
not affect weighing for very rapidly or very slowly reacting compounds,
but they do affect weighing factors for compounds of intermediate reac-
tivity. A limited number of test calculations we have carried out with
one-cell box models suggests that in practice the results of model simula-
tions will not be highly sensitive to estimates of INTOH provided they are
within reasonable limits. However, the recommended factors in Table 8 can
be readily re-calculated for use in processing VOC emissions input for
model simulations where an INTOH of 110 ppt-min is inappropriate. Smaller
values may be more appropriate for 1-day urban simulations, while larger
values may be appropriate for simulations of extended time periods.
The major parameters affecting the reactivity weighing factors are
the OH radical rate constants assigned to the emissions groups and to the
model species. The average T=300 K OH radical rate constants calculated
for the emissions groups where reactivity weighing is recommended, and for
the model species used to represent them in the RADM mechanism, are shown
in Table 8. The former were calculated using the distributions of
emissions in the current NAPAP total U.S. inventory, together with our
assignments of OH radical rate constants for the various types of emitted
compounds (Carter, 1988). The rate constants for several of the lumped
model species in the RADM gas-phase mechanism vary depending on the
version of the mechanism, since these were modified as a results of the
recommendations made during this program. The table gives the rate
constants for the version employing our final recommended values for the
lumped model species which are discussed in the following section, since
it is our understanding that our recommendations in this regard will be
implemented into RADM.
This reactivity weighing approach is clearly not appropriate for
lumping together species with different types of reaction mechanisms,
since different types of mechanisms have been shown to have significantly
different effects on results of model simulations, even if differences in
reaction rates are taken into account (Carter and Atkinson, 1989).
Furthermore, reactivity weighing is an approximation even for species with
81
-------
very similar reaction mechanisms, since differences in rates of reaction
affect the distribution of where and when the compound reacts in the
simulation. Rapidly reacting compounds react closer to the source areas,
and slower reacting compounds have greater effects on remote regions. For
this reason, it is not a good approximation to lump together species with
very large differences in reaction rates. In addition, since the frac-
tions of emitted compounds or model species which react will depend on the
conditions of the model simulation, the weighing factors appropriate for
one set of conditions may not necessarily be appropriate for others.
Clearly, the use of separate model species for compounds with appro-
priate reaction rates would be preferable to using reactivity weighing.
However, when the number of species which can be used in the model is
limited, as is the case for regional modeling applications, the use of
reactivity weighing is preferable to options which are even more approxi-
mate. In particular, it is preferable to the options of either ignoring
the effects of less reactive compounds for which there are no correspond-
ing model species, or of representing emitted compounds with equal number
of moles of model species of significantly different reactivity.
Representation of Alkanes and "Others" in RADM. The RADM mechanism
employs five model species to represent reactions of the alkanes and
species which are assumed to have alkane-like reaction mechanisms. Two of
these species (CHM and ETH) are used exclusively for methane and ethane,
and the remaining three (HC3, HC5, and HC8) are lumped species represent-
ing the higher alkanes and the various types of species which are judged
to be better represented by alkane model species than other species in the
RADM mechanism. The nomenclature for HC3-HC8 is derived from the average
number of carbons associated with each, as discussed in the following
section.
As shown in Table 8, the miscellaneous groups of compounds aggregated
into the "others" emissions groups are represented in RADM by lumped
alkane model species. These "others" consist primarily of alcohols,
ethers, and esters, though a variety of other reactive compounds, which
are not considered appropriate for the other emissions groups, are
included. Although except for the simple alcohols the reaction mechanisms
of these compounds are uncertain, the major compounds in these groups are
82
-------
believed to react analogously to alkanes, reacting primarily with OH
radicals, and probably forming similar types of oxygenated products.
Alkane model species are also used to represent acetylene and the
haloalkenes. Although these are unsaturated compounds, the use of alkene
model species to represent them is considered inappropriate because they
react much slower with OH radicals than any of the alkene model species
(Atkinson, 1986), and because their reaction with ozone is relatively much
less important than is the case for most alkenes (Atkinson and Carter,
1984). They are thus lumped as alkanes by default because no other RADM
model species is appropriate for them. The specific reaction mechanisms
for these compounds are highly uncertain.
The allocation of lumped higher alkane model species to the various
alkane and "others" reactivity groupings is shown in Table 8. These
allocations were made such that each alkane emissions group (other than
the "alkane/aromatic mix" group) is represented by only one RADM model
species. In addition, the allocations were determined, as much as
possible given the emissions groupings, (a) to minimize the range of
reactivities of emissions groups represented by RADM species, and (b) to
have the three higher lumped alkane RADM groups similar in their estimated
numbers of moles reacted when representing the NAPAP U.S. inventory in
RADM simulations. Since all three of the higher alkane RADM species
represent two or more emissions groupings of differing reactivities, reac-
tivity weighing is used in determining numbers of moles of RADM species
representing these groups.
The group "alkane/aromatic mix" is used to represent the emissions in
the NAPAP inventory which are characterized as mixtures with designations
such as "mineral spirits," "naphtha," etc. Based on an analysis of
detailed composition data for one type of mineral spirits (Weir et al.,
1988), we estimate that these consist of approximately 91 % Cy and higher
alkanes, with the remainder being aromatics. The average OH radical rate
constant for the alkanes in this group is within the range used for the
RADM species HC8, and thus these 91/1 of the molar emissions lumped in the
"alkane/aromatic mix" group are represented by HC8, with reactivity weigh-
ing being employed. However, the extent to which the analysis used as a
basis for this assignment is appropriate in representing the actual
83
-------
emissions lumped into this group is highly uncertain. This is an area
where the emissions inventory needs to be improved.
Representation of Alkenes in RADM. The RADM mechanism employs four
species to represent alkenes. Ethene (ETH) and isoprene (ISO) are repre-
sented explicitly, OLT represents propene and those alkenes which are
lumped into the "alkenes (terminal)" emissions group, and OLI represents
those lumped into "alkenes (internal)." As indicated above, the "alkenes
(terminal)" group is defined as those alkenes, other than ethene, which
U -1
react with OH radicals with rate constants less than 7.5 x 10 ppm
min"1 and includes most higher alkenes with only terminal double bonds.
The "alkenes (internal)" group consists of alkenes, other than isoprene,
which react with higher OH radical rate constants and consists of internal
alkenes, dialkenes, and terpenes.
Other than isoprene and the "alkene (mixed)" emissions groups, which
are special cases discussed below, the allocation of alkenes in the RADM
mechanism corresponds almost exactly with that used in the emissions
grouping. The only other exception is that, because of its high
emissions, propene is treated separately in the emissions grouping, while
in RADM it is lumped with the other "primary" alkenes. These correspond-
ences are indicated in Table 8. As indicated there, no reactivity weigh-
ing is employed for alkene species. This is because the only case where
more than one emissions group is aggregated into a RADM species are those
represented by OLT, and these compounds react so rapidly that the reac-
tivity weighing factor would be near unity (see Table 7).
The "alkene (mixed)" group consists of those emissions which are
identified in the emissions inventories with terms such as "isomers of
pentene," "C-iQ olefins," etc. It is not known to what extent these
emissions consist of terminal alkenes (represented by OLT in RADM) or
internal alkenes (represented by OLI), and we arbitrarily assume that they
consist of equal amounts of each. Note, however, that in making alloca-
tions for the Carbon Bond and the CAL (SAPRC/ERT) mechanisms, Jeffries et
al. (1989c) arbitrarily assumed that they were all terminal. This is
another area where work is needed to improve the quality of the emissions
inventories.
Because of the importance of isoprene in biogenic emissions, combined
with the importance of biogenic emissions in regional model simulations,
-------
isoprene is treated explicitly in the RADM mechanism. However, its
emissions are essentially negligible in the NAPAP anthropogenic inventory,
and thus the RADM isoprene species (ISO) is not used in representing any
of the anthropogenic emissions. (The small amount of isoprene in the
anthropogenic inventory, consisting of only 0.001% of the mass emissions,
is represented by OLI.) Thus, the RADM model species ISO is used only for
biogenically emitted isoprene.
On the other hand, there is no RADM model species for terpenes,
despite the fact that these are comparable in importance as isoprene in
the biogenic inventory (Middleton, private communication, 1989). Thus,
the biogenically emitted terpenes are represented by the same model
species, OLI, which is used to represent anthropogenically emitted alkenes
of similar reactivity. However, despite the potential importance of
terpenes in regional model simulations, it is not clear whether there is
any significant advantage to representing them separately in the model.
Relatively little is known about their reaction mechanisms, other than
their initial rates of reaction, which are in the range appropriate for
OLI. Note that the model simulations of the chamber experiments
(discussed in Section 5) indicate that representing at least one terpene,
alpha-pinene, by OLI may not be inappropriate. As far as we are aware,
there are no chamber data available to indicate how appropriate OLI is in
representing the other terpenes.
Representation of Aromatics in RADM. The RADM mechanism uses three
species to represent aromatic compounds. The most important of these are
TOL, representing toluene, other monoalkylbenzenes, and lesser reactive
aromatics, and XYL, representing xylenes, other di- and polyalkylbenzenes,
and other aromatics which react more rapidly than the xylenes. In addi-
tion, the RADM mechanism includes the model species CSL, primarily to
represent the phenol and cresol products formed in the reaction of TOL and
XYL, but which can also be used to represent phenols and cresols which are
emitted directly.
The allocations of the aromatic emissions groups to these three
aromatic model species in RADM are shown in Table 8. Since there is no
separate model species for benzene, the "benzene/halobenzene" emissions
group is represented by the RADM species TOL, with reactivity weighing
being employed to account for the differences in reaction rates for these
85
-------
species. The small amount of styrene emissions are represented by two
RADM groups, with TOL being used to represent reaction at the aromatic
group, and OLT being used to represent reaction at the double bond. (This
is the only case where a version of the "lumped structure" approach is
used with the RADM mechanism.) Most of the aromatics emitted are in the
groups "aromatics (<2 react)," and "aromatics (>2 react)," which are
represented by TOL and XYL, respectively. Since the mechanisms used for
TOL and XYL are based primarily on those for the major species in these
two groups (see following section), no reactivity weighing is employed in
these cases.
As indicated above, the group "alkane/aromatic mix" is used to
represent the emissions in the NAPAP inventory that are solvent mixtures
which we estimate to consist of approximately 9% aromatics, with the
remainder being alkanes. The average OH radical rate constant for the
aromatics estimated to be in these mixtures is within the range used for
the RADM species XYL, and thus these 9% of the molar emissions lumped in
the "alkane/aromatic mix" group are represented by this species. No
reactivity weighing is employed because these higher aromatic species
react sufficiently rapidly that the reactivity weighing factors would be
near unity.
Representation of Aldehydes. Ketones, and Organic Acids in RADM. The
RADM mechanism includes the species HCHO for formaldehyde, ALD for higher
aldehydes, KET for ketones, ORA1 for formic acid, and ORA2 for acetic
acid. The aldehyde species corresponds directly to the aldehyde alloca-
tion in the emissions group, as shown in Table 8. The reactions used in
the RADM mechanism for KET are based on estimated mechanisms for methyl
ethyl ketone and higher ketones, and therefore KET is used to represent
the "higher ketone" group directly, without reactivity weighing. There is
no species in the RADM mechanism corresponding to acetone, so it is repre-
sented by the higher ketone model species, KET, using reactivity weigh-
ing. The organic acids are represented by ORA2, as discussed below.
Note that the use of reactivity weighing factors calculated using OH
radical rate constants (as is the case for all those shown in Table 8) is
not strictly valid when lumping acetone with the higher ketones, because
these compounds (especially acetone) are consumed to a significant extent
by photolysis. However, reactivity simulations we carried out using the
86
-------
SAPRC mechanism (Carter, unpublished results, 1988) indicate that acetone
is about 1/3 as reactive as the higher ketones in terms of effects on
ozone formation, which is reasonably close to the reactivity difference
estimated using OH radical rate constants. Since the reactivity weighing
factor obtained using OH radical rate constants is reasonably consistent
with the reactivity differences for the ketones indicated by modeling, the
former is retained in our recommendations for consistency with the deriva-
tions of the factors for the other types of emitted species.
The representation of organic acids is the one case where the RADM
mechanism is more detailed than the 32-class emissions aggregations. The
mechanism has a separate species for formic acid and for higher organic
acids, while the emissions scheme aggregates all organic acids into one
group. However, the emissions of formic acid in the NAPAP total U.S.
inventory is minor, accounting for only 2.5% of the mass of the organic
acid emissions. For that reason, all the organic acid emissions are
represented by ORA2 (higher organic acids) in the mechanism.
3.3 Derivation of Kinetic and Mechanistic Parameters for Lumped Model
Species in RADM
As discussed in the previous section, the RADM mechanism includes a
number of model species which are intended to represent aggregate mixtures
of many different emitted VOCs. These include the lumped alkane species
HC3, HC5, and HC8, the lumped aromatic species TOL and XYL (which, despite
their names, actually represent a mixture of compounds), the lumped alkene
species OLT and OLI, and the lumped oxygenate species ALD, KET, CSL, and
ORA2, which are also used to represent various types of photooxidation
products. As part of the RADM mechanism evaluation study, we1 evaluated
the appropriateness of the reaction mechanisms used for these species to
the specific groups of emitted compound each represents and made several
recommendations in this regard. These recommendations, which have been
adopted for the latest version of the RADM mechanism, are summarized in
this section.
Lumped model species can be represented in molecularly-based mechan-
isms such as RADM either using the "surrogate species" or the "lumped
molecule" approach. In the "surrogate species" approach, the mechanism of
a typical or representative compound in the group is used to represent the
87
-------
reactions of all members of the group. For example, in the RADM mechanism
(as in many others), the mechanism used for the species ALD, which repre-
sents the higher aldehydes, is based on that for acetaldehyde. In the
"lumped molecule" approach, the reactions of a mixture of compounds is
represented by a pseudo-molecule whose mechanisms are derived to represent
an aggregate average of those of the sets of compounds they are used to
represent. The former approach is most useful when one compound dominates
in the lumped group, when there is only one compound in the group for
which there is enough kinetic and mechanistic data available to derive or
estimate a mechanism, or when the distribution of compounds in the group
is unknown. The latter approach has the greatest potential for chemical
accuracy, provided that there is sufficient information available concern-
ing the distribution of compounds represented by the lumped group and
relevant aspects of the reaction mechanisms of those compounds.
In the current RADM mechanism, the reactions of the lumped model
species ALD, KET, CSL, ORA2, TOL, and (in part) XYL are based on the
surrogate species approach, primarily because there is only limited
information available concerning the mechanisms for more than a few
examples of these types of compounds. On the other hand, the reactions
used for the lumped alkane species HC3, HC5, and HC8, the lumped alkene
species OLI and OLT, and (in part) the lumped higher aromatic species XYL
are derived based on the lumped molecule approach. The derivation of the
kinetic and mechanistic parameters for these lumped molecules was carried
out at SAPRC as part of this RADM evaluation study, and were incorporated
into the final recommended version of the RADM mechanism.
Tables 9 through 11 give the averages of the OH radical rate constant
and other mechanistic parameters of relevance to RADM for the emitted
species in the various emissions groups which are lumped as C3+ alkanes
(Table 9), aromatic hydrocarbons (Table 10) and non-ethene alkenes (Table
11). The averages for the emissions groups were derived using the NAPAP
total contiguous U.S. emissions inventory, and were computed based on the
relative number of moles of the various types of species in each emissions
group. No reactivity weighing of any kind was employed in computing these
averages, because reactivity weighing is not employed when the emissions
are aggregated into these groups. The tables also show the total number
-------
Table 9.
Averages of Parameters for Emissions Groups Lumped as CU Alkanes and
for RADM Lumped Alkane Model Species, Derived Using the NAPAP Total
U.S. Emissions Inventory
K Moles
Emit
kOH
(a)
% Ret
(b)
nf
(c)
ON IT X02 HCHO ALD KET
Lumped as HC3
23.9
16.7
14.0
2.8
2.0
2.1
0.
0.
0.
0.
0.
0.
0.
0.
369
496
115
180
137
025
386
376
45.
40.
9.
2.
1.
0.
6
1
5
9
6
3
4
2
2
3
1
2
2
.02
.10
.00
.00
.88
.00
.99
0.07
0.00
0.00
0.04
0.00
0.00
0.03
0.03
0.48
-0.87
-0.30
0.00
-0.63
0.00
-0.17
-0.10
0.16
0.02
0.00
0.00
0.69
0.00
0.09
0.14
0.61
0.98
0.70
0.30
0.12
1.00
0.75
0.69
0.50
0.04
0.00
0.33
0. 19
0.00
0.25
0.26
Alkanes (0.25-0.50)
Others (0.25-0.50)
Acetylene
Propane
Others (<0.25)
Haloalkenes
Recommended HC3 (d)
Initial RADM HC3 (e)
Lumped as HC8
64.3 5.52 0.11 0.74 0.01 0.51 0.80
35.7 3.43 0.02 -0.63 0.03 0.15 0.50
4.77 0.08 0.25 0.02 0.38 0.69
0.10 0.27 0.00 0.36 0.76
Alkanes (0.50-1.00)
Others (0.50-1.00)
Recommended HC5 (d)
Initial RADM HC5 (e)
21.9
3.1
3.3
1.6
1
2
1
2
1
1
.32
.34
.55
.02
.51
.66
70.5
12.0
11.5
6.0
7.
12.
2.
10.
7.
74
38
87
04
87
0.25
0.34
0.02
0.30
0.24
0.24
0.88
1.10
-0.76
1.34
0.75
0.78
0.04
0.06
0.01
0.12
0.04
0.05
0.32
0.20
0.74
0.37
0.35
0.44
1.11
1.39
0.30
1.31
1.06
1.05
Alkanes (1.00-2.00)
Alkanes O2.00)
Others (>2.00)
Alkane/Aromatic Mix
Recommended HC8 (d)
Initial RADM HC8 (e)
(a) OH Radical rate constant in units of 10 ppm min at T = 300 K.
(b) Percent of estimated moles reacted for each lumped RADM group. Estimated by
moles reacted = moles emitted x [ 1 - exp(-kOH x INTOH) ], where INTOH = 110
ppt-min.
(c) Average number of carbons.
(d) Derived from averages for the emissions groups, as described in the text.
(e) Used in the March 1988 version of the RADM mechanism.
89
-------
Table 10. Averages of Parameters for Emissions Groups Lumped as Aromatics
and for RADM Lumped Aromatic Model Species, Derived Using the
NAPAP Total U.S. Emissions Inventory
K. Moles
Emit
16.1
7.3
0.8
16.3
0.2
kOH
(a)
0.919
0.185
0.902
0.918
0.902
0.902
4.09
3.73
4.09
4.54
% Ret. nC
(b) (c)
Lumped as TOL
84.9 7.26
11.1 6.00
4.0 7.36
7.13
Lumped as XYL
99.0 8.88
1.0 8.26
8.87
Description
Aromatics (<2.00)
Benzene, Halobenzenes
Styrenes
Average for RADM TOL (d)
Recommended RADM TOL (e)
Initial RADM TOL (f)
Aromatics (>2.00)
Alkane/Aromatic Mix
Recommended for RADM XYL (d)
Initial RADM XYL (f)
(a) OH Radical rate constant in units of 10 ppm"1 min"' for T = 300 K.
(b) Percent contribution to the RADM group Reacted, estimated using
Moles Reacted = Moles Emitted x [ 1 - exp(-kOH x INTOH) ], where
INTOH = 110 ppt min"1.
(c) Average number of carbons.
(d) Derived from averages for the emissions groups.
(e) The OH radical rate constant for TOL used in the initial RADM
mechanism is based on that for toluene. This is sufficiently
close to the average for the emissions represented by TOL that
no change is recommended.
(f) Used in the March 1988 version of the RADM mechanism.
90
-------
Table 11. Averages of Parameters for Emissions Groups Lumped as Non-Etnene Alkenes and
for RADM Lumped Non-Ethene Alkene Model Species, Derived Using the NAPAP Total
U.S. Emissions Inventory
K
.Moles
Emit
*
Ret.
(a)
nC
(b)
kOH
(c)
k03
(d)
kM03
(e)
Structural
=CH2 =CHR
Groups
=CR2
Description
Lumped as OLT
10.6
1.6
0.9
0.8
62.6
27.3
5.5
1.5
3.00
1.71
5.89
3.50
3.82
1.72
5.11
1.99
1.19
1.19
3.82
1.71
1.67
1.70
78
1
1.73
1.71
1.71
11.1 1.00 1.00 0.00
21.1 1.00 0.99 0.01
68.1 1.00 0.89 0.11
159-1 1.00 0.68 0.32
26.5 1.00 0.98 0.02
26.5 1.00 1.00 0.00
23.3 1.00 1.00 0.00
Propene
Primary Alkenes
Mixed Alkenes
Styrenes
Average for OLT (f)
Recommended for OLT (g)
Initial RADM OLT (h)
Lumped as OLI
11.
0.
0
9
92
7
.2
.8
1.85
5.89
1.93
1.93
9.88
9.11
9.82
9.82
9.21
23.9
29.0
21.3
21.3
30.2
1911.
578.
1837.
1837.
1837.
0.30
0.00
0.28
0.28
0.28
1.10
2.00
1.11
1.15
1.15
0.11
0.00
0.10
0.10
0.10
Internal Alkenes
Mixed Alkenes
Average for OLI (f)
Recommended for OLI (g)
Initial RADM OLI (h)
(a) Percent contribution to the RADM group reacted, estimated assuming all emitted alkenes
are consumed through reaction.
(b) Average number of carbons.
(c) OH radical rate .constant in units of 10^ pom"1 min~' for T = 300 K.
(d) Ozone rate constant in units of ppm min"^ for T = 300 K.
(e) NO^ radical rate constant in units of ppm" min for T = 300 K.
(f) Derived from averages for the emissions groups.
(g) The average values for the structural group parameters are so close to the initial RADM
values that no change is recommended.
(h) Used in the March 1988 version of the RADM mechanism.
-------
of moles of each emissions group which were derived from the NAPAP total
emissions inventory.
The estimated percent contributions of the various emissions groups
to the reactivities of the lumped RADM groups used to represent them are
also shown in Tables 9 through 11. In the case of the alkanes and
aromatics, these contributions were computed based on the approximate
amounts of each group expected to react in a model simulation, estimated
as described above in the discussion of reactivity weighing [i.e.,
estimated using Equation (II), with INTOH = 110 ppt-min]. For non-ethene
alkenes, these were estimated based on assuming all the emitted alkenes
react in the model simulation as expected based on their relatively high
OH radical rate constants and the fact that they are consumed non-negli-
gibly by reaction with ozone as well. These percent contributions are
then used to compute the recommended averages for the OH radical rate
constants and for the other mechanistic parameters for the lumped RADM
species.
It should be noted that as indicated in Section 2 the parameters for
the evaluated RADM mechanism incorporated earlier recommendations we made
for kinetic and mechanistic parameters for these species, based on an
analysis similar to that described here. However, the emissions parameter
data base at SAPRC was still undergoing development at the time these
initial recommendations were made, and thus the final recommendations are
slightly different. The parameter values used in the initially evaluated
mechanism, derived based on this earlier analysis, are shown in the tables
as "Initial RADM." Since the final recommendations have been implemented
into the new version of the RADM-II mechanism, the derivation of the
earlier recommendations is not discussed here.
The derivations of the values for the kinetic and mechanistic param-
eters shown in Tables 9 through 11, and the ways they are used in the RADM
mechanism for the various types of lumped species are discussed below. In
most cases, the parameters are based on the kinetic and mechanistic
assignments we previously made for representing complex mixtures of
emitted species in the SAPRC mechanism, which is documented elsewhere
(Carter, 1988). However, in some cases, the product yield parameters had
to be modified to be consistent with the set of product species used in
the RADM mechanism.
92
-------
3.3.1 Mechanistic Parameters for Lumped Alkanes
Alkanes react in the atmosphere primarily with OH radicals to
form organic peroxy radicals. The peroxy radicals can react with NO to
form N02, H02, and a variety of oxygenated products or they can react with
NO to form organic nitrates. Techniques exist for estimating the rates of
initial OH radical attack at various positions on the alkanes (Atkinson,
1986, 1987; Carter and Atkinson, 1985), and the branching ratios of the
various reactions of the radicals subsequently formed (Carter and
Atkinson, 1985). We have used these estimates to derive kinetic and
mechanistic parameters for the major alkane species in the NAPAP inventory
(Carter, 1988a). If the distribution of alkanes in the available NAPAP
total U.S. emissions inventory can be considered to be generally represen-
tative, then these estimates can be used to derive average values for
these parameters for the three alkane model species in RADM.
The mechanistic parameters for the alkanes, the ways in which they
are estimated, and the ways in which they are implemented into the RADM
mechanism, are as follows:
The OH Radical Rate Constants. Estimates of the OH radical rate
constants at 300 K were made for the various types of emitted alkanes
using the OH radical rate constant estimation technique of Atkinson
(1987). These values were then used to derive rate constants at 300 K for
the alkane species used in the model. No recommendation is made concern-
ing the temperature dependence for these reactions, and the recommended
modified RADM mechanism uses the same activation energies for these reac-
tions as employed in the original RADM-II mechanism (NCAR 1987).
Organic Nitrate Yields. Organic nitrates are formed in alkane photo-
oxidations by the reaction of peroxy radicals with NO, competing with the
alternative reaction route involving conversion of NO to N02 and ultimate
regeneration of radicals. The nitrate yield is an important parameter in
alkane mechanisms since its formation represents a radical and NOX sink.
Note that by conservation of radicals and NOX, the overall radical yield
in the alkane photooxidation reaction is 1 minus the nitrate yield. These
are estimated for the various types of alkanes in the inventory as
described by Carter and Atkinson (1985), with parameters updated as given
by Atkinson (1988) and Carter and Atkinson (1989b). Note that these
estimates are based on assuming that alkyl nitrate formation occurs only
93
-------
during reaction of NO with unsubstituted peroxy radicals such as those
formed immediately following the initial alkane + OH reaction, and not in
the reactions of the hydroxy- or oxy-substituted radicals formed following
the alkoxy radical isomerizations involved in the photooxidations of the
long-chain alkanes (Carter and Atkinson, 1985). This is highly uncertain,
and thus the simulated nitrate yields in the reactions of the higher
alkanes may be underestimated.
The RADM mechanism represents the formation of alkyl nitrates in
alkane systems by a two-step process. The first is the initial reaction
with OH to form a peroxy radical. This radical then reacts with NO to
form the lumped alkyl nitrate species (ONIT) in variable yields and H02 in
yields of 1 minus the alkyl nitrate yield. For HC5 and HC8, the peroxy
radical yield in the first reaction is unity, and thus the nitrate yield
in the second reaction is the same as the nitrate yield derived in the
emissions analysis. However, to represent the appropriate numbers of NO
to N02 conversions as discussed below, the peroxy radical in the initial
reaction of the lumped alkane HC3 has a less-than-unit yield, and thus the
nitrate yield in the reaction of HCS's peroxy radical with NO is increased
accordingly, to obtain the overall yield indicated in Table 9. The
specifics of how nitrate formation and NO to N02 conversion is represented
in the RADM mechanism is discussed below.
NO-to-NOo Conversions. The alkane photooxidation reactions involve
the formation of organic peroxy radicals which convert varying amounts of
NO to N02- The total amount of NO converted to N02 by these reactions is
determined by a number of factors, including the alkyl nitrate yields and
the overall number of isomerizations and decompositions of radicals which
are involved in the photooxidation process. The number of these conver-
sions have been estimated for the major types of alkanes in the emissions
inventory (Carter, 1988) by using a computer program employing the peroxy-
and alkoxy-radical branching ratio estimates of Carter and Atkinson
(1985), as updated by Atkinson (1988). Analogous methods can be used for
estimating NO-to-NO^ conversions in species such as alcohols, esters, and
others aggregated in the "others" categories which are lumped with alkanes
in RADM, since generally they are predicted to form the same types of
radical intermediates. Note, however, that the photooxidations of simple
alcohols involve fewer NO-to-N02 conversions than do alkanes, since,
-------
unlike alkanes, photooxidations of alcohols do not involve formation of
peroxy radicals.
In the RADM mechanism, the representation of NO to N02 conversions by
peroxy radicals in the lumped alkane mechanism depends on the net number
of conversions that are estimated to occur. If the emissions data
indicate that an average of more than one NO is consumed per alkane
reacted, then the reactions can be represented by
HCn + OH > HCnP + a X02
HCnP + NO --> b ONIT + (1-b) H02 + (1-b) N02 + oxygenated products
X02 * NO > N02
where HCn is the lumped model species representing alkanes, HCnP is the
lumped peroxy radical species associated with HCn, and X02 is a "chemical
operator" used to represent excess NO-to-N02 conversions. The parameter
"a" in the above reaction is shown in Table 9 as the X02 yield, and the
parameter "b" indicates the organic nitrate (ONIT) yield. However, if the
emitted species represented by the lumped alkane species "HCn" includes a
large fraction of alcohols, whose reactions generate H02 and organic
products without producing peroxy radicals and NO-tc-N02 conversions, then
there may be on the average less than one NO converted to N02 per alkane
reacted. This is the case for the lumped species "HC3" in the present
mechanism, and is indicated in Table 9 by a negative number for the
parameter "a," the X02 yield. In this case the reactions are represented
by
HCn + OH > (Ua) HCnP (where a < 0)
HCnP + NO > b/(Ua) ONIT + (1-[b/( Ua)]) (H02 + N02)
+ oxygenated products.
Note that in both cases the parameter "b" reflects the total organic
nitrate (ONIT) yield, and the total number of molecules of NO reacting per
molecule of HCn consumed is a+1.
The above discussion concerns the case where reaction with NO is the
major fate for organic peroxy radicals. The RADM mechanism also includes
reactions of peroxy radicals with each other and with H02. The products
95
-------
formed in these reactions are not affected by the X02 and the ONIT yields,
though they are affected by the yields of the oxygenated organic products
as discussed below.
Oxygenated Organic Products. The atmospheric photooxidation reac-
tions of the alkanes in the presence of NO involve the formation of a
variety of oxygenated products, including complex bi- and poly-functional
species as well as simple aldehydes and ketones. The yields of these
oxygenates can be estimated using the estimates of the relative rates for
alkoxy radical decomposition, isomerization, and reaction with Oo given by
Carter and Atkinson (1985), and updated by Atkinson (1988). The computer
program used to derive these estimates expresses the yields in terms of
the oxygenated products used in the most detailed version of the current
SAPRC mechanism. These products are HCHO (formaldehyde), CCHO (acetalde-
hyde), RCHO (C^ and higher aldehydes), ACET (acetone), and MEK (methyl
ethyl and higher ketones) (Carter, 1988). The RADM mechanism represents
alkane oxygenated products using the model species HCHO (formaldehyde),
ALD (higher aldehydes), and KET (ketones). The SAPRC representation is
converted to RADM model species by using ALD to represent both CCHO and
RCHO, using KET to represent MEK, and using 1/2 KET to represent ACET.
The yields shown in Table 9 for the products HCHO, ALD, and KET refer
to their overall yields under conditions where reaction with NO is the
major fate for organic peroxy radicals. As indicated above, the RADM
mechanism also includes reactions of peroxy radicals with H02 and with
other peroxy radicals. The RADM mechanism represents all peroxy + H02
reactions as forming lumped hydroperoxide species, which do not depend on
the set of oxygenated products assumed when the reaction with NO
dominates. When the peroxy radicals react with other peroxy radicals, the
set of oxygenated organic products will in general be different from those
formed in the reaction with NO, and organic nitrates will not be formed.
However, no separate estimates were made concerning the set of organic
products formed from peroxy + peroxy reactions; instead they were repre-
sented by the same set of products formed in the NO reaction, with the
same relative yields, but their overall yields increased by a factor of
b/(1-b) (where "b" is the ONIT yield) to account for the fact that organic
nitrates are not formed. This is consistent with the way these reactions
are represented in the SAPRC mechanism (Carter, 1988).
96
-------
Mechanistic Parameters for Lumped Alkanes in the RADM-M Mechanism.
As discussed in Section 2, one of the modifications we recommend for the
RADM mechanism is that the two higher lumped alkane classes, HC5 and HC8,
be combined into a single model species. This is because, as shown in
Section 6.4, this condensation in the representation of the alkanes is
unlikely to have significant impacts on predictions of species of interest
in RADM applications and reduces the size of the mechanism by two
species. The kinetic and mechanistic parameters for this new lumped
species were derived in the same manner as discussed above. Table 12
gives the averages for the parameters for the emissions groups represented
by this new lumped higher alkane species and the resulting averages
calculated for the model species. As indicated in the table, the average
number of carbons of the compounds it represents in the NAPAP inventory is
6.3, and thus the model species is designated "HC6." The parameters used
for HC6 in the recommended RADM-M mechanism are based on the results of
this analysis.
3.3.2 Mechanistic Parameters for Lumped Aromatics
The aromatic hydrocarbons are also consumed in the atmosphere
primarily by reaction with OH radicals, and the rate constants for the
reactions of most of the emitted aromatic species are known or can be
estimated based on those for chemically similar species (Atkinson,
1986). OH radical rate constants for T=300 K were assigned for the major
aromatic hydrocarbon species in the NAPAP total U.S. emissions inventory,
and these values, combined with the relative molar emissions of aromatics
in that inventory, were used to derive the weighed average rate constants
shown in Table 10 for the various aromatic emissions groups and for the
two aromatic model species in RADM. Note that the weighed average rate
constant for the RADM species TOL is so close to that for toluene that use
of the toluene rate constant is not inappropriate. However, because of
polyalkyl benzenes, naphthalenes, and other higher aromatics represented
by the RADM species XYL, the average rate constant derived for that
species is higher than that for the xylenes.
We could not derive average values for the other mechanistic
parameters for the aromatics in a manner analogous to that done for the
alkanes because the mechanisms for aromatic reactions are highly
uncertain. We must rely on chamber data, as discussed in Section 2.2.1,
to derive these parameters.
97
-------
Table 12. Averages of Parameters for Emissions Represented by the Combined Higher
Alkane (HC5 + HC8) Model Species Recommended for the RADM-M Mechanism
K. Moles kOH
Emit (a)
Lumped
21.9
25.5
12.5
3.3
3.1
1.6
as
1
0
0
1
2
2
1
HC3
.315
.668
.801
.552
.339
.018
.088
' Ret.
(b)
37.8
29.9
16.6
6.1
6.4
3.2
nC
(c)
7.74
5.52
3.43
2.87
12.38
10.04
6.26
ON IT
0.25
0.11
0.02
0.02
0.34
0.30
0.15
X02
0.88
0.74
-0.63
-0.76
1.10
1.34
0.51
roduct
HCHO
0.04
0.01
0.03
0.01
0.06
0.12
0.03
ALD
0.32
0.51
0.15
0.74
0.20
0.37
0.38
KET
1.11
0.80
0.50
0.30
1.39
1.31
0.87
Description
Alkanes ( 1 . 00-2
Alkanes (0.50-1
Others (0.50-1.
Others (>2.00)
Alkanes O2.00)
Alkane/Aromatic
Recommended HC6
.00)
.00)
00)
Mix
(d)
(a) OH radical rate constant in units of 10^ ppm"1 min"1 at T = 300 K.
(b) Percent of estimated moles reacted for each lumped RADM group. Estimated by
moles reacted = moles emitted x [ 1 - exp(-kOH x INTOH) ], where INTOH = 110
ppt-min.
(c) Average number of carbons.
(d) Derived from averages for the emissions groups, as described in the text.
98
-------
3.3-3 Mechanistic Parameters for Lumped Alkenes
The RADM mechanism includes the reaction of alkenes with OH
radicals, ozone, and NO^ radicals. Rate constants were assigned for the
major alkene species in the NAPAP total U. S. inventory based on experi-
mental or estimated values (Carter, 1988, and references therein). Using
these and the relative molar emissions in the NAPAP inventory, average
rate constant values for the various lumped alkene emissions groups and
for the RADM lumped alkene species OLT and OLI were derived. The results
are shown in Table 11. The contributions of the emissions groups to the
weighed averages for the lumped RADM species were calculated assuming that
all the emitted alkenes represented in these groups are consumed during a
model simulation.
The appropriate sets of organic products to use in alkene reaction
mechanisms depend on the nature of the substituents around the double bond
(Carter, 1988). For example, reactions of OH or NOo radicals with alkenes
result in formation of formaldehyde from =CH2 groups, higher aldehydes
from =CHR groups (where R = alkyl), and ketones from =CR2 groups, with
analogous considerations being involved in determining the products in the
ozone reactions (Carter, 1988). Table 11 shows the average number of
substituents about the double bonds for the alkenes in the various
emissions groups and for those represented by OLT and OLI in the RADM
mechanism. (The number of substituents does not sum up to two for the
"internal alkenes" emissions group because cycloalkenes are counted as
having only one substituent about the double bond for the purpose of
determining the organic product yields.) The recommended sets of products
for the lumped internal alkene species (OLI) is based on the results of
this analysis. Thus, for example, the reaction of OLI with OH radicals
yields 0.28 HCHO + 1.M5 ALD - 0.10 KET as the organic products, reflecting
the weighed average numbers of sCHj, =CHR, and =CR2 substituents,
respectively. In the case of the terminal alkenes (OLT), this analysis
indicates that the continued use of propene (with one =CH2 and one =CHR
substituent) as the surrogate species for the terminal alkenes is
appropriate.
99
-------
M. EVALUATION AGAINST ENVIRONMENTAL CHAMBER DATA:
CHARACTERIZATION AND LUMPING PROCEDURES
4.1 Introduction
An important step in the model design and evaluation process is
evaluation of the gas phase chemistry against environmental chamber
data. The gas phase chemistry module is one of the few modules in the
atmospheric modeling system that can be independently tested against
experimental data. The testing against chamber data is essential for, but
not necessarily sufficient to, fully establishing the scientific validity
of a mechanism for use in regional acid deposition modeling. The reason
for this is that while a large data base of experiments exists for the
ROG/NOx/Oo system under urban-like conditions, virtually no reliable data
exist to test mechanisms' performance for other important species, such as
sulfuric acid, nitric acid, hydrogen peroxide, organic peroxides, and
organic acids, or for other important concentration regimes, such as the
rural regime which is dominant in regional model applications. Neverthe-
less, the present chamber data base can be used to test the rate of NOX
oxidation and concentrations of ozone, peroxyacetylnitrate, and formalde-
hyde predicted by gas phase chemical mechanisms for urban conditions.
This is important because these species are important in long-range trans-
port and the regional model must simulate the chemistry of emissions as
they are transported from urban areas to remote areas with sensitive
receptors.
It is also important to recognize that even though the present data
base does not allow for direct evaluation of model predictions for
sulfuric acid and nitric acid, the testing protocol provides some indirect
evidence for evaluation of performance for these species. First, accurate
prediction of the rate of NOX oxidation and ozone formation necessitate
accurate predictions of OH radical concentrations. OH, according to
present theory, is the sole gas-phase oxidant that converts SOp to
sulfuric acid and NOp to nitric acid in the daylight hours.. Second, since
NOV is primarily oxidized to nitric acid or PAN, the evaluation results
A
for PAN provide some information on whether the mechanism is oxidizing NOV
A
to the correct species.
100
-------
An evaluation of the March 1988 RADM, recommended modified RADM
(RADM-M), and the recommended modified parameter (RADM-P) mechanisms
against environmental chamber data was performed in this study. The
objective of the evaluation was to assess the accuracy and precision of
the mechanisms' predictions for ozone, PAN, formaldehyde, and the rate of
NOX oxidation. The approach for the evaluation involved simulating a
large number of experiments (over 500) from four different chambers and
statistically evaluating the mechanisms' performance.
Experimental data collected in the following chambers was used for
the evaluation:
1) The SAPRC evacuable indoor chamber (EC),
2) The SAPRC indoor Teflon chamber (ITC),
3) The SAPRC outdoor Teflon chamber (OTC), and
4) The University of North Carolina outdoor chamber (UNC).
Only experiments with relatively complete data were employed in the
evaluation. The specific sets of experiments used were those recommended
for use in model evaluation by the experimentalists (Jeffries et al.,
1985; Carter et al., 1986), with additional experiments added which were
employed in more recent evaluations of the SAPRC mechanisms (Carter et
al., 1987, 1988).
The results of the chamber simulations are described in Section 5.
Before presenting these results, it is first necessary to indicate the
assumptions made and procedures that were followed. Carrying out model
simulations of environmental chamber experiments involves more than just
implementing the chemical mechanism on a computer with the appropriate
software, specifying the initial reactant concentrations, and then running
the calculation. Before such simulations can be carried out, it is
necessary to specify in the model input the appropriate conditions of the
chamber experiments, including appropriate representation of such factors
as temperature, light intensity and spectral characteristics, humidity,
and chamber wall and contamination effects. In addition, for a lumped
mechanism such as RADM, which does not contain explicit representation of
all of the various types of organic compounds which are present in the
experiments, appropriate procedures need to be established for
101
-------
representing the organics used in the chamber experiments with the species
incorporated in the mechanism. These are discussed in this section.
U.2 Characterization of Chamber Conditions
In order to test chemical mechanisms against environmental chamber
results, it is necessary to include in the model appropriate representa-
tions for chamber-dependent effects such as wall reactions and character-
istics of the light source used during the experiments. The specific
chamber-dependent aspects represented in the process of testing mechanisms
are the following:
1) Light characterization;
2) Chamber-dependent radical sources;
3) Offgassing of NOX from the walls;
4) Heterogeneous hydrolyses of N20c and NC^;
5) Wall loss of ozone;
6) Excess NO oxidation caused by background organics;
7) Dilution rates; and
8) Characterization of temperature and humidity.
There is significant uncertainty regarding the most appropriate way to
represent some of these aspects and different modeling groups have made
different assumptions. As in our previous mechanism evaluation studies
(Carter et al., 1986, 1987, 1988), the goal in this study was not to
represent these effects exactly (since this is not possible given the
current state of knowledge), but to represent these effects in the four
chambers in a consistent manner and with the minimum number of empirical
parameters required to be consistent with the available characterization
data.
We had developed a comprehensive set of procedures and assumptions
for representing these chamber- and run-dependent processes during the
course of our previous mechanism evaluation studies (Carter et al., 1986,
1987). However, prior to conducting the evaluation, these procedures and
assumptions were reviewed and updated where new data were available.
Procedures for characterizing the SAPRC chambers were almost entirely
adopted from those used in our previous evaluation studies (Carter et al.,
102
-------
1986, 1987) and therefore are only briefly summarized here. However,
significant improvements were made in the procedures to characterize the
light intensity and spectral distribution in the UNC chamber, based on the
recent work carried out by Jeffries et al. (1989a). These changes in the
light characterization for the UNC chamber necessitated other refinements
in the characterization procedures for that chamber.
4.2.1 Light Characterization
The rates of the photolysis reactions in environmental chamber
simulations are obviously important since these reactions supply the
energy to drive the chemical transformations that the model is attempting
to simulate. For both indoor and outdoor chamber simulations, photolysis
rate constants were calculated using the following equation,
kPnot(t) = I [ J(x,t) c(x) *(X) ] (I)
X
where kPhot(t) is a photolysis rate constant at time t, J(X,t) is the
intensity at time t of the light source in a wavelength region centered at
wavelength X and o(X) and 4>(X) are the absorption coefficient for the
photolyzing species and the quantum yield for the photolysis reaction over
that wavelength region. For each photolysis reaction, the values of the
absorption coefficients and quantum yields are considered to be part of
the homogeneous reaction mechanism (given in Section 2). On the other
hand, the values J(X,t) are determined by the intensity and spectral
distribution of the light source. Thus, specification of this parameter
is part of the chamber-dependent input in the model simulations of the
chamber experiments. The procedure used for deriving J(X,t) depends on
whether the chamber is an indoor or an outdoor chamber.
Light Characterization for Indoor Chambers. The light intensities
and spectral distributions used for the SAPRC EC and ITC are those
discussed by Carter et al. (1986). These are believed to be reasonably
well characterized and are assumed to be constant throughout each experi-
ment. For both chambers, the light intensity is monitored by carrying out
periodic NOp actinometry experiments, and the results of these are used to
assign k1 values for various groups of runs. For most experiments the
values of the NC>2 photolysis rates (k^) were in the range of 0.2 to 0.4
103
-------
min , with the values changing relatively slowly over time. Based on the
observed variability of the k^ measurements, the values assigned for
individual runs are estimated to have an uncertainty of -10/1 or less.
The ratios of rates of the other photolysis reactions to those for
N02 are determined by the spectral distribution of the light source. For
the experiments carried out in the SAPRC EC, the relative spectral distri-
bution was measured during the course of each run, as described by Pitts
et al. (1986). For runs carried out around the same time and with the
same lamp, essentially the same spectral distribution was observed, though
with some run-to-run variability; but over longer periods of time, system-
atic changes in the spectral distribution occurred, with the relative
intensity in the UV region becoming progressively less. In the model
testing, consecutive runs with similar measured spectral distributions
were grouped together, and the averages of their measured spectral
distributions were used for all the runs in the group. The sets of EC
spectral distributions which were employed are summarized by Carter et al.
(1986).
The relative spectral distribution for the blacklights used in the
ITC was measured much less frequently, but the few measurements carried
out indicate that it can be assumed to be essentially constant. The
measured spectral distribution for this chamber is also given by Carter et
al. (1986). Compared to the solar or EC spectral distributions, the ITC
spectral distribution is relatively more intense in the ~330 - 370 micron
region, but is much less intense for wavelengths greater than -380
microns.
Light Characterization for Outdoor Chambers. The intensity and
spectral distribution of the light source for the UNC and SAPRC OTC out-
door chambers, which consists of natural sunlight passing through the
chamber walls, is more difficult to characterize. Both the intensity and
spectral distribution of the light source depend on the time of day, the
time of year, and the local meteorological conditions. Jeffries (1989a)
recently completed an extensive experimental and theoretical study of the
light characteristics inside and outside the UNC chamber and developed a
model and a data base for deriving light intensities and spectral
distributions as a function of time for each of the UNC chamber experi-
ments whose data are available for model evaluation. This new UNC light
104
-------
characterization data base was employed in this study. In principle, the
procedures employed to develop the UNC chamber light characterization data
base could be employed to characterize the light in the SAPRC OTC runs,
but there was insufficient time and resources available in this program to
carry this out. Therefore, the methods used to represent OTC light
characteristics in our previous mechanism evaluation studies, described by
Carter et al. (1986), were also employed in this study.
The overall features of the methods used for light characterization
for the OTC and the UNC chambers are summarized below. However, the
reports of Jeffries et al. (1989a,d) (for the UNC chamber) or of Carter ec
al. (1986) (for the OTC) should be consulted for details.
Light Characterization of the SAPRC OTC. The light characterization
data available for the OTC experiments consist of continual UV radiometer
measurements of sunlight outside the chamber and, for many experiments,
continual N02 actinometry (k^) measurements made underneath the reaction
bag, with the light passing in and out of the chamber before reaching the
actinometer. Such data can be used to estimate how total light intensity
is varying with time outside the chambers, but they give no direct indica-
tion of how the relative spectral distribution is varying, or (except to a
limited extent for the k1 measurements under the OTC) how the light
intensity and spectral distribution inside the chamber may differ from
that outside. Therefore, much of the light characterization input
required for the OTC was derived based on theoretical calculations,
assumptions and approximations which are discussed below.
The light intensities used in calculating the photolysis rate
constants for a given OTC chamber experiment are given by J(X,t), which
was calculated as a function of wavelength (X) and time of day (t) accord-
ing to the equation:
UV(t)
J(X,t) = J (z[t'J) fU.t) (II)
0 UV (z[t'j)
where z[t'] is the solar zenith angle at time t1, which is our best
estimate of the true solar time t of the experiment, J0(z[t']) is the
theoretically-calculated clear-sky light intensity for the region
105
-------
associated with wavelength g, UV(t) is the experimental UV radiometer
reading taken at time t during the experiment, UV0(z[t']) is the theore-
tically calculated clear-sky UV radiometer reading for z[t'], and f(X,t)
is a correction factor to account for the effects of transmission of light
through the chamber walls.
The clear-sky light intensities given by Peterson (1976) for his
"best estimate" surface albedos were used to derive J0(x,z[t']) for the
OTC experiments. The values of JQ used for a given wavelength and zenith
angle were calculated using an empirical function fit to the data tabu-
lated by Peterson (1976) as described by Carter et al. (1986). The solar
zenith angle z(t') was calculated from the true solar time, the date, and
the latitude of the SAPRC OTC (34.06 degrees) using the standard formula
as employed by Carter et al. (1986). Because the clocks used during the
outdoor chamber experiments were not always accurate (Carter et al.,
1986), the difference between the clock time t and the true solar time t'
were derived based on adjusting t-t' so that the experimental and the
theoretically calculated UV curve shapes agreed. As expected, the time
offsets did not vary randomly from run-to-run, but were generally the same
for runs carried out around the same time period. For a few runs the
light quality was so poor that it was not possible to determine t-t1 using
the light intensity data, and those runs were assumed to have the same
time offset as observed in runs carried out at around the same time.
The experimentally measured UV data are used to take into account
local meteorological factors which affect the light intensity entering the
chamber as a function of time during the individual chamber experiments.
As indicated in Equation (II), this is done by assuming that for any wave-
length the ratio of the actual light intensity outside the chamber to JQ,
the theoretical "clear sky" light intensity value, is the same as the
ratio of the measured UV radiation intensity to UVQ, the calculated
estimated "clear sky" UV values, which, like JQ, are assumed to be a func-
tion only of the solar zenith angle. These latter quantities are
calculated from an empirical formula (Carter et al., 1986), which is
expressed as:
expt calc
UV = f (z) lim (UV/k ) (0) k (0) (III)
0 shape z->0 1 1
106
-------
where fsnape is an empirical function, normalized to unity for z=0, which
was derived based on fits to the shapes of experimental UV vs time data
obtained during SAPRC OTC and UNC outdoor chamber experiments on clear
days; lim(z->0) (UV/k1)exPfc(0) is the estimated z=0 ratio of UV intensity
to the N02 photolysis rate (k^ derived based on the empirical relation-
ship between simultaneous experimental UV radiometer and k1 measurements
given as a function of zenith angle by Zafonte et al. (1977); and
k1ca c(0) is the calculated N02 photolysis rate using the JQ values for
z=0.
At both the SAPRC and the UNC outdoor chamber laboratories the UV
instrument is located outside the chamber, and thus does not directly
measure the light intensity inside the chamber. In the case of the SAPRC
OTC, which consists of a flexible Teflon bag held on a rope netting above
a green indoor-outdoor carpet, the major cause for differences in light
intensity between the inside and the outside of the chamber is assumed to
be the attenuation of the light as it passes through the Teflon walls.
The k1 data taken during the OTC runs are used to obtain a rough indica-
tion of the magnitude of this attenuation, since the tube used to obtain
these data is located underneath the reaction bag, and thus light reaching
it passes through the chamber walls twice. A comparison of observed with
theoretically calculated k1 readings for these OTC runs (Carter et al.,
1986) indicates that the light reaching the k1 tube under the OTC is
attenuated by a factor of -1.44. Since the light has to pass through two
Teflon surfaces (the top and the bottom of the chamber) before reaching
the NC>2 actinometer, this corresponds to a suppressipn factor of 1.2 when
passing through each wall of the chamber. Thus in modeling the OTC runs
for this study, we use a wall transmission correction factor of 0.83 in
equation (II) when calculating the light intensity.
This estimation that a simple 0.83 correction factor takes into
account all the differences between the light intensity and spectral
distribution inside the OTC and the theoretically calculated values is
obviously subject to a number of uncertainties, since it does not take
into account, for example, possibilities that the walls, or reflections
from the surfaces and structures around the OTC may be changing the zenith
angle dependences of the light intensity, or that attenuation and/or
scattering of light as it passes through the walls may cause a change in
107
-------
the spectral distribution. These uncertainties are expected to be the
greatest at the higher zenith angles, but for most SAPRC OTC runs the
chamber is covered before 0900 PST and after 1500 or 1600 PST, so uncer-
tainties associated with light characteristics at high zenith angles will
not affect model simulations of most SAPRC OTC runs. Based on these
considerations, we estimate that the light intensity inside the OTC is
probably uncertain by -15J; though this is a subjective estimate. The
uncertainty is probably somewhat less at the wavelength region affecting
the photolysis of MOo, and probably somewhat greater at the lower wave-
length region which affects the photolysis of aldehydes and the formation
of 0(1D) from 0^.
It should be noted that both the k1calc(0) values used in Equation
(III), and the comparison of observed and calculated k1 values used in the
derivation of the wall transmission correction factor depend on the set of
absorption coefficients and quantum yields assumed for N02- However,
these two factors have opposing effects on the calculation of the overall
in-chamber light intensity in Equation (II), and thus the light intensi-
ties used in the model simulations are not sensitive to the sets of N02
absorption coefficients and quantum yields used in this analysis. At the
time of the derivation of the data base used for the modeling of the OTC
runs, we used the N02 absorption coefficient and quantum yields recommend-
ed by NASA (1985), which are somewhat different (yielding, for example, 1%
lower calculated N02 photolysis rates for z=0) than those used in the
current RADM (see Section 2) and SAPRC (Carter et al., 1988) mechanisms.
However, since as indicated above these differences will tend to cancel
out in the derivation of the overall light intensity, and since there are
a number of much more significant uncertainties involved in this deriva-
tion, it was determined not to be productive to expend the significant
effort required to re-derive the OTC light characterization data base
using the updated N02 absorption coefficients and quantum yields. If the
OTC light characterization data base is to be re-derived, it would be more
appropriate to use the updated procedures developed by Jeffries (1989a)
for re-deriving the UNC chamber light characterization data, as discussed
below.
108
-------
Updated Light Characterization of the UNC Chamber. In our previous
mechanism evaluation studies (Carter et al., 1986, 1987), the light
characterization input data used in the model simulations of the UNC out-
door chamber were derived in a manner similar to that described above for
the SAPRC OTC. However, this previous light characterization for the UNC
chamber was considered to be highly uncertain for a number of reasons.
The particular uncertainties in UNC light characterization concerned the
attenuation and reflection of light inside the chamber due to the Teflon
walls of that chamber, the effects of enhancement of light due to the
reflective chamber floor, the fact that (unlike the case for the OTC)
irradiations in UNC chamber experiments begin with the sun very low in the
sky, and evidence that at certain times in the past the UNC UV instrument
was significantly out of calibration. Because of lack of data, highly
uncertain and simplified assumptions had to be made in our previous
mechanism evaluation studies concerning these effects (see Carter et al.,
1986), such as, for example, assuming that effects of attenuation due to
wall absorption and enhancement due to floor reflection exactly canceled
out, and thus using a net chamber transmission factor of 1. In addition,
correction factors for UV data when the instrument was apparently out of
calibration were roughly estimated based on incomplete information. For
these reasons, we determined that the light characterization model for the
UNC chamber should be updated and improved as part of the effort carried
out in this program.
For a number of years, Dr. Harvey Jeffries and co-workers at UNC have
been working on developing new models for solar radiation and for light
characteristics inside the UNC chamber (Jeffries et al., 1989a) and on
developing correction factors for questionable light characterization data
from past UNC chamber runs. Therefore, as a part of this program, the
results of this effort have been utilized to develop a new data base for
light characterization for use in modeling UNC chamber runs. The major
part of the effort at UNC in developing and testing new light character-
ization models for solar radiation and radiation inside the UNC chamber
were carried out under separate funding and is described elsewhere
(Jeffries et al., 1989a). For this program, Jeffries and co-workers
performed quality assurance and applied their best estimate corrections to
the past UNC light characterization data, and, for each UNC chamber
109
-------
experiment whose results are currently available for modeling, derived
sets of wavelength- and time-resolved light flux data for use in modeling
the run. These consist of in-chamber actinic flux values for 91 wave-
length intervals for each 16 minute interval in the runs. These data were
then delivered to SAPRC in computer readable form, and the SAPRC chamber
model simulation software was adopted to incorporate these data when
simulating the UNC chamber runs.
A description of the models, data base, and procedures utilized to
develop these run-specific light characterization data is given by
Jeffries et al. (1989a) and is beyond the scope of this report. Briefly,
the procedure involves the following:
(1) A spectral radiation transfer model is utilized to calculate
clear-sky solar radiation fluxes outside the chamber. Its input includes:
solar zenith angle; air pressure; total atmosphere ozone column; total
atmosphere water vapor column; specification of atmospheric aerosol
density, type, and optical properties; relative humidity; and spectral
albedo of the surrounding surface.
(2) A broadband radiation transfer model is utilized to calculate
clear-sky UV and TSR predictions, which can be directly compared with
measured values taken during an experiment. Its input requirements are
the same as the spectral radiation transfer model.
(3) A relatively simple cloud transmission model is utilized to
account for the effects of clouds on surface level radiation fluxes and UV
and TSR readings. Its input consists of ratios of cloudy-to-clear-sky UV
and TSR values; these are derived by fitting calculated to observed UV and
TSR readings for various time periods in the runs as indicated below.
These ratios then account for the effects of clouds on the near UV and the
higher wavelength spectral regions. No attempt is made to convert the
broadband transmissions derived from the UV and TSR data into wavelength-
dependent spectral transmissions in the current version of the UNC cloud
model. However, Jeffries et al. (1989a) indicate that such a refinement
is unlikely to have a significant effect if the clouds attenuate less than
~HQ% of the light intensity.
(4) Models for spectral transmission, reflection, and absorption of
the FEP Teflon film used in the UNC chamber, for reflection of light off
the floor of the chamber, and for other chamber-specific light effects are
110
-------
utilized to calculate the light fluxes inside the chamber. Its inputs,
besides those specifying characteristics of the chamber which remain
constant, include the actinic fluxes outside the chamber for both the
direct and diffuse components, the solar zenith angle, and transmission
factors for the direct and the diffuse radiation. This model was
developed and validated utilizing an extensive series of spectral flux and
UV and broadband light intensity measurements made inside and outside the
UNC chamber (Jeffries et al., 1989a).
For each run, the procedure for developing the run-specific in-
chamber light fluxes involved the following: (1) Values of run-specific
parameters such as humidity, total column ozone, etc., which are derived
from meteorological observations, measurements made during the run, etc.,
and other available data bases are entered into the data base used by the
models. (2) Other parameters required for calculation of out-of-chamber
light fluxes, such as atmospheric aerosol parameters and broadband cloud
transmission factors, are estimated based on time of year, meteorology,
and past experience with the model and then are adjusted to yield accept-
able fits of observed-to-calculated UV and TSR readings for the individual
run. Cloud adjustment factors are derived for approximately three-hour
intervals, which is a similar time resolution used to derive the analogous
factors in the previous development of the UNC and OTC light characteriza-
tion data base by Carter et al. (1986). The types of fits of calculated-
to-observed UV and TSR data obtained after the application of this proce-
dure are illustrated in sample plots given by Jeffries (1989d). (3) The
UNC chamber transmission, absorption, and reflection model is then used to
calculate the in-chamber light fluxes for each 16-minute interval of the
run. These fluxes are incorporated into computer data sets (one for each
run-day) which were then transmitted to SAPRC for use in modeling.
The updated UNC light characterization data were found to be differ-
ent in a number of respects from the UNC light characterization data of
Carter et al. (1986) which was employed in our previous modeling studies
of this chamber. The major differences result from the use of the new
inside-vs-outside chamber model, which predicts enhancement of light
intensity at longer wavelengths due to floor reflection and suppression at
shorter wavelengths due to light absorption by the walls. [As indicated
above, because of lack of data, the Carter et al. (1986) model assumed
111
-------
that these effects canceled and thus the inside and outside fluxes were
the same.] The magnitude of these chamber light effects depends on the
time of day and on the relative amount of direct and diffuse radiation
calculated by the outside radiation transfer model.
A comparison of the effects of changing from the previous light
characterization of Carter et al. (1986) to this new light characteriza-
tion is shown for several UNC chamber runs in Figures 2 through 4, where
the former shows results for a formaldehyde-NOv-air run, and the latter
A
two show runs employing complex surrogate mixtures. [All these simula-
tions used the Carter et al. (1986) set of assumptions for other chamber
effects, though as discussed below some of the adjustable chamber effects
parameters were subsequently modified in this study to be more consistent
with simulations of characterization runs using the new light model.) As
illustrated in Figure 2, the new characterization model tends to yield
significantly better fits of formaldehyde decay rates in the formaldehyde-
NOX runs. However, in the case of this run the new model does not improve
the fits to the initiation time for NO-to-NOp conversion, though the new
model causes an overprediction of the initiation time, while in the old
model it is underpredicted. These effects can be attributed to the
suppression of UV intensity by the chamber walls at low zenith angles,
which is represented in the new but not the old model. On the other hand,
in many other types of runs, such as that shown in Figure 3, the new light
characterization model results in higher predictions of maximum ozone
yields, presumably due to the effects of assumed enhancement of light
intensity caused by the chamber's reflective floor. This tendency of the
new characterization model to increase predicted initiation times but also
increase maximum ozone yields was the case to varying degrees for most of
the UNC chamber runs modeled. However, for some runs, such as that shown
in Figure M, the effect of using different light characterization models
was relatively small.
4.2.2 Representation of Other Chamber-Dependent Effects
Assumptions regarding chamber-dependent reactions and the
presence of reactive background species have been recognized for a number
of years as being important in affecting results of model simulations of
chamber experiments. For example, it has been recognized for several
years that overall rates of initiation in environmental chambers are
112
-------
OZONE
E
a
cr
UJ
CJ
z
o
o
300
420 540 660 780 900
NITROGEN DIOXIDE
0.16-1
0.12-
0.08-
NITRIC OXIDE
0.04-
300 420 540 660 780 900
FORMALDEHYDE
300
420 540 660 780 900
ELAPSED TIME (minutes)
1.0-
0.8-
0.6-
-i
0.4-
-
0.2-
_
0.0-
-^B ^^ ~"
*^<7 r-
Y\.
\*
\
v. \
\"'"
^s>
300 420 540 660 780 900
Figure 2. Experimental and Calculated Concentration-Time Plots for
Selected Species in the UNC Formaldehyde-NO -Air Run AU0279B,
[All calculations used the March 1988 RADM mechanism and the'
Carter et al. (1986) wall characterization model.]
Calculated Using the Carter et al. (1986) Light Model
- » - Calculated Using the New UNC Light Model
* Experimental Data
113
-------
OZONE
E
Q.
Q.
g
*
LJ
O
2
O
O
0.32-,
0.24-
0.1 6-*
0.08-
NITRIC OXIDE
300 420 540 660 780 900 1020
NITROGEN DIOXIDE
0.00
300 420 540 660 780 900 1020
PAN
"i i 1 r
300 420 540 660 780 900 1020
300 420 540 660 780 900 1020
ELAPSED TIME (minutes)
Figure 3.
Experimental and Calculated Concentration-Time Plots for
Selected Species in the UNC Complex Mixture-NO -Air Run
ST1682B. [All calculations used the March 1988 RADM
mechanism with adjusted OH rate constants for lumped
hydrocarbon species and the Carter et al. (1986) wall
characterization model.]
Calculated Using the Carter et al. (1986) Light Model
Calculated Using the New UNC Light Model
Experimental Data
-------
OZONE
NITRIC OXIDE
E
a
a
O
<
a:
LU
o
z
o
a
300 420 540 660 780 900 1020
NITROGEN DIOXIDE
300 420 540 660 780 900 1020
PAN
, , j f
300 420 540 660 780 900 1020
300 420 540 660 760 90C 1020
ELAPSED TIME (minutes)
Figure 4. Experimental and Calculated Concentration-Time Plots for
Selected Species in the UNC Complex Mixture-NO -Air Run
ST2981R. [All calculations used the March 1980* RADM mechanism
with adjusted OH rate constants for lumped hydrocarbon species
and the Carter et al. (1986) wall characterization model.]
Calculated Using the Carter et al. (1986) Light Model
- - - Calculated Using the New UNC Light Model
Experimental Data
115
-------
higher than can be accounted for by known homogeneous processes (Carter et
al.f 1979a), and successful model simulations of chamber experiments have
had to include provisions for "chamber radical sources," whose exact
physical cause remains unknown or speculative. In addition, formation of
PAN and ozone in acetaldehyde-air runs with no added NOX has been observed
in all four chambers whose data are modeled in this program (e.g., see
Carter et al., 1986), indicating that offgassing of NOX must be occurring
from the walls of these chambers. Excess rates of NO oxidation observed
in N0v-air irradiations, and ozone formation in pure air irradiations
A
indicate that unmeasured reactive organic species are also either off-
gassed from the chamber walls or present in the background air in environ-
mental chamber experiments. Heterogeneous hydrolyses of NOp and NpOc and
wall losses of ozone are also known to occur. All of these need to be
represented in model simulations of chamber experiments.
The methods we used in representing these chamber-dependent effects
in the model simulations carried out in this study are summarized in this
section. Methods used to represent other chamber- or run-specific condi-
tions such as water concentration, temperature, or dilution in modeling
runs from the various chambers are also summarized. The general approach-
es and types of assumptions employed in representing these effects is
essentially the same as those described in Carter et al. (1986). That
report should be consulted for detailed discussions of these effects and
for the reasons why we chose to use the specific types of representations
employed. Since the time of the Carter et al. (1986) report, we have made
some minor modifications in the magnitude of the continuous, light-
dependent portion of the chamber radical source used when modeling SAPRC
ITC and OTC experiments (Carter et al., 1987) and in this study had to
make more extensive modifications in the representation of chamber effects
in the UNC chamber as a result of the changes in the UNC light character-
ization model. These changes to the UNC characterization model are
discussed below, following the summary of the representation of the
chamber-dependent parameters for the SAPRC chambers.
Representation of Chamber- or Run-Dependent Parameters in the SAPRC
Chambers. Tables 13 through 15 summarize the methods and parameters used
to represent chamber-dependent reactions, effects of offgasses or
background species, and other chamber- or run-specific conditions for
116
-------
Table 13. Summary of Chamber-Dependent Parameters Used in the Model
Simulations of the SAPRC EC Chamber Runs
Parameter(s)
Value and Derivation
General Run Conditions
Water Concentration
Temperature
Dilution
Based on measured temperature and
humidity during the final pure air
fill of the chamber. Typically
2.0 x 104 ppm.
303 K. This is the typical average
temperature in EC chamber runs.
3.0 x 10*4 min'1 (Pitts et al.,
1979).
Chamber Radical Sources
Initial Nitrous Acid (MONO)
wall
NO-
0.5 HONO +
0.5 loss of N02
NO-
wall
hv -> 0.
0.
HONO +
loss of NO-
wall
hv -> OH
Initial HONO = 1% of the initial N0?,
based on measured or estimated initial
HONO concentrations in EC tracer - NOX
air irradiations (Carter et al., 1982;
Pitts et al., 1983).
Included to represent the dark
formation of HONO from N02 observed in
this chamber. Measured rate constant in
the EC is 2.8 x 10~4 min"1 (Pitts et
al., 1984).
-1
Included to represent the N02-dependent
component of the continuous, light-
dependent radical source in this chamber
Rate constant given by 2.16 x 10"-^ ppm
x k1f where k« is the N02 photolysis
rate. Derived based on results of
tracer-NOx-air irradiations (Carter et
al., 1982; Pitts et al., 1983).
Included to represent the N02-independent
component of the continuous, light-
dependent radical source. Rate given by
0.39 ppb x k1? where k1 is the N02
photolysis rate. Derived based on
results of tracer-NOx-air irradiations
(Carter et al., 1982; Pitts et al., 1983)
(continued)
117
-------
Table 13 (continued) - 2
Parameter(s)
Value and Derivation
NOX Offgassing
wall + hv -> N02
Background Organics
(none)
Other Wall Reactions
wall
0-> -> lOSS Of Oo
wall, H20
N205 -> Loss of N205
Rate given by 0.5 ppb x k1t the N02
photolysis rate. Based on model
simulations of the acetaldehyde-air run
EC-253 (Carter et al., 1986).
No representation of effects of
background organics is used in model
simulations of EC runs. (Carter et al.,
1986).
First order loss rate of 1.1 x 10"
min is used, based on results of
ozone dark decay experiments (Pitts et
al., 1979; Carter et al., 1986).
Rate constant given by
M.65 x 1CT3 + 7.21 x 10'7 [H20] min'1,
based on measurements of N20c loss rates
in the EC by Tuazon et al. (1983).
118
-------
Table 14. Summary of Chamber-Dependent Parameters Used in the Model
Simulations of the SAPRC ITC Chamber Runs
Parameter(s)
Value and Derivation
General Run Conditions
Water Concentration
Temperature
Dilution
Based on measured temperature and
humidity during the final pure air
flush of the chamber. Typically
2.0 x 1CT ppm.
303 K. This is the typical average
temperature in ITC chamber runs.
No dilution is assumed to occur, due to
flexible nature of chamber walls.
Chamber Radical Sources
Initial Nitrous Acid (HONO)
Initial nitrous acid is assumed to be
negligible in Teflon film chambers
(Carter et al., 1986).
wall
N02 -> 0.2 HONO
0.8 loss of N02
Included to represent the dark formation
of HONO from N02 observed in this
chamber. The pseudo-first-order rate
constant used for the ITC is 1.4 x 10
min . This rate constant, and the
assumed HONO/N02 stoichiometry, is based
on studies of tnis process in a smaller,
1300-liter Teflon film chamber (Pitts et
al., 1984), corrected for differences
in chamber size (Carter et al., 1986).
wall
hv -> OH
This is the only process used to
represent the continuous light-dependent
component of the radical source in Teflon
film chambers (Carter et al., 1986). The
rate used is given by kp<> x k-j, where
k, is the NOo photolysis rate, and
kps is determined by averaging, for
each reaction bag, the radical input
rates measured using tracer-NO -air
irradiations (Carter et al., 1982). The
kRS values used in modeling ITC and OTC
runs are given in Table 16.
(continued)
119
-------
Table 14 (continued) - 2
Parameters}
Value and Derivation
NOH Offgassing
wall + hv -> N02
Rate given by 0.15 ppb x k1? the N02
photolysis rate. Based on model
simulations of acetaldehyde-air runs in
this chamber.
Background Organics
wall
OH -> H02
This "reaction" is used to simulate the
effect of trace contamination by reactive
organics which convert NO to NQ2. The
rate constant used is 250 min , based
on model simulations of pure air
irradiations (Carter et al., 1986).
Other Wall Reactions
wall
0-> -> loss of 0-3
wall, H20
N20c -> Coss of
First order loss rate of 1.3 x 10~
min"' is used, based on results of
ozone dark decay experiments (Carter et
al., 1986).
Rate constant given by
2.5 x 10'3 + 5.0 x 10~° [H20] min'1,
based on measurements of N20c loss rates
in a smaller Teflon bag reactor by Tuazon
et al. (1983), adjusted for differences
in chamber sizes as discussed by Carter
et al. (1986).
120
-------
Table 15. Summary of Chamber-Dependent Parameters Used in the Model
Simulations of the SAPRC OTC Chamber Runs
Parameter(s)
Value and Derivation
General Run Conditions
Water Concentration
Temperature
Derived based on continuous measurements
of dew point and temperature data for
runs where such data are available, as
discussed by Carter et al. (1986). If
such data are not available, a default
water concentration of 5 x 1CH ppm was
used. (OTC runs use unhumidified pure
air, so the typical water concentration
is relatively low.)
Plots of measured temperatures against
time were approximated by straight lines,
which were used to derive the temperature
as a function of time in the model
simulations.
Dilution
No dilution is assumed to occur, due to
flexible nature of the chamber walls.
Chamber Radical Sources
Initial Nitrous Acid (HONO)
wall
N02 -> 0.2 HONO
+0.8 loss of NO-
wall
hv -> OH
Initial nitrous acid is assumed to be
negligible in Teflon film chambers
(Carter et al., 1986).
Included to represent the dark formation
of HONO from NOo observed in Teflon
bag chambers. The pseudo-first-order rate
constant used for the OTC is 9.0 x 10~5
min . This rate constant, and the
assumed HONO/NOp stoichiometry, is based
on studies of this process in a smaller,
4300-liter Teflon film chamber (Pitts et
al., 1984), corrected for differences in
chamber size (Carter et al., 1986).
This is the only process used to
represent the continuous light-dependent
component of the radical source in Teflon
film chambers (Carter et al., 1986). The
(continued)
121
-------
Table 15 (continued) - 2
Parameter(s)
Value and Derivation
x k^, where
NOX Offgassing
wall + hv -> N02
Background Organics
wall
OH -> H02
Other Wall Reactions
wall
0, -> loss of
wall,
- -> Loss of
rate used is given by k
k-| is the N02 photolysis rate, and
kpg is determined by averaging, for
each reaction bag, the radical input
rates measured using tracer-NCL-air
irradiations (Carter et al., 1982). The
kRS values used in modeling ITC and OTC
runs are given in Table 16.
Rate given by 0.1 ppb x k1? the N02
photolysis rate. Based on model
simulations of acetaldehyde-air runs in
this chamber.
This "reaction" is used to simulate the
effect of trace contamination by reactive
organics which convert NO to NOp. The
rate constant is assumed to be the same.
-1
as used for ITC runs, which is 250 min
First order loss rate of 1.67 x 10"
min"1 is used, based on results of
ozone dark decay experiments (Carter et
al., 1986).
Rate constant given by 1.6 x 10"^ + 3.0 x
10 [H20] min , based on measurements of
N20c loss rates in a smaller Teflon film
reactor by Tuazon et al. (1983), adjusted
for differences in chamber sizes as
discussed by Garter et al. (1986).
122
-------
experiments carried out in the SAPRC EC, ITC, or OTC. In the case of the
ITC and the OTC, separate values of k^, the parameter used to determine
the rate of light-induced continuous production of radicals from unknown
sources, were employed for experiments using different reaction bags.
These values were determined by averaging the results of radical tracer-
NO -air experiments carried out with the different bags. [See Carter et
A
al. (1982) for a discussion of tracer-NOx-air experiments and measurements
of chamber radical sources.] The averages of the measured values of these
parameters for the various reaction bags and the k^ values used in the
model simulations are given in Table 16. Except for the use of separate
values of kRS for different Teflon reaction bags for ITC and OTC experi-
ments, the representation of these chamber-dependent parameters and run-
specific conditions for the SAPRC chambers is the same as chat discussed
by Carter et al. (1986).
The parameters shown in these tables are also the same as employed
when evaluating the more recent version of the SAPRC gas-phase mechanism
(Carter et al., 1988).
Representation of Chamber- and Run-Dependent Parameters in the UNC
Chamber. Tables 17 and 18 summarize the methods used to represent the
chamber- and run-dependent effects when modeling the UNC chamber runs
during the course of this program. Table 17 indicates the general forms
used for the representation of these effects and the values of the
parameters which were not varied during this program [being in all cases
the same as used by Carter et al. (1986)], while Table 18 gives the values
of the parameters which were varied during the course.of this study, as
discussed below.
As discussed by Carter et al. (1986), values of several of the
chamber-dependent parameters for the UNC and other chambers could be
derived only as a result of model simulations of characterization runs
which are particularly sensitive to these parameters. These are referred
to as the "adjustable" chamber-dependent parameters. Since run-by-run
optimization of uncharacterized parameters is incompatible with the
objectives of using chamber data for mechanism testing, this adjustment
must be done on a global basis, with the same sets of parameter values
being used in modeling all of the runs from a given chamber. In the case
of the UNC chamber, adjustable chamber-dependent parameters included k^g,
123
-------
Table 16. Values of the Continuous Chamber Radical Input Parameter, kRg>
used in Modeling SAPRC ITC and OTC Chamber Runs, and Averages
of Measured Values Used to Derive Them
Chamber
ITC
OTC
Inclusive
Run Numbers
450
532
621
692
737
793
859
924
183
209
- 503
- 563
- 638
- 734
- 789
- 832
- 893
- 970
- 206
- 253
Average Measured (a) Used in Model
0.16
0.44
0.21
0.61
0.49
0.36
0.28
0.25
0.25
0.20
± 0.03
± 0.13
± 0.03
± 0.20
± 0.20
t 0.12
± 0.09
± 0.25
± 0.19
± 0.09
0.15
0.4
0.2
0.6
0.5
0.35
0.3
0.25
0.25
0.2
(a) Runs carried out with new, unconditioned reaction bags or runs
with anomalous results were not counted in the average.
124
-------
Table 17. Summary of Chamber-Dependent Parameters Used in the Model
Simulations of the UNC Chamber Runs
Parameter(s)
Value and Derivation
General Run Conditions
Water Concentration
Temperature
Dilution
Chamber Radical Sources
Initial Nitrous Acid (HONO)
wall
N02 -> 0.2 HONO
0.8 loss of N02
wall
hv -> OH
Derived based on continuous measurements
of dew point and temperature data for
runs where such data are available, as
discussed by Carter et al. (1986). If
such data are not available, a default
water concentration of 2 x 10 ppm was
used.
Plots of measured temperatures against
time were approximated by straight lines,
which were used to derive the temperature
as a function of time in the model
simulations.
Dilution rates of 2 x 10 min were
assumed, as discussed by Carter et al.
(1986). No "dynamic dilution"
experiments were modeled in this study.
Initial nitrous acid is assumed to be
negligible in Teflon film chambers
(Carter et al., 1986).
Included to represent the dark formation
of HONO from NOp observed in Teflon
bag chambers. The pseudo-first-order rate
constant used for the UNC chamber is 5.0 x
10"-* min . This rate constant, and the
assumed HONO/NOo stoichiometry, is based
on studies of this process in a smaller,
4300-liter Teflon film chamber (Pitts et
al., 1984), corrected for differences in
chamber size (Carter et al., 1986).
This is the only process used to
represent the continuous light-dependent
component of the radical source in Teflon
film chambers (Carter et al., 1986). The
(continued)
125
-------
Table 17 (continued) - 2
Parameter(s)
Value and Derivation
rate used is given by kRo x k,, where
is the
photolysis rate, and
kRS was adJusted based on fits of model
simulations to alkane-NOx-air experiments.
The values used for this parameter are
are given in Table 18.
NOH Offgassing
wall + hv -> N02
Rate given by kEN x k-j, the N02
photolysis rate, where k was adjusted
based on fits of model simulations to
acetaldehyde-air experiments. The
values used for this parameter are given
in Table 18.
Background Organics
wall
OH -> H00
This "reaction" is used to simulate the
effect of trace contamination by reactive
organics which convert NO to N02. The
rate constant for this process, ^XSHC'
was adjusted based primarily on
fits of model simulations of pure air
irradiations. The values used for this
parameter are given in Table 18.
Other Wall Reactions
wall
-> loss of
wall, H20
-> Loss of
First order loss rate of 1.M3 x 10
min is used, based on results of two
ozone dark decay experiments (Carter et
al., 1986).
Rate constant given by 9 x 10" +
2 x 10~tt [H20] min"1, based on
measurements of N20e loss rates in a
smaller Teflon film reactor by Tuazon et
al. (1983), adjusted for differences in
chamber sizes as discussed by Carter
et al. (1986).
126
-------
Table 18. Values of Adjustable Chamber-Dependent Parameters Employed
in the Simulations of UNC Chamber Runs
Characterization Model
Original model
(Carter et al., 1986)
Re-adjusted. No
temperature dependence
Re-adjusted, with
temperature dependence
Temperature
(K)
All
All
T <
290 < T <
T =
300 < T <
T >
290
300
300
310
310
kRS
(ppb)
0.30-
0.30
0.05
(a)
0.10
(b)
0.60
kEN
(ppb)
0.30
0.11
0.025
(a)
0.050
(b)
0.20
kXSHC
(min"1)
500.
250.
750.
750.
750.
750.
750.
(a) Obtained by linear interpolation between values given for T = 290 K
and T = 300 K.
(b) Obtained by linear interpolation between values given for T = 300 K
and T = 310 K.
127
-------
which reflects the magnitude of the continuous radical source and which is
adjusted so model simulations fit observed rates of NO oxidation and ozone
formation in alkane-NOx-air runs; kEN, which reflects the NOX offgassing
rates and which is adjusted so simulations fit observed rates of ozone and
PAN formation in acetaldehyde-air runs, and kx<>Hc» which reflects the
extent of excess NO-to-N02 conversions caused by reactive organic con-
taminants and which is determined based on fits to ozone formation rates
observed in pure air photolyses. Since simulations of these characteriza-
tion runs are affected at least to some extent by the full set of charac-
terization parameters, this optimization has to be carried out in an
iterative manner.
This use of model simulations to determine rates of some of the
chamber-dependent processes means that the values derived will depend, at
least to some extent, on the light characterization model used in the
model simulations. Therefore, since the light characterization model for
the UNC chamber was modified during this study, we also had to re-derive
the adjustable chamber-dependent parameters used for modeling data in this
chamber.
Figure 5 shows distribution plots of discrepancies between model
predictions and experimental results for the chamber characterization runs
used to derive the adjustable chamber-dependent parameters for the UNC
chamber. These include distributions of fits to (1) maximum ozone yields
in pure air irradiations, (2) changes in NO and N02 concentrations in NOX-
air and CO-NOx-air experiments, (3) maximum ozone and PAN yields in
acetaldehyde-air experiments, and (4) NO oxidation and ozone formation
rates observed in alkane-NOx-air runs. Also shown are fits to maximum
ozone yields and NO oxidation and ozone formation rates in propene-NOx-air
runs; although modeling these runs was not used to derive the adjustable
chamber-dependent parameters, the results of modeling these runs give a
good indication of the overall performance and variability of the model in
simulating a relatively large number of similar experiments carried out
over a number of years.
Distributions of fits for five sets of model simulations are shown in
Figure 5. The first four sets were all carried out using the latest SAPRC
mechanism (Carter et al., 1988), but with varying light and chamber
characterization models. The first set (designated "original") employed
128
-------
Set Mechanism
Light Model
Other Chamber Effects
1
2
3
14
5
SAPRC
SAPRC
SAPRC
SAPRC
RADM
Carter et al. (1986)
New UNC Model
New
New
New
Experimental
Set 1
UNC Model
UNC Model
UNC Model
Set 2
Carter et al. (1986)
Carter et al. (1986)
This
This
This
work,
work,
work,
Air Ri in*
Set 3
T- independent kRS
T-dependent k^,
T-dependent k^,
.' k£N
''EN
kEN
Set H
Set
5
-0
33 -
0.27 -
0.21 -
0.15 -
-0.09 -
-0.03 -
0.03 -
0.09 -
0.15 -
0.21 -
0.27 -
-0.33
-0.27
-0.21
-0.15
-0.09
-0.03
0.03
0.09
0.15
0.21
0.27
0.33
0.33
111
1
11
111
11 i
Figure 5. Comparison of Results of UNC Characterization Simulations Using
Various Mechanisms and Chamber and Light Characterization
Models.
(continued)
129
-------
Set Mechanism Light Model
Other Chamber Effects
1 SAPRC
2 SAPRC
3 SAPRC
4 SAPRC
5 RADM
Calculated-
Experimental
< -0.33
-0.33 - -0.27
-0.27 - -0.21
-0.21 - -0.15
-0.15 - -0.09
-0.09 - -0.03
-0.03 - 0.03
0.03 - 0.09
0.09 - 0.15
0.15 - 0.21
0.21 - 0.27
0.27 - 0.33
> 0.33
. .
Experimental)
/Experimental
< -1.00
-1.00 - -0.82
-0.82 - -0.64
-0.64 - -0.45
-0.45 - -0.27
-0.27 - -0.09
-0.09 - 0.09
0.09 - 0.27
0.27 - 0.45
0.45 - 0.64
0.64 - 0.82
0.82 - 1.00
> 1.00
Carter et al. (1986) Carter et
New UNC Model Carter et
New UNC Model This work
New UNC Model This work
New UNC Model This work
- (1) Butane-NOx-Air and (2) Branched
Set 1 Set 2 Set 3
1 1 1
1
1
1 1 11
1
22 - 122 - 12
2
1
12 1 1
2 122 122
_.-.-.-. f 1 \ Ru fa no UfW Air anH O\
Average Initial d((
Set 1 Set 2 Set 3
.
111 111 11
-- 122 - 122 - 1122
12 12
1122 12 1
2
al. (1986)
al. (1986)
, T-independent k^, kEN
, T-dependent k^, kEN
, T-dependent k,^, kEN
Alkane-NOx-Air: Maximum Ozone -
Set 4 Set 5
.
:1 1
;
: 2
: 12
:112 1
-:11 - 112
:2
:1 12
:2
:2
:
1031 - [N0]))/dt
Set 4 Set 5
2
1112 1112
- 1112 - 11122
22
Figure 5 (continued) - 2
130
-------
Set Mechanism
Light Model
Other Chamber Effects
1 SAPRC
2 SAPRC
3 SAPRC
1 SAPRC
5 RADM
Experimental
< -0.33
-0.33 - -0.27
-0.27 - -0.21
-0.21 - -0.15
-0.15 - -0.09
-0.09 - -0.03
-0.03 - 0.03
0.03 - 0.09
0.09 - 0.15
0.15 - 0.21
0.21 - 0.27
0.27 - 0.33
> 0.33
( Pa 1 r*i 1 1 af Art
Experimental)
/Experimental
< -0.50
-0.50 - -0.11
-0.11 - -0.32
-0.32 - -0.23
-0.23 - -0.11
-0.1*4 - -0.05
-0.05 - 0.05
0.05 - 0.11
0.11 - 0.23
0.23 - 0.32
0.32 - 0.11
0.11 - 0.50
> 0.50
Carter et al. (1986) Carter et
New UNC Model Carter et
New UNC Model This work
New UNC Model This work
New UNC Model This work
Set 1 Set 2 Set 3
111
1 - - 1
111 11
11 1111
11
Set 1 Set 2 Set 3 '
11
1
1
1 1
1
1 1
1111 11111
al. (1986)
al. (1986)
, T- independent k^, k^
, T-dependent kR£, kg^
, T-dependent k^, kEN
.
Set 1 Set 5
11
1 11
- 11111 - 11
MS- ₯ay
-------
Set Mechanism
Light Model
Other Chamber Effects
1 SAPRC
2 SAPRC
3 SAPRC
U SAPRC
5 RADM
Calculated-
Experimental
< -0.20
-0.20 - -0.16
-0.16 - -0.13
-0.13 - -0.09
-0.09 - -0.05
-0.05 - -0.02
-0.02 - 0.02
0.02 - 0.05
0.05 - 0.09
0.09 - 0.13
0.13 - 0.16
0.16 - 0.20
> 0.20
Calculated-
Experimental
< -0.20
-0.20 - -0.16
-0.16 - -0.13
-0.13 - -0.09
-0.09 - -0.05
-0.05 - -0.02
-0.02 - 0.02
0.02 - 0.05
0.05 - 0.09
0.09 - 0.13
0.13 - 0.16
0.16 - 0.20
> 0.20
Carter et al. (1986)
New UNC Model
New UNC Model
New UNC Model
New UNC Model
----- (1) NOx-Air and (2)
Set 1 Set 2
22 22
122 122
1 12
22 - 22
11111112 111111112
1112 11
NOx-Air and CO-N
Set 1 Set 2
:22 :22
: :
:
i ;
:
: 1 :111
--:11122 -:1112
:12 -.111222
: 11 12 :112
: 11 1122 :12
: ;
* *
: :
Carter et
Carter et
This work
This work
This work
CO-NOx-Air:
Set 3
-.22
J
;
:2
:12
:122
-:22
al. (1986)
al. (1986)
, T- independent ko<
, T-dependent kj^,
, T-dependent k^,
Change in NO/lni
Set <4
:22
*
J
:2
:2
: 11122
-: 11111122
I' EN
it
EN
I*
kEN
Hal UfW .... .
Set 5
:22
*
;
:2
:2
: 11 122
-: 1111 1122
-.111111111 : 1 1 1 : 1 1 1
:1
:
;
:
j
:
:
*
:
Ox-Air: Change in N02/lnitial
Set 3
22
111
112
11122
- 1112
12
2
Set H
22
112
1111122
- 1122
11
12
*
;
j
*
:
UQv
Set 5
22
112
1111122
- 1122
11
12
Figure 5 (continued) - 4
132
-------
Set Mechanism
Light Model
Other Chamber Effects
1 SAPRC
2 SAPRC
3 SAPRC
14 SAPRC
5 RADM
Calculated-
Experimental
< -0.33
-0.33 - -0.27
-0.27 - -0.21
-0.21 - -0.15
-0.15 - -0.09
-0.09 - -0.03
-0.03 - 0.03
0.03 - 0.09
0.09 - 0.15
0.15 - 0.21
0.21 - 0.27
0.27 - 0.33
> 0.33
(Calculated-
Experimental)
/Experimental
< - 1 . 00
-1.00 - -0.82
-0.82 - -0.611
-0.64 - -0.45
-0.45 - -0.27
-0.27 - -0.09
-0.09 - 0.09
0.09 - 0.27
0.27 - 0.45
0.45 - 0.64
0.64 - 0.82
0.82 - 1.00
> 1.00
Carter et al . (1986) Carter et
New UNC Model Carter et
New UNC Model This work
New UNC Model This work
New UNC Model This work
Set 1 Set 2 Set 3
. f
; ;
: ;
:111 :
:11 :11 111
:11 :1111 11
:1111111 -:111 - 111111
:1111 :11111 11111
:11 :111 111
:1 :1111 11
:1 : 1
: :1
al. (1986)
al. (1986)
, T-independent kRS, k£N
, T-dependent k^, k^N
, T-dependent kR^» ^£y
Mil Y *! flfll 1 fTi f*i*^ rtn A
Set 4 Set 5
11
11111 111
1111 11111
- 11111 - 111
1 111
1111 11
11 1
1 111
Set 1 Set 2 Set 3
1111 1111 1111
11111111 - 11111111111 - 1111111
1111111 111 1111
11 111 111
1 1 1
Set 4 Set 5
;
;
111111 111111
111 - 111111111111- 11 m 1 1 1 1 i 1 1
11 i
1 11
Figure 5 (concluded) - 5
133
-------
the Carter et al. (1986) model for both light characterization and chamber
effects, and reflected our best estimates prior to this program. The
second set employed the new UNC light characterization model but the
original (Carter et al., 1986) model for the other chamber effects. The
third and fourth sets employ the new UNC light model and two new sets of
values for the adjustable chamber-dependent parameters, discussed below.
The fifth set employed the same light and characterization input as the
fourth set, but was carried out using the RADM mechanism.
Figure 5 shows that none of these chamber models yields close fits to
experimental results for all of the characterization experiments modeled;
there were always runs where significant discrepancies are obtained. This
can be attributed to run-to-run variation of chamber effects due to
unknown or uncontrolled factors. Improved fits could obviously be obtain-
ed by carrying out run-to-run optimization of the adjustable parameters,
but such a procedure would not be consistent with our objectives. Until
all significant chamber effects are identified and quantified so that they
can be predictively modeled, the best that can be hoped for is that the
biases and errors in the model simulations of these characterization runs
is as small as possible.
A comparison of the first two sets of distributions in Figure 5 shows
that updating the light characterization model without changing the repre-
sentation of the other chamber effects tends to cause a slight increase in
the tendency of the model to overpredict maximum ozone yields. Although
this effect does not appear to be large, it has a non-negligible effect on
the model performance statistics when simulating ozone yields in the full
set of UNC chamber experiments. The updated light model with the old
values of the adjustable parameters also results in significant overpre-
dictions of PAN yields in acetaldehyde-air irradiations, indicating that
the assumed value for the NOX offgassing rate in the UNC chamber may be
too high if the new light model is correct. The use of an apparently high
NOX offgassing rate in the chamber model could contribute to a tendency of
the model to overpredict maximum ozone yields. On the other hand, use of
the new light characterization model does not necessarily indicate that
the assumed magnitudes of the chamber radical source need to be modi-
fied. The fits of model simulations to NO oxidation rates in alkane-NO -
A
air experiments are sensitive to the representation of the chamber radical
-------
source, but use of the new light model without changing other parameters
was found not to significantly degrade the performance of the model in
simulating these results.
