Atmospheric Environment 35 (2001) 4113–4123
Atmospheric degradation and global warming potentials of
three perfluoroalkenes
G. Acerbonia, J.A. Beukesb, N.R. Jensena,*, J. Hjortha, G. Myhrec,
C.J. Nielsenb, J.K. Sundetc
a
European Commission, Joint Research Centre, Environment Institute, TP 272, I-21020 Ispra (VA), Italy
b
University of Oslo, P.O. Box 1033, Blindern, N-0315 Oslo, Norway
c
Department of Geophysics, University of Oslo, P.O. Box 1022, Blindern N-0315 Oslo, Norway
Received 20 December 2000; accepted 26 March 2001
Abstract
The vapour phase reactions of perfluoropropene, CF3aCF ¼ CF2, and perfluorobuta-1,3-diene,
CF2 ¼ CFaCF ¼ CF2, with OH, NO3 and O3 were studied at 29874 K and 74075 Torr using long-path FT-IR
detection. The reactions with ozone are very slow, kCF3 CFCF2 þO3 ¼ ð6:271:5Þ10@22 and kCF2 CFCFCF2 þO3 ¼
ð6:570:2Þ10@21 cm3 molecules@1 s@1, and upper limits of 3 10@15 cm3 molecules@1 s@1 are reported for the
NO3 reaction rate coefficients. The OH reaction rate coefficients were determined as kCF3 CFCF2 þOH ¼ ð2:670:7Þ10@12
and kCF2 CFCFCF2 þOH ¼ ð1:170:3Þ10@11 cm3 molecules@1 s@1; perfluoropropene gave a nearly quantitative yield of
CF3CFO and CF2O as organic products, while perfluorobuta-1,3-diene gave from 130% to 170% of CF2O.
A chemistry transport model was applied to calculate the atmospheric distributions and lifetimes of the
perfluoroalkenes; the global and yearly averaged lifetimes were calculated as 1.9 day for C2F4 and C4F6 and 6 days
for C3F6.
Quantitative infrared cross-sections of perfluoroethene, perfluoropropene, and perfluorobuta-1,3-diene have been
obtained at 298 K in the region 100–2600 cm@1. Radiative forcing calculations have been performed for these gases
assuming either constant vertical profiles or the distribution derived from the chemistry transport model. The results
show that the Global Warming Potentials are totally negligible for these compounds. r 2001 Elsevier Science Ltd. All
rights reserved.
Keywords: Infrared absorption cross sections; Radiative forcing; OH radical; Atmospheric lifetimes; Global modelling; Perfluoroalkenes
1. Introduction
The Montreal Protocol and amendments (Montreal
Protocol, 1987) led to the phase out of a series of
clorofluorocarbons (CFCs), in industrialised countries,
mainly because of their ozone depletion in the stratosphere, but also because of their high Global Warming
Potentials (GWPs) (Zurer, 1993; IPCC, 1994). An
*Corresponding author. Tel.: +39-0332-789225; fax: +390332-785837.
E-mail address: niels.jensen@jrc.it (N.R. Jensen).
intensive investigation of potential CFC replacements,
mainly organic compounds containing hydrogen, fluorine or chlorine atoms was therefore initiated about 10
years ago, in order to estimate their ozone depletion
potentials and their GWPs (Fisher et al., 1990; Clerbaux
et al., 1993; Ravishankara et al., 1993, 1994; Pinnock
et al., 1995; Christidis et al., 1997; Myhre et al., 1999;
Highwood and Shine, 2000).
Perfluorocarbons (PFCs), were for a period considered as an acceptable class of alternative compounds
to replace CFCs in some applications, because it
was estimated that their ozone depletion potential is
1352-2310/01/$ - see front matter r 2001 Elsevier Science Ltd. All rights reserved.
PII: S 1 3 5 2 - 2 3 1 0 ( 0 1 ) 0 0 2 0 9 - 6
4114
G. Acerboni et al. / Atmospheric Environment 35 (2001) 4113–4123
negligible compared to CFC-11 (Ravishankara et al.,
1994). However, saturated PFCs, such as CF4 and C2F6,
are very stable in the atmosphere, with lifetimes at the
order of several thousand years (Ravishankara et al.,
1993), indicating higher GWP values compared to CFC11 (IPCC, 1994). Unsaturated PFCs, such as C2F4 and
C3F6, also have IR absorption bands in the spectral
regions where radiative forcing is most efficient, but they
react relatively fast with OH radicals (McIlroy and
Tully, 1993; Orkin et al., 1997; Acerboni et al., 1999;
Mashino et al., 2000). Typical atmospheric lifetimes are
of the order of days, which suggests that their GWP
values are low.
