1,1,1,3,3,-pentafluorobutane (HFC-365mfc): Atmospheric degradation and contribution to radiative forcing

Article abstract of DOI:10.1002/(SICI)1097-4601(1997)29:8<607::AID-KIN6>3.0.CO;2-Y

The rate constant for the reaction of the hydroxyl radical with 1,1,1,3,3-pentafluorobutane (HFC-365mfc) has been determined over the temperature range 278-323 K using a relative rate technique. The results provide a value of k(OH + CF₃CH₂CF₂CH₃) = 2.0 × 10^-12 exp(- 1750 ± 400/T) cm³ molecule^-1 s^-1 based on k(OH + CH₃CCl₃) = 1.8 × 10^-12 exp (- 1550 ± 150/T) cm³ molecule^-1 s^-1 for the rate constant of the reference reaction. Assuming the major atmospheric removal process is via reaction with OH in the troposphere, the rate constant data from this work gives an estimate of 10.8 years for the tropospheric lifetime of HFC-365mfc. The overall atmospheric lifetime obtained by taking into account a minor contribution from degradation in the stratosphere, is estimated to be 10.2 years. The rate constant for the reaction of Cl atoms with 1,1,1,3,3-pentafluorobutane was also determined at 298 ± 2 K using the relative rate method, k(Cl + CF₃CH₂CF₂CH₃) = (1.1 ± 0.3) × 10^-15 cm³ molecule^-1 s^-1. The chlorine initiated photooxidation of CF₃CH₂CF₂CH₃ was investigated from 273-330 K and as a function of O₂ pressure at 1 atmosphere total pressure using Fourier transform infrared spectroscopy. Under all conditions the major carbon-containing products were CF₂O and CO₂, with smaller amounts of CF₃O₃CF₃. In order to ascertain the relative importance of hydrogen abstraction from the - CH₂ - and - CH₃ groups in CF₃CH₂CF₂CH₃, rate constants for the reaction of OH radicals and Cl atoms with the structurally similar compounds CF₃CH₂CCl₂F and CF₃CH₂CF₃ were also determined at 298 K k(OH + CF₃CH₂CCl₂F) = (8 ± 3) × 10^-16 cm³ molecule^-1 s^-1; k(OH + CF₃CH₂CF₃) = (3.5 ± 1.5) × 10^-16 cm³ molecule^-1 s^-1; k(Cl + CF₃CH₂CCl₂F) = (3.5 ± 1.5) × 10^-17 cm³ molecule^-1 s^-1; k(C; + CF₃CH2F3) < 1 × 10^-17 cm³ molecule^-1 s^-1. The results indicate that the most probable site for H-atom abstraction from CF₃CH₂CF₂CH₃ is the methyl group and that the formation of carbonyl compounds containing more than a single carbon atom will be negligible under atmospheric conditions, carbonyl difluoride and carbon dioxide being the main degradation products. Finally, accurate infrared absorption cross-sections have been measured for CF₃CH₂CF₂CH₃, and jointly used with the calculated overall atmospheric lifetime of 10.2 years, in the NCAR chemical-radiative model, to determine the radiative forcing of climate by this CFC alternative. The steady-state Halocarbon Global Warming Potential, relative to CFC-11, is 0.17. The Global Warming Potentials relative to CO₂ are found to be 2210, 790, and 250, for integration time-horizons of 20, 100, and 500 years, respectively.

Full text of DOI:10.1002/(SICI)1097-4601(1997)29:8<607::AID-KIN6>3.0.CO;2-Y

1,1,1,3,3,-
Pentafluorobutane (HFC-
365mfc): Atmospheric
Degradation and
Contribution to Radiative
Forcing
JOHN BARRY,1 GARRETT LOCKE,2 DONNCHA SCOLLARD,3 HOWARD SIDEBOTTOM,4 JACK
TREACY,5 CATHY CLERBAUX,6 REGINALD COLIN,7 and JAMES FRANKLIN8
1,2,3,4Department of Chemistry, University College Dublin, Dublin, Ireland; 5Department of Chemistry, Dublin Institute of
Technology, Dublin, Ireland; 6,7Laboratoire de Chimie Physique Moléculaire, Université Libre de Bruxelles, Brussels, Belgium;
8Solvay S.A., Brussels, Belgium
Received 3 June 1996; accepted 2 January 1997
ABSTRACT: The rate constant for the reaction of the hydroxyl radical with 1,1,1,3,3-pentafluo-
robutane (HFC-365mfc) has been determined over the temperature range 278–323 K using a
relative rate technique. The results provide a value of k(OH ϩ CF3CH2CF2CH3) ϭ 2.0 ϫ 10Ϫ12
exp(Ϫ 1750 Ϯ 400/T) cm3 moleculeϪ1 sϪ1 based on k(OH ϩ CH3CCl3) ϭ 1.8 ϫ 10Ϫ12 exp
(Ϫ1550 Ϯ 150/T) cm3 moleculeϪ1 sϪ1 for the rate constant of the reference reaction. Assum-
ing the major atmospheric removal process is via reaction with OH in the troposphere, the
rate constant data from this work gives an estimate of 10.8 years for the tropospheric lifetime
of HFC-365mfc. The overall atmospheric lifetime obtained by taking into account a minor
contribution from degradation in the stratosphere, is estimated to be 10.2 years. The rate
constant for the reaction of Cl atoms with 1,1,1,3,3-pentafluorobutane was also determined
at 298 Ϯ 2 K using the relative rate method, k(Cl ϩ CF3CH2CF2CH3) ϭ (1.1 Ϯ 0.3) ϫ 10Ϫ15
cm3 moleculeϪ1 sϪ1. The chlorine initiated photooxidation of CF3CH2CF2CH3 was investigated
from 273–330 K and as a function of O2 pressure at 1 atmosphere total pressure using
Fourier transform infrared spectroscopy. Under all conditions the major carbon-containing
products were CF2O and CO2 , with smaller amounts of CF3O3CF3 . In order to ascertain the rel-
ative importance of hydrogen abstraction from the !CH2 ! and !CH3 groups in
CF3CH2CF2CH3 , rate constants for the reaction of OH radicals and Cl atoms with the struc-
turally similar compounds CF3CH2CCl2F and CF3CH2CF3 were also determined at 298 K k(OH
ϩ CF3CH2CCl2F) ϭ (8 Ϯ 3) ϫ 10Ϫ16 cm3 moleculeϪ1
s
Ϫ1; k(OH ϩ CF3CH2CF3) ϭ (3.5 Ϯ 1.5) ϫ
Ϫ1];
10Ϫ16 cm3 moleculeϪ1 Ϫ1; k(Cl ϩ CF3CH2CCl2F) ϭ (3.5 Ϯ 1.5) ϫ 10Ϫ17 cm3 moleculeϪ1
s
s
k(Cl ϩ CF3CH2CF3) Ͻ 1 ϫ 10Ϫ17 cm3 moleculeϪ1 sϪ1. The results indicate that the most proba-
ble site for H-atom abstraction from CF3CH2CF2CH3 is the methyl group and that the forma-
tion of carbonyl compounds containing more than a single carbon atom will be negligible un-
der atmospheric conditions, carbonyl difluoride and carbon dioxide being the main
degradation products. Finally, accurate infrared absorption cross-sections have been mea-

