NASA Atmos 6mar00

JPL Publication 00-3
Chemical Kinetics and Photochemical Data

for Use in Stratospheric Modeling
Supplement to Evaluation 12: Update of Key Reactions
Evaluation Number 13
NASA Panel for Data Evaluation:

S. P. Sander A. R. Ravishankara
R. R. Friedl NOAA Environmental
W. B. DeMore Research Laboratory
Jet Propulsion Laboratory
D. M. Golden C. E. Kolb
SRI International Aerodyne Research, Inc.
M. J. Kurylo M. J. Molina
R. F. Hampson Massachusetts Institute of
R. E. Huie Technology
National Institute of Standards and
Technology
G. K. Moortgat
Max-Planck Institute for Chemistry
March 9, 2000
Jet Propulsion Laboratory

California Institute of Technology
Pasadena, California
The research described in this publication was carried out by the Jet Propulsion Laboratory,
California Institute of Technology, under a contract with the National Aeronautics and Space
Administration.
Reference herein to any specific commercial product, process, or service by trade name,
trademark, manufacturer, or otherwise, does not constitute or imply its endorsement by the United
States Government or the Jet Propulsion Laboratory, California Institute of Technology
2
ABSTRACT
This is the thirteenth in a series of evaluated sets of rate constants and photochemical cross
sections compiled by the NASA Panel for Data Evaluation.
The data are used primarily to model stratospheric processes, with particular emphasis on
the ozone layer and its possible perturbation by anthropogenic and natural phenomena.
Copies of this evaluation are available in electronic form and may be printed from the
following Internet URL:
http://jpldataeval.jpl.nasa.gov/
3
TABLE OF CONTENTS
INTRODUCTION ............................................................................................................................6
Basis of the Recommendations............................................................................................. 8
Scope of the Evaluation ........................................................................................................ 8
Format of the Evaluation ...................................................................................................... 9
Computer Access .................................................................................................................. 9
Data Formats......................................................................................................................... 9
Units...................................................................................................................................... 9
BIMOLECULAR REACTIONS ....................................................................................................10
Uncertainty Estimates ......................................................................................................... 11
Notes to Table 1 .................................................................................................................. 15
TERMOLECULAR REACTIONS.................................................................................................22
Low-Pressure Limiting Rate Constant [kxo(T) ] .................................................................. 22
Temperature Dependence of Low-Pressure Limiting Rate Constants: Tn ......................... 23
High-Pressure Limit Rate Constants [k(T)] ...................................................................... 24
Temperature Dependence of High-Pressure Limiting Rate Constants: Tm ....................... 24
Uncertainty Estimates ......................................................................................................... 24
Notes to Table 2 .................................................................................................................. 26
EQUILIBRIUM CONSTANTS......................................................................................................29
Format ................................................................................................................................. 29
Definitions........................................................................................................................... 29
Notes to Table 3 .................................................................................................................. 30
PHOTOCHEMICAL DATA ..........................................................................................................31
Discussion of Format and Error Estimates ......................................................................... 31
HETEROGENEOUS CHEMISTRY ..............................................................................................40
Surface Types...................................................................................................................... 40
Surface Porosity .................................................................................................................. 41
Temperature Dependence ................................................................................................... 42
Solubility Limitations ......................................................................................................... 42
Data Organization ............................................................................................................... 42
Parameter Definitions ......................................................................................................... 43
Evaluation Process .............................................................................................................. 45
Notes to Table 7 .................................................................................................................. 48
APPENDIX.....................................................................................................................................57
Phenomenological Model of ClONO2 Hydrolysis and the Reaction
of ClONO2 and HOCl with HCl in Sulfuric Acid Solutions.............................................. 57
REFERENCES ...............................................................................................................................63
TABLES
Table I-1: Editions of this Publication .............................................................................................6
Table I-2: Panel Members and their Major Responsibilities ...........................................................6
Table 1: Rate Constants for Second Order Reactions..................................................................14
Table 2: Rate Constants for Association Reactions.....................................................................25
Table 3: Equilibrium Constants ...................................................................................................30
Table 4: Parameters for the O3 Quantum Yield Equation ...........................................................33
4
Table 5: Absorption Cross Sections of HOCl................................................................................37
Table 6: Absorption Cross Sections of HOBr................................................................................39
Table 7: Gas/Surface Reaction Probabilities ()............................................................................47
Table A-1: Calculations of H2SO4 wt.% from T and pH2O.............................................................58
Table A-2: Parameters for H2SO4 Solution ...................................................................................59
Table A-3: Uptake Model Parameters for the ClONO2 + H2O and ClONO2 + HCl Reactions ....60
Table A-4: Uptake Model Parameters for the HOCl + HCl Reaction ...........................................62
FIGURES
Figure 1: Symmetric and Asymmetric Error Limits ......................................................................13
Figure 2: Recommended Photolysis Quantum Yield for the Formation of O(1D) from O3. .........34
Figure 3: Recommended Reactive Uptake Coefficients as a Function of Temperature for Key
Stratospheric Heterogeneous Processes on Sulfuric Acid Aerosols. ......................................46
5
INTRODUCTION
This compilation of kinetic and photochemical data is an update to the 12th evaluation prepared
by the NASA Panel for Data Evaluation. The Panel was established in 1977 by the NASA Upper
Atmosphere Research Program Office for the purpose of providing a critical tabulation of the
latest kinetic and photochemical data for use by modelers in computer simulations of stratospheric
chemistry. Table I-1 lists this publications editions:
Table I-1: Editions of this Publication

Edition Reference
1 NASA RP 1010, Chapter 1 (Hudson [1])
2 JPL Publication 79-27 (DeMore et al. [93])
3 NASA RP 1049, Chapter 1 (Hudson and Reed [2])
13 JPL Publication 00-3 (Sander et al. [277])
Panel members and their major responsibilities are listed in Table I-2.
Table I-2: Panel Members and their Major Responsibilities for the Current Evaluation
Panel Members Responsibility
S. P. Sander, Chairman Halogen chemistry
M. J. Kurylo
W. B. DeMore HOx chemistry, NOx chemistry
R. R. Friedl
R. F. Hampson
D. M. Golden Three-body reactions, equilibrium constants
R. E. Huie Aqueous chemistry, thermodynamics
C. E. Kolb Heterogeneous chemistry
A. R. Ravishankara O(1D) chemistry
Photochemical data
M. J. Molina Photochemical data
G. K. Moortgat
6
As shown above, each Panel member concentrates his efforts on a given area or type of
data. Nevertheless, the Panels final recommendations represent a consensus of the entire Panel.
Each member reviews the basis for all recommendations, and is cognizant of the final decision in
every case.
Communications regarding particular reactions may be addressed to the appropriate panel

member:
S. P. Sander D. M. Golden
R. R. Friedl PS-031
Jet Propulsion Laboratory SRI International
M/S 183-901 333 Ravenswood Ave.
4800 Oak Grove Drive Menlo Park, CA 94025
Pasadena, CA 91109 [email protected]
[email protected]
[email protected]
R. E. Huie A. R. Ravishankara
M. J. Kurylo NOAA-ERL, R/E/AL2
National Institute of Standards and 325 Broadway
Technology Boulder, CO 80303
Physical and Chemical Properties Division [email protected]
Gaithersburg, MD 20899
[email protected]
[email protected]
C. E. Kolb M. J. Molina
Aerodyne Research Inc. Department of Earth, Atmospheric, and
45 Manning Rd. Planetary Sciences
Billerica, MA 01821 and Department of Chemistry
[email protected] Massachusetts Institute of Technology
Cambridge, MA 02139
[email protected]
G. K. Moortgat
Max-Planck-Institut fr Chemie
Atmospheric Chemistry Division
Postfach 3060
55020 Mainz
Germany
[email protected]
7
Basis of the Recommendations
The recommended rate data and cross sections are based on laboratory measurements. In order to
provide recommendations that are as up-to-date as possible, preprints and written private
communications are accepted, but only when it is expected that they will appear as published
journal articles.
Under no circumstances are rate constants adjusted to fit observations of atmospheric
concentrations. The Panel considers the question of consistency of data with expectations based
on the theory of reaction kinetics, and when a discrepancy appears to exist this fact is pointed out
in the accompanying note.
The major use of theoretical extrapolation of data is in connection with three-body
reactions, in which the required pressure or temperature dependence is sometimes unavailable
from laboratory measurements, and can be estimated by use of appropriate theoretical treatment.
In the case of important rate constants for which no experimental data are available, the panel
may provide estimates of rate constant parameters based on analogy to similar reactions for which
data are available.
Scope of the Evaluation

In the past, the NASA Panel on Data Evaluation reviewed the entire set of reactions presented in
the previous compilations, updating the recommendations and increasing the scope of the review
in response to changes in the published literature [2,84-93,155].
For the current release, the Panel has focused on a selected subset of the kinetic and
photochemical parameters presented in the JPL 97-4 evaluation [92]. The most important criterion
which guided the scope of the present evaluation was an analysis of the sensitivities and
uncertainties of reactions with respect to ozone depletion.
Guidance in this selection was obtained from several recent sensitivity analysis studies
including those of Dubey et al. [102], Thompson and Stewart [316] and Chen et al. [59]. The
reaction lists from these studies were used to identify those processes which play a particularly
important role in ozone depletion calculations. Reactions were selected for inclusion (somewhat
subjectively) if there were significant uncertainties in the laboratory data or if significant time had
elapsed since the last evaluation.
Another important criterion was interpretating atmospheric field measurements. For
example, the OH + NO2 reaction has a significant effect on the ratio of NOx to NOy, which is
measured by high precision aircraft instruments. On this basis, this reaction and several others
were included in this update.
Because of the significant impact of heterogeneous reactions in the polar and mid-latitude
lower stratosphere and rapid progress in laboratory investigations of these processes, several
heterogeneous reactions were included in the present evaluation.
We currently lack guidance from multi-dimensional model sensitivity analyses as to which
heterogeneous processes contribute the largest degrees of uncertainty to current stratospheric
chemistry models. However, available box model calculations [102] indicate that uncertainties in
heterogeneous reactions can lead to significant uncertainties in calculated ozone levels.
8
The six reactions identified as key heterogeneous processes most often included in current
stratospheric photochemical models are listed below:
N2O5 + H2O 2 HNO3
ClONO2 + H2O HOCl + HNO3
ClONO2 + HCl Cl2 + HNO3
HOCl + HCl Cl2 + H2O
BrONO2 + H2O HOBr + HNO3
HOBr + HCl BrCl + H2O
While each of these six reactions occurs to a greater or lesser extent on the full range of
stratospheric aerosol surfaces, we have restricted this review to the three most frequently studied,
and/or believed to be the most likely present in the stratosphere:
Water ice,
Nitric acid trihydrate, and
Liquid sulfuric acid/water mixtures (typically ~40 to 80 wt.% H2SO4).
This selection of aerosol surface compositions covers those found in most current
stratospheric models.
Format of the Evaluation

Changes or additions to the tables of data are indicated by shading. A new entry is completely
shaded, whereas a changed entry is shaded only where it has changed. In some cases only the
note has been changed, in which case the corresponding note number in the table is shaded. In
the Photochemistry section, changed notes are indicated by shading of the note heading.
Computer Access
To ensure universal availability, this document is available online as a file in Adobe PDF
(Portable Data File), Microsoft Word, and Postscript format. Files may be downloaded from
http://jpldataeval.jpl.nasa.gov/. This document is not available in printed form from JPL.
Individuals who wish to receive notice when the web page is revised should submit email
addresses in the appropriate reply box on the web page.
For more information, query Stanley Sander ([email protected]).
Data Formats
In Table 1 (Rate Constants for Second Order Reactions) the reactions are grouped into the classes
O(1D), HOx, NOx, Hydrocarbon Reactions, ClOx and BrOx. The data in Table 2 (Rate Constants
for Association Reactions) are presented in the same order as the bimolecular reactions. The
presentation of photochemical cross section data follows the same sequence.
Units
The rate constants are given in units of concentration expressed as molecules per cubic centimeter
and time in seconds. Thus, for first-, second-, and third-order reactions the units of k are s-1, cm3
molecule-1 s-1, and cm6 molecule-2 s-1, respectively. Cross sections are expressed as cm2
molecule-1, base e.
9
BIMOLECULAR REACTIONS
Some of the reactions in Table 1 are actually more complex than simple two-body reactions. To
explain the pressure and temperature dependences occasionally seen in reactions of this type, it is
necessary to consider the bimolecular class of reactions in terms of two subcategories, direct
(concerted) and indirect (nonconcerted) reactions.
A direct or concerted bimolecular reaction is one in which the reactants A and B proceed to
products C and D without the intermediate formation of an AB adduct that has appreciable
bonding, i.e., there is no bound intermediate; only the transition state (AB) lies between
reactants and products.
A + B (AB) C + D
The reaction of OH with CH4 forming H2O + CH3 is an example of a reaction of this class.
Very useful correlations between the expected structure of the transition state [AB] and the
A-Factor of the reaction rate constant can be made, especially in reactions that are constrained to
follow a well-defined approach of the two reactants in order to minimize energy requirements in
the making and breaking of bonds. The rate constants for these reactions are well represented by
the Arrhenius expression k = A exp(-E/RT) in the 200-300 K temperature range. These rate
constants are not pressure dependent.
The indirect or nonconcerted class of bimolecular reactions is characterized by a more
complex reaction path involving a potential well between reactants and products, leading to a
bound adduct (or reaction complex) formed between the reactants A and B:
A + B [AB]* C + D
The intermediate [AB]* is different from the transition state [AB], in that it is a bound
molecule which can, in principle, be isolated. (Of course, transition states are involved in all of
the above reactions, both forward and backward, but are not explicitly shown.) An example of
this reaction type is ClO + NO, which normally produces Cl + NO2. Reactions of the
nonconcerted type can have a more complex temperature dependence and can exhibit a pressure
dependence if the lifetime of [AB]* is comparable to the rate of collisional deactivation of [AB]*.
This arises because the relative rate at which [AB]* goes to products C + D vs. reactants A + B is
a sensitive function of its excitation energy. Thus, in reactions of this type, the distinction
between the bimolecular and termolecular classification becomes less meaningful, and it is
especially necessary to study such reactions under the temperature and pressure conditions in
which they are to be used in model calculation, or, alternatively, to develop a reliable theoretical
basis for extrapolation of data.
The rate constant tabulation for second-order reactions (Table 1) is given in Arrhenius
form:
10
k(T) = A exp ((-E/R)(1/T)) and contains the following information:
Reaction stoichiometry and products (if known). The pressure dependences are included,
where appropriate.
Arrhenius A-factor.
Temperature dependence and associated uncertainty ("activation temperature" E/RE/R).
Rate constant at 298 K.
Uncertainty factor at 298 K.
Note giving basis of recommendation and any other pertinent information.
Uncertainty Estimates
For second-order rate constants in Table 1, an estimate of the uncertainty at any given temperature
may be obtained from the following expression:
E 1 1
f(T) = f(298) exp
R ( T
-
298 )
Note that the exponent is an absolute value. An upper or lower bound (corresponding
approximately to one standard deviation) of the rate constant at any temperature T can be
obtained by multiplying or dividing the value of the rate constant at that temperature by the factor
f(T). The quantity f(298) is the uncertainty in the rate constant at 298 K. E/R is related to the
uncertainty in the Arrhenius activation energy but is not strictly the one standard deviation
uncertainty in the Arrhenius temperature coefficient. Rather, it has been defined in this evaluation
for use with f(298) in the above expression to obtain the rate constant uncertainty at different
temperatures.
This approach is based on the fact that rate constants are almost always known with
minimum uncertainty at room temperature. The overall uncertainty normally increases at other
temperatures, because there are usually fewer data and it is almost always more difficult to make
measurements at other temperatures. It is important to note that the uncertainty at a temperature T
cannot be calculated from the expression exp(E/RT). The above expression for f(T) must be
used to obtain the correct result.
The uncertainty represented by f(T) is normally symmetric; i.e., the rate constant may be
greater than or less than the central value, k(T), by the factor f(T). In a few cases in Table 1
asymmetric uncertainties are given in the temperature coefficient. For these cases, the factors by
which a rate constant is to be multiplied or divided to obtain, respectively, the upper and lower
limits are not equal, except at 298 K where the factor is simply f(298 K). Explicit equations are
given below for the case where the temperature dependence is (E/R +a, -b):
For T > 298 K, multiply by the factor
f(298 K)e[a(1/298-1/T)]
and divide by the factor
f(298 K)e[b(1/298-1/T)]
For T < 298 K, multiply by the factor
11
f(298 K)e[b(1/T-1/298)]
and divide by the factor

f(298 K)e[a(1/T-1/298)]
Examples of symmetric and asymmetric error limits are shown in Figure 1.

The assigned uncertainties represent the subjective judgment of the Panel. They are not
determined by a rigorous, statistical analysis of the database, which generally is too limited to
permit such an analysis. Rather, the uncertainties are based on a knowledge of the techniques, the
difficulties of the experiments, and the potential for systematic errors.
There is obviously no way to quantify these "unknown" errors. The spread in results among
different techniques for a given reaction may provide some basis for an uncertainty, but the
possibility of the same, or compensating, systematic errors in all the studies must be recognized.
Furthermore, the probability distribution may not follow the normal Gaussian form. For
measurements subject to large systematic errors, the true rate constant may be much further from
the recommended value than would be expected based on a Gaussian distribution with the stated
uncertainty. As an example, in the past the recommended rate constants for the reactions HO2 +
NO and Cl + ClONO2 changed by factors of 30-50. These changes could not have been allowed
for with any reasonable values of in a Gaussian distribution.
12
Figure 1: Symmetric and Asymmetric Error Limits
13
Table 1: Rate Constants for Second Order Reactions
(Shaded areas indicate changes or additions since the last version)
Reaction A-Factora E/RE/R k(298 K)a f(298)b Notes
O(1D) Reactions
-10 -10
O(1D) + H2O OH + OH 2.2x10 0100 2.2x10 1.2 A6
-11 -11
O(1D) + N2O N2 + O2 4.9x10 0100 4.9x10 1.3 A7
-11 -11
NO + NO 6.7x10 0100 6.7x10 1.3 A7
HOx Reactions
O + HO2 OH + O2 3.0x10-11 -(20050) 5.9x10-11 1.1 B2
OH + O3 HO2 + O2 1.5x10-12 880100 7.8x10-14 1.2 B6
OH + HO2 H2O + O2 4.8x10-11 -(250100) 1.1x10-10 1.3 B10
HO2 + O3 OH + 2O2 2.0x10-14 680 +200/-100 2.0x10-15 1.2 B12
NOx Reactions
O + NO2 NO + O2 5.6x10-12 -(18050) 1.0x10-11 1.1 C1
OH + HNO3 H2O + NO3 (See Note) 1.2 C9
NO + O3 NO2 + O2 3.0x10-12 1500200 1.9x10-14 1.1 C20
ClOx Reactions
O + ClO Cl + O2 3.0x10-11 -(7070) 3.8x10-11 1.15 F1
OH + ClO Cl + HO2 7.4x10-12 -(270100) 1.8x10-11 1.4 F10
HCl + O2 3.2x10-13 -(320150) 9.5x10-13 3.0 F10
OH + HCl H2O + Cl 2.6x10-12 350100 8.0x10-13 1.1 F12
Cl + O3 ClO + O2 2.3x10-11 200100 1.2x10-11 1.15 F49
Cl + CH4 HCl + CH3 9.6x10-12 136075 1.0x10-13 1.05 F55
BrOx Reactions
BrO + ClO Br + OClO 9.5 x10-13 -(550150) 6.0x10-12 1.25 G36
Br + ClOO 2.3x10-12 -(260150) 5.5x10-12 1.25 G36
BrCl + O2 4.1x10-13 -(290150) 1.1x10-12 1.25 G36
a Units are cm3 molecule-1 s-1