Better fits of model simulation to results of acetaldehyde-air
experiments were obtained if the magnitude of the NOX offgassing parameter
kg-jj was reduced by approximately a factor of three. This adjustment of
kEN required reducing by a factor of two the value of kXSHC, the parameter
measuring the amount of excess NO-to-NOp conversion caused by background
or contaminant reactive organics, so the model would still fit the maximum
ozone yields observed in the pure air irradiations. The third set of
distributions shows the performance of the model after these minor
modifications are made. It can be seen that these changes result is a
significant improvement in terms of the overall biases in fits to ozone
and PAN yields in the acetaldehyde-air runs and also a slight reduction of
the tendency of the model to overpredict ozone yields in the propene-NOx-
air experiments.
The original chamber characterization model of Carter et al. (1986)
neglected any effects of temperature on the magnitudes of the chamber
radical source or the NOX offgassing rate parameters, and these effects
are also ignored in the re-adjusted model discussed above. However, in
the process of re-assessing this chamber characterization model, we noted
that there was a consistent tendency for the model to overpredict reac-
tivity in runs with relatively low average temperatures (below -290-300
K). This suggests that the chamoer radical source in the UNC chamber may
be temperature dependent. In addition, the modeling of the lower tempera-
ture acetaldehyde-air runs suggests a temperature dependence in the NOX
offgassing rates in that chamber. Therefore, we investigated whether
using a modified chamber effects model with temperature dependences for
the parameters k and k^ might result in reduced discrepancies in model
simulations of the UNC chamber experiments.
The number of UNC characterization runs carried out at a variety of
temperatures is not sufficient to unambiguously derive temperature
dependencies for the radical source or for NOX offgassing. Therefore, we
tried a variety of assumed temperature dependencies to determine which
would minimize the overall average error in the distribution of the fits
to the runs used to characterize these effects. The model simulations
135
-------
suggested that runs with average temperatures below -300 had significantly
lower radical sources and NOX offgassing rates than runs with the more
typical average temperatures of -300-305, but that runs with still higher
average temperatures did not appear to have correspondingly higher radical
sources. This suggests that these chamber effects increase relatively
rapidly with temperature at temperatures below around 300 K, but increase
slower at higher temperatures. To represent this, we assumed that k^ and
knij increased linearly with temperature both above and below 300 K, but
that the rate of change is less in the higher temperature range. The
dependences which we found gave the best fits of those we tried are given
in Table 18. To obtain satisfactory fits to maximum ozone yields in the
pure air irradiations, it was also necessary to use a larger value of
kXSHC' The *ast set of< Pl°ts in figure 5 show the distribution of fits
obtained for the UNC characterization runs using this set of temperature-
dependent chamber characterization parameters.
Figure 5 shows that using the temperature-dependent model for k and
k yields somewhat better fits to ozone yields observed in the acetalde-
hyde-air experiments, and significantly better fits to NO oxidation rates
and ozone yields in alkane-NOx-air runs, compared to using the adjusted
model where temperature dependences of these parameters are ignored. It
also gives better distributions of fits to NO oxidation rates and ozone
yields in the simulations of the propene runs. However, it should be
noted that the number of runs useful for deriving temperature dependencies
for the radical source and the NOV offgassing rates in the UNC chamber is
A
extremely limited, and thus the temperature dependencies derived for them
in this study must be considered to be highly uncertain. In addition,
there is uncertainty in the temperature dependencies of the reactions
involved in PAN formation (Gery et al., 1988), and use of different
assumptions in the mechanism in this regard, such as those employed in the
latest Carbon Bond mechanism (Gery et al., 1988) may result in differences
in the apparent temperature dependencies of k being derived by model
simulations of the acetaldehyde-air experiments. . (On the other hand, this
uncertainty in the temperature dependence of PAN formation reactions will
have less of an impact on the values of kRg which are derived from the
model simulations of the alkane-NO-air experiments.) Nevertheless,
because of this significantly better performance in the temperature-
136
-------
dependent chamber effects model in our simulations of these data, this is
the model which was used in this study in simulating the UNC chamber runs
for the evaluation of the RADM mechanism.
It should be emphasized that this derivation of adjustable chamber-
dependent parameters by model simulation means that the parameters derived
will depend on the chemical mechanism used for this purpose. In this
study these parameters were derived primarily based on calculations using
the latest SAPRC mechanism (Carter et al., 1988). This is because the
SAPRC mechanism was used in our previous chamber characterization modeling
work, and because the SAPRC mechanism has a more explicit representation
of the reactions of the alkanes than does the RADM mechanism. In addi-
tion, since the RADM mechanism is being evaluated in this study, it was
not considered to be in its final form at the time these chamber charac-
terization calculations were being carried out. Ideally, the chamber
characterization runs used to derive the adjustable parameters should only
involve the "known" reactions, so differences in current mechanisms would
not affect modeling of such runs. However, this is not yet entirely the
case.
The differences between the RADM and the current SAPRC mechanism in
modeling these UNC runs used for chamber characterization is shown in
Figure 5, where the distributions of the fits calculated using these two
mechanisms with the new UNC light and chamber characterization models can
be compared. It can be seen that, as expected, there are essentially no
differences between the mechanisms in the simulations of the pure air,
N0x-air, and CO-NOx-air runs, but there are slight differences in the
simulations of the propene-NOx runs, and relatively larger differences in
the simulations of the alkane-NOx-air and the acetaldehyde-air runs. The
differences in the simulations of the alkane-NOx-air runs is expected
since the RADM mechanism uses a more lumped representation of the alkanes
than the SAPRC mechanism, and strictly speaking is not designed to be used
for modeling such runs. The differences in the simulations of the
acetaldehyde-air runs are due to differences between the mechanisms in the
rate constants used for the decomposition and formation of PAN, which are
such that PAN is predicted by the RADM mechanism to be more stable
relative to N02 + acetyl peroxy radicals. This results in the RADM mech-
anism predicting lower ozone yields and higher PAN yields in acetaldehyde-
137
-------
air irradiations than predicted by the SAPRC mechanism. With the kRS
parameters adjusted to give optimum fits using the SAPRC mechanism, this
results in the RADM mechanism underpredicting ozone and overpredicting
PAN, relative to the experimental results. However, either increasing or
decreasing kRS from the optimum for the SAPRC mechanism would cause
greater discrepancies for either PAN or ozone, respectively, with the RADM
model. Because of this, it is considered to be unlikely that using the
RADM mechanism to derive the adjustable chamber-dependent parameters would
have yielded significantly different values than those used in this study.
M.3 Representation of Organics in Chamber Experiments
Many of the environmental chamber experiments used in this study
include runs containing organic compounds, either singly or in mixtures,
which are not represented explicitly in the RADM mechanism. The appro-
priate procedures for representing such compounds when simulating chamber
experiments depends on the intended application of the mechanism being
evaluated. If the mechanism is intended to be a fully detailed mechanism
to test our ability to understand and predictively model the chemical
system being simulated, or if the mechanism is to be used as a standard or
"master mechanism" against which more condensed mechanisms will be
compared, or if the mechanism is to be used to assess relative reactivi-
ties of individual organic compounds, then the approach should be to
adjust the mechanism so that it represents the chemical conditions of the
mixture being simulated as closely as possible. For example, the SAPRC
detailed mechanism is designed to serve both as a basis for developing
more condensed mechanisms and for use in reactivity assessment studies,
and thus its evaluation involved adjusting the kinetic and mechanistic
parameters for lumped model species based on the compounds in the experi-
ments being simulated (Carter, 1988). If, on the other hand, the mechan-
ism being evaluated is a condensed mechanism which is intended to be used
as a component of a larger model such as RADM, then the appropriate
procedure is to represent the compounds in the-chamber experiments in the
same way they would be represented when the mechanism is implemented in
the larger model. Otherwise, the mechanism being evaluated in the chamber
simulations is not the same as the mechanism which will be used in its
intended application. Since the mechanism being evaluated in this study
138
-------
is intended to be a condensed mechanism for use as a component of the
RADM-II model, and since as far as we are aware it is not anticipated it
will be used as a basis for developing more condensed mechanisms or for
assessing relative reactivities of different organics, it is obviously the
latter approach that is appropriate for this study.
Therefore, the approach we used when simulating chamber experiments
to evaluate the RADM mechanism was to process the organic input data in
the chamber simulations in the same way that we anticipate the organic
emissions data will be processed for input into the RADM-II model. The
RADM emissions input processing procedure we anticipate will be used is
the two-step procedure described in Section 3 of this report. This
involves no adjustment of rate constants or mechanistic parameters in the
mechanism these are the same regardless of the mixture being simu-
lated. On the other hand, in some cases model species concentrations are
adjusted to account for differences in reactivities between them and the
input species they represent. The application of this procedure to the
simulations of the chamber runs is briefly summarized below.
The first step in processing the organic input data involves aggre-
gating, on a mole-per-mole basis, the individual chemicals into the 32-
class emissions grouping discussed in Section 3.2.1. Table 7 in that
section indicates how various types of chemicals are aggregated in this
step. As indicated there, some of the groupings are determined based on
the OH radical rate constants assumed for the individual organic com-
pounds, and the specific sets of OH radical rate constants used in making
these assignments in this study are those listed by Carter (1988a). Other
than the UNC auto exhaust experiments, no experiments modeled in this
study contained unknown or unidentified compounds. The initial organic
concentrations used in the UNC auto exhaust experiments in our 1986 mech-
anism evaluation study (Carter et al., 1986) were also employed in this
study. [The alkanes contained in these experiments are given in terms of
the two lumped alkane classes, C^-C^ and C^. The lumped class C1|-C5 was
aggregated into emissions class "Alkanes (0.25-0.5 react)," and the lumped
class C6+ was aggregated into "Alkanes (0.5-1 react)", based on the OH
radical rate constants used for them in the SAPRC/ERT mechanism (Lurmann
et al., 1987).] Therefore, there are no cases where chemicals in the
chamber experiments were aggregated to the "unknown" or "unassigned"
categories.
139
-------
The organics aggregated into the emissions classes were then aggre-
gated into the species in the RADM mechanism as discussed in Section 3.2.3
and as indicated in Table 8 in that section. Some emissions classes are
aggregated on a mole-per-mole basis, while others are aggregated using
"reactivity weighing," based on estimates of how much of the emissions
class and the model species will react in the model simulation. Note,
however, that the aggregation of classes based on "reactivity weighing"
require use of the parameter "INTOH" in the derivation of the weighing
factors. Since INTOH reflects the effective integrated radical levels in
the simulation, it is a model input variable and not a parameter in the
mechanism. As such, the weighing factors shown in Table 8, calculated
using the INTOH value of 110 ppt-min, which was estimated to be
appropriate for use in three-day RADM simulations (Stockwell, private
communication, 1988), are not necessarily appropriate for use in the
chamber simulations. Therefore, separate estimates for INTOH values were
made for each chamber.
The INTOH values used for calculating reactivity weighing factors in
the simulations of the chamber experiments carried out in this study were
as follows:
Chamber INTOH (ppt-min)
SAPRC EC 35
SAPRC ITC 40
SAPRC OTC MO
UNC 70
These were derived based on a rough analysis of average integrated OH
radical levels in model simulations of one-day chamber experiments employ-
ing various complex mixtures of organics. (The model simulations employed
either the SAPRC detailed mechanism, or the RADM mechanism with the rate
constants of the lumped species adjusted based on the compounds present in
the experiments the results were similar in either case.) The higher
INTOH value for the UNC chamber can be attributed to the longer irradia-
tion time used in that chamber, where runs start at sunrise and end at
sunset, and are usually carried out in the summer. The SAPRC runs are
usually limited to six hours.
140
-------
Strictly speaking, the integrated OH radical levels, and thus the
appropriate INTOH values, will depend on the type of experiment, with
lower values for runs with high levels of radical inhibitors, etc., but
this type of run-by-run analysis was not carried out. The simulations of
only a relatively small number of runs are affected to any significant
extent by reactivity weighing, since, of the experiments modeled in this
study, reactivity weighing is of importance only for those containing
significant amounts of methanol or benzene. For the other compounds in
these experiments, the weighing factors are either unity or very close to
unity. For this reason, we did not consider a comprehensive investigation
of the most appropriate INTOH value to use in the chamber simulations to
be worthwhile.
To provide an indication of the level of inaccuracy introduced into
the model simulations by using this aggregation approach, a separate
series of simulations of the chamber experiments was carried out where the
rate constants for the lumped model species were adjusted for each experi-
ment based on the compounds present. The rate constants which were
subject to adjustment when this procedure was employed were those for the
reactions of OH radicals with HC3, HC5, HC8, TOL, XYL, OLI, and OLT, and
for the reactions of ozone and NO? radicals with OLI ana OLT. The rate
constants for the individual organics used when this procedure was
employed are those assigned to the detailed model species in the latest
SAPRC mechanism, as given by Carter (1988). In most cases, the results
were not significantly different than when the standard RADM aggregation
approach was employed, tending to indicate that the aggregation approach
proposed for RADM is probably satisfactory. The results of these compari-
sons are given in more detail in Section 5.6.
-------
5. EVALUATION AGAINST ENVIRONMENTAL CHAMBER DATA: RESULTS
The results of the simulations of the chamber experiments using the
March 1988 and the recommended modified versions of the RADM mechanisms
are described in this section. The organization of this evaluation
against the chamber data is discussed first, then the measures of model
performance are described, and finally the results of the evaluations are
summarized and discussed. The results of the simulations using the March
1988 version of the RADM mechanism (designated simply as the "RADM" mech-
anism) are described first and in the most detail. Then the differences
in the results of the simulations using the recommended modified (RADM-M)
and the recommended modified parameter (RADM-P) versions of the mechanism
are discussed.
Because of the large number (over 550) of chamber runs which were
used in this evaluation, the discussion of the results of these simula-
tions focuses primarily on statistical measures of model performance for
the various groups of experiments. However, concentration-time plots for
selected experiments are also shown for illustrative purposes. For a
listing of all the individual experiments used in this evaluation, and for
summaries of the performances of the mechanisms on a run-by-run basis, the
reader is referred to Appendices A through C of this report. Appendix A
lists all of the chamber experiments which were modeled in this study,
together with their initial NOX and organic concentrations, their average
temperatures and water concentrations, and their maximum ozone yields and
average initial rates of NO oxidation and ozone formation. (An "*" after
the temperature or water concentration indicates that it was was not
measured or that the data was unreliable, and the default for that chamber
was used in the model calculations.) Appendix A also gives for each
experiment the performance of the March 1988 RADM mechanism in simulating
the maximum ozone yields and the average rates of NO oxidation and ozone
formation. Appendices B and C give similar listings of the run-by-run
performance of the RADM-M and RADM-P mechanisms, respectively, for all of
the experiments whose results differed from those obtained with the RADM
mechanism.
-------
5.1 Organization of the Evaluation
The mechanism testing was organized in accordance with the hierarchy
of species shown in Figure 6. This schematic shows the hierarchical
nature and interdependence of the major components of atmospheric photo-
chemical mechanisms. This hierarchy was recommended by the EPA Workshop
on Evaluation and Documentation of Chemical Mechanisms (EPA, 1987) and has
been used successfully in past evaluations. The mechanism testing
followed this hierarchy, beginning with the lowest levels (i.e., the
inorganic species) and proceeding to the highest level by stepwise addi-
tion of species with increasingly complex chemistry. The stepwise feature
of this approach is designed to facilitate identification of the uncertain
components of chemical mechanisms, and minimize the occurrence of
compensating errors in the mechanisms and fortuitous agreement between
models and chamber data. In concept, IT a substantial disagreement
between a model and data occurs at a high level in the hierarchy and the
model has shown to perform well for experiments involving species lower in
the hierarchy, then it is most likely that the new higher level chemistry
is in error and not the whole mechanism.
The chamber data base was subdivided into six major groups for test-
ing purposes. The first group included -20 irradiations of pure air, NOX-
air. and carbon monoxide-MOv-air mixtures. The purpose of simulating
A
these experiments was to test the accuracy and precision of both the
inorganic reactions in the mechanism and the assumptions used to represent
chamber-dependent parameters. Since the level of organics is very low in
these experiments, the rate of NOV oxidation is very sensitive to chamber
A
characteristics, such as the intensity and spectral distribution of
radiation, the rates of radical offgassing from chamber walls, and the
levels of background organic contaminants.
The second group of runs consisted of formaldehyde - air and
acetaldehyde - air mixtures. These runs were simulated primarily to
further test the submodels for the treatment of chamber effects. As
discussed earlier, the acetaldehyde-air runs are particularly useful for
testing how well NOV offgassing from chamber walls is represented in the
A
model.
-------
Ve'.-'C'e
£ inane
>C2 Ai^c^es ! ' >C2 A^er-es
n-gr.e-
Ccrbor
I NO, N02, N03, N205
! HN05. HCNC
i OH, H02, -202
Figure 6. Hierarchy of Species for Mechanism Testing.
144
-------
The third group consisted of -20 single carbonyl - NO mixtures.
These included runs with formaldehyde, acetaldehyde, propionaldehyde,
acetone, or methyl ethyl ketone. These were simulated next to test the
carbonyl chemistry, since the chemistry of the remaining compounds depends
on these species.
The fourth groups consisted of the approximately 200 single hydro-
carbon - NOX experiments. The 91 alkene data sets included runs for
ethene, propene, 1-butene, trans-2-butene, isobutene, 1-hexene, isoprene,
and alpha-pinene. The 45 single alkane-NOx data base included runs for
ethane, n-butane, three branched alkanes (2,3-dimethylbutane, isopentane,
and iso-octane), six long chain alkanes (n-pentane, n-hexane, n-heptane,
n-octane, n-nonane, and n-pentadecane), and cyclohexane. The aromatic
experiments included runs for benzene, toluene, m-xylene, o-xylene,
mesitylene, tetralin, naphthalene, and dimethyl naphthalene. These were
simulated to test the performance of the subgroups of hydrocarbon reac-
tions incorporated into the mechanisms.
The fifth group consisted of -40 runs for simple mixtures of organics
and NOV. By "simple mixtures" we mean mixtures of two or more organic
A
species either from a given class (alkenes, alkanes, and aromatics) or
from two classes, but not from all three classes. The simple mixture runs
were simulated to test for compensating errors that might occur when
compounds from different hydrocarbon classes in the mechanism were
present.
The sixth group consisted of over 200 runs involving complex mixtures
of organic species and NOV, where one or more hydrocarbon species from
A
each major class was included in the mixtures. These runs contained at
least one alkene, alkane, and aromatic i.e., at least one representa-
tive from each of the major classes of organics. The majority of the
mixtures included 3 to 8 organic species. However, some mixtures, such as
those employing actual automobile exhaust, included a large number of
compounds. The complex mixtures used in these experiments are surrogates
for typical atmospheric mixtures of organics. (For this reason, they are
also designated as "surrogate" runs.) The purpose of simulating a large
number of runs of this type was to assess model performance as robustly as
possible for the type of mixtures expected in atmospheric applications of
the mechanism.
145
-------
As indicated above, the mechanisms which were evaluated included the
March 1988 version of the RADM mechanism, the recommended modified mech-
anism (RADM-M) and the recommended modified parameter mechanism (RADM-P).
The evaluation of the March 1988 RADM mechanism is discussed in the most
detail since evaluation of this mechanism was the primary focus of this
study. The recommended modifications involve no changes in the lower
portions of the hierarchy where only the inorganic reactions or the reac-
tions of the carbonyls (other than the carbonyls formed only from the
aromatics) are involved; therefore, the results discussed below for the
RADM mechanism concerning the characterization and the carbonyl runs are
applicable to the modified versions of the mechanism. The results of the
evaluation for the runs containing compounds higher in the hierarchy are
discussed separately for the three versions of the mechanism.
5.2 Measures of Model Performance
The rates of decay of precursor species and the rates of formation
and the yields of key product species are important for assessing chemical
mechanism performance. The selection of appropriate performance measures
depends on the type of experiment. The most important distinction among
experiments in this data base is whether or not significant amounts of
ozone were observed. In simulations of the N0x-air and CO-NOx-air runs,
which generally do not generate much ozone, the total changes in NO and
N02 concentrations over the course of the experiments are the most useful
measures of performance. These provide a coarse measure of the rate of
precursor (NO) oxidation and product (N02) formation in these generally
"slow" experiments.
In the organic-NOx-air runs, where significant ozone is formed, the
relative and absolute errors in the maximum ozone concentration and the
average rate of change of [0^] minus [NO] during the first half of the
time period to reach maximum experimental ozone are the most important
measures of performance. The latter parameter (designated "dQOgl-
[N0])/dt" in the tabulations, and referred to variously as the "NO oxida-
tion and ozone formation rate" or simply as the "NO oxidation rate")
provides a relatively direct indication of the rate of NO oxidation by the
peroxy radicals (H02, R02» and RCOq) that are produced by the organics.
This is illustrated using the kinetic rate equations for the important
-------
reactions of NO and ozone, as shown below:
N02 + hv ---- > NO + 03
NO
N0
2 +
NO + R02 ---- > N02 + H02 + Carbonyls
(1)
(2)
(3)
where:
d([03])/dt = ^ [N02] - k2 [N0][03]
d([NO])/dt = k1 [N02] - k2 [N0][03] - k^ [MO][R02]
By subtraction, one obtains the rate of NO oxidation by peroxy radicals:
d([03]-[NO])/dt =
[NO][R0],
We have found this to be a robust measure of the overall timing of the
simulations. When the rate of NO oxidation is slow, the time to the NO-
N02 cross-over, the time to the N02 maximum, and the rates of organic
precursor decay are almost always slow and vise-versa.
This measure is also a useful measure of the rate of ozone formation
in experiments containing no initial NOV, such as the pure air or
X
carbonyl-air runs. However, in this case, it is referred to as d[0-a]/dt,
or the average initial rate of ozone formation.
For experiments where formaldehyde and PAN are formed, the errors for
the maximum predicted concentrations are also reported. These species are
important intermediates and products in the photo-oxidation cycle.
However, when evaluating a condensed mechanism such as RADM, it should be
recognized that the model species called "PAN" actually refers to PAN +
higher acyl peroxynitrates, and thus calculated "PAN" yields are not
always directly comparable with experimental PAN measurements.
147
-------
We have adopted an evaluation approach were we examine the errors in
both the timing parameter and the maximum concentrations unpaired in time,
rather than examining Just the error in the predicted maximum concentra-
tions at the time of the observed maximum (i.e., paired in time). While
our approach is less stringent than the latter, it measures the aspects of
performance which we believe are most important, i.e., the yields and
timing of the simulations. The focus on maxima unpaired in time is also
often used in evaluation of Eulerian photochemical models against ambient
data (Burton, 1988).
In this study, the maximum ozone statistics were only compiled for
those experiments where the maximum experimental ozone concentration
exceeded 40 ppb. In experiments with lower ozone yields, relatively small
errors and interferences in instrumental readings would constitute a
relatively high percentage of the total yield and would not be a meaning-
ful measurement of model performance. In addition, performance statistics
on ozone yields for runs with less than approximately 40 ppb of ozone
(approximately the global background) are not meaningful in the context of
environmental chamber modeling where maximum ozone concentrations usually
range from 100 to 1000 ppb.
The statistical parameters reported for each group and subgroups of
runs are the mean experimental value, the mean calculated value, the mean
bias, the mean normalized bias, the mean error (or gross error), and the
mean normalized error. -The latter are defined below in accordance with
the American Meteorological Society's recommendations for air quality
model evaluation (Fox, 1981).
1
Mean Absolute Bias (ppb) = £ (C* - E*
N i=1,N
1
Mean Absolute Error (ppb) = £ \C= - Ez|
N l=l,N
100 (Ci - E^
Mean Normalized Bias (1) = £ [ ]
N 1=1,N 0.5 (Ci + Ei)
1118
-------
100 \Ci -
Mean Normalized Error (%) = I [
N i=1,N 0.5 (Ci + Et)
where C- = Calculated concentration, Ei = Experimental concentration, and
the summations are over all N experiments in the group or subgroup. The
standard deviations of the mean normalized bias is also reported.
In terms of our philosophy of evaluating models, we believe the test-
ing should cover a wide range of conditions to "stress" the mechanism and
to facilitate identification of situations where the mean bias and/or
error are so large that one would have to reject the mechanism. A mechan-
ism can be Judged acceptable if it doesn't fail any of the tests. The
performance levels at which one decides to accept or reject a mechanism
are subjective. Nevertheless, it is useful to define such a criteria so
that mechanism performance can be judged consistently. Our criteria for
rejection is a mean normalized bias in excess of 20% and a mean normalized
error in excess of 50%. These are referred to below at the minimum
desired performance criteria. In addition, because the results for any
one experiment are not nearly as meaningful as those for a group of
experiments, one needs to determine the minimum number of experiments that
must be simulated to determine whether or not a mechanism is performing
satisfactorily. We believe a minimum of 10 to 15 experiments is needed to
reasonably assess model performance for a particular type of run.
5.3 Performance on Characterization Runs
The performance of the RADM mechanism, combined with our current
chamber effects submodels, in simulating the N0x-air and CO-NOX air runs
is summarized in Table 19. Run-by-run summaries of these results are
given in Appendix A. (These results are also applicable to the recommend-
ed modified versions of this mechanism.) This table (and Appendix A)
shows the performance statistics for the total changes in NO and N02 con-
centrations during the runs. Figures 7 and 8 graphically compare the
change in [NO] and [N02] for each run, respectively. In the N0x-air runs,
the model slightly underpredicts the absolute changes in NO and N02 con-
centrations, but in the CO-NOX air runs, the model overpredicts the
absolute changes in these values. On the average for all of these runs,
the model overpredicts the changes in NO by 9% with an average error of
-------
and underpredicts the changes in NC^ by 10/1 with an average error of
45J. The graphical comparisons illustrate that the model performs well on
the majority of these characterization runs. This suggests the combina-
tion of the chamber characterization procedures and the inorganic
chemistry in the RADM mechanism provides moderately accurate predictions
for these types of experiments. This is despite the fact that the chamber
characterization model was derived using the SAPRC and not the RADM
chemical mechanism. This level of performance is as good (or better) than
reported elsewhere (Carter et al., 1986) and meets our desired performance
criteria.
Table 19. Selected Performance Statistics in the RADM Simulations of the
N0v-Air, CO-NO-Air, Pure Air, and Carbonyl-Air Runs
A A
Run Type
No. Experimental -Absolute-
Runs Avg. (sdev) Bias Err.
Normalized
Bias (%) Err. (%)
N0x-Air
C0-N0v-Air
A
23
23
13
13
1) Change in NO (ppm).
2) Change in N02 (ppm)
0.06 (0.49)
-0.14 (0.56)
-0.11 (0.10)
0.06 (0.09)
-0.004 0.038
-0.005 0.025
-0.016 0.019
0.005 0.018
Pure Air
Formaldehyde-Air
Acetaldehyde-Air
1) Maximum Ozone (ppm)
2) Maximum Pan (ppb)
6 0.11 (0.05) -0.005 0.008 -4 (12) 9
8 0.18 (0.06) 0.042 0.042 20 (13) 20
18 0.16 (0.16) -0.014 0.039 -12 (30) 26
18 19 (13) 3 6 19 (40) 34
150
-------
RADM Mechanism
p
5.
3; 0.1
^ 0
c
X" NOx Air
; *
/' CO-NOx Air
^
t/^m
& -o.m ./"" :
X ft* !
- /. ;
^ I- .«'
0-2
-Q
D
1 -0.3
"5
0 _nA
A x' 1
-
. / \
/
/ m
/ ,.,...,. , . i
-0.4 -0.3 -0.2 -c.* : - o/ :
Experimental Change in [NO] (ppmj
Figure 7. Calculated vs Experimental Changes in [NO] in N0x-Air and
C0-N0x-Air Runs.
RADM Mechanism
NO A>
0.2 r-
-0.2 -0.1 0 0.1 0.2 0.3
Experimental Change in [N02] (ppm)
Figure 8. Calculated vs Experimental Changes in [N02] in N0x-Air and
C0-N0x-Air Runs.
151
-------
The performance of the RADM mechanism and our chamber effects model
in simulating ozone formation in pure air and carbonyl-air experiments is
also summarized in Table 19- Run-by-run listings of these results are
given in Appendix A. (Again, these results are also applicable to the
recommended modified mechanisms.) As discussed in Section M, the pure air
runs primarily test the levels of background reactive organics and the
rate of NOX off gassing used in the chamber effects submodels. The
carbonyl-air runs, especially the acetaldehyde-air runs, are good tests
for the NOX offgassing rates because the ozone and PAN predictions are
very sensitive to NOX offgassing. The performance statistic indicate that
on the average there is relatively little bias and error (9» in maximum
ozone predictions for pure-air runs. Ozone concentrations are somewhat
overpredicted in the formaldehyde-air and underpredicted in the acetalde-
hyde-air runs. The maximum PAN concentrations tend to be overpredicted in
the acetaldehyde-air runs.
Comparisons of the calculated and experimental maximum ozone yields
and ozone formation rates in the pure air and carbonyl-air runs are shown
in Figures 9 and 10, respectively, and a histogram of the normalized
biases in maximum ozone is shown in Figure 11. There is a fair amount of
scatter in these fits with biases for ozone that are fairly evenly
distributed between -40? and +50?. A plot of the error in maximum ozone
vs. the initial HC/NOX ratio is shown in Figure 12. There is no system-
atic dependence of the ozone prediction errors on this important ratio;
this is a positive result since it probably means there is no systematic
source of error in the mechanism. We suspect that the scatter is probably
primarily due to uncontrolled or incompletely characterized chamber
effects (Carter et al., 1986).
Overall, the RADM mechanism's performance on the pure-air and
carbonyl-air is considered to be reasonably acceptable, given the uncer-
tainties in the chamber characterization model.
152
-------
RADM Mechanism
600
/I HCHO Air
C 100 200 300 4-00 ' 500 600
Experimental Ozone Maximum (ppb)
Figure 9. Calculated vs Experimental Maximum Ozone in the Pure Air and
Carbonyl-Air Runs.
RADM Mechanism
100 r
HCHC-Air
100
Experimental d([03]-[NO])/dt (ppb/hr)
Figure 10. Calculated vs Experimental Rate of Ozone Formation in the
Pure Air and Carbonyl-Air Runs.
153
-------
RADM Mechanism
/
6
2 5
3
(T
4- 4
O
-------
5.M Performance of the RADM Mechanism on Single Organic Runs
Model performance statistics for maximum ozone and the NO oxidation
rate for the March 1988 RADM mechanism are shown in Table 20 for all of
the organic-NOx runs. Run-by-run listings of these results are given in
Appendix A. In general, these results are applicable only to the March
1988 RADM mechanism. However, except for the aromatics-NOx and (for the
RADM-M mechanism) the C<-+ akane-NOx runs, the performance for the
recommended modified mechanisms was found to be the same or very similar.
The cases where the recommended modified mechanisms provide different
results are discussed in Section 5.7.
In the following discussion, concentration-time plots for selected
representative experiments will be presented. In all cases, the "*"
symbols are used to designate experimental data, and solid lines are used
to designate model calculations using the March 1988 RADM mechanism. (For
experiments where the predictions of the recommended modified mechanisms
are noticeably different, the plots also include results of model simula-
tions using those mechanisms. These are indicated by dashed lines, with
the longer dashes being for the RADM-P mechanism and the shorter dashes
being for RADM-M.)
5.4.1 Carbonyl-NO^ Runs
Tabulations of the performance of the RADM mechanism in
simulations of maximum ozone and NO oxidation rates for carbonyl-NOx runs
are given in Table 20, and graphical displays of these results are shown
in Figures 13 through 16. The majority of runs in these groups are for
formaldehyde. On the average, there is only +2% bias in the maximum ozone
and the NO oxidation rate predictions for the formaldehyde runs. The
error in the maximum ozone and timing parameter predictions are +21? and
+ 13J, respectively, on the average. The performance of the RADM mechan-
ism's formaldehyde chemistry is considered acceptable by the criteria
discussed in Section 5.2.
The results for acetaldehyde runs show underprediction of both the
maximum ozone (-37J mean bias) and the rate of NOX oxidation (-18J mean
bias). An example of one of the UNC Acetaldehyde-N0x runs is shown in
Figure 17. The ozone results for propionaldehyde, acetone, and methyl
ethyl ketone show biases of -31t, +52J, and -19>. There does not appear
155
-------
Table 20. Performance Statistics on Maximum Ozone and on NO Oxidation
and Ozone Formation Rates in the Simulations of the Organic-
N0x-Air Runs Using the RADM Mechanism
Organic
or
Mixture
Formaldehyde
Acetaldehyde
Propionaldehyde
Acetone
Methy lethy 1 ketone
Ethene
Propene
1-Butene
trans-2-Butene
Isobutene
1-Hexene
Isoprene
a-Pinene
1)
2)
No.
Runs
15
15
5
5
2
2
1
1
2
2
14
14
49
49
11
11
4
4
1
1
4
4
13
13
4
4
Statistics for Maximum Ozone (ppra)
Statistics for d( [03]-[NO])/dt (ppb/min)
Experimental
Avg. (sdev)
0.47
12.67
0.60
1.73
0.84
1.79
0.23
0.48
0.61
1.19
0.91
5.23
0.61
3.62
0.42
2.86
0.28
6.17
0.90
8.84
0.34
1.57
0.72
5.66
0.35
0.97
( 0.58)
(27.92)
( 0.40)
( 0.29)
( 0.15)
( 0.05)
( 0.00)
( 0.00)
( 0.06)
( 0.09)
( 0.20)
( 4.10)
( 0.24)
( 2.60)
( 0.26)
( 2.18)
( 0.17)
( 2.80)
( 0.00)
( 0.00)
( 0.25)
( 1.04)
( 0.25)
( 4.06)
( 0.11)
( 0.43)
Absolute
Bias Err.
0.03
0.89
-0.17
-0.29
-0.23
-0.41
0.16
0.04
-0.11
0.27
-0.09
-1.21
0.03
0.26
0.04
0.20
-0.08
-1.50
-0.19
-3.25
0.26
2.95
-0.23
0.86
-0.03
0.09
0.09
1.71
0.17
0.29
0.23
0.41
0.16
0.04
0.11
0.27
0.16
1.21
0.09
0.57
0.13
0.81
0.09
1.68
0.19
3.25
0.26
2.95
0.24
1.81
0.08
0.17
Normalized
Bias (%) Err. (%)
2 (
9 (
-37 (
-18 (
-31 (
-26 (
52 (
9 (
-19 (
19 (
-16 (
-27 (
5 (
6 <
-2 (
0 <
-52 I
-25 I
-23 I
-45 I
57 (
82 i
-42 i
-2
-8
4
30)
22)
17)
11)
20)
1)
0)
0)
1)
17)
30)
: 20)
; 20)
: 19)
; 56)
: 32)
: 38)
; 26)
[ 0)
[ 0)
[ 25)
( 25)
( 3D
( 36)
( 35)
( 19)
21
13
37
18
31
26
52
9
19
19
23
27
15
14
45
25
54
31
23
45
57
82
43
26
28
14
(continued)
156
-------
Table 20 (continued) - 2
Organic
or
Mixture
Ethane
n-Butane
C4+ Br. Alkane
C5+ n-Alkane
Cycloalkane
Benzene
Toluene
Xylene
Mesitylene
Tetralin
Naphthalene
2,3-Dimethylnaph.
Simple Mixtures
Simple Surrogates
1)
2)
No.
Runs
1
1
27
27
7
7
14
14
4
4
6
6
20
20
10
10
8
6
5
5
5
5
4
4
40
39
36
36
Statistics for Maximum Ozone (ppm)
Statistics for d( [Oo]-[NO] )/dt (ppb/min)
Experimental
Avg. (sdev)
0.24 (
1.32 (
0.31 (
1.53 (
0.28 (
0.93 (
0.20 (
1.15 (
0.05 (
0.92 (
0.29 (
3.39 (
0.34 <
3.07 (
0.44 (
5.39 <
0.52 I
9.50 I
0.29 i
1.89 i
0.20 i
1.59
0.32
2.48
0.49
4.10
0.37
2.81
0.00)
0.00)
0.27)
0.98)
0.18)
0.46)
0.15)
0.75)
0.00)
0.12)
: 0.14)
: 2.30)
; 0.15)
: 1.87)
: 0.19)
: 4.14)
[ 0.26)
( 4.97)
I 0.23)
( 0.88)
( 0.08)
( 0.41)
( 0.04)
( 0.51)
( 0.27)
( 4.06)
( 0.15)
( 1.61)
Absolute
Bias Err.
-0.04
0.08
0.02
0.33
-0.05
-0.18
0.10
0.22
0.01
0.06
0.08
1.22
0.09
0.04
-0.02
-0.41
-0.18
-6.64
0.11
12.58
0.28
7.04
-0.01
0.13
-0.05
-0.87
0.02
0.33
0.04
0.08
0.13
0.81
0.10
0.20
0.10
0.35
0.02
0.19
0.08
1.22
0.12
0.83
0.08
1.28
0.22
6.64
0.13
12.73
0.28
7.04
0.05
0.47
0.13
1.09
0.05
0.90
Normalized
Bias (%} Err. (J)
-19 (
6 (
0 (
19 (
-17 (
-22 (
42 (
24 (
33 (
3 (
22 (
36 (
24 <
7 (
-4 (
0 (
-39 I
-93 i
-6 i
106 i
84
134
-6
1
-12
-13
7
8
0)
0)
59)
50)
67)
24)
28)
37)
0)
29)
15)
19)
: 43)
: 34)
: 25)
: 35)
: 50)
[ 32)
( 62)
( 85)
( 40)
( 13)
( 24)
( 24)
( 45)
( 26)
( 24)
( 38)
19
6
46
43
53
26
42
34
33
19
22
36
33
27
18
25
48
93
43
119
84
134
17
19
34
22
15
27
(continued)
157
-------
Table 20 (continued) - 3
Organic
or
Mixture
SAPRC
SAPRC
7-HC Surgs.
8-HC Surgs.
UNC Misc. Surgs.
UNC "
UNC "
Synurban"
Synauto"
UNC Auto Exhaust
Syn.
Syn.
Jet Fuel
Jet Exhaust
1)
2)
No.
Runs
11
11
82
82
21
21
11
11
15
15
28
28
8
8
4
4
Statistics for Maximum Ozone (ppm)
Statistics for d( [03]-[NO])/dt (ppb/min)
Experimental
Avg. (sdev)
0.60 (
7.74 (
0.45 (
2.81 (
0.50 (
1.32 <
0.37 <
1.09 (
0.62 <
1.81 I
0.57 I
1.68 1
0.75 i
3-47 <
0.78 i
6.66 i
: 0.17)
: 5.16)
: 0.27)
: LSI)
: 0.17)
: o.4o)
; o.28)
: 0.47)
; 0.18)
; 0.63)
[ 0.27)
[ 0.61)
[ 0.10)
[ 0.74)
( 0.13)
( 3.48)
Absolute
Bias Err.
-0.01
-0.38
0.04
-0.27
0.06
0.00
0.13
0.07
0.13
0.02
0.06
-0.08
0.11
0.38
-0.09
-0.58
0.04
0.65
0.09
0.54
0.12
0.13
0.13
0.09
0.14
0.28
0.09
0.19
0.11
0.61
0.09
0.58
Normalized
Bias (%) Err. (%)
0 (
-5 (
9 (
-10 (
17 (
2 (
39 (
7 (
23 <
2 l
8 1
-6 1
14 l
11 i
-11 i
-8 i
: n)
: 12)
: 26)
: 23)
: 40)
: 15)
; 3D
: 8)
: 17)
; 18)
; 29)
: 15)
: 12)
; 16)
( 8)
[ 4)
8
10
20
19
27
11
39
8
23
14
19
13
14
16
11
8
158
-------
RADM Mechanism
200
.= 80C i-
X i
I I
a, 50C -
0 200 IOC 6CC 500 1000 12
Experimental Ozone Maximum (ppb
Figure 13. Calculated ys Experimental Maximum Ozone in Carbonyl-N0x
Runs.
R.ADM Mechanism
f;
G
2 30
20
20 30 50 100 200 300
Experimental d([03]-[NO])/dt (ppb/hr)
Figure 14. Calculated vs Experimental Rates of NO Oxidation and Ozone
Formation in the Carbonyl-N0x Runs.
159
-------
RADM Mechanism
4
OT
c
3
a:
-------
1.0-1
OZONE
0.25-,
NITRIC OXIDE
1
300 420 540 660 780 900 1020
NITROGEN DIOXIDE
T 1 1 1 1 1 1 ~] 1 "]
300 420 540 660 780 900 1020
PAN
Q.
CL
o
<
tr
UJ
o
~z.
o
o
1 1 ' [ ' I ' I ' I ' I '
300 420 540 660 780 900 1020
1.0-1
ACETALDEHYDE
0.3-,
0.2-
0.1-
300 420 540 660 780 900 1020
FORMALDEHYDE
1 I 'I ' I ' I ' I ' I '
300 420 540 660 780 900 1020
0.0
I ' I I I ' I l
300 420 540 660 780 900 1020
EUPSED TIME (minutes)
Figure 17. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of UNC Acetaldehyde-NO Run
AU2482B.
161
-------
to be any systematic dependence of the errors on the initial HC/NOX
ratio. While the simulation results for the higher aldehydes and acetone
do not meet our criteria for acceptable performance, there are too few
runs with these compounds to draw meaningful conclusions regarding the
adequacy of the chemistry. Nevertheless, we are concerned with the large
biases for these simulations and recommend additional testing when more
data become available.
5.4.2 Alkene-NCL Runs
The result for tests of the RADM mechanism against the alkene-
NO runs are shown in Table 20 and in Figures 18 through 21. Selected
A
concentrations versus time plots for one of the best simulations of the
UNC ethene-NOx experiments are shown in Figure 22.
The statistics for ethene runs show a tendency for the mechanism to
underpredict maximum ozone. On the average, it underpredicts the ozone by
16? with 23? error. Also, it underpredicts the rate of NOV oxidation in
A
ethene runs by 27% on the average. The RADM mechanism's underpredictions
are primarily a result of its not including ethene's reaction with atomic
oxygen, which we have found is important in these high concentration
experiments. Test calculations where this reaction is included in the
mechanism significantly improves its performance in simulating these
results; with its performance being comparable to that of the SAPRC
mechanisms (Carter et al., 1986; Carter, 1988). However, calculations
using the set of test problems discussed in Section 6 of this report
indicate that neglecting these reactions has a negligible effect under
ambient conditions. For that reason, this reaction is not included in the
recommended modified mechanisms. However, its omission does limit the
range of validity of this mechanism to a concentration regime below that
of these chamber experiments.
There are a large number of propene-NOv runs in the data base and the
A
RADM mechanism performs very well on these runs. Results from two typical
propene-NOx simulations are shown in Figures 23 and 2U. The mechanism
predicts the maximum ozone and rate of NOX oxidation with a +5? bias and
15? error on the average. The results for 1-butene simulations also show
a small mean bias (-2?), but the mean error (45?) is much higher than for
the propene runs. Figure 25 shows a fairly typical simulation of a UNC 1-
butene-NOx experiment where the ozone is overpredicted by -30?. (PAN is
162
-------
RADM Mechanism
1400
p 100C
X
2 800
i
O 600 -
o i
°...vx
T-2-5utene
**
./"
^r
A »
0 200 4CO SCO 300 1CCC '2CO 14QC
Experimental Ozone Moxinaum (ppb)
Figure 18. Calculated vs Experimental Maximum Ozone in the Alkene-N0x
Runs for the RADM Mechanism.
RADM Mecnanism
20 50 100 200 500 1000
Experimental d([03]-NO])/dt (ppb/hr)
Figure 19. Calculated ys Experimental Rates of NO Oxidation and Ozone
Formation in the Alkene-N0x Runs for the RADM Mechanism.
163
-------
RADM Mechanism
OT
C
tr
5 -
-
__
.
,*
\ '-' 1
IP
*.
I 1
T-° ,
vO
.
O
vO
^0
'/,
^
n
IP
//t
/^
*p
{: | Ethene
1^1 Bu tones
\/\ Hexene
Bias in Maximum Ozone (%)
Figure 20. Histogram of Normalized Biases in Maximum Ozone in Alkene Runs
for the RADM Mechanism.
RADM Mechanism
g 100
1)
c
0
N 50
E
3
C n
U
X
0
2
c
". -50
a
5
inn
0
A
a
A
A
*^ ,
sr^'V * * *
. V ":*""-
*
1 *
A *
-
Ethene
A
Propene
*
Butenes
A
Hexene
o
0.2 0.5 1 2 5
HC/NOx Ratio
10
20
Figure 21. Normalized Bias in Maximum Ozone vs HC/NOX in Alkene Runs
for the RADM Mechanism.
-------
OZONE
NITRIC OXIDE
Q.
a
z:
o
I
LJ
O
-z.
o
o
0.30 -,
0.24-
0.18-
0.12-
0.06-
300 420 540 660 780 900 1020
NITROGEN DIOXIDE
0.00
0.8-1
0.6-
0.4-
0.2-
300 420 540 660 780 900 1020
FORMALDEHYDE
0.0-
I ' : ! ' | '' ' I
300 420 540 660 780 900 1020 300 420 540 660 780 900 1020
1.0-1
ETHENE
0.0
I IT I I I
300 420 540 660 780 900 1020
ELAPSED TIME (minutes)
Figure 22. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of UNC Ethene-NO Run
OC0584B.
165
-------
OZONE
NITRIC OXIDE
E
Q.
CL
g
or:
o
120 240 360
NITROGEN DIOXIDE
0.0
120 240 360 480
ELAPSED TIME (minutes)
Figure 23. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of the Propene-N0v Run
EC-216.
166
-------
O.B-i
0.6-
0.4 H
0.2-
OZONE
NITRIC OXIDE
0.0
1
r
300 420 540 660 780 900 1020
NITROGEN DIOXIDE
E
Q.
Q_
Ld
O
z.
o
o
0.04-
-
/
°-QO
I
I
I
300 420 540 660 780 900 1020
PROPENE
0.4
j
0.2 J
0.1 -
0.0
0.15-1
i
0.12-1
0.09-
0.06-
0.03-
I
0.00-i-
300 420 540 660 780 900 1020
PAN
0.2
300 420 540 660 780 900 1020
ACETALDEHYDE
0.2J
0.1-
4
8.0E-02-
4.0E-02-
300 420 540 660 780 900 1020
0.0
^\^ i : : ; : i
300 420 540 660 780 900 1020
ELAPSED TIME (minutes)
Figure 24. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of UNC Propene-NO Run
JL2983B.
167
-------
a
a
Ld
o
z
o
o
0.8-
o.e-
OZONE
300 420 540 660 780 900 1020
NITROGEN DIOXIDE
0.0
' I ' 1 i i
300 420 540 660 780 900 1020
0.8-n
0.6-
0.4-
1 -BUTENE
0.4-,
0.3-
NITRIC OXIDE
0.24-.
0.18-
0.12-
0.06-
300 420 540 660 780 900 1020
PAN
°-0°
//
11 iiri]i
300 420 540 660 780 900 1020
300 420 540 660 780 900 1020
ELAPSED TIME (minutes)
Figure 25. Experimental and Calculated Concentration-Time Plots for
*?& 1 ,d ** « n rf O «*«*.*[.>. i r* t ,
of UNC 1-Butene-NOx Run
168
-------
greatly overpredicted in this simulation because the model species PAN
represents peroxypropionyl nitrate as well as PAN.) Figure 21 clearly
shows that there is a lack of systematic dependence of the biases on the
initial HC/NOX ratios in these runs. There are a small number of runs for
other butenes and 1-hexene. The mechanism's performance on these runs is
poor. The ozone yields and rate of NOV oxidation in the trans-2-butene
A
and isobutene runs are underpredicted. Conversely, they are both over-
predicted by almost 60% on the average in the 1-hexene runs.
Clearly, the propene portion of the mechanism perform well and easily
meet the desired performance criteria. The 1-butene results also meet the
criteria, although they are not as accurate. The internal alkene portion
of the RADM mechanism do not perform well, but once again, this is based
only on five experiments. The mechanism does not perform as well as other
mechanisms in the simulations of the ethene, runs because the concentration
regime in these runs is beyond the range of validity of this mechanism.
Test calculations show that if the 0(^P) + ethene reaction were added, the
mechanism would perform better in simulating the ethene runs.
5.4.3 Alkane-NOH Runs
The model performance results for alkane-NOv runs are shown in
A
Table 20 and in Figures 26 through 29. Typical simulations of n-butane,
n-octane, and 2,3-dimethylbutane experiments are shown in Figures 30
through 32, respectively. The simulation of the one ethane-NOx system
underpredicts the maximum ozone by 18%. For the butane experiments, the
mechanism shows no bias but has a fairly large error of 46/1. For the >C4
branched alkanes, the mean bias and error are -17% and 53%, respective-
ly. However, for the >C4 long chain alkanes, the. mean bias and error are
both 42%. For the cyclohexane runs, the mechanism predicts ozone with a
bias and error of 33% on the average. Overall, for all of the alkane
experiments, the RADM mechanism is able to predict the maximum ozone with
a relatively low bias but a relatively high error. The statistics for the
rate of NOX oxidation are only slightly better. Figures 26 through 29
confirm that there is a large amount of scatter in the results. Figure 29
shows that there does not appear to be any systematic dependence of the
bias on initial HC/NOX ratios, which is reassuring.
169
-------
RADM Mechanism
1000
.a
a.
a.
x
O
c
O
N
O
O
O
800 -
600
400
200
0 200 400 600 800 1000
Experimental Ozone Maximum (ppb)
Figure 26. Calculated vs Experimental Maximum Ozone in the Alkane-N0x
Runs for the RADM Mechanism.
RADM Mechanism
£ 500
\
.C
Q.
vCX 200
^ 200
cT
z
^ 100
I"1
^
T5 50
T3
° 30
O
o
Butane
A
>C- 3rcnct-.ec
>C4 Long Chain
Cyclic Alkanes
A
20
20 30 50 100 200 300 500
Experimental d([03]-[NO])/dt (ppb/hr)
Figure 27. Calculated vs Experimental Rates of NO Oxidation and Ozone
Formation in the Alkane-N0x Runs for the RADM Mechanism.
170
-------
RADM Mechanism
1^
12
£ 10
rr
*- B
O
-------
OZONE
0.0
0.06
0.04
0.02-
0.00
NITRIC OXIDE
0 120 240 360
NITROGEN DIOXIDE
480 600
*_
*
120 240 360 480 600
PAN
E
a
a
z
a:
CJ
2:
o
o
0.00
1 I'I'Ir
120 240 360 480
600
n-BUTANE
FORMALDEHYDE
120 240 360 480 600
- 1 ' 1 r .
0 120 240 360 480 600
ELAPSED TIME (minutes)
Figure 30. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of n-Butane-NOx Run EC-130
-------
OZONE
E
Q.
Q.
or
u
o
z
o
o
NITRIC OXIDE
"I 1 1 1 1 1
0 60 120 180 240 300 360 420
0 60 120 180 240 300 360 420
0.12-
0.09 -I
0.06-
0.03-
0.00-
NITROGEN DIOXIDE
_",
=
'
~1 1 I 1 1 1 1
60 120 180 240 300 360 420
0.003-,
0.002-
0.001 -
O.OOO-i
PAN
I
60
120 180 240 300 360 420
EUVPSED TIME (minutes)
Figure 31. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of n-Octane-NO Run ITC-552.
A
173
-------
OZONE
Q_
Q.
UJ
.
o
-------
We do not believe the large error in the results indicates that there
is necessarily a problem in the mechanism because model simulations of
alkane-NOv runs are very sensitive to the chamber effects parameters. The
A
variability in the chamber effects could account for a significant amount
of the variability in the mechanism's predictions for alkane runs. As
noted in previous evaluations (e.g., Carter et al., 1986), it is not
possible to unambiguously evaluate the alkane portion of the mechanism on
chamber data because of the uncertainties in chamber effects and vari-
ability. Therefore, within the limitations of the testing methodology,
the alkane reactions appear to give acceptable predictions for ozone and
the NOV oxidation rate.
A
5.4.4 Aromatic-N
-------
RADM Mechanism
O
0 200 400 600 800 1000
Experimental Ozone Maximum (ppb)
Figure 33. Calculated ys Experimental Maximum Ozone in the Aromatic-
NCL Runs for the RADM Mechanism.
A
RADM Mechanism
000
.O
CL
_ 1000
V
-O
^ 500
rt
o
z
-U 200
C^
T^ 100
«-
^ 50
Benzene
*
Toluene
Xylenes
Mesitylene
A
Other
O 50 100 200 500 1000 2000
Experimental d([03]-[NO])/dt (ppb/hr)
Figure 31*. Calculated vs Experimental Rates of NO Oxidation and Ozone
Formation in the Aromatic-N0.x Runs for the RADM Mechanism.
176
-------
RADM Mechanism
12
cr
*-
o
^
I
"1
;-;-;-:';l Mesitylene
Xylenes
Toluene
Benzene
Bias in Maximum Ozone (%)
Figure 35. Histogram of Normalized Biases in Maximum Ozone in Aromatic
Runs for the RADM Mechanism.
RADM Mechanism
TL/U
g
-------
OZONE
E
Q.
Q.
g
i
LJ
O
2
O
O
0.5-,
NITRIC OXIDE
0 60 120 180 240 300 360
NITROGEN DIOXIDE
0.0
60 120 180
240 300 360
PAN
0 60 120 180 240 300 360
60 120 180 240 300 360
ELAPSED TIME (minutes)
Figure 31. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of the Benzene-NO Run
ITC-562 ( RADM, RADM-P, RADM-M^.
178
-------
OZONE
CL
Q.
v-x
z
o
I
uj
o
o
0.4-,
0.3 -I
0.2-
NITRIC OXIDE
0.0
"I I
0 60 120 180 240 300 360 420
0.3-
NITROGEN DIOXIDE
] - 1 - r
0 60 120 180 240 300 360 420
PAN
0.0-
0.6-
0.5 -,
0.4-
0.3-
0.2-
0.1-
0.0
i i 1 r
60 120 180 240 300 360 420
TOLUENE
0.06 -,
0.05-
0.04-
0.03-
0.02-
0.01 -
6 60 120 180 240 300
FORMALDEHYDE
360
~~1I I 1 1 1 1
60 120 180 240 300 360 420
0.00
6 60 120 180 240 300 360 420
ELAPSED TIME (minutes)
Figure 38. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of the Toluene-NO Run
179
-------
0.5 -,
OZONE
0.16-
0.12-
0.08-
0.04-
NITRIC OXIDE
. "' I'I'I
300 420 540 660 780 900 1020 1140
0.00-
i
300 420 540 660 780 900 1020 1140
NITROGEN DIOXIDE
PAN
a
a
or
UJ
o
z
o
o
1 I ' I' I ' I ' I ' I '1
300 420 540 660 780 900 1020 1140
TOLUENE
1 'I'I'1'
300 420 540 660 780 900 1020 1140
FORMALDEHYDE
I ' I ' I ' I ' I ' | ' ,
300 420 540 660 780 900 1020 1140
'I'I'I'
300 420 540 660 780 900 1020 1140
ELAPSED TIME (minutes)
Figure 39. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of the UNC Toluene-NO Run
JL3080R. ^
130
-------
OZONE
NITRIC OXIDE
E
Q.
a
g
i
LU
a
o
o
120 180 240 300
0 60 120
M-XYLENE
0.0
ELAPSED TIME (minutes)
Figure 40. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of the m-Xylene-NO Run
ITC-702.
181
-------
0.6-
0.5-
0.4-
0.3-
0.2-
0.1 -
0.0 -<
OZONE
NITRIC OXIDE
0.12-
0.08-
0.04-
300 420 540 660 780 900 1020 1140
0.00
300 420 540 660 780 900 1020 1140
ex
Q.
g
*
UJ
o
z
o
o
0.15-.
NITROGEN DIOXIDE
0.00
0.10-1
0.08-
0.06-
0.04-
0.02-
PAN
: I ;^ ; ! : ; T :
300 420 540 660 780 900 1020 1140
0.00-- i',-.
300 420 540 660 780 900 1020 1140
0.3-1
0.2-
0-XYLENE
0.20-,
FORMALDEHYDE
I T I ' ' ' : 'I f :
300 420 540 660 780 900 1020 1140
300 420 540 660 780 900 1020 1140
ELAPSED TIME (minutes)
Figure 41. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of the UNC o-Xylene-NO Run
JL3080B.
182
-------
0.5-,
OZONE
0.5-,
0.4
NITRIC OXIDE
0.0-
60 120 180 240 300 360 6 60 120 180 240 300 360
NITROGEN DIOXIDE
Q.
CL
g
S
UJ
o
~z.
o
o
:
V v« ..,...
0.0
0.3-,
0.2-
60 120 180 240 300 360
PAN
0.1-
60 120 180 240 300 360
0.3-»
0.2-
0.1 -
MESITYLENE
0.0-
0.08 -,
0.06-
0.04-
0.02-
FORMALDEHYDE
n 1 1
60 120 180 240 300 360
-f T T o.oo-
60 120 180 240 300 360
ELAPSED TIME (minutes)
Figure U2. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of the Mesitylene-NO Run
EC-901. x
183
-------
Considering the uncertainties in the yields and subsequent reactions
of the aromatic ring-opening products, this level of performance might be
considered acceptable. However, unlike the alkanes, the modeling of
aromatic runs should not be highly sensitive to uncertain chamber effects,
and cases of generally poor performance can indicate problems with the
mechanism. As discussed in Section 2, other mechanisms perform better in
simulating results of these experiments, and there are several changes
that can be made to the RADM aromatic mechanism which can reduce the error
considerably and reduce the systematic bias in ozone yields between
benzene and mesitylene. The modified RADM mechanisms, discussed in
Section 5.7, perform better in simulating the results of these runs.
5.4.5 Natural Hydrocarbon Runs
The result of tests on experiments with two natural hydro-
carbon species, isoprene and alpha-pinene, are shown in Table 20 and in
Figures 43 through 46. Comparisons of predicted and observed concentra-
tion profiles for two isoprene runs and one alpha-pinene run are shown in
Figures 47 through 49. Although there are some minor differences between
the recommended modified and the March 1988 versions of the RADM mechan-
isms in their reactions of the natural hydrocarbons, these differences
have no significant effect on the predictions for isoprene and only minor
effects on predictions for alpha-pinene. (The most important difference
is a slight change in the rate constant used for the reaction of ozone
with OLI, the lumped model species which is used to represent alpha-
pinene.) Therefore, the results discussed in this section are applicable
to all three versions of the mechanism.
In the case of isoprene, the mechanism has a significant tendency to
underpredict ozone yields, with an average bias of -42/L This level of
performance does not meet the desired performance criteria. On the other
hand, it has very little bias (-2%) in simulating the NO oxidation rate,
and the error (26/t) is only slightly higher than in the simulations of the
propene runs. This mechanism represents isoprene explicitly, so in
principle one should expect it to perform reasonably well in simulating
this compound. However, although the RADM mechanism uses the appropriate
rate constants for the initial atmospheric reactions of isoprene, it
represents the products of these reactions with the same set of species
184
-------
RADM Mechanism
Isoprene
A
Alpho-Pinene
200 400 600 800 1000 1200 1400
Experimental Ozone Maximum (ppb)
Figure 43. Calculated vs Experimental Maximum Ozone in the Natural
Hydrocarbon Runs for the RADM Mechanism.
RADM Mechanism
g- 1000
\ 500
o1
2^
JL, 200
c^
*T3 100
"5 50
c?
o
Isoprene
A
,50 100 200 500 1000
Experimental d([03]-[NO])/dt (ppb/hr)
Figure ^4. Calculated vs Experimental Rates of NO Oxidation and Ozone
Formation in the Natural Hydrocarbon Runs for the RADM
Mechanism.
185
-------
RADM Mechanism
c
o:
4-
o
-------
E
Q.
Q.
ce
h-
LJ
o
z
o
u
1.0-1
OZONE
0.5-.
0 60 120
NITROGEN DIOXIDE
180
240
0.4-,
0.3-
0.2-
0.1-
0.0-
0.15-,
NITRIC OXIDE
120
180
240
PAN
0.0-
60
ISOPRENE
120
180
240
0 60 120 180
FORMALDEHYDE
60
120
180 240 0 60 120
ELAPSED TIME (minutes)
180
240
Figure 47. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of the Isoprene-NO Run
ITC-811.
187
-------
1.0-,
OZONE
0.15-
NITRIC OXIDE
300 420 540 660 780 900 1020 300 420 540 660 780 900 1020
Q.
a
z.
or
I
bJ
O
Z
O
O
0.16-,
0.12-
0.08-
0.04-
NITROGEN DIOXIDE
0.00
0.12-1
0.08-
0.04-
PAN
0.00-
300 420 540 660 780 900 1020 300 420 540 660 780 900 1020
ISOPRENE
0.4-,
0.3-
0.2-
0.1-
FORMALDEHYDE
0.0
-iiIir"~iI'I'T "'" "1 ' T T" 1 ' I ' I ' I ' I
300 420 54-0 660 780 900 1020 300 420 540 660 780 900 1020
ELAPSED TIME (minutes)
Figure 48. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of the UNC Isoprene-N0x Run
JL1680B.
188
-------
OZONE
E
Q_
Q.
tr
.
UJ
o
-z.
o
u
0.16
0.12-
0.08-
0.04-
NITRIC OXIDE
0.00
300 420 540 660 780 900 1020 300 420 540 660 780 900 1020
0.16-,
0.12-
0.08-
0.04-
NITROGEN DIOXIDE
0.00
0.12-,
0.08^
0.04-
0.00
PAN
300 420 540 660 780 900 1020 300 420 540 660 780 900 1020
ELAPSED TIME (minutes)
Figure 49. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of the UNC a-Pinene-NO Run
JL1580B.
189
-------
used to represent the terminal alkenes. Preliminary results from our
laboratories indicate that using mechanisms with more explicit representa-
tions of the reactions of isoprene's major products will result in
somewhat better model performance, although there are still inconsisten-
cies between the model and the data (Carter and Atkinson, unpublished
SAPRC results). [In addition, approximately 40* of the products in the
isoprene-OH reaction system have not been unidentified (Atkinson,
unpublished SAPRC results).] Better performance of the RADM mechanism in
simulating isoprene runs would probably require adding species to the
mechanism to represent the reactions of isoprene's oxidation products, but
the mechanism would still be uncertain.
The performance of the RADM mechanism in simulating the limited
number of alpha-pinene runs, all carried out in the UNC chamber in July of
1980, is surprisingly good. The mean bias and error in the ozone maximum
are -8% and 28%, respectively, while the bias and error in the NO oxida-
tion rate are only +H% and 1MJ. (These four statistics are -8/t, 27%, -»-2%,
and 13%, respectively, for the RADM-P and RADM-M mechanisms.) This level
of performance is perhaps fortuitous, since the atmospheric chemistry of
alpha-pinene is not well understood and is represented by the same lumped
model species (OLD which is used to represent the internal olefins.
Indeed, as discussed in Section 3.3, the rate constants and product yield
parameters for this lumped species are derived based on analysis of
anthropogenic emissions data, where the contribution of terpenes is
negligible. Nevertheless, the alpha-pinene results for these four runs
meet the desired performance criteria. Of course, these results do not
necessarily indicate that the performance would be equally good for beta-
pinene or the other terpenes.
5.4.6 Performance in Simulations of Formaldehyde Yields
Model performance statistics for the maximum formaldehyde
concentrations are shown in Table 21. Comparisons of the individual
calculated and experimental values for all of the single organic runs are
shown in Figure 50. The maximum formaldehyde concentrations are
underpredicted on the average in carbonyl, aromatic, and natural
hydrocarbon runs and overpredicted by small amounts on the average in
alkene and alkane runs. The mean errors in the prediction are large: on
the order of 30 to 80% for most types of runs. Thus, there clearly is a
190
-------
Table 21. Performance Statistics on Maximum Formaldehyde in the
Simulations of the Organic-N0x-Air Runs Using the RADM
Mechanism
Compound or
Mixture
Acetaldehyde
Propionaldehyde
Ethene
Propene
1-Butene
trans-2-Butene
Isobutene
1-Hexene
Isoprene
a-Pinene
n-Butane
C4-t- Br. Alkane
C5+ n-Alkane
Cycloalkane
Toluene
Xylene
Mesitylene
Naphthalene
Simple Mixtures
Simple Surrogates
SAPRC 7-HC Surgs.
SAPRC 8-HC Surgs.
UNC "Synurban"
UNC "Synauto"
UNC Auto Exhaust
Syn. Jet Fuel
No.
Runs
5
2
14
39
8
4
1
4
13
2
21
5
10
4
20
10
8
5
34
7
11
82
11
13
28
8
Experimental(a)
Avg. (sdev)
0.17 (
0.10 (
0.86 (
0.36 (
0.32 (
0.08 (
0.64
0.33 (
0.30 <
0.08 (
0.04 (
0.03 (
0.01 I
0.07 I
0.09 i
0.09 i
0.09 i
0.01 -
0.25 '
0.13
0.38
0.18
0.12
0.16
0.18
0.05
0.
0.
0.
0.
0.
0.
: o.
! o.
: o.
; o.
; o.
: o.
[ o.
( o.
15)
01)
41)
20)
27)
06)
27)
16)
04)
03)
02)
.01)
.12)
.09)
( 0.05)
[ 0.04)
( 0.01)
( 0.20)
( 0.05)
( 0
( 0
( o
( o
( o
( o
.23)
.12)
.04)
.08)
.10)
.03)
Absolute(a)
Bias Err.
-0.
0.
-0.
0.
' 0.
0.
-0.
0.
0.
-0.
0,
-0.
-0
-0
-0
-0
-0
0
0
-0
0
-0
-0
0
-0
-0
05
01
02
00
10
04
,02
30
,01
,01
.01
.02
.01
.06
.04
.05
.05
.02
.02
.05
.09
.02
.05
.02
.10
.03
0.
0.
0.
0.
0.
0.
0.
0.
07
01
23
10
10
06
02
30
0.12
0.01
0.03
0.02
0.
0,
0
0
0
0
0
0
0
0
0
0
0
0
.01
.06
.04
.05
.05
.02
.09
.06
.09
.05
.05
.03
.10
.03
Normalized
Bias (%) Err. (%}
-5 (
10 (
-9 (
4 (
21 (
60 (
-3
76 (
-6 <
-24 1
35 <
-75 i
-26 i
-32
-48
-71
-75
100
-14
-45
18
-1
-55
13
-68
-58
56)
17)
41)
45)
; 18)
: 58)
: 33)
: 59)
; 12)
( 73)
( 68)
(110)
(133)
( 56)
( 51)
( 45)
( 64)
( 56)
( 45)
( 18)
( 47)
( 18)
( 3D
( 3D
( 62)
44
12
34
33
22
71
3
76
46
24
70
91
95
101
54
76
75
100
47
56
21
35
55
24
70
67
(a) Concentrations in ppm.
191
-------
RADM Mechanism
o
2_
Si
c
5
E
X
C
2
g
c
|
~
-
1
0.3
0.1
0.03
0.01
3.003
4 '' '
. *.'i< *
-IxV"***
* *Si 9^/1} * ^
5»» *
*1 /'*"""
a A"»,^tf 44 *
o o ^ 2 ° 4^/^:° °
* ^Vi * = 0
"x **** a
/
RCHC Air
A
Aldehydes
Alkenes
*
Alkanes
A
Aromotics
Biogenics
OOC1 0.003 0.01 003 O.1 C.3 1
Experimental HCHO Maximum (pom)
Figure 50. Calculated vs Experimental Maximum Formaldehyde in the
Single Organic-N0x Runs for the RADM Mechanism.
large amount of scatter in the formaldehyde predictions. While the
accuracy of the measurements is questionable, especially at the low levels
observed in some runs, it is unlikely that this is the primary cause of
the poor comparisons in all cases. Formaldehyde is difficult to predict
because it is both formed and destroyed in the photooxidation cycle. The
RADM mechanism and other mechanisms (e.g., Carter et al., 1986) are not
able to reproduce the observed formaldehyde levels well. If the same
desired performance criteria that were used for ozone and rate of NCL
oxidation were used for formaldehyde, the mechanism would clearly be
rejected. Fortunately, accurate prediction of formaldehyde is not the
primary purpose of the mechanism.
5.4.7 Performance in Simulations of PAN Yields
Model performance statistics for the maximum PAN concentra-
tions are shown in Table 22, and comparisons of the individual calculated
and experimental values for all of the single organic runs are shown in
Figure 51. These results show that the RADM mechanism predicts PAN well
192
-------
Table 22. Performance Statistics on Maximum PAN in the Simulations of
the Organic-N0x-Air Runs Using the RADM Mechanism
Compound or
Mixture
Acetaldehyde
Propionaldehyde
Methylethylketone
Propene
1-Butene
trans-2-Butene
Isobutene
Isoprene
n-Butane
C4+ Br. Alkane
C5+ n-Alkane
Toluene
Xylene
Mesitylene
2,3-Dimethylnaph.
Simple Mixtures
Simple Surrogates
SAPRC 7-HC Surgs.
SAPRC 8-HC Surgs.
UNC Misc. Surgs.
UNC "Synurban"
UNC "Synauto"
UNC Auto Exhaust
Syn. Jet Exhaust
No.
Runs
5
2
1
46
7
4
1
12
21
6
11
20
10
8
4
38
32
11
68
12
10
15
28
4
Experimental(a)
Avg. (sdev)
0.141 (0.088)
0.052 (0.036)
0.070
0.142 (0.084)
0.039 (0.016)
0.080 (0.044)
0.137
0.133 (0.082)
0.296 (1.012)
0.226 (0.506)
0.247 (0.814)
0.051 (0.036)
0.125 (0.102)
0.407 (0.189)
0.046 (0.019)
0.081 (0.059)
0.048 (0.020)
0.095 (0.047)
0.044 (0.030)
0.055 (0.046)
0.024 (0.024)
0.054 (0.039)
0.047 (0.029)
0.080 (0.03D
Absolute(a)
Bias Err.
-0.002 0.020
0.129 0.129
-0.001 0.001
0.044 0.051
0.054 0.055
-0.004 0.014
0.141 0.141
-0.029 0.080
-0.257 0.294
-0.207 0.224
-0.243 0.247
0.019 0.037
-0.041 0.043
-0.317 0.317
-0.001 0.015
0.018 0.033
0.004 0.018
0.035 0.035
0.039 0.040
0.007 0.024
0.012 0.012
0.014 0.029
-0.003 0.009
0.098 0.098
Normalized
Bias (%) Err. (%)
-1 ( 18)
110 ( 63)
-1
28 ( 39)
64 ( 54)
-14 ( 25)
68
-19 ( 85)
17 (104)
-44 (123)
7 (126)
39 ( 65)
-24 ( 47)
-124 ( 40)
-2 ( 44)
14 ( 56)
12 ( 57)
33 ( 13)
66 ( 34)
33 ( 69)
48 ( 30)
30 ( 43)
-14 ( 55)
78 ( 13)
14
110
1
34
69
21
68
61
89
104
98
61
26
124
33
48
39
33
67
55
49
44
36
78
(a) Concentrations in ppm.
193
-------
RADM Mechanism
,-v t /I RCHO Air
E 1 . ,-"
g; 0.3 f . **\"JL
"-' 1- _ * ,**+^ °
1 o.
E
X
i 0.03
Z
0 » >^ " o °
0 ^i^°0^*ft 0 0
° &* °0*^«* ^ °° *
" . * *>s*/ '
... .*/ °*
< ^ ^ ^,
& A. V
CL 0.01
n
\"'X** *
/ A
A
Aldehydes
.
Alkenes
Aikones
^
Arornatics
o
Biogenics
a>
5 0.003
o i
O O.OOi t
O.OC1 0.003 0.01 0.03 0.1 0.3
Experimental PAN Maximum (ppm)
Figure 51. Calculated vs Experimental Maximum PAN in the Single
Organic-N0x Runs for the RADM Mechanism.
in acetaldehyde runs, but poorly in most other runs. The model over-
predicts the PAN concentrations in most but not all types of runs, and
there is also a large amount of scatter in the PAN predictions. The
graphical comparisons confirm this. Cases of positive bias are at least
partially explained by the fact that PAN is used in the mechanism as a
surrogate for PAN plus many analogous organic peroxyacyl nitrate
species. However, this is not expected to be a factor for the propene,
isobutene and toluene runs. On the other hand, PAN is significantly
underpredicted in the mesitylene runs, and is slightly underpredicted in
the trans-2-butene. xylene, isoprene, and C4+ branched alkane runs (though
with large mean errors in the latter two cases.) If the same desired
performance criteria that were used for ozone and rate of NOX oxidation
were used for PAN, the mechanism would clearly be rejected. Given the
partial mismatch between the species that mechanism predicts and what is
measured, this criteria is probably too strict.
194
-------
5.5 Performance of the RADM Mechanism on Runs with Organic Mixtures
Model performance statistics for maximum ozone and the NO oxidation
rate for the March 1988 RADM mechanism are shown in Table 20 for the runs
containing mixtures of organics. Run-by-run listings of these results are
given in Appendix A. Graphical displays of these results are shown in
Figure 52 through 55 for the simple mixture runs, and in Figures 56
through 61 for the complex and surrogate mixture runs. Concentration
versus time profiles for a typical simple mixture run are shown in Figure
62, and similar plots are shown in Figures 63 through 75 for a variety of
representative complex mixture simulations, including several multi-day
simulations of ITC and OTC experiments. [These plots employ the same
notation as those in the previous section the "*" symbols designate
experimental data, solid lines designate data calculated using the March
1988 RADM mechanism, longer dashed lines designate results for the RADM-P
mechanism, and shorter dashed lines are results for the RADM-M
mechanism.] The results discussed in this section are applicable only to
the March 1988 RADM mechanism, though the performance of the modified
mechanisms in simulating experiments not containing aromatics are similar.
The results for the simple mixture runs show the mechanism tends to
underpredict the maximum ozone by 12£ on the average with a mean error of
34J. The NOX oxidation rates are underpredicted by a similar amount on
the average in these runs. The results for the surrogate mixture runs
show overprediction of the maximum ozone by approximately 10J on the
average with a mean error of approximately 2Q%. The mean bias and errors
in the NO oxidation rates for these runs are slightly better than this on
the average.
Since this data set has over 200 experiments for mixtures that
include a wide range of organic compounds and range of hydrocarbon and NOX
concentrations, the evaluation statistics are fairly robust. The model
performance statistics for ozone and NOX oxidation rate are good,
especially for the surrogate mixture runs. The mean error in ozone in the
surrogate mixture simulations is less than that for most of the single
organic compound simulations particularly the aromatics and alkanes.
This suggests that there are some compensating errors in the mechanism,
which is true of most photochemical reaction mechanisms. (Also, for runs
with mixtures of organics, any error in the mechanism for any particular
195
-------
RADM Mechanism
1200
0 200 400 600 800 1000 1200
Experimental Ozone Maximum (ppb)
Figure 52. Calculated vs Experimental Maximum Ozone in the Simple
Mixture Runs for the RADM Mechanism.
RADM Mechanism
20 50 100 200 500 1000
Experimental d([03j-[NO])/dt (ppb/hr)
Figure 53 Calculated ys Experimental Rates of NO Oxidation and Ozone
Formation in the Simple Mixture Runs for the RADM Mechanism.
196
-------
RADM Mechanism
12
10
in
§ 8
a:
Bias in Maximum Ozone (%)
Figure 5^. Histogram of Normalized Biases in Maximum Ozone in the Simple
Mixture Runs for the RADM Mechanism.
RADM Mechanism
100
50
V
c
a
o
E
D
E
X
O
.£ -so
-------
RADM Mechanism
0 200 400 600 800 1000 1200
Experimental Ozone Maximum (ppb)
Figure 56. Calculated vs Experimental Maximum Ozone in the Surrogate
Mixture Runs for the RADM Mechanism.
RADM Mechanism
_u
O
(J
20 50 100 200 500 1000
Experimental d([03]-[NO])/dt (ppb/min)
Figure 57. Calculated vs Experimental Rates of NO Oxidation and Ozone
Formation in the Surrogate Mixture Runs for the RADM
Mechanism.
198
-------
RADM Mechanism
~HJ
40
OT
C
K 30
"o
i_
z
10
n
.
i } j
j£i
r~
H
*v>-
/,
^
-_
x>
^
//
^
_^.
^
/ /
\^
^
\\
//
//
^
v\
A
«
^
^\
^
i .
J"T
s\
\^
\
^->
[: j 3 Comp.
jivx::] 7 Comp.
JH:KKJ 8 Comp
^ UNC Mixes
p^Auto
|y^] Syn Fuels
Bias in Maximum Ozone (%)
Figure 58. Histogram of Normali2ed Biases in Maximum Ozone in the
Surrogate Mixture Runs for the RADM Mechanism.
RADM Mechanism
-------
RADM Mechanism
£S
0)
c
o
N
O
E
1
"x
O
2
C
~~
C/)
a
in
&JU
150
100
50
-50
inn
&
a
A ,
a
A
^° °
Q
' D*°.
*** * "*» £i^*^4if °
* ^. o "^ a?a * * ° °
4 o o o
. , . . ,
3 Comp.
A
7 Comp.
8 Comp.
a
UNC Mixes
4
Auto
*
Syn. Fuel
0.1 0.2 0.5 1
Initial NOx Concentration (ppm)
Figure 60. Normalized Bias in Maximum Ozone vs Initial NO in the
Surrogate Mixture Runs for the RADM Mechanism.
RADM Mechanism
^- 150
0)
0
N 100
E
P 50
X
O
C
S -50
CD
inn
a
* *
a
* a *
o
a
° *
A 4* A*«^° *
*A. * f * 4* "I*lM^Vt ;. v
O O Q
-
3 Corn p.
A
7 Comp.
8 Cornp.
o
UNC Mixes
Auto
*
Syn. Fuel
0.1 0.3 1 3 10 30 100
Initial ROG Concentration (pprnC)
Figure 61. Normalized Bias in Maximum Ozone vs Initial HC in the
Surrogate Mixture Runs for the RADM Mechanism.
200
-------
OZONE
0.4-1
0.3-
0.2-
0.1 -
NITRIC OXIDE
240 360 480 600 720 840 960
0.0-
240 360 480 600 720 840 960
Q.
CL
o
£
UJ
o
z
o
CJ
0.4-,
0.3-
0.2-
0.1 -
NITROGEN DIOXIDE
0.0
240 360 480 600 720 840 960
0.10-,
0.08-
0.06-
0.04-
0.02^
PAN
0.00
240 360 480 600 720 840 960
0.10-,
0.08-
0.06-J
0.04-J
0.02-
ETHENE
0.00
' I I ! II I1^
240 360 4«0 600 720 840 960
EU\PSED TIME (minutes)
Figure 62. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of the UNC Mixed Alkene Run
OC1278R.
201
-------
OZONE
E
CL
O_
O
UJ
O
-z.
O
O
60 120 180 240 300 360
NITRIC OXIDE
120 180 240 300 360
0.4-,
0.3-
NITROGEN DIOXIDE
0.0-
0.15-,
PAN
60
120 180 240 300 360 0
EUPSED TIME (minutes)
r~
60 120 180 240 300 360
Figure 63. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of the 7 Hydrocarbon
Surrogate Run EC-237.
20:
-------
E
Q.
Q.
o
LU
O
o
o
0.5 -i
OZONE
NITRIC OXIDE
/ 0.2
0.3-,
0.2-
720 1440 2160 2880 3600
NITROGEN DIOXIDE
720 1440 2160 2880 3600
PAN
0.0-
I I I I I | I IT IT | I I I I I ] I
720 1440 2160 2880 3600
i i i | i i i i i | i i i i i | i i i i i | i i i i i
720 1440 2160 2880 3600
ELAPSED TIME (minutes)
Figure 64. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of the 8 Hydrocarbon
Surrogate, Multi-Day Run ITC-630.
-------
Q.
CL
O
z
O
0.5 -i
OZONE
720 1440 2160 2880
0.0
NITROGEN DIOXIDE
I I I I I I I I IT i Ti i i i
720 1440 2160 2880
0.5
NITRIC OXIDE
0.0
0 720 1440 2160 2880
PAN
0.08-
0.04-
o.oo
i i i i i I i i i i i I i i i i i I i i i
720 1440 2160 2880
ELAPSED TIME (minutes)
Figure 65. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of the 8 Hydrocarbon
Surrogate, Multi-Day Run ITC-633.
-------
OZONE
E
a
a.
g
t
<
LU
a
z.
o
o
0.3 -,
0.2-
0.1 -
NITRIC OXIDE
0.4-,
0.3-
0.2- !
0 1440 2880
NITROGEN DIOXIDE
4320
0.0-
0.0-
0.15-,
1440 2880 4320
PAN
0 1440 2880
0 1440
ELAPSED TIME (minutes)
2880
4320
Figure 66. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of the 8 Hydrocarbon
Surrogate, Multi-Day Run ITC-637.
205
-------
OZONE
E
Q_
Q_
LJ
O
-z.
O
o
0.3-,
NITRIC OXIDE
0.0
480
960
1440 0 480
ELAPSED TIME (minutes)
Figure 67. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of the 8 Hydrocarbon
Surrogate, Multi-Day Run OTC-192B
206
-------
OZONE
Q.
Q.
o
<
Ld
CJ
-z.
o
0
0.25-,
0.20-
0.15-
0.10-
0.05-
NITRIC OXIDE
0.00-
0 480 960 1440 1920 2400 2880
NITROGEN DIOXIDE
0.05 -,
480 960 1440 1920 2400 2880
PAN
i i i j I I I | ! I I I1
0 480 950 1440 1920 2400 2880
i | i i i | i i i | i i i | i i i | i i
480 960 1440 1920 2400 2880
ELAPSED TIME (minutes)
Figure 68. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of the 8 Hydrocarbon
Surrogate, Multi-Day Run OTC-195A.
-------
OZONE
Q.
a
g
<
UJ
O
-2L
O
O
NITRIC OXIDE
480 960 1440 1920 2400 2880
NITROGEN DIOXIDE
: j-pr-i-i-y-r Ty-r T-p
480 960 1440 1920 2400 2880
0.1 -
0.0
0.02-
0.00
480 960 1440 1920 2400 2880
PAN
| i i i | i i i | i i i | i i i | i i i
0 480 960 1440 1920 2400 2880
ELAPSED TIME (minutes)
Figure 69. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of the 8 Hydrocarbon
Surrogate, Multi-Day Run OTC-202B.
208
-------
E
Q.
Q.
g
£
UJ
o
2:
o
o
1.0-1
OZONE
0.25-
NITRIC OXIDE
1080 1440 1800
T
360
720 1080 1440 1800
0 360 720
ELAPSED TIME (minutes)
1080 1440 1800
Figure 70. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of the 8 Hydrocarbon
Surrogate, Multi-Day Run OTC-224B.
209
-------
E
Q.
CL
Z
o
o:
z
LU
o
o
o
0.7 -,
0.6-
0.5-
0.4-
0.3-
0.2-
0.1 -
0.0
OZONE
NITRIC OXIDE
300 420 540 660 780 900 1020
NITROGEN DIOXIDE
0.08^
0.06-
0.04-
0.02-
300 420 540 660 780 900 1020
PAN
I I .
300 420 540 660 780 900 1020
0.00»
. 1 1 r .
300 420 540 660 780 900 1020
ELAPSED TIME (minutes)
Figure 71. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of the UNC 3 Hydrocarbon
Surrogate Run JN1483R.
210
-------
OZONE
NITRIC OXIDE
E
CL
Q.
g
<
UJ
o
z
o
o
300 420 540 660 780 900 1020
NITROGEN DIOXIDE
300 420 540 660 780 900 1020
PAN
I I I T
300 420 540 660 780 900 1020
i [^ i^ i r
300 420 540 660 780 900 1020
ELAPSED TIME (minutes)
Figure 72. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of the UNC Miscellaneous
Surrogate Run AU2681B.
211
-------
Q.
a
O
LU
O
z
O
O
O.B-i
0.6-
0.4-
0.2-
OZONE
NITRIC OXIDE
0.0
300 420 540 660 780 900 1020
NITROGEN DIOXIDE
I ....
300 420 540 660 780 900 1020
PAN
I 1 ;
300 420 540 660 780 900 1020
I I I
300 420 540 660 780 900 1020
ELAPSED TIME (minutes)
Figure 73. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of the UNC Synurban
Surrogate Run JN2685R.
112
-------
OZONE
NITRIC OXIDE
E
Q.
Q.
g
LJ
O
z.
O
O
300 420 540 660 780 900 1020
NITROGEN DIOXIDE
300 420 540 660 780 900 1020
PAN
300 420 540 660 780 900 1020 300 420 540 660 7SO 900 1020
ELAPSED TIME (minutes)
Figure 74. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of the UNC Synauto
Surrogate Run AU0684R.
213
-------
E
Q_
Q.
g
S
UJ
o
z
o
o
1.0-1
OZONE
0.20-1
NITRIC OXIDE
300 420 540 660 780 900 1020
NITROGEN DIOXIDE
300 420 540 660 780 900 1020
PAN
T |^
300 420 540 660 780 900 1020
300 420 540 660 780 900 1020
EUPSED TIME (minutes)
Figure 75. Experimental and Calculated Concentration-Time Plots for
Selected Species in Simulations of the UNC Auto Exhaust
Run AU1183R.
214
-------
component would have a relatively smaller effect than would be the case in
a run if that component were the only reactive organic present.) There
does not appear to be any systematic bias in the results with respect to
the initial HC to NOV ratios, initial NOV concentration, or initial HC
A X
concentrations. This is reassuring, but is not unexpected, considering
that such a systematic bias was not observed for any of the groups of
individual organics where one might expect such biases to be more evident.
The mechanism's performance on multi-day runs is comparable to that
for single day runs, especially when good performance is achieved on the
first day of the simulation. (See, for example, Figures 64 through 70.)
If poor performance is achieved on the first day, then obviously the
performance on the subsequent days would be even worse, since the starting
conditions would be different.
Figures 76 and 77 compare the calculated and experimental formalde-
hyde maxima for the simple and complex mixture runs. The performance
statistics are shown in Table 21. These results show that the RADM
mechanism underpredicts the maximum formaldehyde on the average, though
for most types of runs the mean biases are smaller in magnitude than the
mean errors, which are typically on the order of 40-50/t. The performance
on formaldehyde in mixture runs is poorer than that found for single
carbonyl and most types of alkene runs, but is better than the results for
alkane and aromatic runs, on the average. The figures clearly show that
for the runs with mixtures as well as those for the single compounds there
is a large amount of scatter in the results and the mechanism is not able
to reproduce the formaldehyde maxima very accurately. As already noted,
however, the accuracy of the formaldehyde measurements is questionable.
Figures 78 and 79 compare the calculated and experimental PAN maxima
for the simple and complex mixture runs. The performance statistics are
shown in Table 22. Except for the UNC auto exhaust runs (whose chemical
compositions are the most uncertain of all the runs), the RADM mechanism
consistently overpredicts PAN in the mixture runs, with average biases
ranging from 12 to over 66^. The mean error of these predictions are
typically on the order of HO-605& or greater. The positive bias and large
error is consistent with the results for single organic compound runs. A
positive bias is expected since the PAN species in the mechanism is used
215
-------
Mechanism
c
,^
X
i
0.5
0.2
0.1
A * /
4 £/
A y/
xi *
\ s*>
- 0.05 (
u
0.01
0.01 0.02 0.05 0.1 0.2 0.5
Experimental HCHO Maximum (ppm)
Figure 76. Calculated vs Experimental Maximum Formaldehyde in the
Simple Mixture Runs for the RADM Mechanism.
RADM Mechanism
' f _ *,"' : ~" '"^
C » , ! A
5- f X i , ,
i <"- i -;'-
§ 0.3
w
X.
O
5 0.1
o
5
| 0.03
_o
3
Q 0.01
. :^-* .
. «J*j4?*t* °£
9 . 7V» * "°
x°»° 8*°8 *
« / 0*0 »
/. * * « *
./
X i . i . i
8 Comp.
UNC Mixes
a
Auto
o
Syn. Fuel
*
Figure 77.
0.01 0.03 0.1 0.3 1
Experimental HCHO Maximum (ppm)
Calculated vs Experimental Maximum Formaldehyde in the
Surrogate Mixture Runs for the RADM Mechanism.
216
-------
RADM Mechanism
,-, 0.3
E
Q_
Q.
^ 01
E
D
E
2 0.03
CO1
' 0 003
O
o
0.001
Figure 78.
0.001 0.003 0.01 0.03 01 0.3
Experimental PAN Maximum (ppm)
Calculated vs Experimental Maximum PAN in the Simple
Mixture Runs for the RADM Mechanism.
0.3
RADM Mechanism
C.i
' °-°3
D
0,01
CL
0)
0.003
D
±j
o.ooi -
A
8 Comp.
«
UNC Mixes
I
Syn. Fue!
0.001 0.003 0.01 003 0.1 0.3
Experimental PAN Maximum (ppm)
Figure 79. Calculated vs Experimental Maximum PAN in the Surrogate
Mixture Runs for the RADM Mechanism.
217
-------
to represent PAN plus numerous analogous organic nitrates. However, the
magnitude of the bias, especially in the surrogate mixture runs, suggests
the mechanism overpredicts the PAN and analogous species concentrations.
This is of concern for regional acid deposition modeling because the model
may underpredict nitric acid in the first few 100 km downwind of sources
(i.e., if NOX is oxidized disproportionately to PAN rather than HNO^) and
overestimate the distances for which nitrogen species are transported
(since PAN is the long-range transport product of NOX emissions).
5.6 Tests of the RADM VOC Aggregation Approach
As discussed in Section U.3, the evaluation of the RADM mechanism
involved processing organic input data in the simulations of the chamber
experiments in a manner analogous to the expected procedure for processing
of emissions input data for the RADM model. This involves, for example,
representing some compounds by model species on a mole-per-mole basis, and
others using "reactivity weighing," and not adjusting any parameters in
the mechanism on a run-to-run basis. This is the appropriate procedure
for evaluating the mechanism to be used in RADM, but in general it is not
the one which would be expected to give the optimum fits of the model
simulations to the experimental results. An alternative approach, which
is appropriate for evaluations of detailed mechanisms or mechanisms to be
used for assessment of relative reactivities, is to adjust parameters in
the mechanism based on the mixtures being simulated. This latter proce-
dure should in general result in the optimum performance of the model in
simulating the experimental data; provided, of course, that errors in
other aspects of the mechanism do not compensate for errors introduced
using a more approximate representation.
Because of this, a comparison of the performance of the RADM mechan-
ism using run-by-run adjustment of parameters with its performance using
the standard evaluation procedure can give an indication of the level of
inaccuracy introduced by using the RADM VOC aggregation procedure. To
evaluate this, a separate set of simulations of the chamber experiments
employing organics not explicitly represented in the RADM mechanism was
carried out where the rate constants of the lumped model species were
adjusted on a run-to-run basis to correspond to those of the compounds
present in each experiment. The averages of the percent errors and biases
218
-------
in predictions of ozone yields and NO oxidation rates for groups of runs
calculated for the RADM mechanism using this procedure are compared with
the corresponding results for the standard evaluation in Table 23.
Differences in ozone predictions for individual organic mixture runs are
shown in Figure 80, where the maximum ozone yields calculated using this
modified procedure are plotted against the yields calculated using the
standard evaluation procedure.