608
BARRY ET AL.
sured for CF3CH2CF2CH3 , and jointly used with the calculated overall atmospheric lifetime of
10.2 years, in the NCAR chemical-radiative model, to determine the radiative forcing of cli-
mate by this CFC alternative. The steady-state Halocarbon Global Warming Potential, relative
to CFC-11, is 0.17. The Global Warming Potentials relative to CO2 are found to be 2210, 790,
and 250, for integration time-horizons of 20, 100, and 500 years, respectively. © 1997 John Wiley
& Sons, Inc. Int J Chem Kinet 29: 607–617, 1997.
INTRODUCTION
EXPERIMENTAL
1,1,1,3,3-pentafluorobutane is envisioned as an even-
tual replacement for CCl2FCH3 (HCFC-141b) which
is used as a blowing agent for polyurethane foams.
HCFC-141b is itself the main substitute for CFC-11
(CCl3F) in this application, but under the provisions
of the Montreal Protocol on Substances that Deplete
the Ozone Layer, its uses will be phased down start-
ing in 2004, with virtual elimination occurring by
2020. Earlier phase-out will be required in the USA,
the European Union, and Japan. The long atmos-
pheric lifetimes of CFCs allow them to be transported
to the stratosphere, where they release chlorine, re-
sulting in the catalytic destruction of ozone [1–3].
Hydrochlorofluorocarbons (HCFCs) and hydrofluoro-
carbons (HFCs) resemble CFCs in their physical and
chemical characteristics, but the presence of at least
one hydrogen atom in these compounds means that
they are degraded largely in the troposphere follow-
ing reaction with the hydroxyl radical [4–6]. Al-
though the replacement of CFCs by HCFCs results in
considerably reduced ozone loss, the presence of
chlorine in these compounds means that the small
amounts entering the stratosphere give rise to some
ozone depletion. Since HFCs do not contain chlorine,
their contribution to stratospheric ozone depletion is
expected to be negligible; however, it is important to
determine their environmental impact with respect to
their Global Warming Potential (GWP) and degrada-
tion products.
Rate Constants and Degradation
Mechanisms
Relative rate and product studies were carried out at
atmospheric pressure in an FEP Teflon cylindrical re-
action vessel (volume ca. 50 litre) surrounded by flu-
orescent lamps. Photolysis experiments in the range
300–450 nm were carried out using 10 sunlamps
(Philips TL 20/09) and 10 blacklamps (Philips TL
20/08). Twenty Philips TUV 15W germicidal lamps
were used as a source of radiation at 254 nm. The
vessel and lamps were enclosed in a thermostatically
controlled chamber which allowed reactions to be
studied over the temperature range 273–330 K. Elec-
tric fans positioned below the reaction vessel helped
provide a uniform reaction temperature, which was
monitored by two chromel/alumel thermocouples
placed within the chamber. Light intensity was varied
by switching off various sets of lamps. All pressure
measurements were made using MKS Baratron ca-
pacitance manometers. Measured amounts of
reagents were flushed from calibrated Pyrex bulbs
into the reaction vessel by a stream of zero-grade ni-
trogen (Air Products); the vessel was then filled to its
full capacity with either ultra-pure air (Air Products)
or mixtures of zero-grade nitrogen and zero-grade
oxygen (Air Products). Reaction mixtures were al-
lowed to mix for at least 30 min prior to the start of
photolysis. Quantitative analyses were carried out us-
ing gas chromatography (Shimadzu Model 14A, with
flame ionization detection) and FTIR spectroscopy
(Mattson Polaris). Samples of reaction mixtures were
drawn through a Valco gas sampling valve for gas
chromatographic analysis. Infrared spectra were ob-
tained using an evacuable 2 litre Teflon-coated Wilks
cell, containing a multipass White mirror arrange-
ment, mounted in the cavity of the spectrometer. Af-
ter various periods of photolysis, the reaction mix-
tures were expanded into the cell through 3 mm inner
diameter Teflon tubing. Spectra were recorded using
a 10 m path length at 5 minute intervals, the spectral
range was 500 to 3500 cmϪ1 with a resolution of 4
cmϪ1 and the spectra derived from 128 scans. Refer-
ence spectra and calibration curves for each com-
The purpose of this work was to determine the rate
constant for the reaction of OH radicals with
CF3CH2CF2CH3 as a function of temperature, to
identify the products arising from its oxidation under
atmospheric conditions, to measure the infrared ab-
sorption cross-sections of CF3CH2CF2CH3 and to cal-
culate its Global Warming Potential. These data are
required in order to evaluate the environmental im-
pact of CF3CH2CF2CH3 .
Correspondence to: H. Sidebottom
Contract grant Sponsor: Belgian Federal Office for Scientific,
Technical and Cultural Affairs (Global Change Programme)
© 1997 John Wiley & Sons, Inc.
CCC 0538-8066/97/080607-11