Perfluoroalkenes are widely used as water repellents
(as polymers) and as building blocks in the production
of perfluorinated polymers. The global annual production of perfluoroalkenes is significant, and is estimated
to be approximately 50, 20 and 0.01 Tg yr@1, for C2F4,
C3F6 and C4F6, respectively (Marchionni, 2000). However, the amount actually emitted into the atmosphere is
unknown.
The aim of this investigation is to calculate the GWPs
for C2F4, C3F6 and C4F6, and to this purpose infrared
absorption cross sections and reaction rate constants
have been determined. Further, degradation products of
these fluorinated alkenes have been measured in order to
contribute to the understanding of possible environmental impact of these compounds. Perfluoroalkenes
react relatively fast in the atmosphere, indicating that
these components are not homogeneously distributed in
the troposphere. In previous radiative forcing and GWP
studies of CFCs and CFC replacements it has been
assumed that the gases have a homogeneous mixing
ratio in the troposphere as the lifetime of such gases are
typically many years. For gases with a short lifetime the
abundance will decrease strongly with altitude. Greenhouse gases are most efficient in longwave radiation
trapping around the tropospause level, and therefore the
calculated radiative forcing (and furthermore also the
GWP) will be reduced substantially for short-lived
components, compared to results assuming a homogeneous mixing ratio in the atmosphere.
Only limited information of atmospheric relevance is
available on perfluoroalkenes: Heicklen (1966) used
infrared spectroscopy to study the gas phase reaction
between C2F4 and O3 at room temperature and 1–
24 Torr total pressure; carbonyl fluoride, CF2O, was the
only oxidation product found. Toby and Toby (1976)
also studied the gas phase reaction between C2F4 and O3
in the temperature range from 273 to 383 K and in the
pressure range from 0.3 to 15 Torr using gas chromatography. Acerboni et al. (1999) used Fourier Transform
Infrared Spectroscopy (FTIR) to study the reaction of
C2F4 with OH and NO3 radicals and with O3 at 298 K
and 740 Torr total pressure in a static reactor. They
found CF2O as the only fluorine containing product
with molar yields of 8575%, 135725% and 19274%
for the OH, NO3 and O3 reactions, respectively. For
the reaction between C2F4 and O3 there is a large
discrepancy between the reaction rate coefficients
obtained in the three previous studies; only in the
investigation by Acerboni et al. (1999) was an OH
radical scavenger, cyclohexane, added to the reaction
system. Orkin et al. (1997) used a flow system with a
flash photolysis resonance fluorescence technique to
study the reactions of C2F4 and C3F6 with OH; for C2F4
the reaction was studied at 298 K and 100 Torr total
pressure, for C3F6 the reaction was studied in the
temperature range from 287 to 370 K.
McIlroy and Tully (1993) used a slow-flow system
with pulsed-laser photolysis/laser-induced fluorescence
technique to study the reaction between C3F6 and the
OH radical in the temperature range from 293 to 831 K
and 750 Torr total pressure. Mashino et al. (2000) used a
static system, equipped with FTIR detection, to study
the reaction between C3F6 and the OH radical at
700 Torr and 296 K; CF2O and CF3CFO were observed
as reaction products with molar yields of 9877% and
9076%, respectively. They also determined upper limits
for the O3 reaction with C3F6.
To the knowledge of the authors no kinetic or product
data are available for C4F6 reactions of atmospheric
relevance in the literature.
2. Experimental details and model description
2.1. Kinetic measurements and product study
The kinetic measurements and product studies were
performed in purified air at 74075 Torr total pressure
and 29874 K in a 480 l Teflon coated reaction chamber.
The chamber is surrounded by 18 UV/VIS lamps (black
lamps, lX300 nm) and equipped with a multiple
reflection mirror system, with a total optical pathlength
of 81.23 m, for on-line FTIR. A detailed description of
the experimental set-up has been published previously
(Jensen et al., 1991). Typical initial volume fractions
were: C3F6, 2–5 ppmV; C4F6, 3–5 ppmV; reference
compounds, 8–25 ppmV; CH3ONO, 10–30 ppmV; NO,
8–15 ppmV; N2O5, 10–25 ppmV; O3, 25–600 ppmV;
Cyclohexane, 20–60 ppmV.
The infrared spectra were obtained in the region 600–
4000 cm@1 with a Bruker IFS 113v at a nominal
resolution of 1 cm@1 by co-adding 50 scans. The
following IR bands were used for spectral subtraction:
1142–1253 and 1357–1424 cm@1 for C3F6 , 1121–
1148 cm@1 for C4F6, 2985–2990 and 907–915 cm@1
for ethene, 942–954 cm@1 for propene and 1453–1460
and 2840–2895 cm@1 for cyclohexane.