b f(298) is the uncertainty factor at 298 K. To calculate the uncertainty at other
temperatures, use the expression:
E 1 1
f (T ) = f (298) exp
R T 298
Note that the exponent is absolute value.
14
Notes to Table 1
A6. O(1D) + H2O. Measurements of the O2 + H2 product yield were made by Zellner et al. [356]
(1 +0.5 or -1)% and by Glinski and Birks [117] (0.6 +0.7 or -0.6)%. The yield of O(3P) from
O(1D) + H2O is reported to be <(4.93.2)% by Wine and Ravishankara [346] and (21)% by
Takahashi et al. [312].
To calculate the rates of OH production via O(1D) reactions in the atmosphere, the
quantities of interest are the ratios of the rate coefficients for the reaction of O(1D) with H2O
to those with N2 and with O2. The ratios of the rate coefficients for O(1D) reactions measured
using the same method (and often the same apparatus) are more accurate (and precise) than
the individual rate constants that are quoted in Table 1. Ratio data are given in the original
references for this reaction.
A7. O(1D) +N2O. The branching ratio for the reaction of O(1D) with N2O to give N2 + O2 or
NO + NO is an average of the values reported by Davidson et al. [78]; Volltrauer et al. [333];
Marx et al. [213] and Lam et al. [190], with a spread in R=k(NO + NO)/k(Total) = 0.52 -
0.62. Cantrell et al. [55] reported a measurement of R=0.57 and an analysis of all
measurements from 1957-1994 leads them to recommend a value of R=0.610.06, where
the uncertainty indicates their 95% confidence interval. The recommended branching ratio
agrees well with earlier measurements of the quantum yield from N2O photolysis (Calvert
and Pitts [52]). The O(1D) translational energy and temperature dependence effects are not
clearly resolved. Wine and Ravishankara [346] have determined that the yield of O(3P) from
O(1D) + N2O is <4.0%. The uncertainty for this reaction includes factors for both the overall
rate coefficient and the branching ratio. A direct measurement by Greenblatt and
Ravishankara [119] of the NO yield from the O(1D) + N2O reaction in synthetic air agrees
very well with the value predicted using the recommended O(1D) rate constants for N2, O2,
and N2O and the O(1D) + N2O product branching ratio. These authors suggest that their
results support the recommendations and reduce the uncertainty in the collected rate
parameters by over a factor of two.
To calculate the rates of NO production via O(1D) reactions in the atmosphere, the
quantities of interest are the ratios of the rate coefficients for the reaction of O(1D) with N2O
to those with N2 and with O2. The ratios of the rate coefficients for O(1D) reactions
measured using the same method (and often the same apparatus) are more accurate and
precise than the individual rate constants that are quoted in Table 1. Ratio data are given in
the original references for this reaction.
B2. O + HO2. The recommended values are based on the results of studies over a range of
temperatures by Keyser [170] and Nicovich and Wine [236] and the room temperature
studies of Sridharan et al. [301], Ravishankara et al. [264], and Brune et al. [39]. Earlier
studies by Hack et al. [121] and Burrows et al. [44,46] are not considered, because the OH +
H2O2 reaction was important in these studies and the value used for its rate constant in their
analyses has been shown to be in error. Data from Ravishankara et al. [264] at 298 K show
no dependence on pressure between 10 and 500 torr N2. The ratio k(O + HO2)/k(O + OH)
measured by Keyser [171] agrees with the rate constants recommended here. Sridharan et al.
15
[299] showedthat the reaction products correspond to abstraction of an oxygen atom from
HO2 by the O reactant. Keyser et al. [174] reported <1% O2(1) yield.
B6. OH + O3. Recommended values are based on the results of studies over a range of
temperatures by Ravishankara et al. [263] and Smith et al. [294] and the room temperature
measurements of Kurylo [187], Zahniser and Howard [352], and Kulcke et al. [186]. In the
study of Kulcke et al. [186] the two rate constants sum, k(OH + O3) + k(HO2 + O3), which is
dominated by the first term, was determined to be (8.4 0.8) x 10-14 cm3 molecule-1 s-1.
B10. OH + HO2. A study by Keyser [173] appears to resolve a discrepancy between low-pressure
discharge flow experiments that all gave rate coefficients near 7 x 10-11 cm3 molecule-1 s-1 :
Keyser [169], Thrush and Wilkinson [318], Sridharan et al. [300,302], Temps and Wagner [315],
and Rozenshtein et al. [274], and atmospheric pressure studies that gave rate coefficients near
11 x 10-11: Lii et al. [201], Hochanadel et al. [148], DeMore [82], Cox et al. [71], Burrows et
al. [45], and Kurylo et al. [189]. Laboratory measurements using a discharge flow experiment
and a chemical model analysis of the results by Keyser [173] demonstrate that the previous
discharge flow measurements were probably subject to interference from small amounts of
O and H. In the presence of excess HO2 these atoms generate OH and result in a rate
coefficient measurement that falls below the true value.
The temperature dependence is from Keyser [173], who covered the range 254 to 382 K.
A flow tube study by Schwab et al. [283] reported k = (8.0 +3/-4) x 10-11. These workers
measured the concentrations of HO2, OH, O, and H and used a computer model of the
relevant reactions to test for interference. A flow tube study by Dransfeld and Wagner [100]
18
employing an isotope labelled OH reactant obtained k = (112) x 10-11 in good agreement
with the recommendation. They attributed about half of the reactive events to isotope
16
scrambling because control experiments with OH gave k = 6 x 10-11. It should be noted
that their control experiments were subject to the errors described by Keyser [173] due to the
18
presence of small amounts of H and O, whereas their OH measurements were not. Kurylo
et al. [189] found no evidence of significant scrambling in isotope studies of the OH and HO2
reaction. An additional careful study of the reaction temperature dependence would be
useful. Hippler and Troe [146] have analysed data for this reaction at temperatures up to
1250K. In summary, this has historically been a difficult reaction to study. Earlier
problems appear to have been resolved, as discussed above, and results now tend to
converge on a central value, but the recommended value is still subject to a large
uncertainty.
B12. HO2 + O3. The recommended values are based on results of studies over a range of
temperatures by Zahniser and Howard [352] at 245 to 365 K, Manzanares et al. [210] at 298
K, Sinha et al. [291] at 243 to 413 K, Wang et al. [337] at 233 to 400 K and the steady-state
photolysis experiments of DeMore [80] at 231 to 334 K. The temperature dependence
studies show varying degrees of curvature in the Arrhenius plots, with the E/R decreasing at
lower temperature. However, the fit to data in the stratospherically relevant 230 K to 300 K
range is not significantly affected by inclusion of higher temperature data in the
recommendation. Additional temperature dependence data are needed for this reaction to
more fully characterize the non-linear behavior of the rate constant at low temperature. The
18
mechanism of the reaction has been studied using O labelled HO2 by Sinha et al. [291], who
16
reported that the reaction occurs 7510% via H atom transfer at 297K and by Nelson and
Zahniser [233], who reported branching ratios for H transfer vs O transfer over the range
226-355K. They report that the H atom transfer decreases from 945% at 22611K to
885% at 3558K.
C1. O + NO2. The recommended values are based on the results of studies over a range of
temperatures by Gierczak et al. [116], Ongstad and Birks [239], Slanger et al. [292] and Geers-
Muller and Stuhl [114] and the room temperature study of Paulson et al. [246]. In the most
recent study of Gierczak et al. [116], special emphasis was placed on accurate measurement
of the NO2 concentration and on measurements at low temperatures. The results of earlier
studies by Davis et al. [79] and Bemand et al. [28] were not used in deriving the
recommended values either because of possible complications from decomposition of NO2
at higher temperatures or lack of direct NO2 detection.
C9. OH + HNO3. The recent study of Brown et al. [38] furnishes the most comprehensive set of
rate measurements for N2 as the bath gas over a significant range of temperature (200-350 K)
and pressure (20-500 torr). They analyzed their results in terms of the mechanism proposed by
Smith et al. [294], involving the formation of a bound, relatively long-lived HOHNO3 complex,
as well as the direct reaction channel. Thus, the P dependence can be represented by
combining a low pressure (bimolecular) limit, k0, with a Lindemann-Hinshelwood
expression for the p-dependence:
k o = 2.4 x1014 exp(460/T)
k [M]
k([M],T)=k o + 3 k 2 = 2.7 x10 17 exp(2199 / T)
k [M]
1+ 3 k = 6.5 x10 34 exp(1335/T)
k2 3
The coefficients k3 and k2 are the termolecular and high pressure limits for the "association"
channel. The value of k at high pressures is the sum k0 + k2.
This expression for k([M],T) and the values of the Arrhenius parameters for k0, k2, and
k3 derived by Brown et al. [38] for N2 as the bath gas constitute the recommended values for
this rate coefficient. These recommended values are derived from a fit to the data of Brown
et al. [38], Stachnik et al. [303], Devolder et al. [96] and Margitan and Watson [212].
The reaction as written assumes that the yield of NO3 (per OH removed) is unity at all
temperatures and that the yield has the same value for both reaction channels. However, a
reanalysis of previously published data leads to NO3 yields of 0.75 at 298 K and 0.89 at 251
K (Brown et al. [38]). Better quantification of this yield as a function of pressure and
temperature is needed.
17
C20. NO + O3. The recommended values are based on the results of studies over a range of
temperatures by Birks et al. [33], Lippmann et al. [204], Ray and Watson [267], Michael et al.
[216], Borders and Birks [34] and Moonen et al. [228] and the room temperature studies of
Stedman and Niki [306] and Bemand et al. [28]. The six temperature-dependent studies were
given equal weighting in the recommendation by averaging over the E/Rs from each
individual data set. Following the Moonen et al. recommendation, the 200 K data point
from their study has been excluded from the fit. All of the temperature dependence studies
show some curvature in the Arrhenius plot at temperatures below 298K. Increasing scatter
between the data sets is evident at the lower temperatures. Clough and Thrush [63], Birks et
al., Schurath et al. [282], and Michael et al. have reported individual Arrhenius parameters
for the two primary reaction channels producing ground and excited molecular oxygen.
F1. O + ClO. There have been five studies of this rate constant over an extended temperature
range using a variety of techniques: Leu [194]; Margitan [211]; Schwab et al. [284]; Ongstad
and Birks [239]; and Nicovich et al. [237]. The recommended value is based on a least
squares fit to the data reported in these studies and in the earlier studies of Zahniser and
Kaufman [353] and Ongstad and Birks [238]. Values reported in the early studies of Bemand
et al. [27] and Clyne and Nip [64] are significantly higher and were not used in deriving the
recommended value. Leu and Yung [200] were unable to detect O2(1) or O2( ) and set
1
upper limits to the branching ratios for their production of 4.4 x 10 and 2.5 x 10-2
-4
respectively.
F10. OH + ClO. The reaction has two known product channels under atmospheric conditions:
OH + ClO Cl + HO2 and OH + ClO HCl + O2. Most studies measure the rate
coefficients for the overall reaction (OH + ClO products) which is presumably the sum of
the two channels. The recommendation for the Cl + HO2 channel is obtained from the
difference between a critical assessment of the measurements of the overall reaction and the
recommendation for the HCl + O2 channel. The assessment of the overall reaction (OH +
ClO products) is based on a fit to the 219-373 K data of Hills and Howard [144], the 208-
298 K data of Lipson et al. [206], the 234-356 K data of Kegley-Owen et al. [167] and the 298
K data of Poulet et al. [252]. Data reported in the studies of Burrows et al. [50], Ravishankara
et al. [260], and Leu and Lin [197] were not used in deriving the recommended value because
ClO was not measured directly in these studies and the concentration of ClO was
determined by an indirect method. The fraction of total reaction yielding HO2 + Cl as
products has been determined by Leu and Lin (>0.65); Burrows et al. (0.850.2); Hills and
Howard (0.860.14); and Poulet et al. (0.980.12). Temperature-dependent rate constants
for the HCl + O2 channel have been directly measured by Lipson et al. [205] by observing
the HCl product and the recommendation is based on this work.
18
F12. OH + HCl. The recommended value is based on a least squares fit to the data over the
temperature range 240-300 K reported in the studies by Molina et al. [225], Keyser [172],
Ravishankara et al. [265] and Battin-Leclerc et al. [23]. In these studies particular attention
was paid to the determination of the absolute concentration of HCl by UV and IR
spectrophotometry. Earlier studies by Takacs and Glass [308], Zahniser et al. [354], Smith
and Zellner [298], Ravishankara et al. [261], Hack et al. [120], Husain et al. [156], Cannon et al.
[53], Husain et al. [157], and Smith and Williams [297] had reported somewhat lower room
temperature values. The data of Sharkey and Smith [288] over the temperature range 138-
216 K and Battin-Leclerc et al. [23] below 240 K depart from normal Arrhenius behavior. It
is unknown whether this is due to an effect such as tunneling at low temperature or a
systematic experimental error. Additional work at low temperature is needed.
F49. Cl + O3. The results reported for k(298 K) by Watson et al. [342], Zahniser et al. [355],
Kurylo and Braun [188], Clyne and Nip [65], Nicovich et al. [235] and Seeley et al. [287] are in
good agreement, and have been used to determine the preferred value at this temperature.
The values reported by Leu and DeMore [196] (due to the wide error limits) and Clyne and
Watson [67] (the value is inexplicably high) are not considered. The six Arrhenius
expressions are in fair agreement within the temperature range 205-300 K. In this
temperature range, the rate constants at any particular temperature agree to within 30-40%.
Although the values of the activation energy obtained by Watson et al. and Kurylo and
Braun are in excellent agreement, the value of k in the study of Kurylo and Braun is
consistently (~17%) lower than that of Watson et al. This may suggest a systematic
underestimate of the rate constant, as the values from the other three agree so well at 298 K.
The two most recent studies (Nicovich et al. and Seeley et al.) obtained significantly smaller
temperature dependences than those observed in the earlier studies. There is no reason to
prefer any one set of data to any other; therefore, the preferred Arrhenius expression shown
above was obtained by computing the mean of the six results between 205 and 298 K.
DeMore [83] directly determined the ratio k(Cl + O3)/k(Cl + CH4) at 197-217 K to be within
15% of that calculated from the absolute rate constant values recommended here.
Vanderzanden and Birks [330] have interpreted their observation of oxygen atoms in this
system as evidence for some production (0.1-0.5%) of O2 ( 1 +g ) in this reaction. The
possible production of singlet molecular oxygen in this reaction has also been discussed by
DeMore [81], in connection with the Cl2 photosensitized decomposition of ozone. However
Choo and Leu [60] were unable to detect O2(1) or O2(1) in the Cl + O3 system and set
-4 -2
upper limits to the branching ratios for their production of 5 x 10 and 2.5 x 10 ,
respectively. They suggested two possible mechanisms for the observed production of
oxygen atoms, involving reactions of vibrationally excited ClO radicals with O3 or with Cl
atoms, respectively. Burkholder et al. [42], in a study of infrared line intensities of the ClO
radical, present evidence in support of the second mechanism. In their experiments with
excess Cl atoms, the vibrationally excited ClO radicals produced in the Cl + O3 reaction can
react with Cl atoms to give Cl2 and oxygen atoms, which can then remove additional ClO
radicals. These authors point out the possibility for systematic error from assuming a 1:1
stoichiometry for [Cl]:[O3]o when using the Cl + O3 reaction as a quantitative source of
ClO radicals for kinetic and spectroscopic studies.
19
F55. Cl + CH4. The values of k at 298 K reported from thirteen absolute rate constant studies
(Manning and Kurylo [209], Whytock et al. [343], Michael and Lee [217], Lin et al. [203],
Zahniser et al. [350], Keyser [168], Ravishankara and Wine [262], Heneghan et al. [140], Dobis
and Benson [97], Sawerysyn et al. [279], Beichert et al. [26], Seeley et al. [287], and Pilgrim et
-13 -13
al. [250]) fall in the range (0.92 - 1.13) x 10 , with a mean value of 0.99 x 10 . An earlier
absolute study by Watson et al. [342] gives rate constant values slightly higher than those of
the aforementioned studies, which may be due to uncertainties in correcting the data for OH
loss via reaction with trace levels of ethane and propane in the methane samples used.
The values derived for k at 298 K from the competitive chlorination studies of Pritchard
et al. [255], Pritchard et al. [256], Knox [183], Knox and Nelson [184], Lee and Rowland [191],
-13
and Lin et al. [203] range from (0.8 - 1.6) x l0 when the original data are referenced to the
presently recommended rate constant values for the reactions of Cl with H2 and C2H6. Of
these relative rate studies, that of Lin et al. [203], yields a room temperature rate constant
-13
(1.07 x 10 ) that is most consistent with the absolute measured values. Thus, the
-13
recommended value for k at 298K (1.0 x 10 ) is derived from an unweighted average of the
rate constants from the thirteen preferred absolute studies and the relative rate study of Lin
et al. [203].
There have been nine absolute studies of the temperature dependence of k in which the
measurements extend below 300K (Watson et al. [342], Manning and Kurylo [209], Whytock
et al. [343], Lin et al. [203], Zahniser et al. [350], Keyser [168], Ravishankara and Wine [262],
Heneghan et al. [140], and Seeley et al. [287]). In general, the agreement among most of
these studies is quite good. However, systematic differences in activation energies are
apparent when calculated using data obtained below 300K versus data from above 300K.
Three resonance fluorescence studies have been performed over the temperature region
between 200K and 500K (Whytock et al. [343], Zahniser et al. [350] and Keyser [168]), and in
each case a strong nonlinear Arrhenius behavior was observed. Ravishankara and Wine
[262] also noted nonlinear Arrhenius behavior over a more limited temperature range. This
behavior tends to partially explain the variance in the values of E/R reported between those
investigators who mainly studied this reaction below 300 K (Watson et al. [342], Manning
and Kurylo [209], and Seeley et al. [287])] and those who only studied it above 300 K (Clyne
and Walker [66], Poulet et al. [253], and Lin et al. [203]). The agreement among all studies
below 300 K is reasonably good, with values of E/R ranging from (1063 - 1320)K, and
k(230K) in the range (2.6 - 3.2) x 10-14. There have not been any absolute studies at
stratospheric temperatures other than those which utilized the resonance fluorescence
technique. Ravishankara and Wine [262] have suggested that the results obtained using the
discharge flow and competitive chlorination techniques may be in error at the lower
temperatures (<240 K) due to a non-equilibration of the 2P1/2 and 2P3/2 states of atomic
chlorine. Ravishankara and Wine observed that at temperatures below 240K the apparent
bimolecular rate constant was dependent upon the chemical composition of the reaction
mixture; i.e., if the mixture did not contain an efficient spin equilibrator, e.g., Ar or CCl4,
the bimolecular rate constant decreased at high CH4 concentrations. The chemical
composition in each of the flash photolysis studies contained an efficient spin equilibrator,
whereas this was not the case in the discharge flow studies. However, the reactor walls in
the discharge flow studies could have been expected to have acted as an efficient spin
equilibrator. Consequently, until the hypothesis of Ravishankara and Wine is proven it is
assumed that the discharge flow and competitive chlorination results are reliable. A
20
composite unweighted Arrhenius fit to all of the temperature dependent absolute studies
with data in the temperature region 300K (with the exception of the data of Watson et al.
[342], which appear to be systematically high due to reactive impurities) yields E/R = 1253K
and k(298K) = 1.0 x 10-13.
The competitive chlorination results differ from those obtained in the absolute studies in
that linear Arrhenius behavior is observed. The values of E/R are consistently larger than
those obtained from the absolute studies, with an average value of approximately 1500K.
Until the hypothesis of Ravishankara and Wine [262] is re-examined, the preferred Arrhenius
expression attempts to best fit the results obtained between 200 and 300 K from all sources.
Thus, using the relative rate results of Lin et al. [203] (referenced to the current
recommendation for the Cl + C2H6 reaction) as representative of the relative rate studies
below 300K, together with the composite fit to the absolute studies given above, we obtain a
recommended E/R value of 1360K. Taken with the recommended value for k(298K) = 1.0
x 10-13, we compute an Arrhenius A factor of 9.6 x 10-12. However, the A-factor thus derived
seems somewhat low (on a per hydrogen atom basis) when compared with the A-factors for
some similar reactions.
G36. BrO + ClO. Friedl and Sander [112], using DF/MS techniques, measured the overall rate
constant over the temperature range 220-400 K and also over this temperature range
determined directly branching ratios for the reaction channels producing BrCl and OClO.
The same authors in a separate study using flash photolysis-ultraviolet absorption
techniques (Sander and Friedl [276]) determined the overall rate constant over the
temperature range 220-400 K and pressure range 50-750 torr and also determined at 220 K
and 298 K the branching ratio for OClO production. The results by these two independent
techniques are in excellent agreement, with the overall rate constant showing a negative
temperature dependence. Toohey and Anderson [321], using DF/RF/LMR techniques,
reported room temperature values of the overall rate constant and the branching ratio for
OClO production. They also found evidence for the direct production of BrCl in a
vibrationally excited state. Poulet et al. [251], using DF/MS techniques, reported room
temperature values of the overall rate constant and branching ratios for OClO and BrCl
production. Overall room temperature rate constant values reported also include those from
the DF/MS study of Clyne and Watson [68] and the very low value derived in the flash
photolysis study of Basco and Dogra [21] using a different interpretation of the reaction
mechanism. The recommended Arrhenius expressions for the individual reaction channels
are taken from the study of Friedl and Sander [112] and Turnipseed et al. [327] . These
studies contain the most comprehensive sets of rate constant and branching ratio data. The
overall rate constants reported in these two studies are in good agreement (20%) at room
temperature and in excellent agreement at stratospheric temperatures. Both studies report
that OClO production by channel (1) accounts for 60% of the overall reaction at 200 K.
Both studies report a BrCl yield by channel (3) of about 8%, relatively independent of
temperature. The recommended expressions are consistent with the body of data from all
studies except those of Hills et al. [143] and Basco and Dogra [21].
21
TERMOLECULAR REACTIONS
M
Rate constants for third order reactions (Table 2) of the type A + B ! [AB]* AB are given
in the form:
-n
T
k o (T ) = k
300
o
6 -2 -1
cm molecule s ,
300
300
(where k o has been adjusted for air as the third body), together with a recommended value of n.
Where pressure fall-off corrections are necessary, an additional entry gives the limiting high-
pressure rate constant in a similar form:
-m
T
k (T ) = k
300

3 -1 -1
cm molecule s .
300
To obtain the effective second-order rate constant for a given condition of temperature and
pressure (altitude), the following formula is used:
-1
2
k (T) M
o
1+ log10

k (T )
k ( T ) [M ]

k f ([M],T)= o

0.6
k o ( T ) [M ]
1 +
k (T )
The fixed value 0.6 that appears in this formula fits the data for all listed reactions
adequately, although in principle this quantity may be different for each reaction, and also
temperature dependent.
Thus, a compilation of rate constants of this type requires the stipulation of the four
parameters, ko(300), n, k(300), and m. These can be found in Table 2. The discussion that
follows outlines the general methods we have used in establishing this table, and the notes to the
table discuss specific data sources.
Low-Pressure Limiting Rate Constant [kxo(T) ]

Troe [323] has described a simple method for obtaining low-pressure limiting rate constants. In
essence this method depends on the definition:
k ox ( T ) x k o,sc
x
Here sc signifies "strong" collisions, x denotes the bath gas, and x is an efficiency
parameter (0 < x <1), which provides a measure of energy transfer.
The coefficient x is related to the average energy transferred in a collision with gas x,
x E x
<E>x, via: =
(
1- x ) FE kT
22
Notice that <E> is quite sensitive to . FE is the correction factor of the energy
dependence of the density of states (a quantity of the order of 1.1 for most species of stratospheric
interest).
For some of the reactions of possible stratospheric interest reviewed here, there exist data in
the low-pressure limit (or very close thereto), and we have chosen to evaluate and unify this data
by calculating kox(T) for the appropriate bath gas x and computing the value of x corresponding
to the experimental value [Troe [323]]. A compilation (Patrick and Golden [245]) gives details for
many of the reactions considered here.
From the x values (most of which are for N2, i.e., N2), we compute <E>X according to
the above equation. Values of <E>N2 of approximately 0.3-1 kcal mole-1 are generally expected.
If multiple data exist, we average the values of <E>N2 and recommend a rate constant
corresponding to the N2 computed in the equation above.
Where no data exist we have sometimes estimated the low-pressure rate constant by taking
N2 = 0.3 at T = 300 K, a value based on those cases where data exist.
Temperature Dependence of Low-Pressure Limiting Rate Constants: Tn

The value of n recommended here comes from measurements or, in some cases, a calculation of
<E>N2 from the data at 300 K, and a computation of N2 (200 K) assuming that <E>N2 is
independent of temperature in this range. This N2 (200 K) value is combined with the computed
value of kosc (200 K) to give the expected value of the actual rate constant at 200 K. This latter,
in combination with the value at 300 K, yields the value of n.
This procedure can be directly compared with measured values of ko (200 K) when those
exist. Unfortunately, very few values at 200 K are available. There are often temperature-
dependent studies, but some ambiguity exists when one attempts to extrapolate these down to 200
K. If data are to be extrapolated beyond the measured temperature range, a choice must be made
as to the functional form of the temperature dependence.
23
There are two general ways of expressing the temperature dependence of rate constants.
Either the Arrhenius expression
ko(T) = Aexp(-E/RT) or the form ko(T) = A' T-n
is employed. Since neither of these extrapolation techniques is soundly based, and since they
often yield values that differ substantially, we have used the method explained earlier as the basis
of our recommendations.
High-Pressure Limit Rate Constants [k(T)]

High-pressure rate constants can often be obtained experimentally, but those for the relatively
small species of atmospheric importance usually reach the high-pressure limit at inaccessibly high
pressures. This leaves two sources of these numbers, the first being guesses based upon some
model, and the second being extrapolation of fall-off data up to higher pressures.
Stratospheric conditions generally render reactions of interest much closer to the low-
pressure limit and thus are fairly insensitive to the high-pressure value. This means that while the
extrapolation is long, and the value of k(T) not very accurate, a "reasonable guess" of k(T) will
then suffice. In some cases we have declined to guess since the low-pressure limit is effective
over the entire range of stratospheric conditions.
Temperature Dependence of High-Pressure Limiting Rate Constants: Tm