ATMOSPHERIC DEGRADATION
609
O(1D) ϩ H2O 9: OH ϩ OH
(4)
pound were obtained by expanding known pressures
of authentic samples of the compounds into the IR
cell. Ozone concentrations were determined by a
Monitor Labs 8810 UV photometric ozone analyzer.
Assuming that substrate and reference are consumed
only by OH radicals it can be shown that:
ln{[CF3CH2CF2CH3]0/[CF3CH2CF2CH3]t}
Infrared Absorption Cross-Sections
ϭ k1/k2ln{[CH3CCl3]0/[CH3CCl3]t} (I)
High-resolution absorption cross-sections for HFC-
365mfc were determined at room temperature in the
infrared atmospheric window spectral range. The in-
frared spectra were recorded using the procedure and
set-up described in detail in previous publications
[7,8]. The gas was introduced into a laboratory-built
thermostated cell [9] maintained at 287 Ϯ 0.5 K. The
cell was placed inside the air evacuated chamber of
the FTIR spectrometer (Bruker IFS120HR). Infrared
spectra were obtained between 600 and 1500 cmϪ1 at
0.03 cmϪ1 spectral resolution. For each of the 6 pres-
sures investigated, ranging from 1 to 5 torr, 100 scans
of the moving mirror were co-added in order to ob-
tain a good signal-to-noise ratio.
where the subscripts 0 and t indicate concentrations
before irradiation and at time t, respectively. Direct
photolytic loss of CF3CH2CF2CH3 and CH3CCl3 was
unimportant and the reaction mixtures were stable in
the dark for at least 5 h.
Mixtures of O3 , H2O, CF3CH2CF2CH3 , and
CH3CCl3 in air were photolyzed for about 25 min
in the temperature range 278–323 K with
[CF3CH2CF2CH3]0 and [CH3CCl3]0 ϭ 1–10 ppm;
[O3]0 ϭ 100–1,000 ppm; and [H2O]0 ϭ 2,000–
10,000 ppm, (1 ppm ϭ 2.46 ϫ 1013 molecule cmϪ3
at 298 K, and 760 torr total pressure). Under the reac-
tion conditions employed the substrate and reference
compound were found to decay by approximately
20% during the experiments. Data for the runs were
plotted in the form of eq. (I), Figure 1, and show the
expected linear relationship. The rate constant
k1/k2 ϭ 0.58 Ϯ 0.04, (error is Ϯ 2␴ and represents
precision only), was independent of relative reactant
concentration, reaction time, and light intensity in
agreement with the proposed mechanism. A possible
source of error is the reaction of O(1D) with
CF3CH2CF2CH3 and CH3CCl3 . The rate constant
for the reaction of O(1D) with H2O,
k4 ϭ 2.2 ϫ 10Ϫ10 cm3 moleculeϪ1 sϪ1 [10] is suffi-
ciently high that under the experimental conditions
employed with H2O Ն 100 [CF3CH2CF2CH3]0 or
[CH3CCl3]0 , O(1D) is efficiently scavenged in reac-
tion (4). Based on the rate of reactant loss in the reac-
Materials
1,1,1,3,3-pentafluorobutane (99 wt.%), 1,1,1,3,3-
hexafluoropropane (Ͼ97 mol%), and 1,1-dichloro-
1,3,3,3-tetrafluoropropane (99.8 mol%) were ob-
tained from Solvay S.A., Belgium and vacuum
distilled. Chlorine, HCl (Matheson, research grade),
CF2O, CClFO, CF3COCF3 , CH3CCl2F, (Fluorochem
Ltd.), and 1,1,1-trichloroethane (Fluka) were used as
received. Ozone was produced by passing zero-grade
oxygen through an ozone generator (Monitor Labs)
directly into the reaction chamber.
RESULTS AND DISCUSSION
Rate Constants and Degradation
Mechanisms
(i) Relative Rate Studies: The rate constant for the re-
action of OH radicals with CF3CH2CF2CH3 was de-
termined relative to that for reaction with CH3CCl3
OH ϩ CF3CH2CF2CH3 9: products
OH ϩ CH3CCl3 9: products
(1)
(2)
Ozone was photolyzed in the presence of water vapor
to generate OH radicals.
Figure 1 Concentration-time data for the reaction of OH
radicals with CF3CH2CF2CH3 and CH3CCl3 at 298 K.
O3 ϩ h
␯
(␭
ϭ 254 nm) 9: O(1D) ϩ O2 (3)

610
BARRY ET AL.
tor, the steady-state concentrations of OH and O(1D)
are estimated to be around 2.5 ϫ 1010 and 2.5 ϫ 102
molecules cmϪ3, respectively. Hence depletion of
CF3CH2CF2CH3 and CH3CCl3 by reaction with
O(1D) will be unimportant compared to loss by reac-
tion with OH. Variations in water vapor pressure from
1–10 torr had no measurable effect on the value of
k1/k2 , thus providing support for this assumption.
Chlorine atoms are generated from the reaction of
OH with CH3CCl3 in the presence of O2 [11], how-
ever, this species will be rapidly removed by reaction
with O3 .
moleculeϪ1 sϪ1, is almost a factor of 1.5 higher than
the rate constant determined in this work. DeMore
[17] has recently described an empirical method for
the estimation of the rate constants for reaction of OH
with halocarbons based on the available experimental
data for a wide range of halogenated methanes and
ethanes. Typically, the rate constants derived from the
method are within a factor of 1.3 of the experimental
values. A value of k1 ϭ 4.9 ϫ 10Ϫ15 cm3 moleculeϪ1
sϪ1 was calculated using the group contributions
given by DeMore [17] and is in reasonable agreement
with the experimental rate constant obtained in this
study. The higher rate constant value determined by
Mellouki et al. [16] is likely to be due to the presence
of reactive impurities in the sample of
CF3CH2CF2CH3 used by these authors. The tropos-
pheric lifetime for HFC-365mfc, which may be as-
sumed to degrade essentially by reaction with the OH
radical in the troposphere, may be calculated by
means of the equation:
Using recent absolute rate constant measurements
for the reaction of OH ϩ CH3CCl3 at 298 K, (k2 ϭ
1.0 ϫ 10Ϫ14 [12], k2 ϭ 0.95 ϫ 10Ϫ14 [13], and
k2 ϭ 1.1 ϫ 10Ϫ14 [14] cm3 moleculeϪ1 sϪ1), pro-
vides a value for k1(OH ϩ CF3CH2CF2CH3) ϭ
(5.9 Ϯ 0.4) ϫ 10Ϫ15 cm3 moleculeϪ1 sϪ1 at 298 K.
The error in the rate constant for the reference reac-
tion probably adds a further 15% to the uncertainty
in k1 . The values of k1/k2 obtained at five different
temperatures over the range 278–323 K show little
variation with temperature and are plotted in
Arrhenius form in Figure 2. A linear least-squares
fit of the data gives k1(OH ϩ CF3CH2CF2CH3)/
k2(OH ϩ CH3CCl3) ϭ 1.13 exp(Ϫ202 Ϯ 200/T).
Taking a value of k2(OH ϩ CH3CCl3) ϭ 1.8 ϫ
10Ϫ12 exp(Ϫ1550 Ϯ 150/T ) cm3 moleculeϪ1 sϪ1 [15],
provides an estimate of k1(OH ϩ CF3CH2CF2CH3) ϭ
2.0 ϫ 10Ϫ12 exp(Ϫ 1750/T ) cm3 moleculeϪ1 sϪ1
with total estimated errors of ⌬(E/R) ϭ Ϯ 400 K.
Mellouki et al. [16] have reported a value of
␶
(HFC-365mfc)/␶(CH3CCl3)
ϭ k(OH ϩ CH3CCl3)/k(OH ϩ HFC-365mfc)
where ␶(HFC-365mfc) is the tropospheric lifetime of
HFC-365mfc, ␶(CH3CCl3) is the lifetime of CH3CCl3
with respect to reaction with OH in the troposphere,
and the OH rate constants refer to a temperature of
277 K [18]. Using the numerical values
␶
(CH3CCl3) ϭ 5.9 years [19], k(OH ϩ CH3CCl3) ϭ
1.8 ϫ 10Ϫ12 exp(Ϫ1550/T ) cm3 moleculeϪ1 sϪ1 [15],
and k(OH ϩ HFC-365mfc) from this work, one can
calculate that the tropospheric lifetime of HFC-
365mfc is 10.8 years. On the other hand, using the
value of k(OH ϩ HFC-365mfc) from [16], the life-
time would be 7.1 years. In order to determine the
overall atmospheric lifetime of HFC-365mfc, ␶, a
correction must be applied to the tropospheric life-
time ␶(HFC-365mfc), to allow for degradation in
k1 ϭ (1.68 Ϯ 0.21) ϫ 10Ϫ12exp(Ϫ1585 Ϯ 80/T ) cm3
Ϫ1
moleculeϪ1
s
using a pulsed laser photolysis-laser
induced fluorescence technique. The rate constant de-
termined at 298 K, k1 ϭ (8.69 Ϯ 0.74) ϫ 10Ϫ15 cm3
the stratosphere using the equation (1/
␶) ϭ
1/ (HFC-365mfc) ϩ 1/ (strat), where ␶(strat) is the
␶
␶
lifetime with respect to transport to the stratosphere
and degradation there. Degradation of HFC-365mfc
in the stratosphere is expected to be mainly initiated
via reaction with OH, since photolysis and reaction
with O(1D) are likely to play only a minor role [20].
An estimate of ␶(strat) ϭ 170 years for HFC-365mfc
was derived from the AER-2D model for structurally
similar HFCs [21] and provides an estimate of 10.2
years for the overall atmospheric lifetime of HFC-
365mfc.
In a similar series of experiments Cl2 was pho-
tolyzed in the presence of CF3CH2CF2CH3/reference
compound/air mixtures with reactant concentrations
Figure 2 Arrhenius plot for the ratio of rate constants for
the reactions of OH radicals with CF3CH2CF2CH3 and
CH3CCl3 .