G. Acerboni et al. / Atmospheric Environment 35 (2001) 4113–4123
2.2. Infrared cross sections
Infrared spectra of the pure gases at room temperature were recorded in the region 100–2600 cm@1 using a
Bruker IFS 113v employing a nominal resolution of
1 cm@1 (boxcar apodization). KBr and Mylars beamsplitters of various thicknesses were used to cover the
spectral region. To ensure optical linearity, only DTGS
detectors were used. Cells of 2.28 and 19 cm length
equipped with windows of CsI and high density
polyethylene were employed, respectively. The partial
pressures of the gasses in the cells were from 4 to
70 mbar and were measured using MKS Baratron
pressure transducers. The cross sections were obtained
from the absorbance spectra assuming that the gas was
ideal and applying a baseline correction. The latter was
performed by subtracting a polynomial function,
obtained by fitting the regions of the spectrum were no
absorption where expected.
2.3. Chemicals
The compounds used in this investigation had the
following purities: C3F6 (>99.9% pure, Ausimont),
C4F6 (99% pure, Ausimont), ethene (>99.5% pure,
AirLiquide), propene (>99.5% pure, Ucar), cyclohexane (99.5% pure, Carlo Erba), CF2O (97% pure,
Matheson), synthetic air (80% N2 and 20% O2:
X99.95% pure, SIO) and O2 (X99.9% pure, SIO).
Hydroxyl radicals were generated from in-situ photolysis of methylnitrite (CH3ONO) in the presence of NO
by using UV/VIS lamps (lX300 nm). NO3 radicals were
produced by mixing O3 with an excess of NO2, to
establish equilibrium between NO3, NO2 and N2O5.
Ozone was prepared by silent discharge of pure oxygen.
CH3ONO was synthesised by adding H2SO4 (50 wt%
aqueous solution) to sodium nitrite dissolved in a
methanol/water mixture as described in detail elsewhere
(Taylor et al., 1980).
2.4. Chemical transport model
The Oslo CTM2 is an off-line chemical transport/
tracer model (CTM) that use pre calculated transport
and physical fields to simulate chemical turnover and
distribution in the atmosphere. The model is valid for
the global troposphere and is three-dimensional (3-D)
with the model domain reaching from the ground up to
10 hPa and with a T21 (5.61 5.61) horizontal resolution.
Advection is done using the second order moment
(Prather, 1986), convection is based on the Tiedtke mass
flux scheme (Tiedtke, 1989), where vertical transport of
species is determined by the surplus/deficit of mass flux
in a column. The chemical scheme is based upon the
QSSA approach (Hesstvedt et al., 1978; Berntsen and
4115
Isaksen, 1997). Photodissociation is done on-line following Wild et al. (2000). Emissions are based upon GEIA
.
and EDGAR for natural emissions and Muller
(1992)
for anthropogenic emissions. Deposition is based upon
Wesley (1989) and the boundary layer is treated
according to the Holtslag K-profile scheme (Holtslag
et al., 1990). Influence of stratospheric ozone is
estimated using a synthetic ozone approach (McLinden
et al., 2000) where ozone flux in the stratosphere is
prescribed and the model transport generates an ozone
distribution that varies with time and space. Wet
removal is done using the 3-D rainfall that is available
in the model data. For water-soluble compounds the
Henry constant is used to estimate the amount of tracer
in cloud water and gas phase. The fraction of cloud
water that is rained out then determines the fraction of
compound that is removed from the grid box.
The model is shown to simulate the seasonal variations of O3 and CO at a number of stations both in the
Northern and Southern Hemisphere well (Sundet, 1997;
Jonson et al., 2001). Transport of ozone done in the
TOPOZ-2 project also show that the model does a good
job in describing the transport in the upper troposphere
and lower stratosphere compared to MOZAIC data.
2.5. Radiative transfer model
An emissivity/absorptivity thermal infrared broad
band model is used in the radiative forcing calculations.
The model includes all the major gases absorbing in the
thermal infrared region and includes altogether around
50 absorption bands (Myhre and Stordal, 1997). The
broad band model is compared to the GENLN2 line-byline model (Edwards, 1992) with radiative forcing due to
CFCs and CFC replacements generally within 5%
(Myhre et al., 1998, 1999). A 101 101 horizontal
resolution is used in the radiative transfer calculations.
Climatological values of water vapour, temperature, and
clouds are used (Myhre and Stordal, 1997; Myhre et al.,
1999).