There are very few data upon which to base a recommendation for values of m. Values in Table 2
are often estimated, based on models for the transition state of bond association reactions and
whatever data are available.
Uncertainty Estimates
For three-body reactions (Table 2) uncertainties are assigned using a procedure that is analogous
to that employed for bimolecular reactions. Uncertainties expressed as increments to ko and k
are given for these rate constants at room temperature. The additional uncertainty arising from
the temperature extrapolation is expressed as an uncertainty in the temperature coefficients n and
m.
24
Table 2: Rate Constants for Association Reactions
Low Pressure Limita High Pressure Limitb

k o (T ) = k 300
o ( T 300 ) k (T ) = k 300 (T 300 )
-n -m
Reaction
k 300
o n k 300
m Notes
M
O + O2 O3 (6.00.5) (-34) 2.40.2 A1
M
O + NO NO2 (9.02.0) (-32) 1.50.3 (3.01.0) (-11) 01.0 C1
M
OH + NO2 HONO2 (2.40.1) (-30) 3.10.2 (1.70.2) (-11) 2.10.3 C4
M
NO2 + NO3 N2O5 (2.00.2) (-30) 4.40.4 (1.40.1) (-12) 0.70.4 C6
M
ClO + NO2 ClONO2 (1.80.3) (-31) 3.40.2 (1.50.4) (-11) 1.90.5 F8
M
ClO + ClO Cl2O2 (2.20.2) (-32) 3.10.2 (3.40.5) (-12) 1.01 F10
M
BrO + NO2 BrONO2 (5.20.4) (-31) 3.20.8 (6.90.4) (-12) 2.90.1 G2
1
2
k (T) M
k o (T) [ M ] 1+ log o
10 k T

k f ([M],T) = ko (T) [M ] 0.6

1+
k (T )

The values quoted are suitable for air as the third body, M.
a Units are cm6 molecule-2 s-1.
b Units are cm3 molecule-1 s-1.
25
Notes to Table 2
A1. O + O2. Temperature dependence re-evaluated. Low pressure limit and T dependence are
an average of Klais, Anderson, and Kurylo [180] and Lin and Leu [202]. The result is in
agreement with most previous work (see references therein) and with the study of Hippler et
al. [Hippler, 1990 #55. Kaye [166] has calculated isotope effects for this reaction, using
methods similar to those discussed in the Introduction (Troe [323], Patrick and Golden [245].)
Croce de Cobos and Troe [75] are in agreement with earlier work. Rawlins et al. [266] report
values in Ar between 80 and 150K that extrapolate to agreement with the recommended
values.
C1. O + NO. Low pressure limit and n from direct measurements of Schieferstein et al. [280]
and their re-analysis of the data of Whytock et al. [344]. Error limits encompass other
studies. High pressure limit and m from Baulch et al. [25] and Baulch et al. [24], slightly
modified. Hippler et al. [145] report higher values for the high pressure limiting rate
constant. Atkinson et al. [12] use Fc =exp(-T/1850). This yields rate constants 10-20%
higher than obtained from Table 2. Shock tube measurements by Yarwood et al. [349] in
argon from 300-1300K are consistent with the values in Table 2.
C4. OH + NO2. The low pressure limit and the high pressure limiting rate constants and their
temperature dependences are from a weighted fit to the data of Anastasi and Smith [8], Wine
et al. [345] and Dohahue et al. [98]. (The values reported in JPL 97-4 used the same data
without weighting, accounting for the different temperature dependence of the low pressure
limit.) Data reported in 1999 by Dransfield et al. [101] and Brown et al. [37] are in excellent
agreement. (Brown et al. report that O2 is about 30% less efficient than N2 as a collider and
suggest that air might therefore have a total efficiency of 0.94 relative to N2) Data from
Anderson et al. [9], Howard and Evenson [151], Burrows et al. [49], and Erler et al. [105] are
in essential agreement. Data of Forster et al [107] and Fulle et al. [113] appear to be too high.
These results and those of Robertshaw and Smith [269], who have measured k in up to 8.6
atmospheres of CF4, suggest that k might be higher than suggested here. This disagreement
might also be due to other causes (i.e., the failure of the simplified fall-off expression as
suggested by Donahue et al., isomer formation, or involvement of excited electronic states).
Burkholder et al. [41] searched for the isomer HOONO and were unable to measure it.
RRKM calculations suggest that the lifetime of the latter species may be too short to
observe under their conditions (248-298 K, 3-850 Torr), but long enough to interfere with
the very high pressure results of Forster et al. [107] and Fulle et al. [113]. The temperature
dependence of both limiting rate constants [345] is consistent with Smith and Golden [296]
and Patrick and Golden [245]. The recommendation here fits all data over the range of
atmospheric interest.
The data of Forster et al. [107] and Fulle et al. [113] may represent the sum of the
pathways forming HONO2 and HOONO. Those experiments were performed on timescales
rapid enough to preclude the dissociation of the HOONO isomer. RRKM calculations
[Golden, D. M., and G. P. Smith submitted to J. Phys. Chem.] calibrated in accord with this
postulate lead to the conclusion that the low temperature laboratory data also measure the
sum of the two pathways. Deconvoluting the data yields a rate constant for nitric acid
formation at stratospheric temperatures that is lower than computed from the Table 2 values.
26
This can be entered into models using parameters that describe the individual steps as given
below.
Fits to individual product species in Table 2 format:
HONO2: ko(300) = 2.40x10-30; n = 2.6; k(300) =1.8x10-11; m = 0.1

HOONO: ko(300) = 2.30x10-31; n = 3.1; k(300) =7.3x10-11; m = 1.3
For the HOONO case, the reverse reaction must be taken into account through the
equilibrium constant. In the format of Table 3:
A=1.2x10-26; B = 8260, K(298K) = 1.3x10-14 and f(298K) = 10
The parameters relating to HOONO are provided for the purposes of model evaluation
only and are not part of the Panels recommendation for reaction C4.
C6. NO2 + NO3. Data with N2 as the bath gas from Kircher et al. [179], Smith et al. [295],
Burrows et al. [48], Wallington et al. [335] and Orlando et al. [243] ranging from 236 to 358K
were used to obtain k0, k, n and m. Values from Croce de Cobos et al. [75] are excluded
due to arguments given by Orlando et al. [243], who point out that a reanalysis of these data
using better values for the rate constant for NO3 + NO 2NO2 yields a negative value for
NO2 + NO3 + M. The study of Fowles et al. [108] is noted, but not used. Johnston et al. [163]
have reviewed this reaction.
A study of the reverse reaction has been carried out by Cantrell et al. [56]. These data are
in excellent agreement with those obtained by Connell and Johnston [69] and Viggiano et al.
[331]. The equilibrium constant recommended in Table 3 is in excellent agreement with the
one given in Cantrell et al. [56], who computed it from the ratio of the rate constant of
Orlando et al. [243] and their rate constants for the reverse reaction.
F8. ClO + NO2. The low pressure limit recommendation and uncertainties are based on
temperature dependent values from Zahniser et al. [351], Lee et al. [193], Birks et al. [32], Leu
et al. [198], Wallington and Cox [336], Cox et al. [70] and Molina et al. [224]. All of these data
were collected in N2 bath gas, except for several points from Lee et al. [193] collected in O2.
The high pressure limit recommendation is based on the RRKM calculations of Smith and
Golden . There are several pressure dependent data sets in the literature, such as Percival et
al. [247], Handwerk and Zellner [122], Dasch et al. [77] and Cox and Lewis [74]; however,
they are too disparate to extract unambiguous values. These data are all reproduced within
two sigma error limits by the current recommendation.
27
F10. ClO + ClO. The recommendation is based on a simultaneous fit to data from Sander et al.
(194 - 247 K) [275], Nickolaisen et al. (260 - 390 K) [234], and Trolier et al. (200 - 263 K)
[324], holding m fixed at 1.0. Latter data have been corrected for the effect of Cl2 as third
body, as suggested by Nickolaisen et al. With this adjustment all the data are in good
agreement. Error limits are from the statistical fit. The ko value for N2 is not in accord with
a Patrick and Golden-type calculation [245]. This may be due to uncertainty in the ClOOCl
thermochemistry, which is based on the equilibrium constants reported by Nickolaisen et al.
and Cox and Hayman [59] (See Table 3). Other previous rate constant measurements, such
as those of Hayman et al. [139], Cox and Derwent [72], Basco and Hunt [22], Walker [334],
and Johnston et al. [164], range from 1-5 x 10-32 cm6 molecule-2 s-1, with N2 or O2 as third
bodies. The major dimerization product is chlorine peroxide (Birk et al. [31], DeMore and
Tschuikow-Roux [94], Slanina and Uhlik [293], Stanton et al. [305] and Lee et al. [192]).
G2. BrO + NO2. Values from a study by Thorn et al. [317] that is in excellent agreement with
Sander et al. [278] are recommended. Error limits are from a reanalysis of the data. Danis et
al. [76] give slightly lower values for the low pressure limiting rate constant and a smaller
temperature dependence as well. This latter study may be hampered by heterogeneous
effects. A theoretical study by Rayez and Destriau [268] suggests that the bond dissociation
energy in BrONO2 is higher than that in ClONO2, thus rationalizing the relative values of the
low pressure limiting rate constants for these two processes. This is confirmed by a more
detailed study by Parthiban and Lee. [244] as well as by Orlando and Tyndall [242], who
measured BrONO2 decomposition and thus an equilibrium constant.
28
EQUILIBRIUM CONSTANTS
Format
Some of the three-body reactions in Table 2 form products which are thermally unstable at
atmospheric temperatures. In such cases the thermal decomposition reaction may compete with
other loss processes, such as photodissociation or radical attack. Table 3 lists the equilibrium
constants, K(T), for several reactions which may fall into this category. The table has three
column entries, the first two being the parameters A and B which can be used to express K(T):
K(T)/cm3 molecule-1 = A exp(B/T) (200 < T < 300 K)
The third column entry in Table 3 is the calculated value of K at 298 K.
The data sources for K(T) are described in the individual notes to Table 3.
Definitions
When values of the heats of formation and entropies of all species are known at the
temperature T, we note that:
o Ho
S T
log 10 [K(T) / cm 3 molecule 1] =
T
+ log10 (T) 21.87
2.303R 2.303RT
Where the superscript "o" refers to a standard state of one atmosphere. In some cases K values
were calculated from this equation, using thermochemical data. In other cases the K values were
calculated directly from kinetic data for the forward and reverse reactions. When available,
JANAF values were used for the equilibrium constants. The following equations were then used
to calculate the parameters A and B:
B/ o K=2.303log10 ( K 200 K 300 ) (300 200 ) (300 200 )
= 1382 log10 ( K 200 K 300 )
log10 A=log10 K ( T ) B ( 2.303T )
The relationships between the parameters A and B and the quantities So(298K) and Ho(298K)
are as follows:
A = (eR'T/Nav) exp(So/R) = 3.7 x 10-22 T exp(So/R)
B = -Ho/R - T(K)
29
Table 3: Equilibrium Constants
Reaction A/cm3 molecule-1 BB/K Keq(298 K) f(298 K)a Note
NO2 + NO3 N2O5 3.0x10-27 10991200 3.1x10-11 1.2 4
ClO + ClO ClOOCl 1.27x10-27 8744500 7.0x10-15 1.3 11
K/cm3 molecule-1 = A exp (B/T) [200 < T/K < 300]

a f(298) is the uncertainty factor at 298 K. To calculate the uncertainty at other
temperatures, use the expression:
1 1
f(T) = f(298 K) exp [ B ( T - 298 ) ].
Notes to Table 3
4. NO2 + NO3. The recommendation is from Pritchard [254] who computed the entropy as a
function of temperature and suggested a Third Law expression. Pritchard [254] examined
the data of Cantrell et al. [54], Burrows et al. [47], Graham and Johnston [118], Wangberg et
al [339], Schott and Davidson [281], and the room temperature data of Tuazon et al. [326],
Perner et al. [249] and Hjorth et al. [147]. He also included the values given by Smith et al.
[295], and Kircher et al. [179], who combined data on the forward reaction, tabulated in Table
2, with decomposition data of by Connell and Johnston [69] and Viggiano et al. [331]. The
recommended value is in excellent agreement with the results of Cantrell et al. [56], who
report rate constants for the decomposition reaction, which they combine with the rate
constants of Orlando et al. [243] to obtain the equilibrium constant. Wangberg et al. [338]
measured the equilibrium constant between 280 and 294K and report results in agreement
with this recommendation.
11. ClO + ClO. The value is from a third-law calculation based on the data from Cox and
Hayman [73] (except for their lowest two points) and Nickolaisen et al. [234]. The entropy of
ClOOCl, the value of which is 72.2 cal mol-1 K-1 at 300K, is calculated from structural and
spectroscopic data given by Birk et al. [31]. The heat of formation at 300K is H0f,300 = 30.8
kcal mol-1. A study of branching ratios of ClO + ClO channels in Cl2/O2/O3 mixtures by
Horowitz et al.[150] also finds the equilibrium constant in O2 at 285 K to be in agreement
with the recommendation. The error limits are chosen to encompass the data.
30
PHOTOCHEMICAL DATA
The photochemical reactions considered in the current evaluation are the following:
O3 + h (1D) + O2
ClOOCl + h products
HOCl + h products
HOBr + h products
Discussion of Format and Error Estimates
Recommended reliability factors for some of the more important photochemical reactions are
given in the Notes. These factors represent the combined uncertainty in cross sections and
quantum yields, taking into consideration the atmospherically important wavelength regions, and
they refer to the total dissociation rate regardless of product identity.
The error estimates are not rigorous numbers resulting from a detailed error propagation
analysis of statistical manipulations of the different sets of literature values; they merely represent
a consensus among the panel members as to the reliability of the data for atmospheric
photodissociation calculations, taking into account the difficulty of the measurements, the
agreement among the results reported by various groups, etc.
The absorption cross sections are defined by the following expression of Beer's Law:
I = Ioexp(-nl),
where Io and I are the incident and transmitted light intensity, respectively; is the absorption
cross section in cm2 molecule-1; n is the concentration in molecule cm-3; and l is the pathlength in
cm. The cross sections are room temperature values at the specific wavelengths listed in the table,
and the expected photodissociation quantum yields are unity, unless otherwise stated.
31
O3 + h O + O2
Quantum yield for production of O(1D)
In the past few years, the quantum yields for the production of O(1D) in the photolysis of O3
in the wavelength range 290 < < 330 nm have been measured using sensitive and selective
methods at 298 K and below [10,14-17,290,309-311,313,314]. The first finding of these studies is that
O(1D) is produced in ozone photolysis at wavelengths longer than 310 nm, which is the threshold
for O(1D) production in photolysis of ground vibrational levels of ozone assuming spin
conservation (i.e., the only co-product of O(1D) is O2(1)). This O(1D) production at wavelengths
longer than 310 nm leads to the existence of the tail. The tail is now known to be due to two
processes: (1) a spin-disallowed channel where O(1D) and O2(3) are produced and (2) the
photolysis of vibrationally excited ozone. The contribution of vibrationally excited ozone
decreases with decreasing temperature such that it is essentially zero by 200 K. These recent
studies have substantially improved our knowledge on the existence of the tail reported in some
earlier studies[11,35,36,325].
The quantum yield in the wavelength range of 248 to 300 nm is recommended to be 0.95,
independent of temperature, with an uncertainty of 5% (1). The recommendation for the
wavelength range of 300 to 345 nm is based on the data of Ball et al.[15], Silvente et al.[290],
Armerding et al.[10], Trolier and Wiesenfeld.[325], Brock and Watson[35,36], Talukdar et
al.[313,314], and Takahashi et al.[309-311] at 298 K. The temperature dependence of the variation
of the quantum yield with wavelength is based on the data of Talukdar et al. [314] and Takahashi
et al [310,311]. The equation used to fit these data between 300 and 340 nm is given below and it
1
attempts to capture the existence of the two unusual O( D) production processes discussed above,
along with the spin-allowed channel above ~310 nm.
To obtain the recommended values given below, all the 298 K data between 300 and 340
nm noted above were first fitted to an equation containing 3 Gaussian functions and a constant
term, using a non-linear least squares method. The highly wavelength resolved data of Takahashi
et al. was binned into 1-nm intervals in this fit. Then, the fit for 298 K was used with the
temperature dependence data of Talukdar and Takahashi et al. (again binned into 1 nm intervals)
to derive the equation for the temperature dependence that is listed below. In this fitting, the
greater-than-unity quantum yields reported by Silvente et al. around 310 nm were excluded.
Also, we did not change any of the normalizations made in individual studies for pinning the
quantum yields to a specific value at a specific wavelength.
( ) 4 T
4
2 ( ) 2
(T , ) = a1 exp 01
+ a2 exp exp
02

1 300 kT 2
( ) 2
+ a3 exp 3 exp + 0.06
03
kT 3
In this equation, is in nm, T in K, and kT is in wavenumbers (i.e., k = 0.695 cm-1 K-1).

This parameterization is valid only for the range 300 330 nm.
32
Table 4 lists the parameters for the O3 quantum yield equation.
Table 4: Parameters for the O3 Quantum Yield Equation
Index ai oi vi i
1 0.887 302 7.9
2 2.35 311.1 820 2.2
3 57 313.9 1190 7.4
For 240 < < 300 nm, we recommend a wavelength and temperature independent value of
0.95. This assumption of constant quantum yield does not significantly alter the calculated O(1D)
production rates in the atmosphere. However, it could lead to errors in laboratory studies. Thus,
for laboratory studies, original literature should be consulted.
There is some evidence for a finite O(1D) quantum yield at > 340 nm. We are
recommending a constant temperature independent value of 0.06 for 330 to 345 nm. However,
the uncertainty in the quantum yields at > 330 nm are much larger than for those below 330 nm.
This constant value of 0.06 is for atmospheric calculations only. Laboratory studies should not
rely on this value for wavelength between 330 and 345 nm.
The uncertainty in the quantum yields between 248 and 300 nm is estimated to be 5%
(1). Between 300 and 320 nm, the uncertainty is estimated to be 20% (1 ) at temperatures
between 200 and 300 K. Between 320 and 340 nm, the uncertainty is estimated to be 25% (1 )
at temperatures between 200 and 300 K.
Figure 2 shows the recommended quantum yield for the formation of O(1D) from the
photodissociation of ozone in the 300-340 nm spectral region over the temperature range 200-300
K. The curves are from the empirical expression in the Note.
33
O3 + h O(1D) + O2
1.0
0.9
0.8
200 K
220 K
240 K
0.7 260 K
280 K
O(1D) Quantum Yield
300 K
0.6
0.5
0.4
0.3
0.2
0.1
0.0
300 310 320 330 340
Wavelength (nm)
1
Figure 2: Recommended Photolysis Quantum Yield for the Formation of O( D) from O3.
34
ClOOCl + h Cl + ClOO
Recommended absorption cross sections in the wavelength range 190-450 nm for