The results shown in Table 23 and Figure 80 indicate that the two
methods give very similar results for all types of mixture runs and for
most single component runs. The differences are obviously greater for the
single component runs; but, except for the one isobutene run and (to a
lesser extent) the trimethylbenzene runs, the statistics for the
simulations using the RADM VOC aggregation method are either about the
same or somewhat better than the simulations where the rate constants are
adjusted. Improvement of fits using the more approximate RADM VOC aggre-
gation method relative to those obtained with adjusted rate constants is
clearly a case of partial cancellation of errors. These results indicate
that for realistic mixtures of organics likely to be encountered in
regional modeling applications, any inaccuracies introduced by using the
RADM VOC aggregation procedure discussed in Section 3 are likely to be
small compared to inaccuracies introduced by other uncertainties in the
mechanism.
5.7 Performance of the Recommended Modified Mechanisms
The recommended modified versions of the RADM mechanism, discussed in
Section 2 of this report, were also evaluated against the chamber data.
These mechanisms are the RADM-M, which incorporates all of the modifica-
tions we recommend as a result of this evaluation study, and the RADM-P,
which primarily incorporates those recommendations involving only changes
in parameter values. These mechanisms are the same as the March 1988 RADM
mechanism in their inorganic and simple carbonyl reaction set, and thus
the evaluation of the RADM mechanism for the characterization and carbonyl
runs is also applicable to these mechanisms. However, both RADM-M and
RADM-P have significant differences in their representation of aromatics,
and in addition the RADM-M has a more condensed representation of the more
rapidly reacting (generally Ce+) alkanes. The two mechanisms also have
219
-------
Table 23. Comparison of Average Normalized Biases and Errors in
Simulations of Chamber Experiments using the RADM Mechanism
with Run-to-Run Adjustment of Rate Constants with Simulations
using the Standard RADM VOC Aggregation Procedure
Compound
or
Mixture
Propene
1-Butene
trans-2-Butene
Isobutene
1-Hexene
a-Pinene
Ethane
n-Butane
C4+ Br. Alkane
C5+ n-Alkane
Cycloalkane
Benzene
Toluene
1) Adjusted Rate Constants
2) Standard VOC Aggregation
Maximum Ozone (?)
Bias (sdev) Err.
4
5
5
-2
-52
-52
-12
-23
64
57
-12
-8
-17
-19
-1
0
-12
-17
49
42
34
33
22
22
24
24
( 19)
( 20)
( 54)
( 56)
( 38)
( 38)
( 0)
( 0)
( 26)
( 25)
( 36)
( 35)
( 0)
( 0)
( 59)
( 59)
( 58)
( 67)
( 34)
( 28)
( 0)
( 0)
( 16)
( 15)
( 43)
( 43)
15
15
45
45
54
54
12
23
64
57
29
28
17
19
46
46
44
53
49
42
34
33
22
22
33
33
d([03]-[NO])/dt (%)
Bias (sdev) Err.
6
6
9
0
-26
-25
10
-45
101
82
-10
4
7
6
18
19
-16
-22
31
24
5
3
28
36
6
7
( 18)
( 19)
( 32)
( 32)
( 26)
( 26)
( 0)
( 0)
{ 30)
( 25)
( 10)
( 19)
( 0)
( 0)
( 50)
( 50)
( 20)
( 24)
( 36)
( 37)
( 28)
( 29)
( 21)
( 19)
( 34)
( 34)
14
14
29
25
31
31
10
45
101
82
10
14
7
6
43
43
19
26
35
34
17
19
31
36
26
27
(continued)
220
-------
Table 23 (continued) - 2
Compound
or
Mixture
Xylene
Mesitylene
Tetralin
Naphthalene
2,3-Dimethylnaph.
Simple Mixtures
Simple Surrogates
SAPRC 7-HC Surgs.
SAPRC 8-HC Surgs.
UNC Misc. Surgs.
UNC "Synurban"
UNC "Synauto"
UNC Auto Exhaust
Syn. Jet Fuel
Syn. Jet Exhaust
1 ) Adjusted Rate Constants
2) Standard VOC Aggregation
Maximum Ozone (%)
Bias (sdev) Err.
-9 i
-4 i
-21 i
-39 '
-2
-6
82
84
18
-6
-12
-12
10
7
1
0
12
9
14
17
36
39
22
23
10
8
10
14
-10
-11
( 30)
( 25)
( 30)
( 50)
( 56)
( 62)
( 44)
( 40)
( 15)
( 24)
( 45)
( 45)
( 28)
( 24)
( 11)
( 11)
( 25)
( 26)
( 39)
( 40)
( 30)
( 3D
( 17)
( 17)
( 29)
( 29)
( 11)
( 12)
( 8)
( 8)
19
18
30
48
41
43
82
84
18
17
34
34
17
15
8
8
21
20
27
27
36
39
23
23
19
19
11
14
10
11
d([03]-[MO])/dt (%)
Bias (sdev) Err.
-19 <
0 i
-60
-93
108
106
118
134
58
1
-13
-13
7
8
-9
-5
-1
-10
1
2
4
7
0
2
-3
-6
35
11
-15
-8
( 3D
( 35)
( 39)
( 32)
( 85)
( 85)
( 16)
( 13)
( 28)
( 24)
( 25)
( 26)
( 43)
( 38)
( 12)
( 12)
( 24)
( 23)
( 16)
( 15)
( 9)
( 8)
( 18)
( 18)
( 15)
( 15)
( 16)
( 16)
( 8)
( 4)
20
25
60
93
121
119
118
134
58
19
22
22
30
27
12
10
17
19
12
11
8
8
14
14
12
13
35
16
15
8
221
-------
C
o
u
cr
TJ
O>
C
ccrd Methoc
Figure 80.
Plots of Maximum Ozone Yields in Simulations of Organic
Mixture Experiments Using the RADM Mechanism with Run-to-Run
Adjustment of Rate Constants Against Yields Calculated Using
the Standard RADM VOC Aggregation Procedure
222
-------
small differences in their reactions for the akenes and the less reactive
Co+ alkanes. The evaluation calculations for these mechanisms were thus
carried out for all the single and mixed organic-NOx-air runs except those
involving only simple carbonyls.
The summary statistics for the performance of the RADM-M and the
RADM-P mechanisms in simulating maximum ozone yields and rates of NO
oxidation and ozone formation are given in Tables 24 and 25; statistics
for predictions of maximum formaldehyde are given in Tables 26 and 27; and
statistics for predictions of maximum PAN are given in Tables 28 and 29.
Table 30 shows comparisons of selected summary statistics for all three
versions of the mechanism which were evaluated in this study. Figures 81
and 82 show plots of calculated vs experimental maximum ozone for the
RADM-M and RADM-P mechanisms for the surrogate mixture runs; Figures 83
through 86 show similar plots for the alkane-NOx and the aromatic-NOx
runs, respectively. Finally, Figures 23-25, 30-32, 37-42, 49 and 62-75
(presented earlier) give concentration-time profiles illustrating the
performance of these mechanisms in simulating results of selected experi-
ments where their predictions differed from the evaluated RADM mechan-
ism. (Solid lines are for the RADM mechanism, longer dashed lines are for
RADM-P, and shorter dashed lines are for RADM-M.) Run-by-run summaries of
the performances of these mechanisms in simulating maximum ozone and NO
oxidation rates are found in Appendix B for RADM-M and in Appendix C for
RADM-P.
The differences between the predictions of the mechanisms for the
alkene-NOv and alkane-NOv runs are relatively minor, as expected due to
X A
the relatively small changes in the mechanisms for these compounds. The
RADM-M mechanism performs slightly worse than the other two for the higher
alkane-NO runs because of its more condensed representation of these
compounds. However, the test calculations discussed in Section 6.3
indicate that this should have relatively little effect in simulations of
more realistic mixtures of organics.
223
-------
Table 24. Performance Statistics on Maximum Ozone and NO Oxidation and
Ozone Formation Rates in the Simulations of Organic-N0x-Air
Runs Using the RADM-M Mechanism
Organic
or
Mixture
Ethene
Propene
1-Butene
trans-2-Butene
Isobutene
1-Hexene
Isoprene
a-Pinene
Ethane
n-Butane
C4+ Br. Alkanes
C5+ n- Alkanes
Cycloalkanes
Benzene
1)
2)
No.
Runs
14
14
49
49
11
11
4
4
1
1
4
4
13
13
4
4
1
1
27
27
7
7
14
14
4
4
6
6
Statistics for Maximum Ozone (ppb)
Statistics for d( [03l-[NO])/dt (ppb/min)
Experimental
Avg. (sdev)
0.91 (
5.23 (
0.61 (
3.62 {
0.42 (
2.86 (
0.28 {
6.17 (
0.90 (
8.84 (
0.34 (
1.57 (
0.72 (
5.66 (
0.35 (
0.97 <
0.24 1
1.32 !
0.31 <
1.53 I
0.28 I
0.93 <
0.20 i
1.15 i
0.05
0.92 *
0.29
3-39
0.20)
4.10)
0.24)
2.60)
0.26)
2.18)
0.17)
2.80)
0.00)
0.00)
: 0.25)
: 1.04)
; o
; 4
; o
; o
; o
; o
[ o
I o
[ o
[ o
( o
( o
( o
( o
( o
( 2
.25)
.06)
.11)
.43)
.00)
.00)
.27)
.98)
.18)
.46)
.15)
.75)
.00)
.12)
.14)
.30)
Absolute
Bias Err.
-0.09
-1.21
0.04
0.53
0.05
0.41
-0
-1
-0
-2
0
3
-0
0
-0
0
-0
0
-0
0
-0
-0
0
0
.08
.68
.17
.81
.28
.32
.23
.86
.03
.06
.04
.09
.01
.18
.05
.21
.16
.42
0.05
0.36
0.07
6.11
0.16
1.21
0.
0.
0.
0.
0.
1.
0.
2.
0.
3.
0.
1.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
09
76
14
96
09
76
17
81
28
32
24
81
08
15
04
09
12
72
10
23
16
56
0.05
0.41
0.07
6.11
Normalized
Bias (») Err. (%)
-16 (
-27 (
8 (
11 (
2 (
5 (
-53 (
-29 (
-21 <
-38 (
59 (
88 (
-42 <
-2 (
-8 {
2 (
-18 <
7 I
-10 <
12 I
-17 I
-25 i
58 i
38 i
85
25
20
92
30)
20)
20)
19)
55)
32)
38)
24)
: o)
0)
: 26)
: 26)
: 3D
; 36)
: 3*0
: 18)
: o)
; o)
[ 59)
! 49)
( 73)
( 25)
( 39)
( 43)
( 0)
( 41)
( 13)
( 23)
23
27
16
17
45
27
54
32
21
38
59
88
43
26
27
1.3
18
7
47
41
58
29
58
46
85
31
20
92
(continued)
224
-------
Table 24 (continued) - 2
Organic
or
Mixture
Toluene
Xylene
Mesitylene
Tetralin
Naphthalene
2 , 3-Dimethylnaph .
Simple Mixtures
Simple Surrogates
SAPRC 7-HC Surgs.
SAPRC 8-HC Surgs.
UNC Misc. Surgs.
UNC "Synurban"
UNC "Synauto"
UNC Auto Exhaust
Syn. Jet Fuel
Syn. Jet Exhaust
1)
2)
No.
Runs
20
20
10
10
8
6
5
5
5
5
4
4
40
39
36
36
11
11
82
82
21
21
11
11
15
15
28
28
8
8
4
Statistics for Maximum Ozone (ppb)
Statistics for d( [O^l-fNO] )/dt (ppb/min)
Experimental
Avg. (sdev)
0.34 (
3.07 (
0.44 (
5.39 (
0.52 (
9.50 (
0.29 (
1.89 (
0.20 (
1.59 (
0.32 (
2.48 (
0.49 (
4.10 (
0.37 (
2.81 (
0.60 (
7.74 <
0.45 I
2.81 I
0.50 I
1.32 I
0.37 i
1.09 i
0.62 i
1.81 <
0.57
1.68
0.75
3.47
0.78
6.66
0.15)
1.87)
0.19)
4.14)
0.26)
4.97)
0.23)
0.88)
0.08)
0.41)
0.04)
0.5D
0.27)
: 4.06)
: 0.15)
: 1.61)
; 0.17)
: 5.16)
; 0.27)
: LSD
[ 0.17)
( 0.40)
( 0.28)
( 0.47)
( 0.18)
( 0.63)
( 0.27)
( 0.61)
( 0.10)
( 0.74)
( 0.13)
( 3.48)
Absolute
Bias Err.
0.03
0.12
-0.01
0.67
-0.13
-4.66
0.23
25.41
0.33
19.56
0.08
3.90
-0.06
-0.89
0.01
1.08
0.00
-0.55
0.02
-0.16
0.02
-0.06
0.06
0.01
0.09
-0.05
0.04
-0.14
0.13
8.22
-0.07
0.06
0.06
0.62
0.05
0.96
0.14
4.66
0.23
25.56
0.33
19.56
0.08
3.90
0.12
1.11
0.05
1.32
0.04
0.72
0.08
0.54
0.10
0.15
0.07
0.06
0.11
0.30
0.10
0.21
0.13
8.22
0.08
0.33
Normalized
Bias (?) Err. (?)
13 (
7 (
0 (
17 (
-26 (
-57 (
57 (
125 (
93 (
170 (
21 (
79 (
-15 (
-15 <
8 (
28 (
0 <
-8 I
3 I
-5 i
11 \
-3 <
22 i
1 i
16
-2
1
-10
16
107
-9
3
3D
3D
16)
20)
19)
22)
50)
9D
37)
8)
17)
; 34)
: 43)
: 25)
: 24)
: 36)
: 12)
; 12)
; 26)
[ 24)
( 40)
( 15)
( 25)
( 9)
( 17)
( 18)
( 34)
( 15)
( 1D
( 14)
( 8)
( 5)
21
21
12
20
29
57
57
137
93
170
21
79
32
23
16
36
8
11
19
19
25
12
24
6
18
16
23
14
16
107
10
5
225
-------
Table 25. Performance Statistics on Maximum Ozone and NO Oxidation and
Ozone Formation Rates in the Simulations of Organic-N0x-Air
Runs Using the RADM-P Mechanism
Organic
or
Mixture
Ethene
Propene
1-Butene
trans-2-Butene
Isobutene
1-Hexene
Isoprene
a-Pinene
Ethane
n-Butane
C4+ Br. Alkane
C5+ n-Alkane
Cycloalkanes
Benzene
1)
2)
No.
Runs
14
14
49
49
11
11
4
4
V
1
4
4
13
13
4
4
1
1
27
27
7
7
14
14
4
4
6
6
Statistics for Maximum Ozone (ppb)
Statistics for d( [03]-[NO])/dt (ppb/min)
Experimental
Avg. (sdev)
0.91 (
5.23 (
0.61 (
3.62 (
0.42 (
2.86 (
0.28 (
6.17 (
0.90 (
8.84 (
0.34 (
1.57 (
0.72 (
5.66 <
0.35 I
0,97 <
0.24 I
1.32 I
0.31 i
1.53 i
0.28
0.93
0.20
1.15
0.05
0.92
0.29
3.39
0.
4.
0.
2.
0.
2.
20)
10)
24)
60)
26)
18)
0.17)
2.80)
0.00)
0.00)
: 0.25)
: i.
.04)
: 0.25)
[ 4.06)
: 0.11)
; 0.43)
[ 0
[ o
( 0
( 0
( 0
( 0
( 0
( o
( o
( o
( o
( 2
.00)
.00)
.27)
.98)
.18)
.46)
.15)
.75)
.00)
.12)
.14)
.30)
Absolute
Bias Err.
-0.
-1.
0.
09
21
04
0.53
0.05
0.41
-0.08
-1,
.68
-0.17
-2.80
0.28
3.33
-0
0
-0
0
-0
0
-0
0
-0
-0
0
0
0
0
0
6
.23
.86
.03
.06
.04
.10
.01
.21
.06
.18
.10
.27
.00
.03
.07
.20
0.16
1.21
0.09
0.76
0.14
0.96
0.09
1.76
0.17
2.80
0.28
3.33
0.24
1.81
0.08
0.15
0.04
0.10
0.12
0.74
0.09
0.20
0.10
0.40
0.02
0.21
0.07
6.20
Bias
-16
-27
8
11
2
5
-53
-29
-21
-38
59
88
-42
-2
-8
2
-18
7
-8
13
-19
-22
41
25
27
-1
19
93
Normalized
(») Err. (J)
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
30)
20)
20)
19)
55)
32)
38)
24)
0)
0)
26)
26)
3D
36)
34)
18)
0)
0)
59)
49)
62)
24)
27)
36)
0)
3D
14)
24)
23
27
16
17
45
27
54
31
21
38
59
88
43
26
27
13
18
7
47
41
50
25
41
35
27
22
19
93
(continued)
226
-------
Table 25 (continued) - 2
Organic
or
Mixture
Toluene
Xylene
Mesitylene
Tetralin
Naphthalene
2,3-Diraethylnaph.
Simple Mixtures
Simple Surrogates
SAPRC 7-HC Surgs.
SAPRC 8-HC Surgs.
UNC Misc. Surgs.
UNC "Synurban"
UNC "Synauto"
UNC Auto Exhaust
Syn. Jet Fuel
Syn. Jet Exhaust
D
2)
No.
Runs
20
20
10
10
8
6
5
5
5
5
4
4
40
39
36
36
11
11
82
82
21
21
11
11
15
15
28
28
8
8
4
4
Statistics for Maximum Ozone (ppb)
Statistics for d( [03]-[NO])/dt (ppb/min)
Experimental
Avg. (sdev)
0.34 (
3.07 (
0.44 (
5.39 (
0.52 (
9.50 (
0.29 (
1.89 (
0.20 (
1.59 (
0.32 (
2.48 (
0.49 <
4.10 (
0.37 I
2.81 <
0.60 I
7.74 I
0.45 I
2.81 I
0.50 l
1.32 I
0.37 i
1.09 i
0.62 i
1.81 i
0.57
1.68
0.75
3.47
0.78
6.66
0.15)
1.87)
0.19)
4.14)
0.26)
4.97)
0.23)
; 0.88)
; 0.08)
: o.4D
: 0.04)
: o.5D
; 0.27)
: 4.06)
; 0.15)
: 1.6D
; 0.17)
: 5.16)
[ 0.27)
( 1.51)
( 0.17)
( 0.40)
( 0.28)
( 0.47)
( 0.18)
( 0.63)
( 0.27)
( 0.61)
( 0.10)
( 0.74)
( 0.13)
( 3.48)
Absolute
Bias Err.
0.05
0.19
0.02
0.61
-0.11
-4.63
0.23
25.16
0.34
19.27
0.10
3.90
-0.06
-0.87
0.02
1.05
-0.01
-0.46
0.03
-0.10
0.02
-0.06
0.07
0.02
0.10
-0.04
0.05
-0.13
0.12
5.54
-0.07
0.00
0.07
0.68
0.06
0.93
0.13
4.63
0.23
25.30
0.34
19.27
0.10
3.90
0.12
1.10
0.05
1.28
0.04
0.77
0.08
0.54
0.10
0.14
0.08'
0.06
0.12
0.29
0.10
0.21
0.12
5.54
0.08
0.34
Normalized --
Bias (?) Err. (%)
17 (
10 (
5 (
17 (
-21 (
-55 (
56 (
125 (
93 (
170 (
26 (
79 (
-14 (
-14 (
9 (
28 (
-1 (
-6 (
5 (
-3 (
10 (
-3 (
24 (
2 (
18 (
-1 (
3 (
-9 (
15 (
87 (
-9 (
2 (
3D
3D
16)
21)
19)
22)
49)
9D
38)
8)
17)
32)
44)
25)
23)
36)
12)
13)
25)
25)
37)
15)
25)
9)
16)
18)
34)
15)
10)
17)
8)
5)
22
23
12
21
25
55
56
137
93
170
26
79
32
22
16
35
8
12
19
19
24
12
25
6
19
15
23
14
15
87
10
5
227
-------
Table 26. Performance Statistics on Maximum Formaldehyde in the
Simulations of Organic-N0x-Air Runs Using the RADM-M
Mechanism
Compound or
Mixture
Ethene
Propene
1-Butene
trans-2-Butene
Isobutene
1 -Hexene
Isoprene
a-Pinene
n-Butane
C4+ Br. Alkane
C5+ n-Alkane
Cycloalkane
Toluene
Xylene
Mesitylene
Naphthalene
Simple Mixtures
Simple Surrogates
SAPRC 7-HC Surgs.
SAPRC 8-HC Surgs.
UNC "Synurban"
UNC "Synauto"
UNC Auto Exhaust
Syn. Jet Fuel
No.
Runs
14
39
8
4
1
4
13
2
21
5
10
4
20
10
8
5
34
7
11
82
11
13
28
8
Experimental(a) Absolute(a)
Avg. (sdev) Bias Err.
0.86 (
0.36 (
0.32 (
0.08 (
0.64
0.33 (
0.30 (
0.08 (
0.04 <
0.03 (
0.01 (
0.07 (
0.09 <
0.09 I
0.09 <
0.01 I
0.25 I
0.13 I
0.38 I
0.18 I
0.12 I
0.16 i
0.18 i
0.05 i
0.41)
: 0.20)
0.27)
; 0.06)
: 0.27)
: 0.16)
: o.o4)
: 0.03)
: 0.02)
: O.OD
: 0.12)
; 0.09)
; 0.05)
; 0.04)
! O.OD
[ 0.20)
[ 0.05)
[ 0.23)
( 0.12)
( 0.04)
( 0.08)
( 0.10)
( 0.03)
-0.02
0.01
0.12
0.04
0.00
0.32
0.01
-0.02
0.01
-0.02
-0.01
-0.06
-0.03
-0.04
-0.03
0.05
0.02
-0.05
0.10
-0.02
-0.05
0.02
-0.10
-0.01
0.23
0.11
0.12
0.06
0.00
0.32
0.12
0.02
0.03
0.02
0.01
0.07
0.04
0.05
0.04
0.05
0.09
0.06
0.10
0.05
0.05
0.04
0.10
0.03
Normalized
Bias (%) Err. (%)
-9
7
24
59
-1
79
-6
-25
14
-57
-19
-31
-26
-45
-30
134
-14
-45
18
1
-54
14
-68
-13
( 41)
( 45)
( 19)
( 58)
( 34)
( 60)
( 12)
( 76)
( 78)
(101)
(135)
( 62)
( 68)
( 49)
( 43)
( 53)
( 45)
( 19)
( 48)
{ 18)
( 3D
( 3D
( 67)
34
34
25
70
1
79
46
25
65
87
89
105
48
67
44
134
45
55
22
35
54
25
70
53
(a) Concentrations in ppm.
228
-------
Table 27. Performance Statistics on Maximum Formaldehyde in the Simulations
of Organic-NOx-Air Runs Using the RADM-P Mechanism
Compound or
Mixture
Ethene
Propene
1 -Butene
trans-2-Butene
Isobutene
1-Hexene
Isoprene
a-Pinene
n-Butane
C4+ Br. Alkane
C5+ n-Alkane
Cycloalkane
Toluene
Xylene
Mesitylene
Naphthalene
Simple Mixtures
Simple Surrogates
SAPRC 7-HC Surgs.
SAPRC 8-HC Surgs.
UNC "Synurban"
UNC "Synauto"
UNC Auto Exhaust
Syn. Jet Fuel
No.
Runs
14
39
8
4
1
4
13
2
21
5
10
4
20
10
8
5
7
11
82
11
13
28
8
Experimental(a)
Avg. (sdev)
0.86 (
0.36 (
0.32 (
0.08 (
0.64
0.33 (
0.30 (
0.08 (
0.04 (
0.03 (
0.01 (
0.07 <
0.09 I
0.09 I
0.09 <
0.01 '
0.25
0.13
0.38
0.18
0.12
0.16
0.18
0.05
0.41)
0.20)
0.27)
0.06)
0.27)
0.16)
0.04)
0.03)
; 0.02)
; O.OD
; 0.12)
( 0.09)
( 0.05)
( 0.04)
( 0.01)
( 0.20)
( 0.05)
( 0.23)
( 0.12)
( 0.04)
( 0.08)
( 0.10)
( 0.03)
Absolute(a)
Bias Err.
-0.02
0.01
0.12
0.04
0.00
0.32
0.01
-0.02
0.01
-0.02
-0.01
-0.06
-0.05
-0.07
-0.06
0.01
0.02
-0.06
0.09
-0.03
-0.05
0.01
-0.10
-0.03
0.23
0.11
0.12
0.06
0.00
0.32
0.12
0.02
0.03
0.02
0.01
0.06
0.05
0.07
0.06
0.01
0.09
0.06
0.10
0.06
.0.05
0.03
0.10
0.03
Normalized
Bias (*) Err. (%)
-9 i
7
24
59
-1
79
-6
-25
14
-74
-35
-41
-90
-109
-101
64
-21
-50
17
-4
-61
9
-74
-75
( 41)
( 45)
( 19)
( 58)
( 34)
( 60)
( 12)
( 76)
( 70)
(106)
(128)
( 58)
( 56)
( 44)
( 75)
( 62)
( 44)
( 19)
( 48)
( 19)
( 3D
( 3D
( 52)
34
34
25
70
1
79
46
25
65
92
95
98
91
114
101
71
52
59
21
36
/ 4
61
24
75
75
(a) Concentrations in ppm.
229
-------
Table 28. Performance Statistics on Maximum PAN in the Simulations of
Organic-N0x-Air Runs Using the RADM-M Mechanism
Compound or
Mixture
Propene
1-Butene
trans-2-Butene
Isobutene
1-Hexene
Isoprene
a-Pinene
n-Butane
C4+ Br. Alkane
C5+ n-Alkane
Cycloalkanes
Toluene
Xylene
Mesitylene
2 , 3-Dimethylnaph .
Simple Mixtures
Simple Surrogates
SAPRC 7-HC Surgs.
SAPRC 8-HC Surgs.
UNC Misc. Surgs.
UNC "Synurban"
UNC "Synauto"
UNC Auto Exhaust
Syn. Jet Exhaust
No.
Runs
46
7
14
1
3
12
2
21
6
11
4
20
10
8
4
38
32
11
68
12
10
15
28
4
Experimental(a)
Avg. (sdev)
0.112 (0.084)
0.039 (0.016)
0.080 (0.044)
0.137
0.003 (0.003)
0.133 (0.082)
0.022 (0.016)
0.296 (1.012)
0.226 (0.506)
0.247 (0.814)
0.002 (0.003)
0.051 (0.036)
0.125 (0.102)
0.407 (0.189)
0.046 (0.019)
0.081 (0.059)
0.048 (0.020)
0.095 (0.047)
0.044 (0.030)
0.055 (0.046)
0.024 (0.024)
0.054 (0.039)
0.047 (0.029)
0.080 (0.03D
Absolute(a)
Bias Err.
0.049 0.055
0.059 0.060
-0.004 0.014
0.140 0.140
0.319 0.319
-0.029 0.080
0.048 0.048
-0.259 0.293
-0.208 0.223
-0.243 0.247
-0.001 0.002
0.045 0.051
0 025 0.059
-0.230 0.230
0.095 0.095
0.023 0.037
0.017 0.020
0.054 0.054
0.042 0.042
0.011 0.019
0.010 0.010
0.018 0.030
0.000 0.011
0.117 0.117
Normalized
Bias (%) Err. (J)
31 ( 39)
70 ( 50)
-14 ( 24)
68
195 ( 4)
-19 ( 85)
100 ( 12)
10 (105)
-46 (119)
21 (129)
-112 ( 14)
66 ( 43)
35 ( 41)
-69 ( 27)
98 ( 28)
18 ( 58)
36 ( 51)
43 ( 12)
66 ( 35)
35 ( 64)
31 ( 28)
29 ( 45)
-13 ( 60)
87 ( 13)
36
72
21
68
195
61
100
90
99
101
112
71
46
69
98
51
41
43
66
48
37
41
41
87
(a) Concentrations in ppm.
230
-------
Table 29. Performance Statistics on Maximum PAN in the Simulations of
Organic-NOx-Air Runs Using the RADM-P Mechanism
Compound or
Mixture
Propene
1-Butene
trans-2-Butene
Isobutene
1-Hexene
Isoprene
a-Pinene
n-Butane
C4+ Br. Alkane
C5+ n-Alkane
Cycloalkane
Toluene
Xylene
Mesitylene
2,3-Dimethylnaph.
Simple Mixtures
Simple Surrogates
SAPRC 7-HC Surgs.
SAPRC 8-HC Surgs.
UNC Misc. Surgs.
UNC "Synurban"
UNC "Synauto"
UNC Auto Exhaust
Syn. Jet Exhaust
No.
Runs
46
7
4
1
3
12
2
21
6
11
4
20
10
8
4
38
32
11
68
12
10
15
28
4
Experimental (a
Avg. (sdev)
0.142
0.039
0.080
0.137
0.003
0.133
0.022
0.296
0.226
0.247
0.002
0.051
0.125
0.407
0.046
0.081
0.048
0.095
0.044
0.055
0.024
0.054
0.047
0.080
(0.084)
(0.016)
(0.044)
(0.003)
(0,
(0,
(1
(0
(0
(0
(0
(0
(0
(0
(0
(0
(0
(0
(0
(0
(0
(0
(0
.082)
.016)
.012)
.506)
.814)
.003)
.036)
.102)
.189)
.019)
.059)
.020)
.047)
.030)
.046)
.024)
.039)
.029)
.031)
) Absolute(a)
Bias Err.
0.
0.
-0.
0.
0.
-0.
0.
-0.
-0.
-0.
-0.
-0.
-0.
049
059
004
140
319
029
048
259
206
243
002
027
073
-0.353
-0.007
0.008
0.002
0.009
0.032
-0.005
0.005
0.003
-0.015
0.097
0.055
0.060
0.014
0.140
0.319
0.080
0.048
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
.293
.224
.248
.002
.035
.073
.353
.009
.027
.017
.019
.033
.023
,005
.027
.016
.097
Normalized
Bias (%) Err. (%)
31
70
-14
68
195
-19
100
11
-46
8
-165
-64
-64
-147
-16
0
9
13
54
5
15
10
-44
78
( 39)
( 50)
( 24)
( 4)
( 85)
( 12)
(105)
(124)
(132)
( 8)
( 62)
( 40)
( 19)
( 34)
( 55)
( 57)
( 21)
( 39)
( 66)
( 29)
( 47)
( 56)
( 14)
36
72
21
68
195
61
100
90
105
105
165
82
64
147
24
44
38
20
57
54
25
41
56
78
(a) Concentrations in ppm.
231
-------
Table 30. Comparisons of Average Normalized Biases and Errors in Maximum
Ozone and NO Oxidation Rates in the Simulations of Organic-
N0v-Air Runs Using the RADM, RADM-M, and RADM-P Mechanisms.
A
Organic or Mixture Mechanism
Ethene
Propene
1-Butene
trans-2-Butene
Isobutene
1-Hexene
Isoprene
a-Pinene
Ethane
n-Butane
C4+ Br. Alkane
RADM
RADM-M
RADM-P
RADM
RADM-M
RADM-P
RADM
RADM-M
RADM-P
RADM
RADM-M
RADM-P
RADM
RADM-M
RADM-P
RADM
RADM-M
RADM-P
RADM
RADM-M
RADM-P
RADM
RADM-M
RADM-P
RADM
RADM-M
RADM-P
RADM
RADM-M
RADM-P
RADM
RADM-M
RADM-P
Maximum Ozone (%)
Bias (sdev) Err.
-16 ( 30)
-16 ( 30)
-16 ( 30)
5 ( 20)
8 ( 20)
8 ( 20)
-2 ( 56)
2 ( 55)
2 ( 55)
-52 ( 38)
-53 ( 38)
-53 ( 38)
-23 ( 0)
-21 ( 0)
-21 ( 0)
57 ( 25)
59 ( 26)
59 ( 26)
-42 ( 3D
-42 ( 3D
-42 ( 3D
-8 ( 35)
-8 ( 34)
-8 ( 34)
-19 ( 0)
-18 ( 0)
-18 ( 0)
0 ( 59)
-10 ( 59)
-8 ( 59)
-17 ( 67)
-17 ( 73)
-19 ( 62)
23
23
23
15
16
16
45
45
45
54
54
54
23
21
21
57
59
59
43
43
43
28
27
27
19
18
18
46
47
47
53
58
50
d([03]-[NO])/dt (?)
Bias (sdev) Err.
-27 ( 20)
-27 ( 20)
-27 ( 20)
6 ( 19)
11 ( 19)
11 ( 19)
0 ( 32)
5 ( 32)
5 ( 32)
-25 ( 26)
-29 ( 24)
-29 ( 24)
-45 ( 0)
-38 ( 0)
-38 ( 0)
82 ( 25)
88 ( 26)
88 ( 26)
-2 ( 36)
-2 ( 36)
-2 ( 36)
4 ( 19)
2 ( 18)
2 ( 18)
6 ( 0)
7 ( 0)
7 ( 0)
19 ( 50)
12 ( 49)
13 ( 49)
-22 ( 24)
-25 ( 25)
-22 ( 24)
27
27
27
14
17
17
25
27
27
31
32
31
45
38
38
82
88
88
26
26
26
14
13
13
6
7
7
43
41
41
26
29
25
(continued)
232
-------
Table 30 (continued) - 2
Organic or Mixture
C5+ n-Alkane
Cycloalkane
Benzene
Toluene
Xylene
Mesitylene
Tetralin
Naphthalene
2 , 3-D imethy Inaph .
Simple Mixtures
Simple Surrogates
Mechanism
RADM
RADM-M
RADM-P
RADM
RADM-M
RADM-P
RADM
RADM-M
RADM-P
RADM
RADM-M
RADM-P
RADM
RADM-M
RADM-P
RADM
RADM-M
RADM-P
RADM
RADM-M
RADM-P
RADM
RADM-M
RADM-P
RADM
RADM-M
RADM-P
RADM
RADM-M
RADM-P
RADM
RADM-M
RADM-P
Maximum Ozone
Bias (sdev)
42 ( 28)
58 ( 39)
41 ( 27)
33 ( 0)
85 ( 0)
27 ( 0)
22 ( 15)
20 ( 13)
19 ( 14)
24 ( 43)
13 ( 3D
17 ( 3D
-4 ( 25)
0 ( 16)
5 ( 16)
-39 ( 50)
-26 ( 19)
-21 ( 19)
-6 ( 62)
57 ( 50)
56 ( 49)
84 ( 40)
93 ( 37)
93 ( 38)
-6 ( 24)
21 ( 17)
26 ( 17)
-12 ( 45)
-15 ( 43)
-14 ( 44)
7 ( 24)
8 ( 24)
9 ( 23)
(I) d([03]-[NO])/dt (%)
Err. Bias (sdev) Err.
42
58
41
33
85
27
22
20
19
33
21
22
18
12
12
48
29
25
43
57
56
84
93
93
17
21
26
34
32
32
15
16
16
24 ( 37)
38 ( 43)
25 ( 36)
3 ( 29)
25 ( 41)
-1 ( 3D
36 ( 19)
92 ( 23)
93 ( 24)
7 ( 34)
7 ( 3D
10 ( 3D
0 ( 35)
17 ( 20)
17 ( 21)
-93 ( 32)
-57 ( 22)
-55 ( 22)
106 ( 85)
125 ( 9D
125 ( 9D
134 ( 13)
170 ( 8)
170 ( 8)
1 ( 24)
79 ( 34)
79 ( 32)
-13 ( 26)
-15 ( 25)
-14 ( 25)
8 ( 38)
28 ( 36)
28 ( 36)
34
46
35
19
31
22
36
92
93
27
21
23
25
20
21
93
57
55
119
137
137
134
170
170
19
79
79
22
23
22
27
36
35
(continued)
233
-------
Table 30 (continued) - 3
Organic or Mixture
SAPRC 7-HC Surgs.
SAPRC 8-HC Surgs.
UNC Misc. Surgs.
UNC "Synurban"
UNC "Synauto"
UNC Auto Exhaust
Syn. Jet Fuel
Syn. Jet Exhaust
Mechanism
RADM
RADM-M
RADM-P
RADM
RADM-M
RADM-P
RADM
RADM-M
RADM-P
RADM
RADM-M
RADM-P
RADM
RADM-M
RADM-P
RADM
RADM-M
RADM-P
RADM
RADM-M
RADM-P
RADM
RADM-M
RADM-P
Maximum Ozone (%)
Bias (sdev) Err.
0 ( 11)
0 ( 12)
-1 ( 12)
9 ( 26)
3 ( 26)
5 ( 25)
17 ( 40)
11 ( 40)
10 ( 37)
39 ( 3D
22 ( 25)
24 ( 25)
23 ( 17)
16 ( 17)
18 ( 16)
8 ( 29)
1 ( 34)
3 ( 34)
14 ( 12)
16 ( 11)
15 ( 10)
-11 ( 8)
-9 ( 8)
-9 ( 8)
8
8
8
20
19
19
27
25
24
39
24
25
23
18
19
19
23
23
14
16
15
11
10
10
d([03]-[NO])/dt (%)
Bias (sdev) Err.
-5 ( 12)
-8 ( 12)
-6 ( 13)
-10 ( 23)
-5 ( 24)
-3 ( 25)
2 ( 15)
-3 ( 15)
-3 ( 15)
7 ( 8)
1 ( 9)
2 ( 9)
2 ( 18)
-2 ( 18)
-1 ( 18)
-6 ( 15)
-10 ( 15)
-9 ( 15)
11 ( 16)
107 ( 14)
87 ( 17)
-8 ( 4)
3 ( 5)
2 ( 5)
10
11
12
19
19
19
11
12
12
8
6
6
14
16
15
13
14
14
16
107
87
8
5
5
234
-------
RADMM Mechanism
_o
CL
Q.
X
0
0)
c
o
N
0 200 400 600 800 1000 1200
Experimental Ozone Maximum (ppb)
Figure 81. Calculated vs Experimental Maximum Ozone in the Surrogate
Mixture Runs for the RADM-M Mechanism.
RADM P Mechanism
UNC Mixes
Syn. Fuel
0 200 4QO 600 800 1000 1200
Experimental Ozone Maximum (ppb)
Figure 82. Calculated vs Experimental Maximum Ozone in the Surrogate
Mixture Runs for the RADM-P Mechanism.
235
-------
RADM-M Mechanism
1000
c 300 *-
X
o
o
NJ
o
600 -
400 -
0 200 400 600 800 'COO
Experimental Ozone Maximum (ppo)
Figure 83. Calculated vs Experimental Maximum Ozone in the Alkane-N0>
Runs for the RADM-M Mechanism.
RADM-P Mechanism
1000
c.
Q.
- SOO
x
O
0)
c
o
N
o
o
o
600
400 -
200
0 200 400 600 800 1000
Experimental Ozone Maximum (ppb)
Figure 84. Calculated vs Experimental Maximum Ozone in the Alkane-N0x
Runs for the RADM-P Mechanism.
236
-------
RADM M Mechamism
1000
_Q
CL
CL
X
0
0)
c
o
N
O
800
60C -
400 r
200 -
0 200 400 600 ' 800 1000
Experimental Ozone Maximum (ppb)
Figure 85. Calculated vs Experimental Maximum Ozone in the Aromatic-
NOV Runs for the RADM-M Mechanism.
A
RADM-P Mechanism
Mesitylene
A
0 200 400 600 800 1000
Experimental Ozone Maximum (ppb)
Figure 86. Calculated vs Experimental Maximum Ozone in the Aromatic-
NOV Runs for the RADM-P Mechanism.
A
237
-------
The greatest differences between the March 1988 and the recommended
modified mechanisms concern the aromatic chemistry. The RADM-M and RADM-P
have significantly less scatter in the simulations of the toluene, xylene,
and mesitylene-NOx runs, and their biases in the simulations of the
toluene and the mesitylene runs are also less. The apparently better
biases of the RADM mechanism relative to RADM-M and RADM-P (Table 30) in
the simulations of ozone yields and NO oxidation rates in the xylene runs
are due in large part to cancellation of large and opposite errors in
simulations of xylene runs carried out in the ITC compared to those
carried out in other chambers. (See Figure 40 for the relative perform-
ances of these mechanisms in simulating an ITC xylene run, and Appendix A
for a run-by-run listing showing the poor performance of the unmodified
mechanism in simulating the xylene runs.) It is interesting to note that
the differences between the mechanisms are the least in simulations of
runs in the UNC chamber and the greatest in simulations of runs in the
SAPRC ITC (see Figures 39 through 41 for examples).
Although the recommended modified mechanisms perform significantly
better in the simulations of runs employing alkylbenzenes, they perform
significantly worse in simulations of runs employing benzene or the
naphthalenes, where they consistently overpredict reactivity. The over-
prediction of reactivities of these compounds is due to the fact that they
are represented in the RADM mechanisms by model species (TOL and XYL)
which are optimized to simulate toluene and xylenes. Toluene and xylenes
are mechanistically much more reactive than benzene and naphthalenes (see
Carter et al., 1987 for a discussion of this). The fact that benzene
reacts slower than TOL is corrected for using "reactivity weighing," but,
this does not correct for differences in mechanistic aspects of reac-
tivity. The naphthalenes react more rapidly than XYL, and yet produce
less ozone (Carter et al., 1987). Note that all of the runs with benzene
and the naphthalenes were carried out in the SAPRC ITC (see Appendices A-
C), where the unmodified mechanism significantly underpredicts the
reactivities of alkylbenzene-NOx runs. (Figure 40 is typical in this
regard.) This underprediction for alkylbenzenes in the ITC tends to
"correct" for the inappropriate use of TOL to represent benzene and XYL to
represent naphthalenes, as shown, for example, in Figure 37 in the case of
benzene. Modifying the TOL and XYL mechanism so they perform
238
-------
satisfactorily in simulations of ITC toluene and xylene runs thus causes
the mechanism to be too reactive in the simulations of the ITC runs with
the mechanistically less reactive aromatics. This modification is still
appropriate, however, because alkylbenzenes are relatively more important
than benzene or the naphthalenes in the ambient atmosphere.
With two exceptions, the RADM-M and RADM-P mechanisms perform
comparably or better than the unmodified RADM mechanism in simulations of
runs employing organic mixtures. The major exception concerns the simula-
tions of the "synthetic Jet fuel" runs, which were ITC runs containing
relatively large amounts of naphthalenes. The other exception concerns
the "simple surrogate" runs. This series is dominated by the large number
of "mini-surrogate" runs carried out in the SAPRC ITC in 1982-83 (Atkinson
et al., 1983), and all versions of the RADM mechanism, and the SAPRC
mechanism as well (Carter, 1988), tend to .consistently overpredict the NO
oxidation rate. However, in most cases the statistical measures for
performance on the runs employing mixtures is not significantly different
for the three versions of the RADM mechanism.
The RADM-P and RADM-M mechanisms are comparable to the unmodified
version in the predictions of formaldehyde. The mechanisms have greater
differences in their predictions of PAN, with the RADM-M predicting
slightly higher, and RADM-P predicting slightly lower, yields of this
product for most, but not all, types of runs. The higher PAN predictions
for RADM-M are probably mostly due to the lumping of the RADM's aromatic
PAN-analog (TPAN) with PAN in the RADM-M mechanism. The causes for the
different PAN predictions for RADM-P relative to RADM are less obvious,
but must be due primarily to changes in the aromatics mechanisms, since
that is the area where the two mechanisms differ the most.
Comparison of the ozone predictions of the modified mechanisms
relative to the unmodified version is shown in Figures 87 and 88. These
give plots of ozone yields calculated using RADM-M and RADM-P, respective-
ly, against those calculated using the RADM mechanism for all of the
chamber runs employing organic mixtures. It can be seen that while there
are many experiments where the mechanisms give the same ozone predictions,
there are cases where the modified mechanisms predict less ozone, and
practically no cases where they tend to predict more. The tendency
towards predicting less ozone in simulating these chamber experiments with
239
-------
RADM-M vs. RADM
0.2 C.
-------
organic mixtures is slightly greater with RADM-M than RADM-P, though
except for one run this difference is slight. This suggests that the
recommended modifications on the average result in a slightly less
reactive mechanism with respect to ozone formation. Differences between
these mechanisms in simulations more directly representative of regional
modeling applications are discussed in Section 6.
211
-------
6. EVALUATION OF ALTERNATIVE CHEMICAL ASSUMPTIONS
An important task in the overall evaluation of a chemical mechanism
is investigation of alternative versions of the mechanism that incorporate
differing chemical approximations or degrees of chemical fidelity. In
this study, we developed a series of 90 test problems representing a wide
range of chemical conditions likely to be encountered in regional model
simulations and used them to assess the effects of selected modifications
of the mechanism on model predictions using the RADM mechanism. These
included calculations to examine effects of using varying levels of
details in representing peroxy radical reactions in the mechanism,
calculations to determine the effects of using a more condensed represen-
tation of the reactions of the higher alkanes, and calculations to deter-
mine the differences in the predictions of the recommended modified
versions of the RADM mechanism relative to the March 1988 version. The
results of these tests are discussed in this section. Tests of several
other mechanistic options were carried out during the course of this study
and are referred to where appropriate in other portions of this report.
The test problems developed in this study can serve as a basis for further
studies of the effects of alternative chemical assumptions on regional
model predictions and the future development of more condensed or effi-
cient mechanisms for use in such models.
6.1 Test Problems for Sensitivity Analysis
A set of test problems that span the conditions of importance in
regional acid deposition modeling is an essential component to any
investigation of sensitivities of mechanism predictions to alternative
formulations. If the set of conditions is not sufficiently comprehensive,
then conditions where a set of alternative assumptions may be important
may not be represented, resulting in the possibility of misleading conclu-
sions regarding the sensitivity of model predictions to these assump-
tions. Several aspects of regional modeling must be considered in the
selection of test problems. First, regional modeling involves transport
of pollutants over long distances, so it is essential to use multi-day
test problems. Second, the vertical domain of the regional models extends
from the earth's surface to the tropopause. Atmospheric temperature,
242
-------
pressure, solar radiation intensity, and the radiation's spectral distri-
bution vary significantly from the lower to the upper troposphere. The
test conditions must address these variations. Third, the regional models
are designed to simulate the chemical and physical evolution of emissions
as they are transported from source areas to remote areas. Therefore, it
is important for the test conditions to span the range of pollutant con-
centrations encountered in both urban and rural regions. This can be
accomplished by varying the VOC and NOV concentrations, VOC-to-NOv ratio,
A A
and VOC composition.
The ideal framework in which to test alternative chemical mechanisms
is in a three-dimensional transport model such as RADM using actual
meteorological conditions. However, this approach is not practical
because of the large computing resources needed for the transport model
and the large effort required to both install an alternative mechanism in
a transport model and develop mechanism specific speciated VOC emissions
inventories. Even if this were practical, the fact that the meteoro-
logical inputs have only been developed for a small number of episodes
limits the utility of this approach.
An alternative approach adopted here is to use a simple photochemical
box model and exercise the model for a large number of cases that span the
expected range of conditions encountered in actual regional modeling. The
box modeling approach can sometimes produce unrealistic pollutant concen-
trations because of its simplistic treatment of transport. Nevertheless,
it is well suited to investigating the relative differences in alternative
chemical mechanisms. In fact, most mechanism comparisons have been
performed with photochemical box models (Jeffries et al., 1981; Dunker,
1984; Leone and Seinfeld, 1985; Schafer and Seinfeld, 1985; Stockwell,
1986; Hough, 1987; Milford, 1988; Stockwell and Lurmann, 1989; Dodge,
1989).
The matrix of the test problems selected for the analysis involves a
relatively small number of environmental conditions and a large number of
chemical conditions. The rationale for this approach is that alternative
mechanisms are expected to respond similarly to changes in environmental
conditions and differently to variations in the chemical mixtures. The
three sets of environmental conditions considered are shown in Table 31.
These constant temperature cases include summer conditions at surface
243
-------
Table 31. Environmental Conditions Used in Test Problems for Sensitivity
Calculations
Parameter
Simulation Duration
Start Time (LDT) (a)
End Time on Last Day (LDT)
Summer
Surface
0500
2000
Number of Days in Simulation (b) 2-5
Light Conditions (c)
Date
Latitude (deg)
Elevation (km)
Meteorological Conditions
Temperature (K)
Pressure (atm)
Rel. Humidity (%)
Water Vapor (ppm)
Inversion Height Schedule
Background Pollutant
Concentrations (ppb) (e)
Oo
CO
CH4
Daytime Dry Deposition
Velocities (cm/sec):
Q
so2
Sulfate
HpOp
N02
June 21
40
0
298
1
50
15,600
LDT (m)
(d) 0539 800
1116 1116
1501 1200
20-40 (f)
200
1800
0.6
0.8
0.4
1.6
0.4
PAN and Organic Nitrates (ONIT) 0.4
Organic peroxides (OP1
Organic Acids (ORA1 and
HNO,
and OP2) 0.8
ORA2) 0.8
3.0
Winter
Surface
0500
2000
3-5
December 21
40
0
273
1
50
3,000
LDT (m)
0805 800
1133 958
1352 1000
20
200
1800
0.3
0.4
0.2
0.8
0.2
0.2
0.4
0.4
1.5
Summer
Aloft
0500
2000
3-5
June 21
40
6
248
0.5
10
160
(Constant)
0
200
1800
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
(continued)
244
-------
Table 31 (continued) - 2
(a) LDT = Local Daylight Time.
(b) Number of days of simulation depended on chemical conditions. See
Table 34.
(c) Solar flux data output by the RADM radiation module was provided
by Stockwell (private communication, 1988). The data were given in
terms of actinic fluxes for 93 wavelength intervals between 280 and
735 nm for zenith angles of 0, 10, 20, 30, 40, 50, 60, 70, 78, and
86. These fluxes are available in computer readable form from the
authors upon request. The zenith angle dependencies of these
actinic fluxes for the various wavelength intervals were fit using a
parameterized function similar to that used by Carter et al. (1986),
which was then used to calculate the actinic flux at any given
zenith angle in the simulation.
(d) Inversion height is constant at initial value shown at beginning
of the day, then changed linearly with time to the values at the
times indicated. The inversion height was constant at the final
value for the remainder of the day. The inversion height schedule
is based on a simplified approximation of the "characteristic
curve" used in the OZIPM software (Gipson, 1984). The final
inversion heights were determined to obtain the desired amount of
dilution for each simulated day.
(e) "Background" species concentrations were added to the initial
pollutant concentrations in the simulations, and (for the ground
level simulations) are entrained into the air mass as the inversion
height increases.
(f) 20 ppb background 0^ was used in the "rural" summer surface
simulations with initial NCL levels of 0.5 ppb. 40 ppb background
ozone was used in all the other summer surface simulations.
245
-------
elevation, winter conditions at surface elevation, and summer conditions
at 6 km above the surface. The solar actinic fluxes used in the simula-
tions of these conditions were obtained from the RADM radiation module
(Stockwell, personal communication, 1988). Solar radiation for the
simulations is based on the diurnal cycle at MO degrees latitude in the
summer and winter solstices. A small amount of dilution (25% per day in
the winter case and 50? per day in the summer case) with relatively clean
background air is included along with very approximate dry deposition
velocities to inhibit build-up of unrealistically high concentrations.
Only daytime dry deposition is included in the simulations. Nighttime
deposition is assumed to be negligible.
All the winter surface and summer aloft simulations were "static"
with respect to pollutant injection. All pollutants were assumed to be
present initially, with no subsequent injection of pollutants other than
background species entrained from aloft as the inversion height
increased. "Static" and "dynamic" simulations were carried out for the
summer surface conditions. In the "dynamic" simulations, no pollutants
were present initially (other than the background species), and pollutants
were injected at a constant flux (moles per unit area) throughout the
simulation. Thus, four different types of environmental conditions are
used in the simulations:
(1) Summer Surface - Static (designated "Summer Surface")
(2) Summer Surface - Dynamic (designated "Summer Dynamic")
(3) Winter Surface - Static (designated "Winter Surface")
(4) Sunnier Aloft - Static (designated "Summer Aloft")
The chemical conditions for these runs can be described as "rural" or
"urban," with the former having the relatively low pollutant levels
characteristic of rural conditions and the latter having the higher
pollutant levels more typical of urban areas. For each type of chemical
condition, the total NOX and reactive organic (ROG) levels were varied
over relatively wide ranges, and several different mixtures representing
ROG pollutants were employed. The ROG mixtures employed in these simula-
tions were chosen to represent typical ambient mixtures of varying
2U6
-------
composition and reactivity, with three different compositions used in the
urban simulations and two used in the rural runs.
The three urban ROG composition profiles are based on 6-9 AM ambient
urban samples collected by EPA/AREAL (under Bill Lonneman's direction) in
41 U.S. cities during 1984-1986. These hydrocarbon data have been
thoroughly analyzed by Jeffries et al. (1989c) to derive the recommended
carbon fractions for the "Carbon Bond" and the "CAL" mechanisms. The
specific profiles used in these simulations are the average of the 773
samples collected in all of the cities (referred to as the "allcity"
average), the average of the 8 samples collected in Washington, DC, during
1986 and the average of 9 samples collected in Beaumont, Texas during
1984. The latter two were selected because their composition and reac-
tivity are most different from the average. The Beaumont profiles have
unusually high alkane fractions and low aromatic and C3+ alkene frac-
tions. The Washington DC profiles have unusually high levels of aromatics
and low levels of alkanes. This data base does not contain aldehyde
measurements, and for all three profiles we followed the standard assump-
tion employed by the EPA and its contractors (e.g., EPA, 1984; Lurmann et
al., 1987; Jeffries et al., 1989c) that urban ROG profiles consist of 5%
aldehydes (on a carbon basis). This is based on an upper limit estimate
derived from data collected in Los Angeles (Grosjean, 1983). It was
assumed that these urban aldehydes consist of 60% formaldehyde and 40?
acetaldehyde (on a carbon basis - as did Lurmann et al., 1987), which is
slightly different than the 40£ formaldehyde-60/& acetaldehyde assumption
employed by Jeffries et al. (1989c). Both assumptions are consistent with
the available data. These three urban ROG surrogates are similar to those
used in the mechanism comparison study recently carried out by Dodge
(1989).
The ROG profiles employed in the rural simulations are based on 53
morning samples collected above the mixed layer by aircraft in four cities
(Westburg and MacGregor, 1987). These measurements include aldehydes.
These data have also been extensively analyzed by Jeffries et al. (1989c)
for determination of typical ROG concentrations aloft. One of the two
rural profiles is based on the average composition of these data. This is
referred to as the "aloft" surrogate. The other rural surrogate is
designed to represent air masses with significant biogenic hydrocarbon
247
-------
input. It has 46% isoprene (on a carbon basis), the same aldehyde levels
as the aircraft data, and the remaining hydrocarbons apportioned based on
the aircraft data. This is referred to as the "aloft+isoprene" surrogate.
The detailed compositions used for these five surrogate mixtures are
given in Table 32. These were taken from the detailed data sets provided
by Jeffries (private communication, 1988). As is the case with most
ambient ROG samples, numerous assumptions were needed to assign the
unidentified and ambiguously classified carbon into known species. These
detailed profiles were converted to profiles of specific compounds (or
groups of compounds assumed to have the same reactivity) based on (1) the
assumptions employed by Jeffries et al. (1989c) concerning ambiguous
classes of olefins and aromatics and concerning the classifications of
species such as alkynes and styrenes, and (2) assuming the "unknown" data
were hydrocarbons with the same distribution of species as the identified
species in the profiles. These individual species were then converted
into species in the RADM mechanism using the same set of procedures used
in processing organic emissions input into the RADM-II model, discussed in
Section 3. The distribution of RADM-II species for these five surrogates
is given in Table 33. Note that in many cases the molar concentrations of
the lumped RADM species do not add up exactly to the totals for the
species they represent because of the use of "reactivity weighting" in the
RADM emissions processing system (see Section 3).
Table 34 summarizes the conditions of all the test calculations,
including the environmental conditions, the duration of the simulation,
the ROG surrogate mixture, and the ROG and NOX levels employed for each
run. The minimum duration of the simulations was two days. The maximum
duration was either five days or one day more than needed for complete
consumption of the NOV, whichever was less. With the exception of a few
A
simulations at very low ROG/NOX ratios, the simulations were carried out
for a sufficient amount of time that the effects of alternative assump-
tions concerning the radical-radical reactions occurring in the absence of
NOX could be adequately tested. For the rural simulations, the ROG
concentrations ranged from 10 to 100 ppbC, the NOX concentrations from 0.5
to 5 ppb, and the initial ROG-to-NOx ratio from 2 to 200. For the urban
simulations, the ROG concentration ranged from 200 to 2000 ppbC, the NOX
concentration from 20 to 200 ppb, and the initial ROG-to-NOx ratio from 2
248
-------
Table 32. Compositions of ROG Surrogate Mixtures Used in Test
Calculations
Compound and i
Emissions Group (a)
Ethane
Propane
Alkanes (0.25-0.50) (b)
n- Butane
Isobutane
Neopentane
Unspeciated C4-C5 Alkanes
Alkanes (0.50-1.00)
n-Pentane
Iso-Pentane
Cyclopentane
n-Hexane
2-Methyl Pentane
3-Methylpentane
2 , 3-Dimethyl Butane
Branched C6 Alkanes
Alkanes (1.00-2.00)
Cyclohexane
C6 Cycloalkanes
n-Heptane
Branched C7 Alkanes
C7 Cycloalkanes
n-Octane
Branched C8 Alkanes
C8 Cycloalkanes
n-Nonane
Branched C9 Alkanes
n-Decane
Branched C10 Alkanes
Alkanes (>2.00)
Branched C11 alkanes
Branched C12 Alkanes
Branched C13 Alkanes
Composition (ppb Species/ppmC ROG Surrogate)
Allcity
Average
20.78
15.47
16.39
6.91
0.39
0.01
6.81
14.09
0.63
2.94
3.82
2.74
1.01
0.89
0.84
1.66
1.12
5.17
0.84
0.59
4.89
0.10
0.46
1.92
0.63
1.38
0.55
0.13
0.00
Washing-
ton DC
24.33
11.35
19.92
5.78
0.20
0.00
5.37
16.31
0.50
1.98
3.51
3.03
1.17
1.84
0.27
1.43
0.89
4.46
0.58
0.58
4.69
0.05
0.42
1.82
0.90
0.96
1.09
0.18
0.00
Beaumont
TX
38.15
27.40
24.62
12.49
0.57
0.00
8.50
14.48
1.11
4.03
4.36
3.00
1.08
0.72
1.38
2.15
0.95
3.65
1.17
0.39
3.54
0.10
0.33
1.71
0.33
1.26
0.20
0.11
0.00
Aloft
111.06
48.47
23.64
12.69
0.00
0.00
7.14
11.59
2.29
2.02
2.44
1.32
0.57
0.00
0.45
0.86
0.00
2.57
0.00
0.35
0.90
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
Aloft
+ Isop.
54.42
23.91
11.70
6.28
0.00
0.00
3.54
5.75
1.14
1.00
1.21
0.66
0.28
0.00
0.22
0.43
0.00
1.28
0.00
0.17
0.45
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
(continued)
249
-------
Table 32 (continued) - 2
Compound and
Emissions Group (a)
Ethene
Propene
Other Terminal Monoalkenes
1-Butene
1-Pentene
1-Hexene
2-Methyl-1-Butene
C4 Terminal Alkanes
C5 Terminal Alkanes
C6 Terminal Alkanes
C7 Terminal Alkanes
C8 Terminal Alkanes
C9 Terminal Alkanes
C10 Terminal Alkanes
C11 Terminal Alkanes
Internal Alkenes, Dienes,
cis-2-Butene
trans-2-Butene
1,3-Butadiene
2-Methyl-2-Butene
C5 Terminal Alkenes
C6 Terminal Alkenes
C7 Terminal Alkenes
C8 Terminal Alkenes
a-Pinene
Benzene
Aromatics (<2.00)
Toluene
Ethyl Benzene
Isopropyl Benzene
n-Propyl Benzene
Monoalkyl Benzenes
Composition (ppb Species/ppmC ROG Surrogate)
Allcity Washing-
Average ton DC
19.18
4.93
2.49
0.54
0.21
0.82
1.72
1.30
0.51
1.14
0.40
0.71
0.03
0.40
and Pinenes
0.79
1.02
0.67
0.11
1.99
0.93
0.10
0.02
0.61
3.43
8.19
1.23
0.14
0.28
1.67
16.33
3.25
2.56
0.43
0.13
0.81
0.25
1.42
0.31
0.82
0.40
0.19
0.15
0.01
0.64
0.74
0.26
0.00
1.25
0.61
0.05
0.00
0.18
3.58
11.18
1.61
0.07
0.39
2.05
Beaumont
TX
23.27
5.16
1.78
0.53
0.09
0.72
1.33
0.98
0.47
0.71
0.20
0.41
0.01
0.76
0.81
1.09
0.44
0.03
1.70
0.72
0.04
0.03
0.28
2.31
4.09
0.71
0.05
0.18
0.93
Aloft
14.98
2.57
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
4.07
4.87
0.88
0.00
0.00
0.00
Aloft
+ Isop.
7.34
1.26
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
2.02
2.42
0.44
0.00
0.00
0.00
(continued)
250
-------
Table 32 (continued) - 3
Compound and
Emissions Group (a)
Composition (ppb Species/ppmC ROG Surrogate)
Allcity Washing- Beaumont Aloft Aloft
Average ton DC TX + Isop.
Aromatics (>2.00)
m-Xylene
o-Xylene
p-Xylene
Dialkyl Benzenes
1 ,2,3-Trimethyl Benzene
Trialkyl Benzenes
Tetraalkyl Benzenes
Acetylene
Formaldehyde
Acetaldehyde
Isoprene
1.93
1.53
1.93
4.80
2.88
1.92
0.18
11.43
30.00
10.00
0.67
2.54
2.37
2.54
5.22
4.03
1.89
0.11
10.55
30.00
10.00
0.61
1.08
0.72
1.08
2.98
1.41
1.44
0.10
4.82
30.00
10.00
0.78
0.72
2.03
0.72
0.00
0.00
0.00
0.00
8.99
74.86
10.75
3.42
0.36
1.01
0.36
0.00
0.00
0.00
0.00
4.41
74.86
10.75
92.75
(a) "Emissions Group" is the group the individual compounds are lumped
into during processing of emissions data (see Section 3). If a
compound is given by itself (e.g., ethane), it is in an emissions
group by itself. If more than one compound (or group of compounds)
is lumped into an emissions group, [e.g., n-butane, isobutane, etc.,
in group "Alkanes (0.25-0.50)"], the individual compounds or groups
are listed below the emissions group name. .
(b) Quantities in parentheses refer to range of OH radical rate constants
(in units of 10 ppm min ) for compounds in the emissions group.
251
-------
Table 33. Compositions of ROG Surrogates Used in the Test Calculations
for the RADM Mechanism
RADM Description
Species (a)
Name
ETH
HC3
HC5
HC8
OL2
OLT
OLI
ISO
TOL
XYL
HCHO
ALD
Ethane
Alkanes (0.25-0.50)
Alkanes (0.50-1.00)
Alkanes (>1.00)
Ethene
Terminal Alkenes
Internal Alkenes
Isoprene
Aromatics (<2.00)
Aromatics (>2.00)
Formaldehyde
Higher Aldehydes
Composition (ppb RADM Species/ppmC ROG)
Allcity
Average
20.78
35.54
29.60
18.64
19.18
15.21
6.24
0.67
12.54
15.16
30.00
10.00
Washing-
ton DC
24.33
35.22
30.31
16.95
16.33
10.73
3-72
0.61
16.37
18.71
30.00
10.00
Beaumont Aloft
TX
38.15
53-33
33.53
15.81
23.27
13.15
5.13
0.78
6.65
8.81
30.00
10.00
111.1
64.63
24.61
4.68
14.98
2.65
0.0
3.42
6.97
3.48
74.86
10.75
Aloft
+ I sop.
54.42
31.93
12.21
2.33
7.34
1.27
0.0
92.75
3.47
1.73
74.86
10.75
(a) Quantities in parentheses for lumped higher alkanes and. aromatics
refer to OH radical rate constant ranges in units of 10 ppm
min"1.
252
-------
Table 34. Summary of Conditions of Test Calculations Used for
Sensitivity Studies
No. Environmental ROG
Condition Surrogate
Reactant Levels (a)
ROG
NOX ROG/NOX
No.
Days
Rural Simulations
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
Summer
Summer
Summer
Summer
Summer
Summer
Summer
Summer
Summer
Winter
Winter
Winter
Summer
Summer
Summer
Summer
Summer
Summer
Summer
Summer
Summer
Summer
Summer
Summer
Summer
Summer
Summer
Winter
Winter
Winter
Summer
Surface
Surface
Surface
Surface
Surface
Surface
Surface
Surface
Surface
Surface
Surface
Surface
Aloft
Aloft
Aloft
Dynamic
Dynamic
Dynamic
Surface
Surface
Surface
Surface
Surface
Surface
Surface
Surface
Surface
Surface
Surface
Surface
Aloft
Summer Aloft
Summer Aloft
Aloft
Aloft
Aloft
Aloft
Aloft
Aloft
Aloft
Aloft
Aloft
Aloft
Aloft
Aloft
Aloft
Aloft
Aloft
Aloft
Aloft
Aloft
Aloft-i-lsoprene
Aloft+Isoprene
Aloft+Isoprene
Aloft+Isoprene
Aloft+Isoprene
Aloft+Isoprene
Aloft-i-Isoprene
Aloft+Isoprene
Aloft+Isoprene
Aloft+Isoprene
Aloft+Isoprene
Aloft+Isoprene
Aloft+Isoprene
Aloft+Isoprene
Aloft+Isoprene
10
33
100
10
33
100
10
33
100
10
33
100
10
33
100
10
33
100
10
33
100
10
33
100
10
33
100
10
33
100
10
33
100
0.5
0.5
0.5
1.7
1.7
1.7
5.0
5.0
5.0
1.7
1.7
1.7
1.7
1.7
1.7
1.7
1.7
1.7
0.5
0.5
0.5
1.7
1.7
1.7
5.0
5.0
5.0
1.7
1.7
1.7
1.7
1.7
1.7
20
66
200
6
20
60
2
7
20
6
20
60
6
20
60
6
20
60
20
66
200
6
20
60
2
7
20
6
20
60
6
20
60
3
3
3
2
2
2
3
3
3
5
4
3
5
4
3
3
4
4
5
5
5
2
2
2
3
3
3
5
4
3
5
4
3
(continued)
253
-------
Table 34 (continued) - 2
No.
34
35
36
Environmental
Condition
Summer
Summer
Summer
Dynamic
Dynamic
Dynamic
ROG
Surrogate
Aloft+Isoprene
Aloft+Isoprene
Aloft+Isoprene
Reactant Levels (a)
ROG
10
33
100
NOX
1.7
1.7
1.7
ROG/ NO v
A
6
20
60
No.
Days
3
4
4
Urban Simulations
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
Summer
Summer
Summer
Summer
Summer
Summer
Summer
Summer
Summer
Winter
Winter
Winter
Summer
Summer
Summer
Summer
Summer
Summer
Summer
Summer
Summer
Summer
Summer
Summer
Summer
Summer
Summer
Winter
Winter
Winter
Surface
Surface
Surface
Surface
Surface
Surface
Surface
Surface
Surface
Surface
Surface
Surface
Aloft
Aloft
Aloft
Dynamic
Dynamic
Dynamic
Surface
Surface
Surface
Surface
Surface
Surface
Surface
Surface
Surface
Surface
Surface
Surface
Allcity Avg
Allcity Avg
Allcity Avg
Allcity Avg
Allcity Avg
Allcity Avg
Allcity Avg
Allcity Avg
Allcity Avg
Allcity Avg
Allcity Avg
Allcity Avg
Allcity Avg
Allcity Avg
Allcity Avg
Allcity Avg
Allcity Avg
Allcity Avg
Washington
Washington
Washington
Washington
Washington
Washington
Washington
Washington
Washington
Washington
Washington
Washington
,
.
m
.
^
»
.
.
.
DC
DC
DC
DC
DC
DC
DC
DC
DC
DC
DC
DC
200
667
2000
200
667
2000
200
667
2000
200
667
2000
200
667
2000
200
667
2000
200
667
2000
200
667
2000
200
667
2000
200
667
2000
67.
67.
67.
20.
20.
20.
100.
200.
200.
67.
67.
67.
67.
67.
67.
67.
67.
67.
67.
67.
67.
20.
20.
20.
100.
200.
200.
67.
67.
67.
3
10
30
10
33
100
2
3
10
3
10
30
3
10
30
3
10
30
3
10
30
10
33
100
2
3
10
3
10
30
3
3
2
3
2
2
5
3
3
5
5
4
5
4
3
5
5
5
3
3
2
3
2
2
5
3
3
5
5
4
(continued)
254
-------
Table 34 (continued) - 3
No.
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
Environmental
Condition
Summer Aloft
Summer Aloft
Summer Aloft
Summer Dynamic
Summer Dynamic
Summer Dynamic
Summer Surface
Summer Surface
Summer Surface
Summer Surface
Summer Surface
Summer Surface
Summer Surface
Summer Surface
Summer Surface
Winter Surface
Winter Surface
Winter Surface
Summer Aloft
Summer Aloft
Summer Aloft
Summer Dynamic
Summer Dynamic
Summer Dynamic
ROG
Surrogate
Washington DC
Washington DC
Washington DC
Washington DC
Washington DC
Washington DC
Beaumont TX
Beaumont TX
Beaumont TX
Beaumont TX
Beaumont TX
Beaumont TX
Beaumont TX
Beaumont TX
Beaumont TX
Beaumont TX
Beaumont TX
Beaumont TX
Beaumont TX
Beaumont TX
Beaumont TX
Beaumont TX
Beaumont TX
Beaumont TX
Reactant Levels (a)
ROG
200
667
2000
200
667
2000
200
667
2000
200
667
2000
200
667
2000
200
667
2000
200
667
2000
200
667
2000
NOX
67.
67.
67.
67.
67.
67.
67.
67.
67.
20.
20.
20.
100.
200.
200.
67.
67.
67.
67.
67.
67.
67.
67.
67.
ROG/NOX
3
10
30
3
10
30
3
10
30
10
33
100
2
3
10
3
10
30
3
10
30
3
10
30
No.
Days
5
4
3
5
5
5
3
3
2
3
2
2
5
3
3
5
5
4
5
4
3
5
5
5
(a) For the static calculations ("summer surface," "winter surface,"
and "summer aloft"), the reactant levels reTer to initial
concentrations in ppbC (for ROG) or ppb (for NOX), excluding
background species. For the "summer dynamic" concentrations, the
reactant levels refer to the increase in concentrations of ROG (in
ppbC) or NOX (in ppb) that would result following one day of
continuous emissions if there were no chemical reaction and the
inversion height remained constant at its initial level of 800
meters.
255
-------
to 100. These combinations of ROG and NOV adequately bracket probable
A
atmospheric concentrations in most rural and urban areas. This set
consists of a total of 90 simulations, 36 representing rural conditions
and 54 representing urban pollution.
6.2 Presentation of Sensitivity Test Results
The results of test calculations using alternative mechanisms can be
presented in a number of ways. Many sensitivity studies have used plots
to compare concentrations of species of interest. While this is a useful
approach when a relatively small number of test problems are used, it is
less practical when the number of test cases is relatively large, as is
the case in this study. This is analogous in some ways to the problems of
summarizing the performance of a mechanism in simulating the results of a
large number of chamber experiments. Therefore, in presenting the results
of the test calculations, we use an approach similar to that used in the
evaluation against the chamber data the results are presented in the
form of tabulations of statistical measures of the comparisons of the
mechanisms, with examples given for results of individual calculations
which are of interest.
Statistics are tabulated for the following species, which are of
known or potential interest in regional acid deposition model simulations
(RADM model species name used in the tabulations are given in
parentheses):
1. Ozone (03);
2. Hydrogen Peroxide (H202);
3. Sulfuric Acid or Sulfates (SULF);
4. Nitric Acid (HN03);
5. Formic Acid (ORA1);
6. Higher Organic Acids (ORA2);
7. Methyl Hydroperoxide (OP1);
8. Higher Organic Hydroperoxides (OP2);
9. Peroxy Acetic Acid (PAA); and
10. PAN and Higher Peroxy Nitrates (PAN).
256
-------
Given the results of model calculations using two different models (one
designated the "standard" calculation, and the other generally having a
modification to the "standard" mechanism whose effect is being examined
designated the "test" calculation), the following statistics are tabulated
for each of these species for each of the test problems:
1. Maximum concentration for the standard calculation;
2. Percent change in the maximum concentration in the test
calculation relative to the standard calculation;
3. Average concentration for the "standard" calculation;
4. Percent change in the average concentration in the test
calculation relative to the standard calculation;
5. Percent change in the average concentration, relative to the
maximum average concentration for,the set (see below); and
6. The "Percent fit" statistic, calculated as discussed below.
The statistics are calculated for the entire time duration of these
multi-day calculations. The first four of the above statistics are
derived in the usual manner. A positive number means that the test
calculation predicts a larger maximum or average than the standard
calculation, while a negative number means it predicts a lower value. The
tabulations indicate which of the mechanisms is used as the "standard" and
which is used as the "test."
The fifth statistic, designated "% /Max" on the tabulations, is
calculated for a given species in a given test calculation as follows:
(Average Cone., (Average Cone.
Test Calc.) - Standard Calc.)
"% /Max" = 100 x
(Maximum of the Averages for the Group)
The purpose of this statistic is to obtain an indication of the changes in
absolute concentrations of the species for an entire set of calcula-
tions. These are normalized to the maximum value for the set to allow for
meaningful comparisons with predictions obtained for other species whose
concentration may be quite different. Thus, if there is a large positive
or negative magnitude in this statistic, it means that the two models are
257
-------
significantly different in their calculations of this species, and it is
not a case of a large relative difference in the calculation of an
insignificant concentration. For the purpose of calculating this
statistic, the calculations are grouped into "rural" and "urban"
problems. This is to account for the fact that concentration regimes for
at least some of the species of interest might be different for these
types of problems.
The sixth statistic, designated as "% Fit" on the tabulations,
measures the overall differences in the simulations of the species on a
point-by-point basis, paired in time. It is calculated by
i |T, - s,i
'% Fit" =
0.5 x Z (Ti + S^
where T* and S. are the concentrations of the species for the test and the
standard calculation, respectively, at time "i," and the summations are
over all the times where data are saved in the calculations, which is once
every hour. The purpose of this statistic is to make evident differences
in curve shapes which might still have similar average or maximum values.
6.3 Treatment of Peroxy Radical Reactions
Alkyl and acyl peroxy radicals are important intermediates in the
gas-phase atmospheric reactions of organic compounds. In the presence of
NOX their major fates are reaction with NO and (for acyl peroxy radicals)
with NOo, and these reactions are important in affecting ozone forma-
tion. When NOX levels are low, their major fates are reaction with H02 or
with other peroxy radicals, and these reactions are important in affecting
predictions of H202 and organic hydroperoxy species. Therefore, the
appropriate representation of the reactions of these species is important
in acid deposition models, where accurate predictions of both oxidant and
hydroperoxide formation are essential. This includes appropriate repre-
sentations of peroxy-peroxy radical reactions, since sensitivity calcula-
tions we have carried out previously (Carter et al., 1986) and more recent
calculations carried out by Dodge (1989) indicate that they cannot be
ignored. However, because of the relatively large number of organic
peroxy species which must be included in even relatively condensed gas-
258
-------
phase atmospheric representations, almost all current mechanisms represent
these peroxy-peroxy radical reactions in an approximate manner. The RADM
mechanism is no exception in this regard.
As indicated in Section 2, the RADM mechanism has a relatively
detailed representation of peroxy radicals, using separate species to
represent peroxy radicals formed from most of the organic model species,
and representing explicitly their reactions with NO, NOp (for acylperoxy
radicals), HC^, methyl peroxy radicals (M02), and acetyl peroxy radicals
(AC03). However, this mechanism ignores all the reactions of organic
peroxy radicals with organic peroxy radicals other than M02 or AC03.
Stockwell (private communication) conducted test calculations which
indicated that using only reactions with M02 and AC03 should be suffi-
cient. However, the description and results of these tests have not been
included with the available documentation of the mechanism. In view of
the importance of peroxide and hydroperoxide predictions in acid
deposition models, we felt that an independent test of the validity of
this approximation in the RADM mechanism would be a useful component of
this evaluation study.
On the other hand, it may well be that the RADM mechanism is more
detailed in its representation of peroxy radical reactions than is
necessary. The latest Carbon Bond (Gery et al., 1988) and SAPRC (Carter,
1989a) mechanisms, and the chemical mechanism for the ADOM model (Lurmann
and Karamchandani, 1987) employ much more condensed representations of
these reactions. These mechanisms do not have separate reactions for the
many kinds of peroxy radical species; instead they use a limited number of
chemical "operators" to represent the net effects of their reactions. The
representations used in these other current mechanisms are too condensed
to permit the separate representation of methyl hydroperoxide and the
higher hydroperoxides which are utilized in the RADM mechanism. Since
Stockwell (private communication) considers separate representation of
methyl and the higher hydroperoxides to be important for RADM, the more
condensed approaches which require that these be lumped together were not
examined in this study.
However, an earlier version of the SAPRC mechanism (Carter et al.,
1986) uses a less condensed representation which allows for separate
representation of the two types of hydroperoxide species, and also allows
259
-------
for all the possible organic peroxy + organic peroxy reactions to be
represented without using many more reactions than used for this purpose
in the RADM mechanism. A detailed discussion of this approach is beyond
the scope of this report (see Carter et al., 1986), but briefly it
involves using chemical operators to account for total organic peroxy
("R02") and acylperoxy ("RC03") radical concentrations, and then using
these to calculate rates of reactions of the individual organic peroxy
radicals with themselves and all other organic peroxy radicals. This
allows for all the possible radical combinations to be represented, but
requires using the same rate constants for all these reactions. By
contrast, the RADM mechanism is more approximate in the sense that it
represents only a subset of the possible peroxy + peroxy reactions, but is
more detailed in the sense that it uses separate estimated rate constants
for the various individual reactions which it does include. Therefore,
the effect of modifying the RADM mechanism so that it incorporates the
SAPRC approach was also evaluated in this study.
In order to test effects of alternative approximation methods of
representing the organic peroxy radical reactions, it is necessary to have
a detailed representation against which to compare them. For this
purpose, Stockwell (private communication, 1988) provided SAPRC with a
"detailed radical" version of the RADM mechanism, which included all the
possible combination reactions of the organic peroxy and acylperoxy
species in the RADM mechanism. Other than the additional reactions of the
organic peroxy and acylperoxy radical species with peroxy radical species
other than M02 and AC03 (and the inclusion of the OLN + M02 and OLN + AC03
reactions), this mechanism was exactly the same as the March 1988 version
of the RADM mechanism which was evaluated in this study. The approach
Stockwell used to estimate the rate constants and products for these
additional reactions was consistent with the approach used for the reac-
tions in the March 1988 mechanism. Almost all of these additional reac-
tions involve estimates which are highly uncertain, and in some cases we
may have made different estimates were we required to independently derive
such a mechanism. However, because this detailed mechanism and the March
1988 RADM mechanism are consistent with each other in their approaches to
the chemistry, any differences between their predictions would reflect
only the effects of the approximations in the RADM mechanism regarding the
260
-------
peroxy radical reactions. Therefore, the detailed radical mechanism
provided by Stockwell was used without modification as the standard in
this evaluation.
The organic peroxy + organic radical reactions in Stockwell's
detailed radical mechanism, which was used as the standard mechanism in
the tests of peroxy radical representations, are listed in Table 35. All
the other reactions are the same as those listed in Table 2 in Section 2
of this report, excluding those indicated as "Organic Peroxy + M02,"
"Organic Peroxy + AC03," "Other Organic Peroxy + Peroxy" and "XN02"
reactions.
The reactions of organic species in the version of the RADM mechanism
employing the approach of the 1986 SAPRC mechanism (Carter et al., 1986)
are given in Table 36. All organic reactions have to be shown because
there are modifications made throughout . this portion of the mechanism.
The inorganic reactions are the same as those used for all versions of the
RADM mechanism and are given in Table 2. Except for the reactions involv-
ing organic peroxy radicals, this mechanism is the same as the March 1988
RADM mechanism. The major difference between this and the March 1988 RADM
mechanism is that the chemical operators "[R02]" and "[RC03]" were added
to represent total amounts of organic peroxy and acylperoxy radicals,
respectively, and instead of reacting with M02 and AC03, the individual
organic peroxy radicals are represented as reacting with these species.
Reactions forming individual peroxy radicals were represented as also
forming [R02]. Reactions forming acetyl peroxy radicals (AC03 or TC03)
were represented as also forming [RC03]. Reactions of [RC03] with N02 and
[R02] and [RC03J with NO, H02, [R02], and [RCQ3] were added. The rate
constants used for the [R02] + [R02], [R02] + [RC03], and [RC03] -» [RC03]
reactions were the same as those used in the SAPRC mechanisms (Carter et
al., 1986; Carter, 1988) the RADM rate constants could not be used
because the RADM reactions they replace have several different rate
constants. The rate constants for the other [R02] and [RC03] reactions
corresponded to those used in the RADM mechanism. The reactions for the
individual peroxy radicals in this mechanism with NO, NOp (where appli-
cable), and H02 are the same as in the RADM mechanism. The reactions of
the peroxy radicals with [R02] and [RC03] are analogous to their reactions
with M02 and AC03, respectively, in the RADM mechanism, except that (1)
261
-------
Table 35. Organic Peroxy + Organic Peroxy Reactions Used in Stockwell's
"Detailed Radical" Version of the RADM Mechanism
Rxn.
Label
Kinetic Parameters (a)
k(300)
Ea
B
Reactions (a,b)
R001 5.81E+02 2.79E+02 -0.44 -1.00
R002 4.28E+02 2.06E+02 -0.44 -1.00
R003 1.25E+02 6.02E+01 -0.44 -1.00
R004 1.04E+02 4.99E+01 -0.44 -1.00
R005 8.86E+01 4.26E+01 -0.44 -1.00
R006 4.28E+02 2.06E+02 -0.44 -1.00
ROOT 4.28E+02 2.06E+02 -0.44 -1.00
R008 5.19E+01 2.50E+01 -0.44 -1.00
R009 5.19E+01 2.50E+01 -0.44 -1.00
R010 5.19E+01 2.50E+01 -0.44 -1.00
R011 5.19E+01 2.50E+01 -0.44 -1.00
R012 5.19E+01 2.50E+01 -0.44 -1.00
R013 5.19E+01 2.50E+01 -0.44 -1.00
R014 5.19E+01 2.50E+01 -0.44 -1.00
R015 2.93E+03 1.41E+03 -0.44 -1.00
R016 2.93E+03 1.41E+03 -0.44 -1.00
R017 7.64E+01 3.67E+01 -0.44 -1.00
R018 4.58E+01 2.20E+01 -0.44 -1.00
R019 3.67E+01 1.76E+01 -0.44 -1.00
M02 + M02 = 11.5 HCHO + H02 +
i.5 ROM
M02 + ETHP =4.75 HCHO + H02 +
1.75 ALD + t.5 ROH
M02 + HC3P = #.75 HCHO + H02 +
1.15 ALD * *.6 KET + #.5 ROH
M02 * HC5P = #.75 HCHO + H02 +
#.105 ALD + #.645 KET * #.5 ROH
M02 + HC8P = #.75 HCHO + H02 +
#.075 ALD + #.675 KET + #.5 ROH
M02 * OL2P = #1.550 HCHO * H02 +
#.350 ALD + #.5 ROH
M02 + OLTP = #1.25 HCHO + H02 +
1.75 ALD + #.5 ROH
M02 « OLIP = #.890 HCHO + H02 *
#.725 ALD + #.3 KET + #.5 ROH
M02 t- KETP = #.75 HCHO + H02 +
#.75 MGLY +1.5 ROH
M02 + TOLP = #.75 HCHO +
#1.405 H02 + #.071 ALD +
#.543 MGLY + #.274 ROH +
#.181 GLY -i- #.905 DCB
M02 + XYLP = #.75 HCHO -»
#1.5 H02 + #.75 MGLY +
#.25 ROH + DCB
M02 + OLN = #1.75 HCHO + #.5 H02 +
ALD + #.25 ROH + N02
M02 + XN02 = #.75 HCHO + #.5 H02 +
#.25 ROH
M02 + X02 = #.75 HCHO + #.5 H02 +
#.25 ROH
M02 f AC03 = HCHO + #.5 H02 +
#.5 ORA2 + #.5 M02
M02 + TC03 = HCHO + #.96 H02 +
#.055 MGLY + #.5 ORA2 +
#.445 GLY + #.025 AC03 + X02 +
#.475 CO
ETHP + ETHP = H02 + #1.5 ALD +
#.5 ROH
ETHP + HC3P = H02 + #.9 ALD H-
#.6 KET + #.5 ROH
ETHP + HC5P = H02 + #.855 ALD *
#.645 KET -i- #.5 ROH
(continued)
262
-------
Table 35 (continued) - 2
Rxn.
Label
Kinetic Parameters (a)
k(300)
Ea B
Reactions (a,b)
R020 3.05E+01 1.47E+01 -0.44 -1.00
R021 1.53E+02 7.34E+01 -0.44 -1.00
R022 1.53E+02 7.34E+01 -0.44 -1.00
R023 1.83E+01 8.81E+00 -0.44 -1.00
R024 1.83E+01 8.81E+00 -0.44 -1.00
R025 1.83E+01 8.81E+00 -0.44 -1.00
R026 1.83E+01 8.81E+00 -0.44 -1.00
R027 1.83E+01 8.81E+00 -0.44 -1.00
R028 1.83E+01 8.81E+00 -0.44 -1.00
R029 1.83E+01 8.81E+00 -0.44 -1.00
R030 1.04E+03 4.99E+02 -0.44 -1.00
R031 1.04E+03 4.99E+02 -0.44 -1.00
R032 7.64E+01 3.67E+01 -0.44 -1.00
R033 1.10E+01 5.28E+00 -0.44 -1.00
R034 9.47E+00 4.55E+00 -0.44 -1.00
R035 4.58E-I-01 2.20E+01 -0.44 -1.00
R036 4.58E+01 2.20E+01 -0.44 -1.00
R037 5.50E+00 2.64E+00 -0.44 -1.00
R038 5.50E+00 2.64E+00 -0.44 -1.00
R039 5.50E+00 2.64E+00 -0.44 -1.00
ETHP + HC8P =
#.675 KET +
ETHP i- OL2P =
11.1 ALD +
ETHP -K OLTP =
#1.5 ALD +
ETHP + OLIP =
#1.475 ALD
ETHP + KETP =
#.75 MGLY +
ETHP + TOLP =
#.821 ALD *
#.274 ROH +
#.905 DCB
ETHP * XYLP =
#.75 ALD +
#.25 ROH +
ETHP + OLN =
#1.75 ALD +
ETHP + XN02 =
#.25 ROH
ETHP + X02 =
#.25 ROH
ETHP + AC03 =
1.5 ORA2
ETHP i- TC03 =
#.055 MGLY
#.445 GLY +
#.475 CO
HC3P + HC3P =
#1.200 KET
HC3P + HC5P =
# 1 . 245 KET
HC3P + HC8P =
#1.275 KET
HC3P + OL2P
#.5 ALD +1
HC3P + OLTP =
#.9 ALD +1
HC3P + OLIP
#.875 ALD +
HC3P + KETP =
#.6 KET + #
H02 + #.825 ALD +
1.5 ROH
1.8 HCHO + H02 +
#.5 ROH
1.5 HCHO + H02 +
1.5 ROH
#.14 HCHO + H02 +
* #.3 KET + #.5 ROH
H02 + #.75 ALD +
#.5 ROH
#1.405 H02 +
#.543 MGLY +
1.181 GLY *
11.5 H02 +
1.75 MGLY +
DCB
HCHO + #.5 H02 +
1.25 ROH + N02
1.5 H02 + #.75 ALD
#.5 H02 + #.75 ALD
#.5 H02 + ALD +
*.5 M02
#.96 H02 + ALD +
+ #.5 ORA2 +
#.025 AC03 + X02 +
H02 * 1.3 ALD +
+ #.5 ROH
H02 1- #.255 ALD +
+ # . 5 ROH
H02 + #.225 ALD +
+ #.5 ROH
= #.8 HCHO + H02 +
1.6 KET +1.5 ROH
= #.5 HCHO + H02 +
.6 KET + #.5 ROH
=1.14 HCHO + H02 +
+ #.9 KET + #.5 ROH
= H02 + 1.15 ALD +
.75 MGLY + #.5 ROH
HC3P + TOLP = #1.405 H02 +
#.221 ALD + #.6 KET +
(continued)
263
-------
Table 35 (continued) - 3
Bxn.
Label
Kinetic Parameters (a)
k(300)
Ea
Reactions (a,b)
R040 5.50E+00 2.64E+00 -0.44 -1.00
R041 5.50E+00 2.64E+00 -0.44 -1.00
R042 5.50E+00 2.64E+00 -0.44 -1.00
R043 5.50E+00 2.64E+00 -0.44 -1.00
R044 3.06E+02 1.47E+02 -0.44 -1.00
R045 3.06E+02 1.47E+02 -0.44 -1.00
R046 7.64E+01 3.67E+01 -0.44 -1.00
R047 7.64E+00 3-67E+00 -0.44 -1.00
R048 3.67E+01 1.76E+01 -0.44 -1.00
R049 3.67E+01 1.76E+01 -0.44 -1.00
R050 4.58E+00 2.20E+00 -0.44 -1.00
R051 4.58E+00 2.20E+00 -0.44 -1.00
R052 4.58E+00 2.20E+00 -0.44 -1.00
R053 4.58E+00 2.20E+00 -0.44 -1.00
R054 4.58E-.-00 2.20E+00 -0.44 -1.00
R055 4.58E+00 2.20E+00 -0.44 -1.00
R056 4.58E+00 2.20E-I-00 -0.44 -1.00
#.543 MGLY -t- #.274 ROH +
#.181 GLY + #.905 DCB
HC3P + XYLP = #1.5 H02 +
#.15 ALD + #.6 KET + #.75 MGLY *
#.25 ROH + DCB
HC3P + OLN = HCHO + #.5 H02 +
#1.150 ALD * #.6 KET +
#.25 ROH + N02
HC3P + XN02 = #.5 H02 + #.15 ALD *
#.6 KET H- #.25 ROH
HC3P + X02 = #.5 H02 * #.15 ALD +
#.6 KET * #.25 ROH
HC3P + AC03 = *.5 H02 + #.2 ALD +
#.8 KET + #.5 ORA2 + #.5 M02
HC3P + TC03 = #.96 H02 * #.2 ALD »
#.8 KET + #.055 MGLY +
#.5 ORA2 + #.445 GLY +
#.025 AC03 + X02 + #.475 CO
HC5P + HC5P = H02 + #.21 ALD +
#1.290 KET -t- #.5 ROH
HC5P * HC8P = H02 + #.180 ALD +
#1.320 KET + #.5 ROH
.8 HCHO + H02 +
.645 KET + #.5 ROH
.5 HCHO * H02 +
.645 KET + #.5 ROH
.14 HCHO + H02 >
.945 KET + #.5 ROH
HC5P + OL2P =
#.455 ALD +
HC5P + OLTP =
#.855 ALD +
HC5P * OLIP =
#.830 ALD
HC5P + KETP = 02+1.105 ALD +
#.645 KET
HC5P + TOLP =
.176 ALD +
.543 MGLY ^
.181 GLY +
HC P + XYLP =
.105 ALD +
.75 MGLY +
.75 MGLY + #.5 ROH
1.405 H02 +
.645 KET +
#.274 ROH +
.905 DCB
1.5 H02 *
.645 KET +
.25 ROH + DCB
HC P + OLN = HCHO + #.5 H02 +
1.105 ALD + #.645 KET +
.25 ROH + N02
HC5P + XN02 = #.5 H02 +
#.105 ALD + #.645 KET + #.25 ROH
HC5P + X02 = #.5 H02 + 1.105 ALD +
#.645 KET + #.25 ROH
(continued)
264
-------
Table 35 (continued) - 4
Rxn.