The three gases are included with 2, 5, and 5 bands for
C2F4, C3F6, and C4F6, respectively. The annual mean
latitudinal and height distribution as calculated with
the CTM is used for the species. Calculations are
performed for a surface abundance of 0.1 ppbV and then
scaled linearly to 1 ppbV to ensure weak line limit, in
accordance with previous similar studies (e.g. Pinnock
et al., 1995; Myhre and Stordal, 1997)
3. Results and discussion
3.1. Kinetic study and degradation products
To determine the atmospheric chemical lifetime for
perfluoroalkenes, we have investigated the rate constants
4116
G. Acerboni et al. / Atmospheric Environment 35 (2001) 4113–4123
of the reactions of C3F6 and C4F6 with OH and NO3
radicals as well as O3:
þ
9
>
=
CF2 ¼ CF2CF ¼ CF2
CF3 CF ¼ CF2
CF2 ¼ CF2
>
;
ð1Þ
ð2Þ
ð3Þ
OH
NO3
O3
- Products
The OH and NO3 reaction rate coefficients were
determined by the ‘‘relative rate’’ method while the O3
reaction rate was measured under ‘‘pseudo-first-order’’
conditions.
Using the ‘‘relative rate’’ method, the ratio of the rate
constants kA =kB is found as the slope of a plot of
lnðA0 =At Þ@kwA t versus lnðB0 =Bt Þ@kwB t, where A0 and
B0 are the initial concentrations, At and Bt are the
concentrations at time t. kwA and kwB are rate constants
describing additional first-order loss processes, if any
such are present, for A and B, respectively (see e.g.
Fig. 1). The pseudo-first-order method: Loss of C3F6
and C4F6 in an excess of O3 was measured to obtain a
pseudo-first-order decay rate constant k0 ð¼ k½O3 Þ from
the plot of ln[C3F6]t and ln[C4F6]] versus time t. A plot
of measured k0 -values versus [O3] then yield the
bimolecular rate constant, k, as the slope. In the O3
experiments large amounts of cyclohexane (20–
60 ppmV) were added to scavenge OH radicals or other
radicals and atoms which may be formed by secondary
chemistry.
In purified air, C3F6 and C4F6 showed a first-order
decay, kw , in the reaction chamber with the UV/VISlamps turned on (lX300 nm), but without the presence
of CH3ONO; this additional loss process was attributed
to wall loss and losses due to photolytical generation of
radicals: kw was measured to be 1.2 10@5 and
1.0 10@5 s@1 for C3F6 and C4F6, respectively. For
the reference compounds: kw was measured to be
1.2 10@5, 1.1 10@5 and 5.7 10@6 s@1 for C2H4,
C3H6 and cyclohexane, respectively. Without UV/VIS
lamps on, none of the compounds showed any
significant loss and a value o2 10@6 s@1 was
assumed for kw . That is, these additional losses are
relatively small for the duration of an experiment (10–
60 min). The additional loss processes (wall loss+photolysis) were in no case responsible for more than 20% of
the overall decay during the experiments.
The values of the OH and NO3 reaction rate constants
and their uncertainties were calculated in the following
way: first, rate constants and s values were obtained
from four individual experiments for each of the
reference compounds, then the uncertainty of the rate
constant of the reference compound was incorporated
by using standard propagation of error. Two or three
reference compounds were used, so the value of the rate
constant was calculated as the mean of different values
weighted by their standard deviations. The uncertainty
is given as 2s, where s is the weighted average of the
standard deviation obtained in the series of experiments.
The uncertainty of the O3 rate constant was determined
by taking to 2s of the slope k0 versus [O3].
Fig. 1 shows a typical example of the relative loss of
C4F6 and C2H4 due to the presence of OH radicals. As
shown in Table 1, an average value of kOHþC4 F6 ¼
ð1:170:3Þ10@11 cm3 molecules@1 s@1 was determined. This rate constant has not been reported before,
Fig. 1. Plot of ln{[CF2CFCFCF2]0/[CF2CFCFCF2]t} corrected for wall-losses and photolysis versus ln[ethene]0/[{[ethene]t} for the
decay of CF2CFCFCF2 and ethene during the photolysis of methylnitrite producing OH radicals. kw =losses on the wall and/or
photolysis.
4117
G. Acerboni et al. / Atmospheric Environment 35 (2001) 4113–4123
but it compares well with the rate coefficient of the C2F4
reaction with OH (Orkin et al., 1997; Acerboni et al.,
1999), Table 1.
For the reaction between C3F6 and the OH radical,
our value of k2 =(2.670.7) 10@12 cm3 molecules@1
s@1 is slightly higher than the value determined by
McIlroy and Tully (1993); Orkin et al. (1997) and
Mashino et al. (2000), but all of the measurements are
within the uncertainty ranges (see Table 1).
For reactions 1 and 2, carbonyl fluoride was the main
end product identified with molar yields of 132–170%
and 90–100%, respectively; for reaction 2, CF3C(O)F
was also identified as a product, with a molar yield of
about 100%. In addition, spectral features around 790,
1300 and 1720 cm@1 were observed in the product
spectrum from all three reactions, indicating the
presence of peroxynitrate compounds (Niki et al.,
1986). We have tentatively attributed these bands to
compounds with the structure RaCF2O2NO2. These
peroxynitrate compounds were not available to us and
therefore we cannot measure their concentrations, but
some of the missing carbon in the carbon balance is
within that product. For reaction 3, CF2O, was
identified as the main product with a molar yield of
110–160% (Acerboni et al., 1999).