ClOOCl are listed in Table 30 of JPL 97-4 [92]. The values for the wavelength range 200 - 360
nm are the average of experimental results reported by Cox and Hayman [73], DeMore and
Tschuikow-Roux [94], Permien et al. [248], and Burkholder et al. [43]. For the 190 - 200 nm range
the data are from DeMore and Tschuikow-Roux [94], these being the only data available in that
range. Data at wavelengths greater than 360 nm were obtained from a linear extrapolation of the
logarithm of the cross sections, using the expression log[1020 (cm2)] = -0.01915 x (nm) +
7.589. For > 360 nm the extrapolated data are considered to be more reliable than the
experimental measurements because of the very small dimer cross sections in this region.
While the results of Cox and Hayman, DeMore and Tschuikow-Roux, Permien et al., and
Burkholder et al. are in good agreement at wavelengths below 250 nm, there are significant
discrepancies at longer wavelengths, which may be attributed to uncertainties in the spectral
subtraction of impurities such as Cl2O, Cl2 and Cl2O2. Huder and DeMore [154] measured
ClOOCl cross sections over the 190 - 310 nm range using a method that minimized the
corrections required for impurities such as Cl2O. The cross sections from this study are
significantly smaller (up to a factor of 2) than the current recommendation, particularly when
extrapolated beyond 400 nm. Additional measurements are needed, particularly at the longer
wavelengths, to check the results of Huder and DeMore.
These studies also indicate that only one stable species is produced in the recombination
reaction of ClO with itself, and that this species is dichlorine peroxide, ClOOCl, rather than
ClOClO. Using submillimeter wave spectroscopy, Birk et al. [31] have further established the
structure of the recombination product to be ClOOCl. These observations are in agreement with
the results of quantum mechanical calculations (McGrath et al. [214,215]; Jensen and Odershede
[162]; Stanton et al. [305]).
The experiments of Cox and Hayman [73] indicate that the main photodissociation products
at 253.7 nm are Cl and ClOO. Molina et al. [222] measured the quantum yield for this channel
to be unity at 308 nm, with no ClO detectable as a product, with an experimental uncertainty in
of about 25%. Moore et al. [230] measured the quantum yield for the Cl + ClOO channel at 248
and 308 nm to be 0.9 0.1 and 0.1 0.1 for the 2 ClO channel, and thus supported the study of
Molina et al. [222]. These results are also supported by quantum mechanical calculations (Stanton
et al. [305]; Stanton and Bartlett [304]). In contrast, Eberstein [103] suggested a quantum yield of
unity for the production of two ClO radicals, based merely on an analogy with the photolysis of
H2O2 at shorter wavelengths. For atmospheric photodissociation calculations the recommended
quantum yield value is based on the work of Molina et al. [222] and Moore et al. [230], i.e., a
quantum yield 0.90 for the Cl + ClOO channel.
35
HOCl + h OH + Cl
The absorption cross sections of HOCl vapor have been measured by several groups.
Molina and Molina [221] and Knauth et al. [182] produced this species using equilibrium mixtures
with Cl2O and H2O; their results provided the basis for the earlier recommendation. More
recently, Mishalanie et al. [220] and Permien et al. [248] used a dynamic source to generate the
HOCl vapor. The cross section values reported by Molina and Molina [221], Mishalanie et al.
[220], and Permien et al. [248] are in reasonable agreement between 250 and 330 nm. In this
wavelength range, the values reported by Knauth et al. [182] are significantly smaller, e.g., a factor
of 4 at 280 nm. Beyond 340 nm, the cross sections of Mishalanie et al. are much smaller than
those obtained by the other three groups. At 365 nm, the discrepancy is about an order of
magnitude.
Burkholder [40] has remeasured the absorption spectrum of HOCl over the wavelength
range 200 to 380 nm, following photolysis of equilibrium mixtures of Cl2O-H2O-HOCl. The
obtained spectrum displays two absorption maxima at 242 and 304 nm, and is in excellent
agreement with the work of Knauth et al. [182], but in poor agreement with the measurements of
Mishalanie et al. [220] and Permien et al. [248]. The discrepancies can be attributed mostly to
difficulties in correcting the measured absorptions for the presence of Cl2 and Cl2O. In the study
by Burkholder, several control experiments were carried out in order to check the internal
consistency of the data. Moreover, Barnes et al. [19] examined the near-UV spectrum of HOCl by
monitoring the OH fragments resulting from photodissociation, and revealed a third weak band
centered at 387 nm extending down to 480 nm. The recommended cross sections up to 420 nm,
calculated from an analytical expression provided by Barnes et al. [19] and based on the values of
Burkholder [40] and Barnes et al. [19], are listed in Table 5. The work by Jungkamp et al. [165]
yields cross section values in excellent agreement with this recommendation for wavelengths
shorter than 350 nm.
Molina et al. [223] observed production of OH radicals in the laser photolysis of HOCl
around 310 nm, and Butler and Phillips [51] found no evidence for O-atom production at 308 nm,
placing an upper limit of ~0.02 for the primary quantum yield for the HCl + O channel. Vogt and
Schindler [332] used broadband photolysis in the 290 - 390 nm wavelength range, determining a
quantum yield for OH production of >0.95.
36
Table 5: Absorption Cross Sections of HOCl
1020 1020 1020
(nm) (cm2) (nm) (cm2) (nm) (cm2)
200 7.18 274 5.26 348 1.55
202 6.39 276 4.94 350 1.43
204 5.81 278 4.74 352 1.33
206 5.46 280 4.64 354 1.24
208 5.37 282 4.62 356 1.17
210 5.54 284 4.68 358 1.11
212 5.98 286 4.79 360 1.06
214 6.68 288 4.95 362 1.02
216 7.63 290 5.13 364 0.985
218 8.81 292 5.33 366 0.951
220 10.2 294 5.52 368 0.919
222 11.6 296 5.71 370 0.888
224 13.2 298 5.86 372 0.855
226 14.7 300 5.99 374 0.822
228 16.2 302 6.08 376 0.786
230 17.5 304 6.12 378 0.748
232 18.7 306 6.12 380 0.708
234 19.6 308 6.07 382 0.667
236 20.2 310 5.97 384 0.624
238 20.5 312 5.84 386 0.580
240 20.6 314 5.66 388 0.535
242 20.3 316 5.45 390 0.491
244 19.8 318 5.21 392 0.447
246 19.0 320 4.95 394 0.405
248 18.1 322 4.67 396 0.364
250 17.0 324 4.38 398 0.325
252 15.8 326 4.09 400 0.288
254 14.6 328 3.79 402 0.254
256 13.3 330 3.50 404 0.222
258 12.1 332 3.21 406 0.194
260 10.9 334 2.94 406 0.168
262 9.73 336 2.68 410 0.144
264 8.68 338 2.44 412 0.124
266 7.75 340 2.22 414 0.105
268 6.94 342 2.02 416 0.089
270 6.25 344 1.84 418 0.075
272 5.69 346 1.69 420 0.063
37
HOBr + h Products
The absorption spectrum of HOBr has been measured by Orlando and Burkholder [241],
Deters et al. [95], Benter et al. [29], Rattigan et al. [259], and Ingham et al. [159]. The spectra cluster
in two groups. Orlando and Burkholder [241], Deters et al. [95], and Benter et al. [29] observe
between 240 and 400 nm two absorption bands with maxima near 284 and 351 nm; the spectra
agree reasonably well in their shape, but show a sharp decrease in cross section above 400 nm. In
contrast, the cross sections reported by Rattigan et al. [259] and Ingham et al. [159] are roughly 50
% larger between 300 and 400 nm.
In addition, the spectrum obtained by Rattigan et al. shows a pronounced tail extending to
520 nm, whereas Ingham et al. observe unambiguously a third weaker absorption band ranging to
550 nm with a maximum at 457 nm. These last two studies confirm the observations of Barnes et
al. [18], who showed that laser photolysis of HOBr between 440-600 nm gives rise to OH
fragments. The presence of a weak band beyond 400 nm is attributable to the presence of a
forbidden transition from the ground electronic to a triplet state predicted by the ab initio
calculations of Francisco et al. [110]. The differences in the spectral shapes are probably
attributable to impurities such as Br2O and Br2, and/or the use of different Br2O cross sections.
However, the presence of impurities alone cannot explain the large difference in cross sections at
the peak of the absorption bands.
The recommended absorption cross sections are listed in Table 6, in the range from 250 to
550 nm; below 250 nm the data are uncertain and no recommendation is given. The cross section
values in the table are based on the latest study by Ingham et al. [159]. These authors generated
HOBr in situ by laser photolytic production of OH in the presence of Br2, and determined the
HOBr spectrum using a gated diode camera shortly after the pulse, circumventing the problem
associated with the presence of the strong absorbing impurity Br2O, which was encountered in
previous studies. The calibration of the absorption cross sections was made relative to the
established cross sections of Br2.
The data presented in Table 6 are computed with the following expression taken from
Ingham et al. [159], which is based on a combination of three Gaussian fits, one for each
absorption band:
284.01
2
350.57
2
457.38
2
( ) = 24.77 exp 109.80 ln + 12.22 exp 93.63 ln

+ 2.283exp 242.40 ln

(): 10-20 cm2 molecule-1; 250 < < 550 nm.
Benter et al. [29] measured quantum yields for HOBr photolysis at 261 and 363 nm (near the
peaks of the second absorption bands). The observed quantum yield for Br formation at 363 nm
was greater than 0.95, and an unity quantum yield into the product channel OH + Br is
recommended. The other channel O + HBr was not observed. The laser photofragment study of
Barnes et al. [18] claimed that OH was the major photolysis product at wavelengths beyond 400
nm. Lock et al. [207] found that at 490 and 510 nm OH and Br fragments are in their respective
vibrational and spin-orbit ground states. The assumption of unit quantum yield of OH formation
should be confirmed experimentally.
38
Table 6: Absorption Cross Sections of HOBr
20 2 20 2 20 2
(nm) 10 (cm ) (nm) 10 (cm ) (nm) 10 (cm )
250 4.15 355 12.1 460 2.28
255 6.19 360 11.5 465 2.14
260 10.5 365 10.5 470 1.91
265 14.6 370 9.32 475 1.62
270 18.7 375 7.99 480 1.30
275 22.1 380 6.65 485 0.993
280 24.3 385 5.38 490 0.723
285 25.0 390 4.22 495 0.502
290 24.0 395 3.23 500 0.333
295 21.9 400 2.43 505 0.212
300 19.1 405 1.80 510 0.129
305 16.2 410 1.36 515 0.076
310 13.6 415 1.08 520 0.042
315 11.8 420 0.967 525 0.023
320 10.8 425 0.998 530 0.012
325 10.5 430 1.15 535 0.0059
330 10.8 435 1.40 540 0.0029
335 11.3 440 1.68 545 0.0013
340 11.9 445 1.96 550 0.0006
345 12.3 450 2.18
350 12.4 455 2.29
39
HETEROGENEOUS CHEMISTRY
We have evaluated and tabulated the currently available information on heterogeneous

stratospheric processes. In addition, because of the increasing level of interest in tropospheric
processes with a direct bearing on the fluxes of reactive species into the stratosphere, such as
heterogeneous loss processes for partially oxidized degradation products of hydrohalocarbons and
heterogeneous contrail and cloud processing of exhaust species from aircraft, we have included
kinetic data for selected heterogeneous interactions relevant to modeling cloud droplet and
aqueous aerosol chemistry in the free troposphere.
However, both stratospheric and tropospheric heterogeneous chemistry are relatively new
and rapidly developing fields, and further results can be expected to change our quantitative and
even our qualitative understanding on a regular basis. The complexity is compounded by the
difficulty of characterizing the chemical and physical properties of atmospheric heterogeneous
surfaces and then reproducing suitable simulations in the laboratory [185].
Surface Types
To a first approximation there are three major types of surfaces believed to be present at
significant levels in the stratosphere. They are: 1) Type I - polar stratospheric clouds (PSCs),
nominally composed of nitric acid trihydrate (HNO3 3H2O); 2) crystals of relatively pure water
ice, designated as Type II PSCs because they form at lower temperatures than Type I and are
believed to be nucleated by Type I (similar surfaces may form as contrails behind high-altitude
aircraft under some stratospheric conditions); and 3) sulfuric acid aerosol, which is nominally a
liquid phase surface generally composed of 60 - 80 weight percent H2SO4 and, concomitantly,
40-20 weight percent H2O.
While PSCs, as their name suggests, are formed primarily in the cold winter stratosphere at
high latitudes, sulfuric acid aerosol is present year round at all latitudes and may influence
stratospheric chemistry on a global basis, particularly after large injections of volcanic sulfur
episodically increase their abundance and surface area. There is also increasing evidence that
ternary H2SO4/HNO3/H2O liquid solutions may play a significant role in PSC formation.
In addition to the major stratospheric surface types noted above, several other types of
heterogeneous surfaces are found in the stratosphere and may play a significant role in some
stratospheric processes. For instance, recent laboratory work has indicated that nitric acid
dihydrate (NAD) may play an important role in the nucleation of Type I PSCs (Worsnop et al.
[348]; Fox et al. [109]) and that mixtures of solid nitric acid hydrates and sulfuric acid tetrahydrate
(SAT) (Molina et al. [227]; Zhang et al. [360]) and/or a more complex sulfuric acid/nitric acid
hydrate (Fox et al. [109]) may also be key to understanding Type I PSC nucleation and evolution.
Analyses of the range of atmospheric conditions possible in the polar stratosphere have also
led to interest in solid SAT surfaces and possibly other forms of frozen sulfuric acid aerosols
(Toon et al. [322]; Middlebrook et al. [219]), as well as liquid sulfuric acid aerosols significantly
more dilute than the 60-80 weight percent normally present at lower latitudes (Wolff and
Mulvaney [347]; Hofmann and Oltmans [149]; Toon et al. [322]).
Some modeling studies also suggest that certain types of major volcanic eruptions transport
significant levels of sodium chloride into the stratosphere (Michelangeli et al. [218]), so studies of
stratospheric trace species interacting with solid NaCl or similar salts, as well as salt solutions,
have also been included.
40
Finally, aircraft and rocket exhausts contribute small but measurable amounts of
carbonaceous soot (Pueschel et al. [257]) and aluminized solid propellant rocket exhausts and
spacecraft debris produce increasing levels of alumina (Al2O3) and similar metal oxide particles
(Zolensky et al. [361]) in the stratosphere. In the free troposphere the primary heterogeneous
surfaces of interest are liquid or solid water (cloud droplets, contrails) or aqueous sulfate solutions
representative of background aerosols.
The detailed composition and morphology of each surface type are uncertain and probably
subject to a significant range of natural variability. Certain chemical and physical properties of
these surfaces, such as their ability to absorb and/or solvate HCl and HNO3, are known to be
strongly dependent on their detailed chemical composition. Moreover, most heterogeneous
processes studied under laboratory conditions (and in some cases proceeding under stratospheric
conditions) can change the chemical composition of the surface in ways that significantly affect
the kinetic or thermodynamic processes of interest.
Thus, a careful analysis of the time-dependent nature of the active surface is required in the
evaluation of measured uptake kinetics experiments. Experimental techniques which allow the
measurement of mass accommodation or surface reaction kinetics with high time resolution
and/or with low trace gas fluxes are often more credible in establishing that measured kinetic
parameters are not seriously compromised by surface saturation or changing surface chemical
composition.
The measured kinetic uptake parameters, mass accommodation coefficients, and surface
reaction probabilities are separately documented for relevant atmospheric trace gas species for the
major and, where available, the minor stratospheric surfaces noted above.
Since these parameters can vary significantly with surface composition (e.g., the
H2SO4/H2O ratio for sulfate aerosol or the HNO3/H2O ratio for Type I PSC) the dependence of
these parameters on surface composition is reviewed where sufficient data are available.
Furthermore, data are also compiled for liquid water for several reasons.
First, this surface is one asymptote of the H2SO4/H2O aerosol continuum; second, the
interactions of some trace species with liquid water and water ice (Type II PSC) surfaces are often
similar; and third, the uptake of some trace species by water surfaces in the troposphere can play a
key role in understanding their tropospheric chemical lifetimes and, thus, the fraction that may be
transported into the stratosphere.
Surface Porosity
The experimental techniques utilized to measure mass accommodation, heterogeneous
reaction, and other uptake coefficients generally require knowledge of the surface area under
study. For solid surfaces, and most particularly for water and acid ice surfaces formed in situ, the
determination of how the molecular scale ice surface differs from the geometrical surface of the
supporting substrate is not easy.
Keyser, Leu, and coworkers have investigated the structure of water and nitric acid ice
films prepared under conditions similar to those used in their flow reactor for uptake studies
(Keyser et al. [178]; Keyser and Leu [176]; Keyser and Leu [175]). They have demonstrated that ice
films grown in situ from the vapor can have a considerably larger available surface than that
represented by the geometry of the substrate; they have also developed a simple model to
attempt to correct measured uptake rates for this effect (Keyser et al. [178]; Keyser et al. [177]).
41
This model predicts that correction factors are largest for small uptake coefficients and thick
films.
The application of the model to experimental uptake data remains controversial (Keyser et
al. [177]; Hanson and Ravishakara [133]; Kolb et al. [185]). Some experimenters prefer to attempt
growing ice surfaces as smooth as possible and to demonstrate that their measured uptake co-
efficients are only weakly dependent on surface thickness (Hanson and Ravishankara, [132]).
Similar issues arise for uptake experiments performed on powered, fused and single crystal
salt or oxide surfaces (Fenter et al.[106]; Hanning-Lee et al. [123]). The degree to which measured
uptake parameters must be corrected for porosity effects will remain in some doubt until a method
is devised for accurately determining the effective surface area for the surfaces actually used in
uptake studies. Most studies evaluated in this review assume that the effective ice or salt surface
area is the geometrical area.
Temperature Dependence
A number of laboratory studies have shown that mass accommodation coefficients and, to
some extent, surface reaction probabilities can be temperature dependent. While these
dependencies have not been characterized for many systems of interest, temperature effects on
kinetic data are noted where available. More work that fully separates heterogeneous kinetic
temperature effects from temperature controlled surface composition is obviously needed.
Solubility Limitations
Experimental data on the uptake of some trace gases by various stratospherically relevant
surfaces can be shown to be governed by solubility limitations rather than kinetic processes. In
these cases properly analyzed data can yield measurements of trace gas solubility parameters
relevant to stratospheric conditions.
In general, such parameters can be strongly dependent on both condensed phase
composition and temperature. Such parameters may be very important in stratospheric models,
since they can govern the availability of a reactant for a bimolecular heterogeneous process (e.g.,
the concentration of HCl available for the HCl + ClONO2 reaction on sulfuric acid aerosols) or
the gas/condensed phase partitioning of a heterogeneous reaction product (e.g., the HNO3 formed
by the reaction of N2O5 on sulfuric acid aerosols).
Data Organization
Data for selected trace gas surface reaction probabilities with relevant condensed phase
surfaces are tabulated in Table 7. Evaluated data for mass accommodation coefficients and
solubilities are given in Tables 63 and 65 of DeMore et al. [92].
42
Parameter Definitions
Mass accommodation coefficients (), represent the probability of reversible uptake of a
gaseous species colliding with the condensed surface of interest. For liquid surfaces this process
is associated with interfacial (gas-to-liquid) transport and is generally followed by bulk liquid
phase solvation. Examples include: simple surface absorption, absorption followed by ionic
-
dissociation and solvation (e.g., HCl + nH2O H+(aq) + Cl (aq)), and absorption followed by a
reversible chemical reaction with a condensed phase substituent
(e.g., SO2 + H2O H+ + HSO3- or CH2O + H2O CH2(OH)2).
The term "sticking coefficient" is often used for mass accommodation on solid surfaces
where physisorption or chemisorption takes the place of true interfacial mass transport.
Processes involving liquid surfaces are subject to Henrys law, which limits the fractional
uptake of a gas phase species into a liquid. If the gas phase species is simply solvated, a physical
Henrys law constraint holds; if the gas phase species reacts with a condensed phase substituent,
as in the sulfur dioxide or formaldehyde hydrolysis cases noted above, a chemically modified or
effective Henrys law constraint holds (Clegg and Brimblecombe [62]; Schwartz [285]; Watson
et al. [341]). Henry's law constants relate the equilibrium concentration of a species in the gas
phase to the concentration of the same species in a liquid phase, and they have, in this report,
units of M atm-1.
Effective Henry's law constants are designated H*, while simple physical Henry's law
constants are represented by H. Effective Henry's law constants are also employed to represent
decreased trace gas solubilities in moderate ionic strength acid or salt solutions with the use of a
Setchenow coefficient formulation which relates H* to the concentration of the acid or salt [158].
It is presently unclear whether surface solubility effects govern the uptake on nominally solid
water ice or HNO3/H2O ice surfaces in a manner analogous to bulk solubility effects for liquid
substrates.
For some trace species on some surfaces, experimental data suggest that mass
accommodation coefficients untainted by experimental saturation limitations have been obtained.
In other cases experimental data can be shown to be subject to Henrys law constraints, and
Henrys law constants, or at least their upper limits, can be determined. Some experimental data
sets are insufficient to determine if measured uptake coefficients are true accommodation
coefficients or if the measurement values are lower limits compromised by saturation effects.
Surface reaction probabilities () are kinetic values for generally irreversible reactive
uptake of trace gas species on condensed surfaces. Such processes may not be rate limited by
Henrys law constraints; however, the fate of the uptake reaction products may be subject to
saturation limitations. For example, N2O5 has been shown to react with sulfuric acid aerosol
surfaces. However, if the H2SO4/H2O ratio is too high, the product HNO3 will be insoluble, and a
large fraction will be expelled back into the gas phase. Surface reaction probabilities for
substantially irreversible processes are presented in Table 7. Reaction products are identified
where known.
The total experimental uptake coefficient measured in laboratory heterogeneous kinetic
experiments are also often symbolized by the symbol . In those cases where surface and/or bulk
reaction dominate the uptake, the total uptake coefficient (total) and reactive uptake coefficient
(total) may well be identical. More formally, for cases where bulk liquid phase reaction is facile
43
and there are no gas phase diffusion constraints, the total uptake coefficient can be approximated
in terms of rxn and sol as [185]:
1 1 1
= +
+
total sol rxn
where
4HRT D 1 / 2
=
sol 1 / 2 c t
and
4HRT 1/2

rxn
=
c
(Dkrxn )
where t is the time the trace gas is exposed to the liquid surface, R is the gas constant, D is the
liquid phase diffusion coefficient, and c is the mean trace gas molecular speed. In the limit of
low solubility or long exposure time sol becomes negligible and:
1 1 1
= + .

total rxn
Discussion of how to use this approach to model chemical reactions in liquid stratospheric
aerosols can be found in Hanson et al. [138] and Kolb et al. [185]. Note that these formulations are
approximate. In cases where separate terms are competitive, more rigorous solution of kinetic
differential operations may be appropriate.
For solid surfaces, bulk diffusion is generally too slow to allow bulk solubility or bulk
kinetic processes to dominate uptake. For solids, reactive uptake is driven by
chemisorption/chemical reaction at the interface, a process that can also influence trace gas uptake
on liquids. In these cases surface reaction (surf) occurs in parallel, rather than in series with mass
accommodation, thus:
1
1 1
total = surf + +
sol + rxn
Examples where this more complex situation holds for liquid surfaces can be found in Hu et
al. [153] and Jayne et al.[161]. In such cases may be significantly larger than .
The data in Table 7 are organized by trace gas species, since some systematic variation
may be expected for surface accommodation or reaction as the surface composition and/or phase
44
is varied. Data presented for one surface may be judged for reasonableness by comparing with
data for a similar surface. In some cases it is not yet clear if surface uptake is truly reversible
(accommodation) or irreversibly reactive in nature.
Evaluation Process
Publications presenting experimental kinetic parameters for selected reactions in Table 7
appearing since our last evaluation (DeMore et al. [92]) were identified and evaluated.
Laboratories known to be studying these reactions were also contacted and asked to provide
available data prior to publication. All available data, from both new and previously identified
studies were plotted for each reaction/surface combination, as a function of temperature, acid
concentration (for water/acid surfaces), and other pertinent experimental parameters. Data were
evaluated for biases reflecting surface and/or bulk saturation effects.
In general, data from experiments using the lowest possible trace gas concentrations and/or
providing continuously generated fresh surfaces were found to be less affected by surface and/or
bulk saturation, and were retained for rate evaluations. All data were plotted in terms of the
reactive uptake coefficient, , even in cases were can be deconvolved to yield more fundamental
parameters such as mass accommodation coefficients, Henrys law constants, liquid phase
diffusion coefficients, and liquid phase second order rate constants (see Robinson et al. [272] for
examples). Data plots generally incorporated the stated statistical uncertainties reported by the
authors of the publications reporting the data.
Data free from obvious experimental biases were then least squares fit with an appropriate
function, phenomenologically based, if appropriate, and empirical if not. The standard deviation
of the fitting function from the data was then determined to yield an estimate of the statistical
uncertainty. Were appropriate, systematic uncertainty estimates were evaluated and added to the
statistical uncertainty. The results of these evaluations are summarized in Table 7.
Reactions of N2O5, ClONO2, HOCl and BrONO2 on/in sulfuric acid are generally dependent
on the species Henrys law solubility and liquid phase diffusion coefficient in the liquid acid as
well as the surface and/or liquid phase reaction rate parameters. All of these processes are
generally functions of the acid composition and temperature (Hanson et al. [138], Robinson et al.
[272]). Thus, these reactions reactive uptake coefficients must be represented by a complex
phenomenological or empirical model which defy simple entry into Table 7. The notes to Table 7
for these reactions discuss and present the models adopted. However, to aid in visualizing the
resulting reactive uptake parameters the results for several reactions have been plotted in Figure 1
as a function of temperature for a background pressure of 50 mbar and background water vapor
and HCl mixing ratios of 5 ppm and 2 ppb, respectively.
Figure 3 graphs recommended reactive uptake coefficients as a function of temperature for
key stratospheric heterogeneous processes on sulfuric acid aerosols. The calculations assume a
total pressure of 50 mbar, H2O vapor mixing ratio of 5 ppm, and HCl mixing ratio of 2 ppb. For
this calculation, the aerosol reacto-diffusive length for each reaction is assumed to be small and
PHCl>>PClONO2.
45
H2SO4 wt%
50 60 70 80
0
10 BrONO2+H2O
N2O5+H2O
-1
10
ClONO2+HCl
-2
10 p=50 mbar
H2O= 5 ppm
o
HOCl + HCl
HCl= 2 ppb
-3
10
-4
10
ClONO2+H2O
-5
10
190 200 210 220 230
Temperature (K)
Figure 3: Recommended Reactive Uptake Coefficients as a Function of Temperature

for Key Stratospheric Heterogeneous Processes on Sulfuric Acid Aerosols.
46
Table 7: Gas/Surface Reaction Probabilities ()
Gaseous Surface Surface Uncertainty
Species Type Composition
T(K) Factor
Notes
N2O5 + H2O 2HNO3
N2O5 Water Ice H2O(s) 188-195 0.02 2 14
Liquid Water H2O(l) 260-295 See Note* See Note 15
Nitric Acid Ice HNO3 3H2O(s) 200 4 x 10-4 3 16
Sulfuric Acid H2SO4 nH2O(l) 195-300 See Note* See Note 17
HOCl + HCl(s) Cl2 + H2O
HOCl Water Ice H2O(s) HCl(s) 195-200 0.2 2 40