Label
Kinetic Parameters (a)
k(300)
Ea
B
Reactions (a,b)
R057 2.57E+02 1.23E+02 -0.44 -1.00
R058 2.57E+02 1.23E+02 -0.44 -1.00
R059 7.64E+01 3-67E+01 -0.44 -1.00
R060 3.05E+01 1.47E+01 -0.44 -1.00
R061 3.05E+01 1.47E+01 -0.44 -1.00
R062 3.97E+00 1.91E+00 -0.44 -1.00
R063 3.97E-t-00 1.91E+00 -0.44 -1.00
R064 3.97E+00 1.91E+00 -0.44 -1.00
R065 3.97E+00 1.91E+00 -0.44 -1.00
R066 3.97E-I-00 1.91E+00 -0.44 -1.00
R067 3.97E+00 1.91E+00 -0.44 -1.00
R068 3-97E+00 1.91E+00 -0.44 -1.00
R069 2.20E+02 1.06E+02 -0.44 -1.00
R070 2.20E+02 1.06E+02 -0.44 -1.00
R071 7.64E+01 3.67E+01 -0.44 -1.00
R072 1.53E+02 7.34E+01 -0.44 -1.00
R073 1.83E+01 8.81E+00 -0.44 -1.00
HC5P + AC03 = #.5 H02 + #.14 ALD +
#.860 KET + #.5 ORA2 + #.5 M02
HC5P + TC03 = #.96 H02 +
#.14 ALD + #.860 KET +
#.055 MGLY + #.5 ORA2 +
#.445 GLY + #.025 AC03 + X02 +
#.475 CO
HC8P + HC8P = H02 + #.15 ALD +
#1.350 KET -t- #.5 ROM
HC8P + OL2P = #.8 HCHO + H02 +
#.425 ALD + #.675 KET + #.5 ROH
HC8P + OLTP = #.5 HCHO + H02 +
#,825 ALD i- #.675 KET + #.5 ROH
HC8P + OLIP = #.14 HCHO + H02 +
#.8 ALD + #.975 KET + #.5 ROH
HC8P -t- KETP = H02 + #.075 ALD +
#.675 KET + #.75 MGLY +1.5 ROH
HC8P + TOLP = #1.405 H02 +
#.146 ALD -i- #.675 KET +
#.543 MGLY + #.274 ROH +
1.181 GLY + #.905 DCB
HC8P + XYLP = #1.5 H02 +
#.075 ALD -i- #.675 KET +
#.75 MGLY + #.25 ROH + DCB
HC8P + OLN = HCHO + #.5 H02 +
#1.075 ALD + #.675 KET +
#.25 ROH + N02
HC8P + XN02 = #.5 H02 +
#.075 ALD + #.675 KET + #.25 ROH
HC8P + X02 = #.5 H02 + #.075 ALD +
#.675 KET + #.25 ROH
HC8P + AC03 = #.5 H02 + #.1 ALD +
#.9 KET + #.5 ORA2 + #.5 M02
HC8P + TC03 = #.96 H02 + #.1 ALD +
#.9 KET + #.055 MGLY +
#.5 ORA2 + #.445 GLY +
#.025 AC03 + X02 + #.475 CO
OL2P + OL2P = #1.6 HCHO + H02 +
#.7 ALD + #.5 ROH
OL2P + OLTP = #1.300 HCHO + H02 +
#1.1 ALD + #.5 ROH
OL2P + OLIP = #.940 HCHO + H02 +
#1.075 ALD + #.3 KET -t- #.5 ROH
(continued)
265
-------
Table 35 (continued) - 5
Rxn. Kinetic Parameters (a)
Label Reactions (a,b)
k(300) A Ea B
R074 1.83E+01 8.81E+00 -0.44 -1.00
R075 1.83E+01 8.81E+00 -0.44 -1.00
R076 1.83E+01 8.81E+00 -0.44 -1.00
R077 1.83E+01 8.81E+00 -0.44 -1.00
R078 1.83E+01 8.81E+00 -0.44 -1.00
R079 1.83E+01 8.81E+00 -0.44 -1.00
R080 1.04E+03 4.99E+02 -0.44 -1.00
R081 1.04E+03 4.99E+02 -0.44 -1.00
R082 7.64E+01 3.67E+01 -0.44 -1.00
R083 1.83E+01 8.81E+00 -0.44 -1.00
R084 1.83E+01 8.81E+00 -0.44 -1.00
R085 1.83E+01 8.81E+00 -0.44 -1.00
R086 1.83E+01 8.81E+00 -0.44 -1.00
R087 1.83E+01 8.81E+00 -0.44 -1.00
R088 1.83E+01 8.81E+00 -0.44 -1.00
R089 1.83E+01 8.81E+00 -0.44 -1.00
R090 1.04E+03 4.99E+02 -0.44 -1.00
OL2P + KETP = #.8 HCHO + H02 +
#.350 ALD + 1.75 MGLY + 1.5 ROH
OL2P + TOLP =1.8 HCHO +
11.405 H02 -t- 1.421 ALD +
1.543 MGLY + 1.274 ROH +
#.181 GLY + #.905 DCB
OL2P + XYLP = #.8 HCHO +
11.5 H02 t- #.350 ALD +
#.75 MGLY + #.25 ROH * DCB
OL2P + OLN = #1.800 HCHO +
#.5 H02 + #1.350 ALD +
#.25 ROH * N02
OL2P + XN02 = #.8 HCHO + #.5 H02 *
#.350 ALD * #.25 ROH
OL2P + X02 = #.8 HCHO +1.5 H02 +
#.350 ALD + #.25 ROH
OL2P + AC03 = #.8 HCHO + #.5 H02 +
#.6 ALD + #.5 ORA2 + #.5 M02
OL2P + TC03 = #.8 HCHO +
#.96 H02 + #.6 ALD +
#.055 MGLY +1.5 ORA2 +
#.445 GLY + #.025 AC03 + X02 +
#.475 CO
OLTP + OLTP = HCHO + H02 +
#1.5 ALD + #.5 ROH
OLTP + OLIP = #.640 HCHO + H02 +
#1.475 ALD + #.3 KET + #.5 ROH
OLTP + KETP = #.5 HCHO + H02 +
#.75 ALD + #.75 MGLY + #.5 ROH
OLTP + TOLP =1.5 HCHO +
#1.405 H02 + #.821 ALD +
#.543 MGLY + #.274 ROH +
1.181 GLY + #.905 DCB
OLTP + XYLP =1.5 HCHO +
#1.5 H02 + #.75 ALD +
#.75 MGLY + #.25 ROH + DCB
OLTP + OLN = #1.5 HCHO + #.5 H02 +
#1.75 ALD + 1.25 ROH + N02
OLTP + XN02 = #.5 HCHO + #.5 H02 +
1.75 ALD + #.25 ROH
OLTP + X02 = #.5 HCHO +4.5 H02 +
#.75 ALD + #.25 ROH
OLTP + AC03 = #.5 HCHO + #.5 H02 +
ALD + #.5 ORA2 + #.5 M02
(continued)
266
-------
Table 35 (continued) - 6
Rxn.
Label
Kinetic Parameters (a)
k(300)
Ea
Reactions (a,b)
R091 1.04E+03 4.99E+02 -0.44 -1.00
R092 1.10E+00 5.28E-01 -0.44 -1.00
R093 2.23E+00 1.07E+00 -0.44 -1.00
R094 2.23E+00 1.07E+00 -0.44 -1.00
R095 2.23E+00 1.07E+00 -0.44 -1.00
R096 2.23E+00 1.07E+00 -0.44 -1.00
R097 2.23E+00 1.07E+00 -0.44 -1.00
R098 2.23E+00 1.07E+00 -0.44 -1.00
R099 1.28E+02 6.17E+01 -0.44 -1.00
R100 1.28E+02 6.17E+01 -0.44 -1.00
R101 1.10E+00 5.28E-01 -0.44 -1.00
R102 2.23E+00 1.07E+00 -0.44 -1.00
R103 2.23E+00 1.07E+00 -0.44 -1.00
R104 2.23E+00 1.07E+00 -0.44 -1.00
OLTP t- TC03 = #.5 HCHO +
#.96 H02 + ALD + 1.055 MGLY +
#.5 ORA2 + #.445 GLY +
#.025 AC03 + X02 + #.475 CO
OLIP + OLIP = #.280 HCHO + H02 +
#1.450 ALD * #.6 KET + #.5 ROH
OLIP -i- KETP = #.14 HCHO + H02 +
#.725 ALD + #.3 KET +
#.75 MGLY * #.5 ROH
OLIP + TOLP = //.14 HCHO +
#1.405 H02 + #.796 ALD +
#.3 KET + #.543 MGLY +
#:274 ROH + #.181 GLY +
#.905 DCB
OLIP f XYLP = #.14 HCHO +
#1.5 H02 + #.725 ALD + #.3 KET +
#.75 MGLY -f #.25 ROH + DCB
OLIP + OLN = #1.140 HCHO +
#.5 H02 + #1.725 ALD + #.3 KET +
#.25 ROH + N02
OLIP + XN02 = #.14 HCHO +
#.5 H02 * #.725 ALD + #.3 KET +
#.25 ROH
OLIP + X02 = #.14 HCHO + #.5 H02 +
#.725 ALD + #.3 KET + #.25 ROH
OLIP > AC03 = #.14 HCHO +
#.5 H02 + #.725 ALD + #.55 KET +
#.5 ORA2 + #.5 M02
OLIP + TC03 = #.14 HCHO +
#.96 H02 + #.725 ALD +
#.55 KET-+ #.055 MGLY +
#.5 ORA2 + #.445 GLY +
#.025 AC03 + X02 + #.475 CO
KETP + KETP = H02 + #1.5 MGLY +
#.5 ROH
KETP + TOLP = #1.405 H02 +
#.071 ALD + #1.293 MGLY +
#.274 ROH + 1.181 GLY +
#.905 DCB
KETP + XYLP = 11.5 H02 +
#1.5 MGLY + #.25 ROH + DCB
KETP + OLN = HCHO + #.5 H02 *
ALD + #.75 MGLY + #.25 ROH + N02
(continued)
267
-------
Table 35 (continued) - 7
Rxn.
Label
Kinetic Parameters (a)
k(300)
Ea B
Reactions (a,b)
R105 2.23E+00 1.07E+00 -0.44 -1.00
R106 2.23E+00 1.07E+00 -0.44 -1.00
R107 1.28E+02 6.17E+01 -0.44 -1.00
R108 1.28E+02 6.17E+01 -0.44 -1.00
R109 1.10E+00 5.28E-01 -0.44 -1.00
R110 2.23E+00 1.07E+00 -0.44 -1.00
R111 2.23E+00 1.07E+00 -0.44 -1.00
R112 2.23E+00 1.07E+00 -0.44 -1.00
R113 2.23E+00 1.07E+00 -0.44 -1.00
R114 1.28E+02 6.17E+01 -0.44 -1.00
R115 1.28E+02 6.17E+01 -0.44 -1.00
R116 1.10E+00 5.28E-01 -0.44 -1.00
R117 2.23E+00 1.07E-I-00 -0.44 -1.00
KETP + XN02 = #.5 H02 +
#.75 MGLY + #.25 ROM
KETP + X02 = 1.5 H02 + #.75 MGLY +
#.25 ROH
KETP + AC03 = #.5 H02 * MGLY +
#.5 ORA2 * #.5 M02
KETP + TC03 = #.96 H02 +
#1.055 MGLY +1.5 ORA2 +
#.445 GLY + #.025 AC03 + X02 +
#.475 CO
TOLP + TOLP = #1.810 H02 +
#.142 ALD + #1.448 MGLY +
#.047 ROH + #.362 GLY +
#1.810 DCB
TOLP + XYLP = #1.905 H02 +
#.071 ALD + #1.655 MGLY +
#.024 ROH + 1.181 GLY +
#1.905 DCB
TOLP + OLN = HCHO + #.905 H02 +
#1.071 ALD + #.905 MGLY +
#.024 ROH + 1.181 GLY +
#.905 DCB + N02
TOLP + XN02 = #.905 H02 +
#.071 ALD + #.905 MGLY +
#.024 ROH + 1.181 GLY +
#.905 DCB
TOLP + X02 = #.905 H02 +
#.071 ALD + #.905 MGLY +
#.024 ROH + #.181 GLY +
#.905 DCB
TOLP + AC03 = #.905 H02 +
#.095 ALD + #.724 MGLY +
#.5 ORA2 + #.181 GLY +
#.905 DCB + #.5 M02
TOLP + TC03 = #1.365 H02 +
#.095 ALD + #.779 MGLY +
#.5 ORA2 + #.626 GLY +
#.905 DCB + #.025 AC03 + X02 +
#.475 CO
XYLP + XYLP = #2 H02 + 12 MGLY +
#2 DCB
XYLP + OLN = HCHO + H02 + ALD +
#1.25 MGLY + DCB + N02
(continued)
268
-------
Table 35 (continued) - 8
Rxn.
Label
Kinetic Parameters (a)
k(300)
Ea
B
Reactions (a,b)
R118 2.23E+00 1.07E+00 -0.44 -1.00
R119 2.23E+00 1.07E+00 -0.44 -1.00
R120 1.28E+02 6.17E+01 -0.44 -1.00
R121 1.28E+02 6.17E+01 -0.44 -1.00
R122 1.10E+00 5.28E-01 -0.44 -1.00
R123 2.23E+00 1.07E+00 -0.44 -1.00
R124 2.23E+00 1.07E+00 -0.44 -1.00
R125 1.28E+02 6.17E+01 -0.44 -1.00
R126 1.28E+02 6.17E+01 -0.44 -1.00
R127 1.10E+00 5.28E-01 -0.44 -1.00
R128 2.23E+00 1.07E+00 -0.44 -1.00
R129 1.28E+02 6.17E+01 -0.44 -1.00
R130 1.28E+02 6.17E+01 -0.44 -1.00
R131 1.10E+00 5.28E-01 -0.44 -1.00
R132 1.28E+02 6.17E+01 -0.44 -1.00
R133 1.28E+02 6.17E+01 -0.44 -1.00
R134 3.67E+03 1.76E+03 -0.44 -1.00
R135 7.33E+03 3.52E+03 -0.44 -1.00
R136 3.67E+03 1.76E+03 -0.44 -1.00
> #1.25 MGLY
#1.25 MGLY H
» MGLY +
it.5 M02
#.025 AC03
ALD
XYLP + XN02 = H02
DCB
XYLP + X02 = H02 H
DCB
XYLP + AC03 = H02
1.5 ORA2 + DCB 1
XYLP -- TC03 = #1.46 H02
#1.055 MGLY + #.5 ORA2
#.445 GLY + DCB
X02 + #.475 CO
OLN + OLN = #2 HCHO
#2 N02
OLN + XN02 = HCHO
OLN * X02 = HCHO
OLN + AC03 = HCHO
#.5 ORA2 -i- N02 1
OLN + TC03 = HCHO
ALD + #.055 MGLY + #.5 ORA2
#.445 GLY > #.025 AC03 + X02
#.475 CO + N02
XN02 + XN02 =
XN02 + X02 =
XN02 1- AC03 = #.5 ORA2
XN02 + TC03 = #.46 H02
N02
N02
ALD
ALD
ALD +
#.5 M02
#.46 H02
#.5 M02
#,
#
055 MGLY
445 GLY H
> #.5 ORA2 H
1.025 AC03
#.475 CO
X02
X02
X02
#
#
#
+ X02 =
+ AC03 = #.5
+ TC03 = #,
,055 MGLY +
.445 GLY
X02
M02
ORA2 + #.5
46 H02 +
#.5 ORA2 +
#.025 AC03 + X02 4
475 CO
AC03 + AC03 = #2 M02
AC03 + TC03 = #.920 H02 +
#.110 MGLY + #.890 GLY +
#.05 AC03 + #2 X02 + #.950 CO
TC03 + TC03 = #1.84 H02 +
#.220 MGLY + #1.780 GLY +
#.1 AC03 + #4 X02 + #1.900 CO
M02
(a) See Footnotes (b) and (c) in Table 2 for a description of the formats
used in the kinetic parameter and the reaction listings.
(b) Note that ROH, a species not in the standard RADM mechanism, is shown
as a product in some of these reactions. However, no reactions for
it are included in this mechanism, so its formation has no effect.
269
-------
Table 36. Listing of the Organic Reactions in the Modified RADM
Mechanism Incorporating the 1986 SAPRC Approach for
Representing Reactions of Organic Peroxy Radicals
Rxn.
Label
(a)
010
011
012
013
014
015
016
017
018
019
020
021
Kinetic Parameters (b)
k(300)
(Phot.
(Phot.
(Phot.
(Phot,
( Phot ,
(Phot,
(Phot,
(Phot
(Phot
(Phot
(Phot
(Phot
A
Ea
B
, Set = HCHOMR )
, Set = HCHORR )
. Set = ALDR )
, Set = OPR )
. Set = OPR )
. Set = PAAR )
. Set = KETR )
. Set = GLY2R )
. Set = GLY1R )
. Set = MGLYR )
. Set = DCBR )
. Set = ONITR )
Reactions (b)
Organic Photolysis Reactions
HCHO * HV = H2 * CO
HCHO + HV = H02 + H02 + CO
ALD + HV = M02 + H02 + CO + [R02]
OP1 -i- HV = HCHO + H02 + HO
OP2 f HV = ALD + H02 + HO
PAA * HV = M02 + C02 -e HO + [R02]
KET * HV = AC03 + [RC03] + ETHP +
[R02]
GLY + HV = 1.13 HCHO + #1.87 CO
GLY + HV = #.45 HCHO + #1.55 CO +
#,80 H02
MGLY * HV = AC03 + [RC03] f H02 +
CO
DCB + HV = #.98 H02 * #.02 AC03 +
TC03 f #1.02 [RC03]
ONIT + HV = #.20 ALD + #.80 KET +
H02 + N02
Organic + OH Reactions
051
052
053
054
055
056
057
058
074
059
060
063
064
065
1
4
3
7
1
1
3
9
1
9
4
1
2
1
.29E+01
. 11E+02
.76E+03
.86E+03
.66E+04
.24E+04
.82E+04
.24E+04
.46E+05
.02E+03
.54E+04
.32E+04
.37E+04
.47E+03
9.23E+02
1.81E+03
2 . 28E+04
2.79E+04
5.87E+04
3.16E+03
7.12E+03
1.48E+04
3.74E+04
3.08E+03
3.08E+04
1.32E+04
1.01E+04
1.76E+04
2.54
0.88
1.07
0.75
0.75
-0.82
-1.00
-1.09
-0.81
-0.64
-0.23
0.00
-0.51
1.48
1
1
-1
-1
-1
-1
-1
-1
-1
-1
-1
-1
-1
-1
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
CH4
ETH
HC3
#
#
HC5
#
HC8
#
OL2
OLT
OLI
ISO
TOL
#
XYL
#
+ HO = M02 + H20 +
* HO = ETHP
* HO = #.90
+ [R02]
"HC3P *
.10 H02 + #.014 HCHO
.069 ALD + #
+ HO = HC5P
1.27 [R02]
+ HO = HC8P
1.78 [R02]
+ HO = OL2P
+ HO = OLTP
+ HO = OLIP
+ HO = OLTP
+ HO = #.84
. 16 CSL + #.
+ HO = #.83
.17 CSL + #.
HCHO + HO = H02
ALD
KET
+ HO = AC03
* HO = KETP
.026 KET
[R02]
+ H20
[R02]"
+
+ H20
f
+ #.27 X02 + H20 -i
+ #.78
+ [R02]
+ [R02]
+ [R02]
* [R02]
"TOLP +
16 H02
"XYLP +
17 H02
+ CO +
X02 -t- H20 -.
[RQ2]"
[R02]"
H20
+
+
+ [RC03] + H20
+ [R02]
+ H20
(continued)
270
-------
Table 36 (continued) - 2
Rxn.
Label
(a)
066
067
068
061
Kinetic Parameters (b)
k(300)
1.
2.
4.
5.
.69E+04
.50E+04
, 1 1 E+04
.87E+04
1,
2
4
5
A
.69E+04
.50E+04
. 11 E+04
.87E+04
Ea
0.
0.
0.
0.
00
00
00
00
B
-1
-1
-1
-1
.00
.00
.00
.00
Reactions (b)
GLY + HO = H02 + #2 CO +
MGLY + HO = AC03 + [RC03]
H20
DCB + HO = TC03 + [RC03]
CSL + HO = #.1 H02 + #.9
H20
+
CO +
+ H20
"X02 +
[R02]" + #.9 "TC03 + [RC03
062
069
070
071
072
073
093
094
095
096
097
098
099
100
101
102
103
104
105
5.
1,
1.
1.
2
3
9
3
9
3
3
3
1
2
1
8
2
1
3
.28E+04
.47E+04
.47E+04
.47E+04
.06E+02
.76E+03
.23E-01
. 65E+00
.23E-01
. 65E+00
. 65E+00
.23E+04
.72E-01
.33E+01
.84E+03
.53E+02
.72E-03
.74E-02
.02E-01
5
1
1
1
9
2
8
2
8
2
2
3
2
1
4
8
1
1
1
.28E+04
.47E+04
.47E+04
.47E+04
.06E+02
.28E+04
.81E+02
.06E+03
.81E+02
.06E+03
.06E+03
.23E+04
.94E+03
.47E+04
0.
0.
0.
0.
0.
1.
4.
3.
4.
3.
3.
0.
00
00
00
00
88
07
09
78
09
78
78
00
5.81
3.
84
.74E+04 1.94
.53E+02
.76E+01
.94E+01
.33E+01
0.
,00
5.23
4,
.18
2.26
-1
-1
-1
-1
1
-1
-1
-1
-1
-1
-1
-1
-1
-1
-1
-1
-1
-1
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
-1.00
CSL + HO = CSL
1"
OP1 + HO = #.5 "M02 + [R02]" +
#.5 HCHO + #.5 HO
OP2 + HO = #.5 "HC3P + [R02]
#.5 ALD + #.5 HO
PAA + HO = AC03 + [RC03]
" +
+ H20
PAN + HO = HCHO + N03 + X02
[R02]
ONIT + HO = HC3P + [R02]
Organic + N03 Reactions
HCHO + N03 = H02 + HN03 n
+
+ N02
H CO
ALD + N03 = AC03 + [RC03] +
GLY + N03 = HN03 + H02 +
MGLY + N03 = HN03 + AC03
[RC03] + CO
DCB + N03 = HN03 + TC03 -
#2
+
HN03
CO
H [RC03]
CSL + N03 = HN03 + XN02 + #.
OL2 + N03 = OLN + [R02]
OLT + N03 = OLN + [R02]
OLI + N03 = OLN + [R02]
ISO + N03 = OLN + [R02]
Organic + 0-3 Reactions
.5 CSL
OL2 + 03 = HCHO + #.42 CO +
#.4 ORA1 + #.12 H02
OLT + 03 = #.53 HCHO + #
#.33 CO + #.20 ORA1 +
#.20 ORA2 + #.23 H02 +
#.22 "M02 + [R02]" + #
#.06 CH4
OLI + 03 = #.18 HCHO + #
#.10 KET + #.23 CO + #
#.29 ORA2 + #.09 CH4 +
#.26 H02 + #.14 HO + #
[R02]"
.50
.10
.72
.06
.31
ALD +
HO +
ALD +
ORA1 -
"M02
(continued)
271
-------
Table 36 (continued) - 3
Rxn. Kinetic Parameters (b)
(a) k(300) A Ea B
Reactions (b)
106 2.20E-02 1.81E1-01 4.00 -1.00
M01 7.52E+03 4.11E+03 -0.36 -1.00
075 (Same k as M01 )
077 (Same k as M01 )
076 2.90E-02 1.17E+18 26.91 0.00
078 2.90E-02 1.17E+18 26.91 0.00
M02 1.12E+04 6.17E+03 -0.36 -1.00
079 (Same k as M02 )
080 (Same k as M02 )
081 (Same k as M02 )
082 (Same k as M02 )
083 (Same k as M02 )
084 (Same k as H02 )
085 (Same k as M02 )
088 (Same k as M02 )
089 (Same k as M02 )
090 (Same k as M02 )
091 (Same k as M02 )
092 (Same k as M02 )
150 (Same k as M02 )
M03 (Same k as M02 )
086 (Same k as M02 )
087 (Same k as M02 )
ISO + 03 = #.53 HCHO + 1.50 ALD
#.33 CO f #.20 ORA1 +
#.20 ORA2 + #.23 H02 +
1.22 "M02 + [R02]" + if. 10 HO
Organic Peroxy + N02 Reactions
[RC03] * N02 = N02
AC03 + N02 = PAN
TC03 + N02 = TPAN
PAN = AC03 + [RC03]
TPAN = TC03 f [RC03]
N02
N02
Organic Peroxy + NO Reactions
[R02] + NO = NO
M02 + NO = HCHO + H02 + N02
HC3P * NO = 1.69 ALD + 1.26 KET
1.14 HCHO + #.03 ONIT +
*.97 N02 + 1.97 H02
HC5P + NO = #.36 ALD
1.10 ONIT
1.76 KET *
#.90 N02 + #.90 H02
HC8P + NO = #.M4 ALD + #1.05 KET +
#.05 HCHO + #.24 ONIT +
#.76 N02 + 1.76 H02
OL2P + NO = #1.6 HCHO + H02 +
N02 + #.2 ALD
OLTP + NO = ALD + HCHO * H02 + N02
OLIP + NO = H02 + #1.45 ALD +
#.28 HCHO + #.10 KET + N02
TOLP * NO = N02 -K H02 +
#.724 MGLY + #.181 GLY +
#.905 DCB
XYLP + NO = N02 + H02 + MGLY + DCB
ETHP + NO = ALD + H02 + N02
KETP + NO = HGLY + N02 + H02
OLN + NO = HCHO + ALD + #2 N02
X02 + NO = N02
[RC033 + NO = NO
AC03 + NO = M02 + N02 + [R02]
TC03 + NO = N02 + #.92 H02 +
#.89 GLY + 1.11 MGLY +
(continued)
272
-------
Table 36 (continued) - 4
Rxn.
Label
(a) k(300)
Kinetic Parameters (b)
A Ea B
Reactions (b)
MOM
107
108
109
110
111
112
113
IIM
115
117
118
120
1M6
M05
116
119
M06
121
122
123
125
126
127
128
129
#.05 "AC03 + [RC031" + #.95 CO
#2 "X02 + [R02]"
Organic Peroxy + H02 Reactions
3.61E+03 1.13E+02 -2.58 -1.00
(Same k as MOM )
(Same k as MOM )
(Same k as MOM )
(Same k as MOM )
(Same k as MOM )
(Same k as MOM )
(Same k as MOM )
(Same k as MOM )
(Same k as MOM )
(Same k as MOM )
(Same k as MOM )
(Same k as MOM )
(Same k as MOM )
(Same k as MOM )
(Same k as MOM )
(Same k as MOM )
1.47E+00 1.47E+00 0.00 -1.00
(Same k as M06 )
(Same k as M06 )
(Same k as M06 )
(Same k as M06 )
(Same k as M06 )
(Same k as M06 )
(Same k as M06 )
(Same k as M06 )
(Same k as M06 )
H02 + [R02] = H02
H02 + M02 = OP1
H02 + ETHP = OP2
H02 + HC3P = OP2
H02 + HC5P = OP2
H02 + HC8P = OP2
H02 + OL2P = OP2
H02 ,+ OLTP = OP 2
H02 + OLIP = OP2
H02 + KETP = OP2
H02 + TOLP = OP2
H02 + XYLP = OP2
H02 + OLN = ON IT
X02 + H02 = OP2
H02 + [RC03] = H02
H02 + AC03 = PAA
H02 + TC03 = OP2
Organic Peroxy + R02 Reactions
[R02] + [R02] =
M02 + [R02] = [R02] + tf.75 HCHO +
#.5 H02
ETHP + [R02] = [R02] + #.75 ALD +
#.5 H02
HC3P + [R02] = [R02] + #.15 ALD +
#.6 KET + #.5 H02
HC5P + [R02] = [R02] + #.105 ALD +
#.645 KET + #.5 H02
HC8P + [R02] = [R02] + #.075 ALD +
#.675 KET -i- #.5 H02
OL2P + [R02] = [R02] + #.8 HCHO +
#.35 ALD + 1.5 H02
OLTP + [R021 = [R02] + #.5 HCHO +
#.75 ALD -i-f.5 H02
OLIP + [R02] = [R02] +1.14 HCHO +
#.725 ALD + #.55 KET + #.5 H02
KETP + [R02] = [R02] + #.75 MGLY +
#.5 H02
(continued)
273
-------
Table 36 (continued) - 5
Rxn.
Label
(a) k(300)
Kinetic Parameters (b)
A Ea B
131
132
M07
(Same
(Same
(Same
k
k
k
as
as
as
M06
M06
M06
)
)
)
TOLP
#1
#.
XYLP
#1
OLN
+ [R02]
.4 H02 +
181 GLY
+ [R02]
.5 H02 +
+ [R02]
= [R02
#.724
+ #.905
= [R02
MGLY +
= [R02]
] + #
MGLY
DCB
] + #
DCB
.23 HCHO
+
.25 HCHO
* HCHO + ALD
+
+
N02
147
(Same
k
as
M06
)
X02
+ [R02]
= [R02]
M08
M09
135
136
137
138
139
140
141
143
144
M10
148
130
2.93E+03 1.41E+03 -0.44 -1.00
(Same k as M08 )
(Same k as M08 )
(Same k as M08 )
(Same k as M08 )
(Same k as M08 )
(Same k as M08 )
(Same k as M08 )
(Same k as M08 )
(Same k as M08 )
(Same k as M08 )
(Same k as M08 )
(Same k as M08 )
(Same k as M08 )
(Same k as M08 )
Organic Peroxy + RC03 Reactions
[RC03] + [R02] =
M02 + [RC03] = [RC031 *
#.75 HCHO + #.5 H02
ETHP * [RC03] = [RC03] +
#.75 ALD + #.5 H02
HC3P + [RC03J = [RC03] +
#.15 ALD + #.6 KET + #.5 H02
HC5P + [RC03] = [RC03] +
#.105 ALD + #.645 KET + #.5 H02
HC8P + [RC03] = [RC031 +
#.075 ALD + #.675 KET + #.5 H02
OL2P + [RC03] = [RC03] +
#.8 HCHO * #.35 ALD + #.5 H02
OLTP + [RC031 = [RC03] «
#.5 HCHO + #.75 ALD + #.5 H02
OLIP + [RC03] = [RC03] +
#.14 HCHO + #.725 ALD +
#.55 KET + #.5 H02
KETP + [RC03] = [RC03] +
#.75 MGLY + #.5 H02
TOLP + [RC03] = [RC03] +
#.23 HCHO + #1.4 H02 +
#.724 MGLY + 1.181 GLY +
#.905 DCB
XYLP + [RC03] = [RC03] +
#.25 HCHO + #1.5 H02 + MGLY +
DCB
OLN + [RC03] = [RC03] + HCHO +
ALD 1- N02
X02 + [RC03] = [RC03]
AC03 + [R02] = #1.5 [R02] +
#.5 M02 + #.5 ORA2
(continued)
274
-------
Table 36 (continued) - 6
Rxn.
Label
(a) k(300)
Kinetic Parameters (b)
A Ea B
Reactions (b)
133
M11
142
145
151
152
153
154
155
(Same k as M08 )
3.64E+03 1.75E+03 -0.44 -1.00
(Same k as M11 )
(Same k as M11 )
1.12E+04 6.17E+03 -0.36 -1.00
(Same k as M04 )
5.19E+01 2.50E+01 -0.44 -1.00
1.28E-t-02 6.17E+01 -0.44 -1.00
1.10E+00 5.28E-01 -0.44 -1.00
TC03 + [R02] = #2 [R02] -t-
4.5 ORA2 + #.46 H02 +
1.445 GLY + #.055 MGLY +
1.025 "AC03 + [RC03]" +
#.475 CO + X02
[RC03] + [RC03] =
AC03 f [RC031 = [RC03] + M02 +
[R02]
TC03 + [RC03] = [RC03] +
1.92 H02 + 1.890 GLY +
1.11 MGLY + 1.05 "AC03 +
[RC03]" + #.950 CO + #2 "X02
[R02]"
XN02 Reactions
XN02 * N02 = ONIT
XN02 + H02 = OP2
XN02 + [R02] = [R02]
XN02 + [RC03] = [RC03]
XN02 + XN02 =
(a) Reactions which correspond directly with those in the standard RADM
mechanism are given the same reaction labels as used for them in
Table 2. Reactions which are added or significantly modified
relative to that mechanism are indicated by the label "Mxx."
(b) See Footnotes (b) and (c) in Table 2 for a description of the formats
used in the kinetic parameter and the reaction listings.
275
-------
the products from M02 and AC03 are not included, (2) [R02] or [RC03] was
included as a product as well as a reactant so the reactions with the
individual radicals would be not counted as contributing to their consump-
tion rate (since this is already accounted for in their reactions with
[R02] or [RC03]), and (3) the rate constants for the corresponding reac-
tion of [R02] or [RC03] are employed. Thus, other than the approximation
approach used in lumping the organic peroxy + organic peroxy reactions,
this mechanism was made to represent, as closely as possible, the chemical
approaches used in the March 1988 and the detailed radical RADM mechan-
isms. Therefore, any differences found between the predictions using this
and the other mechanisms should be due primarily to differences in the
approximation approach.
Table 37 gives the results of the comparisons of the calculations
using the March 1988 RADM mechanism (the "test set") relative to the
calculations using the detailed radical mechanism (the "standard set") for
the species of known or potential interest in RADM applications. The
table gives the summary statistics for all the rural and the urban
scenarios aggregated together, and the results for the individual calcula-
tions where the "% Fit" or some other percent change statistics is greater
than 5%. (Thus, if the results for an individual test problem are not
shown, then the two mechanisms gave predictions to within 5J.) With very-
few exceptions, the two mechanisms give almost exactly the same predic-
tions. Even in the cases where they are the most different, the two
mechanisms are generally within 5% of each other in predictions of maximum
and average concentrations of (X, H202, SULF, HNO^, ORA1, ORA2, OP1, PAA,
and PAN, and within 10% in predictions of OP2, excluding some cases where
insignificantly low concentrations are predicted.
These results indicate that the approximation used in the evaluated
mechanism, that all the organic peroxy + R02 reactions can be represented
by reactions only with M02, and that all the organic peroxy -^RCO^ reac-
tions can be represented only by reactions with AC03, is justified. This
is consistent with the results of previous unpublished tests carried out
by Stockwell (private communication).
276
-------
Table 37. Results of Sensitivity Test Calculations Using the Evaluated
RADM Mechanism as the Test Mechanism, Relative to the
"Detailed Radical" Mechanism as the Standard
Test Problem (a)
Env Mix NOX C/N
Calculated Concentration (ppm) (b)
Maximum Average
Standard % Chg Standard % Chg % /Max % Fit
Rural O^
Maximum (c) 6.50E-02
Average (d)
Avg. Abs. Value (e)
Urban Oo
Maximum (c) 3.47E-01
Average (d)
Avg. Abs. Value
Rural H202
Maximum (c) 2.41E-03
Average (d)
Avg. Abs. Value (e)
Urban H202
Maximum (c) 1.48E-02
Average (d)
Avg. Abs. Value (e)
SA AC 67 3 2.58E-07
SA DC 67 3 2.73E-07
WS BT 67 3 2.05E-08
Rural SULF
Maximum (c) 2.20E-03
Average (d)
Avg. Abs. Value (e)
Urban SULF
Maximum (c) 6.25E-03
Average (d)
Avg. Abs. Value (e)
0.8
0.0
0.1
-1.6
-0.2
0.2
-2.2
0.1
0.7
-16.5
-1.4
1.5
-14.9
-5.3
-16.5
-1.4
0.0
0.5
-1.9
-0.3
0.4
4.96E-02 0.6
0.1
0.1
2.22E-01 -1.0
-0.1
0.2
1.85E-03 3.2
0.2
0.5
1.20E-02 -13.3
-1.2
1.4
1.06E-07 -13.3
1.15E-07 -11.2
6.64E-09 -5.1
1.53E-03 -1.3
0.1
0.4
4.26E-03 -1-6
-0.2
0.3
0.3
0.0
0.0
-0.6
0.0
0.1
-1.5
0.0
0.1
-2.3
-0.2
0.2
0.0
0.0
0.0
0.6
0.0
0.0
-0.3
0.0
0.0
0.6
0.1
1.1
0.1
3.1
0.3
14.3
1.0
14.3
12.0
12.3
1.3
0.2
1.6
0.2
(continued)
277
-------
Table 37 (continued) - 2
Test Problem (a)
Env Mix NOV C/N
A
Rural HN03
Maximum (c)
Average (d)
Avg. Abs. Value
Urban HN03
Maximum (c)
Average (d)
Avg. Abs. Value
Rural ORA1
Maximum (c)
Average (d)
Avg. Abs. Value
SA AL -2 6
Urban ORA1
Maximum (c)
Average (d)
Avg. Abs. Value
Rural ORA2
Maximum (c)
Average (d)
Avg. Abs. Value
WS AL -2 20
SA AL -2 60
Urban ORA2
Maximum (c)
Average (d)
Avg. Abs. Value
SS DC 300 10
Maxim
Standard
1.96E-03
(e)
8.44E-02
(e)
1.85E-04
(e)
9.81E-08
4.52E-03
(e)
2.68E-04
(e)
2.55E-06
2.12E-05
6.46E-03
(e)
5.87E-03
alculate
urn ---
I Chg
3.1
0.2
0.8
-4.6
-0.5
0.6
5.1
0.1
0.7
5.1
0.6
0.1
0.2
-4.5
-0.7
1.2
0.1
-4.5
-5.3
-1.6
1.6
-5.3
id Concentration (ppm) (b)
Standard % Chg
9.86E-04 2.1
0.2
0.6
3.36E-02 -2.9
-0.5
0.5
1.38E-04 4.7
0.1
0.7
7.85E-08 4.7
3.00E-03 0.5
0.1
0.1
2.10E-04 -8.2
-0.9
1.2
2.12E-06 -8.2
1.02E-05 -5.6
3.52E-03 -3.9
-1.3
1.3
3.16E-03 -3.9
% /Max
0.4
0.0
0.0
-0.6
0.0
0.1
-0.7
0.0
0.0
0.0
0.2
0.0
0.0
-0.6
0.0
0.0
-0.1
-0.3
-3.5
-0.3
0.3
-3.5
% Fit
2.2
0.3
2.9
0.4
4.6
0.3
4.6
0.5
0.1
8.7
0.6
8.7
5.8
4.0
0.8
4.0
(continued)
278
-------
Table 37 (continued) - 3
Test Problem (a)
Env Mix NOX C/N
Calculated Concentration (ppm) (b)
Maximum Average
Standard % Chg Standard % Chg % /Max % Fit
Rural OP1
Maximum (c)
Average (d)
Avg. Abs. Value
Urban OP1
Maximum (c)
Average (d)
Avg. Abs. Value
SA AC 67
SA DC 67
WS BT 67
Rural OP2
Maximum (c)
Average (d)
3
3
3
Avg. Abs. Value
WS AL -2
SA AL -2
WS AI -2
Urban OP2
Maximum (c)
Average (d)
6
6
6
Avg. Abs. Value
SS AC 67
SS AC 300
WS AC 67
WS AC 67
WS AC 67
SA AC 67
SS DC 67
SS DC 300
WS DC 67
WS DC 67
WS DC 67
SA DC 67
SS BT 67
SS BT 300
10
10
3
10
30
3
10
10
3
10
30
3
10
10
1.72E-03
(e)
1.38E-02
(e)
7.44E-07
5.29E-07
6.42E-07
1.59E-03
(e)
6.40E-06
5.32E-06
4.72E-06
3.42E-02
(e)
3.66E-03
1.39E-02
4.29E-07
1.74E-03
7.97E-03
7.20E-07
4.28E-03
1.62E-02
4.36E-07
2.07E-03
9.08E-03
6.50E-07
2.31E-03
7.86E-03
-3.1
-0.2
0.8
73.4
-0.6
3.6
-5.4
73.4
-54.1
-8.4
0.3
1.4
5.9
5.5
-8.4
2246.7
56.8
61.9
-9-9
-12.8
63.6
-5.4
-7.3
83.7
-11.1
-13-1
51.2
-6.7
-9.1
173.7
-9.3
-13.6
1.25E-03
8.38E-03
3.58E-07
2.55E-07
1.20E-07
1.28E-03
3.78E-06
2.12E-06
3.44E-06
2.63E-02
1.65E-03
6.38E-03
9.43E-08
1.03E-03
5.63E-03
2.41E-07
1.88E-03
7.33E-03
8.53E-08
1.24E-03
6.40E-03
2.34E-07
1.29E-03
4.38E-03
-3.3
-0.1
0.6
57.3
-0.6
2.8
-6.3
57.3
-33.1
-4.1
0.2
1.1
2.2
3.1
-4.1
1623.2
39.8
42.6
-3.6
-5.3
54.2
-4.5
-6.0
101.7
-4.4
-5.5
50.3
-5.7
-7.8
149.8
-2.9
-4.6
-0.4
0.0
0.0
-1.4
-0.2
0.2
0.0
0.0
0.0
1.1
0.0
0.0
0.0
0.0
0.0
-1.9
-0.2
0.2
-0.2
-1.3
0.0
-0.2
-1.3
0.0
-0.3
-1.5
0.0
-0.3
-1.9
0.0
-0.1
-0.8
3.3
0.3
58.3
1.9
6.5
44.5
58.3
9.3
0.7
2.3
3.4
9.3
178.7
6.9
3.7
5.5
44.9
4.7
6.2
67.2
4.5
5.7
40.9
5.9
8.1
86.8
3.0
4.7
(continued)
279
-------
Table 37 (continued) - 4
Test Problem (a)
Env Mix NOV C/N
A
WS BT 67 3
SA BT 67 3
Rural PAA
Maximum (c)
Average (d)
Avg. Abs. Value
Urban PAA
Maximum (c)
Average (d)
Avg. Abs. Value
Rural PAN
Maximum (c)
Average (d)
Avg. Abs. Value
Urban PAN
Maximum (c)
Average ( d )
Avg. Abs. Value
Maximum
Standard % Chg
1.10E-07 2246.7
7.64E-08 582.8
1.87E-03 -2.9
0.2
(e) 0.9
1.05E-02 3.4
0.5
(e) 0.9
1.87E-03 2.1
0.0
(e) 0.4
4.10E-02 0.9
0.1
(e) 0.2
;d Concentration (ppm) (b)
f» W W » U^V-
Standard % Chg
2.18E-08 1623.2
4.90E-08 241.5
1.24E-03 3.2
0.2
0.6
6.31E-03 3.5
0.4
0.8
1.26E-03 2.3
0.3
0.6
3.28E-02 1.3
0.1
0.3
I /Max
0.0
0.0
-0.4
0.0
0.0
1.1
0.1
0.1
-0.5
0.0
0.0
-0.4
0.0
0.0
% Fit
178.7
109.7
3.1
0.3
3.5
0.6
2.5
0.4
2.3
0.3
(a) The test problems are as indicated in Table 34. The following
notation is employed: (1) "Env" = environmental condition, where
SS = Summer surface, WS = Winter Surface, SA = Summer Aloft, and
SD = Summer Dynamic. (2) "Mix" = ROG surrogate, where "AL" = aloft
mixture, "AI" = aloft + isoprene, "AC" = allcity average, "DC" =
Washington DC mixture, and "BT" = Beaumont, Texas, mixture. (3)
"NO" is the effective NOV level in ppb. (4) C/N is the ROG/NOV ratio.
Ji A A
(b) See Section 6.2 for definitions of the statistical measures.
(c) The values given in the "standard" column are the highest maximum
or average concentrations in the set of standard calculations. The
values given in the "J Chg," "% /Max" and "% Fit" columns are the
comparison statistics for the calculations where the two mechanisms
are the most different.
(d) Average of the comparison statistics for all the calculations.
(e) Average of the absolute value for the comparison statistics for
all the calculations.
280
-------
The results of the test calculations comparing the version of the
RADM mechanism employing the 1986 SAPRC peroxy radical representation with
the detailed radical version are summari2ed in Table 38. As with the
previous comparison, the detailed radical mechanism is used as the
standard, and the results for individual test problems are shown only for
those cases where differences are greater than 5%. For most of the
species whose results are compared, this version of the RADM mechanism
also gave predictions very close to those for the detailed radical
version. However, this approximation did not perform as well in
simulating the predictions of the detailed radical model for the higher
organic peroxides (OP2) or the higher organic acids (ORA2) in some of the
test problems, particularly for some of the urban cases. In the case of
OP2, the approximate mechanism had a slight tendency (generally less than
5%) towards predicting lower levels of OP2 in those simulations where
significant quantities are predicted, with a few cases of much higher
relative predictions for some test problems where very low (sub-ppt)
levels of OP2 are predicted. Since the large relative differences are
generally restricted to cases where OP2 predictions are generally
negligible, this is probably not a significant discrepancy. On the other
hand, the disagreements for ORA2 are significant, since the approximate
model predicted from 0-40J higher ORA2 levels in cases where its formation
is non-negligible.
The discrepancies in the higher organic acid and (to a lesser extent)
higher organic peroxide predictions can be attributed to the fact that the
SAPRC approximation requires assuming the same rate constants for similar
types of peroxy + peroxy reactions while the detailed model uses a variety
of rate constants for these reactions. In particular, the most signifi-
cant discrepancies for ORA2 are due to the fact that the RADM mechanism
uses different rate constants for RCO^ + R02 reactions depending on the
nature of R02. These differences can be represented in the detailed RADM
mechanism, but not in the version using the SAPRC approximation. If it is
assumed that the rate constants for all RCOo + R02 reactions are the same,
then the mechanism using the SAPRC approximation performs much better in
simulating the ORA2 and OP2 results than the detailed version.
281
-------
Table 38. Results of Sensitivity Test Calculations for the Version of
the RADM Mechanism Using the 1986 SAPRC Peroxy Radical
Representation as the Test Mechanism, Relative to the
"Detailed Radical" Mechanism as the Standard
Test Problem (a)
Env Mix NOX C/N
Calculated Concentration (ppm) (b)
Maximum Average
Standard % Chg Standard % Chg % /Max % Fit
Rural Oo
Maximum (c)
Average (d)
Avg. Abs. Value
Urban 0^
Maximum (c)
Average (d)
Avg. Abs. Value
Rural H202
Maximum (c)
Average (d)
Avg. Abs. Value
Urban H202
Maximum (c)
Average (d)
Avg. Abs. Value
SA AC 67 3
WS DC 67 3
SA DC 67 3
WS BT 67 3
Rural SULF
Maximum (c)
Average (d)
Avg. Abs. Value
Urban SULF
Maximum (c)
Average (d)
Avg . Abs . Value
6.50E-02
(e)
3.47E-01
(e)
2.41E-03
(e)
1.48E-02
(e)
2.58E-07
3.12E-07
2.73E-07
2.05E-08
2.20E-03
(e)
6.25E-03
(e)
-1.0
0.0
0.2
-2.3
-0.7
0.8
-4.4
-0.2
0.8
-21.7
-1.1
2.1
-21.7
-15.8
-16.1
2.6
-2.4
-0.3
0.6
-2.2
-0.1
0.4
4.96E-02 -1.1
0.0
0.2
2.22E-01 -2.1
-0.7
0.7
1.85E-03 3.6
-0.3
0.7
1.20E-02 -18.7
-0.8
2.1
1.06E-07 -17.3
3.93E-08 -12.4
1.15E-07 -18.7
6.64E-09 12.9
1.53E-03 -1.5
-0.2
0.5
4.26E-03 1.3
0.1
0.5
-0.5
0.0
0.0
-1.6
-0.2
0.2
-1.6
-0.1
0.1
2.4
0.0
0.2
0.0
0.0
0.0
0.0
0.3
0.0
0.0
-0.7
0.0
0.1
1.1
0.1
2.1
0.5
3.5
0.4
20.5
1.4
19.0
13-7
20.5
12.3
1.5
0.2
1.4
0.3
(continued)
282
-------
Table 38 (continued) - 2
Test Problem (a)
Env Mix NOX C/N
Calculated Concentration (ppm) (b)
Maximum Average
Standard % Chg Standard % Chg % /Max % Fit
Rural HNO^
Maximum (c)
Average (d)
Avg. Abs. Value
Urban HN03
Maximum (c)
Average (d)
Avg. Abs. Value
SD AC 67
SD DC 67
SD BT 67
Rural ORA1
Maximum (c)
Average (d)
30
30
30
Avg. Abs. Value
SA AL -2
SD AI -2
SD AI -2
Urban ORA1
Maximum (c)
Average (d)
20
6
20
1.
(e)
8.
(e)
5.
5.
5.
1.
(e)
5.
3.
2.
96E-03
44E-02
45E-03
59E-03
83E-03
85E-04
12E-07
58E-06
03E-05
4.52E-03
-4.
0.
1.
9.
0.
1.
9.
6.
8.
-6.
-0.
1.
-5.
-6.
-6.
1
2
1
3
2
0
3
6
0
6
3
1
1
1
,6
-1.4
9.
3.
2.
2.
2.
1.
3.
1,
1,
86E-04
36E-02
61E-03
68E-03
79E-03
38E-04
.79E-07
.99E-06
. 11E-05
3.00E-03
-0.4
Avg. Abs. Value
SA AL -2
SD AI ~2
SD AI -2
Rural ORA2
Maximum (c)
Average (d)
20
6
20
(e)
5,
0.4
. 12E-07
3.58E-06
2.03E-05
2.68E-04
Avg. Abs. Value
SS AL < 1
S3 AL < 1
20
66
(e)
3
1
.80E-06
.34E-05
-5
-6
-6
102
25
25
12
33
.1
.1
.6
.8
.5
.5
.3
.5
3
1
1
2
2
1
.79E-07
.99E-06
. 11E-05
. 10E-04
.92E-06
.04E-05
-2.
0.
0.
5.
0.
0.
5.
3.
5.
-5.
-0.
1.
-4,
-5,
-4
-1
-0
. 0
-4
-5
-4
116
24
24
13
34
9
0
8
6
0
7
6
8
0
,7
.2
.0
.1
.7
.1
.9
.3
.4
.1
.7
.1
.1
.8
.8
.2
.9
-0.5
0.0
0.1
-1.1
-0.1
0.1
0.4
0.3
0.4
1.3
0.0
0.0
0.0
-0.1
-0.3
-0.6
-0.1
0.1
0.0
-0.1
-0.3
20.0
1.5
1.5
0.2
1.7
3.3
0.4
5.5
0.5
5.5
3.8
4.9
5.9
0.4
4.2
5.9
4.3
1.1
0.2
4.2
5.9
4.3
73.4
8.2
12.3
29.7
(continued)
283
-------
Table 38 (continued) - 3
Test Problem (a)
Env Mix NOX C/N
SS
SS
SS
SS
SS
SS
SS
us
ws
SA
SA
SA
SD
SD
SD
SS
SS
SS
SS
SS
SS
SS
WS
WS
us
SA
SA
SA
SD
SD
SD
AL
AL
AL
AL
AL
AL
AL
AL
AL
AL
AL
AL
AL
AL
AL
AI
AI
AI
AI
AI
AI
AI
AI
AI
AI
AI
AI
AI
AI
AI
AI
<1
-2
-2
-2
5
5
5
-2
_2
-2
-2
-2
~2
-2
-2
<1
<1
<1
-2
-2
5
5
~2
-2
~2
-2
-2
-2
-2
-2
-2
200
6
20
60
2
7
20
20
60
6
20
60
6
20
60
20
66
200
20
60
7
20
6
20
60
6
20
60
6
20
60
Maxin
Standard
4.14E-05
7.68E-06
2.47E-05
7.19E-05
6.21E-06
2.23E-05
7.44E-05
2.55E-06
1.18E-05
1.01E-07
1.33E-06
2.12E-05
2.44E-06
9.98E-06
3.04E-05
1.22E-05
5.39E-05
1.88E-04
6.62E-05
2.68E-04
4.92E-05
2.18E-04
1.06E-05
3.42E-05
1.80E-04
1.94E-07
6.55E-06
1.62E-04
6.37E-06
3.36E-05
1.24E-04
Calculate
mm
* Chg
71.2
7.4
25.7
54.3
10.5
21.3
44.9
13.4
45.2
16.5
55.3
102.8
8.0
23.8
50.5
9.2
17.4
26.8
12.8
22.6
9.3
19.6
5.0
8.9
10.7
7.9
3L9
39.3
31.9
31.9
30.6
;d Concentra
a
Standard
3.15E-05
4.63E-06
1.58E-05
4.79E-05
4.24E-06
1.56E-05
5.31E-05
2.12E-06
8.46E-06
4.03E-08
6.82E-07
1.02E-05
1.13E-06
5.02E-06
1 .52E-05
6.37E-06
2.80E-05
1.06E-04
5.23E-05
2.10E-04
3.37E-05
1.37E-04
6.48E-06
2.50E-05
1.37E-04
1.59E-07
5.31E-06
7.78E-05
3.90E-06
1.94E-05
6.94E-05
ition (p
iverage
* Chg
71.7
6.3
28.5
59.0
10.8
20.3
42.3
11.1
33.3
24.9
63.7
116.1
7.6
22.0
46.1
8.5
17.1
27.8
10.7
20.0
7.8
16.9
4.9
7.9
9.8
10.8
31.3
38.4
16.9
21.1
24.5
pm) (b)
% /Max
10.8
0.1
2.1
13.4
0.2
1.5
10.7
0.1
1.3
0.0
0.2
5.6
0.0
0.5
3.3
0.3
2.3
14.1
2.7
20.0
1.2
11.0
0.2
0.9
6.4
0.0
0.8
14.2
0.3
1.9
8.1
% Fit
52.7
6.2
24.9
45.5
10.2
18.4
34.9
10.6
28.7
22.1
48.3
73.4
7.3
19.8
37.5
8.2
15.8
24.4
10.2
18.2
7.5
15.6
4.8
7.7
9.4
10.2
27.1
32.2
15.6
19.1
21.8
Urban ORA2
Maximum (c)
Average (d)
Avg. Abs. Value (e)
SS AC 20 10
SS AC 20 33
SS AC 20 100
SS AC 67 3
SS AC 67 10
SS AC 67 30
6.46E-03
4.22E-04
1 .44E-03
4.19E-03
2.05E-04
1.79E-03
4.94E-03
82.4
37.4
37.4
39.9
59.5
40.7
6.6
50.1
57.1
3.52E-03
2.53E-04
1.03E-03
3.22E-03
1.50E-04
1.00E-03
3.39E-03
76.9
31.5
31.5
33.6
54.4
38.5
14.8
47.1
52.8
60.7
7.3
7.3
2.4
15.8
35.2
0.6
13.4
50.8
55.6
15.5
28.8
42.9
32.4
13.9
38.2
41.9
(continued)
284
-------
Table 38 (continued) - 4
Test Problem
Env Mix NOV
A
SS
ss
SS
ws
W2
SA
SA
SD
SD
SD
SS
SS
SS
SS
SS
SS
SS
SS
SS
ws
ws
SA
SA
SD
SD
SD
SS
SS
SS
SS
SS
SS
SS
SS
SS
ws
ws
SA
SA
SD
SD
SD
AC
AC
AC
AC
AC
AC
AC
AC
AC
AC
DC
DC
DC
DC
DC
DC
DC
DC
DC
DC
DC
DC
DC
DC
DC
DC
BT
BT
BT
BT
BT
BT
BT
BT
BT
BT
BT
BT
BT
BT
BT
BT
100
20Q
300
67
67
67
67
67
67
67
20
20
20
67
67
67
100
200
300
67
67
67
67
67
67
67
20
20
20
67
67
67
100
200
300
67
67
67
67
67
67
67
(a)
C/N
2
3
10
10
30
10
30
3
10
30
10
33
100
3
10
30
2
3
10
10
30
10
30
3
10
30
10
33
100
3
10
30
2
3
10
10
30
10
30
3
10
30
._ _ c
Maxim
Standard
1.
.94E-04
6.42E-04
6.46E-03
7
.27E-04
3.39E-03
4
4
9
4
2
3
1
3
1
1
4
1
5
5
4
2
3
3
6
3
1
3
1
3
1
1
4
1
6
5
5
2
3
.89E-04
.84E-03
.29E-05
.79E-04
.04E-03
.71E-04
.25E-03
.25E-03
.38E-04
.64E-03
.23E-03
.25E-04
.50E-04
.87E-03
.69E-04
.35E-03
.57E-04
.70E-03
.87E-05
.78E-04
.58E-03
.58E-04
.21E-03
.44E-03
.73E-04
.52E-03
.22E-03
.63E-04
.38E-04
-53E-03
.71E-04
.59E-03
.45E-04
3.50E-03
7
4
1
.94E-05
.02E-04
.68E-03
alculat<
urn
* Chg
0.6
46.9
54.7
1.9
8.0
37.7
47.5
26.4
68.4
66.6
53.0
76.4
58.6
43.4
60.8
75.2
27.1
59.1
67.4
18.8
17.4
60.6
64.4
31.0
78.2
82.4
34.5
51.2
39.0
10.7
40.9
51.5
0.1
34.0
45.3
0.2
4.9
22.4
44.1
24.9
65.7
65.6
3d Concentra
Standard
1.
4.
3.
5.
2.
3.
3.
5.
2.
1.
2.
8.
2.
1.
9.
2.
8.
3.
04E-04
66E-04
52E-03
02E-04
29E-03
79E-04
09E-03
28E-05
55E-04
09E-03
17E-04
51E-04
44E-03
07E-04
02E-04
78E-03
09E-05
45E-04
3.16E-03
3.46E-04
1.64E-03
2.34E-04
2.11E-03
3.84E-05
1.98E-04
8.33E-04
2.20E-04
8.41E-04
2.63E-03
1.31E-04
8.62E-04
2.78E-03
8,
. 17E-05
3.99E-04
3.00E-03
3
1
2
2
4
2
8
.81E-04
.75E-03
.98E-04
.24E-03
.49E-05
. 12E-04
.83E-04
tion (p
verage
% Chg
9.2
24.9
54.6
12.5
16.0
11.9
22.8
18.6
53.5
59.2
45.2
73.7
57.0
22.5
58.1
73.2
16.5
36.7
67.6
22.0
26.4
23.9
36.6
23.5
65.2
76.9
30.7
47.3
36.5
11.2
. 40.6
47.7
6.9
16.3
47.5
6.3
11.9
6.1
19.6
17.2
50.1
56.9
>pm) (b)
% /Max
0.3
3.3
54.6
1.8
10.4
1.3
20.0
0.3
3.9
18.3
2.8
17.8
39.5
0.7
14.9
57.8
0.4
3.6
60.7
2.2
12.3
1.6
21.9
0.3
3.7
18.2
1.9
11.3
27.3
0.4
9.9
37.5
0.2
1.8
40.4
0.7
5.9
0.5
12.5
0.2
3.0
14.3
% Fit
8.8
22.3
42.9
11.8
14.9
11.4
20.7
17.0
42.2
45.7
36.9
53.9
44.5
20.4
45.0
53.7
15.3
31.2
50.6
19.9
23.4
21.6
31.2
21.0
49.2
55.6
26.7
38.4
31.1
10.8
33.8
38.6
6.8
15.3
38.4
6.2
11.3
6.1
18.1
15.9
40.1
44.3
(continued)
285
-------
Table 38 (continued) - 5
Test Problem (a)
Env Mix NOX C/N
Calculated Concentration (ppm) (b)
Maximum Average
Standard % Chg Standard J Chg % /Max % Fit
Rural OP1
Maximum (c)
Average (d)
Avg.
SS
SS
SS
SS
US
SA
SS
SS
SS
WS
US
Abs.
AL
AL
AL
AL
AL
AL
AI
AI
AI
AI
AI
. Value <
<1
<1
5
5
~2
-2
<1
5
5
-2
_2
20
66
2
7
20
6
20
2
7
6
60
1.72E-03
:e)
4.
4.
6.
6.
4.
2.
4.
6.
6.
3.
2.
01E-04
03E-04
79E-04
61E-04
24E-05
73E-05
08E-04
97E-04
95E-04
43E-05
63E-04
7.9
2.9
3.5
5.8
5.0
7.8
7.5
4.6
-6.5
7.0
7.9
6.9
2.2
5.2
1.
2.
2.
3.
3.
2.
8.
3.
3.
4.
25E-03
58E-04
85E-04
62E-04
84E-04
48E-05
98E-06
12E-04
79E-04
06E-04
1.89E-05
1.60E-04
56.9
2.8
3.2
3.7
3.0
4.3
4.3
5.5
-5.5
4.3
4.1
3.4
6.9
15.0
1.9
0.2
0.2
0.8
0.7
1.3
1.3
0.1
0.0
1.1
1.3
1.1
0.1
1.9
13.8
1.3
3.6
3.0
4.5
4.2
5.4
6.6
4.5
4.1
3.4
6.7
13.8
Urban OP1
Maximum
Average
Avg.
SS
SS
WS
US
US
SA
SA
SS
SS
SS
us
SA
SA
SS
US
us
us
SA
SA
Abs
AC
AC
AC
AC
AC
AC
AC
DC
DC
DC
DC
DC
DC
BT
BT
BT
BT
BT
BT
(c)
(d)
1.38E-02
. Value
67
300
67
67
67
67
67
67
100
300
67
67
67
300
67
67
67
67
67
10
10
3
10
30
3
30
10
2
10
3
3
30
10
3
10
30
3
30
(e)
1.67E-03
4.82E-03
1
5
2
7
1
1
6
4
1
5
1
3
6
3
1
.72E-06
.24E-04
.21E-03
.44E-07
.38E-02
.66E-03
.50E-04
.74E-03
.34E-06
.29E-07
.27E-02
.68E-03
.42E-07
.68E-04
.78E-03
2.77E-07
1
.09E-02
404.3
29.7
30.0
7.5
12.5
78.5
6.3
4.0
286.3
9.0
8.6
6.1
14.7
197.9
237.4
9.1
6.0
258.6
8.7
3.1
404.3
8.5
8.38E-03
1
2
3
3
1
3
8
1
2
3
2
2
7
2
1
1
1
1
. 11E-03
.95E-03
.74E-07
.21E-04
55E-03
.58E-07
.38E-03
. 12E-03
.88E-04
.OOE-03
.92E-07
.55E-07
.52E-03
.37E-03
.20E-07
.79E-04
.18E-03
.53E-07
6.42E-03
417.2
3L3
31.8
-1.1
-0.7
125.4
10.4
5.3
273.2
3.9
-1.4
2.3
-0.5
202.6
233.4
4.0
-0.6
417.2
17.9
5.4
367.6
3.6
3.9
0.1
0.2
-0.1
-0.3
0.0
0.4
1.0
0.0
3.9
-0.2
0.1
-0.2
0.0
0.0
3.6
-0.2
0.0
0.4
0.8
0.0
2.7
135.1
8.9
3.7
6.1
78.2
9.8
5.1
115.5
3.9
4.3
2.8
6.6
100.6
107.6
4.0
5.5
135.1
16.4
5.3
129.5
3.6
(continued)
286
-------
Table 38 (continued) - 6
Test Problem (a)
Env Mix NOV C/N
A
Calculated Concentration (ppra) (b)
Maximum Average
Standard % Chg Standard f. Chg % /Max % Fit
Rural OP2
Maximum (c)
Average (d)
Avg.
SA
WS
Urban
Abs
AL
AI
. Value
-2
-2
6
6
1.
(e)
.59E-03
5.32E-06
4.72E-06
-7.5
-0.8
1.6
-7.5
6.2
1.
2.
3-
28E-03
12E-06
44E-06
17.5
-0.6
2.1
-7.7
17.5
-1.3
-0.1
0.1
0.0
0.0
16.0
0.9
8.3
16.0
OP2
Maximum
Average
Avg.
SS
SS
SS
SS
SS
SS
SS
WS
WS
WS
SA
SA
SA
SS
SS
SS
SS
SS
SS
SS
SS
WS
WS
WS
SA
SA
SS
SS
SS
Abs
AC
AC
AC
AC
AC
AC
AC
AC
AC
AC
AC
AC
AC
DC
DC
DC
DC
DC
DC
DC
DC
DC
DC
DC
DC
DC
BT
BT
BT
(c)
(d)
. Value
20
20
20
67
67
200
300
67
67
67
67
67
67
20
20
20
67
67
100
200
300
67
67
67
67
67
20
20
20
10
33
100
10
30
3
10
3
10
30
3
10
30
10
33
100
10
30
2
3
10
3
10
30
3
30
10
33
100
3
(e)
8
5
1
3
1
1
1
4
1
7
7
4
3
9
6
1
4
2
2
1
1
4
2
9
6
3
6
4
1
.42E-02
. 10E-04
.97E-03
.76E-02
.66E-03
.87E-02
.55E-03
.39E-02
.29E-07
.74E-03
.97E-03
.20E-07
.08E-03
.34E-02
.23E-04
.68E-03
.86E-02
.28E-03
.09E-02
.22E-04
.78E-03
.62E-02
.36E-07
.07E-03
.08E-03
.50E-07
.42E-02
.05E-04
.41E-03
.44E-02
408.0
7.4
24.7
-15.0
-8.3
-8.6
-27.4
-14.8
-3.1
-3^.6
32.5
-12.2
-13.8
99.8
-2.9
-6.3
-16.8
-10.2
-9.3
-28.5
-15.3
-4.2
-3.5
-3M.O
62.2
-13.2
-15.2
140.9
-6.3
-12.7
-6.4
-6.8
2.63E-02
4.26E-04
4.23E-03
1.
1 ,
1.
.34E-02
.65E-03
.30E-02
6.95E-04
6.38E-03
9.43E-08
1.
5
2
2
2
4
4
1
1
1
9
8
7
8
1
6
2
2
3
3
1
.03E-03
.63E-03
.41E-07
.43E-03
.53E-02
.67E-04
.69E-03
.42E-02
.88E-03
.45E-02
.35E-05
.07E-04
.33E-03
.53E-08
.24E-03
.40E-03
.34E-07
.63E-02
.81E-04
.17E-03
.10E-02
311.2
10.1
23.2
-7.5
-8.3
-7.9
-14.8
-11.5
-6.2
-21.3
36.5
-11.1
-12.2
125.5
-0.9
-3.8
-7.7
-9.0
. -8.5
-15.7
-12.0
-6.2
-6.5
-21.5
73.6
-12.1
-13.6
136.6
-4.1
-6.1
-7.3
-7.0
-6.6
-0.8
0.8
-0.1
-1.3
-4.0
-0.9
-5.7
-0.2
-5.2
0.0
-0.4
-2.6
0.0
-0.1
-3.7
-0.1
-1.6
-4.6
-1.1
-6.6
0.0
-0.2
-6.0
0.0
-0.6
-3.3
0.0
-4.1
-0.1
-0.9
-2.9
121.8
9.5
7.8
8.6
8.3
16.0
12.2
6.8
23.8
36.5
11.8
13.0
77.1
0.9
3.9
8.0
9.5
8.9
17.1
12.8
6.4
7.0
24.1
53.8
12.9
14.6
81.3
4.2
6.3
7.6
7.3
(continued)
287
-------
Table 38 (continued) - 7
Test Problem (a)
Env Mix NOX C/N
SS BT 67
SS BT 67
SS BT 200
SS BT 300
WS BT 67
WS BT 67
WS BT 67
SA BT 67
SA BT 67
Rural PAA
Maximum (c)
Average (d)
10
30
3
10
3
10
30
3
30
Avg. Abs. Value
WS AL -2
SA AL -2
Urban PAA
Maximum (c)
Average (d)
6
6
p
Maxim
Standard
2.
1.
8.
7.
1.
1.
5.
7.
2.
1.
(e)
31E-03
39E-02
26E-04
86E-03
10E-07
02E-03
41E-03
64E-08
76E-02
87E-03
3.29E-06
3.34E-06
1.05E-02
Avg. Abs. Value
WS AC 67
WS AC 67
SA AC 67
HS DC 67
WS DC 67
WS DC 67
SA DC 67
SS BT 300
WS BT 67
SA BT 67
Rural PAN
Maximum (c)
Average (d)
3
30
3
3
10
30
3
10
30
30
Avg. Abs. Value
SD AL -2
SD AI -2
60
6
(e)
3
3
6
5
7
2
7
4
2
7
1
(e)
.31E-08
.01E-03
.38E-08
.56E-08
. 17E-04
.96E-03
.26E-08
.07E-03
,51E-03
.94E-03
.87E-03
7.41E-05
2.04E-05
alculate
ium
% Chg
-28.1
-14.1
-6.8
-30.8
122.5
-8.5
-11.0
408.0
-5.5
-7.3
-0.2
1.8
-7.3
-7.0
-12.3
-1.0
3.0
-6.3
9.0
-6.3
-12.3
5.6
11.8
-9.7
-5.9
6.5
-6.2
-5.3
-0.8
1.0
-2.2
-5.3
id Concentra
Standard
1.
9.
3-
4.
2.
4.
3.
4.
2.
29E-03
68E-03
29E-04
38E-03
18E-08
80E-04
56E-03
90E-08
01E-02
1.24E-03
1.35E-06
1,
6
4
1
3
7
4
1
3
2
1
4
1
. 13E-06
.31E-03
.95E-09
.91E-03
. 11E-08
.11E-09
.32E-04
.93E-03
.28E-08
.54E-03
.52E-03
.45E-03
.26E-03
3.37E-05
7.42E-06
.tion (pi
iverage
% Chg
-13.8
-10.3
-9.4
-21.1
215.1
-7.2
-10.2
311.2
-2.7
-5.0
-0.2
1.4
-5.0
-0.4
11.8
-1.0
2.4
-2.5
8.8
-2.8
-9-2
5.4
11.8
-4.4
-4.2
6.6
-2.2
-7.1
-0.8
2.2
-5.6
-6.7
3m)
(b)
% /Max
-0.
-3.
-0.
-3.
0.
-0.
-1.
0.
-2.
7
8
1
5
0
1
4
0
1
2.0
0.0
% Fit
14.
10.
9.
23.
103.
7.
10.
121.
2.
8
9
9
5
7
7
7
8
8
5.1
0.6
0.1
0.0
0.0
3
-0
0
0
2
0
0
0
3
0
-1
1
-1
-1
.6
.1
.3
.0
.6
.0
.0
.4
.6
.0
.7
.6
.6
.8
0.0
5.1
1.7
11
1
3
8
2
9
5
11
4
5
6
2
.1
.6
.8
.8
.9
.8
.4
.1
.5
.4
.5
.3
9.3
1
.0
0.1
-0.1
0.0
5.8
7.3
(continued)
288
-------
Table 38 (continued) - 8
Test Problem
Env Mix NOV
A
SD AI
SD AI
Urban PAN
Maximum (
Average (
Avg. Abs.
SD AC
SD DC
SD BT
-2
-2
c)
d)
(a)
C/N
20
60
Value
67
67
67
30
30
30
Maxin
Standard
6
1
4
(e)
5
4
5
33E-05
.34E-04
.10E-02
. 18E-03
.97E-03
.32E-03
^alculat
num
% Chg
1.8
1.1
-6.6
-1.3
1.3
-6.6
-5.0
-5.1
ed i
Concentre
Standard
2
5
3
3
3
3
.57E-05
.91E-05
.28E-02
.67E-03
.51E-03
.74E-03
ition (f
Average
% Chg
-7.1
-8.5
-8.5
-1.0
1.4
-7.0
-5.5
-6.5
jpm) (b)
% /Max
-0.1
-0.4
-2.3
-0.1
0.2
-0.8
-0.6
-0.7
%
8
9
7
1
7
5
6
Fit
.5
.3
.2
.0
.2
.6
.7
(a) See Footnote (a) in Table 37 for a description of the notation used
to indicate the test problems.
(b) See Section 6.2 for definitions of the statistical measures.
(c) The values given in the "standard" column are the highest maximum
or average concentrations in the set of standard calculations. The
values given in the "% Chg," "% /Max" and "% Fit" columns are the
comparison statistics for the calculations where the two mechanisms
are the most different.
(d) Average of the comparison statistics for all the calculations.
(e) Average of the absolute value for the comparison statistics for
all the calculations.
289
-------
The significance of this apparent failure of the SAPRC approximation
to reproduce detailed model predictions of the higher organic acids
depends upon the extent to which the detailed mechanism taken as the
standard is assumed to represent reality. Had we developed the detailed
radical mechanism for use in this evaluation, we would have estimated,
based on the limited laboratory data available (Atkinson, 1989), that the
rate constants for all the RC^ + RCOg reactions were approximately the
same, and thus the SAPRC radical approximation would have performed much
better in reproducing the detailed model. It is clear that if there are
variations in the RCO-j + RC^ rate constants to the extent assumed in the
RADM mechanism, then the mechanisms utilizing the SAPRC approximation to
represent peroxy radical + peroxy radical reactions will not perform
satisfactorily in predicting yields of the higher organic acids. However,
if the detailed model uses incorrect estimates for these rate constants
which may well be the case since few have been measured the predictions
of the detailed model may be even less reliable than the approximate
one. No suitable data are available to test model predictions of organic
acids.
In summary, the test calculations on the representation of peroxy +
peroxy reactions tend to support the approach which is currently employed
in the evaluated RADM mechanism. This does not mean, however, that it may
not be possible to employ more efficient approaches involving fewer reac-
tions and possibly fewer species. However, at the present time no alter-
natives in this regard are recommended for the RADM mechanism.
6.4 Lumping of Higher Alkanes
One area where the RADM mechanism is more detailed than most other
current condensed mechanisms is its representation of the higher (C3-O
alkanes. However, test calculations previously carried out using the
latest SAPRC mechanism (Carter, 1988) indicate that relatively severe
condensations of alkanes have relatively little effect on model predic-
tions if the parameters and rate constants for the lumped species are
adjusted appropriately based on the mixtures being represented. There-
fore, as part of this study we carried out test calculations to determine
whether a moderate level of condensation of the RADM mechanism, employing
two lumped species to represent the C3+ alkanes rather than three, would
290
-------
have a significant effect on model predictions relative to the current
version.
The condensation examined was to lump the species HC5 and HC8 and
represent them by a single species, designated HC6. For direct comparison
with the March 1988 RADM mechanism, the OH radical rate constant and
product yield parameters for the new lumped species were determined by the
averages of those for HC5 and HC8 in the March 1988 mechanism, weighed by
the estimated relative numbers of moles of each reacting in RADM applica-
tions. These were derived using the total detailed U.S. emissions, and
assuming an integrated OH radical parameter (INTOH) value of 110 ppt-min,
using the methods and data bases discussed in Section 3.2. The reaction
used for HC6 was,
HC6 + HO > HC6P * 0.54 X02 + H20
where k = 3.02 x 10~11 exp (-380/T) cm3 molecule'1 sec"1. The reactions
used for HC6P, its associated peroxy peroxy radical, were
HC6P + NO > 0.4 ALD + 0.91 KET +1.17 ONIT + 0.03 HCHO +
0.83 N02 + 0.83 H02
HC6P + H02 > OP2
HC6P -t- AC03 > 0.12 ALD + 0.88 KET + 0.5 H02 + 0.5 M02 + 0.5 ORA2
HC6P + M02 > 0.75 HCHO + H02 + 0.089 ALD +0.66 KET
where their rate constants were the same as those used for HC5P and
HC8P. The only other modifications made to the RADM mechanism was the
removal of the reactions of HC5, HC5P, HC8, and HC8P.
(Note that the kinetic and product yield parameters used for HC6 and
HC6P in the modified mechanism for the purpose of this test are slightly
different from those used for this species in the RADM-M mechanism. The
parameters in the latter mechanism are based on a more recent analysis of
the emissions data than used to derive the parameters for the March 1988
version of the RADM mechanism used as the standard in this test. The
291
-------
parameters used in the test were derived to be consistent with those used
in the standard mechanism for more direct comparison.)
A comparison of the predictions using this "condensed alkane" RADM
mechanism with those of the standard mechanism in simulations of the 90
test problems is presented in Table 39. The results for individual test
problems are shown only for those cases where differences are greater than
5%. These results show that despite the fact that these scenarios used
five different ROG surrogates with four different profiles for the
alkanes, this level of condensation had an almost insignificant effect on
the simulations of the species of interest in regional oxidant or acid
deposition models. There are only a few cases where relative discrepan-
cies greater than 3% were observed, and in essentially all these cases the
predicted concentrations of the species were very low.
It is on the basis of these results that we recommend that the RADM
mechanism be condensed in its representation of these higher alkanes. It
is possible that it may also be appropriate to lump HC3 with the higher
alkane classes. However, more significant condensations of the RADM
mechanism was considered to be beyond the scope of this evaluation
program.
6.5 The Recommended vs. March 1988 RADM Mechanisms
As discussed in Section 2, as a result of the tests carried out in
this program, we have recommended modifications in the RADM mechanism,
relative to the March 1988, version which was evaluated in this study.
The most important modifications in terms of effects on model performance
concern changes in the representation of the unknown aspects of the aroma-
tic photooxidation mechanism to improve the performance of the mechanism
in simulating the results of aromatic - NOX - air chamber experiments.
Minor changes in rate constant and product yields are recommended for some
of the lumped species based on the results of an updated analysis of
emissions data as discussed in Section 3. Other minor changes are recom-
mended for the representation of phenol and nitrophenol + NO^ reactions to
improve chemical reasonableness. In addition, as a result of the tests
discussed in the previous section, we recommended that the two higher
alkane species in the RADM mechanism be lumped together. All the
292
-------
Table 39. Results of Sensitivity Test Calculations Using the Version of
the RADM Mechanism with the Higher Alkane Species HC5 and HC8
Lumped Together, Relative to the Standard RADM Mechanism
Test Problem (a)
Env Mix NOX C/M
Calculated Concentration (ppm) (b)
Maximum Average
Standard % Chg Standard % Chg % /Max % Fit
Rural O^
Maximum (c) 6.49E-02
Average (c)
Avg. Abs. Value (c)
Urban Cs
Maximum (c) 3.47E-01
Average (c)
Avg. Abs. Value (c)
Rural HoC^
Maximum (c) 2.41E-03
Average (c)
Avg. Abs. Value (c)
Urban H202
Maximum (c) 1.48E-02
Average (c)
Avg. Abs. Value (c)
SA DC 67 3 2.59E-07
WS BT 67 3 1.71E-08
SA BT 67 3 4.36E-08
Rural SULF
Maximum (c) 2.21E-03
Average (c)
Avg. Abs. Value (c)
Urban SULF
Maximum (c) 6.25E-03
Average (c)
Avg. Abs. Value (c)
-1.9
-0.1
0.2
1.6
0.0
0.2
2.8
-0.1
0.6
27.4
1.1
1.5
5.1
27.4
20.2
-3.6
-0.7
0.9
-1.1
-0.2
0.4
4.97E-02 -1.4
-0.2
0.2
2.23E-01 1.3
0.1
0.2
1.82E-03 3.2
-0.1
0.6
1.20E-02 11.1
0.5
0.9
1.02E-07 0.8
6.31E-09 11.1
3.47E-08 4.5
1.53E-03 -3-9
-0.7
0.8
4.26E-03 -1.1
-0.1
0.3
-0.6
0.0
0.0
0.2
0.0
0.0
-1.5
0.0
0.1
-0.5
0.0
0.1
0.0
0.0
0.0
-1.0
0.0
0.1
-0.5
0.0
0.1
1.4
0.1
1.3
0.1
3.3
0.3
13.2
0.6
2.3
13.2
4.6
3-9
0.4
1.1
0.2
(continued)
293
-------
Table 39 (continued) - 2
Toot- D*»r»K1 ARI ^2^ __ P
Maxim
Env Mix NOX C/N Standard
Rural HN03
Maximum (c) 1.95E-03
Average (c)
Avg. Abs. Value (c)
SS AL <1 200 6.48E-05
Urban HN03
Maximum (c) 8.46E-02
Average (c)
Avg. Abs. Value (c)
Rural ORA1
Maximum (c) 1.84E-04
Average (c)
Avg. Abs. Value (c)
Urban ORA1
Maximum (c) 4.53E-03
Average (c)
Avg. Abs. Value (c)
Rural ORA2
Maximum (c) 2.68E-04
Average ( c )
Avg. Abs. Value (c)
SA AL -2 20 1.29E-06
SA AL -2 60 2.03E-05
Urban ORA2
Maximum (c) 6.19E-03
Average (c)
Avg. Abs. Value (c)
alculatc
urn
% Chg
-3.8
-1.3
1.4
-3.6
-1.0
-0.1
0.3
3.2
0.4
0.7
1.2
0.3
0.4
5.2
0.9
1.5
4.8
5.2
2.7
0.3
0.5
;d Concentration (ppra) (b) -
Standard \ Chg
9.87E-04 -6.9
-1.3
1.4
2.68E-05 -6.9
3.36E-02 -2.3
-0.1
0.2
1.37E-04 2.7
0.4
0.7
3.00E-03 1.0
0.3
0.3
2.11E-04 5.2
1.0
1.3
6.60E-07 5.2
9.61E-06 5.0
3.42E-03 1.1
0.3
0.4
% /Max
-1.7
-0.1
0.1
-0.2
-0.6
0.0
0.0
0.4
0.0
0.0
0.7
0.1
0.1
1.1
0.1
0.1
0.0
0.2
0.7
0.1
0.1
* Fit
7.1
0.6
7.1
0.9
0.2
2.6
0.3
1.0
0.2
5.1
0.6
5.1
4.9
1.1
0.3
(continued)
294
-------
Table 39 (continued) - 3
Test Problem (a)
Env Mix NOV C/N
A
Calculated Concentration (ppm) (b)
Maximum Average
Standard % Chg Standard % Chg % /Max % Fit
Rural OP1
Maximum (c) 1.70E-03
Average (c)
Avg. Abs. Value (c)
5.4
0.2
0.8
1.25E-03 5.0
0.5
0.7
0.9
0.0
0.0
4.9
0.4
SA AL ~2 20
Urban OP1
8.48E-05 5.4 4.61E-05 5.0
0.2
Rural OP2
Urban OP2
4.9
Maximum
Average
Avg.
SA
SA
WS
SA
Abs
AC
DC
BT
BT
(c)
(c)
1
.37E-02
. Value (c)
67
67
67
67
3
3
3
10
7
9
2
9
.04E-07
.17E-07
.95E-07
.OOE-04
124.0
1.7
3.8
-5.4
-40.2
124.0
5.4
8
3
4
8
3
-35E-03
.36E-07
.01E-07
.04E-08
.05E-04
56.5
0.8
2.5
-2.5
-35.0
56.5
6.1
0.8
0.0
0.0
0.0
0.0
0.0
0.2
56.8
1.6
2.6
42.4
56.8
5.9
Maximum
Average
Avg.
SS
ss
SS
WS
SA
SA
SD
WS
SA
Abs
AL
AL
AL
AL
AL
AL
AL
AI
AI
(c)
(c)
1
.61E-03
. Value (c)
<1
5
5
-2
-2
-2
~2
-2
-2
200
2
7
20
6
60
6
6
6
6
2
9
3
5
7
2
4
2
.31E-04
.92E-05
. 38E-05
.87E-05
.61E-06
.45E-04
.33E-05
.33E-06
.71E-06
16.2
1.4
3.6
5.7
-5.0
-4.4
8.1
-7.8
6.5
2.1
16.2
-5.9
1
4
2
7
2
2
6
1
3
1
.30E-03
.39E-04
.07E-05
. 10E-05
.56E-05
.19E-06
.60E-04 -
.37E-05
.30E-06
.40E-06
13.7
1.4
3.2
5.9
-1.7
-2.6
7.8
-5.6
6.2
5.0
13.7
-7.5
3.2
0.1
0.2
2.0
0.0
-0.1
0.2
0.0
3.2
0.1
0.0
0.0
13.0
1.5
5.7
5.1
2.8
7.5
7.2
6.0
4.9
13.0
8.3
Maximum
Average
Avg.
SS
WS
SS
(c)
(c)
3
.40E-02
Abs. Value (c)
AC
AC
DC
100
67
100
2
3
2
2
7
2
. 17E-04
.03E-07
.20E-04
-85.6
-2.2
4.3
-5.8
9.3
-6.9
2
8
1
9
.61E-02
.62E-05
.45E-07
.25E-05
-90.6
-2.3
4.6
-7.0
18.7
-7.9
1.5
0.1
0.1
0.0
0.0
0.0
165.4
4.3
7.2
17.0
8.3
(continued)
295
-------
Table 39 (continued) - 4
Test Problem (a]
Env Mix NOV C/N
A
WS DC 67 3
SA DC 67 3
SS BT 200 3
WS BT 67 3
SA BT 67 3
SA BT 67 10
Rural PAA
Maximum (c)
Average (c)
Avg. Abs. Value
SA AL -2 6
Urban PAA
Maximum (c)
Average (c)
Avg. Abs. Value
SA BT 67 3
SA BT 67 10
Rural PAN
Maximum (c)
Average (c)
Avg. Abs. Value
SS AL <1 200
Urban PAN
Maximum (c)
Average (c)
Avg. Abs. Value
I (
Maxiu
Standard
6.60E-07
1.78E-06
8.21E-04
2.58E-06
5.22E-07
3.10E-03
1.86E-03
(c)
3.35E-06
1 . 05E-02
(c)
1.65E-08
6.45E-04
1.87E-03
(c)
1.19E-04
4.10E-02
(c)
'alculat<
num
* Chg
6.5
-17.1
-5.5
-85.6
-34.3
3.8
-5.9
0.6
1.3
-5.9
6.6
0.4
1.3
6.5
6.6
-3.5
0.0
0.7
-3.5
2.1
0.1
0.4
2d Concentre
Standard
1.28E-07
5.85E-07
3.24E-04
3.76E-07
1.67E-07
1.42E-03
1.24E-03
1.17E-06
6.33E-03
1.37E-08
2.02E-04
1.26E-03
2.89E-05
3.28E-02
it ion (p
Werage
% Chg
7.4
-13.7
-0.8
-90.6
-31.0
5.3
4.8
0.9
1.4
-2.6
6.9
0.3
0.9
2.1
6.9
-6.1
-0.2
0.9
-6.1
-1.7
0.0
0.3
pm) (b)
* /Max
0.0
0.0
0.0
0.0
0.0
0.3
1.1
0.0
0.1
0.0
0.6
0.0
0.1
0.0
0.2
-0.8
0.0
0.0
-0.1
0.4
0.0
0.0
% Fit
11.1
23.1
3.9
165.4
38.1
5.2
4.6
0.6
3.4
6.6
0.7
2.1
6.6
6.4
0.5
6.4
2.6
0.3
(a) See Footnote (a) in Table 37 for a description of the notation used
to indicate the test problems.
(b) See Section 6.2 for a discussions of the specific measures of
differences between the mechanisms.
(c) See footnotes (c-e) in Table 37 or 38 for a description of these
quantities.
296
-------
recommended changes are implemented in the RADM-M mechanism. The changes
involving the aromatics mechanism and the updated parameters for the
lumped species were implemented in the RADM-P mechanism, a version where
only parameter values (for the most part) are changed. The modifications
involved in the RADM-P mechanism have been accepted by the RADM team, and
are being implemented into the RADM model (Stockwell, private communica-
tion, 1989).
Table 40 gives the summary statistics of the differences between the
two modified RADM mechanisms and the March 1988 version in their predic-
tions of the species of interest in the 90 test problems. (The March 1988
mechanism is taken as the "standard" for the purpose of this tabulation.)
Tables 11 and 42 give the differences in the calculations of ozone and
1^02 for each of the individual sets of urban test calculations, and
Figures 89 and 90 show comparisons , of concentration-time plots for
selected compounds for two such calculations. Figure 89 shows results
calculated for the urban summer surface static simulation with initial NOV
A
of 67 ppb and ROG/NOX ratio of 10, and Figure 90 shows the results
calculated for winter surface conditions with the same pollutant levels.
As with the plots in Section 5, the solid line designates results of
calculations using the RADM mechanism, the longer dashes designate results
calculated using RADM-P, and the shorter dashes indicate RADM-M results.
The results shown in Table 40 indicate that the recommended modified
mechanisms can in some cases give quite different predictions for
compounds of potential interest in RADM applications, especially for urban
conditions. The greater sensitivities in the urban cases are expected
since the most important modifications in the mechanism concerned the
aromatics, which are relatively more important in urban conditions. The
magnitudes of the differences depend on the conditions of the specific
test problems, as shown in Tables 41 and 42 for the predictions of ozone
and ^2^2 *>or a1^ the ur^an cases, and as shown in Figures 89 and 90 for
summer vs. winter environments for one set of pollution conditions. In
general, the modified mechanisms had a tendency to predict lower levels of
ozone, hydroperoxides and organic acids than the March 1988 mechanism,
with the effects on the peroxides and the acids being greater than the
effects on ozone. In general, the test cases in Tables 41 and 42 that
show the greatest differences in predictions for H202 are also the cases
297
-------
Table 40. Summary of Results of Sensitivity Test Calculations Using the
Recommended Modified (RADM-M) and the Recommended Modified
Parameter (RADM-P) Mechanisms, Relative to the March 1988
RADM Mechanism as the Standard
Test Cases
and
Compound
Rural 03
Maximum
Average
Avg. Abs. Value
Urban 0^
Maximum
Average
Avg. Abs. Value
Rural H202
Maximum
Average
Avg. Abs. Value
Urban H202
Maximum
Average
Avg . Abs . Value
Rural SULF
Maximum
Average
Avg. Abs. Value
Urban SULF
Maximum
Average
Avg. Abs. Value
Rural HN03
Maximum
Average
Avg. Abs. Value
Urban HN03
Maximum
Average
Avg. Abs. Value
RADM-M vs RADM (a)
Max. - Average - %
*Chg. JChg. J/Max Fit
3.9
0.0
0.4
-3^.8
-3.0
6.7
-9.3
-1.7
1.9
-85.3
-17.5
17.6
5.9
0.7
1.0
-42.4
1.5
10.8
7.8
0.8
1.8
-32.0
-0.1
9.7
3.
-0.
0.
-32.
-5.
3
2
5
7
5
7.3
-13.9
-2.1
2.3
-72.9
-19-8
19.
6
0
1
-32
1
8
6
0
1
-31
0
7
.8
.1
.6
.0
.9
.0
.6
.8
.4
.3
.3
.8
.8
1.5
-0.1
0.1
-9.0
-0.9
1.5
-4.0
-0.2
0.2
-18.1
-1.8
1.8
2.0
0.1
0.1
-13.1
0.0
1.2
3.0
0.1
0.1
-17.5
-0.1
1.6
3.2
0.2
39.2
5.4
14.8
1.0
115.8
14.9
5.9
0.4
39.3
5.8
6.5
0.7
37.1
6.2
RADM-P vs RADM (a)
Max. - Average - %
JChg. *Chg. %/Max Fit
6.
0.
0.
-34.
-1.
6.
-10.
-1.
1.
-89.
-16.
17.
7,
1,
1,
2
2
5
6
4
0
7
5
9
3
,9
,0
.4
.2
.4
-42.8
0.4
9
10
2
2
27
0
8
.4
.0
.0
.1
.8
.9
.9
5.
0.
0.
3
1
4
-27.9
-3.4
6.1
-15.1
-2.0
2,
-73
-19.