The reactions of CF3CFCF2 and CF2CFCFCF2 with
O3 were found to be very slow with reaction rate
coefficients of (6.271.5) 10@22 and (6.570.2) 10@21 cm3 molecules@1 s@1, respectively. These low
values are in good agreement with Mashino et al.
(2000), who obtained an upper value of 3 10@21 cm3
molecules@1 s@1 for the reaction between C3F6 and O3.
Also the NO3 reactions were very slow and only
upper limits are given here: kC3 F6 þNO3 o310@15 and
kC4 F6 þNO3 o310@15 cm3 molecules@1 s@1.
A rough estimate of the atmospheric lifetimes of the
three perfluoroalkenes is readily obtained on the basis of
their reactions with OH, NO3 and O3. The chemical
lifetime with respect to OH is for all three compounds a
few days, assuming an [OH] of 1 106 molecules cm@3
(Prinn et al., 1995). With respect to NO3 the lifetimes
are all more than 5 months, assuming [NO3]=2.5 107 molecules cm@3 (Noxon, 1983; Platt and Heintz,
1994). Finally, for a 24-h average ozone concentration of
7.4 1011 molecules cm@3 (Logan, 1985), the lifetimes
are of the order of several years.
3.2. Infrared absorption cross section
The infrared spectra of C2F4, C3F6 and C4F6 are
shown in Fig. 2. The integrated absorption cross
sections (base e) for all three compounds are also
included in the legends. The quality of our cross section
data has been investigated by comparing with the earlier
Table 1
Reaction rate constants at 298 K and Arrhenius parameters for the reaction of perfluoro alkenes with OH radicals. All values are given
with 2s uncertainties
Compound
k298/10@12 cm3 molecules@1 s@1
C2F4
11.373.3
10.270.5
2.470.3a
3.571.3b
2.670.7c
2.470.3
2.270.1
2.170.1
10.371.4d
12.473.3e
13.472.8f
11.373.0c
C3F6
C4F6
A/10@13 cm3 molecules@1 s@1
(5.771.5)
(9.970.6)
E=R=K
@(407785)
@(244722)
Reference
Acerboni et al. (1999)
McIlroy and Tully, 1993
This work
This work
This work
Mashino et al. (2000)
Orkin et al. (1997)
McIlroy and Tully, 1993
This work
This work
This work
This work
a
Measured relative to ethene using kOH+ethene=(7.771.0) 10@12 cm3 molecules@1 s@1, where a rate constant ratio of
(0.315170.0552) was found.
b
Measured relative to propene using kOH+propene=(3.070.8) 10@11 cm3 molecules@1 s@1, where a rate constant ratio of
(0.115370.0321) was found.
c
Average value, see text.
d
Measured relative to ethene using kOH+ethene=(7.771.0) 10@12 cm3 molecules@1 s@1, where a rate constant ratio of
(1.339470.0253) was found.
e
Measured relative to propene kOH+propene=(3.070.8) 10@11 cm3 molecules@1 s@1, where a rate constant ratio of
(0.411670.0116) was found.
f
Measured relative to cyclohexane kOH+cyclohexane=(7.271.5) 10@12 cm3 molecules@1 s@1, where a rate constant ratio of
(1.862370.0249) was found. General comment: The reference k-values for the OH-kinetics have been taken from Atkinson et al. (1992)
and Atkinson and Aschmann (1992).
4118
G. Acerboni et al. / Atmospheric Environment 35 (2001) 4113–4123
Fig. 2. Quantitative vapour-phase infrared spectra of C2F4, C3F6 and C4F6 100–2500 cm@1 at 298 K. The absorption cross section is
given in units of 10@17 cm2 molecules@1 (base e). Estimated uncertainty 710%, see text. Integrated absorption cross sections: C2F4 in
the 1080–1400 cm@1 region, (1.2970.13) 10@16 cm molecules@1; C3F6 in the 970–1850 cm@1 region, (2.3570.24) 10@16 cm molecules@1; C4F6 in the 900–1850 cm@1 region, (2.1870.22) 10@16 cm molecules@1 (base e).
well-studied CClF2H, HCFC-22 (Ballard et al., 2000).
The integrated cross sections obtained with our Bruker
113v is within 8% of the average value reported in this
intercomparison. This, in turn, suggests that the error in
our measurements be in the order of 710%, of which
the largest source is unknown.