Nitric Acid Ice HNO33H2O(s)HCl(s) 195-200 0.1 2 40
Sulfuric Acid H2SO4nH2O(l) 198-209 See Note* See Note 41
ClONO2 + H2O(s) HOCl + HNO3
ClONO2 Water Ice H2O(s) 180-200 0.3 3 43

Nitric Acid Ice HNO3 3 H2O (s) 185-202 0.004 3 44
Sulfuric Acid H2SO4 n H2O (l) 200-265 See Note* See Note 45
ClONO2 + HCl(s) Cl2 + HNO3
ClONO2 Water Ice H2O(s) 180-200 0.3 3 47

Nitric Acid Ice HNO33 H2O HCl 185-210 0.2 2 48
Sulfuric Acid H2SO4nH2O(l)HCl(l) 195-235 See Note* See Note 49
HOBr + HCl(s) BrCl + H2O
HOBr Water Ice H2O(s) HBr(s) 228 0.3 3 57

Sulfuric Acid H2SO4 n H2O
(60-69 wt.% H2SO4) 198-218 See Note 57
BrONO2 + H2O HOBr + HNO3
BrONO2 Water Ice H2O(s) 190-200 0.3 2 59

Sulfuric Acid H2SO4 nH2O 210-300 See Note See Note 60
* is temperature dependent
47
Notes to Table 7
14. N2O5 + H2O(s) - Leu [195] and Hanson and Ravishankara [129] measured nearly identical
values of 0.028 (0.011) and 0.024 (30%) in the 195-202 K range on relatively thick ice
films in coated wall flow tubes. Quinlan et al. [258] measured a maximum value for on ice
surfaces at 188 K of 0.03 in a Knudsen cell reactor. The average of these three studies is
0.027 with a standard deviation of 0.003. Hanson and Ravishankara [131,132] presented new
and re-analyzed data as a function of ice thickness, with a value of ~0.008 for the thinnest ice
sample, rising to 0.024 for the thickest. From these data there would appear to be no strong
dependence on temperature, at least over the 188-195 K range. It is unclear whether the
measured dependence on ice film thickness is due to added porosity surface area in the
thicker films or decreased ice film integrity in thinner films. The error estimate in Table 7 is
driven by the possible systematic error due to unresolved film thickness effects rather than
the small statistical error among the thick film values from the three groups.
15. N2O5 + H2O(l) - Reaction on liquid water has a negative temperature dependence. Van
Doren et al. [328] measured s of 0.057 0.003 at 271 K and 0.036 0.004 at 282 K using a
droplet train uptake technique. George et al. [115] also used a droplet train technique to
measure s of (3.0 0.2) x 10-2 (262 K), (2.9 1.2) x 10-2 (267 K), (2.0 0.2) x 10-2 (273 K),
(1.6 0.8) x 10-2 (276 K), and (1.3 0.8) x 10-2 (277 K) on pure water, while Schweitzer et al.
[286] used the same approach for pure water and salt solutions between 262 and 278 K,
obtaining similar results. Mozurkewich and Calvert [231] studied uptake on NH3/H2SO4/H2O
aerosols in a flow reactor. For their most water-rich aerosols (RH = 76%) they measured s
of 0.10 0.02 at 274 and 0.039 0.012 at 293 K. However, similar studies by Hu and Abbatt
[152] on (NH3)2SO4 aerosols at 297 K showed that uptake rises with decreasing relative
humidity (RH); their 94% RH results agree very well with the temperature trend measured by
Van Doren et al. Msibi et al.[232] measured a smaller of 2.5 x 10-3 for water adsorbed on a
denuder flow tube well under 66-96% relative humidity conditions at room temperature. The
higher values of Van Doren et al., Mozurkewhich and Calvert, and Hu and Abbatt are quite
consistent when temperature and RH effects are factored in. The lower values from the Louis
Pasteur (George et al.; Schweitzer et al.) and Birmingham (Msibi et al.) groups appear to
have a much less pronounced temperature dependence and are inconsistent with the other
measurements. The same function used to fit the N2O5 uptake on sulfuric acid as a function
of temperature and concentration, discussed in note 17 below, has been extended to the Van
Doren et al. and Hu and Abbatt data for pure water and very high RH aerosols. See note 17
for the functional fit and its error discussion.
16. N2O5 + HNO33H2O(s) - Hanson and Ravishankara [130] have measured = 0.0006 (30%)
near 200 K. They presented re-analyzed and additional data as a function of ice thickness
(Hanson and Ravishankara [131,132]), deriving a value of 3 x 10-4 for the thinnest NAT
covered ice layer, with values up to three times higher for thicker NAT-covered ice layers.
As in the case of uptake on water ice this may be due to increased surface area from porosity
in the thicker films, or less integrity in the thinner films. The uncertainty listed in Table A-1
48
is driven by this observed effect. All of the Hanson et al. data are in very poor agreement
with the = 0.015 (0.006) reported by Quinlan et al. [258] from their Knudsen cell
measurements; this measurement may have been biased by formation of a super-cooled
aqueous nitric acid surface and is judged to be unreliable.
17. N2O5 + H2SO4n H2O (l) - This reaction has been intensively studied between 195 and 296
K for a wide range of H2SO4 wt. % values using four complementary experimental
techniques. Data are available from aerosol flow tube studies (Fried et al. [111], Hanson and
Lovejoy [126], and Hu and Abbatt [152]), coated wall flow tube studies (Hanson and
Ravishankara [129], Zhang et al. [359]), a stirred Knudsen cell (Manion et al. [208]) and droplet
train studies (Van Doren et al. [329], Robinson et al. [272]). All studies have yielded s
between ~0.05 and 0.20 with modest dependence on surface H2SO4 wt. % and temperature.
The Knudsen cell studies, aerosol flow tube studies at higher N2O5 exposure and the ternary
H2SO4/HNO3/H2O studies of Zhang et al. all illustrate that significant levels of HNO3 in the
H2SO4/ H2O solutions will reduce measurably; this fact explains some of the scatter in
aerosol flow tube studies and the surface saturation evident in the Knudsen cell studies. The
effects of 5.0 x 10-7 Torr HNO3 for (N2O5) as a function of temperature at two water vapor
concentrations are plotted in Zhang et al.; the decrease in is greatest at low temperatures,
approaching a factor of 2-5 between 200 and 195 K.
Experimental data on sulfuric acid surfaces between 40 and 80 wt.% sulfuric acid deemed
to be free of saturation effects, plus the pure water uptake data of Van Doren et al. [328] and
high relative humidity ammonium sulfate aerosol uptake data of Hu and Abbott [152] were all
fit to a polynomial expression to yield a single function describing for N2O5 uptake valid
between 0 and 80 wt.% H2SO4 and 180 to 300 K. The form of this function is:
o=exp(ko+k1/T+k2/T2), where T is the temperature K. The parameters ko, k1, and k2 obtained
from the best-fit are:
ko = -25.5265 0.133188wt + 0.00930846wt2 9.0194x10-5wt3

k1 = 9283.76 + 115.345wt 5.19258wt2 + 0.0483464wt3
k2 = -851801- 22191.2wt + 766.916 wt2 6.85427wt3
where wt is the weight percentage of H2SO4.
The overall error of applying the uptake function provided here consists of two components.
One is the standard deviation of model-calculated value with respect to measured data, m,
which is given by
2
N
m = 1 i

i = 1
model .
N 1
49
The other is the standard deviation of relative experimental measurement error from the
mean, d, which is given by
2
N
d = i

i = 1 model

N ( N 1)
The overall error is
2 2
= m + d .
These formulations are also applied below in the error estimation for the ClONO2 + H2O
and HCl, BrONO2 + H2O, and HOCl + HCl reactions. For N2O5, the error is estimated to be
15% (one sigma), with m=14.7% and d=2.9% .
40. HOCl + HCl + H2O(s) and HNO33H2O(s) - Hanson and Ravishankara [131] and Abbatt
and Molina [5] have investigated the HOCl + HCl reaction on water ice and NAT-like
surfaces, and Chu et al. [61] studied the reaction on water ice. Product yield measurements
support the identification of Cl2 and H2O as the sole products. The measured yield of
product Cl2 is 0.87 0.20 and was stated to be similar on both surfaces [Abbatt and Molina].
Within the accuracy of the experiments, the reaction probability does not depend on the gas
phase HCl and HOCl densities. Only Abbatt and Molina investigated at more than one
temperature, their data indicates that increases at lower temperatures. A plot of data from
the three studies does show a weak temperature trend, with increasing about a factor of
two as the temperature drops from 202 to 188 K. However, the data are too sparse to assign
a definitive temperature dependence. The average of all three studies yields = 0.26 0.08
for data based on the geometrical area of the flow tube surfaces. Chu et al. indicate that a
porosity correction for their data would reduce their value by ~2.6. The real uncertainty
would appear to be dominated by systematic uncertainties in porosity corrections and a
potential temperature dependence. Given the fact that any porosity correction must reduce
the value, a central value of 0.2 is adopted with an uncertainty factor of 2. The high reaction
probabilities measured for water ice indicate that this reaction may play a significant role in
release of reactive chlorine from the HCl reservoir.
Two studies [Hanson and Ravishankara, [131]; Abbatt and Molina, [5] have measured the
reaction probability of HOCl + HCl on NAT surfaces. These data show increases as the
ambient water pressure increases and then reaches a plateau. At relatively high water
pressure, the two studies averaged o=0.135 0.049, with no porosity correction. The
reaction probability on water poor NAT-like surfaces falls off dramatically (a factor of 10).
A recommendation of 0.1 with and uncertainty factor of 2 is shown in Table A-1. Carslaw
and Peter [58] have published a model of this reaction and its dependence on HCl uptake.
50
41. HOCl + HCl + H2SO4nH2O - This process has been studied in coated flow tubes over
~200-260 K by Zhang et al. [357], Hanson and Ravishankara [134], Donaldson et al. [99], and
Hanson and Lovejoy [128]. Hanson and Lovejoy also made measurements in an aerosol flow
tube from 251 to 276 K. A model of this and related sulfuric acid aerosol reactions tailored to
stratospheric conditions has been published by Hanson et al. [138]. Zhang et al. held the
water vapor partial pressure at 3.8 x 10-4 torr and showed increased by a factor of 50 as the
temperature was lowered from 209 to 198 K, showing that the reaction rate is strongly
dependent on water activity.
A detailed kinetic uptake model has been developed to fit the experimental data. The
formulation for the is given as:
1 1 1
= + rxn
HOCl
where
4 H HOCl RT
( DHOCl k HOCl _ HCl )
1/ 2
rxn
HOCl =
c
Over the low temperatures of interest, for HOCl was assumed to be unity consistent with
the value for HCl measured at 240 K and below ([271]). The individual formulations for
HHOCl, DHOCl and kHOCl-HCl are given in the Table A-4 below. Reaction of HOCl with HCl is
considered to be acid catalyzed. It is known that the reaction rate for HOCl + HCl in pure
water is low [Donaldson et al. [99]]. Experimental data noted above indicated that the
reaction rate of HOCl + HCl increases with acidicity of H2SO4 solution. The data from the
experimental studies noted above were fit to the model without bias. Using the same error
analysis discussed in note 17 for N2O5, a detailed kinetic model yields a 33.4% error (one
sigma fit to the available data set, with m=33.3% and d=3.0%).
In the cold stratosphere where T<190 K, the reaction of ClONO2 + HCl is so fast that HCl
is depleted which slows down the reaction of HOCl + HCl. As shown in Table A-4, the
effect of HCl depletion on the HOCl reactive uptake coefficient (due to reaction with
ClONO2 inside/on the surface of particles) is taken into account via the factor FHCl (also see
note 49).
43. ClONO2 + H2O(s) - Measurement of = 0.3 (+0.7, -0.1) (Hanson and Ravishankara [130])
significantly exceeds previous measurements of Molina et al. [226], Tolbert et al. [320], Leu
[195] and Moore et al. [229] but agrees reasonably well with subsequent measurements by Chu
et al. [61] and Zhang et al. [358] when geometrical surface areas are assumed for analysis.
Previous measurements were probably complicated by NAT formation on the surface
(Hanson and Ravishankara [131]; Chu et al. [61]). Lower levels of ClONO2 (g) used by
Hanson and Ravishankara [130] minimized this surface saturation problem. Also, using lower
ClONO2 concentrations, Zhang et al. obtained a reaction probability of 0.08 0.02 at 195 K,
in fair agreement with the range of 0.03 to 0.13 measured by Chu et al. More recent Knudsen
cell measurements at 180 and 200 K by Oppliger et al. [240] showed initial uptake s in the
51
0.2 to 0.4 range. Measured reaction products were HNO3 and HOCl. All of the HNO3 and
much of the HOCl is retained on the surface under polar stratospheric conditions (Hanson
and Ravishankara [130,131]). Hanson [124] deposited ClONO2 on H218O enriched ice and
detected H18OCl showing the Cl-ONO2 bond is broken at 191 K.
Data plots confirm a trend showing that at high density of ClONO2, the product HNO3 covers
the ice surface preventing the further reaction of ClONO2 with H2O molecules on the surface.
Therefore, data obtained at high ClONO2 densities (>1014 molecules/cm3) are excluded from
further evaluation. A recent experiment [Berland et al. [30]] using a laser-induced thermal
desorption technique yielded a much lower value of ClONO2 reaction probability at 190K
(about 3 orders of magnitude lower) after extrapolating the results obtained at temperatures of
140 K and below. We also exclude this point in the averaging of data since the physical
characteristics of ice surfaces at these very low temperatures may not be very representative
of those found at stratospheric temperatures. Selected data show no temperature dependence
between T=180 and 200 K and averaged o= 0.28 0.25. Again, within the experimental
accuracy, the Hanson and Ravishankara [131,132] and Chu et al. [61] data show that uptake
measurements are nearly independent of ice substrate thickness. See Henson et al. [141,142]
for discussion of a model which accounts for the effect of HNO3 on the reaction ClONO2 on
water and nitric acid ice surfaces.
44. ClONO2 + HNO3nH2O Hanson and Ravishankara [130] report a value of 0.006 at 201 K
for the ClONO2 reaction with the water on NAT (HNO3nH2O). However, these authors
present re-analyzed and additional data with 0.001 at 191 K in Hanson and Ravishankara
[131,132]. Similar experiments (Moore et al. [229], Leu et al. [199]) report a larger value of
0.02 0.01 which falls very rapidly as slight excesses of H2O above the 3/1 H2O/HNO3 ratio
for NAT are removed. They measure of less than 10-6 for slightly water poor NAT
surfaces. The inconsistency between Hanson and Ravishankara and the JPL group (Moore et
al. [229]; Leu et al., [199]) has not been resolved. Abbatt and Molina [6] report values
reaching 0.002 at 202 K and high RH. Hanson and Ravishankara [131] reported that for this
reaction increases by a factor of 4 as the surface temperature increases from 191 to 211 K.
However, Knudsen cell measurements at 185 K by Barone et al. [20] reported = 0.004 at a
relative humidity (RH) of 100%, rising to 0.007 near RH = 120%, indicating a possible mild
negative temperature dependence when high RH values from this and other studies are
compared. Excluding the JPL data, the other data obtained at high RH (~90%) were
averaged, assuming no temperature dependence, to yield = 0.0043 0.0021. The strong
dependence on RH and the possible temperature dependence suggest that systematic error
probably exceeds the calculated statistical error. Within the experimental accuracy, the data
of Hanson and Ravishankara [131,132] show that measured uptake coefficients are
independent of ice substrate thickness. Barone et al. report very similar uptake coefficients
for nitric acid dihydrate (NAD) as for NAT as a function of RH at 202 K. See Henson et al.
[141,142] for discussion of a model which accounts for the effect of HNO3 on the reaction of
ClONO2 on water and nitric acid ice surfaces.
52
45. ClONO2 + H2SO4nH2O(l) - Results from wetted-wall flow tube [Hanson and
Ravishankara [135]] Knudsen cell reactor [Manion et al. [208]] aerosol flow tube [Hanson
and Lovejoy [127]], and droplet train uptake [Robinson et al. [272]] experiments supplement
older wetted-wall flow tube [Hanson and Ravishankara, [129]] and Knudsen cell
measurements [Rossi et al. [273]], [Tolbert et al [319]]. Although earlier Knudsen cell
measurements probably suffered from surface saturation, more recent results compare well
with those from other techniques. Saturation free results, available over a temperature range
of 200-265 K and a H2SO4 concentration range of 39 to 75 wt. %, were fit to a
phenomenological model developed by Robinson et al. [270]. Measured values depend
strongly on H2SO4 concentration and vary modestly with temperature, with a trend to
somewhat higher values for the 210 - 220 temperature range. The temperature-dependent
uptake model takes into account the temperature and composition dependence of the
effective Henry's Law constant, liquid phase diffusion coefficient, and the liquid phase
hydrolysis rate constant. The hydrolysis reaction was treated by modeling two reaction
channels, a direct hydrolysis process dominating reaction at low H2SO4 concentrations with
a reaction rate proportional to water activity and a proton-catalyzed reaction with a rate
proportional to H+ activity, which dominates at higher acid concentrations.
The data fit to the original Robinson et al. model have been supplemented by additional
wetted-wall flow tube and aerosol flow tube data from Hanson [125] and aerosol flow tube
data from Ball et al. [13]. A revised kinetic model based on Robinson et al. [272] adding
these additional data has been developed based on
1 1 1
= + H 2O
b
where,
4 H ClONO2 RT
(D )
1/ 2
bH 2O = ClONO2 khydr
c
The detailed parameterization for HClONO2, DClONO2 and khydr are given in the Appendix.
As was the case for N2O5 hydrolysis khydr is seen to have a direct and an acid catalyzed
channel. Using the same error analysis approach as in note 17 on N2O5 uptake, the error of
using the model is about 32.4% (one sigma), with m=32.2% and d=4.0%.
In the calculation of the chlorine activation (Cl2 production) rate under stratospheric
conditions, one needs to take into account the competition between the reactions of ClONO2
+ H2O and ClONO2 + HCl. The presence of HCl will depress the reaction probability of
ClONO2 with H2O (see note 49).
47. ClONO2 + HCl + H2O(s) - Reaction probabilities of 0.27 (+0.73, -0.13) [Leu [195]] and
0.05 to 0.1 [Molina et al. [226]] were reported at 195 and 185 K, respectively. Abbatt and
Molina [6] and Hanson and Ravishankara [129] report that a portion of the reaction may be
due to HOCl + HCl Cl2 + H2O, with HOCl formed from ClONO2 + H2O(s) HOCl +
53
HNO3(s). Hanson and Ravishankara [130] saw no enhancement of the ClONO2 reaction
probability when H2O(s) is doped with HCl. Their preferred value at 192 K is = 0.3, but
this is consistent with = 1. Chu et al. [61] also report a value of 0.27 (0.19) at 188 K,
assuming no correction for porosity, but suggest the true value is 0.10 (0.08). Using a
Knudsen cell technique and looking at initial uptake, Oppliger et al. [240] measured = 0.7
at 180 K and 0.2 at 200 K with HCl in excess. Eliminating the Molina et al. points, which
were taken at much higher ClONO2 concentrations than the others, data plots of the
remaining data show no obvious bias when plotted as a function of reactant concentration or
temperature (180-200 K). Their average value = 0.26 0.06. The Oppliger et al. data were
presented for two HCl concentrations, differing by a factor of three. All points from both
HCl concentrations were included since all the data were generally consistent with previous
measurements, although the higher HCl concentrations did tend to produce modestly higher
uptake coefficients. Until a fuller model is available, a single temperature independent
value with a moderate uncertainty due to surface porosity seems appropriate.
48. ClONO2 + HCl + HNO33H2O - Measurements by Hanson and Ravishankara [130,131], Leu
and co-workers in Moore et al. [229] and Leu et al. [199], and Abbatt and Molina [6] all report
high values (>0.1) on NAT for temperatures between 192 and 202 K. Hanson and
Ravishankara indicate that reaction probabilities on NAD are similar to those on NAT. The
most recent NAT studies [Abbatt and Molina [6]] show a strong fall-off with relative
humidity from >0.2 at 90% RH to 0.002 at 20% RH, indicating the necessity of sufficient
water to solvate reactants. Within the limited measurements, data plots show no indication
that the reaction probability of ClONO2 + HCl depends on HCl and ClONO2 gas phase
concentrations or temperature between 191 and 202 K. Averaged data yield is = 0.23
0.10. Carslaw and Peter [58] have published a model of this reaction and its dependence on
HCl uptake.
49. ClONO2 + HCl + H2SO4nH2O - Early work by Tolbert et al. [319] and Hanson and
Ravishankara [129] indicated that the presence of HCl had little effect on the reaction of
ClONO2 with concentrated sulfuric acid (>65 wt.% H2SO4). Subsequent realization that
HCl would be more soluble, and therefore a more potent reactant, in the colder, more dilute
sulfuric acid aerosols characteristic of the polar stratosphere led to additional investigations
by Hanson and Ravishankara [135], Zhang et al. [357], Elrod et al. [104] and Hanson [125].
All these measurements show a strong dependence of reactivity on HCl solubility, which in
turn depends on water activity. The solubility of HCl in a wide range of sulfuric acid
solutions has been experimentally determined by a range of techniques which agree well
with current thermodynamic models. See Robinson et al. [271] for a review. Hanson and
Lovejoy [127] measured a reacto-diffusive length, " , of only 0.0090.005 m for 60 wt.%
H2SO4 in an aerosol flow reactor. (See Hanson et al. [138] for a definition of " .) This is a
factor of four lower than the value for the hydrolysis reaction of ClONO2 showing the
significant enhancement of ClONO2 uptake due to HCl.
Since the effect of HCl on the ClONO2 uptake is to increase the ClONO2 pseudo first
order reaction rate, the model of ClONO2 uptake (see note 45) should include the pseudo
first order reaction rate, kHCl. The formulation of kHCl is found in the Appendix. It is likely
that the ClONO2 reaction with HCl, like the ClONO2 hydrolysis reaction, is acid catalyzed
54
via protonated HClONO2+, where Cl+ is activated as in the case of HOCl + HCl. For the
ClONO2 + HCl reaction, there is also a surface reaction [Hanson [125]]. Hanson proposed
that s is linearly proportional to water activity; however, the calculated value of o at 250 K
and 60 wt% H2SO4 using his formulation is 0.02 (here o~s), which is contradictory to his
aerosol flow reactor result, which yielded o=0.0079 (here o~b) [Hanson and Lovejoy
[127]]. In the model presented in Appendix, it is assumed that that s is linearly proportional
to Henrys law constant of ClONO2, rather than the water activity. The temperature
dependence of s is determined, based on two measured values of s at 203K [Hanson,
[125]] and 250K [Hanson and Lovejoy, [127]]. The proposed model yields a value of
o~0.011 (here o~b), which is close to the measured value. The competition between
ClONO2 + HCl and ClONO2 + H2O is discussed above in note 45. See Table A-3 in the
Appendix for the detailed model.
In the stratosphere, when the reaction rate of ClONO2 with HCl exceeds the flux of HCl
to the particle surface, HCl is depleted. This, in turn, will depress the rate of both the
ClONO2 and HOCl + HCl reactions, and increase the ClONO2 hydrolysis rate. Shi et al.
[289] have proposed a model in which this effect is taken into account by including a factor
FHCl (see Table A-3). The formulation of FHCl is based on scaling HCl reaction and
accommodation fluxes. This flux correction is not exact (i.e. it does rigorously calculate
the HCl surface or bulk concentration) but provides a good approximation to expected
reduction in HCl + ClONO2/ HOCl reactivity and, just as importantly, the effective increase
in ClONO2 + H2O reactivity when pClONO2 > pHCl. This is particularly relevant during cold
Cl activation events when HCl can be virtually removed (i.e., see Jaegle et al. [160]).
Using the same error analysis approach as in note 17 on N2O5 uptake, the error of using
the model in the Appendix is about 40.0% (one sigma), with m=39.8% and d=4.0%.
57. HOBr + HCl + H2O(s) and H2SO4 nH2O - Abbatt [3] measured = 0.25 (+0.10/-0.05) for
this reaction on ice at 228 K. The BrCl product was observed by mass spectrometry. Since
only one value is available, a significantly larger uncertainty than indicated by the stated
experimental error is appropriate. No data on NAT surfaces is currently available.
For the sulfuric acid reaction, Abbatt [4] measured s of ~ 0.1 to 0.2 for [HCl] > 1 x 1012
cm over 68.8 wt. % H2SO4 at 228 K; yielding an estimated kIIHCl+HOBr = 1.4 x 105 M-1 s-1
-3
with a factor of 2 uncertainty. Hanson and Ravishankara [136] also measured