ATMOSPHERIC DEGRADATION
611
CF3CH2CF2CH2O2 ϩ NO 9:
CF3CH2CF2CH2O ϩ NO2 (9)
CF3CHCF2CH3 ϩ O2 ϩ M 9:
CF3CHO2CF2CH3 ϩ M (10)
CF3CHO2CF2CH3 ϩ NO 9:
CF3CHOCF2CH3 ϩ NO2 (11)
An accurate study of product formation from the re-
action of OH radicals with CF3CH2CF2CH3 using
FTIR spectroscopy proved impossible on account of
interfering bands from the high concentrations of O3
and H2O required for the kinetic investigations of the
reaction. However, the reaction of Cl atoms with the
halocarbon can be used to produce the required
haloalkyl radicals.
Figure 3 Concentration-time data for the reaction of Cl
atoms with CF3CH2CF2CH3/CH3CCl2F and
CF3CH2CCl2F/CF3CH2CF2CH3 mixtures at 298 K.
Cl ϩ CF3CH2CF2CH3 9:
CF3CH2CF2CH2 ϩ HCl (6a)
in the range 1–20 ppm. Chlorine atoms react with
CF3CH2CF2CH3 and the reference, here CH3CCl2F:
Cl ϩ CF3CH2CF2CH3 9:
CF3CHCF2CH3 ϩ HCl (6b)
Cl2 ϩ h
␯
(␭
ϭ 300– 450 nm) 9: 2Cl
(5)
Cl2 (20–30 ppm) and CF3CH2CF2CH3 (5–10 ppm)
in mixtures of N2 and O2 were photolyzed over the
temperature range 278–330 K. The major products
identified in all the runs were CF2O, CO2 , and HCl
together with small amounts of CF3O3CF3 . Figure 4
gives the infrared spectra obtained following the pho-
tolysis for 30 min of 30 ppm Cl2 with 10 ppm
CF3CH2CF2CH3 in air at 760 torr and 298 K after
successive subtraction of reference spectra of
CF3CH2CF2CH3 and CF2O, respectively. The residual
spectrum indicates the presence of very small
Cl ϩ CF3CH2CF2CH3 9: products (6)
Cl ϩ CH3CCl2F 9: products (7)
and
ln{[CF3CH2CF2CH3]0/[CF3CH2CF2CH3]t}
ϭ k6/k7ln{[CH3CCl2F]0/[CH3CCl2F]t} (II)
Rate data plotted in the form of eq. (III) gave a good
straight line, Figure 3, with slope k6 /k7 ϭ 0.52 Ϯ 0.06.
Taking
a
value
of
k7(Cl ϩ CH3CCl2F) ϭ
(2.2 Ϯ 0.2) ϫ 10Ϫ15 cm3 moleculeϪ1 sϪ1 [22–25]
k6(Cl ϩ CF3CH2CF2CH3) ϭ1.1 ϫ 10Ϫ15 cm3 mole-
culeϪ1 sϪ1 at 298 K with an estimated uncertainty of
around 25%.
(ii) Degradation Studies: The reaction of OH radi-
cals with CF3CH2CF2CH3 takes place via hydrogen
atom abstraction:
OH ϩ CF3CH2CF2CH3 9:
CF3CH2CF2CH2 ϩ H2O (1a)
OH ϩ CF3CH2CF2CH3 9:
CF3CHCF2CH3 ϩ H2O (1b)
Under atmospheric conditions the alkyl radicals
formed will react consecutively with O2 and NO [4,5]
to give the corresponding alkoxy radicals:
Figure 4 (a) Infrared spectra obtained following photoly-
sis for 30 min of 30 ppm Cl2 with 10 ppm CF3CH2CF2CH3
in air at 760 torr and 298 K and after successive subtraction
of reference spectra of (b) reactants and (c) CF2O.
CF3CH2CF2CH2 ϩ O2 ϩ M 9:
CF3CH2CF2CH2O2 ϩ M (8)