G. Acerboni et al. / Atmospheric Environment 35 (2001) 4113–4123
3.3. CTM2 model results
The Oslo CTM2 (described in Section 2.4) was used to
model the atmospheric distributions of C2F4, C4F6 and
C3F6. It is assumed that these are removed by reaction
with OH only, and that the subsequent components are
unimportant to the chemistry in the model. The reaction
rates used in the model integration are those given in
Table 1. The surface (boundary) values were fixed to
1 pptV and the model was integrated for 15 months with
full chemistry. The last 12 months were used in the
analysis of the distributions and lifetimes of the
compounds. Due to the very fast reactions of the three
perfluoroalkenes with the OH radical and the fact that
only a very small negative temperature dependence have
been observed, no temperature dependence was included
in this study. It should be mentioned that at lower
temperature the reaction between the OH radical and
the perfluoroalkenes becomes even slightly faster (see
Table 1).
The global and yearly averaged lifetime for C3F6 was
calculated to be 6.0 days with the shortest lifetime at low
latitudes. This is seen from the distribution of the
compound in Fig. 3. The distribution is a result of the
competition between emission and OH-loss, modified by
transport. At high latitudes the winter season has low
OH and thus the removal is slow, at low latitudes the
OH concentration is high year round and the removal of
the compound is efficient. In particular the area between
301S and 301N is affected by the high OH concentra-
4119
tions. In the vertical the highest concentration of OH is
found in the lower part of the free troposphere in the
crossover between JO3 and water vapour. At low
latitudes convection is an efficient transport agent that
reaches high up in the troposphere, to about 16 km, and
is mainly responsible for the elevated C3F6 concentration in the vertical column reaching values of about 0.2
(i.e. 20% of the boundary value is found at about
277 hPa and 301N).
Since OH is the only loss reaction for C2F4 and C4F6
the distributions and lifetimes for these compounds will
be identical and only the C2F4 result is shown. The zonal
and yearly averaged distribution of C2F4 is shown in
Fig. 4. The C2F4 lifetime is only 1.9 day. Since OH is the
only removal agent for this component, as is the case for
C3F6, the same features are seen for C2F4 as for C3F6,
but with less C2F4 compound in the troposphere and
specifically in the tropospause region. With a different
scale in Fig. 4 it would be possible to see the column of
elevated compound concentration at low latitudes, with
values about 0.05 (i.e. 5% of the fixed surface values).
3.4. Radiative transfer calculations
Using the radiative forcing concept as in IPCC (1996),
(see also Hansen et al., 1997; Shine and Forster, 1999)
we have calculated radiative forcing of the three
perfluoroalkenes. The results are shown in Table 2,
both for a constant vertical profile in the atmosphere
and for the CTM calculated distribution. The radiative
Fig. 3. Zonal and yearly averaged C3F6. Vertical scale is pressure, from 984 to 50 hPa and the horizontal scale goes from 901S to 901N
latitude. The OH generated tracer hump is clearly seen between 301S and 301N. An area column with elevated tracer concentration is
seen around 101N and 400–150 hPa.
4120
G. Acerboni et al. / Atmospheric Environment 35 (2001) 4113–4123
Fig. 4. Zonal and yearly averaged C2F4. Vertical scale is pressure, from 984 to 50 hPa and the horizontal scale goes from 901S to 901N
latitude. The OH generated tracer hump is clearly seen between 301S and 301N.
forcings using the CTM calculated distribution are very
small compared to other CFCs and CFC replacements,
whereas they are rather similar for a constant vertical
profile (Pinnock et al., 1995; Myhre and Stordal, 1997;
Good et al., 1998; Myhre et al., 1999; Highwood and
Shine, 2000; Jain et al., 2000; Shira et al., 2001).
Actually, using the CTM calculated distribution instead
of the constant vertical profile reduces the radiative
forcing by a factor 8 for C3F6 and around 23 for C2F4
and C4F6. This is considerably higher than for other
CFCs and CFC replacements, for which this ratio is
usually below 10% (Myhre and Stordal, 1997; Freckleton et al., 1998) and up to 40% found in Jain et al.
(2000).
A normalised radiative forcing can be defined as the
ratio of the radiative forcing and the integrated band
strength (in 1017 W molecule m@3). For a constant
vertical profile the normalised radiative forcing is rather
similar for the three species with values ranging from 1.0
to 1.2 with C2F4 having the lowest and C4F6 the highest
values, respectively. Compared to other CFCs and CFC
replacements the normalised forcing of the three
perfluoroalkenes with a constant vertical profile are
low, as for CFC-11 the normalised forcing is about 2.5
and for hydrofluoroethers studied in Myhre et al. (1999)
the values were between 1.6 and 1.8. The reason for the
lower normalised forcing of the perfluoroalkenes compared to other components is that strong absorption
bands of the gases are at high wavenumbers where there
is strong overlap with water vapour and the thermal
infrared energy is low at these wavenumbers. For further
details of the spectral variation in radiative forcing per
cross section see Pinnock et al. (1995) and Highwood
and Shine (2000). The normalised radiative forcing using
the CTM calculated distribution is only approximately
0.15 for C3F6 and around 0.05 for C2F4 and C4F6.