= 0.2 [+0.2, - 0.1] for 60 wt. % H2SO4 at 210 K. However, both of these measurements
were based on significant underestimation of the solubility of HOBr in the relevant sulfuric
acid solutions. More recent measurements by Waschewsky and Abbatt [340] indicate that H
for HOBr varies slightly with acidity between 60 to 70 wt.% H2SO4 and more strongly with
temperature between 208 and 238 K. (For 59.7 wt.% H2SO4, H (M atm-1) = 1.2x106 at 208
K and and 2.2x105 at 228 K.) The HOBr + HCl second order liquid phase rate constant,
kIIHCl+HOBr, varies between 2x105 and 3x108 (M-1s-1) between 213 and 238 K over the same
composition range (60-70 wt.% H2SO4). Such a strong dependence on acid composition for
the reaction rate of HOBr + HCl and the very small acid composition dependence for HOBr
solubility in H2SO4 solution might be partially due to the formation of H2OBr+ in the acidic
solution as discussed in their paper. However, this acid catalyzed reaction, i.e. H2OBr+ +
55
HCl, alone does not completely account for measured reaction rates over the acid
composition range studied.
Using the Henrys Law data for HOBr reported by Waschewsky and Abbatt [340], the
limiting reagent will vary depending on atmospheric temperature (H2SO4 wt.%) and the
concentrations of HOBr and HCl. For stratospheric conditions where [HOBr] is 10 pptv and
[HCl] 1ppbv, they predict dissolved HOBr will be in excess above 204 K and HCl in excess
below 204 K for a H2O vapor partial pressure of 3x10-7 atm. From their coated wall flow
reactor uptake measurements, Waschewsky and Abbatt [340] derived expressions for
kIIHCl+HOBr and predicted uptake coefficients. For temperature between 204 and 218 K
where HOBr is likely to be in excess, they calculated HCl uptake coefficients, HCl, which
range between 7x10-5 and 9x10-5. For temperatures in the 202-198 K range, where
dissolved HCl is likely to be excess, the calculated uptake coefficients for HOBr, HOBr, of
1x10-2. Clearly, the HOBr + HCl reaction will be difficult to parameterize in a simple
manner. Potential inconsistencies in their kIIHCl+HOBr values, as discussed by Waschewsky
and Abbatt, indicate that further measurements will be required before this reaction can be
definitively modeled.
59. BrONO2 + H2O(s) - Hanson and Ravishankara [132] investigated these reactions in an ice-
coated flow reactor at 200 (10) K. The reaction of BrONO2 with H2O(s) proceeded at a
rate indistinguishable from the gas phase diffusion limit, implying that the reaction
probability may be as high as one; the product BrNO(g) was observed. Allanic et al [7]
used a Knudsen cell reactor to measure BrONO2 uptake between 190-200 K. Values of
initial s in the 0.2-0.3 range were observed. An average = 0.26 0.05 was obtained from
all of the appropriate data from both experiments.
60. BrONO2 + H2SO4nH2O Hanson and co-workers used both coated flow tube and aerosol
flow tube techniques to show that the reaction of BrONO2 with 45-70 wt. % H2SO4 is
extremely facile at temperatures from 210 to 298 K. Hanson and Ravishankara [136]
measured s of 0.5 (+ 0.5, - 0.25) (45 wt. % H2SO4, 210 K, 0.4 (+0.6, -0.2) (60 wt. %, 210
K) , and 0.3 (+ 0.7, -0.1) (70 wt. %, 220 K) in a coated-wall flow tube experiment. Hanson
et al. [137] measured ~ 0.8 (20 to 40% error) for submicron aerosols at temperatures
between 249 and 298 K and H2SO4 concentrations of 45 to 70 wt. %; they did observe a
sharp fall off in for H2SO4 concentrations between 73 and 83 wt. %. Hanson has analyzed
these combined data sets, the data indicated that is a function of sulfuric acid
concentration, but independent of temperature. He has fit an empirical expression for for
BrONO2 + H2O of: = exp(a+b*wt.)+c [Hanson, priv. comm.]. The data have been fitted
to the formulation 1/=1/+1/rxn, yielding =0.805, and a=29.24, b=-0.396, c=0.114.
Additional unpublished measurements using both techniques at higher temperatures
performed by Hanson [priv. comm.] also fit this functional form. Using the same approach
as detailed in note 17 for N2O5, the error for BrONO2+ H2O is 27.3% (one sigma), with
m=26.6% and d=6.3%.
56
APPENDIX
Phenomenological Model of ClONO2 Hydrolysis and the Reaction

of ClONO2 and HOCl with HCl in Sulfuric Acid Solutions
In this section, a model is presented which provides a formulation for uptake coefficients for
ClONO2 hydrolysis and the reactions of ClONO2 and HOCl with HCl in sulfuric acid solutions.
This model is based on the previous work of Robinson et al. [272] and Donaldson et al. [99].
Both ClONO2 and HOCl are protonated (HClONO2+ and H2OCl+) before they react with
HCl as was suggested for the reaction of HOCl + HCl by Donaldson et al. [99]. In addition to the
normal hydrolysis reaction of ClONO2 with H2O, there is an acid catalyzed reaction of ClONO2
with H2O in concentrated H2SO4, which involves the intermediate, HClONO2+. Assuming a
steady state for the protonated species, the functional forms for the acid catalyzed reaction rate
coefficients of ClONO2 with H2O, HCl, and HOCl + HCl can be derived. These rate coefficients
along with the rate coefficient for ClONO2+ H2O are obtained via a model fit to experimental data
and are listed in the Tables A-3 and A-4. The Henrys Law constant for HOCl in H2SO4/H2O
solution is assumed to be dependent on the molarity of H2SO4 rather than on the molality of
H2SO4 as suggested by Huthwelker et al. [158]. The model of Huthwelker et al. underestimated
the recently measured values of Donaldson et al. [99]. The new expression for the Henrys law
constant for HOCl in Table A-4 was obtained via fitting to previous HOCl solubility data [134]
and agrees well with the measurements of Donaldson et al. [99]. A similar expression for ClONO2
solubility is obtained via the model fit to the experimental uptake data. The expressions for HOCl
and ClONO2 diffusion coefficients are given by Klassen et al. [181]; however, the viscosity of
H2SO4/H2O solution shown in Table A-2 is a new fit.
The following tables present the data needed to calculate the uptake coefficient of ClONO2
+ H2O, HCl, and HOCl + HCl under the stratospheric conditions. Data in Table A-1 were
obtained from Tabazadeh et al. [307] and used to calculate the H2SO4 weight percentage for a
specified temperature, T, and water vapor partial pressure, pH2O. Table A-2 is used to calculate the
H2SO4 parameters, which are required in the calculation of uptake coefficients. Table A-3
contains the complete uptake model for the reaction of ClONO2 + H2O and HCl. Table A-4
contains the complete uptake model for the reaction of HOCl + HCl.
57
Table A-1: Calculations of H2SO4 wt.% from T and pH2O (from Tabazadeh et al.[307])
Parameter Expression Note
saturation
3505.1578807 330918.55082 12725068.262 water vapor
pHo 2O exp 18.452406985 +
T T2 T3 pressure,
mbar
pH 2 O water
aw
p o
H 2O
activity
(T 190)( y2 y1 ) H2SO4
m y1 + molality,
70 mol/kg
y1 a1 awb1 + c1 aw + d1
y2 a2 awb2 + c2 aw + d 2
9800m Weight
wt. percentage,
98m + 1000 %
aw 0.05 0.05 < aw < 0.85 aw 0.85
a1=12.37208932 a1=11.820654354 a1=-180.06541028

b1=-0.16125516114 b1=-0.20786404244 b1=-0.38601102592
c1=-30.490657554 c1=-4.807306373 c1=-93.317846778
d1=-2.1133114241 d1=-5.1727540348 d1=273.88132245
a2=13.455394705 a2=12.891938068 a2=-176.95814097
b2=-0.1921312255 b2=-0.23233847708 b2=-0.36257048154
c2=-34.285174607 c2=-6.4261237757 c2=-90.469744201
d2=-1.7620073078 d2=-4.9005471319 d2=267.45509988
58
Table A-2: Parameters for H2SO4 Solution
Parameter Value or Expression References and Note
wt
M H2SO4 H2SO4 molarity, mol/l
9.8
H2SO4 solution
1 + Z1m+Z2 m1.5 + Z 3 m 2
density [158], g/cm3
Z1 0.12364 5.6x107 T 2
Z2 0.02954 + 1.814x107 T 2
Z3 2.343x10 3 1.487x10 6 T-1.324x10-8 T 2
wt
X (100 wt)98 H2SO4 mole fraction
wt + 18

448 Viscosity of H2SO4
A T -1.43 exp
(T-T ) solution,
# cp
A 169.5 + 5.18 wt-0.0825 wt + 3.27x103 wt 3
2
To 144.11 + 0.166 wt-0.015 wt 2 + 2.18x104 wt 3
exp[60.51 0.095 wt+0.0077 wt 2 1.61x105 wt3
aH
( )
1.76 + 2.52x104 wt 2 T 0.5 Acid activity in
+
(805.89 + 253.05 wt )]0.076 molarity
T 0.5
Note: If T and wt. are known, water activity can be calculated from:
1 26.9033
aw=exp (-69.775 X-18253.7 X 2 + 31072.2 X3 25668.8 X 4 )
T T 2
59
Table A-3: Uptake Model Parameters for the ClONO2 + H2O and ClONO2 + HCl Reactions
Parameter Value or Expression References and Notes
4H ClONO2 RT ( DClONO2 k hydr )
0.5
bH 2O R=0.082 atm K -1M -1

CClONO2
0.5
CClONO2 8RT -1
1474 T 0.5
c= , cm s
MW
4710
1.6 x10 6 exp
H ClONO2 T Shi et al. [289], M atm-1
(
exp SClONO2 M H 2SO4 )
24.0 Setchenow coefficient, [289]
SClONO2 0.306 +
T M-1
8
5 x10 T
DClONO2 Klassen et al.[181], cm2 s-1

k hydr k H 2O a w + k H a H a W Shi et al. [289], s-1
k H 2O 1.95x1010 exp(2800 / T) Shi et al. [289], s-1
kH 1.22x1012 exp(6200 / T) Shi et al. [289], M-1 s-1
-1
k HCl 7.9 x1011 aH DClONO2 M HCl [289], s
M HCl H HCl p HCl p HCl (atm) , M
(0.094 0.61X+1.2X ) 2
H HCl

exp 8.68 +
(8515 10718 X ) 0.7
parameterization of Carslaw et
T al. [57], M atm-1

0.5
DClONO
" ClONO2 2
reacto-diffusive length, cm
( k hydr + k HCl )

1 "
ClONO2
f ClONO2 r r r=aerosol radius, cm
tanh
" ClONO
2
0.5
k
rxn
f ClONO2 1 + HCl
H 2O
Hanson [125]
ClONO2 b k
hydr
k HCl
bHCl ClONO
rxn
2
(kHCl + khydr ) Hanson [125]
1374
s 66.12 HClONO2 MHCl exp Shi et al. [289]
T
60
Table A-3: Uptake Model Parameters for the ClONO2 + H2O and ClONO2 + HCl Reactions (Contd)
1
0.612 ( s + bHCl ) pClONO
0.5
MWHCl
FHCl 1 + 2
= 0.612
p HCl
MWClONO2
s FHCl s
HCl
b FHCl HCl
b
HCl
ClONO
rxn
k
2 hydr
b +
b
(k HCl + khydr )
1

ClONO2 1
1 +

(
s + b

)
s + b
HCl
ClONO2 _ HCl ClONO2 Hanson [125]
s + b
ClONO2 _ H 2O ClONO2 ClONO2 _ HCl Hanson [125]
Note: We have assumed that =1 in Appendix Tables 3 and 4 which is appropriate in the
stratosphere. F accounts for depletion of HCl in the particles due to reaction with ClONO2
inside/on the particle surface. The effect of the HOCl + HCl reaction on the HCl concentration
has been neglected; however, the depletion of HCl in the particles due to reaction with ClONO2
affects the HOCl + HCl reaction probability which is taken into account through FHCl in Table 4
(see HOCl).
61
Table A-4: Uptake Model Parameters for the HOCl + HCl Reaction
Parameter Value or Expression References and Notes
4H HOCl RT ( D HOCl k HOCl_HCl )
0.5
rxn
HOCl R=0.082 atm K -1M -1
C HOCl
0.5
C HOCl 8RT -1
2009 T 0.5
c= , cm s
MW
5862.4
1.91x10 6 exp
H HOCl T Shi et al. [289]
exp(SHOCl M H2SO4 )
59.18 Setchenow coefficient,[289],
SHOCl 0.0776 +
T M-1
8
6.4 x10 T
D HOCl Klassen et al.[181], cm2 s-1

k HOCl _ HCl 1.25x109 a H DHOCl M HCl Shi et al. [289], s-1
0.5
D HOCl
" HOCl reacto-diffusive length, cm
k HOCl_HCl
1 "
HOCl
f HOCl r r r=aerosol radius, cm
tanh
" HOCl
1
HOCl 1 Shi et al. [289]
1 +
( f HOCl HOCl
rxn
F )
HOCl
Note: The factor FHCl here, as shown in Table A-3, accounts for depletion of HCl in the particles
due to reaction with ClONO2 inside/on the particle surface.
62
REFERENCES
1. Chlorofluoromethanes and the Stratosphere. In NASA Reference Publication 1010;