19
7
1
1
-31
-0
7
9
1
1
-28
1
7
.2
.7
.1
.1
.7
.1
.3
.9
.1
.6
.0
.6
.7
.5
.6
.6
2.5
0.0
0.1
9.9
-0.3
1.2
-4.5
-0.2
0.2
-17.8
-1.8
1.8
2.5
0.1
0.1
-13.2
-0.2
1.1
3.9
0.1
0.2
15.9
0.1
1.5
5.2
0.2
32.5
4.5
16.2
1.0
117.9
14.2
7.4
0.5
38.0
5.2
8.6
0.8
33.2
5.7
(continued)
298
-------
Table 40 (continued) - 2
Test Cases
and
Compound
Rural ORA1
Maximum
Average
Avg. Abs. Value
Urban ORA1
Maximum
Average
Avg. Abs. Value
Rural ORA2
Maximum
Average
Avg. Abs. Value
Urban ORA2
Maximum
Average
Avg. Abs. Value
Rural OP1
Maximum
Average
Avg. Abs. Value
Urban OP1
Maximum
Average
Avg. Abs. Value
Rural OP2
Maximum
Average
Avg. Abs. Value
Urban OP2
Maximum
Average
Avg. Abs. Value
Rural PAA
Maximum
Average
Avg. Abs. Value
RADM-M vs RADM (a)
Max. - Average - %
JChg. %Chg. */Max Fit
-6.
-1.
1.
-24.
-8.
8.
7
2
5
4
9
9
-28.5
-4.5
4.7
-40.7
-13.2
13.2
-10.2
-0.7
1
-79
-8
12
-28
-14
14
191
-36
43
-14
-1
2
.6
.9
.7
.8
.1
.1
.3
.3
.3
.4
.8
.5
.1
-6.5
-1.2
1.4
-21.6
-8.1
8.1
-42.5
-5.1
5.2
-29.3
-13.7
13.7
-15.3
-1.2
2.0
-75.7
-15.0
16.6
-23.1
-13.8
13.8
-84.8
-40.5
40.5
-22.1
-1.5
2.7
-0.
0.
0.
-8.
-1.
1.
-1.
-0.
0.
5
0
0
9
2
2
9
1
1
-17.8
-2.2
6.7
0.6
24.2
5.3
53.7
2.3
RADM-P vs RADM (a)
Max. - Average - %
JChg. JChg. */Max Fit
-7.
-1.
1.
-19.
-6.
6.
1
0
3
3
6
6
-20.4
-4.6
4.9
34.6
8.9
2.2
-1.
.6
0.0
16.5
0.9
0.1
-15
-0
0
-9
-0
0
-28
-3
3
-1
0
0
.3
.5
.9
.8
.6
.6
.6
.1
.1
.4
.0
.1
121.8
13.1
26.1
6.1
157.5
33.0
16.9
1.2
-39.1
-13.2
13.2
-12
-1
2
-73
-18
18
35
-5
7
900
-6
39
-17
-5
5
.8
.9
.1
.9
.7
.7
.7
.8
.8
.4
.2
.8
.8
.1
. 1
-6.6
-1.0
1.2
-17.0
-6.0
6.1
-21.1
-5.0
5.1
-26.4
-13.6
13.6
-13.8
-2.5
2.6
-69.0
-22.6
22.6
-17.3
-6.4
7.0
748.8
-10.0
37.7
-21.0
-6.1
6.1
-0.7
0.0
0.0
-6.4
-0.9
0.9
-1.6
-0.1
0.1
-16.4
-2.4
2.4
-3.2
-0.1
0.1
-19.8
-1.8
1.8
-7.6
-0.3
0.3
-19.5
-1.5
1.5
-4.3
-0.2
0.2
6.9
0.5
18.6
4.0
23.6
2.2
30.6
8.9
14.8
1.2
105.2
16.6
18.9
3.0
160.9
20.8
23.4
2.6
(continued)
299
-------
Table 40 (continued) - 3
Test Cases
and
Compound
Urban PAA
Maximum
Average
Avg. Abs. Value
Rural PAN
Maximum
Average
Avg. Abs. Value
Urban PAN
Maximum
Average
Avg. Abs. Value
RADM-M vs RADM (a)
Max. - Average - T>
JChg. JChg. J/Max Fit
-88
-13
15
9.
.2
.4
.8
2
2.4
2.
-61.
3.
15.
8
4
4
9
-73.4
-15.5
17.1
9-5
0.3
1.6
-39.3
-1.6
11.9
-20
-0
1
6.
0.
0.
30.
1.
1.
.3 117.2
.6 12.7
.3
8 9.1
1 1.2
2
7 61.4
1 9.4
6
Max.
JChg
-91
-25
25
-14.
-4.
4.
-64.
-17.
18.
RADM-P vs RADM (a
- Average -
. JChg. J/Max
.1
.1
.1
2
6
6
5
8
5
-82
-27
27
-14.
-4.
4.
-51.
-17.
17.
.0
.2
.2
2
6
6
6
8
8
-37.3
-3.3
3.3
-3.3
-0.2
0.2
-17.3
-1.7
1.7
)
*
Fit
139. 1
20.!
15.3
1.9
69.7
13.0
(a) See Section 6.2 for a discussions of the specific measures of
differences between the mechanisms. See footnotes (c-e) in
Table 37 or 38 for a description of the quantities tabulated.
300
-------
Table 41. Results of Sensitivity Test Calculations of Urban Ozone Formation Using the Recommended
Modified (RADM-M) and the Recommended Modified Parameter (RADM-P) Mechanisms, Relative
to the March 1988 Version of the RADM Mechanism as the Standard
Test Problem (a) Standard Cone (ppm) --- RADM-M vs RADM (b) --- --- RADM-P vs RADM (b) -
Env Mix NOx C/N Maximum Average Max. - Average - % Max. - Average - %
%Chg. flChg. £/Max Fit *Chg. £Chg. J/Max Fit
ss
ss
ss
ss
ss
ss
ss
ss
ss
ws
ws
ws
SA
SA
SA
SD
SD
SD
SS
SS
SS
SS
ss
AC 20
AC 20
AC 20
AC 67
AC 67
AC 67
AC 100
AC 200
AC 200
AC 67
AC 67
AC 67
AC 67
AC 67
AC 67
AC 67
AC 67
AC 67
DC 20
DC 20
DC 20
DC 67
DC 67
10
33
100
3
10
30
2
3
10
3
10
30
3
10
30
3
10
30
10
33
100
3
10
1.00E-01
1.11E-01
1.04E-01
6.10E-02
1.86E-01
1.94E-01
6.62E-02
9.94E-02
3.44E-01
2.00E-02
5.48E-02
1.55E-01
7.93E-02
2.37E-01
2.05E-01
8.46E-02
1.11E-01
1.19E-01
1.00E-01
1.09E-01
1.03E-01
6.10E-02
1.85E-01
7.06E-02
9-32E-02
8.68E-02
4.89E-02
1.25E-01
1.59E-01
3.90E-02
7.00E-02
2.23E-01
8.12E-03
3.76E-02
9.46E-02
5.52E-02
2.13E-01
1.90E-01
6.72E-02
8.56E-02
9.29E-02
7.08E-02
9.18E-02
8.63E-02
4.97E-02
1 .25E-01
-3.8
0.1
3.6
-0.8
-4.2
0.4
-30.3
13.1
-3.1
0.0
9.6
2.7
-26.7
-2.4
7.8
-3.2
-4.2
-3.1
-4.4
0.0
3.4
-1.7
-5.1
-4.5
-0.7
2.4
-7.7
-6.3
-0.7
-19.2
-4.4
-6.6
-19.2
4.8
3.3
-26.8
-5.8
8.1
-2.6
-3.2
-1.6
-5.3
-1.0
2.1
-9.8
-7.7
-1.4
-0.3
0.9
-1.7
-3.5
-0.5
-3.4
-1.4
-6.6
-0.7
0.8
1.4
-6.7
-5.5
6.9
-0.8
-1.2
-0.7
-1.7
-0.4
0.8
-2.2
-4.3
4.6
1.5
3.2
7.9
6.5
2. 1
21.1
18.8
6.9
21.1
10.4
3.4
31.0
5.9
7.9
2.6
3-3
1.7
5.5
2.0
3.3
10.2
8.0
-2.1
2.0
4.5
0.7
-2.0
2.3
-30.1
13.8
-0.7
0.0
10.3
2.8
-22.5
-0.3
9.6
-1.7
-1.4
0.2
-2.4
2.3
4.8
0.1
-2.4
-2.6
1.4
3.7
-5.5
-3.2
1.6
-17.0
-1.6
-2.8
-17.0
5.8
3.5
-22.7
-3.7
9.9
-1.6
-1.3
0.6
-3.1
1.6
4.0
-7.1
-3.8
-0.8
0.6
1.5
-1.2
-1.8
1. 1
-3.0
-0.5
-2.8
-0.6
1.0
1.5
-5.6
-3.6
8.5
-0.5
-0.5
0.2
-1.0
0.6
1.5
-1.6
-2.1
2.7
1.7
3.6
6.1
3.3
2.3
18.4
17.6
2.8
18.4
10.6
3.7
25.6
3.8
9.5
1.6
1.3
0.6
3.1
2.1
3.9
7.3
3.9
(continued)
-------
Table 41 (continued) - 2
u>
o
KJ
Test Problem (a)
Env Mix NOx C/N
Standard Cone (ppm)
Maximum Average
._. RADM-M VS RADM (b) ---
Max. - Average - %
JChg. JChg. %/Max Fit
RADM-P vs RADM (b)
Max. - Average - %
JChg. %/Max Fit
ss
ss
ss
ss
ws
ws
ws
SA
SA
SA
SD
SD
SD
SS
SS
SS
SS
SS
SS
SS
SS
ss
ws
ws
ws
DC 67
DC 100
DC 200
DC 200
DC 67
DC 67
DC 67
DC 67
DC 67
DC 67
DC 67
DC 67
DC 67
BT 20
BT 20
BT 20
BT 67
BT 67
BT 67
BT 100
BT 200
BT 200
BT 67
BT 67
BT 67
30
2
3
10
3
10
30
3
10
30
3
10
30
10
33
100
3
10
30
2
3
10
3
10
30
1.89E-01
7.07E-02
9.65E-02
3.38E-01
2 . OOE-02
5.26E-02
1.50E-01
8.08E-02
2.32E-01
1.98E-01
8.50E-02
1.11E-01
1.17E-01
9.63E-02
1.16E-01
1.11E-01
6.36E-02
1.81E-01
2.11E-01
1.90E-02
1.22E-01
3.17E-01
2. OOE-02
5.12E-02
1.16E-01
1.51E-01
1.06E-02
7.07E-02
2.21E-01
8.52E-03
3.65E-02
9.25E-02
5.65E-02
2.10E-01
1.85E-01
6.76E-02
8.55E-02
9.20E-02
6.77E-02
9.60E-02
9.31E-02
1.51E-02
1.19E-01
1.70E-01
3.19E-02
6.82E-02
2.16E-01
6.19E-03
3.18E-02
8.80E-02
0.1
-31.8
11.7
-1.1
0.0
11.1
3.8
-32.6
-2.3
9.7
-1.1
-5.1
-3.9
-2.7
-0.2
2.6
0.6
-3.6
0.1
6.5
-31.2
-3.5
0.0
-11.1
-2.0
-1.2
-22.8
-6.5
-8.2
-21.0
8.3
1.5
-32.7
-6.1
9.1
-3.3
-3.8
-2.2
-3.0
-0.7
1.6
-1.0
-1.7
-0.7
-3.2
-29.3
-5.5
-9.6
-8.9
-2.3
-0.8
-1.2
-2.1
-8.1
-0.9
1.1
1.9
-8.3
-5.7
7.8
-1.0
-1.5
-0.9
-0.9
-0.3
0.7
-0.8
-2.5
-0.5
-0.5
-9.0
-5.1
-0.3
-1.1
-0.9
2.6
25.5
22.3
8.6
27.1
11.7
1.5
39.2
6.2
9.1
3.3
3.9
2.3
3.0
1. 1
2.5
1.6
1.8
1.5
9.8
31.0
5.7
10.0
9.3
2.1
2.5
-31-6
15.1
-1.0
0.0
11.8
1.0
-27.8
0.5
12.1
-2.3
-1.8
0.2
-1.7
1.2
3.6
1.2
-2.5
1.5
6.5
-25.3
-2.3
0.0
-13.2
-2.5
1.7
-20.2
-3.2
-3.3
-21.3
9.5
1.8
-27.9
-3.1
11.9
-2.1
-1.5
0.6
-1.9
0.8
2.9
-2.9
-3.1
0.9
-2.3
-21.1
-3.5
-8.1
-8.2
-2.8
1.2
-3.7
-1.0
-3.3
-0.8
1.6
2.0
-7.1
-3.2
9.9
-0.6
-0.6
0.2
-0.6
0.3
1.2
-0.6
-1.7
0.7
-0.3
-7.1
-3.1
-0.2
-1.3
-1.1
2.5
22.3
20.6
3.1
23.7
11.8
1.9
32.5
3.8
11.3
2.1
1.5
0.6
1.9
1.1
2.8
1.3
3.2
1.5
8.9
27.1
3.5
8.7
8.5
2.9
(continued)
-------
Table 41 (continued) - 3
o
OJ
Test Problem (a) Standard Cone (ppm) RADM-M vs RADM (b) RADM-P vs RADM (b)
Env Mix NOx C/N Maximum Average Max. - Average - % Max. - Average - %
£Chg. %Chg. £/Max Fit £Chg. £Chg. £/Max Fit
SA
SA
SA
SD
SD
SD
BT
BT
BT
BT
BT
BT
67
67
67
67
67
67
3
10
30
3
10
30
5.95E-02
2.33E-01
2.23E-01
8.37E-02
1.10E-01
1.24E-01
4.00E-0?
1.98E-01
2.07E-01
6.62E-02
8.45E-02
9.56E-02
-17.0
-2.3
6.9
-2.2
-2.8
-2.3
-19.1
-5.8
5.8
-1.8
-2.2
-1.3
-3.4
-5.1
5.4
-0.5
-0.8
-0.6
21.1
5.9
5.8
1.8
2.2
1.4
-15.4
-0.9
7.9
-1.4
-1.1
0.0
-17.0
-4.6
6.8
-1.2
-1.0
0.2
-3.0
-4.1
6.3
-0.4
-0.4
0.1
18.6
4.7
6.7
1.2
1.0
0.2
(a) See Footnote (a) in Table 37 for a description of the notation used to indicate the test
problems.
(b) See Section 6.2 for a discussions of the specific measures of differences between the
mechanisms.
-------
Table 12. Results of Sensitivity Test Calculations of Urban HpOp Formation Using the Recommended
Modified (RADM-M) and the Recommended Modified Parameter (RADM-P) Mechanisms, Relative
to the March 1988 RADM Mechanism as the Standard
Test Problem (a)
Env Mix NOx C/N
Standard Cone (ppm)
Maximum Average
... RADM-M vs RADM (b) ---
Max. - Average - %
JChg. £Chg. 2/Max Fit
RADM-P vs RADM (b)
Max. - Average - %
*Chg. tChg. */Max Fit
ss
ss
ss
ss
ss
ss
ss
ss
ss
ws
ws
ws
SA
SA
SA
SD
SD
SD
SS
SS
SS
SS
SS
AC 20
AC 20
AC 20
AC 67
AC 67
AC 67
AC 100
AC 200
AC 200
AC 67
AC 67
AC 67
AC 67
AC 67
AC 67
AC 67
AC 67
AC 67
DC 20
DC 20
DC 20
DC 67
DC 67
10
33
100
3
10
30
2
3
10
3
10
30
3
10
30
3
10
30
10
33
100
3
10
2.62E-03
5.15E-03
1.17E-02
1.91E-03
1.11E-03
1.22E-02
1.77E-03
3.12E-03
8.35E-03
1.80E-07
1.06E-03
1.26E-03
2.19E-07
1.56E-03
1.18E-02
1.71E-03
3.76E-03
7.11E-03
2.65E-03
5.37E-03
1.10E-02
1.98E-03
1.15E-03
1.86E-03
3.86E-03
8.16E-03
9.12E-01
3.12E-03
8.56E-03
7.61E-01
1.57E-03
5.90E-03
2.92E-08
6.27E-01
3.03E-03
9.15E-08
1.08E-03
1.20E-02
1.03E-03
2.29E-03
1.67E-03
1.89E-03
3.81E-03
7.68E-03
9.28E-01
3.19E-03
-6.1
-5.0
-2.3
-4.8
-11.3
-5.7
-12.6
-18.7
-15.0
-81.9
-27.7
-11.1
-77. 1
-32.8
-16.1
-7.5
-8.2
-6.9
-7.5
-5.5
-3.0
-7.1
-13.3
-8.1
-6.1
-3.8
-10.8
-13.7
-7.3
-13.9
-29.1
-16.9
-61.8
-33.0
-12.8
-58.9
-11.7
-15.3
-7.2
-8.0
-6.1
-9.1
-7.1
-1.7
-12.7
-15.6
-1.3
-2.1
-2.6
-0.8
-3.6
-5.2
-0.9
-3.8
-8.3
0.0
-1.7
-3.2
0.0
-1.0
-15.3
-0.6
-1.5
-2.5
-1.1
-2.3
-3.0
-1.0
-1.2
8.1
6.7
1.0
11.3
11.7
7.7
11.9
33.6
18.5
97.0
39.1
13.7
83.9
57.2
16.6
7.1
8.3
6.7
9.5
7.7
1.9
13.5
16.9
-5.5
-6.0
-3.8
-3.6
-9-8
-7.5
-12.5
-16.2
-12.3
-80.5
-29.2
-13.7
-75.6
-29.7
-16.2
-5.8
-6.5
-5.8
-7.5
-7.0
-1.7
-5.5
-11.1
-7.8
-6.7
-5.0
-9.3
-12.7
-8.1
-12.9
-23.0
-15.1
-65.2
-33.8
-15.5
-51.5
-12.0
-15.2
-6.5
-7.2
-6.1
-9.1
-7.6
-5.8
-11.1
-11.1
-1.2
-2.2
-3.1
-0.7
-3-3
-6.0
-0.8
-3.0
-7.6
0.0
-1.8
-3.9
0.0
-3.8
-15.2
-0.6
-1.1
-2.1
-1.5
-2.1
-3.7
-0.9
-3.8
8.0
6.9
5.1
9.6
13.6
8.8
13.7
25.6
16.6
97.6
10.6
16.8
75.3
52.9
16.1
6.8
7.1
6.3
9.8
7.9
6.0
11.7
15.5
(continued)
-------
Table 12 (continued) - 2
U)
o
Ln
Test Problem (a)
Env Mix NOx C/N
Standard Cone (ppm)
Maximum Average
___ RADM-M vs RADM (b) --
Max. - Average - %
%Chg. %Chg. %/Max Fit
___ RADM-P vs RADM (b)
Max. - Average - %
%Chg. %Chg. J/Max Fit
ss
ss
ss
ss
ws
ws
ws
SA
SA
SA
SD
SD
SD
SS
SS
SS
SS
SS
SS
ss
ss
ss
ws
ws
ws
DC 67
DC 100
DC 200
DC 200
DC 67
DC 67
DC 67
DC 67
DC 67
DC 67
DC 67
DC 67
DC 67
BT 20
BT 20
BT 20
BT 67
BT 67
BT 67
BT 100
BT 200
BT 200
BT 67
BT 67
BT 67
30
2
3
10
3
10
30
3
10
30
3
10
30
10
33
100
3
10
30
2
3
10
3
10
30
1 . 18E-02
1.85E-03
3.17E-03
8.16E-03
3.08E-07
1 . 08E-03
1.09E-03
2.59E-07
1.67E-03
1.16E-02
1.77E-03
3.80E-03
7.36E-03
2.11E-03
1.96E-03
1.09E-02
1.90E-03
3.96E-03
1.09E-02
1.60E-03
2.83E-03
7.29E-03
1.71E-08
7.11E-01
3.10E-03
8.32E-03
8.11E-01
1.60E-03
6.08E-03
3.89E-08
6.55E-01
2.96E-03
1.02E-07
1.22E-03
1.18E-02
1.05E-03
2.32E-03
1.66E-03
1.65E-03
3.51E-03
7.61E-03
8.08E-01
2.51E-03
7.62E-03
6.67E-01
8.98E-01
1.17E-03
6.31E-09
3.57E-01
2.23E-03
-6.5
-16.8
-20.9
-17.9
-85.3
-28.8
-13.3
-81.9
-10.6
-19.7
-9.1
-9-5
-8.1
-4.5
-1.1
-2.1
-2.3
-8.3
-5.2
0.7
-25.1
-11.7
-12.6
-12.0
-1.1
-8.5
-19.5
-30.9
-19.1
-72.9
-33.5
-11.9
-67.2
-53.8
-18.5
-8.3
-9.1
-7.7
-6.2
-5.0
-3.6
-12.1
-11.8
-6.1
-8.3
-65.2
-16.2
-31.1
-16.5
-6.2
-5.9
-1.3
-1.1
-9.8
0.0
-1.8
-3.7
0.0
-5.5
-18.1
-0.7 .
-1.8
-3.0
-0.9
-1.5
-2.3
-0.8
-2.5
-1.1
-0.5
-1.9
-6.0
0.0
-0.5
-1.1
9.0
21.1
36.1
21.5
115.8
10.1
16.1
101.7
73.3
20.3
8.7
9.5
8.0
6.1
5.2
3.7
12.6
12.5
6.6
10.1
91.8
17.6
11.3
17.9
6.1
-8.8
-17.1
-18.3
-11.5
-89.3
-30.8
-16.9
-81.1
-36.9
-19.6
-7.5
-7.7
-7.0
-1.5
-1.1
-2.7
-2.5
-7.8
-5.9
0.9
-22.3
-10.1
-31.7
-12.7
-1.6
-9.8
-18.8
-23.6
-17.1
-73.7
-31.9
-18.6
-61.6
-51.1
-18.2
-7.8
-8.0
-7.2
-6.2
-1.8
-3.1
-11.3
-11.7
-6.1
-7.1
-63.1
-15.2
-29.0
-16.9
-7.0
-6.8
-1.3
-3.1
-8.8
0.0
-1.9
-1.6
0.0
-5.2
-17.8
-0.7
-1.6
-2.8
-0.9
-1.1
-2.2
-0.8
-2.5
-1.1
-0.1
-1.7
-5.7
0.0
-0.5
-1.3
10.3
20.6
26.5
19.0
117.9
12.1
20.1
96.0
68.3
19.9
8.1
8.1
7.1
6.1
1.9
3.5
11.8
12.1
6.7
8.9
90.8
16.1
31.6
18.1
7.3
(continued)
-------
Table 42 (continued) - 3
U)
o
Test
Env
SA
SA
SA
SD
SD
SD
Problem (a)
Mix NOx C/N
BT
BT
BT
BT
BT
BT
67
67
67
67
67
67
3
10
30
3
10
30
Standard Cone (ppm)
Maximum Average
4.36E-08
1.14E-03
1.20E-02
1.67E-03
3.54E-03
6.96E-03
3.47E-08
6.21E-04
9.78E-03
9.66E-04
2.11E-03
4.31E-03
._. RADM-M vs RADM (b) ---
Max. - Average - %
*Chg.
-30.7
-17.5
-13.5
-6.0
-6-.0
-5.0
*Chg.
-37.4
-34.7
-13.5
-5.4
-5.6
-4.8
J/Max
0.0
-1.8
-11.0
-0.4
-1.0
-1.7
Fit
46.2
41.7
14.4
5.5
5.8
4.9
_ RADM-P vs RADM (b)
Max. - Average - %
*Chg.
-30.2
-16.5
-13.2
-5.2
-5.0
-4.1
JChg.
-36.0
-34.8
-13.0
-5.2
-5.3
-4.4
*/Max
0.0
-1.8
-10.6
-0.4
-0.9
-1.6
Fit
44.1
41.8
13.9
5.4
5.4
4.5
(a) See Footnote (a) in Table 37 for a description of the notation used to indicate the test
problems.
(b) See Section 6.2 for a discussions of the specific measures of differences between the
mechanisms.
-------
OZONE
0.005 -,
HYDROGEN PEROXIDE
0.04-
0.00'
l ' I ' I ' I ' I ' I ' I ' I ' I ' 1
5 11 17 23 29 35 41 47 53 59 65
n ' i * i * i » i » i * i * i
5 11 17 23 29 35 41 47 53
i ' i
59 65
Q.
a
O
h-
UJ
O
O
O
0.003 -i
0.002 -
0.001 -
SULFATE
0.000
5 11 17 23 29 35 41 47 53 59 65
FORMIC ACID
0.000'
I i I r I . j T y^T ' I ' I '
5 11 17 23 29 35 41 47 53
METHYL HYDROPEROX1DE
I ' I '
59 65
T ' I ' I ' I ' I ' I ' I ' 1
5 11 17 23 28 35 41 47 53 59 65
I ' I ' I ' I ' 1 ' I ' I
5 11 17 23 29 35 41 47 53
I ' I
59 65
HOURS AFTER MIDNIGHT
Figure 89. Concentration-Time Plots for Selected Species Calculated
Using the RADM, RADM-M, and RADM-P Mechanisms for the Summer
Surface Static Urban Test Problem with Initial NOX of 6? ppb
and a ROG/NOV Ratio of 10.
A
307
-------
OZONE
0.0012-1
0.0009-
0.0006-
0.0003 -J
HYDROGEN PEROXIDE
0.0000
17 29 41 53 65 77 89 101 113
SULFATE
5 17 29 41 53 65 77 89 101 113
NITRIC ACID
I ' I ' ' I ' 'T T M ' ' 1 ' ' I ' ' I ' ' I
17 29 41 53 65 77 89 101 113
FORMIC ACID
17 29 41 53 65 77 89 101 113
METHYL HYDROPEROX1DE
l l ' i i ' ' l ' ' l ' ' l ' ' l 'T T
17 29 41 53 65 77 89 101 113
17 29 41 53 65 77 89 101 113
HOURS AFTER MIDNIGHT
Figure 90. Concentration-Time Plots for Selected Species Calculated
Using the RADM, RADM-M, and RADM-P Mechanisms for the Winter
Surface Static Urban Test Problem with Initial NOX of 67 ppb
and a ROG/NOX Ratio of 10.
308
-------
where the predictions are most different for the other peroxides and the
acid species. On the other hand, the modifications have much less of a
consistent effect on predictions of nitric acid and sulfate, with positive
as well as negative changes, and fewer cases with relatively large differ-
ences being observed.
As expected, the differences between the two versions of the recom-
mended modified mechanisms are much less than the differences between the
modified and the March 1988 mechanisms. The differences between the
modified mechanisms in predictions of ozone, H^Op, sulfate, HNO?, and the
organic acids are generally minor, with discrepancies only being observed
in predictions of peroxy or acid species in a few cases where their
calculated concentrations are very low. There are some cases of non-
negligible differences in predictions of the organic hydroperoxide
species, peroxyacetic acid, and PAN., Since PAN is not represented
explicitly in any of these mechanisms, and since there are no chamber data
available to test predictions of organic hydroperoxide or acid species,
there is no way of determining which of the two modified mechanisms is the
more reliable in their predictions of these species. In view of the many
chemical uncertainties in the mechanisms regarding reactions forming these
species and the sensitivity of predictions to relatively small differences
in the mechanisms, predictions for these species in regional model simula-
tions should be viewed as highly uncertain.
These results indicate that the recommended modifications to the RADM
mechanism may have a non-negligible impact on predictions of species of
interest in regional modeling applications. Because the modified versions
perform somewhat better in simulating the aro.raatic-NOx-air chamber runs,
predictions obtained with these mechanisms probably should be considered
to be somewhat more reliable than those obtained with the March 1988
mechanism. On the other hand, it probably does not matter which of the
two modified versions is adopted, except that the RADM-M version employs
fewer species and thus is more efficient. The magnitude of the differ-
ences between the modified and the March 1988 mechanisms also gives us an
approximate indication of the sensitivity of regional model predictions to
the uncertain aspects of aromatics photooxidation chemistry, since that is
the major aspect of the mechanism that was changed.
309
-------
7. SUMMARY AND CONCLUSIONS
One of the more significant advances in regional acid deposition
modeling in the 1980s has been the incorporation of gas phase chemical
mechanisms that predict the formation of sulfate, nitric acid, ozone,
organic nitrates, hydrogen peroxide, organic peroxide, and organic acids
from emissions of S02, NOX, and reactive organic gases (ROG). The gas-
phase chemical mechanism is a particularly important component of regional
acid deposition models because it controls the rates of radical and
oxidant generation, which in turn controls the production rates of acidic
species. The chemical mechanism is also the major potential source of
nonlinearity in the relationships between acid deposition and emissions of
SOo and NOX. As such, it is one of the major reasons why control
strategies for acid deposition cannot rely on analyses based on simple
rollback, but require use of complex models such as RADM. Thus, it is
particularly important that the gas-phase chemical mechanism used in such
models be evaluated as comprehensively as possible, to assure that it
represents the current state-of-the-science of atmospheric chemistry. In
addition, since the RADM model is to be used for policy development, its
major components such as the gas-phase chemical mechanism must be accepted
as representing the best science by not only the technical community but
by the public as a whole.
Peer review is the method presently used to assure the quality of
scientific research. The essence of peer review is that the research be
evaluated by experts in the field other than those who developed the
research being considered. It is required before results are published in
the most reputable scientific journals, and agencies such as the EPA
periodically have peer review panels to assure the continuing quality of
the research they support. Therefore, peer review is clearly an appro-
priate means to assure the quality and credibility of important components
of the RADM model such as the gas-phase chemical mechanism. However, peer
review as it is normally practiced does not lend itself to in-depth
assessment of highly complex model components such as gas-phase atmos-
pheric reaction mechanisms. Journal space is much too limited to permit
in-depth discussion of the many components involved in such mechanisms,
and thus they cannot be evaluated adequately by the reviewers. Peer
310
-------
review panels of agency research programs generally have relatively
limited time to assess very broad areas of research and usually cannot be
expected to carry out in-depth analyses of individual components. The
study described in this report, however, is an in-depth peer review of an
important model component the RADM gas-phase mechanism. The authors of
this report have been involved for years in the development and evaluation
of gas-phase mechanisms for use in airshed models, but had not parti-
cipated in the development of the RADM gas-phase mechanism. Because of
the many aspects involved in evaluating a chemical mechanism, including
the need to test it against the results of over 500 environmental chamber
experiments, this review took more than a year.
The results of this review indicated that for the most part the RADM
mechanism represents the state of our present knowledge of atmospheric
chemistry and performs as well as reasonably can be expected in simulating
the results of environmental chamber experiments. However, the state of
knowledge of atmospheric chemistry is constantly advancing, and over the
course of this study, several sets of modifications to the RADM mechanism
and its interface to the emissions inventory were identified and
recommended. These recommendations and areas of suggested improvement are
described in detail in the body of this report and are briefly summarized
below. These recommendations do not involve major changes to the basic
nature of the mechanism as conceived by its developers, but instead should
be considered as refinements and enhancements to an already high quality
product. However, we believe that the implementation of these recommenda-
tions will result in improvements in representation of several important
aspects of the atmospheric chemistry of organic compounds.
As stated in the preface to this report, the developers of the RADM
model cooperated fully with this review and actively participated in some
of its elements. The principal developer of the RADM gas-phase mechanism
kept us fully informed of the exact status of the mechanism as it under-
went changes, evaluated our recommendations, and worked directly with us
in those aspects of this evaluation where his participation proved most
valuable. He and other members of the RADM team actively solicited our
participation in the development of the emissions processing and aggrega-
tion methods, re-directing our efforts to this critical aspect of the
implementation of the mechanism in the model. Without the high degree of
311
-------
interest and cooperation by the RADM team, this program could not have
been nearly as successful as we believe it was. Most of the major
recommendations developed during this evaluation were accepted for imple-
mentation in the RADM-JI mechanism.
It should be emphasized that although we believe that the resulting
RADM gas-phase mechanism and its associated emissions processing system
represents the current state-of-the-science, this does not mean that there
are no significant areas of uncertainty. There remain many important
aspects of the gas-phase atmospheric reaction mechanisms of organics which
are unknown and where basic laboratory research is still needed before
reliable mechanisms representing them can be developed. The available
environmental chamber data are not suitable for testing many critical
aspects of the mechanism, particularly those of the peroxy and acid
species which are of concern in acid deposition models. In addition, the
need to represent the highly complex chemistry of the many hundreds of
organic compounds emitted into the atmosphere with mechanisms using only
limited numbers of species and reactions introduces uncertainties in model
predictions whose effects are not fully understood. This is in addition
to the uncertainties in the emissions inventories themselves, which
particularly for biogenic emissions often dwarf the effects of any
uncertainties in the chemical mechanism. Thus, users of regional models,
even those such as RADM that contain the most up-to-date, thoroughly peer-
reviewed gas-phase chemical mechanisms, must view their predictions with a
certain amount of skepticism and with an appreciation of the magnitudes of
the various uncertainties involved.
The scope of work for the study consisted of four major tasks. The
research findings from each task are summarized below. Following that, we
briefly summarize our recommendations for future work.
7.1 Evaluation of Kinetic and Mechanistic Parameters
At the time this study was initiated, the RADM mechanism was in a
state of transition between the RADM-I and RADM-II versions of the mech-
anism. An initial review of the preliminary RADM-II mechanism was
performed, and a number of changes in both the inorganic and organic reac-
tions in the mechanism were recommended. A procedure was developed for
determining rate constants and products for the reactions of lumped
312
-------
organic species based on the detailed speciation of the national emissions
inventory. Many of these initial recommendations were adopted and
incorporated into the March 1988 version of the RADM-II mechanism
(referred to as the RADM). The comprehensive evaluation was carried out
for this version of the RADM mechanism.
On the basis of testing the RADM mechanism against environmental
chamber data, two alternate versions of the mechanism were developed and
fully evaluated. The first alternate mechanism, referred to as the
recommended modified RADM mechanism (RADM-M), offers the following
features:
Changes in the representation of the unknown aspects of the
aromatic photooxidation mechanism which improve its perform-
ance in simulating chamber data and reduce the number of
species needed;
A more condensed representation of the >C3 alkanes which has
comparable performance with fewer species;
Some changes in rate constant and product yield parameters
for lumped species to correspond more closely with the most
recent analysis of national VOC emissions data;
Correction of an error in the methylglyoxal quantum yields
which resulted in overestimation of its photolysis rates;
Addition of several reactions omitted from the NC^ + alkene
reaction system; and
Revised representation of the reactions of nitrophenols
which improves the chemical fidelity of the reactions.
Because implementation of the RADM mechanism was almost completed by
the time the RADM-M was developed and evaluated, a second alternate mech-
anism was proposed that involved primarily changes in rate constants and
313
-------
product yields. This version is referred to as the recommended modified
parameter version of the RADM (RADM-P). This incorporates:
An improved aromatic oxidation mechanism which performs
approximately as well as the RADM-M mechanism, but uses the
larger number of species employed in the RADM mechanism;
Some changes in rate constant and product yield parameters
for lumped species to correspond more closely with the most
recent analysis of national VOC emissions data;
Reduced methylglyoxal quantum yields; and
Addition of several reactions omitted from the NO? + alkene
reaction system.
The RADM developers have adopted the modified parameter version of the
mechanism and this version is now being implemented into the RADM trans-
port model (Stockwell, private communication, 1989).
7.2 Development of the Linkage Between the VOC Emissions Inventory and
the RADM Mechanism
In recent years, modelers have begun to appreciate fully the impor-
tance of carefully interfacing a chemical mechanism with VOC emission
inventories. At the start of this program, only a very preliminary
emission inventory interface had been developed for the RADM mechanism.
At the request of the RADM developers, a large effort in this program was
directed towards developing emissions aggregation procedures for use with
the RADM model. A new two-step emissions aggregation system was developed
for this purpose. The first step involved the aggregation of the detailed
VOX; emissions data into a 32-class mechanism-independent VOC grouping
system based on reactivity and relative contributions to total emissions.
The second step involved further aggregating these emissions groups into
the more limited number of VOC species used in the RADM mechanism. This
effort was carried out jointly as part of this program and as part of the
RADM development effort.
314
-------
The assignment of species in the emissions groups to the surrogate
species used in the RADM mechanism involves use of a new reactivity weigh-
ing scheme. This scheme allows VOCs reacting at different rates, but with
similar mechanisms, to be represented by the same model species. This is
based on estimation of how much of the VOCs and the model surrogate under-
go chemical reaction in the model simulation, and adjusting the amounts of
the latter so they are approximately equal. Although approximate, the use
of reactivity weighing is preferable to the alternatives of either ignor-
ing compounds of low but non-negligible reactivity, or representing them
by much more reactive model surrogates.
Another feature of this aggregation system is the use of detailed
emissions data to derive the rate constants and product yield coefficients
of lumped species in the mechanism based on those for the aggregate of
compounds they will be representing. A software system recently developed
as part of another program and the 1985 NAPAP anthropogenic emissions
inventory were used for this purpose. This is considered a significant
advancement that optimizes the linkage between the chemical mechanism and
the emissions inventory.
7.3 Mechanism Evaluation Against Chamber Data
An important step in the model design and evaluation process is
evaluation of the gas phase chemistry against environmental chamber
data. The gas phase chemistry module is one of the few modules in the
atmospheric modeling system that can be independently tested against
experimental data. The present chamber data base, however, is not suit-
able for testing all aspects of the mechanism, particularly those involv-
ing predictions of acid or peroxy species or simulations of very low NOV
A
conditions that prevail in rural regions. The present chamber data base
is best suited for testing the mechanism's predictions of ozone formation
and rates of NOX oxidation under urban conditions. To a lesser extent,
the data can be used to test the mechanisms' ability to simulate formation
of peroxyacetyl nitrate and formaldehyde under urban conditions.
Simulation of urban-like conditions is important because urban areas
represent the major sources of the pollutants which are transported to
remote areas with sensitive receptors. Ozone, NOX, PAN, and formaldehyde
play critical roles in long-range transport.
315
-------
It is also important to recognize that even though the present data
base does not allow for direct evaluation of mechanisms' predictions for
sulfuric acid and nitric acid, the testing protocol provides some indirect
evidence for evaluation of performance for these species. First, accurate
prediction of the rate of NOX oxidation and ozone formation necessitate
accurate predictions of OH radical concentrations. According to present
theory, OH is the sole oxidant converting $©2 to sulfuric acid and NOp to
nitric acid in the daylight hours. Second, since NOV is primarily
A
oxidized to nitric acid or PAN, the evaluation results for PAN provide
some information on whether the mechanism is oxidizing NOX to the correct
products. In general, if the model cannot at least simulate ozone and NOX
correctly in one-day smog chamber experiments, it certainly cannot be
trusted to simulate accurately concentrations of these and other species
under multi-day, long-range transport conditions.
A comprehensive evaluation of the RADM mechanism and the recommended
modified versions of RADM mechanism against available environmental
chamber data was performed in this study. The approach involved simula-
ting a large number (-550) of experiments from four different chambers and
statistically evaluating the mechanisms' performance. Experimental data
collected in the following chambers was used for the evaluation:
The SAPRC evacuable indoor chamber (EC),
The SAPRC indoor Teflon chamber (ITC),
The SAPRC outdoor Teflon chamber (OTC), and
The University of North Carolina outdoor chamber (UNC).
Prior to conducting the evaluation, the procedures for characterizing
chamber dependent processes were reviewed and updated where new data were
available. A consistent set of procedures for representing these was used
for all experiments from a given chamber. Organic compounds in experi-
ments employing complex mixtures of organics were aggregated using the
types of procedures which will be employed when the mechanism is used in
RADM simulations. Thus there was no run-to-run adjust of chamber-
dependent or mechanistic parameters employed in the evaluation protocol.
The major findings from the evaluation of the RADM, RADM-M, and
RADM-P mechanisms against environmental chamber data are as follows.
316
-------
Testing against experiments using complex organic mixtures, that
are surrogates for atmospheric VOC mixtures, showed that the RADM mechan-
ism performs as well as other chemically up-to-date mechanisms (Lurmann et
al., 1987; Gery et al., 1988; Carter, 1988). The mechanism is able to
predict the maximum ozone and NOX oxidation rate with a small bias and a
mean error of 2Q% on the average. Even though the good performance on
complex mixtures can in some cases be a result of partially compensating
errors in the mechanism, the mechanism does provide ozone and NOX oxida-
tion rate predictions that are consistent with the available data. The
performance of the two alternative versions of the RADM are comparable to
that for the RADM on complex mixture simulations.
Testing of the RADM mechanism against single organic-NOx-air
experimental data produced mixed results. With a few notable exceptions,
the RADM mechanism performed as well as can reasonably be expected against
experiments with the species for which the mechanism was designed. The
areas where the predictions of the mechanism can be considered to be
comparable to those of other current mechanism are as follows:
The predictions for experiments with formaldehyde and pro-
pene were reasonably good.
The predictions for acetaldehyde and methyl ethyl ketone
systematically underpredicted ozone and NOX oxidation rates,
but there are too few experiments to draw definitive conclu-
sions on the adequacy of this portion of the mechanism.
The mechanism has a bias towards underprediction of reac-
tivity in ethene runs. This is due to the omission of a
reaction which we found to be unimportant under conditions
where the model will be applied. If this reaction is added
to the mechanism, its performance in simulating these
experiments is acceptable.
The mean error in the ozone predictions in the 1-butene runs
are greater than is the case for propene, but the perform-
ance is considered acceptable.
The predictions for a limited number of trans-2-butene are
poor. However, there are too few of these experiments to
adequately test this portion of the mechanism.
317
-------
The mechanism had large errors in simulating results of
alkane runs. This is not necessarily due to any failure in
the mechanism, but to the extreme sensitivity of simulations
of alkane-NOx-air experiments to chamber effects which tend
to vary unpredictably from run to run. Compared to the
errors, the biases in the predictions of ozone and NO oxida-
tion rates were relatively low.
The performance of the RADM-M and RADM-P mechanisms in simulating
these experiments was similar. The only difference was that the RADM-M
mechanism performed slightly worse in simulating some of the higher alkane
runs because of its more condensed representation of these compounds.
The performance of the RADM mechanism in tests against single
aromatic species was not as good as that reported for other chemically up-
to-date mechanisms. The testing showed that the aromatic reactions in the
RADM mechanism have a bias towards overpredicting ozone and NOV oxidation
A
rates for the more slowly reacting compounds and underpredicting those
same characteristics for the faster reacting compounds. This results in
fairly large errors in ozone predictions on the average for toluene,
xylenes, and mesitylene.
The RADM-M and RADM-P mechanisms were designed to improve the
performance of the mechanism in simulating the toluene, xylene, and
mesitylene runs. This design objective was achieved. The performance of
these mechanisms in simulating the results of aromatic runs was comparable
to other current mechanisms.
Because of the importance of biogenic emissions in regional model
applications, the RADM mechanism has a separate species to represent
isoprene. However, the performance of this mechanism in simulating
isoprene experiments was not as good as those for some other alkenes.
This is probably because the mechanism assumes that isoprene reacts to
form the same photooxidation products that are formed when propene
reacts. Improving the performance of the RADM mechanism in this regard
would probably require adding additional species to the mechanism to
represent these products. Research in this area is still underway at our
laboratories, and thus no modifications in this area are recommended at
the present time. The RADM-M and RADM-P therefore have the same isoprene
mechanism as RADM and thus perform the same in simulating these
experiments.
318
-------
On the other hand, the RADM mechanism performed surprisingly well
in simulations of the limited number of a-pinene runs, despite the fact
that it does not represent this compound explicitly. The RADM-M and
RADM-P mechanisms perform similarly.
The RADM mechanism's performance for species other than those for
which it was designed, such as propionaldehyde, 1-hexene, benzene,
naphthalenes, and tetralin was generally poor. Much better simulations of
experiments employing those compounds can be obtained in mechanisms
containing species designed to represent them. However, the level of
performance of the RADM mechanism in this regard is certainly comparable
to what one would expect from other condensed mechanisms designed for use
in Eulerian atmospheric models. Satisfactory performance in simulating
these compounds would probably require adding new species to the mechan-
ism. This is probably not worthwhile given the relative contributions of
these compounds to overall VOC reactivity in most regional model applica-
tions.
The RADM, RADM-M, and RADM-P mechanisms do not predict maximum
formaldehyde concentrations as accurately as they predict maximum ozone
concentrations or NO oxidation rates. They tend to overpredict formalde-
hyde yields in alkene and alkane experiments and underpredict formaldehyde
yields in experiments with single aromatic compounds and organic mix-
tures. The mean errors associated with the formaldehyde predictions are
fairly large in alkane, aromatic, and organic mixture experiments. While
one would like to see better formaldehyde predictions, this does not
necessarily present a problem with respect to using the mechanisms for
long-range transport and acid deposition modeling, especially given the
uncertainty of the formaldehyde measurements.
The mechanism consistently overpredicts measured PAN concentra-
tions. This positive bias is expected since the species PAN is used in
the mechanism to represent PAN plus numerous analogous organic nitrates.
However, the magnitude of the bias, especially in the surrogate mixture
runs, suggests the mechanism overpredicts the PAN and analogous species
concentrations. This is of some concern for regional acid deposition
modeling because the model may underpredict nitric acid in the first few
100 km downwind of sources (i.e., if NOX is preferentially oxidized to PAN
rather than HNOo) and overpredict the distances for which nitrogen species
319
-------
are transported (since PAN is the long range transport product of NO
emissions).
Lastly, it is very unlikely that mechanisms could achieve the high
level of performance on ozone and NOX oxidation rates found in complex
mixture runs without accurately simulating the production and destruction
of the OH radical. These results indirectly suggest that the mechanisms
are able to reasonably predict the OH concentrations and, therefore, the
probable yields of nitric acid and sulfuric acid in daylight hours.
7.4 Sensitivity Testing of Alternate Mechanistic Assumption
Given the nonlinear couplings in mechanisms, it is often difficult to
predict a-priori the conditions and species for which changes in mechanis-
tic assumptions will have large effects. For this reason, it is particu-
larly important in evaluating alternative mechanistic assumptions to make
sure the testing is performed over a wide range of conditions. Testing of
alternative mechanistic assumptions employing an insufficiently comprehen-
sive range of conditions can often lead to misleading conclusions.
Unfortunately, such testing can not rely on environmental chamber data
alone because the chamber data do not cover all the species and conditions
of interest for acid deposition modeling. To overcome this problem, a
large number (-90) of test problems were developed in this program to
represent a wide range of chemical conditions for purposes of evaluating
selected mechanistic assumptions. Considerable thought and analysis went
into the design of the test problems, and we believe they will be as use-
ful in future mechanism evaluation programs as they have been in this
program.
Based on our review of the kinetic and mechanistic assumptions
employed in the RADM mechanism and the results of the evaluation against
chamber data, a limited number of alternative chemical assumptions were
identified for evaluation in sensitivity studies. The topics for investi-
gation included:
The effects of various levels of condensation of the peroxy-
peroxy radical reactions in the RADM mechanism;
The effects of condensation of the alkane reactions in the
RADM-M mechanism; and
320
-------
The effects of the modifications incorporated into the
RADM-M and RADM-P mechanisms on the predictions of species
of interest in regional acid deposition models.
Testing was conducted to compare the RADM mechanism to an alternate
version of the mechanism which included all possible peroxy-peroxy radical
combinations. The results showed that even for hydrogen peroxide and
organic peroxides, which are very sensitive to changes in the peroxy-
peroxy radical reactions, the RADM mechanism's treatment of peroxy
radicals is justified. Predicted concentrations of key species were
almost identical for all test problems with the RADM mechanism and the
detailed peroxy radical version of the RADM mechanism.
The method used to represent peroxy radical combinations in the RADM
mechanism is clearly satisfactory, and no changes in this regard are
recommended. However, tests were carried out to determine if a more
condensed representation of peroxy-peroxy radical reactions could be used
in the RADM model. The condensed representation did not yield predictions
which agreed as closely with those obtained with the detailed radical
mechanism, though the differences are probably not significant. The
benefits of changing to the slightly more condensed peroxy-peroxy radical
reactions scheme are not considered great enough to warrant the change.
Tests were carried out to determine if a more condensed representa-
tion of the reactions of alkanes could be employed in the RADM model. The
results indicated there are no significant differences in model predic-
tions if the two higher alkane classes in RADM are combined, provided the
combined class has the appropriate mechanistic parameters. These results
served as the basis for the representation of the alkanes in the
recommended RADM-M mechanism.
Test calculations indicated that the versions of the RADM mechanism
incorporating our recommended modifications (RADM-M and RADM-P) can in
some cases be quite different from the original mechanism in predictions
of species of interest in acid deposition models. The differences between
the mechanisms tend to be the greatest in the simulations of urban condi-
tions. These are due primarily to the recommended changes in the mechan-
isms for the aromatics. The modified mechanisms tended to predict
slightly lower levels of ozone and in some cases significantly lower
321
-------
levels of peroxide or acid species under some conditions. On the other
hand, the modifications had smaller effects on predictions of nitric acid
or sulfate. This indicates that implementing the recommended changes in
the mechanism might in some cases have non-negligible effects on results
of regional model simulations.
The test calculations also indicate that the two recommended modified
versions of the RADM mechanism are very close, but are not identical, to
each other in their predictions of species of interest in regional acid
deposition models. Their predictions of PAN and of the higher organic
hydroperoxides and acids differed in some cases by 25% or more. Differ-
ences in predictions of PAN are expected because of different degrees of
condensation in the two mechanisms PAN in RADM-M is used to represent
the PAN analogues formed from the unknown aromatic photooxidation
products, while RADM-P, like RADM, uses a separate species for these PAN
analogues. In the case of the predictions of higher organic acids and
hydroperoxides, the uncertainties concerning the rates and mechanisms of
their formation are so great and the data suitable to test models for
their predictions so inadequate that both mechanisms should probably be
considered to be equally unreliable in this regard. Discrepancies on the
order of 25% or even 50% are well within the uncertainty bounds for
predictions of such species. Therefore, for all practical purposes the
predictions of these two mechanisms can be considered to be essentially
equivalent.
7.5 Recommendations
As indicated above, although we believe that the RADM gas-phase
chemical mechanism, along with its emissions aggregation system, repre-
sents the current state-of-the-science, there remain significant uncer-
tainties in this important model component. Therefore, continuing
research is needed in areas related to its development and implementa-
tion. Our specific recommendations in this regard are given below. The
longer term critical research needs will be discussed first, followed by a
discussion of specific shorter-term research needs which became evident
during the course of this program.
322
-------
7.5.1 Longer Term Research Needs
The only way the gas-phase chemical mechanisms in airshed
models can be tested in their entirety, but independently of the many
other uncertainties in the models such as emissions, meteorology, etc., is
by use of environmental chamber data. The presently available chamber
data are suitable only for testing the ability of models to predict ozone
formation and NOX oxidation under the relatively high-NOx conditions
representative of the most polluted urban atmospheres. Even for this
purpose, there are problems due to poorly characterized chamber effects.
There is a critical need for chamber data suitable for testing mechanisms
in the low-NOx concentration regime important in regional and global
models, and for testing their abilities to predict formation of peroxide
and acidic species. Chamber wall effects are such that usable data of
this type cannot be obtained using existing environmental chamber tech-
nology. This need can only be addressed by first carrying out research
into new environmental chamber technologies and associated surface
effects. This would require a significant commitment of resources.
Unless such a commitment for long-term research is made, the present
unsatisfactory situation will not change significantly.
There still remain many critical areas of uncertainties in our under-
standing of, and thus our ability to model, the gas-phase atmospheric
chemistry of many classes of organics. The uncertainties associated with
the chemistry of aromatics have been dwelt upon in "Recommendations"
sections in reports such as this for at least a decade. The modifications
we recommended to the RADM mechanism in this area are based on empirical
curve-fits to chamber data which are largely outside the concentration
regime even of urban conditions, and we have no way of knowing how
accurately this largely empirical model will perform when extrapolated to
ambient conditions. The situation with the higher alkanes only seems to
be better because the present chamber data are unsuitable for indicating
Just how good or bad the present mechanisms really are. An additional
area which is now recognized as being important is the mechanisms of the
biogenic hydrocarbons. Recent work concerning isoprene indicates that its
OH reaction mechanism is not as well understood as we once thought it was
roughly ^0% of the reaction routes are not accounted for by identified
products. In addition, there are large and significant uncertainties in
323
-------
the mechanism for its reaction with ozone. Even less is known about the
terpenes the fact that the RADM mechanism happens to be able to
simulate the results of four a-pinene outdoor chamber experiments carried
out within a one-month period is probably largely coincidental. It is
unknown how well the model can simulate the reactions of other biogenic-
ally emitted species such as beta-pinene or other terpenes.
In some cases, important uncertainties can be addressed by applying
presently available techniques to compounds which have not been adequately
studied previously. An example of this concerns the non-isoprene
biogenics, which is discussed in the next section. However, for compounds
such as the aromatics, the higher alkanes, and even to some extent
isoprene, we have probably approached the point where the experimental
techniques traditionally used to study reaction mechanisms are unlikely to
yield significant new information. It is clear that the aromatics and
many other compounds of interest form products which cannot be quanti-
tatively measured using present techniques. Many of these compounds
appear to be highly labile and contribute significantly to the overall
reactivities of the emitted organics. In the case of the higher alkanes,
polyfunctional compounds are formed which probably do not remain in the
gas phase. Until research laboratories interested in atmospheric reac-
tions of organics have the type of equipment necessary to measure and
study such compounds, significant further progress in this area is
unlikely. For example, tandem MS equipment appears to be promising in
this regard, but few research laboratories can afford such equipment, and
those that have them apparently cannot afford to dedicate them to the long
periods of time required for the types of fundamental research projects
that are needed. Research into new types of equipment, and means of
making them available for researchers for use in fundamental studies, is
also needed. This is another area where significant new resources are
needed if further progress is to be made.
The emissions inventories may not be as interesting scientifically as
the subjects discussed above, but they probably represent the greatest
area of uncertainty in the overall area of gas-phase chemistry in the
models. Even if the gas-phase chemical mechanism were perfectly charac-
terized, it would do us no good if we do not know the input rates of the
species whose reactions are being modeled. The biogenic inventories are
324
-------
particularly uncertain; order-of-magnitude errors in the present inventor-
ies may well be the case. Until they are better characterized, it may
well not be worth the effort to add new species in the model to better
represent them. The anthropogenic emissions inventories also have signif-
icant uncertainties, particularly for the VOCs. If there are major errors
in the anthropogenic inventory, it will not matter how accurate the mech-
anism is for the aromatics or other anthropogenic emissions. Improving
the quality of these inventories may be a difficult and expensive process,
but is is absolutely essential if airshed models are to be used for policy
assessment purposes. This is an area which should be given top priority.
7.5.2 Shorter Term Needs
We believe that this program has resulted in an advance in the
processing of emissions data for regional models. The aggregation of the
huge detailed emissions data base into a manageable data set which can be
used by a variety of models will permit for much more flexibility in using
alternative chemical mechanisms and for carrying out model intercomparison
studies. However, there is clearly need for further near-term work in
this area. The first priority should be to improve the software presently
used for processing the NAPAP emissions data bases. The software intro-
duces artificial limits to the level of chemical accuracy and range of
applicability in the emissions aggregation system; for example, the same
aggregation process cannot be used for both the RADM and the widely-used
Carbon Bond mechanism. In addition, some categories presently used in the
VOC speciation source profiles are chemically ambiguous and need to be
replaced by those which are not. Examples include "Mineral Spirits,"
which is not a compound but an uncharacterized mixture, and "C5 Olefins,"
which is a mixture of isomeric compounds with quite different reactivi-
ties. In some cases the speciation of the emissions sources may be highly
uncertain. However, those responsible for compiling emissions profiles
are in a better position to make estimates concerning the nature of these
profiles than the chemical mechanism developers. These are not, strictly
speaking, problems in the area of chemical mechanism development, but are
areas where experts in atmospheric chemistry could well provide much
needed guidance.
Another area where we believe this study has advanced the representa-
tion of VOC emissions in models is the use of model surrogates with param-
325
-------
eters which are adjusted based on the set of compounds they represent.
The RADM mechanism has been able to take advantage of some, but not all,
of the recent advances in this area. Under funding for the California
ARB, we developed software and procedures where mechanism parameters can
be readily re-adjusted as emissions inventory change. However, at present
the parameters in the RADM mechanism are fixed they were estimated
based on the 1985 NAPAP anthropogenic inventory but cannot readily be
changed if this inventory changes. It probably would not be too difficult
to modify the RADM mechanism and the model software so that the parameters
can be changed.
The representation of biogenically emitted alkenes is a particular
concern in regional models. Although the RADM mechanism represents iso-
prene explicitly, its performance in simulating the chamber data is less
than totally satisfactory. Probably it will be necessary to add new
species to the model to represent isoprene's products satisfactorily.
However, we were not able to complete our investigation of how best to
represent isoprene in the mechanism within the time and funding limits of
this program we are continuing work in this area under separate
funding.
The terpenes are also important in the biogenic inventory, and
research is needed on how best to represent them in the model. Chamber
studies on the reactivities of the terpenes and possibly other biogenics
under a wide range of conditions are clearly needed to test mechanism's
abilities to simulate at least their effects on ozone formation and NOX
oxidation. In addition, such data are needed to assess the ozone forma-
tion reactivities of these compounds relative to anthropogenic VOC
emissions. Until more data are available to test mechanisms for these
compounds, no change in how they are represented in the RADM mechanism is
warranted.
The scope of this study was such that we were unable to investigate
effects of condensation in the mechanism as comprehensively as we had
hoped at the time the program was conceived. We believe that research is
still needed to investigate the appropriateness of the tradeoffs between
condensation and chemical accuracy which are employed in RADM and other
condensed chemical mechanisms. We also believe that the test problems
developed in this program are well suited for this purpose. Examples of
326
-------
areas which need to be investigated concern the degree of condensation of
the species used to represent reactive products. Preliminary studies of
condensation methods we carried out for the SAPRC mechanism show that
condensing the number of species used to represent primary emitted
organics has less of an effect on model predictions than condensing the
number of species used to represent reactive organic products. The RADM
mechanism is relatively detailed in the number of species used to repre-
sent primary organic emissions, but is relatively condensed in its
representation of reactive products. We did not investigate this problem
as part of this program because by the time the other necessary tasks were
completed, it was too late for the RADM team to consider implementing
major changes in the mechanism for RADM-II. However, the possibility of
such changes should be considered in time for the next major update of the
RADM gas-phase chemical mechanism.
327
-------
REFERENCES
Atkinson, R., W. P. L. Carter, K. R. Darnall, A. M. Winer, and J. N.
Pitts, Jr. (1980), "A Smog Chamber and Modeling Study of the Gas
Phase N0x-Air Photooxidation of Toluene and the Cresols," Int J.
Chem. Kinet., J12, 779-836.
Atkinson, R., W. P. L. Carter, and A. M. Winer (1983), "Evaluation of
Hydrocarbon Reactivities for Use in Control Strategies," Final
Report, California Air Resources Contract No. AO-105-32, May.
Atkinson, R. and W. P. L. Carter (1984), "Kinetics and Mechanisms of the
Gas-Phase Reactions of Ozone with Organic Compounds Under
Atmospheric Conditions," Chem. Rev., 84, 437-470.
Atkinson, R. (1986), "Kinetics and Mechanisms of the Gas Phase Reactions
of the Hydroxyl Radical with Organic Compounds Under Atmospheric
Conditions," Chem Rev., 86, 69-201.
Atkinson, R. (1987), "A Structure-Activity Relationship for the
Estimation of Rate Constants for the Gas-Phase Reactions of OH
Radicals with Organic Compounds," Int. J. Chem. Kinet., ^9,
799-828.
Atkinson, R. (1988), "Gas-Phase Atmospheric Chemistry of Organic
Compounds," Appendix A of Final Report, California Air Resources
Board Contract No. A5-122-32, October.
Atkinson, R. (1989), "Gas-Phase Tropospheric Chemistry of Organic
Compounds: A Review," Atmos. Environ., in press.
Atkinson, R., S. M. Aschmann, J. Arey, and W. P. L. Carter (1989),
"Formation of Ring-Retaining Products from the OH Radical-
Initiated Reactions of Benzene and Toluene," Int. J. Chem.
Kinet., in press.
Burton, C. S. (1988), "Ozone Air Quality Models - Critical Review
Discussion Papers," J. Air Poll. Cont. Assoc., 38, 1119-1128.
Carter, W. P. L., A. C. Lloyd, J. L. Sprung, and J. N. Pitts, Jr. (1979),
"Computer Modeling of Smog Chamber Data: Progress in Validation
of a Detailed Mechanism for the Photooxidation of Propene and
n-Butane in Photochemical Smog," Int. J. Chem. Kinet, VI,
45-101.
Carter, W. P. L., R. Atkinson, A. M. Winer, and J. N. Pitts, Jr. (1982),
"Experimental Investigation of Chamber-Dependent Radical Sources,"
Int. J. Chem Kinet., _14, 1071-1103.
Carter, W. P. L. and R. Atkinson (1985), "Atmospheric Chemistry of
Alkanes," J. Atmos. Chem., 3, 377-405.
328
-------
Carter, W. P. L., F. W. Lurmann, R. Atkinson, and A. C. Lloyd (1986),
"Development and Testing of a Surrogate Species Chemical Reaction
Mechanism," EPA-600/3-86-031, U.S. Environmental Protection Agency,
Research Triangle Park, NC.
Carter, W. P. L., A. M. Winer, R. Atkinson, S. E. Heffron, M. P. Poe, and
M. A. Goodman (1987), "Atmospheric Photochemical Modeling of
Turbine Engine Fuels. Phase II. Computer Model Development,"
Report, USAF Contract No. F08635-83-0278, Engineering and
Services Laboratory, Air Force Engineering and Services Center,
Tyndall Air Force Base, Florida, August.
Carter, W. P. L. (1988a), "Documentation of a Gas Phase Photochemical
Mechanism for Use in Airshed Modeling," Appendix B to Final Report,
California Air Resources Board Contract No. A5-122-32, October.
Carter, W. P. L. (1988b) "Documentation for the SAPRC Atmospheric
Photochemical Mechanism Preparation and Emissions Processing
Programs for Implementation in Airshed Models," Appendix C to
Final Report, California Air Resources Board Contract No.
A5-122-32, October.
Carter, W. P. L. and R. Atkinson (1989a), "Alkyl Nitrate Formation from
the Atmospheric Photooxidation of Alkanes: A Revised Estimation
Method," J. Atmos. Chem., in press.
Carter, W. P. L. and R. Atkinson (1989b), "A Computer Modeling Study of
Incremental Hydrocarbon Reactivity," Environ. Sci. Technol.,
in press.
Dodge, M. C. (1989), "A Comparison of Three Photochemical Reaction
Mechanisms," J. Geophys. Res., 9^. 5121-5136.
Dunker, A. M., S. Kumar, and P. H. Berzins (1984), "A Comparison of
Chemical Mechanisms Used in Atmospheric Models," Atmos.
Environ., 18, 311-321.
EPA (1984), "Guideline for Using the Carbon Bond Mechanism in City
Specific EKMA," EPA-450/4-84-005, U.S. Environmental Protection
Agency, Research Triangle Park, NC.
EPA (1987), "Workshop on Evaluation/Documentation of Chemical
Mechanisms," EPA-600/9-87-024.
Fox, D. G. (1981), "Judging Air Quality Model Performance," Bull. Am.
Meteorol. Soc., 62, 599-609.
Gery, M. W., D. L. Fox, J. E. Jeffries, L. Stockbuger, and W. S. Weathers
(1985), "A Continuous Stirred Tank Reactor Investigation of the
Gas-Phase Reactions of Hydroxyl Radicals and Toluene," Int. J.
Chem. Kinet., J7, 931-955.
329
-------
Gery, M. W., G. Z. Whitten, and J. P. Killus (1988), "Development and
Testing of the CBM-IV For Urban and Regional Modeling,"
EPA/600/3-88/012, U.S. Environmental Protection Agency, Research
Triangle Park, NC.
Gipson, G. L. (1984), "Users Manual for OZIPM-2: Ozone Isopleth Plotting
With Optional Mechansims/Version 2," EPA-450/4-84-024, U.S.
Environmental Protection Agency, Research Triangle Park, NC.
Grosjean, D. and K. Fung (1984), "Hydrocarbons and Carbonyls in Los
Angeles Air," J. Air Poll. Control Assoc., 34, 537-543.
Hough, A. M. (1988), "An Intercomparison of Mechanism for the Production
of Photochemical Oxidants," J. Geophys. Res., 93, 3789-3812.
Jeffries, H. E., K. G. Sexton, and C. N. Salmi (1981), "The Effects of
Chemistry and Meteorology on Ozone Control Calculations Using
Simple Trajectory Models and the EKMA Procedure," EPA-450/4-81-
034, U.S. Environmental Protection Agency, Research Triangle Park,
NC.
Jeffries, H. E., K. G. Sexton, R. M. Kamens, and M. S. Holleman (1985),
Outdoor Smog Chamber Experiments to Test Photochemical Models:
Phase II. EPA/600/3-85/029, U.S. Environmental Protection Agency,
Research Triangle Park, NC.
Jeffries, H. E., K. G. Sexton, J. R. Arnold, and T. L. Kale (1989a),
"Validation Testing of New Mechanisms with Outdoor Chamber Data.
Volume 3: Calculation of Photochemical Reaction Photolysis Rates
in the UNC Outdoor Chamber," EPA/600/3-89/010c, U.S. Environmental
Protection Agency, Research Triangle Park, NC.
Jeffries, H. E., K. G. Sexton, J. R. Arnold, and T. L. Kale (1989b),
"Validation Testing of New Mechanisms with Outdoor Chamber Data.
Volume 1: Comparison of CB4 and CAL Mechanisms," EPA/600/3-89/010a,
U.S. Environmental Protection Agency, Research Triangle Park, NC.
Jeffries, H. E., K. G. Sexton, and J. R. Arnold (1989c), "Validation
Testing of New Mechanisms with Outdoor Chamber Data. Volume 2:
Analysis of VOC Data for the CB4 and CAL Photochemical
Mechanisms," EPA/600/3-89/010b, U.S. Environmental Protection
Agency, Research Triangle Park, NC.
Jeffries, H. E., K. G. Sexton, J. R. Arnold, and T. L. Kale (1989d),
"Validation Testing of New Mechanisms with Outdoor Chamber Data.
Volume 4: Appendices to Photochemical Reaction Photolysis Rates in
the UNC Outdoor Chamber," EPA/600/3-89/010d, U.S. Environmental
Protection Agency, Research Triangle Park, NC.
Killus, J. P. and G. Z. Whitten (1982), "A New Carbon-Bond Mechanism for
Air Quality Simulation Modeling," EPA-600/3-82-841, U.S.
Environmental Protection Agency, Research Triangle Park, NC.
330
-------
Leone, J. A. and J. H. Seinfeld (1984), "Evaluation of Chemical Reaction
Mechanisms for Photochemical Smog Part II - Quantitative
Evaluation of the Mechanisms," EPA-600/3-84-063, U.S.
Environmental Protection Agency, Research Triangle Park, NC.
Lurmann, F. W., A. C. Lloyd, and R. Atkinson (1986), "A Chemical
Mechanism for Use in Long-Range Transport/Acid Deposition
Modeling," J. Geophys. Res. 9J., 10905-10936.
Lurmann, F. W. and P. K. Karamchandani (1987), "An Updated Gas Phase
Chemical Mechanism with Chlorine Chemistry for the TADAP Regional
Model," Prepared for The Umweltbundesamt, West Germany,
Environmental Research and Technology Document P-E142-201,
September.
Lurmann, F. W., W. P. L. Carter, and L. A. Coyner (1987), "A Surrogate
Species Chemical Reaction Mechanism for Urban-Scale Air Quality
Simulation Models. Volume I - Adaptation of the Mechanism,"
EPA/600/3-87/Ol4a, U.S. Environmental Protection Agency,
Research Triangle Park, NC.
Milford, J. B. (1988), "Photochemical Air Pollution Control Strategy
Development," Ph.D. Thesis, Carnegie Mellon Univ., Pittsburgh, PA.
NASA (1985), "Chemical Kinetics and Photochemical Data for Use in
Stratospheric Modeling. Evaluation Number 7," JPL Publication
85-37, Jet Propulsion Laboratory, Pasadena, CA, July.
NASA (1987), "Chemical Kinetics and Photochemical Data for Use in
Stratospheric Modeling. Evaluation Number 8," JPL Publication
87-41, Jet Propulsion Laboratory, Pasadena, CA, September.
NCAR (1987), "Development and Implementation of Chemical Mechanisms for
the Regional Acid Deposition Model (RADM)," EPA Interagency
Agreement DW49930144-01-4, U.S. Environmental Protection Agency,
Research Triangle Park, NC.
Peterson, J. T. (1976), "Calculated Actinic Fluxes (290 - 700 nm) for
Air Pollution Photochemistry Applications," EPA-600/4-76-025,
U.S. Environmental Protection Agency.
Pitts, J. N., Jr., K. Darnall, W. P. L. Carter, A. M. Winer, and R.
Atkinson (1979), "Mechanisms of Photochemical Reactions in Urban
Air," EPA-600/3-79-110, U.S. Environmental Protection Agency.
Pitts, J. N., Jr., R. Atkinson, W. P. L. Carter, A. M. Winer, and E. C.
Tuazon (1983), "Chemical Consequences of Air Quality Standard and
of Control Implementation Programs," Final Report, California
Air Resources Board Contract No. A1-030-32, April.
Pitts, J. N., Jr., E. Sanhue2a, R. Atkinson, W. P. L. Carter, A. M.
Winer, G. W. Harris, and C. N. Plum (1984), "An Investigation of
the Dark Formation of Nitrous Acid in Environmental Chambers," Int.
J. Chen. Kinet., ^6, 919-939.
331
-------
Plum, C. N., E. Sanhueza, R. Atkinson, W. P. L. Carter, and J. N. Pitts,
Jr. (1983), "OH Radical Rate Constants and Photolysis Rates of
Alpha-Dicarbonyls," Environ. Sci. Technol., V7, 479-484.
Shafer, T. B. and J. H. Seinfeld (1985), "Evaluation of Chemical Reaction
Mechanisms for Photochemical Smog Part III - Sensitivity of EKMA
to Chemical Mechanisms and Input Parameters," EPA-600/3-85-042.
U.S. Environmental Protection Agency, Research Triangle Park, NC.
Shareef, G. S.f W. A. Butler, L. A. Bravo, and M. B. Stockton (1988), "Air
Emissions Species Manual Vol 1. Volatile Organic Compound Species
Profiles," EPA-450/2-88-003a, U.S. Environmental Protection Agency,
Research Triangle Park, NC.
Stockwell, W. R. (1986), "A Homogeneous Gas Phase Mechanism for Use in a
Regional Acid Deposition Model," Atmos. Environ., 20, 1615-1632.
Stockwell, W. R. and F. W. Lurmann (1989), "Intercomparison of the ADOM
and RADM Gas-Phase Chemical Mechanisms," State University of New
York, Draft Report to the Electric Power Research Institute, Palo
Alto, CA.
Tuazon, E. C., R. Atkinson, C. N. Plum, A. M. Winer, and J. N. Pitts,
Jr. (1983), "The Reaction of Gas-Phase N205 with Water
Vapor," Geophys. Res. Lett., H), 953-956.
Wagner, J. K., R. A. Walters, L. J. Maiocco, and D. R. Neal (1986),
"Development of the 1980 NAPAP Emissions Inventory,"
EPA-600/7-86-57a, U.S. Environmental Protection Agency, Research
Triangle Park, NC.
Weir, B. R., A. S. Rosenbaum, L. A. Gardner, G. Z. Whitten and W. Carter
(1988), "Architectural Coatings in the South Coast Air
Basin: Survey, Reactivity, and Toxicity Evaluation," Final Report,
South Coast Management District, SYSAPP-88/137, Systems
Applications, Inc., San Rafael, CA, December.
Westburg, H. and L. MacGregor (1987), "Nonmethane Organic Carbon
Concentrations in Air Masses Advected into Urban Areas in the
United States," EPA-600/3-87/045, U.S. Environmental Protection
Agency, Research Triangle Park, NC.
Whitten, G. Z., H. Hogo, and J. P. Killus (1980), "The Carbon-Bond
Mechanism: A Condensed Kinetic Mechanism for Photochemical Smog,"
Environ. Sci. Technol., 14, 690-287.
Zafonte, L., P. L. Rieger, and J. R. Holmes (1977), Environ. Sci. Technol.
H, 483.
332
-------
APPENDIX A
SELECTED RESULTS OF SIMULATIONS OF ALL EXPERIMENTS MODELED
USING THE RADM MECHANISM
333
-------
Appendix A. Selected Results of Simulations of All E»p»r1ments Modeled Using the RAOM Mechanism.
Page 1
Experiment
1 . Pure A1 r
ITC940
ITC955
ITC100B
JN06B2R
OC0684R
OC06B4B
Group Average
S. Dev.
Avg. Abs. Value
S. Dev.
Initial
Concentrat Ions
NOx
(ppm)
0.02
0.03
0.01
0.00
0.02
0.02
0.02
0.01
Ave
Temp
Ave
H20
HC HC/NOx
( ppmC )
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
(degK)
0.
0.
0.
0.
0.
0.
0.
0.
0
0
0
0
3
3
1
2
303
303
303
296
296
300
3
.0
.0
.0
.5
.5
.4
.6
(ppm)
20300.
20300.
18200.
14200.
20000.
20000.
18833.
2404.
Maximum Concentrat
OZONE
Expt
(ppm)
0.072
O.OB4
O.OBB
0.203
0.097
0. 1 19
0.107
0.051
Calc
(ppm)
0.073
0.073
0.073
0.203
0.094
0.09B
0. 102
0.051
Calc
-Expt
(ppm)
0.001
0.009
-0.014
0.001
-0.003
-0.021
-0.005
0.01 1
O.OOB
O.OOB
1 on
Calc
-Expt
/Avg
0.01
0. 13
-0. IB
0.00
-0.04
-0. 19
-0.04
0. 12
0.09
O.OB
CO
co
(cont1nued)
-------
Appendix A.
Selected Results
(cont1nued).
of Simulations of All Experiments Modeled Using the RADM Mechanism
Page 2
U)
U)
U1
Experiment
Initial
Concent rat Ions
NOx HC HC/NOx
(ppm) (ppmC)
1. N0x-A1r
EC436
EC440
EC442
EC457
EC464
EC597
EC599
ITC695
ITC826
ITCB82
OTC185
JN2782B
AU0282R
AU2082R
AU2282R
AU2382R
OC0882R
OC0882B
ST0582R
JL2483R
JL24B3B
JL2783B
AU06B3B
Group Average
S. Dev.
Avg. Abs. Value
S. Dev.
2. CO-NOx
ITC625
ITC634
OTC188
OTC201A
OTC201B
JN2782R
AU0282B
AU2082B
AU2282B
AU2382B
ST0582B
JL2783R
1 .79
0.76
0.58
0.50
0. 19
0.56
3.40
0.50
0.90
0.70
0.28
0.44
0.39
0.41
0.46
0.43
0.30
0.30
0.50
0.31
0.48
0.43
0.37
0.65
0.68
0.28
0.60
0.34
0.37
0.76
0.45
0.40
0.41
0.46
0.43
0.50
0.47
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.8
0.0
0.0
0.0
0.0
0.0
0.0
o.o
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.1
0.2
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0,0
0.0
0. 1
0. 1
0.2
0. 1
0.0
0. 1
0.9
0.0
0.1
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
o.'o
0.0
0. 1
0.2
0. 1
0. 1
0. 1
0. 1
0. 1
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Ave
1 emp
(degK)
303,0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
298. 1
301 .3
301 .6
302.0
300.0
302.8
294.3
294.3
297 .3
305.0
305 .0
300.7
304.2
301 .6
3.0
303.0
303.0
308.8
310.9
310.8
301 .3
301 .6
302. 1
300.0
302.8
297.3
302.8
Ave
H20
(ppm)
14900.
17500.
21400.
19600.
19300.
22000.
20800.
20300.
17200.
17900.
5000. *
29300.
27700.
21000.
25600.
20200.
18100.
1B100.
14300.
19600.
19600.
20000.
20200.
19548.
4721 .
20300.
20300.
5000. »
3880.
3000.
29300.
27700.
21000.
25600.
20200.
14300.
20000. *
Final - In1t
NO
Expt
(ppm)
-0.098
-0.063
0.069
-0.066
-0.019
-0.061
2.276
-0.021
-0.349
-0.001
-0.002
-0. 1 13
0.035
-0.011
0.000
0.056
-0.024
-0.017
0.070
-0.047
-0.071
-0.090
0.001
0.063
0.490
0. 155
0.468
-0.069
-0.040
-0.090
-0. 160
-0. 123
-0.320
-0.054
-0.111
-0.086
-0.004
0.019
-0.289
Calc
(ppm)
-0.052
-0.049
0.093
-0.042
-0.002
-0.031
1 .878
-0.008
-0. 182
-0.027
-0.004
-0.118
0.039V
-0.010
0.022
0.057
-0.032
-0.032
0.085
-0.063
-0.071
-0.076
-0.004
0.060
0.401
0. 130
0.383
-0.085
-0.061
-0. 136
-0. 157
-0.115
-0.339
-0.062
-0. 146
-0.089
-0.051
0.026
Calc
-Expt
(ppm)
0.046
0.015
0.024
0.023
0.016
0.030
-0.398
0.014
0. 167
-0.027
-0.002
-0.005
0.004
0.002
0.023
0.001
-0.008
-0.014
0.015
-0.015
0.000
0.012
-0.005
-0.004
0.094
0.038
0.085
-0.016
-0.021
-0.046
0.003
0.008
-0.019
-0.008
-0.035
-0.003
-0.047
0.007
Final - In1t
N02-UNC
Expt
(ppm)
-0.071
0.024
-0. 172
0.008
-0.002
0.041
-2.685
0.011
0.263
-0.010
-0.003
0.050
-0. 129
-0.049
-0. 147
-0. 144
0.005
0.009
-0.171
0.011
0.014
0.036
-0.090
-0. 139
0.563
0. 180
0.551
0.068
0.034
0.072
0. 162
0.096
0. 202
-0.002
0.072
0.003
-0.042
-0.086
0.214
Calc
(ppm)
-0.057
0.013
-0. 160
-0.008
-0.012
0.005
-2.536
0.001
0. 135
-0.014
-0.001
0.025
-0. 120
-0.058
-0.116
-0. 14B
-0.006
-0.005
-0. 157
-0.004
-0.010
0,009
-0.088
-0. 144
0.526
0. 160
0.521
0.078
0.052
0. 1 25
0. 143
0. 100
0.239
-0.026
0.076
-0.009
-0.046
-0.098
Calc
-Expt
(ppm)
0.014
-0.01 1
0.012
-0.017
-0.010
-0.036
0. 149
-0.010
-0. 128
-0.003
0.002
-0.025
0.009
-0.009
0.031
-0.005
-0.011
-0.014
0.014
-0.015
-0.024
-0,027
0,002
-0.005
0.045
0.025
0.037
0.010
0.017
0.053
-0.019
0.004
0.038
-0.023
0.004
-0.012
-0.004
-0.012
{cont1nued)
-------
Appendix A. Selected Results of Simulations of All Experiments Modeled Using the RADM Mechanism
(continued).
Page 3
Experiment
Croup
Avg.
Average
S. Dev.
Abe. Value
S. Dev.
Initial Ave Ave
Concentrations Temp H20
NOx
(ppm)
0.45
0.12
HC HC/NOx
(ppmC) (degK) (ppm)
0.0 0.0 303.7 17760.
0.0 0.0 4.1 8754.
Final - I nit
NO
Expt
(ppm)
-0.112
0.099
0. 1 15
0.095
Calc
(ppm)
-0.110
0.092
0.115
0.065
Calc
-Expt
(ppm)
-0.016
0.020
0.019
0.017
Final - In1t
N02-UNC
Expt
(ppm)
0.063
0.089
0.083
0.069
Calc
(ppm)
0.058
0.097
0.090
0.064
Calc
-Expt
(ppm)
0.005
0.024
0.018
0.015
(contInued)
u>
-------
Appendix A. Solected Results of Simulations of All Experiments Modeled Using the RADM Mechanism
(cont1nued) .
Page 4
to
UJ
Experiment
1. HCHO-A1r
EC250
EC255
JL17B2R
JL1782B
OC0784R
OC07B4B
OC1684R
OC1684B
Group Average
S. Dev.
Avg. Abs. Value
S. Dev.
Initial
Concent rat
NOx
(ppm)
0.02
0.02
0.01
0.01
0.00
0.01
0.00
0.00
0.01
0.01
HC
(ppmC)
0.3
0.3
0.5
0.6
0.4
1 .0
0.4
0.3
0.5
0. 2
1ons
HC/NOx
Ave
Temp
Ave
H20
Maximum Concentrat
OZONE
Expt
(degK)
14.5
20.4
71.2
86.0
96. 1
125.7
69.0
43.8
303
303
303
303
292
292
299
299
299
4
.0
.0
.3
.3
.6
.6
.7
.7
.7
.6
(ppm)
17400.
15700.
14700.
14700.
20000.*
20000.*
20000.*
20000.
17B13.
24B3.
(ppm)
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
215
203
239
264
087
138
153
133
179
061
Calc
(ppm)
0.258
0.255
0.329
0.335
0. 124
0. 181
0. 152
0. 134
0. 221
0.085
Calc
-Expt
(ppm)
0.043
0.052
0.090
0.071
0.037
0.043
-0.001
0.000
0.042
0.031
0.042
0.031
1 on
Calc
-Expt
/Avg
0. 18
0.23
0.32
0. 24
0.36
0. 27
-0.01
0.00
0.20
0.13
0. 20
0.13
(contInued)
-------
Appendix A.
Selected Results of Simulations of All Experiments Modeled Using the RADM Mechanism
(cont1nued).
Page 5
CO
Experiment
Initial
Concent rat
NOx HC
(ppm) (ppmC)
1ona
HC/NOx
Ave
Temp
(OegK)
Ave
H20
(ppm)
Maximum Concentrat
OZONE
Expt
(ppm)
Calc
(ppm)
Calc
-Expt
(ppm)
1 on
Calc
-Expt
/Avg
Maximum Concentrat
PAN
Expt
(ppm)
Calc
(ppm)
Calc
-Expt
(ppm)
1 on
Calc
-Expt
/Avg
1. Acetaldehyde-A1r
EC253
ITC627
ITC636
ITCB25
ITC957
ITC974
ITC1009
OTC200A
OTC200B
OTC206A
OTC206B
OTC234A
OTC234B
JL2683R
JL2683B
AU0483R
AU0483B
OC1584R
OC1584B
Group Average
S. Oev.
Avg. Abs . Value
S. Dev.
0.02
0.03
0.03
0.01
0.04
0.03
0.04
0.01
0.01
0.01
0.01
0.04
0.04
0.02
0.01
0.03
0.02
0.02
0.02
0.02
0.01
1 . 1
0.8
0.7
O.B
1 . 1
0.9
0.9
1 . 1
o.e
1.3
1 .0
.0
.5
. 1
. 1
.0
. 1
.0
.0
1 .0
0.2
53.8
26. 1
26.6
55.3
27.6
29.8
27.0
96.9
68.8
136.8
94.0
26.2
39.3
48.3
87.6
34.3
47.1
48.5
45.2
53.6
30.5
303.0
303.0
303.0
303.0
303.0
303.0
303.0
309.5
309.4
302.2
304.5
307.5
307.7
302.8
302.8
304.1
304. 1
296.5
296.5
303.6
3.4
17100.
20300.
20300.
17300.
20300.
19000.
20100.
3950.
2870.
8740.
7B40.
5000.*
5000.*
25900.
25900.
29300.
29300.
15200.
15200.
16242.
8541.
0. 137
0.060
0.047
0.037
0.076
0.085
0.078
O.OBB
0.076
0.030
0.023
0.083
0.084
0.422
0.331
0.548
0.431
0. 140
0. 193
0. 156
0.156
0. 100
0.050
0.050
0.017
0.051
0.048
0.052
0.081
0.080
0.063
0.087
0. 138
0. 138
0.364
0.235
0.456
0.342
0. 149
0. 195
0. 142
0. 124
-0.036
-0.010
0.003
-0.019
-0.025
-0.037
-0.026
-0.007
0.004
0.033
0.064
0.055
0.054
-0.057
-0.096
-0.092
-0.090
0.009
0.002
-0.014
0.048
0.038
0.031
-0.31
-0. 19
-0.40
-0.56
-0.40
-0.08
0.05
0.50
0.49
-0. 15
-0.34
-0. 18
-0.23
0.06
0.01
-0. 12
0.30
0.26
0. IB
0.040
0.013
0.011
0.005
0.013
0.006
0.008
0.008
0.006
0.020
0.013
0.004
0.007
0.026
0.027
0.040
0.040
0.023
0.035
0.018
0.013
0.046
0.014
0.014
0.008
0.016
0.015
0.015
0.007
0.008
0.011
0.010
0.012
0.012
0.035
0.035
0.043
0.047
0.025
0.023
0.021
0.014
0.006
0.001
0.003
0.003
0.004
0.009
0.007
-0.001
0.002
-0.009
-0.003
0.008
0.005
0.010
0.008
0.003
0.007
0.002
-0.012
0.003
0.006
0.005
0.003
0. 15
0.10
0.27
0.47
0.25
0.84
0.59
-0.08
0. 23
-0.58
-0.31
0.97
0.53
0.31
0.26
0.07
0. 16
0.08
-0.40
0.20
0.39
0.35
0.26
(contInued)
-------
Appendl x A .
Selected Results
(cont1nued).
of Simulations of Alt Experiments Modeled Using the RAOM Mechanism
Page 6
to
CO
Experiment
Inlt 1al
Concent rat Ions
NOx HC HC/NOx
(ppm) (ppmC)
1 . HCHO-NOx
EC251
EC252
EC3B9
EC391
EC392
ITCB64
OTC235A
OTC235B
AU0179B
AU0279B
AU0479B
AU0579B
JL2381B
OC0984R
OC0984B
Group Average
S. Dev.
Avg. Aba. Value
S. Dev.
2. Acetaldehyde-NOx
EC164
EC254
AU0179R
JN1482R
AU2482B
Group Average
S. Dev.
Avg. Abs . Value
S. Dev.
0. 1 1
0,49
4.75
4.43
8.05
0.54
0.56
0.58
0.35
0.21
0.23
0.54
0.43
0.56
0.50
1.49
2.33
0.51
0.11
0.36
0.31
0.32
0.32
0. 14
0.2
0.4
9. 1
17.7
9.5
0. 1
0. 1
0. 1
1 .0
1 .0
0.5
1 .2
1 .5
1 .0
1 .0
2.9
5. 1
0.7
1 .0
2.0
3. 1
1 .9
1 .7
1 .0
1 .7
0.7
1 .9
4.0
1 .2
0. 1
0. 1
0. 1
2.8
4.7
2. 1
2.2
3.5
1 .7
1 .9
1 .9
1 .4
1 .4
8.5
5.7
9.9
6.0
6.3
3.2
Ave
Temp
(degK)
303.0
303.0
303.0
303.0
303.0
303.0
299.2
297.9
SOB. 2
307 . 2
304.8
296.5
306. 1
295. 1
295. 1
301 .9
4.2
303.0
303.0
308.3
300. 2*
304.0
303 . 7
2.9
Ave
H20
(ppm)
17600.
17100.
2080.
2310.
877.
16000.
4430. *
3940.*
19400.
20000. *
20800.
12500.
20000.*
10900.
10900.
11922.
7438.
34000.
18900.
31300.
22300.
17400.
24760.
7462.
Maximum Concentration
OZONE
Expt
(ppm)
0.271
0.032
0.002
2.371
0.000
0.001
0.273
0.308
0.618
0.606
0.378
0.508
0.637
0.666
0.301
0.465
0.5BO
0.086
0.264
0.930
0.731
0.972
0.596
0.401
Calc
(ppm)
0.300
0.025
0.031
2. 190
0.008
0.017
0. 143
0.267
0.519
0.612
0.392
0.547
0.792
1 . 187
0.403
0.496
0.573
0.048
0. 185
0.580
0.653
0.665
0.426
0.289
Calc
-Expt
(ppm)
0.029
-0.007
0.029
-0. 181
0.006
0.017
-0. 130
-0.041
-0.099
0.006
0.014
0.040
0. 155
0.521
0. 102
0.031
0.160
0.092
0. 132
-0.037
-0.079
-0.350
-0.078
-0.307
-0. 170
0. 146
0. 170
0. 146
Calc
-Expt
/Avg
0. 10
-0.08
-0.63
-0.14
-0.17
0.01
0.04
0.07
0.22
0.56
0. 29
0.02
0.30
0.21
0. 21
-0.55
-0.35
-0.46
-0.11
-0.37
-0.37
0. 17
0.37
0. 17
Average
d( (03) -
Initial
[NO] )/dt
Calc
Expt Calc -Expt
(ppb/mln) --
3.25
1 .94
50.01
102.89
13.52
0.60
2.00
2. 16
2.06
2.12
1.17
1 .95
2.70
2.11
1.51
12.67
27.92
2.03
1 .33
1 .98
1 .62
1 .68
1 . 73
0. 29
3.77
2.13
50.67
97. 24
30.58
0.71
1 . 74
2.09
1 .96
2.06
1.11
1 .94
2.80
2.73
1 .76
13.55
27.07
1 .49
0.99
1 .69
1 .50
1 .54
1 .44
0.26
0.51
0. 19
0.66
-5.64
17.06
0.11
-0.26
-0.06
-0.10
-0.07
-0.06
-0.01
0.10
0.62
0.25
0.89
4.72
1 .71
4.47
-0.54
-0.33
-0.29
-0.12
-0.14
-0.29
0. 17
0.29
0. 17
Calc
-Expt
/Avg
0. 15
0.09
0.01
-0.06
0. 77
0.17
-0.14
-0.03
-0.05
-0.03
-0.05
0.00
0.04
0. 26
0. 15
0.09
0.22
0. 13
0. 19
-0.31
-0.29
-0. 16
-0.08
-0.09
-0. 18
0.11
0. 18
0.11
3. Propr1onaldehyde-N0x
JN1482B
0.30
3. 1
10.5
300. 1*
18900.
0.733
0.621
-0.112
-0.17
1 .75
1 .34
-0.41
-0.27
(cont 1nued)
-------
Appendix A. Selected Results of Simulations of All Experiments Modeled Using the RADM Mechanism
(contInued).
Page 7
Experiment
Group
Avg.
Average
S. Dev.
Abs. Value
S. Dev.
Initial Ave Ave
Concentrations Temp H20
NOx
(ppm)
0.32
0.02
HC HC/NOx
(ppmC) (degK) (ppm)
2.5 8.0 302.1 18150.
0.9 3.4 2.8 1061.
Maximum Concentrat
OZONE
Expt
(ppm)
0.837
0. 147
Calc
(ppm)
0.608
0.019
Calc
-Expt
(ppm)
-0.229
0. 166
0.229
0. 166
Ion
Calc
-Expt
/Avg
-0.31
0.20
0.31
0.20
Average Initial
d( (031 - (NO) )/dt
Expt Calc
(ppb/mln)
1.79 1.38
0.05 0.05
Calc
-Expt
-0.41
0.00
0.41
0.00
Catc
-Expt
/Avg
-0.26
0.01
0.26
0.01
to
-Cr
O
4. Acetone-NOx
JN0480R
0.17
4.8
27.5 302.2
17700.