Omitting clouds in radiative transfer calculations of
halocarbons usually increase their forcing by up to 40%
(Myhre and Stordal, 1997; Myhre et al., 1999; Jain et al.,
2000) which is similar for the 3 species investigated in
this study assuming a constant vertical profile of the
gases. However, when the CTM calculated atmospheric
distribution is used, omitting clouds increase the
radiative forcing by 57–77%. This is because a
comparatively larger fraction of the perfluoroalkenes is
found below the clouds where the radiative effect is
small or negligible. In calculations of radiative forcing
due to greenhouse gases stratospheric temperature
adjustment is important (see Hansen et al., 1997). The
radiative forcing due to the perfluoroalkenes is only
slightly reduced by omitting the stratospheric temperature adjustment (up to 1%). In contrast, omitting
stratospheric temperature adjustment the radiative
forcing due to halocarbons are usually reduced by up
to 10% (depending on the spectral overlap with ozone)
(Pinnock et al., 1995). This is actually also the case for
the perfluoroalkenes when a constant vertical profile is
assumed. The small effect of stratospheric temperature
adjustment when using the CTM calculated distribution
is mainly due to the rapid drop in the mixing ratio
4121
G. Acerboni et al. / Atmospheric Environment 35 (2001) 4113–4123
Table 2
Radiative forcing of three perfluoroalkenes assuming the CTM calculated atmospheric distribution and a constant vertical profile
Compound
CTM distribution/W m@2 ppbv@1
Constant distribution/W m@2 ppbv@1
C2F4
C3F6
C4F6
0.006
0.035
0.013
0.14
0.28
0.28
Table 3
GWP values of perfluoroalkenes relative to CO2 (and CFC-11) for three time horizons. CFC-11 values are in parenthesis
Compound
Time horizon (yr)
20
C2F4
C3F6
C4F6
100
@5
0.070 (1.1 10 )
0.86 (1.3 10@4)
0.091 (1.4 10@5)
with height and, thus, a very low concentration
in the stratosphere which will give a negligible
heating of the lower stratosphere compared to other
halocarbons.
3.5. Global warming potentials
The GWP is a simple measure to compare the
effectiveness of various climate gases (IPCC, 1990,
1994, 1996). The GWP is given relative to another gas,
either CO2 or CFC-11. The concept has some limitation
(see discussion in Fuglestvedt et al., 2000; Lashof, 2000;
O’Neil, 2000; Smith and Wigley, 2000).
GWP calculations were performed using both CFC-11
and CO2 as reference gases. For calculations using CO2
as a reference gas the expression given in IPCC (1990),
with updated constant in Myhre et al. (1998), is used in
the calculation of the radiative forcing with an
abundance of CO2 of 364 ppmV and an increase of
1 ppmV. This give a radiative forcing 5% lower than the
forcing given in WMO (2000) based on a simplified
expression. For the lifetime for CO2 the expression in
WMO (2000) is used (this expression differs slightly
from the one used in IPCC (1996) and which we used in
Myhre et al. (1999)).
Table 3 shows the GWP for the three perfluoroalkenes investigated for three time horizons for CO2
as reference gas. The perfluoroalkenes also have very
low GWP values compared to various halocarbons
(see WMO, 2000) as the radiative forcing is low and
further the lifetime is low which is an important
factor for the integrated radiative forcing over a time
horizon. Actually, for the three compounds the GWP
values range between 10@4 and 10@6 relative to
CFC-11.
500
@6
0.021 (4.4 10 )
0.25 (5.4 10@5)
0.027 (5.8 10@5)
0.0065 (3.9 10@6)
0.079 (4.8 10@5)
0.0084 (5.2 10@5)
4. Atmospheric implications
Very short atmospheric lifetimes of 1.9, 6.0 and 1.9
days with respect to the OH radical are calculated for
CF2CF2, CF3CFCF2 and CF2CFCFCF2, respectively.
Because of the high reactivity of these compounds, the
lifetime will strongly depend on local and seasonal
conditions, but an estimate of the average GWP of these
compounds can be made as shown in this paper.
Radiative forcing calculations were performed based
on the measured infrared absorption cross sections and
the calculated atmospheric distributions. The GWP
values for all three fluoroalkenes are very small
compared to that of CFC-11 (see Table 3). The GWP
values are even lower than that of CO2 for all three time
horizons calculated (see Table 3). In summary, the GWP
of perfluoroalkenes is totally negligible.