Hudson, R. D., Ed.; NASA: Washington, D.C, 1977.
2. The Stratosphere: Present and Future. In NASA Reference Publication 1049; Hudson, R.
D., Reed, E. I., Eds.; NASA: Washington, D.C, 1979.
3. Abbatt, J. P. D., 1994, Geophys. Res. Lett., 21, 665-668.
4. Abbatt, J. P. D., 1995, Geophys. Res. Lett., 100, 14,009-14,017.
5. Abbatt, J. P. D. and M. J. Molina, 1992, Geophys. Res. Lett., 19, 461-464.
6. Abbatt, J. P. D. and M. J. Molina, 1992, J. Phys. Chem., 96, 7674-7679.
7. Allanic, A., R. Oppliger and M. Rossi, 1997, J. Geophys. Res., 102, 23529-23541.
8. Anastasi, C. and I. W. M. Smith, 1976, J. Chem. Soc. Faraday Trans. 2, 72, 1459-1468.
9. Anderson, J. G., J. J. Margitan and F. Kaufman, 1974, J. Chem. Phys., 60, 3310.
10. Armerding, W., F. J. Comes and B. Schulke, 1995, J. Phys. Chem., 99, 3137-3143.
11. Arnold, I., F. J. Comes and G. K. Moortgat, 1977, Chem. Phys., 24, 211-217.
12. Atkinson, R., D. L. Baulch, R. A. Cox, J. Hampson, R. F., J. A. Kerr, M. J. Rossi and J.
Troe, 1997, J. Phys. Chem. Ref. Data, 26, 1329-1499.
13. Ball, S. M., A. Fried, B. E. Henry and M. Mozurkewich, 1998, Geophys. Res. Lett., 25,
3339-3342.
14. Ball, S. M. and G. Hancock, 1995, Geophys. Res. Lett., 22, 1213-1216.
15. Ball, S. M., G. Hancock, S. E. Martin and J. C. Pinot de Moira, 1997, Chem. Phys. Lett.,
264, 531-538.
16. Ball, S. M., G. Hancock, I. J. Murphy and S. P. Rayner, 1993, Geophys. Res. Lett., 20,
2063-2066.
17. Ball, S. M., G. Hancock, J. C. Pinot de Moira, C. M. Sadowski and F. Winterbottom,
1995, Chem. Phys. Lett., 245, 1-6.
18. Barnes, R. J., M. Lock, J. Coleman and A. Sinha, 1996, J. Phys. Chem., 100, 453-457.
19. Barnes, R. J., A. Sinha and H. A. Michelsen, 1998, J. Phys. Chem. A, 102, 8855-8859.
20. Barone, S. B., M. A. Zondlo and M. A. Tolbert, 1997, J. Phys. Chem. A, 101, 8643-8652.
21. Basco, N. and S. K. Dogra, 1971, Proc. Roy. Soc. A., 323, 417-429.
22. Basco, N. and J. E. Hunt, 1979, Int. J. Chem. Kinet., 11, 649-664.
23. Battin-Leclerc, F., I. K. Kim, R. K. Talukdar, R. W. Portmann, A. R. Ravishankara, R.
Steckler and D. Brown, 1999, J. Phys. Chem. A, 103, 3237-3244.
24. Baulch, D. L., R. A. Cox, P. J. Crutzen, R. F. Hampson, Jr., J. A. Kerr, J. Troe and R. T.
Watson, 1982, J. Phys. Chem. Ref. Data, 11, 327-496.
25. Baulch, D. L., R. A. Cox, R. F. Hampson, Jr., J. A. Kerr, J. Troe and R. T. Watson, 1980,
J. Phys. Chem. Ref. Data, 9, 295-471.
26. Beichert, P., L. Wingen, J. Lee, R. Vogt, M. J. Ezell, M. Ragains, R. Neavyn and B. J.
Finlayson-Pitts, 1995, J. Phys. Chem., 99, 13156-13162.
27. Bemand, P. P., M. A. A. Clyne and R. T. Watson, 1973, J. Chem. Soc. Faraday Trans. 1,
69, 1356-1374.
28. Bemand, P. P., M. A. A. Clyne and R. T. Watson, 1974, J. Chem. Soc. Faraday Trans. 2,
70, 564-576.
29. Benter, T., C. Feldmann, U. Kirchner, M. Schmidt, S. Schmidt and R. N. Schindler, 1995,
Ber. Bunsenges. Phys. Chem., 99, 1144-1147.
30. Berland, B. S., M. A. Tolbert and S. M. George, 1997, J. Phys. Chem. A, 101, 9954-9963.
63
31. Birk, M., R. R. Friedl, E. A. Cohen, H. M. Pickett and S. P. Sander, 1989, J. Chem. Phys.,
91, 6588-6597.
32. Birks, J. W., B. Shoemaker, T. J. Leck, R. A. Borders and L. J. Hart, 1977, J. Chem.
Phys., 66, 4591-4599.
33. Birks, J. W., B. Shoemaker, T. J. Leck and D. M. Hinton, 1976, J. Chem. Phys., 65, 5181-
5185.
34. Borders, R. A. and J. W. Birks, 1982, J. Phys. Chem., 86, 3295-3302.
35. Brock, J. C. and R. T. Watson, 1980, Chem. Phys., 46, 477-484.
36. Brock, J. C. and R. T. Watson, 1980, Chem. Phys. Lett., 71, 371-375.
37. Brown, S. S., R. K. Talukdar and A. R. Ravishankara, 1999, Chem. Phys. Lett., 299, 277-
284.
38. Brown, S. S., R. K. Talukdar and A. R. Ravishankara, 1999, J. Phys. Chem. A, 103, 3031-
3037.
39. Brune, W. H., J. J. Schwab and J. G. Anderson, 1983, J. Phys. Chem., 87, 4503-4514.
40. Burkholder, J. B., 1993, J. Geophys. Res., 98, 2963-2974.
41. Burkholder, J. B., P. D. Hammer and C. J. Howard, 1987, J. Phys. Chem., 91, 2136-2144.
42. Burkholder, J. B., P. D. Hammer, C. J. Howard and A. Goldman, 1989, J. Geophys. Res.,
94, 2225-2234.
43. Burkholder, J. B., J. J. Orlando and C. J. Howard, 1990, J. Phys. Chem., 94, 687-695.
44. Burrows, J. P., D. I. Cliff, G. W. Harris, B. A. Thrush and J. P. T. Wilkinson, 1979, Proc.
Roy. Soc. (London), A368, 463-481.
45. Burrows, J. P., R. A. Cox and R. G. Derwent, 1981, J. Photochem., 16, 147-168.
46. Burrows, J. P., G. W. Harris and B. A. Thrush, 1977, Nature, 267, 233-234.
47. Burrows, J. P., G. S. Tyndall and G. K. Moortgat, 1985, Chem. Phys. Lett., 119, 193-198.
48. Burrows, J. P., G. S. Tyndall and G. K. Moortgat, 1985, J. Phys. Chem., 89, 4848-4856.
49. Burrows, J. P., T. J. Wallington and R. P. Wayne, 1983, J. Chem. Soc. Faraday Trans. 2,
79, 111-122.
50. Burrows, J. P., T. J. Wallington and R. P. Wayne, 1984, J. Chem. Soc. Faraday Trans. 2,
80, 957-971.
51. Butler, P. J. D. and L. F. Phillips, 1983, J. Phys. Chem., 87, 183-184.
52. Calvert, J. G. and J. N. Pitts. . In Photochemistry; John Wiley & Sons, Inc., New York:,
1966; pp 783.
53. Cannon, B. D., J. S. Robertshaw, I. W. M. Smith and M. D. Williams, 1984, Chem. Phys.
Lett., 105, 380-385.
54. Cantrell, C. A., J. A. Davidson, A. H. McDaniel, R. E. Shetter and J. G. Calvert, 1988, J.
Chem. Phys., 88, 4997-5006.
55. Cantrell, C. A., R. E. Shetter and J. G. Calvert, 1994, J. Geophys. Res., 99, 3739-3743.
56. Cantrell, C. A., R. E. Shetter, J. G. Calvert, G. S. Tyndall and J. J. Orlando, 1993, J. Phys.
Chem., 97, 9141-9148.
57. Carslaw, K. S., S. L. Clegg and P. Brimblecombe, 1995, J. Phys. Chem., 99, 11,557-
11,574.
58. Carslaw, K. S. and T. Peter, 1997, Geophys. Res. Lett., 24, 1743-1746.
59. Chen, L., H. Rabitz, D. B. Considine, C. H. Jackman and J. A. Shorter, 1997, J. Geophys.
Res., 102, 16,201-16,214.
60. Choo, K. Y. and M. T. Leu, 1985, J. Phys. Chem., 89, 4832-4837.
61. Chu, L. T., M.-T. Leu and L. F. Keyser, 1993, J. Phys. Chem., 97, 12798-12804.
64
62. Clegg, S. L. and P. Brimblecombe, 1986, Atmos. Environ., 20, 2483.
63. Clough, P. N. and B. A. Thrush, 1967, Trans. Faraday Soc., 63, 915-925.
64. Clyne, M. A. A. and W. S. Nip, 1976, J. Chem. Soc. Faraday Trans. 1, 72, 2211-2217.
65. Clyne, M. A. A. and W. S. Nip, 1976, J. Chem. Soc. Faraday Trans. 2, 72, 838-847.
66. Clyne, M. A. A. and R. F. Walker, 1973, J. Chem. Soc. Faraday Trans. 1, 69, 1547-1567.
67. Clyne, M. A. A. and R. T. Watson, 1974, J. Chem. Soc. Faraday Trans. 1, 70, 2250-2259.
68. Clyne, M. A. A. and R. T. Watson, 1977, J. Chem. Soc. Faraday Trans. 1, 73, 1169-1187.
69. Connell, P. S. and H. S. Johnston, 1979, Geophys. Res. Lett., 6, 553-556.
70. Cox, R. A., J. P. Burrows and G. B. Coker, 1984, Int. J. Chem. Kinet., 16, 445-67.
71. Cox, R. A., J. P. Burrows and T. J. Wallington, 1981, Chem. Phys. Lett., 84, 217-221.
72. Cox, R. A. and R. G. Derwent, 1979, J. Chem. Soc. Far. Trans. 1, 75, 1635-1647.
73. Cox, R. A. and G. D. Hayman, 1988, Nature, 332, 796-800.
74. Cox, R. A. and R. Lewis, 1979, J. Chem. Soc. Faraday Trans. 1, 75, 2649-2661.
75. Croce de Cobos, A. E. and J. Troe, 1984, Int. J. Chem. Kinet., 16, 1519-1530.
76. Danis, F., F. Caralp, J. Masanet and R. Lesclaux, 1990, Chem. Phys. Lett., 167, 450.
77. Dasch, W., K.-H. Sternberg and R. N. Schindler, 1981, Ber. Bunsenges. Phys. Chem., 85,
611-615.
78. Davidson, J. A., C. J. Howard, H. I. Schiff and F. C. Fehsenfeld, 1979, J. Chem. Phys., 70,
1697-1704.
79. Davis, D. D., J. T. Herron and R. E. Huie, 1973, J. Chem. Phys., 58, 530-535.
80. DeMore, W. B., 1979, J. Phys. Chem., 83, 1113-1118.
81. DeMore, W. B. 182nd National Meeting of the American Chemical Society, 1981, New
York.
82. DeMore, W. B., 1982, J. Phys. Chem, 86, 121-126.
83. DeMore, W. B., 1991, J. Geophys. Res., 96, 4995-5000.
84. DeMore, W. B., D. M. Golden, R. F. Hampson, C. J. Howard, M. J. Kurylo, J. J.
Margitan, M. J. Molina, A. R. Ravishankara and R. T. Watson Chemical Kinetics and
Photochemical Data for Use in Stratospheric Modeling, Evaluation Number 7, JPL
Publication 85-37, Jet Propulsion Laboratory, California Institute of Technology,
Pasadena CA, 1985.
85. DeMore, W. B., D. M. Golden, R. F. Hampson, C. J. Howard, M. J. Kurylo, M. J. Molina,
A. R. Ravishankara and S. P. Sander Chemical Kinetics and Photochemical Data for Use
in Stratospheric Modeling, Evaluation Number 8, JPL Publication 87-41, Jet Propulsion
Laboratory, California Institute of Technology, Pasadena CA, 1987.
Laboratory, California Institute of Technology, Pasadena, CA, 1992.
Laboratory, California Institute of Technology, Pasadena, CA, 1994.
65
A. R. Ravishankara and R. T. Watson Chemical Kinetics and Photochemical Data for
Use in Stratospheric Modeling, Evaluation Number 5, JPL Publication 82-57, Jet
Propulsion Laboratory, California Institute of Technology, Pasadena CA, 1982.
A. R. Ravishankara and R. T. Watson Chemical Kinetics and Photochemical Data for
Propulsion Laboratory, California Institute of Technology, Pasadena CA, 1983.
91. DeMore, W. B., D. M. Golden, R. F. Hampson, M. J. Kurylo, J. J. Margitan, M. J. Molina,
L. J. Stief and R. T. Watson Chemical Kinetics and Photochemical Data for Use in
Stratospheric Modeling, Evaluation Number 4, JPL Publication 81-3, Jet Propulsion
92. DeMore, W. B., S. P. Sander, D. M. Golden, R. F. Hampson, M. J. Kurylo, C. J. Howard,
A. R. Ravishankara, C. E. Kolb and M. J. Molina Chemical Kinetics and Photochemical
Data for Use in Stratospheric Modeling, Evaluation Number 12, JPL Publication 97-4,
Jet Propulsion Laboratory, California Institute of Technology, Pasadena, CA, 1997.
93. DeMore, W. B., L. J. Stief, F. Kaufman, D. M. Golden, R. F. Hampson, M. J. Kurylo, J. J.
Margitan, M. J. Molina and R. T. Watson Chemical Kinetics and Photochemical Data for
Propulsion Laboratory, California Institute of Technology, Pasadena, CA, 1979.
94. DeMore, W. B. and E. Tschuikow-Roux, 1990, J. Phys. Chem., 94, 5856-5860.
95. Deters, B., J. P. Burrows, S. Himmelmann and C. Blindauer, 1996, Ann. Geophysicae, 14,
468-475.
96. Devolder, P., M. Carlier, J. F. Pauwels and L. R. Sochet, 1984, Chem. Phys. Lett., 111,
94-99.
97. Dobis, O. and S. W. Benson, 1987, Int. J. Chem. Kinet., 19, 691-708.
98. Donahue, N. M., M. K. Dubey, R. Mohrschladt, K. Demerjian and J. G. Anderson, 1997,
J. Geophys. Res., 102, 6159-6168.
99. Donaldson, D. J., A. R. Ravishankara and D. R. Hanson, 1997, J. Phys. Chem. A, 101,
4717-4725.
100. Dransfeld, P. and H. G. Wagner, 1987, Z. Naturforsch., 42a, 471-476.
101. Dransfield, T. J., K. K. Perkins, N. M. Donahue, J. G. Anderson, M. M. Sprengnether and
K. Demerjian, 1999, Geophys. Res. Lett., 26, 687-690.
102. Dubey, M. K., G. P. Smith, W. S. Hartley, D. E. Kinnison and P. S. Connell, 1997,
Geophys. Res. Lett., 24, 2737-2740.
103. Eberstein, I. J., 1990, Geophys. Res. Lett., 17, 721-724.
104. Elrod, M. J., R. E. Koch, J. E. Kim and M. S. Molina, 1995, Faraday Discuss, 100, 269-
278.
105. Erler, K., D. Field, R. Zellner and I. W. M. Smith, 1977, Ber. Bunsenges. Phys. Chem.,
81, 22.
106. Fenter, F. F., F. Caloz and M. J. Rossi, 1996, J. Phys. Chem., 100, 1008-1019.
107. Forster, R., M. Frost, D. Fulle, H. F. Hamann, H. Hippler, Schlepegreli and J. Troe, 1996,
J. Chem. Phys., 103, 2949-2958.
108. Fowles, M., D. N. Mitchell, J. W. L. Morgan and R. P. Wayne, 1982, J. Chem. Soc.
Faraday Trans. 2, 78, 1239-1248.
109. Fox, L. E., D. R. Worsnop, M. S. Zahniser and S. C. Wofsy, 1994, Science, 267, 351-355.
66
110. Francisco, J. S., M. R. Hand and I. H. Williams, 1996, J. Phys. Chem., 100, 9250-9253.
111. Fried, A., B. E. Henry, J. G. Calvert and M. Mozukewich, 1994, J. Geophys. Res., 99,
3517-3532.
112. Friedl, R. R. and S. P. Sander, 1989, J. Phys. Chem., 93, 4756-4764.
113. Fulle, D. H., H. F. Hamann, H. Hipler and J. Troe, 1998, J. Chem. Phys., 108, 5391-5397.
114. Geers-Muller, R. and F. Stuhl, 1987, Chem. Phys. Lett., 135, 263-268.
115. George, C., J. L. Ponche, P. Mirabel, W. Behnke, V. Sheer and C. Zetzsch, 1994, J. Phys.
Chem., 98, 8780-8784.
116. Gierczak, T., J. B. Burkholder and A. R. Ravishankara, 1999, J. Phys. Chem. A, 103, 877-
883.
117. Glinski, R. J. and J. W. Birks, 1985, J. Phys. Chem., 89, 3449-3453.
118. Graham, R. A. and H. S. Johnston, 1978, J. Phys. Chem., 82, 254-268.
119. Greenblatt, G. D. and A. R. Ravishankara, 1990, J. Geophys. Res., 95, 3539-3547.
120. Hack, W., G. Mex and H. G. Wagner, 1977, Ber. Bunsenges. Phys. Chem., 81, 677-684.
121. Hack, W., A. W. Preuss, F. Temps and H. G. Wagner, 1979, Ber. Bunsenges. Phys.
Chem., 83, 1275-1279.
122. Handwerk, V. and R. Zellner, 1984, Ber. Bunsenges. Phys. Chem., 88, 405.
123. Hanning-Lee, M. A., B. B. Brady, L. R. Martin and J. A. Syage, 1996, Geophys. Res.
Lett., 23, 1961-1964.
124. Hanson, D. R., 1995, J. Phys. Chem., 99, 13,059-13,061.
125. Hanson, D. R., 1998, J. Phys. Chem. A, 102, 4794-4807.
126. Hanson, D. R. and E. R. Lovejoy, 1994, Geophys. Res. Lett., 21, 2401-2404.
127. Hanson, D. R. and E. R. Lovejoy, 1995, Science, 267, 1326-1329.
128. Hanson, D. R. and E. R. Lovejoy, 1996, J. Phys. Chem., 100, 6397-6405.
129. Hanson, D. R. and A. R. Ravishankara, 1991, J. Geophys. Res., 96, 17307-17314.
131. Hanson, D. R. and A. R. Ravishankara, 1992, J. Phys. Chem., 96, 2682-2691.
136. Hanson, D. R. and A. R. Ravishankara, 1995, Geophys. Res. Lett., 22, 385-388.
137. Hanson, D. R., A. R. Ravishankara and E. R. Lovejoy, 1996, J. Geophys. Res., 101, 9063-
9069.
138. Hanson, D. R., A. R. Ravishankara and S. Solomon, 1994, J. Geophys. Res., 99, 3615-
3629.
139. Hayman, G. D., J. M. Davies and R. A. Cox, 1986, Geophys. Res. Lett., 13, 1347-1350.
140. Heneghan, S. P., P. A. Knoot and S. W. Benson, 1981, Int. J. Chem. Kinet., 13, 677-691.
141. Henson, B. F., K. R. Wilson and J. M. Robinson, 1996, Geophys. Res. Lett., 23, 1021-
1024.
142. Henson, B. F., K. R. Wilson and J. M. Robinson, 1999, submitted to J. Phys. Chem. A, .
143. Hills, A. J., R. J. Cicerone, J. G. Calvert and J. W. Birks, 1988, J. Phys. Chem., 92, 1853-
1858.
144. Hills, A. J. and C. J. Howard, 1984, J. Chem. Phys., 81, 4458-4465.
145. Hippler, H., M. Siefke, H. Stark and J. Troe, 1999, Phys. Chem. Chem. Phys., 1, 57-61.
146. Hippler, H. and J. Troe, 1992, Chem. Phys. Lett., 192, 333-337.
67
147. Hjorth, J., J. Nothholt and G. Restelli, 1992, Int. J. Chem. Kinet., 24, 51-65.
148. Hochanadel, C. J., T. J. Sworski and P. J. Ogren, 1980, J. Phys. Chem., 84, 3274-3277.
149. Hofmann, D. J. and S. J. Oltmans, 1992, Geophys. Res. Lett., 22, 2211-2214.
150. Horowitz, A., J. N. Crowley and G. K. Moortgat, 1994, J. Phys. Chem., 98, 11924-11930.
151. Howard, C. J. and K. M. Evenson, 1974, J. Chem. Phys., 61, 1943.
152. Hu, J. H. and J. P. D. Abbatt, 1997, J. Phys. Chem. A, 101, 871-878.
153. Hu, J. H., Q. Shi, P. Davidovits, D. R. Worsnop, M. S. Zahniser and C. E. Kolb, 1995, J.
Phys. Chem., 99, 8768-8776.
154. Huder, K. J. and W. B. DeMore, 1995, J. Phys. Chem., 99, 3905-3908.
155. Hudson, R. D., 1971, Reviews of Geophysics and Space Physics, 9, 305-399.
156. Husain, D., J. M. C. Plane and N. K. H. Slater, 1981, J. Chem. Soc. Faraday Trans. 2, 77,
1949-1962.
157. Husain, D., J. M. C. Plane and C. C. Xiang, 1984, J. Chem. Soc. Faraday Trans. 2, 80,
713-728.
158. Huthwelker, T., T. Peter, B. P. Juo, S. L. Clegg, K. S. Carshlaw and P. Brimblecombe,
1995, J. Atmos. Chem., 21, 81-95.
159. Ingham, T., D. Bauer, J. Landgraf and J. N. Crowley, 1998, J. Phys. Chem. A, 102, 3293-
3298.
160. Jaegle, L., C. R. Webster, R. D. May, D. C. Scott, R. M. Stimpfle, D. W. Kohn, P. O.
Wennberg, T. F. Hansico, R. C. Cohen, M. H. Proffitt, K. K. Kelly, J. Elkins, D.
Baumgardner, J. E. Dye, J. C. Wilson, R. F. Pueschel, K. R. Chan, R. J. Salawitch, A. F.
Tuck, S. J. Hovde and Y. L. Yung, 1997, J. Geophys. Res., 102, 13,235-13,253.
161. Jayne, J. T., D. R. Worsnop, C. E. Kolb, E. Swartz and P. Davidovits, 1996, J. Phys.
Chem., 100, 8015-8022.
162. Jensen, F. and J. Oddershede, 1990, J. Phys. Chem., 94, 2235.
163. Johnston, H. S., C. A. Cantrell and J. G. Calvert, 1986, J. Geophys. Res., 91, 5159-5172.
164. Johnston, H. S., E. D. Morris, Jr. and J. Van den Bogaerde, 1969, J. Am. Chem. Soc., 91,
7712-7727.
165. Jungkamp, T. P. W., U. Kirchner, M. Schmidt and R. N. Schindler, 1995, J. Photochem.
Photobiol. A: Chemistry, 99, 1-6.
166. Kaye, J. A., 1986, J. Geophys. Res., 91, 7865-7874.
167. Kegley-Owen, C. S., M. K. Gilles, J. B. Burkholder and A. R. Ravishankara, 1999, J.
Phys. Chem. A, 103, 5040-5048.
168. Keyser, L. F., 1978, J. Chem. Phys., 69, 214-218.
169. Keyser, L. F., 1981, J. Phys. Chem., 85, 3667-3673.
174. Keyser, L. F., K. Y. Choo and M. T. Leu, 1985, Int. J. Chem. Kinet., 17, 1169-1185.
175. Keyser, L. F. and M.-T. Leu, 1993, J. Colloid Interface Sci., 155, 137-145.
176. Keyser, L. F. and M.-T. Leu, 1993, Micros. Res. Technol., 25, 434-438.
177. Keyser, L. F., M.-T. Leu and S. B. Moore, 1993, J. Phys. Chem., 97, 2800-2801.
178. Keyser, L. F., S. B. Moore and M. T. Leu, 1991, J. Phys. Chem., 95, 5496-5502.
179. Kircher, C. C., J. J. Margitan and S. P. Sander, 1984, J. Phys. Chem., 88, 4370-4375.
180. Klais, O., P. C. Anderson and M. J. Kurylo, 1980, Int. J. Chem. Kinet., 12, 469-490.
68
181. Klassen, J. K., Z. Hu and L. R. Williams, 1998, J. Geophys. Res., 103, 16,197-16,202.
182. Knauth, H. D., H. Alberti and H. Clausen, 1979, J. Phys. Chem., 83, 1604-1612.
183. Knox, J. H., 1955, Chemistry and Industry, 1631-1632.
184. Knox, J. H. and R. L. Nelson, 1959, Trans. Far. Soc., 55, 937-946.
185. Kolb, C. E., D. R. Worsnop, M. S. Zahniser, P. Davidovits, L. F. Keyser, M.-T. Leu, M. J.
Molina, D. R. Hanson, A. R. Ravishankara, L. R. Williams and M. A. Tolbert. Progress
and Problems in Atmospheric Chemistry. In Adv. Phys. Chem. Series, 3; Barker, J. R., Ed.,
1994; pp 771-875.
186. Kulcke, A., B. Blackman, W. B. Chapman, I. K. Kim and D. J. Nesbitt, 1998, J. Phys.
Chem. A, 102, 1965-1972.
187. Kurylo, M. J., 1973, Chem. Phys. Lett., 23, 467-471.
188. Kurylo, M. J. and W. Braun, 1976, Chem. Phys. Lett., 37, 232-235.
189. Kurylo, M. J., O. Klais and A. H. Laufer, 1981, J. Phys. Chem., 85, 3674-3678.
190. Lam, L., D. R. Hastie, B. A. Ridley and H. I. Schiff, 1981, J. Photochem., 15, 119-130.
191. Lee, F. S. C. and F. S. Rowland, 1977, J. Phys. Chem., 81, 86-87.
192. Lee, T. J., C. M. Rohlfing and J. E. Rice, 1992, J. Chem. Phys., 97, 6593-6605.
193. Lee, Y.-P., R. M. Stimpfle, R. A. Perry, J. A. Mucha, K. M. Evenson, D. A. Jennings and
C. J. Howard, 1982, Int. J. Chem. Kinet., 14, 711-732.
194. Leu, M. T., 1984, J. Phys. Chem., 88, 1394-1398.
195. Leu, M. T., 1988, Geophys. Res. Lett., 15, 17-20.
196. Leu, M. T. and W. B. DeMore, 1976, Chem. Phys. Lett. , 41, 121-124.
197. Leu, M. T. and C. L. Lin, 1979, Geophys. Res. Lett., 6, 425-428.
198. Leu, M. T., C. L. Lin and W. B. DeMore, 1977, J. Phys. Chem., 81, 190-195.
199. Leu, M. T., S. B. Moore and L. F. Keyser, 1991, J. Phys. Chem., 95, 7763-7771.
200. Leu, M. T. and Y. L. Yung, 1987, Geophys. Res. Lett., 14, 949-952.
201. Lii, R.-R., R. A. Gorse, Jr., M. C. Sauer, Jr. and S. Gordon, 1980, J. Phys. Chem., 84, 819-
821.
202. Lin, C. L. and M. T. Leu, 1982, Int. J. Chem. Kinet., 14, 417.
203. Lin, C. L., M. T. Leu and W. B. DeMore, 1978, J. Phys. Chem., 82, 1772-1777.
204. Lippmann, H. H., B. Jesser and U. Schurath, 1980, Int. J. Chem. Kinet., 12, 547-554.
205. Lipson, J. B., T. W. Beiderhase, L. T. Molina, M. J. Molina and M. Olzmann, 1999, J.
Phys. Chem. A, 103, 6540-6551.
206. Lipson, J. B., M. J. Elrod, T. W. Beiderhase, L. T. Molina and M. J. Molina, 1997, J.
Chem. Soc. Faraday Trans., 93, 2665-2673.
207. Lock, M., R. J. Barmes and A. Sinha, 1996, J. Phys. Chem., 100, 7972-7980.
208. Manion, J. A., C. M. Fittschen, D. M. Golden, L. R. Williams and M. A. Tolbert, 1994,
Israel J. Chem., 34, 355-363.
209. Manning, R. and M. J. Kurylo, 1977, J. Phys. Chem., 81, 291-296.
210. Manzanares, E. R., M. Suto, L. C. Lee and D. Coffey, 1986, J. Chem. Phys., 85, 5027-
5034.
211. Margitan, J. J., 1984, J. Phys. Chem., 88, 3638-3643.
212. Margitan, J. J. and R. T. Watson, 1982, J. Phys. Chem., 86, 3819-3824.
213. Marx, W., F. Bahe and U. Schurath, 1979, Ber. Bunsenges. Phys. Chem., 83, 225-230.
214. McGrath, M. P., K. C. Clemitshaw, F. S. Rowland and W. J. Hehre, 1988, Geophys. Res.
Lett., 15, 883-886.
69
215. McGrath, M. P., K. C. Clemitshaw, F. S. Rowland and W. J. Hehre, 1990, J. Phys. Chem.,
94, 6126-6132.
216. Michael, J. V., J. E. Allen, Jr. and W. D. Brobst, 1981, J. Phys. Chem., 85, 4109-4117.
217. Michael, J. V. and J. H. Lee, 1977, Chem. Phys. Lett., 51, 303-306.
218. Michelangeli, D. V., M. Allen and Y. L. Yung, 1991, Geophys. Res. Lett., 18, 673-676.
219. Middlebrook, A. M., L. T. Iraci, L. S. McNeil, B. G. Koehler, M. A. Wilson, O. W.
Saastad and M. A. Tolbert, 1993, J. Geophys. Res., 98, 20473-20481.
220. Mishalanie, E. A., J. C. Rutkowski, R. S. Hutte and J. W. Birks, 1986, J. Phys. Chem., 90,
5578-5584.
221. Molina, L. T. and M. J. Molina, 1978, J. Phys. Chem., 82, 2410-2414.
222. Molina, M. J., A. J. Colussi, L. T. Molina, R. N. Schindler and T. L. Tso, 1990, Chem.
Phys. Lett., 173, 310-315.
223. Molina, M. J., T. Ishiwata and L. T. Molina, 1980, J. Phys. Chem., 84, 821-826.
224. Molina, M. J., L. T. Molina and T. Ishiwata, 1980, J. Phys. Chem., 84, 3100-3104.
225. Molina, M. J., L. T. Molina and C. A. Smith, 1984, Int. J. Chem. Kinet., 16, 1151-1160.
226. Molina, M. J., T. L. Tso, L. T. Molina and F. C. Wang, 1987, Science, 238, 1253-1259.
227. Molina, M. J., R. Zhang, P. J. Woolridge, J. R. McMahon, J. E. Kim, H. Y. Chang and K.
D. Beyer, 1993, Science, 261, 1418-1423.
228. Moonen, P. C., J. N. Cape, R. L. Storeton-West and R. McColm, 1998, J. Atmos. Chem.,
29, 299-314.
229. Moore, S. B., L. F. Keyser, M. T. Leu, R. P. Turco and R. H. Smith, 1990, Nature, 345,
333-335.
230. Moore, T. A., M. Okumura, J. W. Seale and T. K. Minton, 1999, J. Phys. Chem. A, 103,
1692-1695.
231. Mozurkewich, M. and J. Calvert, 1988, J. Geophys. Res., 93, 889-896.
232. Msibi, I. M., Y. Li, J. P. Shi and R. M. Harrison, 1994, J. Phys. Chem., 18, 291-300.
233. Nelson, D. D., Jr. and M. S. Zahniser, 1994, J. Phys. Chem., 98, 2101-2104.
234. Nickolaisen, S. L., R. R. Friedl and S. P. Sander, 1994, J. Phys. Chem., 98, 155-169.
235. Nicovich, J. M., K. D. Kreutter and P. H. Wine, 1990, Int. J Chem. Kinet., 22, 399-414.
236. Nicovich, J. M. and P. H. Wine, 1987, J. Phys. Chem., 91, 5118-5123.
237. Nicovich, J. M., P. H. Wine and A. R. Ravishankara, 1988, J. Chem. Phys., 89, 5670-
5679.
238. Ongstad, A. P. and J. W. Birks, 1984, J. Chem. Phys., 81, 3922-3930.
239. Ongstad, A. P. and J. W. Birks, 1986, J. Chem. Phys., 85, 3359-3368.
240. Oppliger, R., A. Allanic and M. J. Rossi, 1997, J. Phys. Chem. A, 101, 1903-1911.
241. Orlando, J. J. and J. B. Burkholder, 1995, J. Phys. Chem., 99, 1143-1150.
242. Orlando, J. J. and G. S. Tyndall, 1996, J. Phys. Chem., 100, 19398-19405.
243. Orlando, J. J., G. S. Tyndall, C. A. Cantrell and J. G. Calvert, 1991, J. Chem. Soc. Far.
Trans., 87, 2345-2349.
244. Parthiban, P. and T. Lee, 1998, J. Chem. Phys., 109, 525-530.
245. Patrick, R. and D. M. Golden, 1983, Int. J. Chem. Kinet., 15, 1189-1227.
246. Paulson, S. E., J. J. Orlando, G. S. Tyndall and J. G. Calvert, 1995, Int. J. Chem. Kinet.,
27, 997-1008.
247. Percival, C. J., G. D. Smith, L. T. Molina and M. J. Molina, 1997, J. Phys. Chem. A, 101,
8830-8833.
70
248. Permien, T., R. Vogt and R. N. Schindler. Mechanisms of Gas Phase-Liquid Phase
Chemical Transformations. In Air Pollution Report #17; Cox, R. A., Ed.; Environmental
Research Program of the CEC.: Brussels, 1988.
249. Perner, D., A. Schmeltekopf, R. H. Winkler, H. S. Johnston, J. G. Calvert, C. A. Cantrell
and W. R. Stockwell, 1985, J. Geophys. Res., 90, 3807-3812.
250. Pilgrim, J. S., A. McIlroy and C. A. Taatjes, 1997, J. Phys. Chem. A, 101, 1873-1880.
251. Poulet, G., I. T. Lancar, G. Laverdet and G. Le Bras, 1990, J. Phys. Chem., 94, 278-284.
252. Poulet, G., G. Laverdet and G. Le Bras, 1986, J. Phys. Chem., 90, 159-165.
253. Poulet, G., G. Le Bras and J. Combourieu, 1974, J. Chim. Physique, 71, 101-106.
254. Pritchard, H. O., 1994, Int. J. Chem. Kinet., 26, 61-72.
255. Pritchard, H. O., J. B. Pyke and A. F. Trotman-Dickenson, 1954, J. Amer. Chem. Soc., 76,
1201-1202.
256. Pritchard, H. O., J. B. Pyke and A. F. Trotman-Dickenson, 1955, J. Amer. Chem. Soc., 77,
2629-2633.
257. Pueschel, R. F., D. F. Blake, A. G. Suetsinger, A. D. A. Hansen, S. Verma and K. Kato,
1992, Geophys. Res. Lett., 19, 1659-1662.
258. Quinlan, M. A., C. M. Reihs, D. M. Golden and M. A. Tolbert, 1990, J. Phys. Chem., 94,
3255-3260.
259. Rattigan, O. V., D. J. Lary, R. L. Jones and R. A. Cox, 1996, J. Geophys. Res., 101,
23021-23033.
260. Ravishankara, A. R., F. L. Eisele and P. H. Wine, 1983, J. Chem. Phys., 78, 1140-1144.
261. Ravishankara, A. R., G. Smith, R. T. Watson and D. D. Davis, 1977, J. Phys. Chem., 81,
2220-2225.
262. Ravishankara, A. R. and P. H. Wine, 1980, J. Chem. Phys., 72, 25-30.
263. Ravishankara, A. R., P. H. Wine and A. O. Langford, 1979, J. Chem. Phys., 70, 984-989.
264. Ravishankara, A. R., P. H. Wine and J. M. Nicovich, 1983, J. Chem. Phys., 78, 6629-
6639.
265. Ravishankara, A. R., P. H. Wine, J. R. Wells and R. L. Thompson, 1985, Int. J. Chem.
Kinet., 17, 1281-1297.
266. Rawlins, W. T., G. E. Caledonia and R. A. Armstrong, 1987, J. Chem. Phys ., 87, 5209-
5213 .
267. Ray, G. W. and R. T. Watson, 1981, J. Phys. Chem., 85 , 1673-1676.
268. Rayez, M. T. and M. Destriau, 1993, Chem. Phys. Lett., 206, 278-284.
269. Robertshaw, J. S. and I. W. M. Smith, 1982, J. Phys. Chem., 86, 785.
270. Robinson, G. N., D. R. Worsnop, J. T. Jayne, C. E. Kolb and P. Davidovits, 1996, J.
Geophys. Res., 102, 3583-3601.
271. Robinson, G. N., D. R. Worsnop, J. T. Jayne, C. E. Kolb, E. Swartz and P. Davidovits,
1998, J. Geophys. Res., 103, 25,371-25,381.
272. Robinson, G. N., D. R. Worsonop, J. T. Jayne, C. E. Kolb and P. Davidovits, 1997, J.
Geophys. Res., 102, 3583-3601.
273. Rossi, M. J., R. Malhotra and D. M. Golden, 1987, Geophys. Res. Lett., 14, 127-130.
274. Rozenshtein, V. B., Y. M. Gershenzon, S. O. Il'in and O. P. Kishkovitch, 1984, Chem.
Phys. Lett., 112, 473-478.
275. Sander, S. P., R. P. Friedl and Y. L. Yung, 1989, Science, 245, 1095-1098.
276. Sander, S. P. and R. R. Friedl, 1989, J. Phys. Chem., 93, 4764-4771.
71
277. Sander, S. P., R. R. Friedl, W. B. DeMore, D. M. Golden, M. J. Kurylo, R. F. Hampson,
R. E. Huie, G. K. Moortgat, A. R. Ravishankara, C. E. Kolb and M. J. Molina Chemical
Kinetics and Photochemical Data for Use in Stratospheric Modeling, Evaluation Number
13, JPL Publication 00-3, Jet Propulsion Laboratory, California Institute of Technology,
Pasadena, CA, 2000.
278. Sander, S. P., G. W. Ray and R. T. Watson, 1981, J. Phys. Chem., 85, 199.
279. Sawerysyn, J.-P., C. Lafage, B. Meriaux and A. Tighezza, 1987, J. Chim. Phys., 84, 1187-
1193.
280. Schieferstein, M., K. Kohse-Hinghaus and F. Stuhl, 1983, Ber. Bunsenges. Phys. Chem.,
87, 361-366.
281. Schott, G. and N. Davidson, 1958, J. Amer. Chem. Soc., 80, 1841-1853.
282. Schurath, U., H. H. Lippmann and B. Jesser, 1981, Ber. Bunsenges. Phys. Chem., 85, 807-
813.
283. Schwab, J. J., W. H. Brune and J. G. Anderson, 1989, J. Phys. Chem., 93, 1030-1035.
284. Schwab, J. J., D. W. Toohey, W. H. Brune and J. G. Anderson, 1984, J. Geophys. Res.,
89, 9581-9587.
285. Schwartz, S. E., 1988, Atmos. Environ., 22, 2331.
286. Schweitzer, F., P. Mirabel and C. George, 1998, J. Phys. Chem. A, 102, 3942-3952.
287. Seeley, J. V., J. T. Jayne and M. J. Molina, 1996, J. Phys. Chem., 100, 4019-4025.
288. Sharkey, P. and I. W. M. Smith, 1993, J. Chem. Soc. Faraday Trans., 89, 631-638.
289. Shi, Q., P. Davidovits, J. T. Jayne, C. E. Kolb and D. R. Worsnop, 2000, submitted to J.
Geophys. Res., .
290. Silvente, E., R. C. Richter, M. Zheng, E. S. Saltzman and A. J. Hynes, 1997, Chem. Phys.
Lett., 264, 309-315.
291. Sinha, A., E. R. Lovejoy and C. J. Howard, 1987, J. Chem. Phys., 87, 2122-2128.
292. Slanger, T. G., B. J. Wood and G. Black, 1973, Int. J. Chem. Kinet., 5, 615-620.
293. Slanina, Z. and F. Uhlik, 1991, Chem. Phys. Lett., 182, 51-56.
294. Smith, C. A., L. T. Molina, J. J. Lamb and M. J. Molina, 1984, Int. J. Chem. Kinet., 16,
41-55.
295. Smith, C. A., A. R. Ravishankara and P. H. Wine, 1985, J. Phys. Chem., 89, 1423-1427.
296. Smith, G. P. and D. M. Golden, 1978, Int. J. Chem. Kinet., 10, 489-501.
297. Smith, I. W. M. and M. D. Williams, 1986, J. Chem. Soc. Faraday Trans. 2, 82, 1043-
1055.
298. Smith, I. W. M. and R. Zellner, 1974, J. Chem. Soc.. Faraday Trans. 2, 70, 1045-1056.
299. Sridharan, U. C., F. S. Klein and F. Kaufman, 1985, J. Chem. Phys., 82, 592-593.
300. Sridharan, U. C., L. X. Qiu and F. Kaufman, 1981, J. Phys. Chem., 85, 3361-3363.
303. Stachnik, R. A., M. J. Molina and L. T. Molina, 1986, J. Phys. Chem., 90, 2777-2780.
304. Stanton, J. F. and R. J. Bartlett, 1993, J. Chem. Phys., 98, 9335-9339.
305. Stanton, J. F., C. M. L. Rittby, R. J. Bartlett and D. W. Toohey, 1991, J. Phys. Chem., 95,
2107-2110.
306. Stedman, D. H. and H. Niki, 1973, J. Phys. Chem., 77, 2604-2609.
307. Tabazadeh, A., O. B. Toon, S. L. Clegg and P. Hammill, 1997, Geophys. Res. Lett., 24,
1931-1934.
308. Takacs, G. A. and G. P. Glass, 1973, J. Phys. Chem., 77, 1948-1951.
72
309. Takahashi, K., M. Kishigami, Y. Matsumi, M. Kawasaki and A. J. Orr-Ewing, 1996, J.
Chem. Phys., 105, 5290-5293.
310. Takahashi, K., M. Kishigami, N. Taniguchi, Y. Matsumi and M. Kawasaki, 1997, J.
Chem. Phys., 106, 6390-6397.
311. Takahashi, K., Y. Matsumi and M. Kawasaki, 1996, J. Phys. Chem., 100, 4084-4089.
312. Takahashi, K., R. Wada, Y. Matsumi and M. Kawasaki, 1996, J. Phys. Chem., 100,
10145-10149.
313. Talukdar, R. K., M. K. Gilles, F. Battin-Leclerc, A. R. Ravishankara, J.-M. Fracheboud, J.
J. Orlando and G. S. Tyndall, 1997, Geophys. Res. Lett., 24, 1091-1094.
314. Talukdar, R. K., C. A. Longfellow, M. K. Gilles and A. R. Ravishankara, 1998, Geophys.
Res. lett., 25, 143-146.
315. Temps, F. and H. G. Wagner, 1982, Ber. Bunsenges. Phys. Chem., 86, 119-125.
316. Thompson, A. M. and R. W. Stewart, 1991, J. Geophys. Res., 96, 13,089-13,108.
317. Thorn, R. P., E. P. Daykin and P. H. Wine, 1993, Int J. Chem. Kinet., 25, 521-537.
318. Thrush, B. A. and J. P. T. Wilkinson, 1981, Chem. Phys. Lett., 81, 1-3.
319. Tolbert, M. A., M. J. Rossi and D. M. Golden, 1988, Geophys. Res. Lett., 15, 847-850.
320. Tolbert, M. A., M. J. Rossi, R. Malhotra and D. M. Golden, 1987, Science, 238, 1258-
1260.
321. Toohey, D. W. and J. G. Anderson, 1988, J. Phys. Chem., 92, 1705-1708.
322. Toon, O., E. Browell, B. Gray, L. Lait, J. Livingston, P. Newman, R. P. P. Russell, M.
Schoeberl, G. Toon, W. Traub, F. P. J. Valero, H. Selkirk and J. Jordan, 1993, Science,
261, 1136-1140.
323. Troe, J., 1977, J. Chem. Phys., 66, 4745 .
324. Trolier, M., R. L. Mauldin, III and A. R. Ravishankara, 1990, J. Phys. Chem., 94, 4896-
4907.
325. Trolier, M. and J. R. Wiesenfeld, 1988, J. Geophys. Res., 93, 7119-7124.
326. Tuazon, E. C., E. Sanhueza, R. Atkinson, W. P. L. Carter, A. M. Winer and J. N. Pitts, Jr.,
1984, J. Phys. Chem., 88, 3095-3098.
327. Turnipseed, A. A., J. W. Birks and J. G. Calvert, 1991, J. Phys. Chem., 95, 4356-4364.
328. Van Doren, J. M., L. R. Watson, P. Davidovits, D. R. Worsnop, M. S. Zahniser and C. E.
Kolb, 1990, J. Phys. Chem., 94, 3265-3269.
329. Van Doren, J. M., L. R. Watson, P. Davidovits, D. R. Worsnop, M. S. Zahniser and C. E.
Kolb, 1991, J. Phys. Chem., 95, 1684-1689.
330. Vanderzanden, J. W. and J. W. Birks, 1982, Chem. Phys. Lett., 88, 109-114.
331. Viggiano, A. A., J. A. Davidson, F. C. Fehsenfeld and E. E. Ferguson, 1981, J. Chem.
Phys., 74, 6113-6125.
332. Vogt, R. and R. N. Schindler, 1992, J. Photochem. Photobiol. A: Chem., 66, 133-140.
333. Volltrauer, H. N., W. Felder, R. J. Pirkle and A. Fontijn, 1979, J. Photochem., 11, 173-
181.
334. Walker, R. W. . In Ph.D. Thesis; Queen Mary College University of London:, 1972.
335. Wallington, T. J., R. Atkinson, A. M. Winer and J. N. Pitts, Jr., 1987, Int. J. Chem. Kinet.,
19, 243-249.
336. Wallington, T. J. and R. A. Cox, 1986, J. Chem. Soc. Faraday Trans. 2, 82, 275-289.
337. Wang, X., M. Suto and L. C. Lee, 1988, J. Chem. Phys., 88 , 896-899.
338. Wangberg, I., T. Etzkorn, I. Barnes, U. Platt and K. H. Becker, 1997, J. Phys. Chem. A,
101, 9694-9698.
73
339. Wangberg, I., E. Ljungstrom, B. E. R. Olsson and J. Davidsson, 1992, J. Phys. Chem., 96,
7640-7645.
340. Waschewsky, G. C. G. and J. P. D. Abbatt, 1999, J. Phys. Chem. A, 103, 5312-5320.
341. Watson, L. R., J. M. V. Doren, P. Davidovits, D. R. Worsnop, M. S. Zahniser and C. E.
Kolb, 1990, J. Geophys. Res., 95, 5631-5638.
342. Watson, R. T., G. Machado, S. Fischer and D. D. Davis, 1976, J. Chem. Phys., 65, 2126-
2138.
343. Whytock, D. A., J. H. Lee, J. V. Michael, W. A. Payne and L. J. Stief, 1977, J. Chem.
Phys., 66, 2690-2695.
344. Whytock, D. A., J. V. Michael and W. A. Payne, 1976, Chem. Phys. Lett., 42, 466-471.
345. Wine, P. H., N. M. Kreutter and A. R. Ravishankara, 1979, J. Phys. Chem., 83, 3191.
346. Wine, P. H. and A. R. Ravishankara, 1982, Chem. Phys., 69, 365-373.
347. Wolff, E. W. and R. Mulvaney, 1991, Geophys. Res. Lett., 18, 1007-1010.
348. Worsnop, D. R., L. E. Fox, M. S. Zahniser and S. C. Wofsy, 1993, Science, 259, 71-74.
349. Yarwood, G., J. W. Sutherland, M. A. Wickramaaratchi and R. B. Klemm, 1991, J. Phys.
Chem., 95, 8771-8775.
350. Zahniser, M. S., B. M. Berquist and F. Kaufman, 1978, Int. J. Chem. Kinet., 10, 15-29.
351. Zahniser, M. S., J. Chang and F. Kaufman, 1977, J. Chem. Phys., 67, 997-1003.
352. Zahniser, M. S. and C. J. Howard, 1980, J. Chem. Phys., 73, 1620-1626.
353. Zahniser, M. S. and F. Kaufman, 1977, J. Chem. Phys., 66, 3673-3681.
354. Zahniser, M. S., F. Kaufman and J. G. Anderson, 1974, Chem. Phys. Lett., 27, 507-510.
355. Zahniser, M. S., F. Kaufman and J. G. Anderson, 1976, Chem. Phys. Lett., 37, 226-231.
356. Zellner, R., G. Wagner and B. Himme, 1980, J. Phys. Chem., 84, 3196-3198.
357. Zhang, R., J. T. Jayne and M. J. Molina, 1994, J. Phys. Chem., 98, 867-874.
358. Zhang, R., M.-T. Leu and L. F. Keyser, 1994, J. Phys. Chem., 98, 13,563-13,574.
359. Zhang, R., M.-T. Leu and L. F. Keyser, 1995, Geophys. Res. Lett., 22, 1493-1496.
360. Zhang, R., P. J. Wooldridge and M. J. Molina, 1993, J. Phys. Chem., 97, 8541-8548.
361. Zolensky, M. E., D. S. McKay and L. A. Kaczor, 1989, J. Geophys. Res., 94, 1047-1056.
74