0.233 0.397 0.164 0.52
0.48 0.52 0.04 0.09
S. Methy l«thyl ketone-NOx
OC2079R
JN0480B
Group Average
S. Dev.
Avg. Aba. Value
S. Oev.
6. Ethene-NOx
EC142
EC143
EC156
EC285
EC286
EC287
ITC926
ITC936
AU0479R
AU0579R
OC0584R
OC1 184R
OC1284R
OC0584B
Group Average
S. Dev.
Avg. Abs. Value
S. Dev.
0.22
0.18
0.20
0.03
0.48
0.50
0.50
1 .01
0.94
0.53
0.51
0.50
0.23
0.64
0.36
0.35
0.72
0.37
0.55
0.22
13.6
3.8
8.7
6.9
1 .9
4. 1
4.0
3.9
7 .5
8.0
7.9
3.9
0.9
4. 1
3.2
2.9
2.7
1 .8
4. 1
2.3
60.5
21 .6
41 .0
27.5
4. 1
8. 1
8.0
3.9
8.0
15.1
15.6
7.8
3.9
6.4
8.8
B.2
3.7
5.0
7.6
3.8
295
302
298
4
303
303
303
303
303
303
303
303
298
292
297
297
295
297
300
3
.3
.2
.8
.9
.0
.0
.0
.0
.0
.0
.0
.0
.9
.4
.3
.2
.6
.3
. 2
.6
27800.
18100.
22950.
6859.
26500.
22700.
29000.
20400.
20500.
19700.
21 100.
20300.
32500.
29 100.
20000.*
12700.
13600.
20000.
22007.
5611 .
0
0
0
0
0
1
1
0
1
0
0
0
0
1
0
0
0
0
0
0
.563
.652
.607
.063
.782
.087
. 105
.840
.081
.965
.982
.940
.729
.294
.856
.858
.495
.675
.906
.204
0.466
0.534
0.500
0.048
0.474
0.791
0.751
0.892
1.171
1 .026
0.920
0.838
0.554
1 . 142
0.971
1 .039
0. 185
0.646
O.B14
0. 275
-0.096
-0. 1 18
-0. 107
0.015
0. 107
0.015
-0.307
-0.296
-0.354
0.052
0.090
0.061
-0.062
-0. 102
-0. 175
-0. 152
0.114
0. 182
-0.310
-0.029
-0.092
0. 179
0. 163
0.111
-0. 19
-0.20
-0. 19
0.01
0. 19
0.01
-0.49
-0.32
-0.38
0.06
0.08
0.06
-0.06
-0. 1 1
-0.27
-0. 12
0. 13
0. 19
-0.91
-0.04
-0. 16
0.30
0.23
0. 24
1 .
1 .
1 .
0.
3.
8.
8.
5.
1 1 .
13.
6.
2.
1 .
3.
2.
2.
1 .
1 .
5.
4.
25
13
19
09
20
SO
89
05
76
89
96
72
60
17
16
22
5B
48
23
10
1 .71
1 .21
1 .46
0.35
2.20
4.51
4.50
4.89
10. 13
13.00
4.91
2.34
0.99
2.27
1 .98
2.01
1 . 17
1 .33
4.02
3.53
0.46
0.08
0.27
0.26
0.27
0.26
-1 .00
-3.99
-4.39
-0.16
-1 .63
-0.88
-2.05
-0.38
-0.61
-0.91
-0. IB
-0.21
-0.41
-0. 15
-1.21
1 .39
1.21
1 .39
0.31
0.07
0. 19
0. 17
0. 19
0. 17
-0.37
-0.61
-0.66
-0.03
-0. 15
-0.07
-0.35
-0.15
-0.47
-0.33
-0.09
-0. 10
-0.30
-0. 1 1
-0. 27
0.20
0.27
0.20
(contInued)
-------
Appendix A. Selected Results of Simulations of All Experiments Modeled Us1n8 the RADM Mechanism
(cont1nued).
Page 8
oo
-fc
Experiment
Concent rat 1 ons
NOx HC HC/NOx
(ppm) (ppmC)
Avc
Temp
(degK)
Ave
H2O
(ppm)
Maximum Concentration
OZONE
Expt
(ppm)
Calc
(ppm)
Calc
-Expt
(ppm)
Calc
-Expt
/Avg
Average
d( [03] -
Initial
(NO) )/dt
Calc
Expt Calc -Expt
(ppb/m1n) --
Calc
-Expt
/Avg
7. Propene-NOx
EC 121
EC177
EC216
EC21 7
EC230
EC256
EC257
EC276
EC277
EC278
EC279
EC314
EC315
EC316
EC317
ITC693
ITC810
ITCB60
ITC925
ITC938
ITC947
ITC960
OTC186
OTC191
OTC210
OTC233
OTC236
JA1078R
OC1278B
OC207BR
OC207BB
OC21 78R
OC2578B
JN1279R
JN1279B
JN1379R
AU0279R
AU27BOB
ST0482B
ST1382B
JL1783R
JL2183R
JL2983B
JL3183R
ST2383B
0.51
0.46
0.52
0.48
0.52
0.56
0.56
0.52
0. 1 1
0.49
0.97
0.93
0.94
0.98
0.54
0.49
0.52
0.52
0.54
0.52
0.53
0.50
0.55
0.54
0.57
0.46
0.53
0.46
0.48
0.46
0.46
0.50
0.44
0.50
0.49
0.45
0.22
0.48
0.23
0.33
0.27
0.22
0.21
0.21
0.38
1.5
1 .5
1 .5
0.6
1 .9
0.4
0.7
1 .6
1 . 7
3. 1
3.5
3. 2
2.9
3.2
1 .5
3.5
2.6
3.0
2.8
2.8
1 .9
2.8
3.6
3.7
2.7
0. 1
3.3
3. 1
1 .4
1 .3
3.5
3.9
1 .3
1 .0
1 .5
2.9
1 .5
1 .9
1 . 1
1 . 1
1 . 1
1 . 1
1 . 1
1 . 1
1 .6
2.9
3.2
3.0
1 .2
3.7
0.7
1 .3
3.2
15.7
6.2
3.5
3.5
3. 1
3.3
2.8
7.2
5.4
5.8
5.2
5.3
3.6
5.5
6.6
6.9
4.8
0.2
6.3
6.9
2.9
2.9
7.7
7.9
2.9
2. 1
3.0
6.5
7.0
4.0
4.9
3.3
3.9
5.0
5.3
5.1
4.3
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303 .0
303.0
303.0
303.0
288. 7
312.0
303.0
303.0
303.0
303.0
303 . 0
303.0
303.0
303.0
299.8
309.3
304.7
303.6
300.0
265. 1
294.3
290.4
290.4
290.3
288.4
293.5
293.5
294.9
307.2
302.2
300.5
302.6
307.2
308.5
303. 1
305.0
292.5
26500.
26900.
18100.
21400.
17200.
20200.
20000.
14500.
15700.
16100.
14200.
24000.
10600.
44900.
24200.
20300.
16800.
17100.
16300.
19000.
20300.
20300.
5000. »
5000. *
3/20. *
5000. »
6510.
20000.*
16300.
9590.
9560.
20000.
12800.
20000. *
20000. *
20000.*
20000.*
28400.
1 1800.
24100.
29700.
23BOO.
23300.
19800.
23400.
0.506
0.540
0.564
0. 149
0.344
0.002
0.068
0.388
0.313
0.625
0.679
0.728
0.344
0.955
0.615
0.779
0.782
0.585
0.779
0.729
0.710
0.721
0.822
0.903
0.972
0.633
0.848
0.363
0.461
0.340
0.727
0.670
0.230
0.382
0.673
0.974
0.788
1 .044
0.658
0. 731
0.848
0.804
0.697
0.719
0.405
0.448
0.466
0.550
0. 227
0.309
0.010
0.094
0.435
0.386
0.685
0.692
0.776
0.638
1 .043
0.563
0.702
0.705
0.665
0.714
0.704
0.701
0.697
0.853
1 .082
0.989
0.916
0.990
0.583
0.387
0.378
0.924
0.893
0.243
0.285
0.508
0.861
0.622
1 . 180
0.607
0.678
0.713
0.725
0.733
0.682
0.654
-0.057
-0.074
-0.014
0.07B
-0.035
0.007
0.025
0.046
0.073
o.oeo
0.013
0. 048
0. 294
0.088
-0.052
-O.O78
-0.078
0.081
-0.065
-0.025
-0.009
-0.024
0.030
0. 179
0.017
0.285
0. 142
0.220
-0.074
0.037
0. 197
0. 223
0.013
-0.097
-0. 165
-0.113
-0. 166
0. 136
-0.052
-0.054
-0. 134
-0.079
0.037
-0.037
0.250
-0. 12
-0.15
-0.03
0.42
-0. 1 1
0.31
0. 1 1
0.21
0.09
0.02
0.06
0.60
0.09
-O.09
-0.11
-0. 10
0.13
-0.09
-0.03
-0.01
-0.03
0.04
0. 18
0.02
0.37
0. 15
0.46
-0.17
0. 10
0.24
0.29
0.05
-0 . 29
-0.28
-0.12
-0. 24
0.12
-0.08
-O.OB
-n. 17
-0.10
0.05
-0.05
0.47
7.46
3.69
4. 18
0.60
3.06
0.9B
3.35
3. 24
8.27
7 .72
6.64
7.21
4.33
10.64
4.06
5.07
4.25
3,57
3.72
3.61
3.34
4.23
5. 16
12.18
6.70
3.69
6.91
0.46
1 .40
1 . 23
2.83
2.53
1 .03
0.85
1 . 24
2. 69
2.41
3. 26
1 .46
1 .69
2.10
1 . 76
1 .66
1.81
1 . 23
3.80
3.71
5. 28
1 . 25
4.41
1 .03
4.06
4. 06
7. 60
9. 03
7. 38
9. 47
5.96
10.99
4. 28
5. 29
4. 22
4. 50
3.81
3.91
3. 82
4.05
7. 22
1 2. 79
5.95
5.12
7.83
0.48
1 .39
1 .33
3.42
3.32
1.10
0.74
1 .09
2. 75
2.01
2.79
.54
. 57
.72
.83
. 70
.62
.62
-3.66
-0. 18
1.10
0. 45
1 . 35
0 . 05
0.71
0.82
-0.67
1 . 30
0 .73
2 . 26
1 . 62
0.35
0. 22
0.23
-0.03
0.94
0.09
0 .30
0.48
-0.18
2.06
0.61
-0. 75
1 .43
0.93
0.01
-0.01
0. 09
0.59
0. 79
0.07
-0.11
-0. 15
0.06
-0. 40
-0.47
0.06
-0.12
-0. 38
0. 07
0. 04
-0.19
0.39
-0.65
-0 . 05
0 . 23
0 . 44
0. 36
0 . 05
0.19
0.22
-0. 08
0.16
0.10
0 . 27
0 . 32
0.03
0 . 05
0.04
-0.01
0 . 23
0. 02
0 . 08
0.13
-0.04
0.33
0. 05
-0.12
0. 32
0.13
0. 03
-0.01
0. 07
0.19
0. 27
0 .06
-0.14
-0.13
0 .02
-0.18
-0.16
0 .06
-0 . 07
-0 . 20
0 .04
0 .03
-0.11
0. 28
(contInued)
-------
Appendix A. Selected Reeulta of Simulations of All Experiments Mod*lad Utlng the RAOM Mechanism
(continued).
Page 9
ru
Experiment
Initial
Concentrat lone
NOx
Ave
Temp
Ave
H20
HC HC/NOx
(ppm) (ppmC)
OC04B4B
0011848
OC12B4B
Group Average
S. Dev.
Avg. Aba. Value
S. Dev.
8. 1-Butene-NOx
EC122
EC 123
EC124
ITC927
ITC928
ITC930
ITC935
ST23B3R
ST2583R
ST25B3B
ST2783R
Group Average
S. Dev.
Avg. Aba. Value
S. Dev.
0.36
0.36
0.68
0.49
0. 18
0.50
O.SI
0.99
0.31
0.67
0.32
0.66
0.40
0.46
0.42
0.45
0.52
0.20
1.0
2.2
2.0
2. 1
1.0
0.9
1.6
1 .7
3.6
3.B
7.2
7.6
1.5
1 .6
2.9
1 .6
3.1
2.3
2.9
6.3
2.9
4.6
2.5
1 .7
3.2
1 .7
12.3
5.7
22.2
1 1 .6
3.7
3,6
6.8
3.6
6.9
6.2
(degK)
296.7
297.2
295.9
299.9
7.5
303.0
303.0
303.0
303.0
303.0
303.0
303.0
292.6
292.3
292.3
295.5
299.4
5.0
(ppm)
20000.*
12700.
13500.
18134.
7349.
22000.
27200.
24500.
19000.
18400.
21 100.
19700.
23400.
18800.
18900.
19400.
21127.
2856.
Maximum Concentration
OZONE
Expt
(ppm)
0.446
0.674
0.432
0.608
0.238
0.227
0.506
0.247
0.646
0.022
0.717
0.872
0.206
0.266
0.594
0.28S
0.417
0.263
Calc
Calc -Expt
(ppm) (ppm)
0.457 0.011
0.757 0.083
0.43B 0.006
0.635 0.026
0.249 0.112
0.087
0.075
0.073 -0.154
0.267 -0.238
0.121 -0.126
0.712 0.066
0.037 0.015
0.807 0.090
0.949 0.077
0.431 0.225
0.427 0.161
0.767 0.174
0.430 0.145
0.457 0.039
0.316 0.150
0. 134
0.068
Calc
-Expt
/Avp
0.02
0.12
0.01
0.05
0.20
0.15
0. 13
-1.02
-0.62
-0.69
0. 10
0. 12
O.OB
0.71
0.46
0.26
0.40
-0.02
0.56
0.45
0.31
Average
d( [03] -
Expt
Calc
Initial
[NO] )/dt
Calc Calc
-Expt -Expt
(ppb/m1n) -- /Avg
1.25
2.38
1 .67
3.62
2.60
2.29
4.16
1 .99
3.24
1 .36
7.91
5.39
0.96
1 . 16
1.78
1 . 19
2.86
2.18
1.17
2.36
1.57
3.88
2.81
1.31
2. 14
1 .58
4. 19
1 .58
10.39
6. 56
1. 12
1.27
2. 13
1 . 27
3.05
2.95
-0.08 -0.07
-0.02 -0.01
-0.10 -0.06
0.26 0.06
0.87 0.19
0.57 0.14
0.70 0.13
-0.98 -0.54
-2.02 -0.64
-0.41 -0.23
0.95 0.26
0.23 0.15
2.49 0.27
1.20 0.20
0.16 0.16
0.11 0.09
0.35 0.18
0.08 0.06
0.20 0.00
1.16 0.32
0.81 0.25
0.81 0. 18
9. trans-2-Butene-NOx
EC146
EC147
EC157
ST27B3B
Group Average
S. Dev.
Av0. Abs . Value
S. Dev.
10. Isobutene-NOx
ITC694
0.51
0.9B
0.53
0.43
0.61
0.25
0.51
0.9
1 .7
0.9
2.0
1 .4
0.6
4,6
1.8
1 .7
1 .7
4.7
2.5
1.5
9. 1
303.0
303.0
303.0
295.5
301 . 1
3.7
303.0
26900.
22300.
27500.
19400.
24025.
3860.
20000.
0.247
0. 154
0.205
0.523
0.282
0. 165
0.900
0.109 -0. 137
0.084 -0.070
0.091 -0.114
0.544 0.021
0.207 -0.075
0.225 0.070
O.OB5
0.051
0.714 -0.186
-0.77
-0.59
-0.77
0.04
-0.52
0.3B
0.54
0.35
-0.23
5.87
9.83
5.96
3.00
6. 17
2.80
8.84
3.99
7.66
3.64
3.36
4.66
2.01
5.59
-1.88 -0.38
-2.18 -0.25
-2.32 -0.48
0.35 0.11
-1.50 -0.25
1.25 0.26
1.68 0.31
0.90 0. 16
-3.25 -0.45
(cont Inued)
-------
Appendix A. Selected Results of Simulations of All Experiments Modeled Using the RADM Mechanism
(continued).
Page 10
4r
U)
Experiment
Initial
Concent rat
Ions
NOx HC HC/NOx
(ppm) (ppmC)
11. 1-Hexene-NOx
ITC929
ITC931
ITC934
ITC937
Group Average
S. Oev.
Avg. Abs. Value
S. Oev.
12. Isoprene-NOx
EC520
EC522
EC524
EC525
EC527
ITC81 1
ITCB1 2
JL1680R
JL1680B
JL1780R
JL1780B
JL2381R
ST09B1R
Group Average
S. Oev.
Avg. Abs. Value
13. a-P1nene-NOx
JL1580R
JL1580B
JL2580R
JL2580B
Group Average
S. Dev.
Avg. Abs. Value
S. Dev.
0.51
0.49
1 .00
0.99
0.75
0.26
0.49
0.92
0.96
0.54
0.50
0.50
0.52
0. 18
0.18
0.46
0.47
0.43
0. 17
0.49
0.25
0. 18
0. 19
0.25
0.25
0.22
0.04
5. 1
10.3
9.7
0. 1
6.3
4.8
2.2
2.3
4.7
4.5
2. 2
3.4
1 .8
4.6
6.4
1 .0
2.6
1 .4
1 .0
2.9
1 .7
1 . 1
2.6
1 .0
0.3
1 .3
1 .0
10.0
21 . 1
9.7
0. 1
10.2
8.6
4.5
2.4
4.9
8.3
4.5
6.8
3.4
26/0
36.6
2. 1
5.5
3.4
6. 1
8.6
10.4
5.8
13.8
4.0
1 .4
6.3
5.3
Ave
Temp
(UegK)
303.0
303.0
303.0
303.0
303.0
0.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
305.0
301 .9
305.6
302.8
306. 1
297.5
303. 1
2. 1
302.4
299.6
302.4
300 . 1
301 . 1
1 .5
Ave
H20
(ppm)
19000.
20100.
20300.
20300.
19925.
624.
13000.
13000.
13000.
13000.
13000.
17300.
17300.
.30200.
26000.
27000.
25000.
20000.*
20000.
19123.
6284.
30300.
29200.
31700.
2630O.
29375.
2291 .
Maximum Concentration
OZONE
Expt
(ppm)
0. 298
0.606
0.428
0.007
0.335
0.252
0.503
0.276
0. 759
0.691
0.547
0.919
0.768
0.652
0.837
0.606
1 .298
0.750
0.506
0.716
0.246
0.201
0.470
0.377
0.334
0.346
0.112
Calc
(ppm)
0.647
0.799
0.876
0.047
0.593
0.376
0.300
0. 1 18
0.429
0.630
0.317
0.577
0.253
0.712
0.600
0.420
0.835
0.691
0.457
0.488
0.208
0.299
0.420
0.326
0.211
0.314
0.086
Calc
-Expt
(ppm)
0.349
0. 193
0.448
0.040
0.258
0. 179
0.258
0. 179
-0.203
-0. 158
-0.330
-0.061
-0. 230
-0.311
-O.M5
0.060
-0. i'37
-0. 386
-0.463
-0.058
-0. 049
-0.229
0. 174
0. 238
0. 160
0.098
-0.051
-0.05 1
-0.124
-0.032
0.093
0.08 1
0.036
Calc
-Expt
/AVH
0.74
0.27
0.69
0.57
0.25
0.57
0.25
-0.50
-0.80
-0.55
-0.09
-0.53
-0.46
-1.01
0.09
-0.33
-0.63
-0.43
-0.08
-0.10
-0.42
0.31
0.43
0.29
0.39
-0. 1 1
-0.14
-0 . 45
-0 .08
0.35
0.28
0.17
Average
d( [03] -
Initial
(NO] )/dt
Calc
Expt Calc -Expt
(ppb/m1n) --
1.27
2.82
1 .84
0.33
1 .57
1 .04
5.30
6.00
15.57
10. 13
5.97
10.00
4.86
3. 18
2.88
1 .96
4. 20
2. 13
1 .46
5.66
4.06
0.61
1 .59
0.84
0.84
0.97
0.43
3.06
9.84
4.59
0.56
4.51
3 .92
7.56
6.75
15.94
22. 16
7.95
8.95
2.42
3.00
2.43
1 .36
2.87
2.01
1 .45
6.53
6.29
0.69
2.02
0.83
0.69
1 .06
0.64
1 .79
7.02
2.75
0.22
2.95
2 9 1
2.95
2.91
2.26
0. 76
0.37
12.03
1 .97
-1 .05
-2.43
-0. IB
-0.44
-0.60
-1 .33
-0.12
-0.01
0.66
3.58
1 .81
3. 18
0.08
0.43
-0.01
-0.15
0 .09
0.25
0.17
0.18
Calc
-Expt
XAvg
0.63
1.11
0.86
0.50
0.62
0 25
0.82
0.25
0.35
0.12
0.02
0.75
0.28
-0 1 1
-0.67
-0.06
-0.17
-0.36
-0.38
-0.06
0.00
-0.02
0.36
0. 26
0. 24
0.12
0. 24
-O.O 1
-O. 2O
0 04
0. 19
0. 14
0.10
(cont1nued)
-------
Appendix A. Selected Results of Simulations of All Experiments Modeled Using the RADM Mechanism
(centInued).
Page 11
Experiment
Initial
Concantrat Ions
NOx HC HC/NOx
(ppm) (ppmC)
14. Ethane-NOx
ITC999
t5. n-Butana-NOx
EC130
EC133
EC 134
EC137
EC 162
EC 163
EC16B
EC 178
EC304
EC305
EC306
EC307
EC30B
EC309
ITC507
ITC533
ITC770
ITC939
ITC946
OTC211
JL2178R
JL2178B
JL2278R
JL227BB
STI87SB
OC0979R
OC1879B
Group Average
S. Dev.
Avg. Abs. Value
S. Dev.
16. C4+ Branched
EC 165
0.09
0. 10
0.50
0.51
0.50
0.51
0.49
0.49
0.10
0.47
0. 10
0. 19
0. 10
0.4B
0.47
0.09
0.12
0.52
0.51
0.26
0.55
0.24
0.24
0.55
0.55
0.21
0.21
0.20
0.34
0. 18
Alkane-NOx
0. 10
45.4 534.1
17.6 179.3
8.6 17.1
8.3 16.3
8.7 17.3
8.2 16.3
9.0 18.3
8.0 16.2
7.B 79.6
17.1 36.7
15.7 159.7
25.6 138.2
25.8 252.9
16.2 33.6
17.2 36.3
15.2 165.0
11.9 99.8
37.9 72.8
14.8 28.9
10.0 38.2
42.8 77.5
7.2 29.9
15.4 63.9
7.9 14.4
17.5 31.5
21.2 103.1
14.6 71.0
14.3 71.8
15.7 69.8
8.8 61.6
11.3 114.3
Avp
Temp
(dagK)
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
302.6
303.0
288.7
310.9
303.0
303.0
303.0
303.0
303.0
300. 1
302.9
302.9
305.2
305.2
298.3
297.2
295.0
302. 1
3.8
303.0
Ave
H20
(ppm)
20500.
26200.
27800.
20000."
20000.*
27300.
28500.
20700.
20900.
26800.
25400.
24200.
29700.
8750.
12400.
20300.
20800.
16800.
20300.
20300.
3530.
20000.
20000.
20000.
20000.
20000.
15300.
28100.
20892.
6033.
1B600.
Maximum Concentration
OZONE
Expt
(ppm)
0.243
0.459
0.249
0.034
0.042
0.112
0.454
0.655
0.384
0.362
0.398
0.535
0.420
0.047
0.545
0. 149
0. 165
0.042
0.017
0.054
0.008
0.763
0.986
0. 166
0.788
0. 185
0. 191
0. 208
0.312
0.268
0.48B
Calc
(ppm)
0.202
0.45B
0.052
0.042
0.050
0.056
0. 158
0.340
0.432
0.494
0.547
0.731
0.606
0. 240
0.678
0.397
0.356
0.059
0.028
0. 159
0.258
0.639
0.910
0. 161
0.484
0. 129
0.320
0. 193
0.332
0.245
0.482
Calc
-Expt
(ppm)
-0.042
-0.001
-0. 197
0.008
o.ooa
-0.056
-0.296
-0.315
0.049
0. 132
0. 149
0. 196
0. 186
0.193
0. 133
0.24B
0. 191
0.017
0.01 1
0. 105
0.250
-0. 124
-0.077
-0.006
-0.304
-0.056
0. 129
-0.015
0.021
0. 163
0. 128
0.101
-0.006
Calc
-Expt
/Avg
-0. 19
0.00
-1.31
-0.67
-0.97
-0.63
0. 12
0.31
0.32
0.31
0.36
0. 22
0.91
0.73
0.99
-0. IB
-0.08
-0.03
-0.48
-0.36
0.51
-0.08
0.00
0.59
0.46
0.36
-0.01
Average Initial
d( (O3] - |NO] )/dt
Calc
Expt Calc -Expt
(ppb/mln) --
1 .32
4.41
2.42
0.94
1 .02
1 .73
3.31
2.03
1 .61
2.09
2.39
2.38
2.61
1 .04
2.00
0.69
0.61
1 .56
0.36
0.53
0.56
1.17
1 . 64
0.9O
1 .55
0.51
0.60
0.60
1 .53
0.98
1 .77
1.41 0.08
2.72 -1.70
1.13 -1.29
1.36 0.42
1.33 0.32
1.27 -0.45
1.74 -1.57
1.21 -0.82
1.80 0.20
2.49 0.40
3.74 1.35
4.59 2.21
6.34 3.74
2.84 1.80
2.69 0.70
.68 1.00
.45 0.84
2.13 0.57
0.68 0.32
.17 0.64
.51 0.94
.09 -0.08
.50 -0.13
0.77 -0.13
1.33 -0.22
0.45 -0.06
0.62 0.03
0.54 -0.06
1.86 0.33
1 .32 1.13
0.81
0.84
1.60 -0.16
Calc
-Expt
/Avg
0.06
-0.48
-0.73
0.37
0.27
-0.30
-0.62
-0.51
0. 12
0. 17
0.44
0.63
0.83
0.93
0.30
0.84
0.82
0.31
0.61
0.75
0.91
-0.07
-O.OB
-0.16
-0. 15
-0. 13
0.04
-0.11
0.19
0.50
0.43
0.29
-0. 10
(cont 1nued)
-------
Appendt x A.
Selected Results
(cont Inued).
of Simulations of All Experiments Modeled Using the RADM Mechanism
Page 12
4=
VJ1
Experiment
Initial
Concentrat Ions
NOx
HC HC/NOx
(ppm) (ppmC)
EC171
OC1B79R
OC2079B
AU1983R
AU1983B
Group Average
S. Dew.
Avg. Abs. Value
S. Dev.
17. C5+ n-Alkane-NOx
EC 135
OC0979B
EC 131
ITC559
ITC538
ITC540
ITC552
ITC761
ITC762
ITC763
ITC797
ST1879R
EC155
ITC1001
Group Average
S. Dev.
Avg. Aba. Value
S. Dev.
0. 10
0.20
0.22
0.38
0.37
0.22
0.11
0. 10
0.21
0. 10
0. 19
0. 1 1
0. 1 1
0. 13
0.52
0.27
0.28
0.52
0.21
0. 10
0. 1 1
0.21
0. 14
3.5 35.7
16.4 81.9
12.6 56.5
4.7 12.6
4.1 10.9
8.2 47.9
5.2 38.7
20.4 212.7
15.1 73.3
24.6 251.1
279.4 1441.1
60.3 529.0
274.8 2421 .3
428.8 3278.4
75.2 145.9
74.7 280.4
7.7 27.7
7.3 14.0
6.3 30.4
37.3 385.1
2.4 22.4
93.9 650.9
133.5 1015.3
Ave
Temp
(degK)
303.0
295.0
295.3
307. 1
307. 1
301 .9
5.0
303.0
297.2
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
298.3
303.0
303.0
302.3
1 .9
Ave
H20
(ppm)
19000.
28100.
27800.
21300.
21300.
24143.
5463.
25000.
15200.
25000.
20300.
18900.
20300.
20300.
17500.
17000.
17100.
12200.
20000.*
26200.
17500.
19464.
3919.
Maximum Concentrat
OZONE
Expt
(ppm)
0.403
0.236
0.217
0.088
0.057
0.283
0.181
0.435
0. 184
0.393
0.377
0. 150
0.360
0.315
0.030
0. 105
0.041
0.004
0. 122
0.264
0.036
0.201
0. 153
Calc
(ppm)
0.318
0. 120
0.056
0.213
0.087
0.231
0. 157
0.549
0.226
0.528
0.601
0.313
0.384
0.358
0.036
0. 186
0.073
0.01 1
0.332
0.524
0.078
0.300
0.204
Calc
-Expt
(ppm)
-0.085
-0. 116
-0. 161
0. 125
0.030
-0.052
0. 106
0.096
0.059
0.114
0.042
0. 135
0. 224
0. 162
0.024
0.042
0.006
0.081
0.032
0.007
0.210
0.260
0.042
0.099
0.086
0.099
0.086
1 on
Calc
-Expt
/Avg
-0.24
-0.65
-1.18
0.83
0.41
-0. 17
0.67
0.53
0.39
0. 23
0.21
0. 29
0.46
0. 70
0.06
0.13
0.55
0.92
0.66
0.42
0. 28
0.42
0. 28
Average
d( (03] -
Expt
Calc
Initial
(NO) )/dt
Calc
-Expt
-- (ppb/m1n)
1 .30
0.64
0.66
0.61
0.53
0.93
0.46
2.92
0.59
1 .92
1 .79
0. 74
1 .85
1 . 19
1 .08
0.83
0.68
0.64
0.40
1 .33
0. 15
1.15
0.75
0.82
0.49
0.38
0.66
0.55
0.75
0.41
2.89
0.53
2.69
2.05
1 .05
1.12
1 .09
1 .27
1 .53
1 .45
0.98
0.45
1 .69
0.43
1 .37
0.76
-0.48
-0. 15
-0.28
0.05
0.02
-0. 18
0. 19
0.20
0. 16
-0.02
-0.06
0.77
0.26
0.31
-0.73
-0. 10
0. 19
0.70
0.77
0.34
0.05
0.36
0.28
0. 22
0.40
0.35
0.28
Calc
-Expt
/Avg
-0.46
-0.27
-0.54
0.08
0.04
-0.22
0.24
0.26
0.19
-0.01
-0. 1 1
0.33
0. 14
0.35
-0.49
-0.08
0. 16
0.59
0.72
0.42
0. 12
0.24
0.97
0.24
0.37
0.34
0.28
18. Methyl eye lohexane-NOx
ITC765
ITC766
ITC767
ITC800
Group Average
S. Dev.
Avg. Abs. Value
S. Dev.
0.53
0.26
0.55
0.54
0.47
0.14
0.0 0.1
0.0 0.1
0.1 0.1
0.0 0.1
0.0 0.1
0.0 0.0
303.0
303.0
303.0
303.0
303.0
0.0
17900.
17100.
17900.
17500.
17600.
383.
0.022
0.121
0.041
0.015
0.050
0.012
0. 169
0.031
0.01 1
0.056
-0.010
0.047
-0.010
-0.004
0.006
0.018
0.33
0.33
0.33
0.86
0.87
1 .09
0.86
0.92
0. 12
0.70
1 .37
1 .08
0.76
0.98
0.31
-0.16
0.50
-0.01
-0.09
0.06
0.30
0. 19
0. 22
-0.20
0.45
-0.01
-0. 1 1
0.03
0.29
0. 19
0. 19
(contInued)
-------
Appendix A. Selected Result* of Simulations of All Experiments Modeled Using the RAOM Mechanism
(cont1nued) .
Page 13
Experiment
19. Benzene-NOx
ITC560
ITC561
ITC562
ITC698
ITC710
ITC831
Group Average
S. Oev.
Av0. Abs. Value
S. Oev.
20. Toluene-NOx
EC264
EC265
EC266
EC269
EC270
EC271
EC272
EC273
EC327
EC336
EC337
EC339
EC340
ITC699
ITC828
JL3080R
AU2780R
AU2782B
OC2782R
AU0183R
Group Average
S. Dev.
Avg . Aba. Value
S. Dev.
Initial
Concent rat Ions
NOx
(ppm)
0.12
0, 11
0.56
0.50
0.55
1 .01
0.47
0.33
0.4B
0.48
0.49
0.47
0.46
0.21
0.48
0. 11
0.45
0.44
0.45
0.44
0.43
0.5t
1.02
0. 18
0.48
0.43
0.39
0.39
0.44
0. 18
HC
(ppmC)
332.3
79.1
83.8
83.5
83.6
12.2
1 12.4
1 t 1 .3
8. 1
7.5
8.4
4.0
4.2
8.0
4. 1
4. 1
4.0
7.2
7.9
5.0
4. 1
10.5
3.0
3.9
2.3
3.0
4.5
4.6
5.4
2.3
HC/NOx
2874.4
694.2
149.7
167.4
151 .0
12. 1
674.6
1103.2
17.0
15.6
17.0
8.4
9.0
37.4
8.5
37.2
8.9
16.3
17.7
11.3
9.5
20.8
3.0
21.3
4.8
7.0
11.7
11 .8
14.7
9.2
Ave
Temp
(degK)
303.0
303.0
303.0
303.0
303.0
303.0
303.0
0.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
304.5
302.2
301 .7
289.8
306. 1
302,5
3. 1
Ave
H20
(ppm)
17100.
17600.
18800.
21900.
21000.
18300.
19117.
1920.
18900.
21900.
19400.
19100.
20000.
20500.
19400.
21200.
22400.
26900.
27000.
26800.
25900.
20300.
17200.
35200.
32000.
24700.
19500.
23700.
23100.
4692.
Maximum Concentration
OZONE
Expt
(ppm)
0.323
0.273
0.412
0.374
0.367
0.021
0.295
0. 142
0.419
0.393
0.405
0.318
0.369
0.296
0.410
0.215
0.376
0.396
0.325
0.225
0.344
0.485
0.021
0.273
0.736
0. 1 16
0. 123
0.458
0.335
0. 154
Calc
(ppm)
0.347
0.285
0.556
0.507
0.542
0.007
0.374
0. 21 1
0.427
0.426
0.432
0.380
0.420
0.307
0.374
0.223
0.423
0,463
0.427
0.404
0.419
0.215
0.006
0.484
0.985
0.487
0.426
0.722
0.423
0. 190
Calc
-Expt
(ppm)
0.024
0.013
0. 144
0. 133
0. 176
-0.014
0.079
0.081
0.084
0.075
0.008
0.032
0.026
0.063
0.051
0.012
-0.036
0.008
0.047
0.067
0. 102
0. 179
0.074
-0. 270
-0.015
0.210
0. 249
0.371
0.303
0.264
0.087
0. 143
0.119
0.116
Calc
-Expt
/Avg
0.07
O.OS
0.30
0.30
0.39
0.22
0. 15
O.22
0.15
0.02
0.08
0.06
0. 16
0. 13
0.04
-0.09
0.04
0.12
0. 16
0. 27
0.57
0. 19
-0.77
0.56
0.29
1 .23
1 . 10
0.45
0.24
0. 43
0. 33
0.36
Average
d( [03] -
Initial
(NO) )/dt
Calc
Expt Calc -Expt
(ppb/mln) --
7.01
4.75
2.85
2. 87
2.70
0.17
3.39
2.30
4.46
3.56
4.62
2.55
3.72
6.56
3.69
5.90
2.49
6.06
2.55
1 .53
2.50
4. 72
0.49
1 .20
1 .76
0.63
0.66
1.75
3.07
1 .87
9.61
4,79
4.40
4.37
4.21
0.30
4.61
2.96
4.94
4.83
5. 10
2.78
4. 16
4.92
2.61
3.09
3.05
6.82
4. 22
3.12
2.90
2.34
0.88
1. 10
1 .65
0. 72
0.88
1 .88
3. 11
1 .68
2.59
0.04
1 .55
1 .49
1.51
0. 13
1 .22
0.97
1 .22
0.97
0.49
1 .27
0.48
0.23
0.44
-I .64
-0.88
-2.81
0.56
0. 76
1 .67
1 .59
0.40
-2.38
0.39
-0. 10
-0. 1 1
0.08
0.22
0. 13
0.04
1.17
0.83
0.80
Calc
-Expt
/Avg
0.31
0.01
0.43
0.41
0 .44
0.56
0.36
0.19
0.36
0. 19
0.10
0.30
0.10
0.09
0.11
-0.29
-0. 27
-0.63
0.20
0.12
0 .49
0.68
0. 15
-0.67
0.57
-0.08
-0.06
0.12
0. 28
0.07
0,07
0.34
0. 27
0.22
21. Xylene-NOx
(cont1nued)
-------
Appendix A.
Selected Results
(continued).
of Simulations of All Experiment* Mod*lad Using the RADM Mechanism
Page 14
U>
Experiment
Initial
Concent rat 1 ons
NOx
HC
HC/NOx
(ppm) (ppmC)
EC343
EC344
EC345
EC346
ITC702
ITC827
JL3080B
AU2782R
OC27B2B
AU0183B
Group Average
S. Oev.
Avg. Abs . Value
S. Dav.
22. Mes1tylene-N0x
EC900
EC901
EC903
ITC703
ITC706
ITC709
ITC742
1TC826
Group Average
S. Dev.
Avg. Abs. Value
S. Dev.
23. Tetral1n-N0x
ITC739
ITC747
ITC748
ITC750
ITC832
Group Average
S. Dev.
Avg. Abs. Value
S. Dev.
0.28
0.67
0.28
0.26
0.52
1 .07
0. 18
0.43
0.39
0.37
0.4S
0.26
0.53
0.51
1 .00
0.50
0.49
0.99
0.48
0.90
0.68
0.24
0.52
0.50
0.22
0.53
1.00
0.55
0.28
4
4
3
3
4
1
2
2
2
2
3
1
5
2
4
5
2
4
4
0
3
1
2
93
84
44
39
52
36
.2
.0
.7
.9
.0
. 2
. 2
.0
.8
.7
. 1
.0
.4
. 7
.7
.3
. 7
.7
.6
.6
.9
. 6
.4
. 2
.0
.5
.4
.7
.7
14.9
5.9
13.3
14.8
7.8
1 . 1
12.4
4.6
7.0
7.2
8.9
4.7
10.2
5.2
4.7
10.6
5.4
4.7
9.7
0.9
6.4
3.4
4.6
187 .4
385. 1
84.6
39.5
140.2
153.2
Ave
Temp
(degK)
303.0
303.0
303.0
303.0
303.0
303.0
304.5
301 .7
289.8
306.0
302.0
4 .4
303.0
303.0
303.0
303.0
303.0
303.0
303,0
303,0
303.0
0.0
303.0
303.0
303.0
303.0
303.0
303.0
0.0
Ave
H20
(ppm)
29500.
28900.
25200.
25900.
20300.
14400.
28000.
24700.
19500.
23700.
24010.
4737.
20000.
20000.*
20000. *
21000.
21000.
21000.
18100.
17200.
19788.
1419.
17900.
17100.
17500.
17100.
16800.
17280.
427.
Maximum Concentration
OZONE
Expt
(ppm)
0.283
0.589
0.396
0.384
0.627
0.021
0.555
0.491
0.396
0.688
0.443
0. 195
0.381
0.384
0.502
0.707
0.641
0.779
0.773
0.022
0.524
0.258
0.002
0.508
0.370
0.482
0.073
0.287
0.235
Calc
(ppm)
0.354
0.515
0.362
0.349
0.338
0.009
0.519
0.601
0.422
0.777
0.425
0.202
0.417
0.402
0.591
0.373
0.318
0.214
0.406
0.007
0.341
0.171
0.202
0.659
0.484
0.599
0.025
0.394
0.271
Calc
-Expt
(ppm)
0.071
-0.073
-0.033
-0.034
-0.2B9
-0.012
-0.036
0.111
0.026
0.089
-0.018
0. 1 13
0.077
0.081
0.036
0.019
0.089
-0.334
-0.322
-0.565
-0.367
-0.015
-0. 183
0.243
0.218
0.206
0.200
0. 150
0.114
0.117
-0.048
0. 107
0.093
0. 126
0.056
Calc
-Expt
/Avg
0.22
-0. 13
-0.09
-0.09
-0.60
-0.07
0.20
0.06
0. 12
-0.04
0. 25
0. IB
0. 17
0.09
0.05
0.16
-0.62
-0.67
-1.14
-0.62
-0. 39
0.50
0.48
0.40
0.26
0. 27
0.22
-0.98
-0.06
0.62
0.43
0.37
Average
d( f03] -
Expt
Calc
Initial
[NO] )/dt
Calc
-Expt
(ppb/mln) --
8. 18
10.72
11 .35
7.65
7.79
1.12
1.97
1.31
1 .43
2.39
5.39
4. 14
3.85
8.88
14.96
14.59
7.20
1 1 .74
13.14
1 .68
9.50
4.97
0.63
2.73
2.60
2.10
1 .38
1 .89
0.88
9.39
9.23
9.13
9.14
3. 10
1 .25
1.96
1 .69
2.04
2.91
4.98
3.69
5.07
8.16
3.49
2. 13
3. 10
2.95
0.85
3.68
2.36
1 .47
31.12
20. 16
18.59
0.99
14.47
13.01
1.21
-1 .50
-2.22
1 .49
-4.69
0.13
-0.01
0.38
0.61
0.52
-0.41
1 .89
1 .28
1 .39
-3.81
-6.80
-11.10
-5.08
-8.64
-10.20
-0.83
-6.64
3.67
6.64
3.67
0.84
28.40
17.56
16.48
-0.38
12.58
12.21
12.73
12.01
Calc
-Expt
/Avg
0.14
-0.15
-0.22
0. 18
-0.86
0.11
0.00
0.26
0.35
0.20
0.00
0.35
0. 25
0. 23
-0.55
-0.59
- 1 . 23
-1 .09
-1.16
-1 .27
-0. 66
-0.93
0.32
0.93
0.32
0.60
1 .68
1 .54
1 .59
-0.33
1 .06
0.85
1 . 19
0.60
(cont\nued )
-------
Appendix A.
Selected Reaults of Simulations of AM
(continued).
Experiment* Modeled Using the RAOM Mechanism
Page IS
Xr
CO
Experiment
Initial
Concent rat
NOx HC
(ppm) (ppmC)
Ions
HC/NOx
Ave
Temp
(degK)
Ave
H20
(ppm)
Maximum Concentrnt
OZONE
Expt
(ppm)
Calc
(ppm)
Calc
-Expt
(ppm)
1 on
Calc
-Expt
/Avg
Average Initial
d( |03] - [NO] )/dt
Ca Ic Ca 1 c
Expt Calc -Expt -Expt
(ppb/m1n) -- /Avg
24. Naphtha lene-NOx
ITC751
ITC755
ITC756
ITC79B
ITC802
Group Average
S. Dev.
Avg. Abe. Value
S. Dev.
0.52
0.24
0.26
0.53
0.53
0.42
0.15
7.5
14. 1
27.4
19.4
8.4
15.4
6.3
14.3
58.3
106.8
36.8
15.9
46.4
38.2
303.0
303.0
303.0
303.0
303.0
303.0
0.0
15100.
17500.
17200.
16400.
16800.
16600.
93b.
0.113
0.259
0.282
0.204
0. 124
0. 196
0.077
0.505
0.392
0.451
0.521
0.503
0.474
0.053
0.392
0.132
0. 168
0.317
0.379
0.278
0. 120
0.278
0. 120
1 .27
0.41
0.46
0.88
1 .21
0.84
0.40
0.84
0.40
1 .08
1.57
2.20
1 .66
1 .42
1.59
0.41
4.76 3.68 1.26
8.68 7.11 1.39
13.91 11.71 1 .45
10.47 8.80 1.45
5.34 3.91 1.16
8.63 7.04 1.34
3.78 3.39 0.13
7.04 1.34
3.39 0.13
25. 2. 3-D1methy1naphtha lene-NOx
ITC775
ITC771
ITC806
ITC774
Group Average
S. Dev.
Avg. Abs. Value
S. Dev.
26. Simple Mixture
EC 144
EC145
EC 160
EC 149
EC150
EC151
EC152
EC153
EC161
OC1278R
OC2578R
AU0180R
AU14BOR
EC166
EC172
ST0682R
0.29
0.26
0.33
0.56
0.36
0. 14
Runs
0.51
0.99
0.99
0.99
1 .00
2.06
0.50
0.97
0.51
0.48
0.44
0.56
0.47
0.10
0. 10
0.46
1 .7
4.8
5.9
4.0
4. 1
1 .8
4.7
3.4
3.2
2.0
3.5
5.2
3.7
6.6
3.2
1 .4
1 .4
0.5
1 .4
9.2
2.8
2.8
5.8
18.0
17.7
7. 1
12.2
6.6
9.3
3.4
3.3
2.0
3.5
2.5
7.3
6.8
6.4
3.0
3. 1
0.8
3.0
92.0
28.9
6.2
303.0
303.0
303.0
303.0
303.0
0.0
303.0
303.0
303.0
303.0
303. U
303.0
303.0
303.0
303.0
294.3
288.4
304.9
305.2
303.0
303.0
299. 1
16800.
20300.
17100.
22700.
19225.
2806.
24400.
24400.
29200.
29300.
29900.
30400.
26400.
25400.
36900.
16100.
12800.
23400.
34700.
27700.
21600.
22200.
0.274
0.293
0.360
0.341
0.317
0.040
1 .065
0.777
0.874
0. 286
0. 799
0. 147
0. 791
1 .050
U.B57
0.260
0. 147
0. 256
0.863
0.462
0.369
0.378
0. 185
0.339
0.388
0.313
0.306
0.087
0.773
0.365
0.316
0. 125
0.327
0.082
0.665
0.820
0.600
0.250
0. 125
0. 107
0.484
0.395
0.310
0.619
-0.089
0.045
0.028
-0.028
-0.011
0.061
0.048
0.029
-0.292
-0.411
-0.559
-0.161
-0.472
-0.064
-0. 126
-0.230
-0.25B
-0.010
-0.022
-0. 148
-0.379
-0.067
-0.059
0.240
-0.39
0. 14
0.08
-0.08
-0.06
0.24
0. 17
0. 15
-0.32
-0.72
-0.94
-0.78
-0.84
-0.56
-0.17
-0.25
-0.35
-0.04
-0.16
-O.B2
-0.56
-0. 16
-0.17
0.48
1 .73
2.67
2.69
2.84
2.48
0.51
10.53
5.11
5.86
10.83
6.30
8.43
10.41
19.23
9.94
1 . 14
0.94
0.98
2.13
2. 1 1
1 .00
1.11
1 .28 -0.44 -0. 29
3.20 0.53 0.18
3.35 0.66 0.22
2.61 -0.23 -0.08
2.61 0.13 0.01
0.94 0.55 0.24
0.47 0.19
0.18 0.09
6.09 -4.44 -0.53
3.71 -1.40 -0.32
3.37 -2.49 -0.54
5.12 -5.71 -0.72
4.07 -2.22 -0.43
6.84 - 1 .58 -0.21
7.93 -2.48 -0.27
12.23 -7.00 -0.44
5.42 -4.53 -0.59
1.22 0.08 0.07
1.01 0.07 0.07
0.64 -0.34 -0.42
1 .39 -0. 75 -0.42
1.50 -0.61 -0.34
0.73 -0.27 -0.31
1.40 0.29 0.23
(cent 1 nued)
-------
Appendix A. Selected Reaults of Simulations of All Experiments Modeled Uslns the RADM Mechanism
(continued).
Page 16
to
jr
vO
Exper Iment
Initial
Concent rat Ions
NOx
Ave
Temp
Ave
H20
HC HC/NOx
(ppm) (ppmC)
EC 106
EC1 13
eci 14
eci 15
ECI 16
EC335
EC329
EC330
EC334
EC338
EC328
JL1581R
JL1881R
ST24B1R
AU2781R
OC0382R
OC0382B
NV1582R
JL2281R
JL21B1B
NV15B2B
DE0782B
JN1379B
Group Average
S. Dev.
Avg. Abs. Value
S. Dev.
27. 3-Component
ITC479
ITC584
ITC579
ITC472
ITC474
ITC581
ITC5B5
ITC47B
ITC482
ITC488
ITC492
ITC494
ITC498
ITC500
ITC502
0.50
0.11
1 .00
0.51
0.49
0.44
0.45
0.29
0.45
0.45
0.45
0.27
0.26
0.23
0.23
0.25
0.25
0. 18
0.26
0.24
0.18
0.19
0 .44
0.49
0.37
and "M1n1"
0.09
0.10
0.10
0.10
0.09
0.09
0.10
0.10
0.10
0.09
0.09
0.09
0. 10
0. 10
0.09
9.2
9.5
17.3
12.7
16.6
7.7
4.2
4.3
8.1
15.0
12.1
2.3
2. 1
1 .6
2. 1
2.7
1 .9
2.6
2.4
1 .3
2.7
3.5
6. 1
5.2
4.5
18.3
85.0
17.3
25.2
37.6
17.4
9.2
14.6
18.2
33.7
27.1
6.4
8.0
6.7
9.2
11.0
7.7
14.4
9. 1
5.4
14.9
1&.7
13.7
15.5
19.2
(degK)
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
304.6
303.2
292.2
299.5
297.2
296.9
283.8
302.3*
304. 2*
283.8
286.4
294.9
300.2
5.7
(ppm)
20900.
19500.
28800.
22000.
27300.
24700.
27700.
26000.
27800.
27800.
27500.
20000.*
20000.*
22800.
30900.
21200.
21 100.
20000.*
20000. *
20000.*
20000.*
13400.
20000.*
24160.
5188.
Maximum Concent rat
OZONE
Expt
(ppm)
0.592
0.352
0. 744
0.590
0.743
0.398
0.403
0,344
0.408
0.484
0.523
0.485
0.658
0.211
0.554
0.247
0. 158
0.061
0.626
0. 237
0. 167
0.093
0.756
0.492
0.270
Calc
(ppm)
0.626
0.441
0.721
0.630
0.786
0.426
0.444
0.380
0.456
0.549
0.564
0.694
0.597
0.269
0.498
0. 145
0.039
0.046
0.635
0.437
0.246
0.304
0.80B
0.443
0.226
Calc
-Expt
(ppm)
0.033
0.089
-0.023
0.040
0.044
0.027
0.041
0.036
0.04B
0.065
0.041
0.209
-0.062
O.U58
-0.056
-0.102
-0.119
-0.015
0.009
0.200
0.079
0.212
0.052
-0.049
0. 183
0. 133
0. 134
1on
Calc
-Expt
/Avg
0.05
0 . 23
-0.03
0.06
0.06
0.07
0. 10
0.10
0.11
0. 13
0.08
0.35
-0. 10
0.24
-0.11
-0.52
-1.21
-0.29
0.01
0.59
0.38
1 .07
0.07
-0.12
0.45
0.34
0.32
Average
d( [03] -
Expt
Calc
Initial
[NO] )/dt
Calc
-Expt
(ppb/m1n) --
3.80
5. 26
6.03
3. 78
8. 28
4.64
3.27
3.91
5.59
5. 19
3.50
1 .08
1.31
0.62
0.95
0.76
0.62
0.42
1 . 18
0.63
0.71
0.55
2.44
4.10
4.06
3.95
6.93
5.89
3.41
8 .99
4.43
3.43
3.47
5. 14
4.96
3. 58
1 .21
1 .21
0.65
0.92
0.55
0.34
1.12
0.64
0.76
0.74
2.77
3.32
2.76
0.15
1 .67
-2.14
-0.36
0.70
-0. 22
0.16
-0 .44
-0.45
-0.24
0.08
0.13
-0. 10
0.03
-0.04
-0.21
-0.08
-0.06
0.01
0.05
0. 19
0.33
-0.87
1.81
1 . 09
1 . 68
Calc
-Expt
/Avg
0.04
0.27
-0.31
-0. 10
O.OB
-0.05
0.05
-0.12
-0.08
-0.05
0.02
0.12
-0.08
0.05
-0.04
-0.32
-0.22
-0.05
0.02
0.07
0.30
0.13
-0.13
0 . 26
0 . 22
0.18
Surrogates
4.0
4.0
3.4
3.6
4.2
4.3
4.8
5.0
1 .2
6.0
8. 1
8.5
3.9
4. 1
4.6
43.4
40.7
34.7
35. 1
45. 1
50.0
50.8
52.6
12.3
66.2
88. 4
92.5
40.2
42.7
49.5
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
20300.
20300.
15800.
13700.
15600.
16100.
17100.
18300.
18300.
19600.
15000.
21 100.
19300.
1B700.
18100.
0.332
0.375
0.356
0. 258
0.293
0.352
0.311
0.320
0. 279
0.293
0.326
0.321
0.320
0.286
0. 271
0.304
0.303
0. 270
0.304
0.320
0.307
0.325
0.344
0. 277
0.315
0.340
0.349
0.317
0.305
0.291
-0.028
-0.071
-0.086
0.046
0.027
-0.045
0.014
0.024
-0.002
0.021
0.014
0.027
-0.003
0.019
0.021
-0.09
-0.21
-0.28
0.16
0.09
-0.14
0.04
0.07
-0.01
0.07
0.04
0.08
-0.01
0.06
0 .07
2.95
3.05
1 .54
1 .72
3.40
3.74
5.11
4.73
2.44
3.06
3. 28
3.30
2. 16
5. 18
7.07
3.47
2.80
1 .25
1 .75
4 .84
4.41
6.10
7.30
3.00
4.00
4.30
4.36
2.21
6.84
12.43
0.52
-0. 25
-0.28
0.03
1 .44
0.67
0.99
2.58
0.55
0.94
1 . 02
1 .06
0.05
1 . 65
5.35
0.16
-0.08
-0.20
0.02
0.35
0.17
0.18
0.43
0 . 20
0 . 27
0 . 27
0. 28
0 .02
0.28
0.55
(cont1nued)
-------
Appendix A.
Selected Results
(contInued).
of Simulations of AH Experiment* Modeled Using the RADM Mechanism
Page 17
U>
Ul
O
Experiment
ITC466
ITC468
ITC451
ITC455
ITC977
ITC9B5
ITC997
ITC979
ITC992
ITC961
ITC993
JL15B1B
ST24818
JN0982B
JN1463R
JN2783B
AU18B3B
AU2683R
JL1B81B
JN09B2R
Group Average
S. Oev.
Avg . Aba. Value
S. Dev.
Initial
Concent rat Ions
NOx
(ppm)
0. 10
0.09
0. 10
0.09
0. 13
0.12
0.12
0.12
0. 12
0. 14
0.11
0.28
0.23
0.2B
0.22
0.26
0.28
0.32
0.27
0.29
0.14
0.08
Ave
Temp
Ave
H20
HC HC/NOx
( ppmC )
6.5
4.6
S.2
4.4
3.2
2.7
2.7
2.9
2.6
6.5
7. 1
2.3
1.9
3.1
2.6
2.9
0.6
2.6
2.3
3.1
4, 1
1 .9
62.7
49.0
52.0
47.8
25. 1
22.9
23.0
23.1
20.9
45.0
66.9
8.0
8.1
11 .0
11.6
11.1
2.0
8.1
8.5
10.7
37.0
23.7
(degK)
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
304.6
292.2
299.7*
301 .2
305. 1
303.0
302.2
303.2
299.7*
302.6
2.0
(ppm)
18200.
18000.
19BOO.
20200.
20900.
20100.
20100.
20100.
20900.
19000.
20100.
20000.*
14200.
14300.
26100.
24400.
23100.
26400.
20000.*
14700.
19061 .
3028.
Maximum Concentration
OZONE
Expt
(ppm)
0. 131
0.176
0.312
0.304
0.329
0.336
0.316
0.404
0.384
0.32B
0.276
0.474
0.246
0.667
0.585
0.511
0.556
0.646
0.693
0.714
0.366
0. 148
Calc
Calc -Expt
(ppm) (ppm)
0.274 0.143
0.290 0.114
0.295 -0.016
0.292 -0.012
0.312 -0.017
0.282 -0.054
0.281 -0.035
0.401 -0.003
0.393 0.009
0.360 0.033
0.347 0.071
0.657 0.184
0.342 0.095
0.604 -0.063
0.558 -0.027
0.633 0.123
0.578 0.023
0.647 0.001
0.644 -0.049
0.647 -0.067
0.383 0.016
0.134 0.066
0.049
0.047
Calc
-Expt
/Avg
0.71
0.49
-0.05
-0.04
-0.05
-0. 17
-0. 12
-0.01
0.02
0. 10
0. 23
0.32
0.32
-0. 10
-0.05
0.21
0.04
0.00
-0.07
-0. 10
0.07
0.24
0. 15
0. 19
Average Initial
d( [03) - (NO) )/dt
Expt
Calc
Calc -Expt
(ppb/mln) --
0.77
1 . 19
3.95
3.52
3.94
3.95
3.89
4.B8
5.57
2.56
2.45
1 . 10
0. 71
.37
. 10
.01
0.70
. 24
.91
.77
2.81
1 .61
2.57 .80
2.80 .61
3.92 -0.03
3.55 0.03
2.36 - .58
2.17 - .78
2.15 - .74
3.21 - .66
3.59 - .99
2.43 -0.13
2.06 -0.39
1.17 0.07
0.73 0.02
.22 -0.15
.10 -0.01
.16 0.15
0.66 -0.04
.21 -0.03
.81 -0.10
1.60 -0.17
3.13 0.33
2.28 1.37
0.90
1.07
Calc
-Expt
/Avg
1 .08
0.81
-0.01
0.01
-0.50
-0.56
-0.5B
-0.41
-0.43
-0.05
-0. 17
0.06
0.03
-0. 12
0.00
0. 13
-0.05
-0.03
-0.05
-0. 10
0.08
0.38
0.27
0.26
2B, SAPRC 7-Component Surrogates
EC231
EC232
EC233
EC237
EC23B
EC241
EC242
EC243
EC245
EC246
EC247
Group Average
S. Dev.
Avg. Aba. Value
S. Oev.
0.49
0.49
0. 10
0.48
0.95
0.49
0.50
0.50
1.00
0.51
0.51
0.55
0.24
13.2
9.3
9.5
10.5
10. 1
5.0
12.9
9.7
12.9
8.6
6.2
9.8
2.6
26.9
18.9
92.5
21.6
10.6
10.2
25.6
19.5
12.9
17.0
12.2
24.4
23.3
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
0.0
17400.
17100.
21400.
23200.
25400.
20700.
24600.
16400.
20400.
22100.
22700.
21036.
3020.
0.620
0.342
0.330
0.652
0.691
0.40B
0.6B2
0.716
0.894
0.574
0.657
0.597
0. 173
0.698 0.07B
0.408 0.066
0.381 0.051
0.639 -0.013
0.708 0.017
0.407 -0.001
0.65B -0.074
0.700 -0.016
0.821 -0.073
0.474 -0.100
0.606 -0.052
0.591 -0.006
0.149 0.056
0.044
0.032
0.12
0. 18
0. 14
-0.02
0.02
0.00
-0.04
-0.02
-0.08
-0. 19
-0.08
0.00
0. 1 1
0.08
0.07
6.99
3. 12
4.07
7.36
5. 10
2.87
17.53
14.50
13.41
2.65
7.49
7.74
5. 16
7.72 0.73
3.04 -0.08
4.67 0.60
6.06 -1.30
4.98 -0.12
3.03 0.16
16.87 -0.66
13.10 -1.41
12.91 -0.50
2.09 -0.57
6.49 -1.00
7.36 -0.38
4.85 0.71
0.65
0.45
0. 10
-0.03
0. 14
-0. 19
-0.02
0.06
-0.04
-0. 10
-0.04
-0.24
-0.14
-0.05
0. 12
0. 10
0.07
(cont1nued)
-------
Appendix A. Selected Results of Simulations of All Experiments Modeled Using the RAOM Mechanism
(continued).
Page 18
UJ
VJl
Experiment
Initial
Concentrat
1 ons
NOx HC HC/NOx
(ppm) (ppmC)
Ave
Temp
(degK)
Ave Maximum Concentration
H20 OZONE
Expt
(ppm) (ppm)
Calc
(ppm)
Calc
-Expt
(ppm)
Calc
-Expt
/Avg
Average
d( (03] -
In1t 1al
[NO] )/dt
Calc
Expt Calc -Expt
(ppb/mln) --
Calc
-Expt
/AvB
29. SAPRC 8-Component Surrogates
ITC626
ITC630
ITC631
ITC633
ITC635
ITC637
1TC665
ITC867
ITC868
ITC871
ITC872
ITCB73
ITC874
ITCB77
ITC880
ITCSBl
:TC885
ITC886
ITC888
ITC891
OTC189A
OTC189B
OTC190A
OTC190B
OTC192A
OTC192B
OTC194A
OTC194B
OTC195A
OTC195B
OTC196B
OTC197A
OTC197B
OTC19BA
OTC19BB
OTC199A
OTC199B
OTC202A
OTC202B
OTC203A
OTC203B
OTC204A
OTC204B
OTC205A
OTC205B
0.30
0.31
0.32
0.64
1 .21
0.31
0.28
0. 28
0.37
0.37
0.38
0.39
0.38
0.38
0.73
0.73
0.64
0.73
0.33
0.32
0.45
0.45
0.41
0.42
0.47
0.47
0.38
0.38
0.41
0.41
0.41
0.38
0.77
0.64
0.40
0.36
0.37
0.73
0.40
0.39
0. 19
0.35
0. 17
0.84
0. 14
4.0
1 .9
1 .0
4.0
4,0
4.0
8.4
4.8
2.9
1 .7
2. 1
1 .3
2. 1
2.3
2.2
2.3
1 .5
2.3
4.7
4.4
3.3
4.0
3.8
3.8
4. 2
4. 2
7.4
3.7
1 .8
4. 1
3.7
3.6
3. 1
5.4
3.7
3.5
3.5
6.0
2.6
3.5
3,5
3.6
3.5
3.8
3.6
13.4
6.3
3.2
6.2
3.3
12.8
29.9
17.2
7.8
4.6
5.7
3.4
5.7
6.2
3.0
3.1
2.4
3.1
14.5
13-. 7
7.4
8.9
9.2
9.0
8.9
8.9
19.3
9.7
4.5
9.9
9. 1
9.5
4. 1
6.4
9.4
9.6
9.6
8.2
6.7
9.0
18.5
10.2
20.3
4.5
26.5
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
295.4
298.0
302.7
299.5
297.8
298. 4
296. 1
295.7
351 .6
360.0
337.5
353.2
358.0
353 .5
347.5
299.7
298 .9
325. 1
324. 1
293.3
293.7
322.2
348.8
464.9
433.4
14300. 0.618
16300. 0.284
17500. 0.043
16600. 0.231
17500. 0.006
16900. 0.617
15100. 0.632
14600. 0.631
17100. 0.518
14600. 0.376
14600. 0.213
14600. 0.160
13200. 0.191
14800. 0.250
13700. 0.031
14000. 0.012
14600. 0.012
20000.* 0.012
18000. 0.579
20000.
5000.
5000.
5400.
5410.
5000.
5000.
5000.
5000.
4660.
3180.
5000.
5000.
5000.
5000.
5000.
5000.
5000.
5000.
4560.
4850.
4880.
4330.
4330.
5000.
5000.
0.602
0.576
0.712
0.702
0.669
0.735
0.754
0.737
0.603
0. 164
0.681
0.597
0.621
0. 152
0.704
0.674
0.591
0.606
0.714
0.522
0.231
0.394
0.302
0.381
0.039
0.392
0.531
0.265
0.069
0.267
0.022
0.529
0.491
0.532
0.409
0.259
0.234
0. 1 19
0. 164
0.243
0.051
0.037
0.029
0.024
0.505
0.494
0.789
0.869
0.867
0.818
0.874
0.798
0.890
0.713
0. 126
0.789
O.B30
0.563
0. 130
0.575
0.640
0.670
0.654
0.509
0.441
0.291
0.436
0.373
0.460
0.075
0.542
-0.087
-0.019
0.027
0.036
0.016
-0.088
-0. 141
-0.099
-0. 109
-0. 1 16
0.022
-0.041
-0.026
-0.008
0,020
0.025
0.018
0.012
-0.074
-0. 108
0.214
0. 157
0. 164
0. 149
0. 139
0.045
0. 152
0.110
-0.038
0. 108
0.233
-0.058
-0.021
-0. 129
-0.034
0.079
0.048
-0.205
-0.080
0.060
0.041
0.071
0.079
0.035
0. 150
-0.15
-0.07
0.14
-0. 15
-0.25
-0. 17
-0.24
-0.37
0.10
-0.29
-0.15
-0.03
-0. 14
-0. 20
0.31
0. 20
0.21
0. 20
0.17
0.06
0. 19
O. 17
-0. 27
0. 15
0.33
-0. 10
-0. 15
-0. 20
-0.05
0.12
0.08
-0. 34
-0.17
0. 23
0.10
0.21
0. 19
0.32
2. 18
0.86
0.47
1 .26
0.95
2.31
2.60
2.23
2. 10
1 . 18
1 .20
0.85
0.83
1 . 14
0.85
1 .03
0.44
1 .24
1 .65
2.67
3.79
3.70
4.24
3.73
4 .25
3.84
7.85
2.95
2.05
4.14
3.86
4.12
2.86
6.09
3.86
3 .86
3.40
6.95
3.82
2.34
2. 20
3.27
2.62
4.85
4.49
1 .57
0.80
0.44
1.12
0.78
1 .57
2.44
2.75
1 .20
0.97
1 .20
0.71
0.79
1 . 19
0.86
0.99
0.68
0.63
1.53
2.07
3.48
3.63
3.27
2.88
3.32
2.77
7.07
2.93
1.21
3.66
3. 26
2.59
2.07
3.54
2.50
2.95
2.84
1 1 .67
3.10
2.21
2.46
3.20
2.71
2. 16
3.45
-0.60
-0.06
-0.03
-0.15
-0. 17
-0.74
-0. 15
0.52
-0.90
-0.21
0.01
-0. 13
-0.04
0.05
0.01
-0.04
0.25
-0.61
-0.12
-0.60
-0.31
-0.07
-0.98
-0.85
-0.93
-i .07
-0.78
-O.O2
-0.84
-0.48
-0.60
-1 .52
-0.80
-2.56
-1 .36
-0.91
-0.56
4.71
-0.72
-0.13
0.26
-0.06
0.09
-2.69
-1 .04
-0.32
-0.07
-0.06
-0. 12
-0.20
-0.3B
-0.06
0.21
-0.55
-0.20
0.01
-0. 17
-0.05
0.04
0.01
-0.04
0.44
-0.65
-0.08
-0. 25
-0.09
-0.02
-0. 26
-0. 26
-0. 25
-0.32
-0.10
-0.01
-0.51
-0.12
-0.17
-0.45
-0.32
-0.53
-0.43
-0.27
-0.18
0.51
-0.21
-0.06
0.11
-0.02
0.03
-0.77
-0.26
(cont1nu«d)
-------
Appendix A. Selected Result* of Simulation* of All Experiments Modeled Using the RADM Mechanism
(continued).
Page 19
rv>
Experiment
Initial
Concentrat Ions
NOx
Ave
Temp
HC HC/NOx
(ppm) (ppmC)
OTC215B
OTC217A
OTC217B
OTC221A
OTC221B
OTC222A
OTC222B
OTC223A
OTC223B
OTC224A
OTC224B
OTC226A
OTC228A
OTC228B
OTC229A
OTC229B
OTC230A
OTC230B
OTC237A
OTC237B
OTC238A
OTC238B
OTC239A
OTC239B
OTC240A
OTC240B
OTC241A
OTC241B
OTC242A
OTC242B
OTC243A
OTC243B
OTC248A
OTC248B
OTC249A
OTC249B
Group Average
S. Dev.
Avg. Abs. Value
S. Dev.
30. UNC Mlsce) 1
ST2081R
DE0782R
0.44
0.50
0.51
0.41
0.42
0.44
0.43
0.36
0.40
0.34
0.34
0.45
0.41
0.41
0.46
0.46
0.41
0.41
0.52
0.52
0.50
0.50
0.49
0.50
0.50
0.50
0.32
0.31
0.45
0.46
0.47
0.38
0.48
0.46
0.48
0.46
0.45
0.16
4. 1
4.9
4.6
2.4
2.0
3.4
2.6
4.8
3.5
4.4
4.3
2.5
2.4
2.3
3.0
1 .7
3.1
1 .7
4.3
4.4
2.9
3.9
2.7
2.5
1 .8
2.0
4.7
4.0
2.6
2.1
4.7
3.7
3.7
2.9
5.6
4.9
3.5
1 .3
9.3
9.6
9.0
5.8
4.7
7.8
6.0
13.4
8.8
12.7
12.6
5.5
6.0
5.6
6.6
3.8
7.6
4.2
8.2
8.5
5.8
7.8
5.4
5.0
3.6
4. 1
14.8
12.9
5.8
4.7
10.0
9.7
7.7
6.3
1 1.8
10.7
8.7
4.9
(degK)
293.4
295.5
294.2
290.6
288.8
298.5
296. 1
314.9
314.7
303.5
303.8
312.5
293.2
294.3
295.7
296.0
297.9
298.2
292.3
292.8
292.5
292.0
290.2
290.9
93.4
104.2
290.7
291 . 1
302.0
302. 1
276.5
277.4
271 .6
268.9
286.9
286.2
302.3
43.6
Ave Maximum Concentrat
H20 OZONE
Expt
(ppm) (ppm)
4420.
7460.
7460.
4460.
4460.
5000.
5000.
4270.
4270.
6410.
6410.
6840.
5000.
5000.
5000.
5000.
5000.
5000.
3810.
3810.
4810.
4810.
5150.
5150.
6210.
6210.
4760.
4760.
4670.
4670.
4720.
4720.
5000.
5000.
3540.
0.866
0.483
0.831
0.235
0.333
0.909
0.940
0.953
0.771
0.776
0.813
0.751
0.246
0. 296
0.253
0. 168
0.489
0.271
O.B07
0.757
0.406
0.702
0.343
0.234
0.034
0.217
0.671
0.674
0. 182
0.639
0.142
0. 152
0.056
0.081
0.348
3540. 0.325
7630. 0.454
4865. 0.272
Calc
(ppm)
0.937
0.633
0.937
0.312
0.372
1.030
1.169
0.893
0.756
0.804
0.847
0.598
0.483
0.507
0.517
0.460
0.873
0.624
0.865
0.739
0.441
0.738
0.449
0.212
0.049
0.153
0.694
0.696
0.21 1
0.694
0. 202
0. 180
0.069
0. 103
0.360
0.392
0.498
0.291
Calc
-Expt
(ppm)
0.069
0. 151
0.105
0.077
0.039
0. 121
0.229
-0.060
-0.015
0.029
0.035
-0. 153
0.237
0.211
0.264
0.292
0.384
0.353
0.058
-0.018
0.035
0.036
0. 106
-0.022
0.015
-0.064
0.023
0.022
0.029
0.055
0.060
0.028
0.013
0.022
0.012
0.067
0.044
0.111
o.oea
0.081
1on
Calc
-Expt
/Avg
0.08
0.27
0.12
0.28
0.11
0.12
0.22
-0.06
-0.02
0.04
0.04
-0.23
0.65
0.53
0.69
0.93
0.56
0.79
0.07
-0.02
0.08
0.05
0.27
-0. 10
-0.35
0.03
0.03
0. 15
0.08
0.35
0.17
0. 21
0. 24
0.03
0.19
0.09
0. 26
0. 20
0.18
Average Initial
d( [03] - [NO] )/dt
Expt
Calc
Calc -Expt
(ppb/m1n)
5.33
2.55
4.64
1 .63
1 .63
3.83
5.55
4.55
3. 15
3.67
4.94
3.12
1 .49
1 .79
1.56
1 .38
2.43
1 .66
4.10
3.48
2.58
4. 06
2.b6
1.85
1 .08
1 .79
3.26
3.30
1 .47
3.04
1.41
1 .40
1 .40
1 .61
2.55
2.83
2.81
1.51
5.33 0.00
2.51 -0.05
4.10 -0.54
1.48 -0.15
1.57 -0.06
3.80 -0.03
6.22 0.67
3.94 -0.61
2.83 -0.32
2.92 -0.76
4.19 -0.75
2.40 -0.72
1.92 0.43
2.57 0.78
1.98 0.42
1.82 0.44
3.05 0.62
2.16 0.50
3.41 -0.69
2. 98 -0.49
2.21 -0.37
3.58 -0.47
2.53 -0.12
1.51 -0.34
0.94 -0.14
1.42 -0.37
2.68 -0.38
3.23 -0.07
1.38 -0.09
2.72 -0.32
1.73 0.32
2.11 0.71
1.21 -0.19
1.70 0.09
2.31 -0.24
2.99 0.16
2.54 -0.27
1.59 0.82
0.54
0.67
Calc
-Expt
/Avg
0.00
-0.02
-0. 12
-0.09
-0.04
-0.01
0.11
-0. 14
-0.11
-0.23
-0. 16
-0.26
0.25
0.36
0.24
0.28
0.23
0. 26
-0. 18
-0.15
-0. 15
-0.12
-0.05
-0.20
-0. 14
-0.23
-0.12
-0.02
-0.07
-0. 1 1
0.21
0.40
-0. 14
0.05
-0. 10
0.06
-0. 10
0. 23
0. 19
0. 16
aneous Surrogates
0.23
0. 19
2.3
3.4
10.0
18.3
294. 1
286.4
21900. 0.403
13400. 0.076
0.479
0.369
0.076
0. 293
0. 17
1 .31
0.95
0.52
1.11 0.15
0.78 0.26
0. 15
0.40
(cont1nu«d)
-------
Appendix A.
Selected Results
(continued).
of Simulations of All Experiments Modeled Using the RADM Mechanism
Page 20
oo
ui
CO
Experiment
Initial
Concent rat Ions
Ave
Temp
Ave
H20
NOx HC HC/NOx
(ppm) (ppmC)
AU2681B
AU2781B
ST03B1R
ST1081R
ST2081B
JL208IB
ST1682R
JL2081R
JL2281B
OC1481R
ST1682B
ST2981R
ST2981B
OC1481B
ST0381B
ST1081B
JL0882R
JL0882B
Group Average
S. Dev.
Avg. Abs. Value
S. Dev.
31. UNC "Synurban"
AU2284R
AU25B4R
AU2584B
ST0184R
ST0184B
ST0284R
ST02B4B
JN2685R
JN2685B
JN2885R
JN2885B
Group Average
S. Oev.
Avg. Abs. Value
S. Dev.
0.24
0.23
0.24
0.25
0.23
0.42
0.43
0.41
0.26
0.28
0.43
0.24
0.24
0.29
0.23
0.24
0.29
0.28
0.28
0.08
Surrogate
0.32
0.34
0.33
0.31
0.30
0.34
0.33
0.30
0.30
0.38
0.39
0.33
0.03
2.0
2.0
1 .6
2.8
2. 1
1 .8
3.2
2.7
2.9
3.3
3. 1
2.5
2.5
2.9
2.0
1 .0
2. 1
2. 1
2.4
0.6
0.4
0. 1
0. 1
0.2
0.2
0. 1
0. 1
0.2
0.3
0. 1
0.2
0.2
0. 1
8.5
B.8
7.6
11.3
9.2
4.3
7.5
6.6
11.2
1 1 .9
7.2
10.3
10.4
9.9
8.6
4. 1
7.3
7.4
9.0
3.0
1 .2
0.3
0.4
0.6
0.8
0.4
0.3
0.6
0.9
0.4
0.5
0.6
0.3
(degK)
295.7
296.4
301 .0
297.9
292.6
305.0*
304 .4
305.0*
302.3*
291 .3
304.4
293.7
291 .6
288 .9
297.9
295.3
304 .2
304.2
297.7
5.8
302.7
302.0
302.0
302.5
302.5
304.6
304.5
303.6
303.6
299.8
299.8
302.5
1 .6
(ppm)
27500.
25000.
23200.
22300.
15200.
20000. *
23000.
20000.*
20000.
9470.
23000.
18700.
14400.
10300.
16800.
17700.
16000.
16000.
1901B.
4902.
8470.
22600.
22600.
9380.
9360.
19900.
19900.
16800.
16800.
1 1800.
1 1800.
15403.
5427.
Maximum Concentrat
OZONE
Expt
(ppm)
0.544
0.623
0.541
0.610
0.414
0. 165
0.410
0.635
0.722
0.462
0.840
0.294
0.485
0.458
0.61 1
0.626
0.598
0.541
0.503
0. 175
0.657
0.075
0.096
0.546
0.646
0.119
0.020
0.629
0.788
0.23B
0.275
0.372
0.284
Calc
(ppm)
0.468
0.493
0.474
0.596
0.449
0.581
0.584
0.765
0.665
0.542
0.817
0.460
0.479
0.550
0.513
0.463
0.754
0.743
0.561
0. 122
0.802
0. 145
0. 181
0.725
0.743
0.343
0. 192
0.721
0.791
0.48B
0.353
0.499
0.265
Calc
-Expt
(ppm)
-0.077
-0. 130
-0.067
-0.015
0.035
0.417
0. 174
0. 130
-0.057
0.080
-0.023
0. 166
-0.006
0.091
-0.098
-0. 163
0. 156
0.201
0.058
0. 145
0.118
0.099
0. 145
0.070
0.085
0. 179
0.097
0.224
0. 172
0.093
0.003
0.250
0.078
0. 127
0.074
0. 127
0.074
1 on
Calc
-Expt
/Avg
-0. 15
-0.23
-0.13
-0.02
0.08
1.12
0.35
0. 19
-0.08
0.16
-0.03
0.44
-0.01
0.18
-0. 17
-0.30
0.23
0.31
0. 17
0.40
0.27
0.33
0.20
0.64
0.61
0.28
0. 14
0.97
0. 14
0.00
0.69
0.25
0.39
0.31
0.39
0.31
Average
d( (03) -
Expt
Calc
Initial
[NO] )/dt
Calc
-Expt
(ppb/m1n) --
1 .22
1 .27
1 .48
1 .59
1 .06
0.82
1 .28
1 .55
1 .80
1 .53
2.39
0.76
1.35
1 .48
1 .70
1 .36
1 .25
1.12
1 .32
0.40
1.61
0.70
0.69
1 .32
1 . 72
0.73
0.40
1 .34
1 .74
0.85
0.87
1 .09
0.47
1 .07
1 . 13
1 . 27
1 .60
1 .05
0.97
1 .21
1 .53
1 . 76
1 .64
2. 17
0.91
1 .36
1.58
1 .57
1 .06
1 .41
1 .36
1 .32
0.33
1 .70
0.66
0.68
1 .39
1 .66
0.78
0.51
1 .54
1 .93
0.94
0.92
1 . 16
0.50
-0. 15
-0. 14
-0.21
0.00
-0.01
0.16
-0.07
-0.02
-0.04
0.11
-0. 22
0.15
0.01
0. 10
-0.13
-0.30
0.16
0.24
0.00
0.16
0. 13
0.09
0.09
-0.04
-0.01
0. 07
-0.06
0.05
0. 1 1
0. 20
0. 19
0.09
0.06
0.07
0.08
0.09
0.06
Calc
-Expt
/Avg
-0.13
-0. 12
-0. 15
0.00
-0.01
0. 17
-0.06
-0.01
-0.02
0.07
-0. 10
0. 18
0.01
0.07
-0.08
-0.25
0.12
0.19
0.02
0. 15
0. 1 1
0.10
0.06
-0.05
-0.01
0.05
-0.03
0.07
0.24
0. 14
0.10
0. 10
0.06
0.07
0.08
0.08
0.06
32. UNC "Synauto" Surrogate
(cont\nued)
-------
Appendix A. Selected Results of Simulations of All Experiments Modeled Using the RADM Mechanism
(continued).
Page 21
LO
Experiment
Initial
Concent rat Ions
NOx
Ave
Temp
Ave
H20
HC HC/NOx
(ppm) (ppmC)
AU0484R
AU04B4B
AU0584R
AU05B4B
AU0684R
AU06B4B
AU0784R
AU0884B
AU0984R
ST0884R
ST06B4B
ST1784R
ST17B4B
ST2184R
ST21B4B
Group Average
S. Oev.
Avg. Abs. Value
S. Dev.
0.37
0.36
0.36
0.35
0.35
0.35
0.38
0.34
0.39
0.34
0.33
0.34
0.34
0.36
0.36
0.35
0.02
0.4
0.3
0.3
0.4
0.7
1 . 1
0.4
1 . 1
0.4
0.6
1 .0
1 . 1
0.8
0.6
0.7
0.7
0.3
1.0
O.B
0.7
1.0
2. 1
3.2
1.0
3.1
0.9
1.9
2.9
3.3
2.3
1 .7
2.0
1.9
0.9
(degK)
307.5
307.5
307. 1
307.2
307.6
307.6
308.0
307.3
307.8
29B.5
298 . 5
294.4
294.4
302.2
302.2
303.9
5. 1
(ppm)
34900.
34700.
44900.
44900.
36800.
36BOO.
34700.
29000.
31500.
6920.
6920.
24BOO.
24800.
7350.
7350.
27089.
13697.
Maximum Concentrat
OZONE
Expt
(ppm)
0.515
0.328
0.335
0.595
0.887
0.940
0.602
0.834
0.521
0.566
0.750
0.539
0.484
0.671
0.721
0.619
0. 182
Calc
(ppm)
0.701
0.622
0.535
0.700
0.891
0.935
0.727
0.846
0.648
0.769
0.877
0.687
0.685
0.833
0.847
0.754
0. 1 14
Calc
-Expt
(ppm)
0. 186
0.294
0.200
0. 105
0.004
-0.005
0. 125
0.012
0. 127
0.203
0.127
0. 146
0.201
0. 162
0. 126
0. 134
0.082
0. 135
0.081
1on
Calc
-Expt
/Avg
0.31
0.62
0.46
0. 16
0.00
0.00
0. 19
0.01
0.22
0.30
0. 16
0.24
0.34
0.22
0. 16
0.23
0. 17
0.23
0. 17
Average
d( 103] -
Expt
Calc
Initial
INO] )/dt
Calc
-Expt
(ppb/m1n) --
1.45
1 . 16
1 .08
1.48
2.53
3. 16
1 .61
2.92
1 .51
1 .27
1 .90
1 .59
1 .38
1.78
2.28
1 .81
0.63
1.46
1.17
1 .03
1 .44
2.17
2.58
1 .55
2.44
1 .33
1 .44
1 .73
2.35
2.03
2.05
2.64
1 .83
0.53
0.01
0.01
-0.06
-0.04
-0.36
-0.58
-0.06
-0.48
-0. IB
0.17
-0. 18
0.76
0.65
0.27
0.35
0.02
0.38
0.28
0.25
Calc
-Expt
/Avg
0.01
0.01
-0.05
-0.03
-0. 15
-0. 20
-0.04
-0. 18
-0. 13
0.12
-0. 10
0.39
0.38
0. 14
0. 14
0.02
0. 18
0. 14
0. 12
33. UNC Auto Exhaust
OC0483B
OC0783R
OC0783B
JN2582R
JN2582B
JN2982R
JN2982B
JN3082R
JN3082B
JL0283B
JL0883B
ST2982B
OC0682R
AU1 183R
AU1 183B
JL0182R
JL01B2B
AU03B2R
AU0382B
ST1782R
ST1782B
ST29B2R
OC0682B
0.25
0.33
0.34
0.65
0.65
0.24
0.25
0.32
0.32
0. 19
0.37
0.39
0.46
0.22
0.23
0.37
0.35
0.44
0. 16
0.26
0.25
0.39
0.46
0.4
2.7
2.7
0.6
0.6
2.5
2.5
2.8
0.6
1 .7
1 .7
1 .7
2.0
2.2
0.7
3.5
3.6
2.5
1 .2
2.2
2.4
1 .7
2.0
1 .7
8.1
7.9
0.9
0.9
10.3
9.8
8.7
1 .7
9. 1
4.6
4.3
4.3
9.9
3.1
9.6
10.2
5.7
7.3
8.6
9.5
4.4
4.3
299.7
295.5
295.4
301.1*
301 . 2*
303. 1
303. 1
302.5*
302.3*
304.0
299.5
295.7
300.2
306.8
306.9
299.5
299.3
302.6
302.6
300. 1
SOU. 1
295.7
300.2
18800.
21600.
21500.
22700.
22900.
30500.
30500.
20000.*
20000.*
24500.
21200.
24800.
16900.
21400.
21400.
27700.
27400.
27400,
27400.
27400.
27400.
24800.
16900.
0.642
0. 178
0.451
0.003
0.003
0.704
0.766
0.81 1
0.840
0.697
0.879
0.205
0.355
0.850
0.601
0.740
0.759
0.344
0.576
0.562
0.637
0.073
0.450
0.706
0.469
0.637
0.016
0.018
0.814
0.817
0.965
0.973
0.850
0.898
0.079
0.348
0.887
0.696
0.998
0.946
0.527
0.513
0.598
0.621
0.063
0.406
0.064
0.291
0.186
0.014
0.014
0. 109
0.052
0. 154
0. 133
0. 153
0.019
-0. 126
-0.007
0.036
0.096
0.258
0.187
0. 183
-0.063
0.036
-0.017
-0.010
-0,043
0. 10
0.90
0.34
0.14
0.07
0.17
0. 15
0.20
0.02
-0.89
-0.02
0.04
0. 15
0.30
0.22
0.42
-0.11
0.06
-0.03
-0.14
-0.10
1 .86
0.94
1.41
0.64
0.68
2.32
2.67
2.55
2.68
1 .65
1 .76
1.12
1 .37
2.55
1 . 26
2.31
2.41
1 .26
1 .06
1 .63
1.72
0.86
1 .49
2.02
1 . 18
2.00
0.59
0.59
2. 14
2.21
2.46
2.46
1 .75
1 .52
0.82
1.23
2. 19
1 . 29
2.47
2.59
1 .27
0.95
1 . 19
1.30
0.70
1 .34
0.17
0.24
0.59
-0.05
-0.09
-0. 19
-0.46
-0.09
-0.22
0.09
-0.24
-0.30
-0.14
-0.35
0.03
0.16
0.18
0.01
-0.12
-0.43
-0.41
-0. 16
-0. 15
0.09
0. 23
0.34
-0.08
-0. 14
-0.08
-0. 19
-0.04
-0.08
0.06
-0. 15
-0.31
-0. 11
-0. 15
0.03
0.07
0.07
0.01
-0. 12
-0.31
-0.27
-0.20
-0. 11
(cont tnued)
-------
Appendix A. Selected Results of Simulations of AM Experiments Modeled Using the RADM Mechanism
(concluded).
Page 22
uo
ui
VJ1
Experiment
Initial
Concentrat
NOx
(ppm)
JL0683R
JL1583R
JL1583B
OC0483R
Group Average
S. Dev.
Avg. Abs. Value
S. Dev.
34. SAPRC Synthetic
ITC781
ITC784
ITC785
ITC805
ITC795
ITC796
ITC799
ITC801
Group Average
S. Dev.
Avg. Abs. Value
S. Dev.
35. SAPRC Synthetic
ITC963
ITC965
ITC967
ITC968
Group Average
S. Dev.
Avg. Abs. Value
S. Dev.
0.37
0.35
0.35
0.25
0.34
0. 12
Jet
0.51
0.50
0.26
0.52
0.50
0.54
0.51
0.55
0.49
0.09
Jet
0.49
0.46
0.26
0.49
0.42
0.11
HC
(ppmC)
1 .7
2.2
2.3
2.6
2.0
0.8
Fuel
43.0
88.0
45.0
98.0
45.0
97.0
94.0
41 .0
68. 9
27.3
Exhaust
4.4
5.2
4.4
8.7
5.7
2. 1
Ions
HC/NOx
Ave
Temp
(degK)
4.6
6.2
6.6
10.3
6.5
3. 1
83.5
177.6
170.7
189.5
89. B
178.9
184. 1
75. 1
143.7
50.8
9.1
11.3
17.2
17.8
13.8
4.3
299
306
306
299
301
3
303
303
303
303
303
303
303
303
303
0
303
303
303
303
303
0
.5
. 2
.2
. 7
.2
.3
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
Ave
H20
(ppm)
21200.
25700.
25700.
18800.
23607.
3802.
16600.
16600.
17BOO.
17200.
17900.
17200.
17200.
17000.
17200.
499.
19400.
20000.
18300.
19400.
19275.
709.
Maximum Concentration
OZONE
Expt
(ppm)
0.756
0.866
0.922
0.603
0.567
0. 272
0.751
0.746
0.598
0.791
0.761
0.597
0.840
0.881
0.746
0. 102
O.B22
0.863
0.586
0.852
0.781
0. 131
Calc
(ppm)
0.778
0.800
0.896
0.668
0.632
0.299
O.B47
0.905
0.625
0.934
0.804
0.876
0.953
0.885
0.854
0. 104
0.705
0.714
0.580
0.777
0.694
0.082
Calc
-Expt
(ppm)
0.022
-0.066
-0.026
0.066
0.065
0. 101
0.090
0.078
0.096
0.160
0.027
0. 144
o.oin
0.279
0.113
0.004
0. 108
O.OU8
0.108
0.008
-0.117
-0. 149
-0.006
-0.075
-0.087
0.062
0.087
0.062
Calc
-Expt
/Avg
0.03
-0.08
-0.03
0.10
0.08
0.29
0. 19
0.23
0.12
0.19
0.04
0. 17
0.06
0.3B
0. 13
0.00
0.14
0. 12
0.14
0. 12
-0. 15
-0. 19
-0.01
-0.09
-0. 1 1
0.08
0.11
0.08
Average
d( [03] -
Expt
Calc
Initial
[NO] )/dt
Calc
-Expt
(ppb/m1n) --
1 .52
2. 10
2.20
1 .76
1 .68
0.61
2.77
3.93
2.74
3.27
3.05
4.43
4.59
2.98
3.47
0.74
4.00
5. 10
5.79
1 1 .76
6.66
3.48
1 .43
1 .87
2.08
1.81
1 .60
0.60
3.33
4. 15
3.05
4.02
4.30
5.31
3.75
2.94
3.86
0.78
3.72
4.73
5.59
10.28
6.08
2.90
-0.09
-0. 23
-0. 12
0.05
-0.08
0.23
0. 19
0. 14
0.56
0.22
0.31
0.75
1 .25
0.88
-0.84
-0.05
0.38
0.64
0.61
0.40
-0.29
-0.37
-0.20
-1 .48
-0.58
0.60
0 .58
0.60
Calc
-Expt
/Avg
-0.06
-0. 1 1
-0.06
0.03
-0.06
0. 15
0.13
0.09
0. 18
0.05
0. 11
0.21
0.34
0. 18
-0.20
-0.02
0. 1 1
0.16
0.16
0. 10
-0.07
-0.08
-0.04
-0. 13
-0.08
0.04
0.08
0.04
-------
APPENDIX B
SELECTED RESULTS OF SIMULATIONS OF ALL ORGANIC-NOX EXPERIMENTS
MODELED USING THE RADM-M MECHANISM
356
-------
Appendix B. Selected Results of Simulations of Individual Organlc-NOx-AIr Experiments Using the
RADM-M Mechanism.