612
BARRY ET AL.
2CF3CHO2CF2CH3 9:
2CF3CHOCF2CH3 ϩ O2 (13)
By comparison with the reported data for alkoxy radi-
cals of general structure R!CH2O (where R is a
halogenated alkyl group) [11,26–28], reaction with
O2 is likely to be the only significant loss for the
CF3CH2CF2CH2O radical in both laboratory experi-
ments and the atmosphere:
CF3CH2CF2CH2O ϩ O2 9:
CF3CH2CF2CHO ϩ HO2
Since aldehydes are expected to react with Cl atoms
with rate constants which are orders of magnitude
higher than for the parent halocarbons [29],
CF3CH2CF2CHO formed in reaction (14) will be
rapidly removed by reaction with Cl atoms under the
experimental conditions employed.
Figure 5 Concentration-time profiles of CF2CH2CF2CH3
and CF2O in the photolysis of Cl2 (30 ppm) and
CF3CH2CF2CH3 (10 ppm) in air at 760 torr and 298 K.
amounts of CF3O3CF3 . The results confirm the ab-
sence of carbon containing products other than CF2O,
CF3O3CF3 , and CO2 . Concentration-time profiles of
the reactant, CF3CH2CF2CH3 , and CF2O from an ex-
periment in which 30 ppm Cl2 and 10 ppm
CF3CH2CF2CH3 were photolyzed in air at 760 torr
and 298 K are shown in Figure 5 and the yield of
CF2O is plotted as a function of the loss of
CF3CH2CF2CH3 in Figure 6. The results shown that
for conversions of CF3CH2CF2CH3 Յ 60%, the yield
of CF2O is close to 200%, and that of the minor prod-
uct, CF3O3CF3 , less than 5%, within the experimental
error (estimated to be Ϯ5%).
Cl ϩ CF3CH2CF2CHO 9:
CF3CH2CF2CO ϩ HCl (15)
CF3CH2CF2CO ϩ O2 ϩ M 9:
CF3CH2CF2C(O)O2 ϩ M (16)
2CF3CH2CF2C(O)O2 9:
2CF3CH2CF2C(O)O ϩ O2 (17)
CF3CH2CF2C(O)O 9:
CF3CH2CF2 ϩ CO2 (18)
The experimental data can be rationalized in terms
of the following reactions:
CF3CH2CF2 ϩ O2 ϩ M 9:
CF3CH2CF2O2 ϩ M (19)
2CF3CH2CF2CH2O2 9:
2CF3CH2CF2O2 9:
2CF3CH2CF2CH2O ϩ O2 (12)
2CF3CH2CF2O ϩ O2 (20)
Carbon-carbon bond scission may be assumed to be
the only important reaction pathway for the
CF3CH2CF2O radical by comparison with the data for
CF3CF2O [30–32], CHF2CF2O [33], CH2FCF2O
[34], and CH3CF2O [30,31] radicals.
CF3CH2CF2O 9: CF3CH2 ϩ CF2O
Further reactions of the CF3CH2 radical give rise to
the formation of CF3CHO [11,26–28] which subse-
quently reacts with Cl in the presence of O2 to form
CF3 radicals and CO2 [28]. CF3 radicals will lead to
the production of CF2O or CF3O3CF3 .
CF3 ϩ O2 ϩ M 9: CF3O2 ϩ M
2CF3O2 9: 2CF3O ϩ O2
(22)
(23)
Figure 6 Plot of the yield of CF2O against the loss of
CF3CH2CF2CH3 from the photolysis of Cl2 (30 ppm) and
CF3CH2CF2CH3 (10 ppm) in air at 760 torr and 298 K.

ATMOSPHERIC DEGRADATION
613
yield close to 200% in the Cl-atom initiated oxidation
of CF3CH2CF2CH3 involves a series of carbon-carbon
bond scission steps. It is possible, however, that
the absence of the carbonyl compound,
CF3C(O)CF2CH3 , from the products may reflect the
low reactivity of the !CH2 ! site in the parent com-
pound rather than the dominance of C!C bond scis-
sion in the CF3CHOCF2CH3 radical compared to its
reaction with O2 . In order to examine the reactivity of
the methylene group in CF3CH2CF2CH3 with respect
to reaction with OH and Cl, rate constants were de-
termined for the reaction of these species with
CF3CH2CCl2F and CF3CH2CF3 . The !CH2 !
groups in these compounds have similar molecular
environments to the methylene group in
CF3CH2CF2CH3 and the rate data should provide in-
formation on the reactivity of the !CH2 !
group in 1,1,1,3,3-pentafluorobutane. Kinetic data
from the photolysis of ozone in the presence of mix-
tures of CF3CH2CCl2F/CF3CH2CF2CH3/H2O and
CF3CH2CF3/CF3CH2CF2CH3/H2O in air at 298 K are
shown plotted in the form of eq. (I) in Figure 7. The
data provide values of k(OH ϩ CF3CH2CCl2F)/
CF3O2 ϩ CF3O 9: CF3O3CF3
(24)
(25)
several steps
CF3O 999:
CF2O
The yield of CF3O3CF3 was relatively low and hence
the conversion of CF3O2 radicals to CF2O must be
virtually quantitative. It is possible that CF3O radicals
formed in reaction (23) produce CF3OH by hydrogen
atom abstraction from CF3CH2CF2CH3 , followed by
heterogeneous loss of CF3OH to give CF2O [35]. Re-
action of CF3O with fluorocarbons has been shown to
be relatively slow [36] and is unlikely to be the major
source of CF2O. Previous work from this laboratory
has shown that the generation of CF3O from the pho-
tolysis of CF3O2CF3 in air gives high yields of CF2O
in the presence of Cl atoms [37]. Generation of CF2O
in the present system is consistent with the exother-
mic reactions (26) or (27).
CF3O2 ϩ Cl 9: CF2O ϩ ClOF
CF3O ϩ Cl 9: CF2O ϩ ClF
(26)
(27)
Abstraction of a hydrogen atom may also occur from
the !CH2 ! group in CF3CH2CF2CH3 , reaction
(6b), leading to the formation of the alkoxy radical
CF3CHOCF2CH3 via reactions (10) and (13). This
species may either decompose by C!C bond cleav-
age or react with O2 :
k(OH ϩ CF3CH2CF2CH3) ϭ 0.13 Ϯ 0.04
and
k(OH ϩ CF3CH2CF3)/k(OH ϩ CF3CH2CF2CH3) ϭ
0.06 Ϯ 0.02. Combination of the rate constant ratios
with an estimate of k1(OH ϩ CF3CH2CF2CH3) ϭ
(5.9 Ϯ 0.4) ϫ 10Ϫ15 cm3 moleculeϪ1 sϪ1 obtained in
this work at 298 K gives k(OH ϩ CF3CH2CCl2F) ϭ
(8 Ϯ 3) ϫ 10Ϫ16
cm3
moleculeϪ1
sϪ1
and
k(OH ϩ CF3CH2CF3) ϭ (3.5 Ϯ 1.5) ϫ 10Ϫ16
cm3
CF3CHOCF2CH3 9:
moleculeϪ1 sϪ1 at 298 K. Agreement between the
room temperature rate constant obtained for the reac-
tion of OH with CF3CH2CF3 and the values recently
CF3 ϩ CH3CF2CHO (28a)
CF3CHOCF2CH3 9:
CF3CHO ϩ CH3CF2 (28b)
reported
by
Hsu
and
DeMore
[38],
k ϭ (3.4 Ϯ 1.3) ϫ 10Ϫ16 cm3 moleculeϪ1 sϪ1 in a rel-
CF3CHOCF2CH3 ϩ O2 9:
CF3C(O)CF2CH3 ϩ HO2 (29)
Further reactions of the products of reaction (28) will
yield CF2O and CF3O3CF3 in a series of reactions
analogous to those described above. An authentic
sample of CF3C(O)CF2CH3 , the product of the bi-
molecular reaction of CF3CHOCF2CH3 with O2 , was
not available, however, the infrared spectrum of this
compound is expected to exhibit similar features to
that of CF3C(O)CF3 . There was no evidence for the
formation of CF3C(O)CF2CH3 even at 273 K and 760
torr of O2 , the most favorable conditions employed
for the reaction of the alkoxy radical with O2 relative
to dissociation by carbon-carbon bond cleavage.
(iii) Position of Attack in the Reactions of OH and
Cl with CF3CH2CF2CH3 : The reaction scheme pro-
posed to rationalize the formation of CF2O with a
Figure 7 Concentration-time data for the reaction of OH
radicals with CF3CH2CCl2F/CF3CH2CF2CH3 and
CF3CH2CF3/CF3CH2CF2CH3 mixtures at 298 K.