NASA Atmos 6mar00

Uploaded by

Copyright:

Available Formats

NASA Atmos 6mar00

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

NASA Atmos 6mar00

Uploaded by

Copyright:

Available Formats

JPL Publication 00-3

Chemical Kinetics and Photochemical Data

NASA Panel for Data Evaluation:

Jet Propulsion Laboratory

Table I-1: Editions of this Publication

Communications regarding particular reactions may be addressed to the appropriate panel

Scope of the Evaluation

Format of the Evaluation

For T < 298 K, multiply by the factor

and divide by the factor

Examples of symmetric and asymmetric error limits are shown in Figure 1.

a Units are cm3 molecule-1 s-1

Low-Pressure Limiting Rate Constant [kxo(T) ]

Temperature Dependence of Low-Pressure Limiting Rate Constants: Tn

High-Pressure Limit Rate Constants [k(T)]

Temperature Dependence of High-Pressure Limiting Rate Constants: Tm

Low Pressure Limita High Pressure Limitb

k f ([M],T) = ko (T) [M ] 0.6

HONO2: ko(300) = 2.40x10-30; n = 2.6; k(300) =1.8x10-11; m = 0.1

A=1.2x10-26; B = 8260, K(298K) = 1.3x10-14 and f(298K) = 10

K(T)/cm3 molecule-1 = A exp(B/T) (200 < T < 300 K)

The third column entry in Table 3 is the calculated value of K at 298 K.

log10 A=log10 K ( T ) B ( 2.303T )

A = (eR'T/Nav) exp(So/R) = 3.7 x 10-22 T exp(So/R)

NO2 + NO3 N2O5 3.0x10-27 10991200 3.1x10-11 1.2 4

ClO + ClO ClOOCl 1.27x10-27 8744500 7.0x10-15 1.3 11

K/cm3 molecule-1 = A exp (B/T) [200 < T/K < 300]

In this equation, is in nm, T in K, and kT is in wavenumbers (i.e., k = 0.695 cm-1 K-1).

Recommended absorption cross sections in the wavelength range 190-450 nm for

( ) = 24.77 exp 109.80 ln + 12.22 exp 93.63 ln

We have evaluated and tabulated the currently available information on heterogeneous

Figure 3: Recommended Reactive Uptake Coefficients as a Function of Temperature

HOCl + HCl(s) Cl2 + H2O

HOCl Water Ice H2O(s) HCl(s) 195-200 0.2 2 40

ClONO2 + H2O(s) HOCl + HNO3

ClONO2 Water Ice H2O(s) 180-200 0.3 3 43

ClONO2 + HCl(s) Cl2 + HNO3

ClONO2 Water Ice H2O(s) 180-200 0.3 3 47

HOBr + HCl(s) BrCl + H2O

HOBr Water Ice H2O(s) HBr(s) 228 0.3 3 57

BrONO2 + H2O HOBr + HNO3

BrONO2 Water Ice H2O(s) 190-200 0.3 2 59

ko = -25.5265 0.133188wt + 0.00930846wt2 9.0194x10-5wt3

where wt is the weight percentage of H2SO4.

The overall error is

with a factor of 2 uncertainty. Hanson and Ravishankara [136] also measured

Phenomenological Model of ClONO2 Hydrolysis and the Reaction

a1=12.37208932 a1=11.820654354 a1=-180.06541028

To 144.11 + 0.166 wt-0.015 wt 2 + 2.18x104 wt 3

exp[60.51 0.095 wt+0.0077 wt 2 1.61x105 wt3

bH 2O R=0.082 atm K -1M -1

1. Chlorofluoromethanes and the Stratosphere. In NASA Reference Publication 1010;

You might also like