Page 1
uo
VJ1
Experiment
Initial
Concent rat Ions
NOx HC HC/NOx
(ppm) (ppmC)
1. Ethene-NOx
EC142
EC143
EC 156
EC265
EC2B6
EC2B7
ITC926
ITC936
AU0479R
AU0579R
OC0584R
OC1 184R
OC1284R
OC05B40
Group Average
S. Dev.
Avg. Abs. Value
S. Dev.
2. Propene-NOx
EC121
EC177
EC216
EC217
EC230
EC256
EC257
EC276
EC277
EC278
EC279
EC314
EC315
EC316
EC3I7
ITC693
ITC810
ITC860
ITC925
ITC938
ITC947
ITC960
0.48
0.50
0.50
1 .01
0.94
0.53
0.51
0.50
0.23
0.64
0.36
0.35
0.72
0.37
0.55
0.22
0.51
0.46
0.52
0.48
0.52
0.56
0.56
0.52
0.11
0.49
0.97
0.93
0.94
0.98
0.54
0.49
0.52
0.52
0.54
0.52
0.53
0.50
1
4
4
3
7
B
7
3
0
4
3
2
2
1
4
2
1
1
1
0
1
0
0
1
1
3
3
3
2
3
1
3
2
3
2
2
1
2
.9
. 1
.0
.9
.5
.0
.9
.9
.9
. 1
.2
.9
.7
.8
. 1
.3
.5
.5
.5
.6
.9
.4
.7
.6
.7
. 1
.5
.2
.9
.2
.5
.5
.8
.0
.8
.8
.9
.8
4. 1
B. 1
8.0
3.9
8.0
15. 1
15.6
7.8
3.9
6.4
8.8
8.2
3.7
5.0
7.6
3.8
2.9
3.2
3.0
1 . 2
3.7
0.7
1 .3
3.2
15.7
6.2
3.5
3.5
3. 1
3.3
2.8
7.2
5.4
5.8
5.2
5.3
3.6
5.5
Ave
Temp
(degK)
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
298.9
292.4
297.3
297.2
295.6
297.3
300. 2
3.6
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
288.7
312.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303. 0
Ave
H20
(ppm)
26500.
22700.
29000.
20400.
20500.
19700.
21 100.
20300.
32500.
29100.
20000.*
12700.
13600.
20000.*
22007 .
561 1 .
26500.
26900.
18100.
21400.
17200.
20200.
20000.
14500.
15700.
16100.
14200.
24000.
10600.
44900.
24200.
20300.
16800.
17100.
16300.
19000.
20300.
20300.
Maximum Concentration
OZONE
Expt
(ppm)
0.782
1 .087
1 . 105
0.840
1 .081
0.965
0.982
0.940
0.729
1 . 294
0.856
0.858
0.495
0.675
0.906
0. 204
0.506
0.540
0.564
0. 149
0.344
0.002
0.068
0.3B8
0.313
0.625
0.679
0.728
0.344
0.955
0.615
0.779
0.782
0.585
0.779
0.729
0.710
0. 721
Calc
(ppm)
0.475
0.791
0.751
0.892
1 . 172
1 .026
0.920
0.83B
0.554
1 . 142
0.970
1 .039
0. 185
0.646
0.814
0.275
0.460
0.471
0.555
0.230
0.329
0.01 1
0.098
0.447
0.395
0.702
0.706
0.786
0.661
1 .051
0.570
0.718
0.721
0.680
0.728
0.719
0.715
0.712
Calc
-Expt
(ppm)
-0.307
-0.296
-0.354
0.052
0.091
0.061
-0.062
-0. 102
-0. 175
-0. 152
0.114
0. 182
-0.310
-0.029
-0.092
0. 179
0. 163
0.111
-0.046
-0.069
-0.009
0.081
-0.015
O.OOB
0.029
0.059
0.082
0.076
0.027
0.05B
0.316
0.096
-0.046
-0.062
-0.061
0.096
-0.051
-0.010
O.OOb
-0.009
Calc
-Expt
/Avg
-0.49
-0.32
-0.3B
0.06
0.08
0.06
-0.06
-0. 1 1
-0.27
-0. 12
0.12
0. 19
-0.91
-0.04
-0.16
0.30
0.23
0. 24
-0.09
-0.14
-0.02
0.43
-0.05
0.35
0.14
0.23
0.12
0.04
0.08
0.63
0.10
-0.08
-0.08
-0.08
0. 15
-0.07
-0.01
0.01
-0.01
Average
d( 103] -
Initial
[NO] )/dt
Calc
Expt Calc -Expt
(ppb/m1n) --
3.20
8.50
8.89
5.05
1 1 . 76
13.89
6.96
2.72
1 .60
3.17
2. 16
2. 22
1.58
1.48
5.23
4.10
7.46
3. 89
4. 18
0.80
3.06
0.98
3.35
3.24
8. 27
7.72
6.64
7.21
4.33
10.64
4.06
5.07
4.25
3.57
3.72
3.61
3.34
4.23
2.20
4.51
4.50
4.89
10. 13
13.00
4.91
2.34
0.99
2.27
1 .98
2.01
1.17
1 .33
4.02
3.53
4. 10
4.00
5.69
1 .30
4.78
1 .08
4.20
4.41
8.06
9.83
8.08
10.36
6.42
1 1 .93
4.71
5.72
4.57
4. 88
4. 14
4. 24
4.14
4.40
-1 .00
-3.99
-4.39
-0. 15
-1 .63
-0.88
-2.05
-0.3B
-0.61
-0.91
-0. 18
-0.21
-0.41
-0. 15
-1.21
1 .39
1.21
1 . 39
-3.36
0.11
1.51
0.50
1 .72
0. 10
0.84
1 . 16
-0. 21
2.11
1 .43
3. 15
2.09
1 .29
0.65
0.65
0.31
1.31
0.42
0.63
0.80
0. 17
Calc
-Expt
/Avg
-0.37
-0.61
-0.66
-0.03
-0. 15
-0.07
-0.35
-0. 15
-0.47
-0.33
-0.09
-0.10
-0.30
-0.11
-0.27
0.20
0.27
0.20
-0.58
0.03
0.31
0.48
0.44
0.10
0.22
0.30
-0.03
0.24
0. 19
0.36
0.39
0.11
0.15
0. 12
0.07
0.31
0.11
0.16
0.21
0.04
(cont1nued)
-------
Appendix B.
Selected Results of Simulations of
RAOM-M Mechanism (continued).
Individual Organ1c-N0x-A1r Experiments Using the
Page 2
ui
CD
Experiment
Initial
Concent rat
NOx
Ions
Ave
Temp
Ave
H20
HC HC/NOx
(ppm) (ppmC)
OTC186
OTC191
OTC210
OTC233
OTC236
JA107BR
OC1278B
OC2078R
OC2078B
OC2178R
OC2578B
JN1279R
JN1279B
JN1379R
AU0279R
AU2780B
ST0482B
ST1382B
JL1783R
JL2183R
JL29B3B
JL3183R
ST2383B
OC0484R
OC0484B
OC1184B
OC1284B
Group Average
S. D»v.
Avg. Abs . Value
S. Oev.
3. 1-Butene-NOx
EC122
EC 123
EC124
ITC927
ITC928
ITC930
ITC935
ST2383R
ST2583R
ST2583B
ST2783R
0.55
0.54
0.57
0.46
0.53
0.46
0.48
0.46
0.46
0.50
0.44
0.50
0.49
0.45
0.22
0.46
0.23
0.33
0.27
0.22
0.21
0.21
0.38
0.36
0.36
0.36
0.68
0.49
0. 18
0.50
0.51
0.99
0.31
0.67
0.32
0.66
0.40
0.46
0.42
0.45
3.6
3.7
2.7
0. 1
3.3
3. 1
1 .4
1 .3
3.5
3.9
1 .3
1 .0
1 .5
2.9
.5
.9
,
t
f
t
,
.6
2. 1
1 .0
2.2
2.0
2.1
1 .0
0.9
1 .6
1 .7
3.8
3.8
7.2
7.6
1 .5
1 .6
2.9
1 .6
6.6
6.9
4.8
0.2
6.3
6.9
2.9
2.9
7.7
7.9
2.9
2. 1
3.0
6.5
7.0
4.0
4.9
3.3
3.9
5.0
5.3
5. 1
4.3
5.9
2.9
6.3
2.9
4.6
2.5
1 .7
3.2
1 .7
12.3
5.7
22.2
1 1 .6
3.7
3.6
t'.e
3.6
(degK)
299.8
309.3
304.7
303.6
300.0
265. 1
294.3
290.4
290.4
290.3
286.4
293.5
293.5
294.9
307.2
302.2
300.5
302.6
307.2
306.5
303. 1
305.0
292.5
296.8
296.7
297.2
295.9
299.9
7.5
303.0
303.0
303.0
303.0
303.0
303.0
303.0
292.6
292.3
292.3
295.5
(ppm)
5000.
5000.
3720.
5000.
6510.
20000.
16300.
9590.
9560.
20000.*
12800.
20000.*
20000.*
^0000.*
20000.*
28400.
1 1800.
24100.
29700.
23800.
23300.
19800.
23400.
20000. »
20000.*
12700.
13500.
18134.
7349.
22000.
Z7200.
24500.
19000.
18400.
21 10U.
19700.
23400.
18600.
16900.
19400.
Maximum Concentration
OZONE
Expt
(ppm)
0.822
0.903
0.972
0.633
0.848
0.363
0.461
0.340
0.727
0.670
0.230
0.382
0.673
0.974
0.788
1 .044
0.658
0. 731
0.848
0.804
0.697
0.719
0.405
0.645
0.446
0.674
0.432
0.608
0. 238
0.227
0.506
0. 247
0.646
0.022
0.717
0.872
0.206
0.266
0.594
0.285
Calc
(ppm)
0.939
1 . 155
1 .006
0.934
1 .015
0.604
0.413
0.403
0.948
0.913
0.269
0.306
0.515
0.874
0.632
1 . 183
0.614
0. 607
0.720
0.731
0.739
0.6B7
0.666
0.759
0.474
0.772
0.471
0.651
0.253
0.081
0.280
0.132
0.727
0.045
O.B27
0.972
0.451
0.449
0.783
0.452
Calc
-Expt
(ppm)
0.117
0.252
0.034
0. 301
0. 167
0.240
-0.048
0.063
0.221
0.243
0.039
-0.076
-0. 158
-0.099
-0. 156
0. 139
-0.044
-0.045
-0. 127
-0.072
0.042
-0.032
0.261
0. 1 14
0.029
0.098
0.039
0.043
0.117
0.093
0.081
-0. 147
-0.226
-0.115
0.08 1
0.023
0.110
0. 100
0.244
0. 182
0. 190
0. 167
Calc
-Expt
/Avg
0. 13
0.24
0.03
0.38
0. 18
0.50
-0.11
0. 17
0.26
0.31
0. 16
-0.22
-0.27
-0.11
-0.22
0. 12
-0.07
-0.06
-0. 16
-0.09
0.06
-0.05
0.49
0. 16
0.06
0. 14
0.09
0.08
0.20
0. 16
0. 14
-0.95
-0.58
-0.61
0.12
0.14
0. 1 1
0. 74
0.51
0.28
0.45
Average
d( [03] -
Expt
Calc
Initial
{NO] )/dt
Calc
-Expt
(ppb/mln) --
5.16
12. 18
6.70
3.69
6.91
0.46
1 .40
1 .23
2.83
2.53
1 .03
0.85
1 .24
2.69
2.41
3.26
1 .46
1 .69
2. 10
1 .76
1 .66
1 .81
1 .23
2.07
1 .25
2.38
1 .67
3.62
2.60
2.29
4. 16
1 .99
3. 24
1.36
7.91
5.39
0.96
1 . 16
1.7B
1 . 19
7.81
13.94
6.36
5.50
8.41
0.48
1 .45
1 .38
3.53
3.44
1 .15
0.77
1 . 10
2.83
2.06
2.84
1 .59
1 .61
1 .75
1 .87
1 . 74
1 .67
1 .67
2. 15
1 .21
2.45
1 .64
4. 15
3.08
1 .40
2.30
1 . 71
4. 54
1 .69
11.13
7. 15
1 . 15
1.31
2. 19
1.31
2.65
1 .77
-0.34
1 .81
1.50
0.02
0.05
0. 15
0.70
0.91
0. 12
-0.09
-0. 14
0.14
-0.35
-0.42
0. 13
-O.OB
-0.35
0. 10
0.08
-0.14
0.44
0.08
-0.04
0.06
-0.03
0.53
1 .01
0.76
0.85
-0.89
-1 .86
-0.27
1.31
0.34
3.22
1 .76
0.20
0. 15
0.41
0.12
Calc
-Expt
/Avg
0.41
0. 14
-0.05
0.39
0.20
0.04
0.03
0.12
0.22
0.31
0. II
-0.11
-0.12
0.05
-0. 16
-0. 14
0.08
-0.05
-0. 18
0.06
0.05
-0.08
0.31
0.04
-0.04
0.03
-0.02
0. 11
0. 19
0. 17
0.14
-0.48
-0.58
-0. 15
0. 34
0. 22
0. 34
0. 26
0. 19
0.12
0.21
0. 10
(contInued)
-------
Appendix B. Selected Results of Simulations of Individual
RADM-M Mechanism (continued).
Orgun1c-N0x-A1r Experiments Using the
Page 3
to
VJI
Experiment
Initial
Concent rat 1 ons
NOx HC HC/NOx
(ppm) (ppmC)
Group Average
S. Dev.
Avg. Abs . Value
S. Dev.
0.52
0.20
3. 1
2.3
6.9
6.2
Ave
Temp
(degK)
299.4
5.0
Ave
H20
(ppm)
21 127.
2856.
Maximum Concentration
OZONE
Expt
(ppm)
0.417
0.263
Calc
(ppm)
0.472
0.321
Calc
-Expt
(ppm)
0.055
0. 154
0. 144
0.066
Calc
-Expt
/Avg
0.02
0.55
0.45
0.29
Average
d( [03] -
Initial
[NO] )/dt
Calc
Expt Calc -Expt
(ppb/mln)
2.86
2. 18
3.26
3.18
0.41
1.34
0.96
0.99
Calc
-Expt
/Avg
0.05
0.32
0.27
0.15
4. t rans-2-Butene-NOx
EC146
EC147
EC157
ST2783B
Group Average
S. Dev.
Avg. Abs. Value
S. Dev.
5. Isobutene-NOx
ITC694
6. 1-Hexene-NOx
ITC929
ITC931
ITC934
ITC937
Group Average
S. Dev.
Avg. Abs. Value
S. Dev.
7. Isoprene-NOx
EC520
EC522
EC524
EC525
0.51
0.9B
0.53
0.43
0.61
0.25
0.51
0.51
0.49
1 .00
0.99
0.75
0.28
0.49
0.92
0.96
0.54
0.9
1 . 7
0.9
2.0
1 .4
0.6
4.6
5. 1
10.3
9.7
0. 1
6.3
4.8
2.2
2.3
4.7
4.5
1 .8
1 .7
1 .7
4.7
2.5
1.5
9. 1
10.0
21 . 1
9.7
0. 1
10.2
8.6
4.5
2.4
4.9
8.3
303.0
303.0
303.0
295.5
301 . 1
3.7
303.0
303.0
303.0
303.0
303.0
303.0
0.0
303.0
303.0
303.0
303,0
26900.
22300.
27500.
19400.
24025.
3860.
20000.
19000.
20100.
20300.
20300.
19925.
624.
13000.
13000.
13000.
13000.
0.247
0. 154
0.205
0.523
0.282
0. 165
0.900
0.296
0.606
0.428
0.007
0.335
0.252
0.503
0.276
0.759
0.691
0. 109
0.083
0.091
0.543
0. 207
0.224
0.731
0.670
0.819
0.894
0.060
0.61 1
0.379
0.300
0. 1 IB
0.430
0.630
-0. 138
-0.071
-0. 1 14
0.019
-0.076
0.069
O.OB5
0.052
-0. 169
0.372
0.213
0.466
0.053
0.276
0. 182
0.276
0. 182
-0.203
-0. 158
-0.330
-0.061
-0.77
-0.60
-0.77
0.04
-0.53
0.38
0.54
0.35
-0.21
0.77
0.30
0.71
0.59
0.26
0.59
0. 26
-0.50
-0.80
-0.55
-0.09
5.87
9.83
5.96
3.00
6.17
2.80
8 .84
1 .27
2.82
1 .84
0.33
1 .57
1 .04
5.30
6.00
15.57
10. 13
3.84
7.39
3.54
3. 16
4.48
1.96
6.03
3.32
10.67
5.00
0.57
4.89
4.26
7.57
6.75
15.94
22. 16
-2.02
-2.45
-2.42
0. 15
-1 .68
1 .24
1 .76
1 .09
-2. 81
2.05
7.84
3. 16
0.24
3.32
3.25
3.32
3.25
2.26
0.76
0.37
12.03
-0.42
-0.28
-0.51
0.05
-0.29
0.24
0.32
0.20
-0.38
0.89
1.16
0.92
0.52
0.88
0.26
0.88
0.26
0.35
0.12
0.02
0.75
(cont1nued)
-------
Appendix B. Selected Results of Simulations of Individual Organ1c-N0x-A1r Experiments Using the
RADM-M Mechanism (continued).
Page 4
U)
CT>
O
Experiment
EC527
ITC811
1TC812
JL16BOR
JL1680B
JL1780R
JL1780B
JL23B1R
ST0981R
Croup Average
S. Oev.
Avg. Abs. Value
S. Dev.
8. a-P1nene-NOx
JL1580R
JL1580B
JL2580R
JL2580B
Group Average
S. Dev.
Avg. Abs. Value
S. Dev.
9. Ethane-NOx
ITC999
10. n-Butane-NOx
EC130
EC133
EC 134
EC137
EC162
EC163
EC 168
EC178
EC304
Initial
Concentrat Ions
NOx
(ppm)
0.50
O.SO
0.52
0. IB
0. IB
0.46
0.47
0.43
0. 17
0.49
0.25
0.1B
0. 19
0.25
0.25
0.22
0.04
0.09
0. 10
0.50
0.51
0.50
0.51
0.49
0.49
0. 10
0.47
HC
( ppmC )
2.2
3.4
1 .8
4.6
6.4
1 .0
2.6
1 .4
1 .0
2.9
1 .7
1 . 1
2.6
1.0
0.3
1 .3
1 .0
45.4
17.6
6.6
8.3
8.7
8.2
9.0
B.O
7.8
17.1
HC/NOx
4.5
6.8
3.4
26.0
36.6
2. 1
5.5
3.4
6.1
8.6
10.4
5.8
13.8
4.0
1 .4
6.3
5.3
534. 1
179.3
17.1
16.3
17.3
16.3
18.3
16. 2
79.6
36.7
Ava
Temp
(degK)
303.0
303.0
303.0
305. 0
301 .9
305.6
302.8
306. 1
297.5
303. 1
2. 1
302.4
299.6
302.4
300. 1
301 . 1
1 .5
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
Ave
H20
(ppm)
13000.
17300.
17300.
30200.
26BOO.
27000.
25000.
20000.
20000.*
19123.
6284.
30300.
29200.
31700.
26300.
29375.
2291 .
20500.
26200.
27800.
20000."
20000.*
27300.
28500.
20700.
20900.
26800.
Maximum Concentration
OZONE
Expt
(ppm)
0.547
0.919
0.768
0.652
0.837
0.806
1 .298
0.750
0.506
0.716
0.246
0.201
0.470
0.377
0.334
0.346
0.112
0.243
0.459
0.249
0.034
0.042
0.112
0.454
0.655
0.384
0.362
Calc
(ppm)
0.317
0.577
0.253
0.712
0.600
0.420
0.835
0.691
0.457
0.488
0.208
0.297
0.421
0.327
0.211
0.314
0.067
0.203
0.442
0.043
0.036
0.043
0.048
0. 151
0.308
0.418
0.422
Calc
-Expt
(ppm)
-0.230
-0.342
-0.515
0.060
-0.237
-0.3B6
-0.463
-0.058
-0.049
-0.229
0. 174
0.238
0.160
0.095
-0.049
-0.050
-0. 123
-0.032
0.092
0.079
0.036
-0.041
-0.017
-0.206
0.002
0.001
-0.064
-0.304
-0. 347
0.034
0.060
Calc
-Expt
/Avg
-0.53
-0.46
-1 .01
0.09
-0.33
-0.63
-0.43
-0.08
-0. 10
-0.42
0.31
0.43
0.29
0.3B
-0. 11
-0. 14
-0.45
-O.OB
0.34
0.27
0. 17
-0. 18
-0.04
-1.41
-0.80
-1 .00
-0.72
0.09
0. 15
Average
dC [03J -
Expt
Calc
In1t 1al
[NOJ )/dt
Calc
-Expt
(ppb/m1n) --
5.97
10.00
4.86
3. 18
2.88
1 .96
4.20
2. 13
1 .46
5.66
4.06
0.61
1.59
0.84
0.84
0.97
0.43
1 .32
4.41
2.42
0.94
1.02
1 .73
3.31
2.03
1 .61
2.09
7.95
8.95
2.42
3.00
2.43
1 .36
2.87
2.01
1 .45
6.53
6.29
0.67
1 .96
0.83
0.6B
1 .03
0.62
1 .42
2.44
1 .05
1 .27
1 . 24
1 . 19
1 .69
1.12
1 .65
2.26
1 .97
- 1 .05
-2.43
-0. IB
-0.44
-0.60
-1 .33
-0.12
-0.01
0.86
3.58
1.81
3. 18
0.06
0.37
-0.02
-0. 16
0.06
0. 22
0. 15
0. 16
0.09
-1.97
-1 .37
0.33
0.23
-0.54
-1.62
-0.91
0.04
0. 17
Calc
-Expt
/Avg
0.28
-0.11
-0.67
-0.06
-0.17
-0.36
-0.38
-0.06
0.00
-0.02
0.36
0.26
0.24
0. 10
0.21
-0.02
-0.21
0.02
0. IB
0. 13
0.09
0.07
-0.58
-0.79
0.30
0.20
-0.37
-0.65
-0.58
0.03
0.08
(cont1nued)
-------
Appendix B. Selected Results of Simulations of Individual Organ1c-N0x-A1r Experiments Using the
RADM-M Mechanism (continued).
Page 5
ON
Experiment
Initial
Concentrations
Ave
Temp
Ave
H20
Maximum Concentrat
OZONE
Calc
NOx HC HC/NOx
(ppm) (ppmC)
EC305
EC306
EC307
EC30B
EC309
ITC507
ITC533
ITC770
ITC939
ITC948
OTC21 1
JL2178R
JL2178B
JL2278R
JL2278B
ST1879B
OC0979R
OC1879B
Group Average
S. Dev.
Avg. Abs. Value
S. Dev.
1 1 . C4+ Branched
EC 165
EC 169
EC171
OC1879R
OC2079B
AU1983R
AU1983B
Group Average
S. Dev.
Avg. Abs. Value
S. Dev.
0. 10
0. 19
0. 10
0.48
0.47
0.09
0.12
0.52
0.51
0.26
0.55
0. 24
0.24
0.55
0.55
0.21
0.21
0.20
0.34
0. IB
15.7
25.8
25. B
16.2
17.2
15.2
11.9
37.9
14. B
10.0
42.8
7. 2
15.4
7.9
17.5
21 -2
14.6
14.3
15.7
8.8
159.7
138.2
252.9
33.6
36.3
165.0
99.6
72.8
28.9
38.2
77.5
29.9
63.9
14.4
31 .5
103. 1
71.0
71 .8
69. B
61 .6
(degK)
303.0
302.6
303.0
288. 7
310.9
303.0
303.0
303.0
303.0
303.0
300. 1
302.9
302.9
305. 2
305.2
298.3
297.2
295.0
302. 1
3.8
(ppm)
25400.
24200.
29700.
8750.
1 2400.
20300.
20BOO.
16800.
20300.
20300.
3530.*
20000.*
20000.*
20000. *
20000.*
20000. *
15300.
28100.
20892.
6033.
Expt
(ppm)
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
398
535
420
047
545
149
165
042
017
054
008
763
986
166
788
185
191
208
312
268
Calc
(ppm)
0.528
0.705
0.585
0.204
0.601
0.350
0.321
0.048
0.025
0. 137
0.181
0.586
0.861
0. 132
0.406
0.111
0.266
0. 153
0.300
0.233
-Expt
(ppm)
0
0
0
0
0
0
0
0
0
0
0
-0
-0
-0
-0
-0
0
-0
-0
0
0
. 130
. 170
. 165
. 157
.056
.201
. 156
.006
.008
.083
. 173
. 177
. 125
.034
.382
.075
.075
.055
.011
. 161
121
. 104
1 on
Calc
-Expt
/Avg
0.28
0.27
0.33
0. 10
0.81
0.64
0.87
-0.26
-0.14
-0.23
-0.64
-0.50
0.33
-0.30
-0.10
0.59
0.47
0.36
Average
d( [03] -
In1 t lal
[NO] )/dt
Calc
Expt Calc -Expt
(ppb/mln) --
2.39
2.38
2.61
1 .04
2.00
0.69
0.61
1 .56
0.36
0.53
0.56
1.17
1 .64
0.90
1 .55
0.51
0.60
0.60
1 .53
0.98
3.38
4.12
5.81
2.54
2.42
1 .52
1 .33
1 .94
0.63
1 .08
1 .37
1 .05
1 .45
0. 75
1 .27
0.43
0.59
0.51
1.71
1.19
0.99
1 . 74
3.20
1.51
0.42
0.83
0.72
0.38
0.27
0.55
0.81
-0.12
-0. 19
-0. 15
-0.28
-0.08
-0.01
-0.09
0. 18
1 .04
0.72
0.76
Calc
-Expt
/Avg
0.34
0.54
0.76
0.84
0.19
0.75
0.75
0.22
0.55
0.68
0.83
-0.11
-0. 12
-0.18
-0.20
-0.17
-0.01
-0.16
0.12
0.49
OA 1
. *1 I
0.28
Alkane-NOx
0. 10
0. 19
0. 10
0.20
0.22
0.38
0.37
0.22
0.11
11.3
4.5
3.5
16.4
12.6
4.7
4. 1
8.2
5.2
114.3
23.5
35.7
81 .9
56.5
12.6
10.9
47.9
38.7
303.0
303.0
303.0
295.0
295.3
307. 1
307. 1
301 .9
5.0
18600.
32900.
19000.
28100.
27BOO.
21300.
21300.
24143.
5463.
0.
0.
0.
0.
0.
0.
0.
0.
0.
488
493
403
236
217
088
057
283
181
0.4B1
0.340
0.31 1
0. 106
0.051
0. 231
0.098
0.231
0. 156
-0
-0
-0
-0
-0
0
0
-0
0
0
.008
.153
.092
. 130
. 165
. 143
.041
.052
. 1 15
1 05
!oeo
-0.02
-0.37
-0.26
-0.76
-1 . 23
0.90
0.53
-0. 17
0.73
0. 58
0.41
1 .77
1 .00
1 .30
0.64
0.66
0.61
0.53
0.93
0.46
1 .46
0. 74
O.B1
0.45
0. 36
0.66
0.56
0.72
0.36
-0.30
-0.26
-0.50
-0.18
-0.30
0.05
0.03
-0.21
0.20
0 . 23
0. 16
-0. 19
-0. 30
-0.47
-0.34
-0.59
0.09
0.06
-0.25
0.25
0 . 29
0.20
12. C5 + n-Alkane-NOx
EC135
OC0979B
EC 131
ITC559
0. 10
0.21
0. 10
0. 19
20.4
15. 1
24.6
279.4
212.7
73.3
251 . 1
1441 . 1
303.0
297.2
303.0
303.0
25000.
15200.
25000.
20300.
0.
0.
0.
0.
435
184
393
377
0.543
0.205
0.524
0.516
0
0
0
0
. 108
.021
.131
. 139
0. 22
0. 11
0.29
0.31
2.92
0.59
1 .92
1 .79
2. 44
0.51
2. 26
1 .58
-0.47
-0.08
0.33
-0.21
-0. 18
-0. 15
0. 16
-0.12
(contInued)
-------
Appendix B.
Selected Result* of Simulation* of Individual Organlc-NOx-AIr Experiment* Us1no the
RADM-M Mechanism (continued).
Page 6
oo
cr>
ro
Experiment
Initial
Concentrat 1on»
ND»
HC HC/NOx
(ppm) (ppmC)
ITC53B
ITC540
ITC552
ITC761
ITC762
ITC763
ITC797
ST1879R
EC 155
ITC1001
Group Average
S Oev.
Avg. Ab«. Value
S. Dev.
0.11
0. 1 1
0.13
0.52
0.27
0.28
0.52
0.21
0. 10
0. 11
0.21
0. 14
60.3 529.0
274.8 2421 .3
428.8 3278.4
75.2 145.9
74.7 280.4
7.7 27.7
7.3 14.0
6.3 30.4
37.3 385.1
2.4 22.4
93.9 650.9
133.5 1015.3
Ave
Temp
(degK)
303.0
303.0
303.0
303.0
303.0
303.0
SOo.O
298.3
303.0
303.0
302.3
1 .9
Ave
H20
(ppm)
18900.
20300.
20300.
17500.
17000.
17100.
12200.
20000.*
26200.
17500.
19464.
3919.
Maximum Concentration
OZONE
Expt
(ppm)
0. 150
0.360
0.315
0.030
0. 105
0.041
0.004
0.122
0.264
0.036
0.201
0. 153
Calc
(ppm)
0.457
0.546
0.532
0.079
0.327
0.088
0.012
0.532
0.558
0.084
0.357
0.215
Calc
-Expt
(ppm)
0.306
0. 186
0.217
0.049
0.221
0.047
0.007
0.410
0.294
0.049
0. 156
0. 122
0. 156
0. 122
Calc
-Expt
/Avg
1 .01
0.41
0.51
1 .02
1 .25
0.72
0.58
0.39
0.58
0.39
Average
d( [03] -
Expt
Calc
Initial
fNO] )/dt
Calc
-Expt
-- (ppb/m1n)
0.74
1.85
1. 19
1.08
0.83
0.68
0.64
0.40
1 .33
0. 15
1 . 15
0.75
1.50
1 .62
1.59
1 .85
2.30
1 .78
1 . 10
0.51
2.42
0.49
1.57
0.69
0.76
-0. 24
0.39
0.77
1 .47
1 . 10
0.46
0. 1 1
1 .09
0. 34
0.42
0.57
0.56
0.42
Calc
-Expt
/Avg
0.68
-0.14
0.28
0.53
0.94
0.89
0.53
0.23
0.58
1 .06
0.38
0.43
0.46
0.33
13. Methy 1 cycl ohexane-NOx
ITC765
ITC766
ITC767
ITC800
Group Average
S Dev.
Avg. Aba. Value
S. Oev.
14. Benzene-NOx
ITC560
ITC561
ITC562
ITC698
ITC710
ITC831
Group Average
S. Dev.
Avg. Abs. Value
S. Dev.
0.53
0.26
0.55
0.54
0.47
0. 14
0.12
0. 1 1
0.56
0.50
0.55
1 .01
0.47
0.33
0.0 0.1
0.0 0.1
0.1 0.1
0.0 0.1
0.0 0.1
0.0 0.0
332.3 2874. 4
79.1 694.2
83.8 149.7
83.5 167.4
83.6 151.0
12.2 12.1
112.4 674.8
111.3 1 103.2
303.0
303.0
303.0
303.0
303.0
0.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
0.0
17900.
17100.
17900.
17500.
17600.
363.
17100.
17600.
18800.
21900.
21000.
18300.
19117.
1920.
0.022
0.121
0.041
0.015
0.050
0.323
0.273
0.412
0.374
0.367
0.021
0.295
0. 142
0.012
0.301
0.064
0.012
0.097
0.335
0.299
0.528
0.497
0.512
0.009
0.363
0. 199
-0.010
0.180
0.022
-0.003
0.047
0.054
0.012
0.026
0.116
0. 123
0. 145
-0.012
0.068
0.067
0.072
0.062
0.85
0.85
0.85
0.04
0.09
0.25
0.28
0.33
0.20
0. 13
0.20
0.13
0.86
0.87
1 .09
0.86
0.92
0.12
7.01
4.75
2.85
2.87
2.70
0. 17
3.39
2.30
0.77
2.03
1.48
0.85
1 .28
0.59
15.99
10.63
10.21
10. 16
9.68
0.31
9.50
5.08
-0.09
1. 16
0.39
0.00
0.36
0.57
0.41
0.53
8.98
5.88
7.36
7.29
6.98
0.14
6.11
3.08
6.11
3.08
-0. 12
0.80
0.30
0.00
0.25
0.41
0.31
0.35
0.78
0.76
1.13
1.12
1.13
0.60
0.92
0.23
0 .92
0.23
(contInued)
-------
Appendix B. Selected Results of Simulations of Individual Organ1c-N0x-A1r Experiments Using the
RADM-M Mechanism (continued).
Page
U)
Exper 1ment
Initial
Concentrat 1 ons
NOx HC HC/NOx
(ppm) (ppmC)
15. Toluene-NOx
EC264
EC265
EC266
EC269
EC270
EC271
EC272
EC273
EC327
EC336
EC337
EC339
EC340
ITC699
ITC828
JL3080R
AU2780R
AU2782B
OC2782R
AU0183R
Group Average
S. Dev.
Avg . Abs. Value
S. Dev.
16. Xy 1 ene-NOx
EC343
EC344
EC345
EC346
ITC702
ITC827
JL3080B
AU2782R
OC2782B
AU01B3B
Group Average
S. Dev.
Avg. Abs. Value
S. Dev.
0.48
0.48
0.49
0.47
0.46
0,21
0.48
0.11
0.45
0.44
0.45
0.44
0.43
0.51
1 .02
0. 18
0.46
0.43
0.39
0.39
0.44
0. 18
0.28
0.67
0.28
0. 26
0.52
1 .07
0. 18
0.43
0. 39
0.37
0.45
0.26
8. 1
7.5
B.4
4.0
4. 2
8.0
4. 1
4. 1
4.0
7. 2
7.9
5.0
4. 1
10.5
3.0
3.9
2.3
3.0
4.5
4.6
5.4
2.3
4.2
4.0
3.7
3.9
4.0
1 .2
2.2
2.0
2.8
2.7
3. 1
1 .0
17.0
15.6
17.0
8.4
9.0
37.4
8.5
37.2
8.9
16.3
17.7
11.3
9.5
20.8
3.0
21 .3
4.8
7.0
11.7
11,8
14.7
9.2
14.9
5.9
13.3
14.8
7.8
1 . 1
12.4
4.6
7.0
7.2
8.9
4.7
Ave
Temp
(degK)
303
303
303
303
303
303
303
303
303
303
303
303
303
303
303
304
302
301
2B9
306
302
3
303
303
303
303
303
303
304
301
289
306
302
4
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.5
. 2
.7
.8
. 1
.5
. 1
.0
.0
.0
.0
.0
.0
.5
. 7
.8
.0
.0
.4
Ave
H20
(ppm)
18900.
21900.
19400.
19100.
20000.
20500.
19400.
21200.
22400.
26900.
27000.
26800.
25900.
20300.
17200.
35200.
32000.
24700.
19500.
23700.
23100.
4692.
29500.
28900.
25200.
25900.
20300.
14400.
28000.
24700.
19500.
23700.
24010.
4737.
Maximum Concentrat
OZONE
Expt
(ppm)
0.419
0.393
0.405
0.318
0.369
0.296
0.410
0.215
0.376
0.396
0.325
0.225
0.344
0.485
0.021
0.273
0.736
0.116
0. 123
0.458
0.335
0. 154
0.283
0.589
0.396
0.384
0.627
0.021
0.555
0.491
0.396
0.688
0.443
0. 195
Calc
(ppm)
0.387
0.383
0.391
0.293
0.362
0.316
0. 288
0.236
0.354
0.426
0.375
0.325
0.341
0.430
0.007
0.414
0.666
0.261
0.334
0.537
0.366
0. 156
0.380
0.524
0.390
0.375
0.480
0.012
0.503
0.552
0.438
0.706
0.436
0. 179
Calc
-Expt
(ppm)
-0.032
-0.010
-0.014
-0.025
-0.006
0.021
-0. 122
0.021
-0.022
0.030
0.050
0. 101
-0.004
-0.055
-0,014
0. 140
0. 130
0. 146
0.212
0.079
0.031
0.081
0.062
0.060
0.097
-0.065
-0.006
-0.009
-0. 147
-0.009
-0.052
0.061
0.042
0.01B
-0.007
0.069
0.050
0.045
1on
Cole
-Expt
/Avg
-0.08
-0.03
-0.04
-0.08
-0.02
0.07
-0.35
0.09
-0.06
0.07
0.14
0.37
-0.01
-0.12
0.41
0. 16
0.77
0.93
0. 16
0.13
0.31
0.21
0.26
0. 29
-0. 12
-0.01
-0.02
-0.26
-0.10
0. 12
0. 10
0.03
0.00
0. 16
0. 12
0.10
Average
d( [03] -
Initial
[NO] )/dt
Calc
Expt Calc -Expt
(ppb/m1n) --
4.46
3.56
4.62
2.55
3.72
6.56
3.69
5.90
2.49
6.06
2.55
1 .53
2.50
4.72
0.49
1 .20
1 .76
0.63
0.66
1 .75
3.07
1.87
8. 18
10.72
1 1 .35
7.65
7.79
1.12
1 .97
1.31
1 .43
2.39
5.39
4.14
5.13
4.95
5.30
2.54
3.94
5.59
2.52
3.47
2.63
6.61
3.99
2.88
2.47
4.86
1 . 15
1 .03
1.51
0.65
0.79
1 .70
3. 19
1 .79
10.74
9.84
10.79
10.71
8. 20
1.97
1 .95
1 .63
1 .97
2. 76
6.06
4.29
0.67
1.39
0.68
-0.01
0.21
-0.97
-1.17
-2.43
0.14
0.55
1 .44
1 .35
-0.03
0.14
0.66
-0.16
-0.25
0.02
0.13
-0.06
0.12
0.91
0 .62
0.65
2.56
-0.88
-0.55
3.06
0.42
0.84
-0.02
0.32
0 .54
0.37
0.67
1 .25
0.96
1.01
Calc
-Expt
/Avg
0.14
0.33
0.14
0.00
0.06
-0.16
-0.38
-0.52
0.05
0.09
0.44
0.61
-0.01
0.03
0.80
-0. 15
-0.15
0.02
0.18
-0.03
0.07
0.31
0.21
0. 23
0.27
-0.09
-0.05
0.33
0.05
0.55
-0.01
0.22
0. 32
0.14
0. 17
0.20
0. 20
0.17
(cont1nued)
-------
Appendix B. Selected Results of Simulations of Individual Organ1c-NOx-A1r Experiments Using the
RAOM-M Mschanlsm (continued).
Page B
Experiment
Initial
Concantrat
1ons
NOx HC HC/NOx
(ppm) (ppmC)
17. Mesltylene-NOx
EC900
EC901
EC903
ITC703
ITC706
ITC709
ITC742
ITC826
Group Avsrsgs
S. D»v.
Avg. Abs. Value
S. Dev.
18. TetraHn-NOx
ITC739
ITC747
ITC746
ITC750
ITC832
Group Avarage
S. Dev.
Avg. Abs. Value
S. Dav.
19. Naphthalana-NOx
ITC751
ITC755
ITC756
ITC798
ITC802
Group Avaraga
S. Dev.
Avg. Abs. Valua
S. Oev.
0.53
0.51
1.00
0.50
0.49
0.99
0.48
0.90
0.68
0. 24
0.52
0.50
0.22
0.53
1 .00
0.55
0. 28
0.52
0.24
0.26
0.53
0.53
0.42
0. 15
54
2.7
4. 7
5.3
2.7
4.7
4.6
0.8
3.9
1 .6
2.4
93.2
84.0
44.5
39.4
52.7
36.7
7.5
14. 1
27.4
19.4
8.4
15.4
8.3
10.2
5.2
4.7
10.6
5.4
4.7
9.7
0.9
6.4
3.4
4.6
187.4
385.1
84.6
39.5
140.2
153.2
14.3
58.3
106.8
36.8
15.9
46.4
38.2
Ave
Temp
(deuK)
303.0
303.0
303 .0
303.0
303.0
303.0
303.0
303.0
303.0
0.0
303.0
303.0
303.0
303.0
303.0
303.0
0.0
303.0
303.0
303.0
303.0
303.0
303.0
0.0
Ave
H20
(ppm)
20000.*
20000.*
20000.*
21000.
21000.
21000.
18100.
17200.
18788.
1419.
17900.
17100.
17500.
17100.
16800.
17280.
427.
15100.
17500.
17200.
16400.
16800.
16600.
935.
Maximum Concantrat
OZONE
Expt
(ppm)
0.381
0.384
0.502
0.707
0.641
0.779
0.773
0.022
0.524
0.258
0.002
0.508
0.370
0.482
0.073
0.287
0.235
0.113
0. 259
0.282
0.204
0. 124
0. 196
0.077
Calc
(ppm)
0.419
0.323
0.403
0.478
0.475
0.528
0.481
0.008
0.389
0. 166
0.325
0.726
0.478
0.706
0.352
0.517
0. 190
0.534
0.453
0.491
0.634
0.542
0.531
0.068
Calc
-Expt
(ppm)
0.038
-0.061
-0.099
-0.229
-0. 165
-0.251
-0.293
-0.013
-0. 134
0. 1 19
0. 144
0. 106
0.322
0.217
0. 108
0.224
0.279
0.230
0.081
0.230
O.OB1
0.421
0. 193
0. 209
0.430
0.418
0.334
0. 122
0.334
0. 122
1 on
Avarage
d( [03] -
Calc
-Expt
/Avg
0
-0
-0
-0
-0
-0
-0
-0
0
0
0
0
0
0
1
0
0
0
0
1
0
0
1
1
0
0
0
0
.09
. 17
.22
.39
.30
.38
.47
.26
. 19
.29
. 13
.35
.26
.38
.31
.57
.50
.57
.50
.30
.54
.54
.03
.26
.93
.37
.93
.37
In1t
[NO]
1al
)/dt
Calc
Expt Calc -Expt
(ppb/m1n) --
3.85
8.88
14.96
14.59
7.20
1 1 .74
13.14
1 .68
9.50
4.97
0.63
2.73
2.60
2. 10
1 .38
1 .89
0.88
1 .08
1 .57
2.20
1 .66
1 .42
1 .59
0.41
3.57
5.63
9.62
4.45
7.09
8.01
1. IB
5.65
2.86
2.57
54.85
31 . 70
46.35
1 .00
27.30
24.72
1 1 .24
22.32
29.67
29.98
12.52
21.15
9.01
-5.
-9.
-4.
-2.
-4.
-5.
-0.
-4.
2.
4 .
2.
1 .
52.
29.
44.
-0.
25.
23.
25 .
23.
10.
20.
27.
28.
1 1 .
19.
8.
19.
8.
32
33
97
75
65
13
49
66
69
bb
69
95
12
1 1
25
37
41
96
bb
77
15
75
47
32
1U
56
67
bb
67
Calc
-Expt
/Avg
-0.85
-0.91
-0.41
-0.47
-0.49
-0.48
-0.34
-0.57
0.22
0.57
0.22
1 .22
1 .81
1 .70
1 .83
-0.31
1 .25
0.91
1 .37
0.64
1 .65
1 .74
1 .72
1 .79
1 .59
1 .70
0.08
J .70
o.oe
(cont1nued)
-------
Appendix B. Selected Results of Simulations of Individual Organlc-NOx-A1r Experiments Using the
RADM-M Mechanism (continued).
Page 9
Experiment
Initial
Concent rat 1 ons
NOx HC HC/NOx
(ppm) (ppmC)
20.
Ave
Temp
(degK)
Ave
H20
(ppm)
Maximum Concentrat
OZONE
Expt
(ppm)
Calc
(ppm)
Calc
-Expt
(ppm)
1 on
Calc
-Expt
/Avg
Average
d( [03] -
In1 t lal
[NO] )/dt
Calc
Expt Calc -Expt
(ppb/m1n) --
Calc
-Expt
/Avg
2,3-D1methylnapnthBlene-NOx
ITC775
ITC771
ITCB06
ITC774
Grouo Average
Avg
21 .
S. Dev.
. Abs. Value
S. Dev.
Simple Mixture
EC 144
EC145
EC160
EC149
EC 150
EC151
EC152
EC153
EC161
OC1278R
OC2578R
AU0180R
A 1)1 48 OR
EC 166
EC172
ST0682R
ST0682B
EC 106
EC1 13
EC1 14
EC1 15
EC 11 6
EC335
EC329
EC330
EC334
EC33B
EC328
JL1581R
JL1881R
ST2481R
AU27B1R
0.29
0.26
0.33
0.56
0. 36
0. 14
Runs
0.51
0.99
0.99
0.99
1 .00
2.06
0.50
0.97
0.51
0.46
0.44
0.56
0.47
0. 10
0. 10
0.46
0.45
0.50
0. 1 1
1.00
0.51
0.49
0.44
0.45
0. 29
0.45
0.45
0.45
0. 27
0.26
0. 23
0.23
1 .7
4.B
5.9
4.0
4. 1
1 .8
4.7
3.4
3.2
2.0
3.5
5.2
3.7
6.6
3.2
1 .4
1 .4
0.5
1 .4
9.2
2.8
2.8
2.9
9. 2
9.5
17.3
12.7
18.6
7. 7
4.2
4.3
8. 1
15.0
12. 1
2.3
2. 1
1 .6
2. 1
5.8
18.0
17.7
7. 1
12.2
6.6
9.3
3.4
3.3
2.0
3.5
2.5
7.3
6. '8
6.4
3.0
3. 1
0.8
3.0
92.0
28.9
6.2
6.5
18.3
85.0
17.3
25.2
37.6
17.4
9.2
14.6
18.2
33.7
27. 1
8.4
8.0
6.7
9.2
303.0
303.0
303.0
303.0
303 .0
0.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
294.3
288.4
304.9
305.2
303.0
303.0
299. 1
299. 1
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303 .0
303.0
303.0
303.0
304. G
303.2
292.2
299.5
16BOO.
20300.
17100.
22700.
19225.
2806.
24400.
24400.
29200.
29300,
29900.
30400.
26400.
25400.
36900.
16100.
12800.
23400.
34700.
27700.
21600.
22200.
22200.
20900.
19500.
28800.
22000.
27300.
24700.
27700.
26000.
27800.
27800.
27500.
20000. *
20000.*
22800.
30900.
0.274
0.293
0.360
0.341
0.317
0.040
1 .065
0.777
0.874
0.286
0.799
0. 147
0.791
1 .050
0.857
0.260
0. 147
0.256
0.863
0.462
0.369
0.378
0.47B
0.592
0.352
0.744
0.590
0.743
0.398
0.403
0.344
0.408
0.484
0.523
0.485
0.658
0.211
0.554
0.275
0.383
0.432
0.514
0.401
0. 100
0.772
0.376
0.326
0.131
0.329
0.086
0.665
0.817
0.599
0.250
0. 125
0. 107
0.474
0.377
0.297
0.487
0.522
0.614
0.441
0.702
0.606
0.782
0.370
0.384
0.348
0.409
0.4B7
0.507
0.691
0.594
0.276
0.498
0.001
0.090
0.072
0. 174
0.084
0.071
0.084
0.071
-0.293
-0.401
-0.548
-0. 155
-0.470
-0.060
-0. 127
-0.233
-0.259
-0.010
-0.022
-0. 149
-0.389
-0.085
-0.072
0. 108
0.044
0.021
0.089
-0.042
0.016
0.039
-0.028
-0.019
0.004
0.001
0.003
-0.016
0.206
-0.065
0.065
-0.056
0.00
0.27
0. 18
0.41
0.21
0.17
0.21
0.17
-0.32
-0.70
-0.91
-0. 74
-0.83
-0.52
-0. 17
-0.25
-0.36
-0.04
-0. 16
-0.82
-0.58
-0.20
-0.22
0.25
0.09
0.04
0. 22
-0.06
0.03
0.05
-0.07
-0.05
0.01
0.00
0.01
-0.03
0.35
-0.10
0.27
-0. 1 1
1 .73
2.67
2.69
2. 84
2.48
0.51
10.53
5.11
5.86
10.83
6.30
8.43
10.41
19.23
9.94
1.14
0.94
0.98
2.13
2.11
1 .00
1.11
1 .35
3.80
5.26
8.03
3.78
8.28
4.64
3.27
3.91
5.59
5. 19
3.50
1 .08
1.31
0.62
0.95
2.47
8.32
8.91
5.82
6.38
2.93
6. 15
3.84
3.48
5.20
4.12
7.04
7.84
11.91
5.43
1 .22
1 .01
0.64
1 .37
1 .37
0.70
1 .27
1 .58
4.05
7.06
6.07
3.40
9.25
4.21
3. 16
3.44
4.99
4.71
3.23
1 .23
1 .23
0.67
0.92
0.74
5.65
6.22
2.98
3.90
2.54
3.90
2.54
-4.39
-1 .27
-2.39
-5.63
-2.17
-1 .38
-2.57
-7.32
-4.51
0.08
0.07
-0.35
-0.76
-0.74
-0.29
0.16
0.23
0.25
1.81
-1 .95
-0.38
0.97
-0.43
-0. 10
-0.47
-0.59
-0.49
-0.27
0. 15
-0.08
0.05
-0.03
0.35
1 .03
1 .07
0.69
0.79
0.34
0. 79
0. 34
-0.53
-0.28
-0.51
-0. 70
-0.42
-0. 18
-0.28
-0.47
-0.59
0.07
0.07
-0.43
-0.44
-0.42
-0. 35
0.13
0.15
0.06
0. 29
-0.28
-0. 10
0. 1 1
-0.10
-0.03
-0.13
-0.11
-0. 10
-0.08
0.13
-0.06
0.08
-0.03
(cont1nued)
-------
Appendix B. Selected Results of Simulations of Individual Organ1c-N0*-A1r Experiments Using the
RADM-M Mechanism (continued).
Page 10
j
T\
Experiment
Initial
Concentrations
NOx
Ave
Temp
Ave
H20
HC HC/NOx
(ppm) (ppmC)
OC03B2R
OC03BZB
NV15B2R
JL2281R
JL2181B
NV1582B
DE0782B
JN1379B
Group Average
S. Dev.
Avg. Abs. Value
S. Dev.
22. 3-Component
ITC479
ITC5B4
ITC579
ITC472
ITC474
ITC581
ITC5B5
ITC47B
ITC4B2
ITC4BB
ITC492
ITC494
ITC498
ITC500
ITC502
ITC462
ITC466
ITC468
ITC451
ITC455
ITC977
ITC985
ITC997
JTC979
ITC992
ITC981
ITC993
JL1581B
ST2481B
0.25
0. 25
0. IB
0. 26
0.24
0. IB
0. 19
0.44
0.49
0.37
and "Mini"
0.09
0. 10
0. 10
0. 10
0.09
0.09
0. 10
0. 10
0. 10
0.09
0.09
0.09
0. 10
0. 10
0.09
0. 11
0.10
0.09
0. 10
0.09
0. 13
0. 12
0. 12
0. 12
0. 12
0. 14
0. 1 1
0.28
0.23
2.7
1 .9
2.6
2.4
1 .3
2.7
3.5
6. 1
5.2
4.5
11.0
7.7
14.4
9. 1
5.4
14.9
18.7
13.7
15.5
19.2
(degK)
297.2
296.9
283.8
302.3*
304.2*
283.8
286.4
294.9
300.2
5.7
(ppm)
21200.
21100.
20000.*
20000.*
20000.*
20000.*
13400.
20000.*
24160.
5188.
Maximum Concent rat
OZONE
Expt
(ppm)
0.247
0. 158
0.061
0.826
0.237
0.167
0.093
0.756
0.492
0.270
Calc
(ppm)
0. 137
0.038
0.042
0.638
0.441
0.245
0.303
0.809
0.428
0.220
Cslc
-Expt
(ppm)
-0.110
-0. 120
-0.019
0.012
0.204
0.079
0.210
0.053
-0.065
0. 172
0. 123
0. 136
Ion
Calc
-Expt
/Avg
-0.57
-1 .22
-0.36
0.02
0.60
0.38
1 .06
0.07
-0.15
0.43
0.32
0.32
Average Initial
d( (03) - [NO] )/dt
Expt
Calc Calc
Calc -Expt -Expt
(ppb/m1n) -- /Avg
0.76
0.62
0.42
1. 18
0.63
0.71
0.55
2.44
4. 10
4.06
0.54 -0.22 -0.33
0.33 -0.09 -0.24
1.12 -0.06 -0.05
0.65 0.01 0.02
0.74 0.03 0.05
0.71 0.17 0.27
2.78 0.33 0.13
3.30 -0.89 -0.15
2.76 1.82 0.25
1.11 0.23
1 . 69 0.18
Surrogate*
4.0
4.0
3.4
3.6
4. 2
4.3
4.6
5.0
1 . 2
6.0
B. 1
B.5
3.9
4. 1
4.6
7.5
6.5
4.6
S. 2
4.4
3.2
2.7
2.7
2.9
2.6
6.5
7. 1
2.3
1.9
43.4
40.7
34.7
35.1
45. 1
50.0
50. B
52.6
12.3
66.2
BB.4
92.5
40.2
42.7
49.5
71.3
62.7
49.0
52.0
47.8
25. 1
22.9
23.0
23. 1
20.9
45.0
66.9
8.0
B.I
303.0
303.0
303.0
303.0
303.0
303.0
30J.O
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303. 0
303.0
303.0
303.0
303.0
303.0
303 . 0
303.0
303.0
304.6
292. 2
20300.
20300.
15800.
13700.
15600.
16100.
17100.
18300.
18300.
19600.
15000.
21 100.
19300.
18700.
18100.
18300.
18200.
18000.
19800.
20200.
20900.
20100.
20100.
20100.
20900.
19000.
20100.
20000. *
14200.
0.332
0.375
0.356
0.25B
0.293
0.352
0.311
0.320
0.279
0.293
0.326
0.321
0.320
0.286
0.271
0.099
0.131
0. 176
0.312
0.304
0.329
0.336
0.316
0.404
0.384
0.328
0.276
0.474
0.246
0.317
0.314
0.292
0.319
0.327
0.313
0.332
0.351
0.292
0.327
0.350
0.358
0.330
0.318
0.302
0.267
0.280
0.299
0.294
0.295
0.324
0.294
0.292
0.414
0.404
0.377
0.365
0.619
0.274
-0.015
-0.060
-0.064
0.061
0.034
-0.039
0.021
0.031
0.013
0.034
0.024
0.036
0.009
0.031
0.031
0. 168
0. 149
0. 123
-0.018
-0.008
-0.006
-0.042
-0.024
0.010
0.020
0.049
0.089
0. 145
0.028
-0.05
-0.18
-0.20
0.21
0.11
-0.12
0.07
0.09
0.05
0.11
0.07
0.11
0.03
0. 10
0. 1 1
0.92
0.72
0.52
-0.06
-0.03
-0.02
-0. 13
-0.08
0.02
0.05
0. 14
0.28
0.27
0. 1 1
2.95
3.05
1.54
1 .72
3.40
3.74
5. 11
4.73
2.44
3.06
3.28
3.30
2. 16
5. 18
7.07
0.70
0.77
1 . 19
3.95
3.52
3.94
3.95
3.89
4.88
5.57
2.56
2.45
1 . 10
0.71
4.81 1.B7 0.4B
3.95 0.90 0.26
2.14 0.61 0.33
3.01 1.30 0.55
6.10 2.70 0.57
5.30 1.56 0.35
7.00 1.89 0.31
8.20 3.48 0.54
4.41 1.97 0.57
5.18 2.12 0.51
5.40 2.12 0.49
5. 34 2 .04 0.47
3. 63 1 . 47 0.51
7.42 2.23 0.35
12.32 5.25 0.54
2.49 1 . 79 1.12
2.93 2.16 1.17
3.46 2.27 0.98
5.68 1.73 0.36
5.08 1 .56 0.36
3.51 -0.43 -0.12
3.32 -0.63 -0.17
3.27 -0.61 -0.17
4.11 -0.77 -0.17
4.49 -1 .08 -0.22
3.73 1.17 0.37
3.21 0.76 0.27
1.15 0.05 0.05
0.70 -0.01 -0.01
(cont 1 nued)
-------
Appendix B. Selected Results of Simulations of Individual Organlc-NOx-A1r Experiments Using the
RAOM-M Mechanism (continued).
Page 11
Experiment
Initial
Concentrat Ions
NOx
Ave
Temp
Ave
H20
HC HC/NOx
(ppm) (ppmC)
JN0982B
JN1483R
JN2783B
AU1883B
AU2683R
JL18B1B
JN09B2R
Group Average
S. Dev.
Avg. Abs. Value
S. Dev.
0.28
0.22
0.26
0.28
0.32
0.27
0.29
0. 14
0.08
3. 1
2.6
2.9
0.6
2.6
2.3
3. 1
4. 1
1 .9
11 .0
11.6
11.1
2.0
a. i
a. 5
10.7
37.0
23.7
(deBK)
299.7*
301 .2
305. 1
303.0
302.2
303.2
299.7*
302.6
2.0
(ppm)
14300.
26100.
24400.
23100.
26400.
20000.*
14700.
19061 .
3028.
Maximum Concentration
OZONE
Expt
(ppm)
0.667
0.585
0.51 1
0.556
0.646
0.693
0.714
0.366
0. 148
Calc
(ppm)
0.578
0.516
0.579
0.530
0.596
0.620
0.634
0.380
0.117
Calc
-Expt
(ppm)
-0.089
-0.069
0.069
-0.025
-0.050
-0.073
-0.080
0.014
0.065
0.051
0.042
Calc
-Expt
Average Initial
d( [03] - [NO] )/dt
Expt
Calc Calc
Calc -Expt -Expt
/Av9 -- (ppb/m1n) /Avg
-0. 14
-0. 13
0.13
-0.05
-0.08
-0. 1 1
-0.12
0.08
0.24
0.16
0. 19
1 .37
1 . 10
1 .01
0.70
1 .24
1 .91
1 .77
2.81
1 .61
1.19 -0.18 -0. 14
1.06 -0.04 -0.04
1.12 0.11 0.10
0.65 -0.05 -0.07
1.17 -0.07 -0.06
1.76 -0.15 -0.08
1.55 -0.22 -0.13
3.88 1.08 0.28
2.44 1.36 0.36
1.32 0.36
1.13 0.28
23. SAPRC 7-Component Surrogates
EC231
EC232
EC233
EC237
EC238
EC241
EC242
EC243
EC245
EC246
EC247
Group Average
S. Dev.
Avg. Abs. Value
S. Dev.
0.49
0.49
0.10
0.48
0.95
0.49
0.50
0.50
1 .00
0.51
0.51
0.55
0.24
13.2
9.3
9.5
10.5
10. 1
5.0
12.9
9.7
12.9
8.6
6.2
9.8
2.6
26.9
IB. 9
92.5
21 .6
10.6
10. .2
25.6
19.5
12.9
17.0
12.2
24.4
23.3
303.0
303.0
303.0
303.0
303.0
303.0
303 .0
303.0
303.0
303.0
303.0
303.0
0.0
17400.
17100.
21400.
23200.
25400.
20700.
24600.
16400.
20400.
22100.
22700.
21036.
3020.
0.620
0.342
0.330
0.652
0.691
0.408
0.682
0.716
0.894
0.574
0.657
0.597
0. 173
0.715
0.374
0.3B6
0.648
0.701
0.37B
0.686
0.722
0.856
0.434
0.627
0.593
0. 170
0.094
0.032
0.056
-0.004
0.010
-0.030
0.004
0.006
-0.038
-0. 140
-0.030
-0.004
0.060
0.041
0.043
0.14
0.09
0.16
-0.01
0.01
-0.08
0.01
0.01
-0.04
-0.28
-0.05
0.00
0. 12
0.08
0.08
6.99
3. 12
4.07
7.36
5. 10
2.87
17.53
14.50
13.41
2.65
7.49
7.74
5.16
7.67 0.68 0.09
2.91 -0.21 -0.07
4.34 0.27 0.06
5.89 -1 .47 -0. 22
4.71 -0.39 -0.08
2.90 0.03 0.01
16.65 -0.88 -0.05
12. 77 -1 .74 -0.13
12.82 -0.59 -0.05
1.95 -0.70 -0.30
6.49 -1.00 -0.14
7.19 -0.55 -0.08
4.83 0.73 0.12
0.72 0.11
0.53 0.09
24. SAPRC 8-Component Surrogates
ITC626
ITC630
ITC631
ITC633
ITC635
ITC637
ITC865
0.30
0.31
0.32
0.64
1.21
0.31
0.2B
4.0
1 .9
1 .0
4.0
4.0
4.0
8.4
13.4
6.3
3.2
6.2
3.3
12.8
29.9
303.0
303.0
303.0
303.0
303.0
303.0
303.0
14300.
16300.
17500.
16600.
17500.
16900.
15100.
0.618
0.284
0.043
0. 231
0.006
0.617
0.632
0.540
0.270
0.059
0. 279
0.018
0.537
0.513
-0.078
-0.014
0.016
0.046
0.013
-0.080
-0. 120
-0.13
-0.05
0.19
-0.14
-0.21
2. 18
0.86
0 .47
1 .26
0.95
2.31
2.60
2.26 0.09 0.04
1 .00 0.14 0.15
0.52 0.05 0.10
1 . 50 0 . 24 0.17
1.00 0.05 0.05
2.24 -0.07 -0.03
3.43 0.84 0.28
(cent1nued)
-------
Appendix B.
Results of Simulations of Individual Organ1c-N0x-A1r Experiment* lifting the
RADM-M MechanUm (continued).
Page 12
Experiment
Initial
Concentrat lone
NO* HC HC/NOn
(ppm) (ppmC)
ITC867
ITCB6B
ITC871
ITC872
ITC873
ITC874
ITC877
ITC880
ITCB81
ITCB85
ITC8S6
ITCB8B
ITCB91
OTC189A
OTC189B
OTC190A
OTC190B
OTC192A
OTC192B
OTC194A
OTCJ94B
OTC195A
OTCI95B
OTC196B
OTC197A
OTC197B
OTC198A
OTC198B
OTC199A
OTC199B
OTC202A
OTC202B
OTC203A
OTC203B
OTC204A
OTC204B
OTC205A
OTC205B
OTC215A
OTC215B
OTC217A
OTC217B
OTC221A
OTC221B
OTC222A
0.28
0.37
0.37
0.38
0.39
0.38
0.38
0.73
0.73
0.64
0.73
0.33
0.32
0.45
0.45
0.41
0.42
0.47
0.47
0.38
0.38
0.41
0.41
0.41
0.38
0.77
0.84
0.40
0.36
0.37
0.73
0.40
0.39
0.19
0.35
0. 17
0.84
0. 14
0.45
0.44
0.50
0.51
0.41
0.42
0.44
4.8
2.9
1 .7
2. 1
1 .3
2. 1
2.3
2.2
2.3
1 .5
2.3
4.7
4.4
3.3
4.0
3.8
3.8
4.2
4.2
7.4
3.7
1.8
4. 1
3.7
3.6
3.1
5.4
3. 7
3.5
3.5
6.0
2.6
3.5
3.5
3.6
3.5
3.6
3.6
3.5
4. 1
4.9
4.6
2.4
2.0
3.4
17.2
7.8
4.6
5.7
3.4
5.7
6.2
3.0
3. 1
2.4
3. 1
14.5
13.7
7.4
8.9
9.2
9.0
8.9
8.9
19.3
9.7
4.5
9.9
9. 1
9.5
4. 1
6.4
9.4
9.6
9.6
8.2
6.7
9.0
18.5
10.2
20.3
4.5
26.5
7.8
9.3
9.6
9.0
5.8
4.7
7.8
Ave
Temp
(degK)
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
295.4
298.0
302.7
299.5
297. 6
298.4
296. 1
295.7
351 .6
360.0
337.5
353.2
358.0
353.5
347.5
299.7
298.9
325. 1
324. 1
293.3
293.7
322.2
34B.8
464.9
433.4
293. 1
293.4
295.5
294.2
290.6
288. 8
298.5
Ave Maximum Concentration
H20 OZONE
Expt
(ppm) (ppm)
14600. 0.631
17100. 0.518
14600. 0.376
14600. 0.213
14600. 0.160
13200. 0.191
14600. 0.250
13700. 0.031
14000. 0.012
14600. 0.012
20000.* 0.012
18000. 0.579
20000.
5000.
5000.
5400.
5410.
5000.
5000.
5000.
5000.
4660.
3180.
5000.
5000.
5000.
5000.
500O.
5000.
5000.
5000.
4560.
4850.
4880.
4330.
4330.
5000.
5000.
4420.
4420.
7460.
7460.
4460.
4460.
5000.
0.602
0.576
0.712
0.702
0.669
0.735
0.754
0.737
0.603
0. 164
0.6B1
0.597
0.621
0. 152
0.704
0.674
0.591
0.606
0.714
0.522
0. 231
0.394
0.302
0. 381
0.039
0.392
0.830
0.868
0.483
0.831
0.235
0.333
0.909
Calc
(ppm)
0.539
0.423
0.255
0.220
0. 109
0. 152
0.229
0.042
0.030
0.025
0.021
0.505
0.511
0. 757
0.833
0.028
0.782
O.B41
0. '47
0 . B97
0.695
0. 105
0.764
0.7B9
O.B49
0.093
0.568
0.627
0.645
0.630
0.475
0.413
0.279
0.447
0.374
0.465
0.054
0.546
0.850
0.931
0.570
0.922
0.279
0.349
0.945
Calc
-Expt
(ppm)
-0.092
-0.095
-0.121
0.007
-0.051
-0.038
-0.022
0.011
0.018
0.013
0.009
-0.074
-0.091
0.181
0. 121
0. 126
0. 1 13
0. 106
-0.007
0. 160
0.091
-0.059
0.083
0. 192
-0.072
-0.059
-0. 136
-0.047
0.054
0.024
-0.239
-0. 109
0.048
0.053
0.071
0.084
0.015
0. 154
0.020
0.063
0.087
0.091
0.044
0.016
0.036
Calc
-Expt
/Avg
-0. 16
-0.20
-0.38
0.03
-0.38
-0. 22
-0.09
-0. '4
-0 . 16
0 27
0. 16
0.16
0 16
0 13
-0.01
0 20
0. 14
-0. 44
0.11
0. 28
-0. 12
-0.48
-0. 21
-0. 07
0.09
0. 04
-0. 40
-0.23
0. 19
0. 12
0.21
0. 20
0.33
0.02
0.07
0.17
0. 10
0.17
0.05
0.04
Average
d( (03] -
Initial
INO] )/dt
Calc
Expt Calc -Expt
(ppb/m1n) --
2.23
2. 10
1. 18
1 . 20
0.85
0.63
1 . 14
0.85
1 .03
0.44
1 .24
1 .65
2.67
3.79
3.70
4.24
3.73
4.25
3.84
7.85
2.95
2.05
4.14
3.66
4. 12
2.86
6.09
3.66
3.B6
3.40
6.95
3.62
2.34
2.20
3.27
2.62
4.85
4.49
4.33
5.33
2.55
4 .64
1 .63
1.63
3.B3
3.37
1 .41
1 .09
1 .33
0.78
0.85
1 .31
0.96
1 .08
0.76
0.69
1 .84
2.84
3.55
3.67
3.30
2.90
3.35
2.75
7.55
3.00
1 . 16
3.73
3.30
2.63
2.02
3.63
2.55
3.00
2.88
12.47
3.10
2.29
2.61
3.36
2.86
2.09
3.64
3.61
5.48
2.47
4. 24
1.48
1.59
3.67
1. 15
-0.6B
-0. 10
0. 13
-0.07
0.02
0. 17
0. 1 1
0.04
0.33
-0.55
0. 19
0. 17
-0.24
-0.03
-0.95
-0.83
-0.91
-1 .09
-0.31
0.05
-0.89
-0.40
-0.56
-1.48
-0.84
-2.46
-1.31
-0.86
-0.52
5.52
-0.72
-0.05
0.41
0.09
0.24
-2.76
-0.85
-0.72
0. 15
-0.08
-0.40
-0. 14
-0.04
-0. J6
Calc
-Expt
/Avg
0.41
-0.39
-0.09
0. 10
-0.09
0.03
0.14
0.12
0.04
0.54
-0.57
0.11
0.06
-0.07
-0.01
-0.25
-0.25
-0.24
-0.33
-0.04
0.02
-0.55
-0. 10
-0.16
-0.44
-0.34
-0.51
-0.41
-0.25
-0.17
0.57
-0. 21
-0.02
0.17
0.03
0.09
-0.80
-0.2J
-0. IB
0.03
-0.03
-0.09
-0.09
-0.03
-0.04
(contInued)
-------
Appendix B. Selected Results of Simulations of Individual Organ1c-NO»-A1r Experiments Using the
RAOM-M Mechanism (continued).
Page 13
Experiment
Initial
Concentrat Ions
NOx
Ave
Temp
HC HC/NOx
(ppm) (ppmC)
OTC222B
OTC223A
OTC223B
OTC224A
OTC224B
OTC226A
OTC22BA
OTC228B
OTC229A
OTC229B
OTC230A
OTC230B
OTC237A
OTC237B
OTC23BA
OTC23BB
OTC239A
OTC239B
OTC240A
OTC240B
OTC241A
OTC241B
OTC242A
OTC242B
OTC243A
OTC243B
OTC24BA
OTC248B
OTC249A
OTC249B
Group Average
S. Oev.
0.43
0.36
0.40
0.34
0.34
0.45
0.41
0.41
0.46
0.46
0.41
0.41
0.52
0.52
0.50
0.50
0.49
0.50
0.50
0.50
0.32
0.31
0.45
0.46
0.47
0.38
0.48
0.46
0.48
0.46
0.45
0. 16
2.6
4.8
3.5
4.4
4.3
2.5
2.4
2.3
3.0
1 .7
3. 1
1 .7
4.3
4.4
2.9
3.9
2.7
2.5
1 .8
2.0
4.7
4.0
2.6
2. 1
4.7
3.7
3.7
2.9
5.6
4.9
3.5
1 .3
6.0
13.4
8.6
12.7
12.6
5.5
6.0
5.6
6.6
3.8
7.6
4.2
8.2
8.5
5.8
7.8
5.4
5.0
3.6
4. 1
14.8
12.9
5. .8
4.7
10.0
9.7
7.7
6.3
1 1 .8
10.7
8.7
4.9
(degK)
296. 1
314.9
314.7
303.5
303.8
312.5
293 . 2
294.3
295.7
296.0
297.9
298.2
292.3
292.8
292.5
292.0
290.2
290.9
93.4
104 . 2
290. 7
291 . 1
302.0
302. 1
276.5
277.4
271 .6
268 .9
286.9
286.2
302.3
43.6
Ave Maximum Concentration
H2O OZONE
Expt
(ppm) (ppm)
5000.
4270.
4270.
6410.
6410.
6840.
5000.
5000.
5000.
5000.
5000.
5000.
3810.
3810.
4810.
4610.
5150.
5150.
6210.
6210.
4760.
4760.
4670.
4670.
4720.
4720.
5000.
5000.
3540.
0.940
0.953
0. 771
0.776
0.813
0. 751
0. 246
0. 296
0.253
0. 168
0.489
0.271
0.807
0.757
0.406
0.702
0.343
0.234
0.034
0.217
0.671
0.674
0. 1H2
0.639
0.142
0. 152
0.056
0.081
0.34B
3540. 0.325
7630. 0.454
4865. 0.272
Calc
(ppm)
1 . 107
0.866
0.714
0.774
0.836
0.530
0.421
0.456
0.446
0.391
0.628
0.550
0.625
0.676
0.393
0.696
0.390
0. 159
0.034
0. 1 10
0.693
0.679
0. 148
0.623
0.201
0.171
0.055
O.OB5
0.330
0.360
0.474
0.284
Avg. Abs. Value
S. Oev.
25. UNC Mlscel
ST20B1R
OE0782R
AU2681R
AU2681B
AU2781B
ST0381R
ST1081R
ST2081B
Calc
-Expt
(ppm)
0. 167
-0.086
-0.057
-0.002
0.023
-0.221
0. 175
0. 160
0. 193
0.223
0.339
0.279
0.018
-0.081
-0.013
-0.006
0.047
-0.075
0.000
-0. 107
0.022
0.005
-0.034
-0.016
0.059
0.019
-0.001
0 . 004
-0.018
0.035
0.020
0. 102
0.076
0.070
Calc
-Expt
/Avg
0. 16
0.10
0.08
0.00
0.03
-0.35
0.52
0.43
0.55
0.80
0.52
0.68
0.02
-0. 1 1
-0.03
-0.01
0.13
-0.38
-0.65
0.03
0.01
-0.20
-0.03
0.34
0. 1 1
-0.03
0.05
-0.05
0.10
0.03
0.26
0. 19
0. 17
Average
d( [03] -
Expt
Calc
Initial
[NO] )/dt
Calc Calc
-Expt -Expt
(ppb/mln) -- /Avg
5.55
4.55
3. 15
3.67
4.94
3.12
1 .49
1 .79
1 .56
1.38
2.43
1 .66
4. 10
3.48
2.58
4.06
2.66
1 .85
1 .08
1 .79
3.26
3.30
1 .47
3.04
1 .41
1 .40
1 .40
1 .61
2.55
2.83
2.81
1 .51
6. 17
4.06
2.83
2.91
4.41
2.32
1 .92
2.56
1 .96
1.83
3.07
2. 16
3.39
2.92
2. 18
3.56
2.49
1 .46
0.90
1 .36
2.94
3.26
1.31
2.64
1.78
2. 14
1 .20
1 .69
2.29
2.97
2.65
1 .65
0.62 0.1)
-0.47 -0.11
-0.33 -0.11
-0.76 -0.23
-0.53 -0.11
-0.80 -0.29
0.43 0.25
0.77 0.36
0.40 0.23
0.45 0.28
0.64 0.23
0.50 0.26
-0.71 -0.19
-0.56 -0.17
-0.40 -0.17
-0.49 -0.13
-0.16 -0.06
-0.39 -0.24
-0. IB -0. 18
-0.43 -0.28
-0.32 -0, 10
-0.05 -O.01
-0.17 -0.12
-0.40 -0. 14
0.37 0.23
0.74 0.42
-0.20 -0.16
O.OB 0.05
-0.26 -0.11
0.14 0.05
-0.16 -0.05
0.89 0.24
0.54 0.19
0.73 0.16
laneous Surrogates
0.23
0. 19
0.24
0.24
0. 23
0. 24
0. 25
0.23
2.3
3.4
2.0
2.0
2.0
1 .8
2.8
2. 1
10.0
18.3
8.4
8.5
8.8
7.6
11.3
9.2
294. 1
286.4
29B.B
295.7
296.4
301 .0
297.9
292.6
21900. 0.403
13400. .0.076
25500. 0.506
27500. 0.544
25000. 0.623
23200. 0.541
22300. 0.610
15200. 0.414
0.463
0.361
0.514
0.442
0.469
0.453
0.598
0.415
0.059
0. 285
0.007
-0. 102
-0. 153
-O.OBB
-0.012
0.000
0.14
1 .30
0.01
-0.21
-0.28
-0. 18
-0.02
0.00
0.95
0.52
. 18
.22
.27
.48
.59
.06
1 .05
0.74
.08
.00
.07
.22
.54
0.98
0.10 0.10
0.22 0.35
-0.11 -0.09
-0.22 -0.20
-0.21 -O.18
-0.26 -0. 19
-0.05 -0.03
-0.07 -0.07
( cont 1 niied )
-------
Appendix B.
Selected Results of Simulations of Individual Organ1c-N0x-A1r Experiments Using the
RADM-M Mechanism (continued).
Page 14
J
«1
Experiment
Initial
Concentrat Ions
Ave
Temp
Ave
H20
NOx HC HC/NOx
(ppm) (ppmC)
JL2081B
ST1682R
JL2081R
JL2281B
OC14S1R
ST16B2B
ST2981R
ST2981B
OC1481B
ST038IB
ST1081B
JLOB82R
JL0882B
Group Average
S. Oev.
Avg. Aba. Value
S. Dev.
26. UNC "Synurban"
AU2284R
AU25B4R
AU2584B
ST0184R
ST0184B
ST0284R
ST0284B
JN2685R
JN2685B
JN2685R
JN2BB5B
Group Average
S. Dev.
Avg. Aba. Value
S. Dev.
0.42
0.43
0.41
0.26
0.2B
0.43
0.24
0. 24
0.29
0.23
0.24
0.29
0.28
0.28
0.08
Surrogate
0.32
0.34
0.33
0.31
0.30
0.34
0.33
0.30
0.30
0.38
0.39
0.33
0.03
1.8
3.2
2.7
2.9
3.3
3. 1
2.5
2.5
2.9
2.0
1 .0
2. 1
2. 1
2.4
0.6
0.4
0. 1
0. 1
0.2
0.2
0. 1
0. 1
0.2
0.3
0. 1
0.2
0.2
0. 1
4.3
7.5
6.6
11.2
11.9
7.2
10.3
10.4
9.9
8.6
4. 1
7.3
7.4
9.0
3.0
1.2
0.3
0.4
0.6
0.8
0.4
0.3
0.6
0.9
0.4
0.5
0.6
0.3
(degK)
305.0*
304.4
305.0*
302.3*
291 .3
304.4
293.7
291 .6
288.9
297.9
295.3
304.2
304.2
297.7
5.8
302.7
302.0
302.0
302.5
302.5
304.6
304.5
303.6
303.6
299.8
299.8
302.5
1 .6
(ppm)
20000.*
23000.
20000.*
20000.*
9470.
23000.
18700.
14400.
10300.
16800.
17700.
16000.
16000.
19018.
4902.
8470.
22600.
22600.
9380.
9380.
19900.
19900.
16800.
16800.
1 18OO.
11800.
15403.
5427.
Maximum Concentrat
OZONE
Expt
(ppm)
0.165
0.410
0.635
0.722
0.462
0.840
0.294
0.485
0.458
0.61 t
0.626
0.598
0.541
0.503
0.175
0.657
0.075
0.096
0.546
0.646
0. 1 19
0.020
0.629
0.788
0. 238
0.275
0.372
0.284
CalC
(ppm)
0.473
0.508
0.666
0.652
0.570
0.784
0.427
0.492
0.564
0.517
0.376
0.668
0.651
0.527
0.110
0.756
0.105
0. 130
0.653
0.687
0.261
0. 140
0.662
0.742
0.387
0.271
0.436
0.266
Calc
-Expt
(ppm)
0.308
0.098
0.031
-0.070
0. 108
-0.055
0. 134
0.008
0. 106
-0.093
-0.250
0.070
0. 1 10
0.024
0. 134
0. 102
0.086
0.099
0.030
0.034
0. 107
0.041
0. 142
0.120
0.034
-0.046
0. 148
-0.004
0.064
0.063
0.073
0.051
1 on
Calc
-Expt
/Avg
0.97
0.21
0.05
-0.10
0.21
-0.07
0.37
0.02
0.21
-0. 17
-0.50
0.11
0. 18
0. 11
0.40
0.25
0.32
0. 14
0.33
0.30
0. IB
0.06
0.74
0.05
-0.06
0.47
-0.01
0.22
0.25
0.24
0.23
Average
d( [03] -
Expt
Calc
Initial
[NO] )/dt
Calc
-Expt
-- (ppb/m1n) --
0.82
.28
.55
.80
.53
2.39
0.76
.35
.48
.70
.36
.25
. 12
1.32
0.40
1 .61
0.70
0.69
1 .32
1 .72
0.73
0.40
1 .34
1 .74
0.85
0.87
1 .09
0.47
0.94
. 17
.43
.70
.56
2.08
0.88
.29
.50
.55
0.95
.33
.29
1 .25
0.32
1 .63
0.64
0.65
1 .30
1 .57
0.75
0.49
1 .45
1 .84
0.8B
0.87
1 . 10
0.47
0.12
-0.12
-0. 12
-0. 10
0.04
-0.32
0. 12
-0.06
0.02
-0.15
-0.41
0.09
0. 17
-0.06
0.17
0. 15
0. 10
0.02
-0.06
-0.04
-0.02
-0. 15
0.02
0.09
0.11
0. 10
0.03
0.00
0.01
0.08
0.06
0.05
Calc
-Expt
/Avg
0. 14
-0.09
-0.08
-0.06
0.02
-0. 14
0. 15
-0.04
0.02
-0.09
-0.35
0.07
0. 14
-0.03
0.15
0. 12
0.09
0.01
-0. 10
-0.06
-0.01
-0.09
0.02
0.21
0.08
0.06
0.04
0.01
0.01
0.09
0.06
0.06
(cortt 1 nued)
-------
Appendix B. Selected Results of Simulations of Individual Organ1c-N0x-A1r Experiments Using the
RADM-M Mechanism (continued).
Page 15
Experiment
Initial
Concent rat Ions
NOx HC HC/NOx
(ppm) (ppmC)
Ave
Temp
(degK)
Ave
H20
(ppm)
Maximum Concentration
OZONE
Expt
(ppm)
Calc
(ppm)
Calc
-Expt
(ppm)
Calc
-Expt
/Avg
Average
d( [03] -
Initial
(NO] )/dt
Calc
Expt Calc -Expt
(ppb/m1n) --
Calc
-Expt
/Avg
27. UNC "Synauto" Surrogate
AU0484R
AU0484B
AU0584R
AU0584B
AU0684R
AU06B4B
AU0784R
AU08B4B
AU0984R
ST08B4R
STOB84B
ST17B4R
ST1784B
ST2184R
ST21B4B
Group Average
S. Dev.
Avg. Abs. Value
S. Dev.
2B. UNC Auto Exhaust
OC0483B
OC07B3R
OC07B3B
JN2582R
JN25B2B
JN2982R
JN29B2B
JN3082R
JN3082B
JL0283B
JL0883B
ST2982B
OC0682R
AU1 183R
AU1 183B
JL0182R
JL01B2B
0.37
0.36
0.36
0.35
0.35
0.35
0.38
0.34
0.39
0.34
0.33
0.34
0.34
0.36
0.36
0.35
0.02
0.25
0.33
0.34
0.65
0.65
0.24
0.25
0.32
0.32
0. 19
0.37
0.39
0.46
0.22
0.23
0.37
0.35
0.4
0.3
0.3
0.4
0.7
1 . 1
0.4
1 . 1
0.4
0.6
1 .0
1 . 1
0.8
0.6
0.7
0.7
0.3
0.4
2.7
2.7
0.6
0.6
2.5
2.5
2.8
0.6
1 .7
1 .7
1 .7
2.0
2. 2
0.7
3.5
3.6
1.0
0.8
0.7
1 .0
2. 1
3.2
1 .0
3. 1
0.9
1 .9
2.9
3.3
2.3
1 .7
2.0
1 .9
0.9
1 . 7
B. 1
7.9
0.9
0.9
10.3
9.8
B.7
1 .7
9. 1
4.6
4.3
4.3
9.9
3. 1
9.6
10.2
307.5
307.5
307. 1
307. 2
307. 6
307.6
308.0
307.3
307.8
298.5
298.5
294.4
294.4
302. 2
302. 2
303.9
5. 1
299. 7
295.5
295.4
301 . 1»
301 . 2»
303. 1
303. 1
302.5*
302.3*
304.0
299.5
295. 7
300. 2
306.8
306.9
299.5
299.3
34900.
34700.
44900.
44900.
36800.
36BOO.
34700.
29000.
31500.
6920.
6920.
24800.
24800.
7350.
7350.
27089.
13697.
18800.
21600.
21500.
22700.
22900.
30500.
30500.
20000. *
20000. *
24500.
21200.
24800.
16900.
21400.
21400.
27700.
27400.
0.515
0.328
0.335
0.595
0.887
0.940
0.602
0.834
0.521
0.566
0.750
0.539
0.484
0.671
0.721
0.619
0. 182
U.642
0. 178
0.451
0.003
0.003
0.704
0.766
0.81 1
O.B40
0.697
0.879
0.205
0.355
0.850
0.601
0.740
0.759
0.627
0.548
0.450
0.636
0.823
O.B70
0.651
0.798
0.541
0.740
0.869
0.710
0.696
O.B11
O.B29
0.707
0. 130
0.708
0.365
0.640
0.012
0.013
0.823
O.B29
0.985
0.992
0.839
0.638
0.054
0.267
0.878
0.665
1 .032
0.989
0.112
0.220
0.115
0.042
-0.064
-0.070
0.049
-0.035
0.021
0. 174
0.119
0.171
0.212
0.141
0. 108
0.086
0.094
0.110
0.064
0.066
0.207
0. 190
0.010
0.010
0. 1 19
0.063
0. 174
0. 151
0.141
-0.041
-0. 151
-0.088
0.028
0.065
0.292
0.230
0.20
0.50
0.29
0.07
-0.07
-0.08
0.08
-0.04
0.04
0.27
0.15
0.27
0.36
0. 19
0. 14
0. 16
0.17
0. 18
0.13
0. 10
0.73
0.35
0.16
0.08
0. 19
0. 17
0. 18
-0.05
-1.17
-0. 28
0.03
0.10
0. 33
0. 26
1 .45
1 .16
1 .08
1 .48
2.53
3. 16
1 .61
2.92
1 .51
1 .27
1 .90
1 .59
1 .38
1 .78
2.28
1 .81
0.63
1 .86
0.94
1.41
0.64
0.68
2.32
2.67
2.55
2.68
1 .65
1.76
1 . 12
1 .37
2.55
1 . 26
2.31
2.41
1.37
1.11
0.97
1 .37
2. 10
2.52
1 .46
2.39
1 .25
1 .38
1 .69
2.25
1.95
2.00
2.61
1 .76
0.54
1 .97
1.11
1 .92
0.57
0.57
2. 10
2. IB
2.41
2.41
1 .72
1 .43
0.75
1.17
2.16
1 .25
2.41
2.53
-0.08
-0.06
-0.12
-0.11
-0.43
-0.64
-0. 15
-0.52
-0.26
0. 1 1
-0.21
0.67
0.57
0.22
0.33
-0.05
0.37
0.30
0.21
0.12
0.18
0.51
-0.07
-0. 1 1
-0.22
-0.49
-0.14
-0.27
0.07
-0.33
-0.38
-0. 20
-0.39
-0.01
0.09
0.12
-0.06
-0.05
-0.11
-0.08
-0. 19
-0.22
-0. 10
-0.20
-0. 19
0.08
-0. 12
0.35
0.34
0.1 1
0. 13
-0.02
0. 18
0. 16
0.09
0.06
0.17
0.30
-0.11
-0. 18
-0.10
-0.20
-0.05
-0.11
0.04
-0.21
-0.40
-0. 16
-0.17
-0.01
0.04
0.05
(cont 1nued)
-------
Appendix B.
Selected Results of Simulations of Individual Organlc-NOx-AIr Experiments Using the
RAOM-M Mechanism (continued).
Page 16
Experiment
AU03B2R
AU0382B
ST1782R
ST17B2B
ST2982R
OC0682B
JL0283R
JLOBB3R
JL15B3R
JL15B3B
OC0483R
Group Average
S. Dev.
Avg. Abs. Value
S. Dev.
29. SAPRC Synthetic
ITC781
ITC784
ITC785
ITC805
ITC795
ITC796
ITC799
ITC801
Group Average
S. Dev.
Avg. Abs. Value
S. Dev.
30. SAPRC Synthetic
ITC963
ITC965
ITC967
ITC96B
Group Average
S. Dev.
Avg. Abs. Value
S. Dev.
Initial
Concentrat Ions
NOx
(ppm)
0.44
0.16
0.26
0.25
0.39
0.46
0. 18
0.37
0.35
0.35
0.25
0.34
0. 12
Jet
0.51
0.50
0.26
0.52
0.50
0.54
0.51
0.55
0.49
0.09
HC
( ppmC )
2.5
1.2
2.2
2.4
1 .7
2.0
1 .6
1 .7
2.2
2.3
2.6
2.0
0.8
Fuel
43.0
68. 0
45.0
98.0
45.0
97.0
94.0
41.0
68.9
27.3
HC/NOx
5.7
7.3
6.6
9.5
4.4
4.3
9.0
4.6
6.2
6.6
10.3
6.5
3. 1
83.5
177.6
170.7
189.5
89.6
178.9
184. 1
75. 1
143.7
50. B
Ave
Temp
(degK)
302.6
302.6
300. 1
300. 1
295.7
300.2
304.0
299.5
306.2
306.2
299.7
301 .2
3.3
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
0.0
Ave
H20
(ppm)
27400.
27400,
27400.
27400.
24800.
16900.
24500.
21200.
25700.
25700.
18800.
23607.
3602.
16600.
16600.
17900.
17200.
17900.
17200.
17200.
17000.
17200.
499.
Maximum Concentration
OZONE
Expt
(ppm)
0.344
0.576
0.562
0.637
0.073
0.450
0.595
0.756
0.866
0.922
0.603
0.567
0.272
0.751
0.746
0.598
0.791
0.761
0.597
0.840
0.881
0.746
0. 102
Calc
(ppm)
0.444
0.49B
0.586
0.612
0.044
0.319
0.658
0.683
0.705
O.B36
0.657
0.607
0.310
0.927
0.921
0.611
0.959
0. 799
0.851
0.993
0.974
0.880
0. 126
Calc
-Expt
(ppm)
0.100
-0.078
0.024
-0.025
-0.029
-0. 131
0.063
-0.073
-0. 161
-0.086
0.055
0.040
0. I 19
0. 102
0.072
0. 176
0. 176
0.013
0. 169
0.03B
0.254
0. 152
0.093
0. 134
0.080
0. 134
0.080
Calc
-Expt
/Avg
0.25
-0. 15
0.04
-0.04
-0.49
-0.34
0. 10
-0.10
-0.20
-0. 10
0,09
0.01
0.34
0. 23
0. 25
0. 21
0.21
0.02
0. 19
0.05
0.35
0. 17
0. 10
0.16
0.11
0. 16
0.11
Average
d( [03] -
Expt
Calc
Initial
(NO] )/dt
Calc
-Expt
(ppb/mln) --
1.26
.06
.63
.72
0.86
.49
.37
.52
2. 10
2.20
1 . 76
1 .68
0.61
2.77
3.93
2.74
3.27
3.05
4.43
4.59
2.98
3.47
0.74
1 .22
0.92
1 . 15
1 .27
0.68
1 .26
.37
.31
.73
.98
.74
1 .55
O.S9
9.61
13.15
8.64
12.47
13.25
17.30
11.31
7.82
1 1 .69
3.06
-0.03
-0.14
-0.47
-0.45
-0. 19
-0.23
0.00
-0.21
-0.36
-0.22
-0.02
-0. 14
0.23
0.21
0. 15
6.84
9.22
5.90
9.21
10. 19
12.86
6.72
4.84
8.22
2.63
8.22
2.63
Calc
-Expt
/Avg
-0.03
-0. 14
-0.34
-0.30
-0.24
-0. 17
0.00
-0. 15
-0. 19
-0.10
-0.01
-0. 10
0. 15
0. 14
0.11
. 10
.08
.04
. 17
.25
1 . 18
0.65
0.90
1 .07
0.14
1 .07
0.14
Jet Exhaust
0.49
0.46
0.26
0.49
0.42
0.11
4.4
5.2
4.4
8.7
5.7
2. 1
9. 1
1 1.3
17.2
17. B
13. B
4.3
303.0
303.0
303.0
303.0
303.0
0.0
19400.
20000.
18300.
19400.
19275.
709.
0.822
O.B63
0.586
0.652
0. 781
0. 131
0.722
0. 728
0.591
0. 793
0. 708
O.OB4
-0. 100
-0. 135
0,005
-0.059
-0.072
0.060
0.075
O.OS6
-0.13
-0. 17
0.01
-0.07
-0.09
0.08
0. 10
0.07
4.00
5. 10
5.79
11 .76
6.66
3.48
4. 16
5.31
6.21
11.21
6.72
3.11
0.16
0.21
0.42
-0.55
0.06
0.42
0.33
0. 18
0.04
0.04
0.07
-0.05
0.03
0.05
0.05
0.01
-------
APPENDIX C
SELECTED RESULTS OF SIMULATIONS OF ALL ORGANIC-NOX EXPERIMENTS
MODELED USING THE RADM-P MECHANISM
373
-------
Appendix C.
Selected Reaulta of Simulation* of Individual Organlc-NOx-Atr Experiment* Using the
RADM-M Mechanlem.