614
BARRY ET AL.
ative rate study, by Garland and Nelson [39],
k ϭ (4.15 Ϯ 0.33) ϫ 10Ϫ16 cm3 moleculeϪ1 sϪ1,
using a laser photolysis-laser induced fluorescence
method and by Gierczak et al. [40], k ϭ
(3.2 Ϯ 0.6) ϫ 10Ϫ16 cm3 moleculeϪ1 sϪ1 using a flash
photolysis-laser induced fluorescence technique is
within the stated experimental errors. Thus, the rate
constants for reaction of OH with CF3CH2CCl2F or
CF3CH2CF3 are considerably smaller than that ob-
tained for the overall reaction of OH with
CF3CH2CF2CH3 , suggesting that the major reaction
channel for the reaction with CF3CH2CF2CH3 is via
H-atom abstraction from the !CH3 group. Estimates
of the rate constants for reaction of OH at the
!CH2 ! and !CH3 sites in CF3CH2CF2CH3 may
be obtained using the empirical method developed by
DeMore [17] for determination of rate constants for
reaction with halocarbons. The results indicate that
attack at the terminal methyl group is a factor of five
faster than reaction at the methylene group.
CF3CH2CCl2F provide strong evidence that the at-
mospheric oxidation of CF3CH2CF2CH3 will be initi-
ated predominantly by OH radical attack at the
!CH3 group. Based on the results from the chlorine
atom initiated oxidation of CF3CH2CF3CH3 , it is ex-
pected that the molecule will be broken down under
atmospheric conditions in a series of carbon-carbon
bond breaking processes leading to the formation of
CF2O and CF3 radicals with close to 100% yields. It
is expected that the CF3 radical will be quantitatively
converted to CF3O in the atmosphere by successive
reactions with O2 and NO. Under tropospheric condi-
tions reaction with CH4 and NO are the main sinks
for the CF3O radical leading to the formation of
CF2O [6,35]. Reaction at the methylene group in
CF3CH2CF2CH3 is expected to be of negligible im-
portance but may lead to small amounts of the fluo-
roketone, CF3C(O)CF2CH3 . This ketone is also likely
to give rise to CF2O via photolysis or OH radical ini-
tiated oxidation.
Similar trends are shown by the rate constants
determined for the reaction of Cl atoms with
CF3CH2CCl2F and CF3CH2CF3 . The results from
relative rate experiments with these compounds,
Figure 3, give k(Cl ϩ CF3CH2CCl2F)/k(Cl ϩ
Radiative Forcings and Global Warming
Potentials
The infrared absorption cross-section data, shown in
Figure 8, were integrated over 100 cmϪ1 intervals and
introduced into the NCAR model in order to provide
radiative forcings and Global Warming Potentials
(GWPs) to complete the results obtained for 11 other
CFC replacement compounds [7,8]. Radiative forc-
ings and GWPs have been introduced as convenient
parameters to quantify the time-varying relative con-
tribution of long-lived greenhouse gases to global
warming [42]. The radiative forcing due to the pertur-
bation introduced by adding a certain amount of
greenhouse gas is defined as the net radiative flux
CF3CH2CF2CH3) ϭ 0.032 Ϯ 0.003
and
k(Cl ϩ
CF3CH2CF3)/k(Cl ϩ CF3CH2CF2CH3) Ͻ 0.01 and,
hence, k(Cl ϩ CF3CH2CCl2F) ϭ (3.5 Ϯ 1.5) ϫ10Ϫ17
cm3 moleculeϪ1 sϪ1 and k(Cl ϩ CF3CH2CF3) Ͻ
1 ϫ 10Ϫ17 cm3 moleculeϪ1 sϪ1. The results presented
above suggest that the products observed from the Cl-
atom initiated oxidation of CF3CH2CF2CH3 in this
study arise from Cl atom attack at the !CH3 group
of CF3CH2CF2CH3 .
Møgelberg et al. [41] have recently investigated
the fluorine atom initiated oxidation of CF3CH2CF3 at
298 K and 760 torr total pressure. The major carbon-
containing product formed was CF3C(O)CF3 and the
yield was shown to be independent of O2 partial pres-
sure over the range 5–760 torr. The alkoxy radical,
CF3CHOCF3 , produced in this system, by reactions
analogous to those shown above for CF3CH2CF2CH3 ,
can decompose or react with O2 :
CF3CHOCF3 9: CF3 ϩ CF3CHO
(30)
CF3CHOCF3 ϩ O2 9: CF3C(O)CF3 ϩ HO2 (31)
The results from Møgelberg et al. [41] indicate that
decomposition of the alkoxy radical is negligible and
thus it is likely that ketone formation from the struc-
turally similar CF3CHOCF2CH3 radical will still
dominate over C!C bond scission. Hence, the avail-
able experimental data from CF3CH2CF3 and
Figure 8 Infrared absorption cross-sections (cm2 mole-
culeϪ1) measured between 600 and 1500 cmϪ1 for HFC-
365mfc.