Paga 1
U)
-J
-Cr
Experiment
Initial
Concantrat 1on»
NO* HC HC/NOx
(ppm) (ppmC)
1. Ethene-NOx
EC142
EC 143
EC156
EC285
EC2B6
EC287
ITC926
ITC936
AU0479R
AU0579R
OC0584R
OC11B4R
OC1284R
OC05B4B
Group Average
S. Oev.
Avg. Aba. Value
S. Dav.
2. Propene-NOx
EC121
EC177
EC216
EC217
EC230
EC256
EC257
EC276
EC277
EC278
EC279
ECS 14
EC315
EC316
EC317
ITC693
ITC810
ITC860
ITC925
ITC938
ITC947
ITC960
0.48
0.50
0.50
1.01
0.94
0.53
0.51
0.50
0.23
0.64
0.36
0.35
0.72
0.37
0.55
0.22
0.51
0.46
0.52
0.48
0.52
0.56
0.56
O.S2
0. 1 1
0.49
0.97
0.93
0.94
0.98
0.54
0.49
0.52
0.52
0.54
0.52
0.53
0.50
1 .9
4. 1
4.0
3.9
7.5
8.0
7.9
3.9
0.9
4. 1
3.2
2.9
2.7
1 .8
4. 1
2.3
1 .5
1 .5
1 .5
0.6
1 .9
0.4
0.7
1.6
1 .7
3. 1
3.5
3.2
2.9
3.2
1 .5
3.5
2.6
3.0
2.8
2.8
1 .9
2.8
4. 1
8.1
8.0
3.9
8.0
15. 1
15.6
7.8
3.9
6.4
8.8
8.2
3.7
5.0
7.6
3.8
2.9
3.2
3.0
1 .2
3.7
0.7
1.3
3.2
15.7
6.2
3.5
3.5
3.1
3.3
2.8
7.2
5.4
5.8
5.2
5.3
3.6
5.5
Ave
Tamp
(dagK)
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
298.9
292.4
297.3
297.2
295.6
297.3
300.2
3.6
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
2BB.7
312.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
Ave
H20
(ppm)
26500.
22700.
29000.
20400.
20500.
19700.
21 100.
20300.
32500.
29100.
20000.*
12700.
13600.
20000.*
22007.
561 1 .
26500.
26900.
18100.
21400.
17200.
20200.
20000.
14500.
15700.
16100.
14200.
24000.
10600.
44900.
24200.
20300.
16800.
17100.
16300.
19000.
20300.
20300.
Maximum Concantrat
OZONE
Expt
(ppm)
0.782
1.087
1.105
0.840
1 .081
0.965
0.982
0.940
0.729
1 .294
0.856
0.858
0.495
0.675
0.906
0.204
0.506
0.540
0.564
0. 149
0.344
0.002
0.068
0.388
0.313
0.625
0.679
0.728
0.344
0.955
0.615
0.779
0.782
0.585
0.779
0.729
0.710
0.721
Calc
(ppm)
0.474
0.791
0.751
0.892
1.171
1 .026
0.921
0.839
0.554
1 142
0.970
1.039
0. 185
0.646
O.B14
0.275
0.460
0.471
0.555
0.230
0.329
0.011
0.098
0.447
0.395
0.702
0.706
0.786
0.661
1 .051
0.570
0.718
0.721
0.681
0.728
0.719
0.715
0.712
Calc
-Expt
(ppm)
-0.307
-0.296
-0.354
0.052
0.090
0.061
-0.061
-0. 101
-0. 175
-0.152
0.114
0. 181
-0.310
-0.029
-0.092
0. 179
0. 163
0.111
-0.046
-0.069
-0.009
0.081
-0.015
0.008
0.029
0.059
0.082
0.076
0.027
0.058
0.316
0.096
-0.045
-0.061
-0.061
0.096
-0.051
-0.010
0.005
-0.009
1 on
Calc
-Expt
/Avg
-0.49
-0.32
-0.38
0.06
0.08
0.06
-0.06
-0.11
-0.27
-0.12
0.12
0. 19
-0.91
-0.04
-0. 16
0.30
0.23
0.24
-0.09
-0.14
-0.02
0.43
-0.05
0.35
0. 14
0.23
0.12
0.04
0.08
0.63
0. 10
-0.08
-0.08
-0.08
0. 15
-0.07
-0.01
0.01
-0.01
Avaraga
d( [03] -
Initial
[NO] )/dt
Calc
Expt Calc -Expt
(ppb/m1n) --
3.20
8.50
8.89
5.05
1 1 .76
13.89
6.96
2.72
1 .60
3. 17
2. 16
2.22
1 .58
1 .48
5.23
4.10
7.46
3.89
4. 18
0.80
3.06
0.98
3.35
3.24
8.27
7.72
6.64
7.21
4.33
10.64
4.06
5.07
4.25
3.57
3.72
3.61
3.34
4. 23
2.20
4.51
4.50
4.89
10. 14
13.01
4.92
2.35
0.99
2.27
1.98
2.01
1 . 17
1.33
4.02
3.53
4. 10
4.00
5.69
1 .30
4.78
1 .08
4.20
4.41
8.06
9.84
8.08
10.36
6.43
1 1 .93
4.71
5.72
4.57
4.88
4.14
4.24
4. 15
4.40
-1 .00
-4.00
-4.39
-0. 15
-1 .63
-0.88
-2.05
-0.37
-0.61
-0.91
-0.18
-0.21
-0.41
-0. 15
-1 .21
1.39
1.21
1 .39
-3.36
0. 1 1
1 .51
0.50
1 . 72
0. 10
0.84
1 . 16
-0.21
2. 1 1
1 .43
3. 15
2.09
1 . 29
0.65
0.66
0.32
1.31
0.42
0.63
0.81
0. 17
Calc
-Expt
/Avg
-0.37
-0.61
-0.66
-0.03
-0. 15
-0.07
-0.34
-0. 15
-0.47
-0.33
-0.09
-0. 10
-0.30
-0.11
-0.27
0.20
0.27
0.20
-0.58
0.03
0.31
0.48
0.44
0. 10
0.22
0.30
-0.03
0.24
0.19
0.36
0.39
0.11
0. 15
0.12
0.07
0.31
0.11
0. 16
0.22
0.04
( con t1nu*d)
-------
Appendix C.
Selected Results of Simulations of Individual Organ1c-N0x-A1r Experiments Using the
RADM-M Mechanism (continued).
Page 2
VJ1
Experiment
Initial
Concent rat Ions
NOx
Ave
Temp
Ave
H20
HC HC/NOx
(ppm) (ppmC)
OTC186
OTC191
OTC210
OTC233
OTC236
JA107BR
OC1278B
OC207BR
OC207BB
OC217BR
OC2578B
JN1279R
JN1279B
JN1379R
AU0279R
AU27BOB
ST04B2B
ST1382B
JL17B3R
JL2183R
JL2983B
JL3183R
ST23B3B
OC0484R
OC0484B
OC1 184B
OC1284B
Group Average
S. Dev.
Avg. Abs . Value
S. Dev.
3. 1-Butene-NOx
EC122
EC123
EC 124
ITC927
ITC92B
ITC930
ITC935
ST2383R
ST25B3R
ST25B3B
ST2783R
0.55
0.54
0.57
0.46
0.53
0.46
0.48
0.46
0,46
0.50
0.44
0.50
0.49
0.45
0.22
0.4B
0.23
0.33
0.27
0.22
0.21
0.21
0.36
0.36
0.36
0.36
0.68
0.49
0.18
0.50
0.51
0.99
0.31
0.67
0.32
0.66
0.40
0.46
0.42
0.45
3.6
3.7
2.7
0. 1
3.3
3. 1
1 .4
1 .3
3.5
3.9
1 .3
1 .0
1 .5
2.9
1 .5
1 .9
. 1
. 1
. 1
. 1
. 1
. 1
1 .6
2. 1
1 .0
2. 2
2.0
2. 1
1 .0
0.9
1 .6
1 .7
3.8
3.8
7.2
7.6
1 .5
1 .6
2.9
1 .6
6.6
6.9
4.8
0.2
6.3
6.9
2.9
2.9
7.7
7.9
2.9
2.1
3.0
6.5
7.0
4.0
4.9
3.3
3.9
5.0
5.3
5. 1
4'. 3
5.9
2.9
6.3
2.9
4.6
2.5
1 .7
3.2
1 . 7
12.3
5.7
22.2
11.6
3.7
3.6
6.8
3.6
(degK)
299.8
309.3
304.7
303.6
300.0
265. 1
294.3
290.4
290.4
290.3
288.4
293.5
293.5
294.9
307.2
302.2
300.5
302.6
307.2
308.5
303. 1
305.0
292.5
296.8
296.7
297.2
295.9
299.9
7.5
303.0
303.0
303.0
303.0
303.0
303.0
303.0
292.6
292. 3
292.3
295.5
(ppm)
5000.
5000.
3720.
5000.
6510.
20000.
16300.
9590.
9560.
20000.*
12800.
20000.
20000.*
20000.*
20000.*
2B400.
1 1600.
24100.
29700.
23800.
23300.
19800.
23400.
20000.*
20000.*
12700.
13500.
18134.
7349.
22000.
27200.
24500.
19000.
1B400.
21 100.
19700.
23400.
18800.
18900.
19400.
Maximum Concentration
OZONE
Expt
(ppm)
0.822
0.903
0.972
0.633
O.B4B
0.363
0.461
0.340
0.727
0.670
0.230
0.382
0.673
0.974
0.788
1 .044
0.65B
0.731
O.B4B
0.804
0.697
0.719
0.405
0.645
0.446
0.674
0.432
0.608
0. 238
0. 227
0.506
0. 247
0.646
0.022
0.717
0.872
0.206
0.266
0.594
0.285
Calc
(ppm)
0.939
1 . 155
1.006
0.934
1.015
0.604
0.413
0.403
0.948
0.913
0. 269
0.306
0.515
0.874
0.632
1 . 183
0.614
0.687
0. 720
0.731
0.739
0.687
0.666
0.759
0.474
0.772
0.471
0.651
0. 254
0.081
0.280
0. 132
0. 728
0.045
0.827
0.972
0.451
0.449
0.783
0.452
Calc
-Expt
(ppm)
0.117
0.252
0.034
0.301
0. 167
0.240
-0.049
0.063
0.221
0.243
0.039
-0.076
-0. 158
-0.099
-0. 156
0. 139
-0.044
-0.045
-0.127
-0.073
0.042
-0.032
0.261
0.114
0.029
0.098
0.039
0.043
0.117
0.093
O.OB1
-0. 147
-0.226
-0.115
0.082
0.023
0.110
0. 100
0.245
0. 182
0. 190
0. 167
Calc
-Expt
/Avg
0. 13
0. 24
0.03
0.38
0.18
0.50
-0. 11
0. 17
0.26
0.31
0. 15
-0.22
-0.27
-0. 11
-0.22
0. 12
-0.07
-0.06
-0.16
-0.09
0.06
-0.05
0.49
0. 16
0.06
0. 14
0.09
O.OB
0.20
0. 16
0. 14
-0.95
-0.58
-0.61
0. 12
0.14
0.11
0.74
0.51
0.28
0.45
Average
d( 103] -
Expt
Calc
Initial
(NO) )/dt
Calc
-Expt
(ppb/mln) --
5. 16
12. 18
6.70
3.69
6.91
0.46
1 .40
1.23
2.63
2.53
1 .03
0.85
1 .24
2.69
2.41
3.26
.46
.69
2.10
.76
.66
.81
1 .23
2.07
1 .25
2.38
1 .67
3.62
2.60
2. 29
4. 16
1 .99
3.24
1 .36
7.91
5.39
0.96
1.16
1 .78
1 . 19
7.81
13.94
6.36
5.50
8.41
0.46
1 .45
1.38
3.53
3.44
1 . 15
0.77
1 . 10
2.83
2.06
2.84
1.59
1.61
1 . 75
1 .87
1 . 74
1.67
1 .67
2. 15
1 .21
2.45
1 .64
4. 15
3.08
1 .40
2.30
1.71
4.54
1 .69
11.12
7.16
1 . 15
1 .31
2. 19
1 .31
2.65
1 .77
-0.34
1.81
1 .50
0.02
0.05
0. 15
0.70
0.91
0. 12
-0.09
-0. 14
0.14
-0.35
-0.42
0. 13
-0.08
-0.35
0.10
0.08
-0. 14
0.44
O.OB
-0.04
0.06
-0.03
0.53
1 .01
0.76
0.85
-0.89
-1 .86
-0.27
1.31
0.34
3.21
1 .77
0.20
0.15
0.41
0. 12
Calc
-Expt
/Avg
0.41
0.14
-0.05
0.39
0.20
0.04
0.03
0.12
0.22
0.31
0.11
-0.11
-0.12
0.05
-0.16
-0. 14
0.08
-0.05
-0.18
0.06
0.05
-0.08
0.31
0.04
-0.04
0.03
-0.02
0.11
0. 19
0.17
0.14
-0.48
-0.58
-0. 15
0.34
0.22
0.34
0. 28
0. 19
0. 12
0.21
0. 10
(cont1nued)
-------
u>
-4
en
Appendix C. Selected Reault* of Simulation* of Individual Organlc-NOx-AIr Experiment* Using the
RADM-M Mechanism (continued).
Page 3
Experiment
Initial
Concent ret Ions
NOx HC HC/NOx
(ppm) (ppmC)
Group Average
S. Oev.
Avg. Abe. Value
S. Dev.
0.52
0.20
3. 1
2.3
6.9
6.2
Ave
Temp
(degK)
299.4
f. 0
Ave
H20
(ppm)
21 127.
2656.
Maximum Concent rat
OZONE
Expt
(ppm)
0
0
.417
.263
Calc
(ppm)
0
0
.473
.321
Celc
-Expt
(ppm)
0.055
0. 154
0. 144
0.066
1 on
Average
d( [03] -
Calc
-Expt
/Avg
0
0
0
0
.02
.55
.45
.29
Initial
[NO] )/dt
Calc
Expt Calc -Expt
(ppb/mln) --
2.
2.
86
IB
3.26
3. IB
0.41
1 . 34
0.96
0.99
Calc
-Expt
/Avg
0.05
0.32
0.27
0.15
4. trans-2-Butane-NOx
EC146
EC147
EC157
ST2783B
Croup Average
S. Dev.
Avg. Abe. Value
S. Dev.
0.51
0.98
0.53
0.43
0.61
0.25
0.9
1 .7
0.9
2.0
1 .4
0.6
KB
1 .7
1 .7
4. 7
2.5
1.5
303.0
303.0
303.0
295.5
301 . 1
3.7
26900.
22300.
27500.
19400.
24025.
3B60.
0
0
0
0
0
0
.247
. 154
.205
.523
.282
.165
0
0
0
0
0
0
. 109
.083
.091
.543
.207
.224
-0. 13B
-0.071
-0.114
0.019
-0.076
0.069
0.085
0.052
-0
-0
-0
0
-0
0
0
0
.77
.60
.77
.04
.53
.38
.54
.35
5.
9.
5.
3.
6.
2.
87
63
96
00
17
80
3.85
7.39
3.54
3.16
4.48
1.96
-2.02
-2.45
-2.42
0. 15
-1.68
1 .24
1 .76
1 .09
-0.42
-0.28
-0.51
0.05
-0.29
0.24
0.31
0.20
5. leobutene-NOx
ITC694 0.51
4.6
9. 1
303.0
20000.
0.900 0.731 -0.169 -0.21
6.84
6.04 -2.80 -0.3B
6. 1-Hexene-NOx
ITC929
ITC931
ITC934
ITC937
Group Average
S. Dev.
Avg. Abs . Value
S. Dev.
7. Isoprene-NOx
EC520
EC522
EC524
EC525
0.51
0.49
1 .00
0.99
0.75
0.28
0.49
0.92
0.96
0.54
5. 1
10.3
9.7
0. 1
6.3
4.8
2.2
2.3
4.7
4.5
10.0
21.1
9.7
0. 1
10.2
8.6
4.5
2.4
4.9
6.3
303
303
303
303
303
0
303
303
303
303
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
19000.
20100.
20300.
20300.
19925.
624.
13000.
13000.
13000.
13000.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
298
606
428
007
335
252
503
276
759
691
0.670
O.B19
O.B95
0.060
0.61 1
0.379
0.300
0. 1 18
0.430
0.630
0
0
0
0
0
0
0
0
-0
-0
-0
-0
.372
.213
.467
.053
. 276
. 182
.276
. 1B2
.203
. 158
.330
.061
0.
0.
0.
0.
0.
0.
0.
-0.
-0.
-0.
-0.
77
30
71
59
26
59
26
50
80
55
09
1
2
1
0
1
1
5
6
15
10
.27
.82
.84
.33
.57
.04
.30
.00
.57
. 13
3.33
10.67
5.00
0.57
4.89
4.26
7.56
6.75
15.93
22. 16
2.06
7.85
3. 16
0. 24
3.33
3. 25
3.33
3.25
2.26
0.76
0.37
12.04
0.89
1 . 16
0.93
0.52
0.88
0.26
0.88
0.26
0.35
0. 12
0.02
0.75
(cont Intied)
-------
Appendix C. Selected Results of Simulations of Individual Organ1c-N0x-A1r Experiments Using
RADM-M Mechanism (continued).
the
Page 4
to
Experiment
EC527
ITC81 1
ITC812
JL1680R
JL1680B
JL17BOR
JL1780B
JL2381R
ST0981R
Group Average
S. Dev.
Avg. Abs. Value
S. Oev.
8. a-P1nene-NOx
JL15BOR
JL15BOB
JL2580R
JL2580B
Group Average
S. Oev.
Avg. Abs. Value
S. Oev.
9. Ethane-NOx
ITC999
10. n-Butane-NOx
EC130
EC133
EC 134
EC137
EC162
EC163
EC 168
EC1 78
EC304
Initial
Concent rat
NOx
(ppm)
0.50
0.50
0.52
0. 18
0. 18
0.46
0.47
0.43
0.17
0.49
0.25
0. 18
0. 19
0.25
0.25
0.22
0.04
0.09
0. 10
0.50
0.51
0.50
0.5t
0.49
0.49
0.10
0.47
HC
(ppmC)
2.2
3.4
1 .8
4.6
6.4
1 .0
2.6
1 .4
1 .0
2.9
1 .7
1 . 1
2.6
1 .0
0.3
1.3
1 .0
45.4
17.6
8.6
8. 3
8. 7
8.2
9.0
8.0
7.8
17.1
1ons
HC/NOx
4.5
6.B
3.4
26.0
36.6
2.1
5.5
3.4
6. 1
8.8
10.4
5.8
13.8
4.0
1 .4
6.3
5.3
534. 1
179.3
17.1
16.3
17.3
16.3
18.3
16.2
79.6
36.7
Ave
Temp
(degK)
303.0
303.0
303.0
305.0
301 .9
305.6
302.8
306. 1
297.5
303. 1
2. 1
302.4
299.6
302.4
300. 1
301 . 1
1 .5
303.0
303.0
303.0
303.0
303 .0
303 .0
303.0
303.0
303.0
303.0
Ave
H20
(ppm)
13000.
17300.
17300.
30200.
26800.
27000.
25000.
20000.*
20000.
19123.
6284.
30300.
29200.
31700.
26300.
29375.
2291 .
20500.
26200.
27800.
20000.
20000.
27300.
28500.
20700.
20900.
26800.
Maximum Concentration
OZONE
Expt
(ppm)
0.547
0.919
0.76B
0.652
0.837
0.806
1 .298
0.750
0.506
0.716
0. 246
0.201
0.470
0.377
0.334
0.346
0.112
0.243
0.459
0.249
0.034
0.042
0.112
0.454
0.655
0.384
0.362
Calc
(ppm)
0.317
0.578
0.253
0.712
0.600
0.420
0.835
0.691
0.457
0.488
0.20B
0. 297
0.421
0.327
0.211
0.314
0.087
0.203
0.444
0.045
0.038
0.044
0.050
0. 152
0.315
0.421
0.434
Calc
-Expt
(ppm)
-0. 230
-0.341
-0.515
0.060
-0. 237
-0.386
-0.463
-0.058
-0.049
-0. 228
0. 174
0.238
0. 160
0.095
-0.049
-0.050
-0. 123
-0.032
0.091
0.079
0.036
-0.040
-0.015
-0. 204
0.004
0.002
-0.062
-0.302
-0.340
0.037
0.073
Calc
-Expt
/Avg
-0.53
-0.46
-1.01
0.09
-0.33
-0.63
-0.43
-0.08
-0.10
-0.42
0.31
0.43
0.29
0.38
-0.11
-0. 14
-0.45
-O.OB
0.34
0.27
0. 17
-0.18
-0.03
-1 .38
-0.76
-0.99
-0. 70
0.09
0. 18
Average
d( [03] -
In1t1a
[NO] )
Ca Ic
Expt Calc -Expt
(ppb/m1n) --
5.97
10.00
4.86
3.18
2.88
1 .96
4.20
2. 13
1 .46
5.66
4.06
0.61
1 .59
0.84
0.84
0.97
0.43
1 .32
4.41
2.42
0.94
1 .02
1 .73
3.31
2.03
1 .61
2.09
7.95
B.96
2.43
3.00
2.43
1 .36
2.87
2.01
1 .45
6.53
6.29
0.67
1 .96
0.83
0.68
1 .03
0.62
1 .42
2.49
1 .07
1 .30
1 .27
1.21
1 .71
1.14
1 .68
2.31
1 .97
-1 .04
-2.43
-0. IB
-0.44
-0.60
-1 .33
-0. 12
-0.01
0.86
3.58
1 .81
3.18
0.06
0.37
-0.02
-0. 16
0.06
0. 22
0. 15
0. 16
0. 10
-1 .93
-1 .35
0.36
0.25
-0.51
-1 .60
-0.89
0.07
0.22
1
/dt
Cal c
-Expt
/Avg
0.28
-0.11
-0.67
-0.06
-0.17
-0.36
-0.38
-0.06
0.00
-0.02
0.36
0.26
0.24
0.10
0.21
-0.02
-0.21
0. 02
0. 18
0. 13
0.09
0.07
-0.56
-0.77
0.32
0.22
-0.35
-0.64
-0.56
0.04
0. 10
(cont1nued)
-------
Appendix C. Selected Results of Simulations of Individual Organ1c-N0x-A1r Experiments Using the
RADM-M Mechanism (continued).
Page
UJ
-J
CD
Experiment
EC305
EC306
EC307
EC30B
EC309
ITC507
ITC533
ITC770
ITC939
ITC94S
OTC211
JL2178R
JL2176B
JL227BR
JL227BB
ST1B7BB
OC0979R
OC1B79B
Group Average
S. Dev.
Avg. Abs. Value
S. Dev.
1 1 . C4+ Branched
EC 165
EC 169
EC171
OC1B79R
OC2079B
AU1983R
AU1983B
Group Average
S. Oev.
Avg. Abs. Value
S. Dev.
Initial
Concentrat Ions
NOx
(ppm)
0. 10
0.19
0. 10
0.48
0.47
0.09
0. 12
0.52
0.51
0.26
0.55
0.24
0.24
0.55
0.55
0.21
0.21
0.20
0.34
0.18
HC
(ppmC)
15.7
25.8
25.8
16.2
17.2
15.2
11.9
37.9
14.8
10.0
42.8
7.2
15.4
7.9
17.5
21.2
14.6
14.3
15.7
8.8
HC/NOx
159.7
138.2
252.9
33.6
36.3
165.0
99.8
72.8
28.9
38.2
77.5
29.9
63.9
14.4
31 .5
103.1
71 .0
71.8
69.8
61 .6
Ave
Temp
(degK)
303.0
302.6
303.0
288.7
310.9
303.0
303.0
303.0
303.0
303.0
300. 1
302.9
302.9
305.2
305.2
298.3
297.2
295.0
302. 1
3.6
Ave
H20
(ppm)
25400.
24200.
29700.
8750.
12400.
20300.
20800.
16800.
20300.
20300.
3530.
20000.
20000.
20000.
20000.
20000.
15300.
28100.
20892.
6033.
Maximum Concentrat
OZONE
Expt
(ppm)
0.398
0.535
0.420
0.047
0.545
0. 149
0. 165
0.042
0.017
0.054
0.008
0.763
0.986
0. 166
0.788
0.1B5
0. 191
0.208
0.312
0.268
Calc
(ppm)
0.531
0.709
0.588
0.210
0.616
0.356
0.327
0.050
0.025
0. 141
0. 190
0.594
0.868
0. 138
0.41B
0. 1 14
0.272
0. 158
0.305
0.235
Calc
-Expt
(ppm)
0.133
0. 174
0. 166
0. 164
0.071
0.207
0.162
O.OOB
0.008
O.OB7
0. 182
-0. 168
-0. 1 19
-0.028
-0.369
-0.072
0.081
-0.050
-0.006
0. 160
0. 122
0. 102
Ion
Calc
-Expt
/Avg
0.29
0.28
0.33
0. 12
0.62
0.66
0.89
-0.25
-0.13
-0. IB
-0.61
-0.48
0.35
-0.27
-0.08
0.59
0.47
0.36
Average
d( (031 -
Expt
Calc
Initial
INO] )/dt
Calc
-Expt
(ppb/m1n) --
2.39
2.38
2.61
1 .04
2.00
0.69
0.61
1 .56
0.36
0.53
0.56
1.17
1 .64
0.90
1.55
0.51
0.60
0.60
1 .53
0.98
3.43
4.20
5.89
2.62
2.48
1.54
1 .36
1 .98
0.65
1. 10
1 .40
1 .06
1 .46
0.75
1.28
0.43
0.59
0.52
1 .74
1 .21
1 .04
1 .82
3.28
1 .59
0.4B
0.85
0.75
0.42
0.28
0.57
O.B3
-0. 1 1
-0. 18
-0. 15
-0.27
-0.08
0.00
-0.09
0.21
1 .05
0.74
0.77
Calc
-Expt
/Avg
0.36
0.55
0.77
0.87
0.22
0.77
0.76
0.24
0.56
0.70
O.B5
-0. 10
-0. 12
-0. IB
-0. 19
-0.16
0.00
-0.16
0.13
0,49
0.41
0.28
Alksne-NOx
0. 10
0. 19
0. 10
0.20
0.22
0.38
0.37
0.22
0.11
11.3
4.5
3.5
16.4
12.6
4.7
4. 1
8.2
5.2
114.3
23.5
35.7
81 .9
56.5
12.6
10.9
47.9
38.7
303.0
303.0
303.0
295.0
295.3
307. 1
307. 1
301.9
5.0
18600.
32900.
19000.
28100.
27800.
21300.
21300.
24143.
5463.
0.488
0.493
0.403
0.236
0.217
0.088
0.057
0.283
0.181
0.479
0.326
0.306
0. 130
0.058
0. 197
0.078
0.225
0. 153
-0.009
-0. 168
-0.097
-0. 106
-0. 158
0. 109
0.020
-0.05B
0. 102
0.095
0.061
-0.02
-0.41
-0.27
-0.58
-1.15
0.77
0.30
-0. 19
0.62
0.50
0.37
1 .77
1 .00
1 .30
0.64
0.66
0.61
0.53
0.93
0.46
1 .65
0.70
0.79
0.50
0.39
0.65
0.55
0.75
0.42
-0.12
-0.30
-0.51
-0. 13
-0.27
0.05
0.02
-0. 18
0.20
0.20
0. 17
-0.07
-0.35
-0.49
-0.23
-0.52
0.07
0.03
-0.22
0.24
0.25
0.20
12. C5 + n-Alkane-NOx
EC 135
OC0979B
EC131
ITC559
0. 10
0.21
0. 10
0. 19
20.4
15. 1
24.6
279.4
212.7
73.3
251 . 1
1441 . 1
303.0
297. 2
303.0
303.0
25000.
15200.
25000.
20300.
0.435
0. 184
0.393
0.377
0.552
0.240
0.530
0.654
0.117
0.056
0. 137
0.277
0. 24
0.26
0.30
0.54
2.92
0.59
1.92
1.79
3. 16
0.55
2.95
2.46
0.25
-0.05
1 .03
0.67
o.oe
-0.08
0.42
0.32
(contInued)
-------
Appendix C. Selected Results of Simulations of Individual Organ1c-N0x-A1r Experiments Using the
RADM-M Mechanism (continued).
Page 6
Experiment
ITC538
ITC540
ITC552
ITC761
ITC762
ITC763
ITC797
ST1879R
EC 155
ITC1001
Group Average
S. Dev.
Avg. Abs. Value
S. Dev.
Initial
Concentrat Ions
NOx
(ppm)
0.11
0.11
0. 13
0.52
0.27
0.26
0.52
0.21
0. 10
0.11
0.21
0. 14
HC HC/NOx
(ppmC)
60.3 529.0
274.8 2421.3
428.8 3276,4
75.2 145.9
74.7 280.4
7.7 27.7
7.3 14.0
6.3 30.4
37.3 385.1
2.4 22.4
93.9 650.9
133.5 1015.3
Ave
Temp
(degK)
303.0
303.0
303.0
303.0
303.0
303.0
303.0
296.3
303.0
303.0
302.3
1 .9
Ave
H20
(ppm)
16900.
20300.
20300.
17500.
17000.
17100.
12200.
20000.*
26200.
17500.
19464.
3919.
Maximum Concentrat
OZONE
Expt
(ppm)
0. 150
0.360
0.315
0.030
0. 105
0.041
0.004
0. 122
0.264
0.036
0.201
0. 153
Calc
(ppm)
0.300
0.370
0.344
0.033
0. 176
0.066
0.010
0.300
0.521
0.073
0.298
0.211
Calc
-Expt
(ppm)
0. 150
0.010
0.029
0.003
0.071
0.025
0.006
0. 179
0. 257
0.037
0.097
0.092
0.097
0.092
Ion
Calc
-Expt
/Avg
0.67
0.03
0.09
0.50
0.85
0.65
0.41
0.27
0.41
0.27
Average
d( [03] -
Inlt lal
[NO] )/dt
r« i r
Expt Calc -Expt
(ppb/m1n) --
0.74
1 .85
1 . 19
1 .08
0.83
0.68
0.64
0.40
1 .33
0. 15
1.15
0.75
1 .03
1 . 10
1 .07
1 . 23
1.51
1 .40
0.93
0.44
1 .65
0.42
1 .42
0.87
0. 29
-0.76
-0.12
0. 16
0. 68
0. 72
0. 29
0.04
0.32
0.27
0. 27
0.44
0.40
0.31
Ca Ic
-Expt
/Avg
0. 33
-0.51
-0. 1 1
0. 14
0. 58
0.69
0. 37
0.10
0. 22
0.95
0. 25
0. 36
0. 35
0.26
13. Methy Icycl ohexane-NOx
ITC765
ITC766
ITC767
ITC800
Group Average
S. Dev.
Avg. Abs. Value
S. Dev.
14. Benzene-NOx
ITC560
ITC561
ITC562
ITC698
ITC710
ITC831
Group Average
S. Dev.
Avg. Abs. Value
S. Dev.
0.53
0.26
0.55
0.54
0.47
0. 14
0. 12
0. 1 1
0.56
0.50
0.55
1 .01
0.47
0.33
0.0 0.1
0.0 0.1
0.1 0.1
0.0 0.1
0.0 0.1
0.0 0.0
332.3 2874.4
79. 1 694.2
83.8 149.7
83.5 167.4
83.6 151.0
12.2 12.1
112.4 674.8
111.3 1 103.2
303.0
303.0
303.0
303.0
303.0
0.0
303 .0
303.0
303.0
303. 0
303.0
303.0
303.0
0.0
17900.
17100.
17900.
17500.
17600.
383.
17100.
17600.
18800.
21900.
21000.
18300.
19117.
1920.
0.022
0. 121
0.041
0.015
0.050
0.323
0.273
0.412
0.374
0. 367
0.021
0. 295
0. 142
O.OM
0. 159
0.029
0.010
0.052
0.329
0.294
0.528
0.495
0.512
0.010
0.361
0. 199
-0.01 1
0.036
-0.012
-0.004
0.002
0.016
0.006
0.021
0. 1 16
0.121
0. 145
-0.011
0.066
0.068
0.070
0.064
0.27
0.27
0.27
0.02
0.08
0.25
0.28
0.33
0.19
0.14
0.19
0. 14
0.86
0.87
1 .09
0.86
0.92
0.12
7.01
4.75
2.85
2.87
2.70
0.17
3.39
2.30
0.66
1 .34
1 .05
0.72
0.95
0.32
15.95
10.53
10.45
10.39
9.92
0.32
9.59
5.08
-0.20
0.48
-0.04
-0.13
0.03
0.31
0.21
0. 19
8 .94
5.78
7 .60
7.52
7. 22
0.15
6. 20
3.13
6.20
3. 13
-0.26
0.43
-0.03
-0. 17
-0.01
0.31
0.22
0. 17
0. 78
0.76
1 . 14
1 . 13
1 . 14
0.61
0.93
0.24
0.93
0.24
(cont1nued)
-------
Appendix C. Selected Result* of Simulations, of Individual Organ1c-N0x-A1r Experiment* Using the
RADM-M Mechanlem (continued).
Page 7
u>
00
O
Experiment
Initial
Concentrat 1on»
NOx HC
(ppm) (ppmC)
IS. Tolutne-NOx
EC264
EC265
EC266
EC269
EC270
EC271
EC272
EC273
EC327
EC336
EC337
EC339
EC340
ITC699
ITC828
JL3080R
AU2780R
AU27B2B
OC2782R
AU0183R
Group Average
S. Dev.
Avg. Abs. Value
S. Dev.
16. Xylene-NOx
EC343
EC344
EC345
EC346
ITC702
ITC827
JL3080B
AU2782R
OC2782B
AU0183B
Group Average
S. Dev.
Avg. Abs. Value
S. Dev.
0.4B
0.48
0.49
0.47
0.46
0.21
0.48
0. 11
0.45
0.44
0.45
0.44
0.43
0.51
1 .02
0. 18
0.4S
0.43
0.39
0.39
0.44
0. 18
0.28
0.67
0.28
0.26
0.52
1 .07
0.1B
0.43
0.39
0.37
0.45
0.26
6. 1
7.5
6.4
4.0
4.2
B.O
4. 1
4. 1
4.0
7.2
7.9
5.0
4. 1
10.5
3.0
3.9
2.3
3.0
4.5
4.6
5.4
2.3
4.2
4.0
3.7
3.9
4.0
1 .2
2.2
2.0
2.8
2.7
3. 1
1 .0
HC/NOx
17.0
15.6
17.0
8.4
9.0
37.4
6.5
37.2
6.9
16.3
17.7
11.3
9.5
20.6
3.0
21.3
4.6
7.0
11.7
1 1 .8
14.7
9.2
14.9
5.9
13.3
14.8
7.8
1 . 1
12.4
4.6
7.0
7.2
8.9
4.7
Ave
Temp
(degK)
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
304.5
302.2
301 .7
289. 8
306. 1
302.5
3. 1
303.0
303.0
303.0
303.0
303.0
303.0
304.5
301 . 7
289.8
306.0
302.0
4.4
Ave
H2O
(ppm)
18900.
21900.
19400.
19100.
20000.
20500.
19400.
21200.
22400.
26900.
27000.
26800.
25900.
20300.
17200.
35200.
32000.
24700.
19500.
23700.
23100.
4692.
29500.
28900.
25200.
ib900.
20300.
14400.
28000.
24700.
19500.
23700.
240)0.
4737.
Maximum Concentration
OZONE
Expt
(ppm)
0.419
0.393
0.405
0.318
0.369
0.296
0.410
0.215
0.376
0.396
0.325
0.225
0.344
0.465
0.021
0.273
0.736
0. 1 16
0. 123
0.458
0.335
0. 154
0.283
0.589
0.396
0.384
0.627
0.021
0.555
0.491
0.396
0.688
0.443
0.195
Calc
(ppm)
0.397
0.395
0.401
0.31 1
0.385
0.318
0.306
0.245
0.376
0.434
0.389
0.337
0.356
0.441
0.008
0.441
0.879
0.295
0.346
0.561
0.382
0.159
0.380
0.552
0.369
0.376
0.509
0.013
0.535
0.606
0.453
0.784
0.460
0.200
Calc
-Expt
(ppm)
-0.022
0.001
-0.004
-0.006
0.016
0.023
-0. 105
0.030
0.000
0.039
0.064
0.112
0.012
-0.043
-0.013
0. 167
0. 143
0. 180
0.223
0. 123
0.047
0.084
0.066
0.069
0.096
-0.037
-0.007
-0.007
-0. 1 18
-0.007
-0.019
0. 1 17
0.057
0.096
0.017
0.073
0.056
0.047
Calc
-Expt
/Avg
-0.05
0.00
-0.01
-0.02
0.04
0.07
-0.29
0. 13
0.00
0.09
0. 18
0.40
0.03
-0.09
0.47
0. 18
0.87
0.95
0.24
0. 17
0.31
0.22
0.28
0. 29
-0.06
-0.02
-0.02
-0.21
-0.04
0.21
0. 13
0. 13
0.05
0. 16
0.12
0. 10
Average
d( [03) -
Initial
(NO) )/dt
Calc
Expt Calc -Expt
(ppb/m1n) --
4.46
3.56
4.62
2.55
3.72
6.56
3.69
5.90
2.49
6.06
2.55
1 .53
2.50
4.72
0.49
1 .20
1 .76
0.63
0.66
1.75
3.07
1 .87
8. 18
10.72
1 1 .35
7.65
7.79
1 . 12
1 .97
1 .31
1 .43
2.39
5.39
4. 14
5. 24
5.07
5.43
2.67
4.12
5.58
2.65
3.46
2.76
6.76
4.11
2.99
2.59
4.88
1.21
1 .04
1 .53
0.67
0.81
1 .72
3.26
1 .61
10.60
9.93
10.59
10.49
6.09
2.08
1 .93
1 .64
1 .97
2.75
6.01
4. 22
0.79
1 .52
0.81
0.12
0.39
-0 .99
-1 .04
-2.44
0.27
0.70
1 .56
1 .46
0.09
0. 15
0.72
-0.16
-0.23
0.03
0. 15
-0.03
0. 19
0.94
0.68
0.66
2.41
-0.79
-0.76
2.84
0.30
0.96
-0.04
0.33
0.54
0.36
0.61
1 . 19
0.93
0.94
Calc
-Expt
/Avg
0. 16
0.35
0. 16
0.05
0. 10
-0,16
-0.33
-0.52
0.10
0.11
0.47
0.64
0.03
0.03
0.85
-0.14
-0.14
0.05
0.20
-0.02
0.10
0.31
0.23
0.23
0. 26
-0.08
-0.07
0.31
0.04
0.60
-0.02
0.22
0 .32
0.14
0.17
0.21
0.21
0. 16
(cont1nued)
-------
Appendix C. Selected Results of Simulations of Individual Organlc-NOx-A1r Experiments Using the
RADM-M Mechanism (continued).
Page B
Lo
OO
Experiment
17. Mes1tylene-N0x
EC900
EC901
EC903
ITC703
ITC706
ITC709
ITC742
ITCB26
Group Average
S. Dev.
Avg. Abs. Value
S. Dev.
18. Tetralln-NOx
ITC739
ITC747
ITC748
ITC750
ITC832
Group Average
S. Dev.
Avg. Abs. Value
S. Dev.
19. Naphthalane-NOx
ITC751
ITC755
ITC756
ITC798
ITC802
Group Average
S. Dev.
Av0. Abs. Value
S. Oev.
In1 1 lal
Concentrat
NOx
(ppm)
0.53
0.51
1 .00
0.50
0.49
0.99
0.48
0.90
0.68
0.24
0.52
0.50
0.22
0.53
1 .00
0.55
0.28
0.52
0.24
0.26
0.53
0.53
0.42
0. 15
HC
(ppmC)
5.4
2.7
4.7
5.3
2.7
4.7
4.6
0.8
3.9
1 .6
2.4
93.2
84.0
44.5
39.4
52.7
36.7
7.5
14. 1
27.4
19.4
8.4
15.4
8.3
Ions
HC/NOx
10.2
5.2
4.7
10.6
5.4
4.7
9.7
0.9
6.4
3.4
4.6
187.4
385. 1
84.6
39.5
140.2
153.2
14.3
58.3
106.6
36.6
15.9
46.4
3B.2
Ave
Temp
(degK)
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
0.0
303.0
303.0
303.0
303.0
303.0
303.0
0.0
303.0
303.0
303.0
303.0
303.0
303.0
0.0
Ave
H20
(ppm)
20000.*
20000.*
20000.*
21000.
21000.
21000.
18100.
17200.
19786.
1419.
17900.
17100.
17500.
17100.
16800.
17280.
427.
15100.
17500.
17200.
16400.
16800.
16600.
935.
Maximum Concentration
OZONE
Expt
(ppm)
0.381
0.384
0.502
0.707
0.641
0.779
0.773
0.022
0.524
0. 258
0.002
0.508
0.370
0.482
0.073
0.287
0.235
0.113
0.259
0.282
0.204
0. 124
0. 196
0.077
Calc
(ppm)
0.434
0.341
0.424
0.496
0.510
0.560
0.502
0.009
0.410
0. 175
0.358
0.720
0.478
0.697
0.336
0.516
0. 182
0.548
0.450
0.486
0.624
0.551
0.532
0.067
Calc
-Expt
(ppm)
0.053
-0.042
-0.078
-0.211
-0. 130
-0.219
-0. 271
-0.013
-0.114
0.113
0. 127
0.096
0.356
0.211
0. 108
0.215
0.263
0.231
0.090
0.231
0.090
0.436
0.191
0.203
0.421
0.427
0.335
0. 127
0.335
0. 127
Calc
-Expt
/Avg
0.13
-0. 12
-0. 17
-0.35
-0.23
-0.33
-0.42
-0.21
0. 19
0.25
0. 12
0.34
0. 26
0. 36
1 .29
0.56
0.49
0.56
0.49
1 .32
0.54
0.53
1.02
1 .27
0.93
0. 38
0.93
0.38
d(
Expt
3.85
8.88
14.96
14.59
7.20
1 1 . 74
13.14
1 .68
9.50
4.97
0.63
2.73
2.60
2. 10
1 .38
1 .89
0.88
1 .08
1 .57
2.20
1.66
1 .42
1 .59
0.41
Average
[03] -
Initial
[NO] )/dt
Calc
Calc -Expt
(ppb/m1n) --
3.72
5.88
9.40
4.48
7.21
7.77
1 .29
5.68
2. 74
2.64
54.42
31 .58
45.57
1 .00
27.04
24.42
1 1 . 27
21.78
29. 17
29.50
12.55
20.85
8. 74
-5. 17
-9.08
-5. 19
-2.72
-4.53
-5.37
-0.39
-4.63
2 . 66
4.63
2.66
2.02
51 .69
28.98
43.46
-0.37
25. 16
23.67
25.30
23.47
10.18
20.22
26.97
27.84
11.13
19.27
8 . 40
19.27
8.40
Calc
-Expt
/Avg
-0.82
-0.87
-0.43
-0 .47
-0.48
-0.51
-0.26
-0.55
0 . 22
0.55
0.22
1 .23
1.81
1 .70
1 .82
-0.31
1 . 25
0.91
1 .37
0.64
1 .65
1 . 73
1 . 72
1 . 79
1 .59
1 . 70
0 . 08
1 . 70
0.08
(cont1nued)
-------
Appendix C. Selected Results of Simulations of Individual Organ1c-N0x-A1r Experiments Using the
RADM-M Mechanism (continued).
Page 9
10
OD
Experiment
Initial
Concentrations
NOx
(ppm)
HC HC/NOx
(ppmC)
Ave
Temp
(degK)
Ave
H20
(ppm)
Maximum Concentration
OZONE
Expt
(ppm)
Catc
(ppm)
Calc
-Expt
(ppm)
Calc
-Expt
/Avg
Average
d( [03] -
Initial
[NO) )/dt
Calc
Expt Calc -Expt
(ppb/ml n) , --
Calc
-Expt
/Avg
20. 2,3-D1mathy1naphtha1ene-NOx
ITC775
ITC771
ITC806
ITC774
Group Average
S. Oev.
Avg . Abs . Value
S. Dev.
21 . Simple Mixture
EC 144
EC 145
EC160
EC149
EC 150
EC151
EC152
EC153
EC161
OC1278R
OC257BR
AU0180R
AU14BOR
EC 166
EC172
ST0682R
ST0682B
EC 106
EC 11 3
EC114
EC1 15
EC1 16
EC335
EC329
EC330
EC334
EC338
EC328
JL1581R
JL1881R
ST2481R
AU2781R
0.29
0.26
0.33
0.56
0.36
0. 14
Runs
0.51
0.99
0.99
0.99
1 .00
2.06
0.50
0.97
0.51
0.48
0.44
0.56
0.47
0. 10
0. 10
0.46
0.45
0.50
0. It
1 .00
0.51
0.49
0.44
0.45
0.29
0.45
0.45
0.45
0.27
0.26
0.23
0.23
1.7
4.8
5.9
4.0
4. 1
1 .8
4.7
3.4
3.2
2.0
3.5
5.2
3.7
6.6
3.2
1 .4
1 .4
0.5
1 .4
9.2
2.8
2.8
2.9
9.2
9.5
17.3
12.7
IB. 6
7. 7
4.2
4.3
8. 1
15.0
12. 1
2.3
2. 1
1 .6
2. 1
5.8
18.0
17.7
7. 1
12.2
6.6
9.3
3.4
3.3
2.0
3.5
2.5
7.3
6.8
6.4
3.0
3.1
0.8
3.0
92.0
28.9
6.2
6.5
18.3
85.0
17.3
25.2
37.6
17.4
9.2
14.6
18.2
33.7
27.1
8.4
8.0
6.7
9.2
303.0
303.0
303.0
303.0
303.0
0.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
294.3
288.4
304.9
305. 2
303.0
303.0
299. 1
299. 1
303.0
303.0
303.0
303.0
303.0
303.0
303 . 0
303.0
303.0
303.0
303.0
304.6
303.2
292.2
299.5
16800.
20300.
17100.
22700.
19225.
2806.
24400.
24400.
29200.
29300.
29900.
30400.
26400.
25400.
36900.
16100.
12800.
23400.
34700.
27700.
21600.
22200.
22200.
20900.
19500.
28800.
22000.
27300.
24700.
27700.
26000.
-'7800.
27800.
27500.
20000.*
20000.*
22800.
30900.
0.274
0.293
0.360
0.341
0.317
0.040
1.065
0.777
0.874
0.286
0.799
0.147
0.791
1.050
0.857
0.260
0. 147
0.256
0.863
0.462
0.369
0.378
0.478
0.592
0.352
0.744
0.590
0.743
0.398
0.403
0.344
0.408
0.484
0.523
0.485
0.65B
0.211
0.554
0.296
0.388
0.438
0.551
0.418
0. 106
0.772
0.376
0.326
0.131
0.329
0.086
0.665
0.817
0.599
0.250
0. 125
0. 107
0.474
0.379
0.297
0.536
0.574
0.617
0.442
0.710
0.612
0.785
0.383
0.406
0.359
0.420
0.499
0.520
0.693
0.596
0.277
0.496
0.022
0.095
0.079
0.210
0. 101
0.079
0. 101
0.079
-0.293
-0.401
-0.548
-0. 155
-0.470
-0.060
-0. 127
-0.233
-0.259
-0.010
-0.022
-0.149
-0.389
-0.083
-0.072
0.158
0.096
0.025
0.090
-0.034
0.021
0.042
-0.015
0.002
0.015
0.012
0.016
-0.002
0.208
-0.063
0.066
-0.05B
0.08
0.28
0.20
0.47
0.26
0.17
0. 26
0.17
-0.32
-0.70
-0.91
-0.74
-0.83
-0.52
-0.17
-0.25
-0.36
-0.04
-0.16
-0.62
-0.58
-0.20
-0.22
0.34
0. 18
0.04
0.23
-0.05
0.04
0.06
-0.04
0.01
0.04
0.03
0.03
0.00
0.35
-0. 10
0.27
-0.11
1.73
2.67
2.69
2.84
2.48
0.51
10.53
5. 11
5.86
10.83
6.30
8.43
10.41
19.23
9.94
1.14
0.94
0.96
2.13
2.11
1 .00
1.11
1.35
3.80
5.26
8.03
3.78
8.28
4 .64
3.27
3.91
5.59
5. 19
3.50
1 .08
1.31
0.62
0.95
2.53
8.26
8.80
5.92
6.38
2.85
6. 15
3.84
3.48
5.21
4. 12
7.05
7.84
11.91
5.43
1 .22
1.01
0.64
1 .37
1.41
0.70
1 .29
1 .59
4.08
7.09
6. 1 1
3.44
9.30
4.31
3.27
3.50
5.09
4.78
3.31
1 .23
1 .23
0.67
0.92
0.81
5.59
6.11
3.09
3.90
2.45
3.90
2.45
-4.39
-1 .27
-2.39
-5.63
-2.18
-1 .38
-2.57
-7.32
-4.52
0.08
0.07
-0.35
-0.76
-0.70
-0.30
0. 18
0.24
0.28
1 .83
-1.91
-0.34
1.01
-0.33
0.00
-0.41
-0.50
-0.41
-0. 19
0. 15
-0.08
0.05
-0.03
0.38
1 .02
1 .06
0.70
0.79
0.32
0.79
0.32
-0.53
-0.28
-0.51
-0.70
-0.42
-0. IB
-0.28
-0.47
-0.59
0.07
0.07
-0.43
-0.44
-0.39
-0.35
0. 15
0. 16
0.07
0.30
-0.27
-0.09
0. 12
-0.07
0.00
-0.11
-0.09
-0.08
-0.06
0. 13
-0.06
0.08
-0.03
(cont1nued)
-------
Appendix C. Selected Results of Simulations
RADM-M Mechanism (continued).
of Individual Organlc-NOx-AIr Experiments Using the
Page 10
U>
oo
wo
Experiment
Initial
Concentrat Ions
NOx
Ave
Temp
Ave
H20
HC HC/NOx
(ppm) (ppmC)
OC03B2R
OC0382B
NV15B2R
JL2281R
JL2181B
NV15B2B
DE07B2B
JN1379B
Group Average
S. Dev.
Avg. Abs. Value
S. Dev.
22. 3-Component
ITC479
ITC584
ITC579
ITC472
ITC474
ITC581
ITC5B5
ITC478
ITC482
ITC4B8
ITC492
ITC494
ITC498
1TC500
ITC502
ITC462
ITC466
ITC46B
ITC451
ITC455
ITC977
ITC985
ITC997
ITC979
ITC992
ITC981
ITC993
JL1581B
ST2481B
0.25
0.25
0. 18
0.26
0.24
0.18
0. 19
0.44
0.49
0.37
and "M1n1"
0.09
0. 10
0. 10
0. 10
0.09
0.09
0. 10
0. 10
0. 10
0.09
0.09
0.09
0. 10
0. 10
0.09
0. 1 1
0. 10
0.09
0. 10
0.09
0. 13
0. 12
0. 12
0. 12
0. 12
0. 14
0. 1 1
0.28
0.23
2.7
1 .9
2.6
2.4
1 .3
2.7
3.5
6. 1
5.2
4.5
11 .0
7.7
14.4
9. 1
5.4
14.9
16.7
13.7
15.5
19.2
(degK)
297.2
296.9
283.6
302.3*
304. 2*
263.8
286.4
294.9
300.2
5.7
(ppm)
21200.
21 100.
20000.*
20000.*
20000.*
20000.
13400.
20000.*
24160.
51B8.
Maximum Concentration
OZONE
Expt
(ppm)
0.247
0. 158
0.061
0.626
0.237
0. 167
0.093
0.756
0.492
0.270
Calc
(ppm)
0.141
0.038
0.045
0.628
0.428
0.240
0.296
0.809
0.432
0.221
Calc
-Expt
(ppm)
-0. 106
-0. 120
-U.016
0.002
0.191
0.073
0.203
0.053
-0.060
0. 175
0. 124
0. 136
Calc
-Expt
/Ave
-0.55
-1 .23
-0.30
0.00
0.58
0.36
1 .05
0.07
-0.14
0.44
0.32
0.32
Average
d( [03] -
Expt
Calc
Initial
[NO] )/dt
Calc Calc
-Expt -Expt
(ppb/m1n) -- /Avg
0.76
0.62
0.42
1.18
0.63
0.71
0.55
2.44
4. 10
4.06
0.55
0.34
1.11
0.64
0.74
O.72
2. 78
3.32
2.77
-0.21 -0.32
-0.08 -0.22
-0.06 -0.06
0.01 0.02
0.04 0.05
0.18 0.28
0.34 0.13
-0.87 -0.14
1.83 0.25
1 . 10 0.22
1.69 0.18
Surrogates
4.0
4.0
3.4
3.6
4.2
4.3
4.B
5.0
1 .2
6.0
8. 1
B.5
3.9
4. 1
4.6
7.5
6.5
4.6
5.2
4.4
3.2
2.7
2.7
2.9
2.6
6.5
7. 1
2.3
1 .9
43.4
40.7
34.7
35. 1
45. 1
50.0
50.8
52.6'
12.3
66.2
68.4
92.5
40.2
42.7
49.5
71.3
62.7
49.0
52.0
47. B
25.1
22.9
23.0
23. 1
20.9
45.0
66.9
8.0
8. 1
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303. 0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
304.6
292. 2
20300.
20300.
15800.
13700.
15600.
16100.
'7100.
18300.
18300.
19600.
15000.
?1 100.
19300.
18700.
IB 100.
18300.
18200.
18000.
19800.
20200.
20900.
20100.
i!0100.
20100.
20900.
19000.
20100.
20000.*
14200.
0.332
0.375
0.356
0.258
0.293
0.352
0.311
0.320
0.279
0.293
0.326
0.321
0.320
0.286
0.271
0.099
0. 131
0. 176
0.312
0.304
0.329
0.336
0.316
0.404
0.384
0.328
0.276
0.474
0.246
0.323
0.320
0.303
0.331
0.329
0.315
0.332
0.352
0.298
0.332
0.355
0.363
0.336
0.322
0.304
0.269
0.283
0.304
0.296
0.300
0.330
0.300
0.298
0.418
0.407
0.375
0.359
0.632
0.283
-0.009
-0.055
-0.05"
0.073
0.036
-0.037
0.021
0.032
0.019
0.039
0.029
0.041
0.016
0.036
0.034
0. 170
0. 152
0. 128
-0.016
-0.004
0.001
-0.036
-0.018
0.014
0.023
0.047
0.083
0. 158
0.037
-0.03
-0. 16
-0.16
0.25
0.12
-0.11
0.07
0.09
0.07
0.12
0.09
0. 12
0.05
0. 12
0. 12
0.92
0.73
0.53
-0.05
-0.01
0.00
-0.11
-0.06
0.03
0.06
0.13
0. 26
0. 29
0.14
2.95
3.05
1 .54
1 .72
3.40
3.74
5.11
4.73
2.44
3.06
3.28
3.30
2. 16
5. 18
7.07
0.70
0.77
1 . 19
3.95
3.52
3.94
3.95
3.89
4.88
5.57
2.56
2.45
1. 10
0.71
4.79
3.93
2.14
3. O2
6.06
5.28
6.98
8. 19
4.39
5. 17
5.39
5.33
3.60
7.40
12.30
2.51
2.96
3. 48
5.63
5.05
3.50
3.31
3.26
4.09
4.47
3.33
2.71
1.16
0.71
1.84 0.48
0.88 0.25
0.61 0.33
1.31 0.55
2.69 0.57
1.54 0.34
1.BB 0.31
3.46 0.54
1.94 0.57
2.10 0.51
2.11 0.49
2.03 0.47
1.45 0.50
2.22 0.35
5.23 0.54
1.81 1.13
2.19 1.17
2.29 0.98
1.68 0.35
1.52 0.36
-0.44 -0.12
-0.65 -0.18
-0.63 -0.18
-0.78 -0.17
-1.10 -0.22
0.77 0.26
0.25 0.10
0.06 0.05
0.00 0.00
(cont1nued)
-------
Appendix C.
Sal acted Results of Simulations of Individual Organlc-NOx-AIr Experiments Using the
RADM-M M»ch«n1sm (continued).
Page 11
10
CO
XT
Experiment
Initial
Concentrat Ions
NOx
Ave
Temp
Ave
H20
HC HC/NOx
(ppm) (ppmC)
JN09B2B
JN1483R
JN2783B
AU1883B
AU2683R
JL18B1B
JN0982R
Group Average
S. Dev.
Avg. Abe. Value
S. Dev.
0.28
0.22
0.26
0.28
0.32
0.27
0.29
0. U
0.08
3. 1
2.6
2.9
0.6
2.6
2.3
3. 1
4. 1
1 .9
1 1 .0
11.6
11.1
2.0
8.1
8.5
10.7
37.0
23.7
(degK)
299.7*
301 .2
305. 1
303.0
302.2
303.2
299.7*
302.6
2.0
(ppm)
14300.
26100.
24400.
23100.
26400.
20000.*
14700.
19061 .
3028.
Maximum Concentrat
OZONE
Expt
(ppm)
0.667
0.585
0.511
0.556
0.646
0.693
0.714
0.366
0.148
Calc
(ppm)
0.586
0.529
0.596
0.545
0.612
0.643
0.647
0.387
0. 121
Calc
-Expt
(ppm)
-0.081
-0.056
0.085
-0.01 1
-0.035
-0.050
-0.068
0.021
0.063
0.050
0.043
1 on
Calc
-Expt
/Avg
-0.13
-0. 10
0. 15
-0.02
-0.05
-0.08
-0. 10
0.09
0.23
0. 16
0. 19
Average Initial
d( (03] - [NO] )/dt
Expt
Calc Calc
Calc -Expt -Expt
(ppb/mln) -- /Avg
.37
. 10
.01
0.70
.24
.91
.77
2.81
1 .61
.20 -0.17 -0.13
.07 -0.03 -0.03
.13 0.12 0.11
0.65 -0.05 -0.07
.18 -0.07 -0.05
.76 -0.15 -0.08
.56 -0.21 -0.13
3.85 1.05 0.28
2.43 1 .36 0.36
1.28 0.35
1.14 0.29
23. SAPRC 7-Component Surrogates
EC231
EC232
EC233
EC237
EC23B
EC241
EC242
EC243
EC245
EC246
EC247
Group Average
S. Dev.
Avg. Abs. Value
S. Dev.
0.49
0.49
0. 10
0.48
0.95
0.49
0.50
0.50
1 .00
0.51
0.51
0.65
0.24
13.2
9.3
9.5
10.5
10. 1
5.0
12.9
9.7
12.9
8.6
6.2
9.8
2.6
28.9
18.9
92.5
21.6
10.6
10.2
25.6
19.5
12.9
17.0
12.2
24.4
23.3
303,0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
0.0
17400.
17100.
21400.
23200.
25400.
20700.
24600.
16400.
20400.
22100.
22700.
21036.
3020.
0.620
0.342
0.330
0.652
0.691
0.408
0.682
0.716
0.894
0.574
0.657
0.597
0. 173
0.711
0.377
0.385
0.645
0.701
0.378
0.678
0.717
0.847
0.433
0.625
0.591
0. 167
0.090
0.035
0.056
-0.007
0.010
-0.030
-0.003
0.001
-0.047
-0.141
-0.032
-0.006
0.060
0.041
0.043
0.14
0. 10
0. 16
-0.01
0.01
-0.08
-0.01
0.00
-0.05
-0.28
-0.05
-0.01
0.12
0.08
0.08
6.99
3. 12
4.07
7.36
5. 10
2.87
17.53
14.50
13.41
2.65
7.49
7.74
5. 16
7.96 0.97 0.13
3.03 -0.09 -0.03
4.73 0.66 0.15
5.98 -1.38 -0.21
4.79 -0.31 -0.06
2.94 0.08 0.03
16.62 -0.91 -0.05
12.74 -1 . 76 -0.13
12.84 -0.57 -0.04
1.97 -0.68 -0.29
6.48 -1.01 -0.14
7.28 -0.46 -0.06
4.78 0.83 0.13
0.77 0.12
0.52 0.08
24. SAPRC 8-Component Surrogates
ITC626
ITC630
ITC631
ITC633
ITC635
ITC637
ITC865
0.30
0.31
0.32
0.64
1 .21
0.31
0.28
4.0
1 .9
1 .0
4.0
4.0
4.0
8.4
13.4
6.3
3.2
6.2
3.3
12.8
29.9
303.0
303.0
303 . 0
303. 0
303. 0
30:1 . o
303.0
14300.
16300.
17500.
16600.
17500.
16900.
15100.
0.618
0.284
0.043
0.231
0.006
0.617
0.632
0.548
0.287
0.066
0.308
0.021
0.545
0.514
-0.070
0.003
0.023
0.077
0.015
-0.072
-0.118
-0.12
0.01
0.28
-0.12
-0.21
2. IB
0.86
0.47
1.26
0.95
2.31
2.60
2.32 0.14 0.06
1 .04 0.18 0.19
0.54 0.07 0.15
1.58 0.31 0.22
1.05 0.09 0.09
2.30 -0.01 0.00
3.51 0.91 0.30
(cont 1nued)
-------
Appendix C. Selected Results of Simulations of Individual Organlc-NOx-A1r Experiments Using the
RAOM-M Mechanism (continued).
Page 12
oo
oo
Experiment
Initial
Concent rat Ions
NOx HC HC/NOx
(ppm) (ppmC)
ITC867
ITC868
ITC871
ITCB72
ITC873
ITC874
ITC877
ITCB80
ITC881
ITC885
1TC886
ITC888
ITC891
OTC189A
OTC1B9B
OTC190A
OTC190B
OTC192A
OTC192B
OTC194A
OTC194B
OTC195A
OTC195B
OTC196B
OTC197A
OTC197B
OTC198A
OTC198B
OTC199A
OTC199B
OTC202A
OTC202B
OTC203A
OTC203B
OTC204A
OTC204B
OTC205A
OTC2058
OTC215A
OTC215B
OTC217A
OTC217B
OTC221A
OTC221B
OTC222A
0.28
0.37
0.37
0.38
0.39
0.38
0.38
0.73
0.73
0.64
0.73
0.33
0.32
0.45
0.4S
0.41
0.42
0.47
0.47
0.38
0.38
0.41
0.41
0.41
0.38
0.77
0.84
0.40
0.36
0.37
0.73
0.40
0.39
0. 19
0.35
0. 17
0.64
0. 14
0.45
0.44
0.50
0.51
0.41
0.42
0.44
4.8
2.9
1 .7
2. 1
1 .3
2. 1
2.3
2.2
2.3
1.5
2.3
4.7
4.4
3.3
4.0
3.8
3.8
4.2
4.2
7.4
3.7
1 .8
4. 1
3.7
3.6
3. 1
5.4
3.7
3.5
3.5
6.0
2.6
3.5
3.5
3.6
3.5
3.8
3.6
3.5
4. 1
4.9
4.6
2.4
2.0
3.4
17.2
7.8
4.6
5.7
3.4
5.7
6.2
3.0
3. 1
2.4
3. 1
14.5
13.7
7.4
8.9
9.2
9.0
8.9
B.9
19.3
9.7
4.5
9.9
9. 1
9.5
4. 1
6.4
9.4
9.6
9.6
8.2
6.7
9.0
18.5
10.2
20.3
4.5
26.5
7.8
9.3
9.6
9.0
5.8
4.7
7.8
Ave
Temp
(degK)
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
303.0
295.4
298.0
302.7
299.5
297.8
298.4
296. 1
295.7
351 .6
360.0
337.5
353.2
356.0
353.5
347.5
299.7
298.9
325. 1
324. 1
293.3
293.7
322.2
34B.6
464.9
433.4
293. 1
293.4
295.5
294. 2
290.6
288.8
298.5
Ave
H20
(ppm)
14600.
17100.
14600.
14600.
14600.
13200.
14600.
13700.
14000.
14600.
20000.*
18000.
20000.*
5000.*
5000.*
5400.*
5410.*
5000.*
5000.*
5000.*
5000.*
4660.*
3180.*
5000.*
5000.*
5000.*
5000.*
5000.*
5000.*
5000.*
5000.*
4560.
4850.*
4880.*
4330.
4330.*
5000.
5000.*
4420.*
4420.*
7460.*
7460.*
4460.*
4460. *
5000.*
Maximum Concentration
OZONE
Expt
(ppm)
0.631
0.518
0.376
0.213
0. 160
0. 191
0.250
0.031
0.012
0.012
0.012
0.579
0.602
0.576
0.712
0.702
0.669
0.735
0.754
0.737
0.603
0. 164
0.681
0.597
0.621
0. 152
0.704
0.674
0.591
0.606
0.714
0.522
0.231
0.394
0.302
0.381
0.039
0.392
0.830
0.868
0.483
0.831
0.235
0.333
0.909
Calc
(ppm)
0.544
0.435
0.265
0.228
0.115
0. 160
0.237
0.046
0.033
0.027
0.022
0.512
0.515
0.766
0.845
0.841
0.793
0.852
0.760
0.905
0.701
0.112
0.775
0.804
0.554
0. 102
0.576
0.633
0.653
0.638
0.494
0.424
0.282
0.447
0.377
0.468
0.060
0.552
O.B57
0.937
0.584
0.935
0.286
0.357
0.964
Calc
-Expt
(ppm)
-0.088
-0.083
-0.111
0.015
-0.045
-0.031
-0.013
0.016
0.021
0.015
0.010
-0.068
-0.087
0. 190
0. 132
0.138
0. 124
0. 1 17
0.006
0. 167
0.098
-0.052
0.095
0.206
-0.068
-0.049
-0. 128
-0.041
0.062
0.032
-0.220
-0.09B
0.051
0.053
O.O75
0.087
0.021
0. 160
0.027
0.069
0.101
0. 104
0.051
0.024
0.055
Calc
-Expt
/Avg
-0. 15
-0. 17
-0.35
0.07
-0.33
-0. 18
-0.05
-0.12
-0. 16
0.28
0.17
0.18
0. 17
0. 15
0.01
0.20
0. 15
-0.38
0. 13
0.29
-0. 12
-0.39
-0.20
-0.06
0.10
0.05
-0.37
-0.21
0.20
0. 13
0.22
0.21
0.34
0.03
0.08
0.19
0.12
0.20
0.07
0.06
Average
d( [03] -
Initial
(NO] )/dt
Calc
Expt Calc -Expt
(ppb/m1n) --
2.23
2. 10
1 . 18
1 .20
0.85
0.83
1.14
0.85
1 .03
0.44
1 .24
1 .65
2.67
3.79
3.70
4.24
3.73
4.25
3.84
7.85
2.95
2.05
4. 14
3.86
4.12
2.86
6.09
3.86
3.86
3.40
6.95
3.B2
2.34
2.20
3.27
2.62
4.85
4.49
4.33
5.33
2.55
4.64
1 .63
1 .63
3.83
3.41
1 .46
1.12
1 .36
0.80
0.88
1 .34
0.99
1 . 10
0.78
0.70
1 .89
2.91
3.62
3.74
3.35
2.94
3.41
2.80
7.71
3.04
1 .20
3.83
3.37
2.66
2.07
3.69
2.58
3.05
2.92
13. 77
3.17
2.36
2.65
3.46
2.91
2. 16
3.71
3.65
5.49
2.52
4. 30
1 .50
1 .60
3.75
1 . 19
-0.64
-0.06
0. 16
-0.05
0.05
0.20
0. 14
0.07
0.35
-0.54
0.24
0.23
-0. 17
0.04
-0.89
-0.79
-0.85
-1 .05
-0. 14
0.09
-0.85
-0.31
-0.49
-1 .46
-0.79
-2.40
-1 .28
-0.81
-0.47
6.81
-0.65
0.02
0.45
0.21
0.29
-2.70
-0.78
-0.68
0.16
-0.03
-0.34
-0.12
-0.03
-0.09
Calc
-Expt
/Avg
0.42
-0.36
-0.05
0. 13
-0.06
0.06
0. 16
0. 15
0.06
0.57
-0.55
0. 14
0.08
-0.05
0.01
-0.23
-0.24
-0.22
-0.31
-0.02
0.03
-0.52
-0.08
-0.14
-0.43
-0.32
-0.49
-0.40
-0.23
-0. 15
0.66
-0. 19
0.01
0. 19
0.06
0. 1 1
-0.77
-0. 19
-0. 17
0.03
-0.01
-0.08
-0.08
-0.02
-0.02
(cont1nued)
-------
Appendix C. Selected Results of Simulations of Individual Organ1c-NOx-A1r Experiments Using the
RADM-M Mechanism (continued).
Page 13
Oa
CTi
Experiment
Initial
Concentrat
Ions
NOx HC HC/NOx
(ppm) (ppmC)
OTC222B
OTC223A
OTC223B
OTC224A
OTC224B
OTC226A
OTC22BA
OTC228B
OTC229A
OTC229B
OTC230A
OTC230B
OTC237A
OTC237B
OTC23BA
OTC23BB
OTC239A
OTC239B
OTC240A
OTC240B
OTC241A
OTC241B
OTC242A
OTC242B
OTC243A
OTC243B
OTC24BA
OTC248B
OTC249A
OTC249B
Group Average
S. Oev.
0.43
0.36
0.40
0.34
0.34
0.45
0.41
0.41
0.46
0.46
0.41
0.41
0.52
0.52
0.50
0.50
0.49
0.50
0.50
0.50
0.32
0.31
0.45
0.46
0.47
0.3B
0.48
0.46
0.48
0.46
0.45
0. 16
2.6
4.8
3.5
4.4
4.3
2.5
2.4
2.3
3.0
1 .7
3. 1
1 .7
4.3
4.4
2.9
3.9
2.7
2.5
1 .8
2.0
4.7
4.0
2.6
2. 1
4.7
3.7
3.7
2.9
5.6
4.9
3.5
1 .3
6.0
13.4
B.B
12.7
12.6
5.5
6.0
5.6
6.6
3.8
7.6
4.2
8.2
8.5
5.8
7.8
5.4
5.0
3.6
4. 1
14.8
12.9
5.8
4.7
10.0
9.7
7.7
6.3
11.8
10.7
8.7
4.9
Ave
Temp
(detiK)
296. 1
314.9
314.7
303.5
303.8
312.5
293.2
294.3
295.7
296.0
297.9
298. 2
292.3
292.8
292.5
292.0
290. 2
290.9
93.4
104. 2
290. 7
291. 1
30?. 0
302. 1
276.5
277.4
271 .6
268.9
286. 9
286.2
302.3
43.6
Ave Maximum Concentration
H20 OZONE
Expt
(ppm) (ppm)
5000.
4270.
4270.
6410.
6410.
6840.
5000.
5000.
5000.
5000.
5000.
5000.
3810.
3810.
4810.
4810.
5150.
5150.
6210.
6210.
4760.
4760.
4670.
4670.
4720.
4720.
5000.
5000.
3540.
0.940
0.953
0.771
0.776
0.813
0.751
0.246
0.296
0.253
0. 168
0.4B9
0.271
0.807
0.757
0.406
0.702
0.343
0.234
0.034
0.217
0.671
0.674
0. 182
0.639
0. 142
0. 152
0.056
O.OB1
0.348
3540. 0.325
7630. 0.454
4865. 0.272
Calc
(ppm)
1 .127
0.883
0.729
0.786
0.850
0.545
0.433
0.464
0.460
0.405
0.842
0.564
0.836
0.691
0.403
0.700
0.401
0. 172
0.037
0.119
0.698
0.683
0. 161
0.640
0.207
0. 174
0.060
0.090
0.340
0.365
0.483
0.286
Avg. Abs. Value
S. Dev.
25. UNC Mlscel
ST20B1R
DE0782R
AU2681R
AU2681B
AU2781B
ST03B1R
ST10B1R
ST2081B
Calc
-Expt
(ppm)
0.187
-0.069
-0.043
0.010
0.038
-0.205
0. 187
0. 168
0.207
0.237
0.353
0.293
0.029
-0.066
-0.003
-0.002
0.058
-0.062
0.003
-0.098
0.027
0.009
-0.021
0.001
0.065
0.022
0.004
0.009
-0.008
0.040
0.029
0. 103
0.078
0.072
Calc
-Expt
/Ava
0.16
-0.08
-0.06
0.01
0.05
-0.32
0.55
0.44
0.58
0.83
0.53
0.70
0.03
-0.09
-0.01
0.00
0.15
-0.31
-0.58
0.04
0.01
-0.12
0.00
0.37
0. 13
0.06
0. 10
-0.02
0. 12
0.05
0.25
0. 19
0. 17
Average
d( (03] -
Expt Calc
(ppb/m1r
5.55
4.55
3.15
3.67
4.94
3.12
1 .49
1 .79
1 .56
1 .38
2.43
1 .66
4. 10
3. 46
2.58
4.06
2.66
1 .85
1 .08
1 .79
3.26
3.30
1 .47
3.04
1.41
1 . 40
1 .40
1 .61
2.55
2.83
2.81
1 .51
6.25
4. 15
2.88
2.96
4.48
2.37
1 .96
2.60
2.01
1 .87
3. 14
2.21
3.46
2.97
2.22
3.60
2.54
1 .49
0.92
1 .39
2.99
3.27
1 .35
2. 70
1 .84
2.18
1 .24
1 .73
2.36
3.02
2.71
1 .76
Initial
[NO] )/dt
Calc Calc
-Expt -Expt
i) /Avg
0.70 0.12
-0.41 -0.09
-0.27 -0.09
-0.71 -0.21
-0.47 -0.10
-0.75 -0.27
0.47 0. 27
0.81 0.37
0.45 0.25
0.49 0.30
0.71 0. 25
0.55 0.29
-0.64 -0.17
-0.50 -0.16
-0.36 -0.15
-0.46 -0.12
-0.12 -0.05
-0.35 -0.21
-0.16 -0.16
-0.40 -0.25
-0.28 -0.09
-0.03 -0.01
-0.13 -0.09
-0. 33 -0.12
0. 43 0.27
0.77 0.43
-0.16 -0.13
0.11 0.07
-0.19 -0.08
0.19 0.07
-0. 10 -0.03
1.00 0.25
0.54 0.19
0.84 0.16
laneoua Surrogates
0.23
0. 19
0.24
0.24
0.23
0.24
0.25
0.23
2.3
3.4
2.0
2.0
2.0
1 .8
2.8
2 . 1
10.0
18.3
8.4
8.5
8.8
7.6
11 .3
9.2
294. 1
286.4
298.8
295.7
296.4
301 .0
297.9
292.6
21900. 0.403
13400. 0.076
25500. 0.506
27500. 0.544
25000. 0.623
23200. 0.541
22300. 0.610
15200. 0.414
0.466
0.357
0.520
0.448
0.476
0.452
0.603
0.414
0.062
0. 281
0.014
-0.096
-0. 147
-0.089
-0.008
0.000
0.14
1 .30
0.03
-0. 19
-0. 27
-U. 18
-0.01
0.00
0.95
0.52
1 . 18
1 . 22
1 .27
1 .48
1 .59
1 .06
1 .06
0.76
1 .08
1.01
1 .07
1 .22
1 .54
0.99
0.10 0.10
0.23 0.36
-0.10 -0.09
-0.21 -0.19
-0.20 -0.17
-0.26 -0.19
-0.05 -0.03
-0.07 -0.07
(cont1nued)
-------
AppencH x C.
Selected Results
RADM-M Mechanism
of Simulations
(cont1nued).
of Individual Organ1c-N0x-A1r Experiments Using the
to
oo
-J
Page 14
Experiment
Initial
Concentrations
NOx HC HC/NOx
(ppm) (ppmC)
JL2081B
ST1682R
JL20B1R
JL2281B
OC1481R
ST1682B
ST2981R
ST29B1B
OC1481B
ST0381B
ST10B1B
JL08B2R
JL0882B
Group Average
S. Dev.
Avg. Abs. Value
S. Oev.
26. UNC "Synurban"
AU22B4R
AU2584R
AU2584B
ST0184R
ST0184B
ST0284R
ST0284B
JN2685R
JN2685B
JN2885R
JN2885B
Group Average
S. Dev.
Avg. Abs. Value
S. Oev.
0.42
0.43
0.41
0.26
0.28
0.43
0.24
0.24
0.29
0.23
0.24
0.29
0.28
0.28
0.08
1 .8
3.2
2.7
2.9
3.3
3. 1
2.5
2.5
2.9
2.0
1 .0
2. 1
2. 1
2.4
0.6
4.3
7.5
6.6
11.2
1 1 .9
7.2
10.3
10.4
9.9
B.6
4. 1
7.3
7.4
9.0
3.0
Ave
Temp
(degK)
305.0*
304.4
305.0*
302.3*
291 .3
304.4
293.7
291 .6
288.9
297.9
295.3
304.2
304.2
297.7
5.8
Ave
H20
(ppm)
20000.*
23000.
20000.*
20000.*
9470.
23000.
18700.
14400.
10300.
16800.
17700.
16000.
16000.
19018.
4902.
Maximum Concentration
OZONE
Expt
(ppm)
0. 165
0.410
0.635
0.722
0.462
0.840
0.294
0.485
0.458
0.611
0.626
0.598
0.541
0.503
0. 175
Calc
(ppm)
0.325
0.508
0.679
0.675
0.569
0.800
0.426
0.494
0.561
0.516
0.382
0.703
0.689
0.527
0. 126
Calc
-Expt
(ppm)
0. 160
0.099
0.043
-0.047
0. 108
-0.040
0.132
0.009
0. 102
-0.095
-0. 244
0. 105
0. 147
0.024
0.121
0.097
0.073
Calc
-Expt
/Avg
0.66
0.22
0.07
-0.07
0.21
-0.05
0.37
0.02
0. 20
-0.17
-0.4B
0.16
0.24
0. 10
0.37
0.24
0.29
Average Initial
d( [03] - [NO] )/dt
Calc
Expt Calc -Expt
(ppb/m1n) --
0.62
1 .28
1 .55
1 .80
1 .53
2.39
0.76
1 .35
1.48
1 .70
1 .36
1 .25
1.12
1 .32
0.40
0.82 0.01
1.17 -0.11
1 .44 -0.11
1.70 -0.10
1.57 0.04
2.08 -0.31
0.88 0.12
1 .30 -0.05
1.51 0.03
1.55 -0.15
0.96 -0.40
1 .35 0.10
1.30 0.18
1.26 -0.06
0.33 0.16
0. 14
0. 10
Calc
-Expt
/Avg
0.01
-0.09
-0.07
-0.06
0.03
-0. 14
0.15
-0.04
0.02
-0.09
-0.34
0.08
0. 15
-0.03
0. 15
0.12
0.10
Surrogate
0.32
0.34
0.33
0.31
0.30
0.34
0.33
0.30
0.30
0.38
0.39
0.33
0.03
0.4
0. 1
0. 1
0.2
0.2
0. 1
0. 1
0.2
0.3
0. i
O.Z
0.2
0. 1
1 .2
0.3
0.4
0.6
0.8
0.4.
0.3
0.6
0.9
0.4
0.5
0.6
0.3
302.7
302.0
302.0
302.5
302.5
304.6
304.5
303.6
303.6
299.8
299.8
302.5
1 .6
8470.
22600.
22600.
9380.
9380.
19900.
19900.
16800.
16800.
1 1800.
1 1800.
15403.
5427.
0.657
0.075
0.096
0.546
0.646
0. 1 19
0.020
0.629
0.788
0.238
0.275
0.372
0.284
0.773
0. 109
0. 135
0.661
0.704
0.265
0. 144
0.677
0.761
0.391
0.278
0.445
0.272
0.116
0.034
0.039
0.115
0.058
0. 146
0. 124
0.049
-0.027
0. 153
0.003
0.074
0.060
0.07B
0.053
0. 16
0.37
0.33
0.19
0.09
0.76
0.07
-0.03
0.49
0.01
0. ?4
O.ilb
0.25
0.24
1.61
0.70
0.69
1 .32
1 .72
0.73
0.40
1 .34
1 .74
0.85
0.87
1 .09
0.47
1.63 0.03
0.64 -0.06
0.66 -0.03
1.31 -0.01
1.58 -0.14
0.76 0.02
0.50 0.10
1 .46 0.12
1 .85 Oil
0.89 0.04
0.88 0.01
1.11 0.02
0.47 0.08
0.06
0.05
0.02
-0.09
-0.05
-0.01
-0.08
0.03
0.22
0.09
0 06
0.05
0.01
0 02
0.09
0.06
0.06
(cont1nued)
-------
Appendix C. Selected Results of Simulations of Individual Organ1c-N0x-A«r Experiments Using the
RADM-M Mechanism (continued).
Page IS
to
00
CD
Experiment
Initial
Concentret
NOx HC
(ppm) (ppmC)
27. UNC "Synauto'
AU04B4R
AU0484B
AU0584R
AU05B4B
AU0684R
AU0684B
AU0784R
AU0884B
AU0984R
ST0884R
STOB84B
ST1784R
ST1784B
ST2184R
ST2184B
Group Average
S. Dev.
Avg. Abs. Value
S. Dev.
Ions
HC/NOx
Ave
Temp
(degK)
Ave
H20
(ppm)
Maximum Concentration
OZONE
Expt
(ppm)
Calc
(ppm)
Calc
-Expt
(ppm)
Calc
-Expt
/Avg
Average
d( [03} -
Initial
{NOJ )/dt
Calc
Expt Calc -Expt
(ppb/m1n) --
Calc
-Expt
/Avg
Surrogate
0.37
0.36
0.36
0.35
0.35
0.35
0.38
0.34
0.39
0.34
0.33
0.34
0.34
0.36
0.36
0.35
0.02
0.4
0.3
0.3
0.4
0.7
1 . 1
0.4
1 . 1
0.4
0.6
1 .0
1 . 1
O.B
0.6
0.7
0.7
0.3
1 .0
0.8
0.7
1.0
2. 1
3.2
1.0
3. 1
0.9
1.9
2.9
3.3
2.3
1 .7
2.0
1 .9
0.9
30/ .S
307.5
307. 1
307.2
307.6
307.6
308.0
307.3
307.8
296.5
298.5
294.4
294.4
302.2
302.2
303.9
5. 1
34900.
34700.
44900.
44900.
36800.
36800.
34700.
29000.
31500.
6920.
6920.
24800.
24800.
7350.
7350.
27089.
13697.
0.515
0.328
0.335
0.595
0.887
0.940
0.602
0.834
0.521
0.566
0.750
0.539
0.484
0.671
0.721
0.619
0. 182
0.646
0.562
0.465
0.656
0.853
0.902
0.672
0.625
0.559
0.751
0.874
0.716
0.705
0.827
0.648
0.724
0.131
0. 133
0.234
0. 130
0.062
-0.035
-0.037
0.070
-0.009
0.039
0. 185
0. 123
0. 177
0.222
0. 157
0. 126
0. 105
0.087
0. 1 16
0.071
0.23
0.53
0.33
0. 10
-0.04
-0.04
0. 1 1
-0.01
0.07
0. 28
0. 15
0.26
0.37
0.21
0. 16
0. 18
0. 16
0. 19
0. 14
1 .45
1 . 16
1 .08
1 .48
2.53
3. 16
1 .61
2.92
1 .51
.27
.90
.59
.36
. 78
2.28
1 .81
0.63
1 .38
1.12
0.98
1.3B
2. 10
2.52
1 .48
2.39
1 . 26
1 . 38
1 .69
2. 26
1 .95
2.01
2.61
1 .77
0.53
-0.06
-0.04
-0.10
-0. 10
-0.43
-0.64
-0.13
-0.52
-0. 25
0.11
-0.21
0.67
0.58
0. 22
0.32
-0.04
0, 37
0.29
0.22
-0.04
-0.04
-0.10
-0.07
-0.18
-0.22
-0.09
-0.20
-0. 18
0. 08
-0.12
0. 35
0. 35
0.12
0. 13
-0.01
0.18
0. 15
0. 10
28. UNC Auto Exhauat
OC0483B
OC0783R
OC0783B
JN2582R
JN2582B
JN2982R
JN2982B
JN3082R
JN3082B
JL0283B
JL0883B
ST2982B
OC0682R
AU1 183R
AU1 183E)
JL0182R
JL0182B
0.25
0.33
0.34
0.65
0.65
0.24
0.25
0.32
0.32
0. 19
0.37
0.39
0.46
0.22
0.23
0.37
0.35
0.4
2.7
2.7
0.6
0.6
2.5
2.5
2.6
0.6
1 .7
1 .7
1 .7
2.0
2.2
0.7
3.5
3.6
1 .7
8.1
7.9
0.9
0.9
10.3
9.8
8.7
1 .7
9. 1
4.6
4.3
4.3
9.9
3. 1
9.6
10.2
299. 7
295.5
295.4
301 . 1
30' . 2*
303. 1
303. 1
302.5*
302.3*
304.0
299.5
295.7
300.2
306.8
306.9
299.5
299.3
18800.
21600.
21500.
22700.
22900.
30500.
30500.
20000.*
20000.*
24500.
21200.
24800.
16900.
21400.
21400.
27700.
27400.
0.642
0. 178
0.451
0.003
0.003
0.704
0.766
0.811
0.840
0.697
0.879
0. 205
0.355
0.850
0.601
0. 740
0. 759
0.719
0.391
0.650
0.013
0.014
0.833
0.839
1 .005
1.011
0.850
0.845
0.057
0. 276
0.890
0.673
1 .038
0.995
0.077
0.213
0. 199
0.010
0.01 1
0. 128
0.073
0. 194
0. 170
0. 152
-0.034
-0. 148
-0.079
0.040
0.072
0.298
0.236
0. 11
0.75
0.36
0. 17
0.09
0.21
0. 18
0. 20
-0.04
-1.13
-0.25
0.05
0. 1 1
0.34
0. 27
1 .86
0.94
1.41
0.64
0. 68
2.32
2. 67
2.55
2. 68
1 .65
1 . 76
1.12
1 .37
2.55
1 . 26
2.31
2.41
1 .97
1.13
1 .93
0.58
0 . 58
2.10
2.18
2.40
2.40
1.72
1 .44
0.76
1.16
2.15
1 . 25
2.40
2.52
0.12
0. 19
0. 52
-0.06
-0.10
-0.22
-0. 50
-0. 15
-0. 28
0.07
-0.32
-0.36
-0.19
-0.39
0.00
0.09
0. 11
0 . 06
0.18
0.31
-0.10
-0.17
-0.10
-0 . 20
-0.06
-0.11
0.04
-0. 20
-0.38
-0. 15
-0.17
0. 00
0.04
0.05
(cont1nued)
-------
Appendix C. Selected Results of Simulations of Individual Organlc-NOx-A1r Experiments Using the
RADM-M Mechanism (concluded).
Page 16
OJ
00
Exper 1ment
Initial
Concentrations
NOx
(ppm)
AU0382R
AU0382B
ST1782R
ST1782B
ST2982R
OC0682B
JL0283R
JL0883R
JL1583R
JL1583B
OC0483R
Group Average
S. Dev.
Avg. Abs . Value
S. Dev.
29. SAPRC Synthetic
ITC781
ITC784
ITC785
ITC805
ITC795
ITC796
ITC799
ITC801
Group Average
S. Dev.
Avg. Abs. Value
S. Dev.
30. SAPRC Synthetic
ITC963
ITC965
ITC967
ITC968
Group Average
S. Dev.
Avg. Abs . Va 1 ue
S. Dev.
0.44
0. 16
0.26
0.25
0.39
0.46
0. 18
0.37
0.35
0.35
0.25
0.34
0. 12
Jet
0.51
0.50
0. 26
0.52
0.50
0.54
0.51
0.55
0.49
0.09
Jet
0.49
0.46
0. 26
0.49
0.42
0. 1 1
Ave
Temp
Ave
H20
HC HC/NOx
(ppmC)
2.5
1 .2
2.2
2.4
1 .7
2.0
1 .6
1 .7
2.2
2.3
2.6
2.0
0.8
Fuel
43.0
88.0
45.0
98.0
45.0
97.0
94.0
41.0
68.9
27.3
Ex haus »
4.4
5.2
4.4
8.7
5.7
2. 1
(degK)
5.7
7.3
8.6
9.5
4.4
4.3
9.0
4.6
6.2
6.6
10.3
6.5
3. 1
83.5
177.6
170.7
189.5
89.8
178 .9
184. 1
75.1
143.7
50.8
9. 1
11.3
17.2
17.8
13.8
4.3
302.
302.
300.
300.
295.
300.
304.
299.
306.
306.
299.
301 .
3.
303.
303.
303.
303.
303.
303.
303 .
303.
303,
0.
303.
303.
303.
303.
303.
0.
b
6
1
1
7
2
0
5
2
2
7
2
3
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
(ppm)
27400.
27400.
27400.
27400.
24800.
16900.
24500.
71200.
25700.
25700.
18800.
23607.
3802.
16600.
16600.
17900.
17200.
17900.
17200.
17200.
17000.
17200.
499.
19400.
20000.
18300.
19400.
19275.
709.
Maximum Concentration
OZONE
Expt
(ppm)
0.344
0.576
0.562
0.637
0.073
0.450
0.595
0.756
0.866
0.922
0.603
0.567
0.272
0.751
0.746
0.598
0.791
0.761
0.597
0.840
0.881
0.746
0. 102
0.822
0.863
0.586
0.852
0.7B1
0.131
Calc
(ppm)
0.450
0.501
0.590
0.616
0.047
0.328
0.675
0.694
0.737
0.860
0.672
0.617
0.314
0.914
0.906
0.602
0.942
0.794
0.840
0.972
0.957
0.866
0. 122
0.722
0.728
0.591
0.791
0.708
0.084
Calc
-Expt
(ppm)
0. 106
-0.074
0.028
-0.021
-0.026
-0.121
O.OBO
-0.062
-0. 129
-0.062
0.069
0.050
0.119
0. 104
0.074
0. 16:)
0 . 1 60
0.004
0 . 1 c> 1
0.033
0.242
0.131
0.076
0. 1 20
0.078
0. 120
0.078
-0. 100
-0. 135
0.005
-0.061
-0.073
0.060
0.075
0.056
Calc
-Expt
/Avg
0.27
-0. 14
0.05
-0.03
-0.44
-0.31
0.13
-0.09
-0.16
-0.07
0.11
0.03
0.34
0.23
0.24
0.20
0.19
0.01
0.17
0.04
0.34
0.15
0.08
0.15
0.10
0. 15
0. 10
-0.13
-0. 17
0.01
-0.07
-0.09
0.08
0.10
0.07
Average
d( [03] -
Expt
Calc
Initial
[NOJ )/dt
Calc
-Expt
(ppb/m1n) --
1 .26
1 .06
1 .63
1 .72
0.86
1 .49
1 .37
1 .52
2.10
2.20
1 .76
1 .68
0.61
2.77
3.93
2.74
3.27
3.05
4.43
4.59
2.98
3.47
0.74
4.00
5. 10
5.79
1 1 . 76
6.66
3.48
1.23
0.93
1 . 16
1 . 27
0.68
1 .27
1 .37
1 .32
1 .75
1 .99
1.75
1 .55
0.59
7.54
9.86
6.65
9.39
10.78
13.31
8.44
6. 14
9.01
2.35
4. 15
5.28
6. 14
1 1 .08
6.66
3.06
-0.02
-0.14
-0.47
-0.45
-0.18
-0.22
0.00
-0.20
-0.35
-0.21
-0.02
-0.13
0.23
0.21
0. 15
4.77
5.93
3.91
6.12
7.72
8.88
3.86
3. 16
5.54
2.01
5.54
2.01
0.14
0. 18
0.35
-0.68
0.00
0.46
0. 34
0.24
Calc
-Expt
/Avg
-0.02
-0.14
-0.34
-0.30
-0.23
-0. 16
0.00
-0.14
-0. 18
-0. 10
-0.01
-0.09
0. 15
0. 14
0. 10
0.92
0.86
0.83
0.97
1 . 12
1 .00
0.59
0.69
0.87
0. 17
0.87
0.17
0.03
0.03
0.06
-0.06
0.02
0.05
0.05
0.01
------- |