ATMOSPHERIC DEGRADATION
615
change (in W/m2) at the tropopause. The GWP of a
well-mixed gas relative to a reference gas is defined
as the product of the instantaneous radiative forcing
associated with the emission of 1 kg of the gas and
the concentration of the gas remaining in the atmos-
phere, integrated over a period of time, divided by the
same quantity for the reference gas.
The two dimensional chemical-radiative model
used in this study is an updated version of the model
described by Brasseur et al. [43]. The infrared radia-
tive transfer is treated as in the NCAR Community
Climate wide band model CCM1 [44,45], which in-
cludes the thermal effects of H2O, CO2 , CH4 , O3 ,
N2O, CFC-11, and CFC-12. Radiative forcings were
determined for a mean cloud cover, using annually-
averaged vertical concentration profiles for the green-
house gases. Mixing ratios were calculated between 0
and 85 km using current ground-level concentrations
[42] as a boundary condition for CO2 , CH4 , N2O,
CFC-11, and CFC-12, and set at 2.5 pptv for HFC-
365mfc.
The absolute instantaneous radiative forcing val-
ues (in W/m2) for CO2 , CFC-11 and HFC-365mfc
were found to be, respectively, 0.652 (corresponding
to a 10% increase in concentration, from 350 to 385
ppmv), 0.227, and 0.209 (values extrapolated to a 1
ppbv increase in concentration). Relative radiative
forcings obtained for HFC-365mfc using CFC-11 and
CO2 as reference gases are reported in Table I. Fi-
nally, the Global Warming Potentials have been ob-
tained for the time horizons corresponding to 20, 100,
and 500 years of integration time using the methodol-
ogy from the 1994 IPCC report [46], and are given in
Table I. In this calculation, the Bern carbon cycle
model was employed to estimate the persistence of
carbon dioxide and an atmospheric lifetime of 50
years was used for CFC-11. The results are presented
using both the calculated tropospheric lifetime of
10.8 years and an estimated overall atmospheric life-
time of 10.2 years for HFC-365mfc. Taking the rate
constant for reaction of OH with CF3CH2CF2CH3 re-
ported by Mellouki et al. [16] gives GWPs a factor of
about 1.5 lower than those given in Table I.
To our knowledge, no other cross-section or radia-
tive forcing data are available in the open literature
for HFC-365mfc. We estimate the total uncertainty
associated with the integrated cross-sections to be
lower than 4% [7], whereas the error made when cal-
culating radiative forcings and Global Warming Po-
tentials is much more difficult to evaluate. Using the
infrared spectra we have measured for several CFC
replacement compounds, Pinnock et al. [47] have re-
cently calculated the associated radiative forcings
with a narrow band model. When normalized to the
respective values for CFC-11, radiative forcings cal-
culated by each of the two groups were found to be
within Ϯ10%. The main discrepancies are thought to
be due to the different resolution, concentration pro-
files and cloud cover used in the models. The use of
an approximate model described by Pinnock et al.
[47] allows one to obtain estimated instantaneous ra-
diative forcings from known cross-sections. We have
carried out the calculation for HFC-365mfc with the
narrow band radiative forcings per unit cross-section
they provide. The absolute forcings for a concentra-
tion change of 1 ppbv were found to be quite differ-
ent (0.256 W/m2 instead of 0.209 W/m2), but the
value relative to CFC-11 was found to be very close
(0.93 instead of 0.92), which validates the results ob-
tained in this study.
CONCLUSION
The results obtained in this work indicate that the tro-
pospheric lifetime of CF3CH2CF2CH3 with respect to
removal by OH radicals is 10.8 years and that CF2O
will be the main oxidation product. Tropospheric loss
Table I Instantaneous Cloudy-Sky Radiative Forcing and Global Warming Potential Values
for HFC-365mfc Relative to CFC-11 and CO2
Atmospheric
Relative
Radiative
Forcing
Lifetimes of
HFC-365mfc,
Years
Global Warming Potential
Reference
Gas
20
100
500
years
CFC-11
CO2
0.92
10.8
10.2
10.8
10.2
0.47
0.45
2300
2210
0.21
0.20
830
790
0.18
0.17
260
250
11200
Note that relative radiative forcings are expressed on a molecular basis, GWPs on a mass basis.

616
BARRY ET AL.
11. L. Nelson, I. Shanahan, H. W. Sidebottom, J. Treacy,
and O. J. Nielsen, Int. J. Chem. Kinet., 22, 577 (1990).
12. B. J. Finlayson-Pitts, M. J. Ezell, T. M. Jayaweera,
H. N. Berko, and C. C. Lai, Geophys. Res. Lett., 19,
1371 (1992).
13. R. K. Talukdar, A. Mellouki, A. M. Schmoltner, T.
Watson, S. Montzka, and A. R. Ravishankara, Science,
257, 227 (1992).
of CF2O is governed by uptake in cloud, rain, or
ocean water followed by hydrolysis to give CO2 and
HF [5,6]. The tropospheric lifetime and an estimate
of the overall atmospheric lifetime of 10.2 years for
HFC-365mfc have been used with the measured in-
frared cross-sections to estimate the radiative forcing
and Global Warming Potential associated with this
compound. Calculations have shown that for all the
integration times investigated, the contribution of this
third generation substitute to the radiative forcing of
climate is considerably lower than that associated
with the release of the same mass of CFC-11. Using
the value of 10.2 years for the overall atmospheric
lifetime of HFC-365mfc, the infinite-time-horizon (or
steady-state) Halocarbon Global Warming Potential,
expressed relative to CFC-11, is 0.17; the Global
Warming Potentials relative to CO2 are 2210, 790,
and 250, for integration time-horizons of 20, 100, and
500 years, respectively.
14. Z. Jiang, P. H. Taylor, and B. Dellinger, J. Phys. Chem.,
96, 8691 (1992).
15. W. B. DeMore, D. M. Golden, R. F. Hampson, C. J.
Howard, C. E. Kolb, M. J. Kurylo, M. J. Molina, A. R.
Ravishankara, and S. P. Sander, Chemical Kinetics and
Photochemical Data for use in Stratospheric Model-
ling, JPL Publication 94-26: Jet Propulsion Laboratory,
California Institute of Technology, 1994.
16. A. Mellouki, S. Téton, and G. Le Bras, Geophys. Res.
Lett., 22, 389 (1995).
17. W. B. DeMore, J. Phys. Chem., 100, 5813 (1996).
18. M. J. Prather and C. M. Spivakovsky, J. Geophys. Res.
95D, 18723 (1990).
19. Intergovernmental Panel on Climate Change, Climate
Change 1995—The Science of Climate Change, Con-
tribution of Working Group I to the Second Assessment
Report of the IPCC, Chap. 2, J. T. Houghton, L. G.
Meira Filho, B. A. Callander, N. Harris, A. Kattenberg,
and K. Maskell, Eds., Cambridge University Press,
1996.
20. A. M. Schmoltner, R. K. Talukdar, R. P. Warren, A.
Mellouki, L. Goldfarb, T. Gierczak, S. A. McKeen, and
A. R. Ravishankara, J. Phys. Chem., 97, 8976 (1993).
21. NASA Reference Publication 1339, Report on Concen-
trations, Lifetimes and Trends of CFCs, Halons and
Related Species, J. A. Kaye, S. A. Penkett, and F. M.
Ormond, Eds., January 1994, pp. 5–30.
Cathy Clerbaux is grateful to Claire Granier (NCAR) for
providing help in using the NCAR model. The authors
thank Dr. Georges Le Bras (CNRS, Orléans, France) for
helpful discussions. Financial support from the Belgian
Federal Office for Scientific, Technical and Cultural Affairs
(Global Change Programme) and Solvay S.A. is gratefully
acknowledged.
BIBLIOGRAPHY
1. M. J. Molina and F. S. Rowland, Nature, 249, 810
(1974).
22. J. P. Sawerysyn, A. Talhaoui, R. Meriaux, and P. De-
volder, Chem. Phys. Lett., 198, 197 (1992).
2. J. D. Farman, B. G. Gardiner, and J. D. Shanklin, Na-
ture, 315, 207 (1985).
23. E. C. Tuazon and R. Atkinson, Int. J. Chem. Kinet., 24,
639 (1992).
3. S. Solomon, Nature, 347, 6291 (1990).
4. Scientific Assessment of Stratospheric Ozone, WMO
Global Ozone Research and Monitoring Project, Re-
port No. 20, 1989, Vol. 2, AFEAS Report.
5. Scientific Assessment of Ozone Depletion, WMO
Global Ozone Research and Monitoring Project, Re-
port No. 25, 1991.
6. Scientific Assessment of Ozone Depletion, WMO
Global Ozone Research and Monitoring Project, Re-
port No. 37, 1994.
24. T. J. Wallington and M. D. Hurley, Chem. Phys. Lett.,
189, 437 (1992).
25. R. F. Warren and A. R. Ravishankara, Int. J. Chem.
Kinet., 25, 833 (1993).
26. R. Zellner, A. Hoffman, V. Mörs, and W. Malms, Pro-
ceedings of the STEP-HALOCSIDE/AFEAS Workshop
on Kinetics and Mechanisms for the Reaction of Halo-
genated Organic Compounds in the Troposphere,
Dublin, Ireland, 23–25 March, 1993, pp. 80–87.
27. E. C. Tuazon and R. Atkinson, Environ. Sci. Technol.,
28, 2306 (1994).
28. O. J. Nielsen, E. Gamborg, J. Sehested, T. J. Walling-
ton, and M. D. Hurley, J. Phys. Chem., 98, 9518
(1994).
7. C. Clerbaux, R. Colin, P. C. Simon, and C. Granier, J.
Geophys. Res, 98D, 10491 (1993).
8. C. Clerbaux and R. Colin, Geophys. Res. Lett., 21,
2377 (1994).
9. D. Hurtmans, R. Pétrisse, D. Lemaitre, C. Clerbaux,
and J. Vander Auwera, EPA Newsl., 44, 40 (1992).
10. R. Atkinson, D. L. Baulch, R. A. Cox, R. F. Hampson,
Jr., J. A. Kerr, and J. Troe, J. Phys. Chem. Ref. Data,
21, 1125 (1992).
29. D. J. Scollard, J. J. Treacy, H. W. Sidebottom, C.
Balestra-Garcia, G. Laverdet, G. Le Bras, H. Macleod,
and S. Téton, J. Phys. Chem., 97, 4683 (1993).
30. E. O. Edney and D. J. Driscoll, Int. J. Chem. Kinet., 24,
1067 (1992).

ATMOSPHERIC DEGRADATION
617
31. E. C. Tuazon and R. Atkinson, J. Atmos. Chem., 17,
179 (1993).
32. J. Sehested, T. Ellermann, O. J. Nielsen, T. J. Walling-
ton, and M. D. Hurley, Int. J. Chem. Kinet., 25, 701
(1993).
33. O. J. Nielsen, T. Ellermann, J. Sehested, and T. J.
Wallington, J. Phys. Chem., 96, 10875 (1992).
34. J. Barry, H. Sidebottom, J. Treacy, and J. Franklin, Int.
J. Chem. Kinet., 27, 27 (1995).
41. T. E. Møgelberg, J. Platz, O. J. Nielsen, J. Sehested,
and T. J. Wallington, J. Phys. Chem., 99, 5373 (1995).
42. Intergovernmental Panel on Climate Change, Climate
Change—The IPCC Scientific Assessment, Chap. 2,
J. T. Houghton, G. J. Jenkins, and J. J. Ephraums,
Eds., Cambridge University Press, 1990.
43. G. Brasseur, M. H. Hitchman, S. Walters, M. Dymek,
E. Falise, and M. Pirre, J. Geophys. Res., 95D, 5639
(1990).
35. J. Chen, T. Zhu, H. Niki, and G. J. Mains, Geophys.
Res. Lett., 19, 2215 (1992).
36. T. J. Wallington and J. C. Ball, J. Phys. Chem., 99,
3201 (1995).
37. C. Kelly, J. J. Treacy, and H. W. Sidebottom, unpub-
lished work.
38. K. J. Hsu and W. B. DeMore, J. Phys. Chem., 99, 1235
(1995).
39. N. L. Garland and H. H. Nelson, Chem. Phys. Lett.,
248, 296 (1996).
40. T. Gierczak, R. K. Talukdar, J. B. Burkholder, R. W.
Portmann, J. S. Daniel, S. Soloman, and A. R. Ravis-
hankara, J. Geophys. Res., 101D, 12903 (1996).
44. J. T. Kiehl, R. J. Wolsky, B. P. Briegleb, and V. Ra-
manathan, Documentation of Radiation and Cloud
Routines in the NCAR Community Climate Model
(CCM1), NCAR Tech. Note, NCAR/TN-288ϩIA,
Boulder, Colorado, 1987.
45. B. P. Briegleb, J. Geophys. Res., 97D, 11475 (1992).
46. Intergovernmental Panel on Climate Change, Climate
Change 1994—Radiative Forcing of Climate Change,
Chap. 5, J. T. Houghton, L. G. Meira Filho, J. Bruce,
Joesung Lee, B. A. Callander, E. Haites, N. Harris, and
K. Maskell, Eds., Cambridge University Press, 1995.
47. S. Pinnock, M. D. Hurley, K. P. Shine, T. J. Wallington,
and T. J. Smyth, J. Geophys. Res., 100D, 23227 (1995).

618
BARRY ET AL.