The main atmospheric degradation product of
CF2CF2, CF3CFCF2 and CF2CFCFCF2 has been identified as CF2O, and from the oxidation of C3F6 also
CF3CFO was identified. CF2O and CF3CFO are rapidly
(5–10 days) incorporated into raindrops/aerosols in the
atmosphere (De Bruyn et al., 1995). In the water phase
CF2O is eventually converted to HF and CO2 (Sehested
and Wallington, 1993), but this additional amount of
fluorine ions in the liquid phase will be negligible
compared to the global F@ budget (Sehested and
Wallington, 1993). CF3CFO in the water phase is
eventually converted to HF and CF3COOH, TFA, (De
Bruyn et al., 1995). TFA is only degraded very slowly in
the hydrosphere (Kanakidou et al., 1995).
One could argue that the oxidation products, CF2O
and CF3CFO, could also contribute with a GWP, but
with atmospheric lifetimes of 5–10 days (De Bruyn et al.,
1995), it can be assumed that the GWPs of these
compounds are also negligible compared to CFC-11.
4122
G. Acerboni et al. / Atmospheric Environment 35 (2001) 4113–4123
Acknowledgements
The authors would like to thanks Dr. T. Wallington
for a spectrum of CF3CFO. G. Acerboni acknowledges
a grant from Ausimont (I) to carry out this work and for
providing the three perfluoroalkenes.
References
Acerboni, G., Jensen, N.R., Rindone, B., Hjorth, J., 1999.
Kinetics and products formation of the gas-phase reactions
of tetrafluoroethylene with OH and NO3 radicals and
ozone. Chemical Physics Letters 309, 364–368.
Atkinson, R., Aschmann, S.M., 1992. OH radical reaction rate
constants for polycyclicalkanes: effect of ring strain and
consequences for estimation methods. International Journal
Chemical Kinetics 24, 983–989.
Atkinson, R., Baulch, D.L., Cox, R.A., Hampsom Jr., R.F.,
Kerr, J.A., Troe, J., 1992. Evaluated kinetics and photochemical data for atmospheric chemistry. Supplement IV.
IUPAC subcommittee on gas kinetics on gas kinetics data
evaluation for atmospheric chemistry. Journal of Physical
and Chemical Reference Data 21, 1125–1568.
Ballard, J., Knight, R.J., Newnham, D.D., Vander Auwera, J.,
Herman, M., Di Leonardo, G., Masciarelli, G., Nicolaisen,
F.M., Beukes, J.A., Christensen, L.K., McPheat, R., Duxbury, G., Freckleton, R., Shine, K.P., 2000. An intercomparison of laboratory measurements of absorption crosssections and integrated absorption intensities for HCFC-22.
Journal of Quantitative Spectroscopy and Radiative Transfer 66, 109–128.
Berntsen, T., Isaksen, I.S.A., 1997. A global 3-D chemical
transport model for the troposphere, 1, Model description
and CO and Ozone results. Journal of Geophysical
Research 102, 21239–21280.
Christidis, N., Hurley, M.D., Pinnock, S., Shine, K.P., Wallington, T.J., 1997. Radiative forcing of climate change by
CFC-11 and possible CFC replacements. Journal of
Geophysical Research 102, 19597–19609.
Clerbaux, C., Colin, R., Simon, P.C., Granier, C., 1993.
Infrared cross sections and global warming potentials of
10 alternatice hydrocarbons. Journal of Geophysical Research 98, 10491–10497.
De Bruyn, W.J., Shorter, J.A., Davidovits, P., Worsnop, D.R.,
Zahniser, M.S., kolb, C.E., 1995. Uptake of haloacetyl and
carbonyl halides by water surfaces. Environmental Science
and Technology 29, 1179–1185.
Edwards, D.P., 1992. GENLN2: a general line-by-line atmospheric transmittance and radiance model. NCAR Technical Note, NCAR/TN-367+STR, National Centre for
Atmospheric Research, Boulder, CO.
Fisher, D.A., Hales, C.H., Wang, W.-C., Ko, M.K.W., Sze,
N.D., 1990. Model calculations of the relative effects of
CFCs and their replacements on global warming. Nature
344, 513–516.
Freckleton, R.S., Highwood, E.J., Shine, K.P., Wild, O., Law,
K.S., Sanderson, M.G., 1998. Greenhouse gas radiative
forcing: effects of averaging and inhomogeneities in trace
gas distribution. Quarterly Journal of the Royal Meteorological Society 124, 2099–2127.
Fuglestvedt, J.S., Berntsen, T.K., Godal, O., Skodvin, T., 2000.
Climate implications of GWP-based reductions in greenhouse gas emissions. Geophysical Research Letter 27, 409–
412.
Good, D.A., Francisco, J.S., Jain, A.K., Wuebbles, D.J., 1998.
Lifetimes and global warming potentials for dimethyl
ether and for fluorinated ethers: CH3OCF3 (E143a),
CHF2OCHF2 (E134), CHF2OCF3 (E125). Journal of
Geophysical Research 103, 28181–28186.
Hansen, J., Sato, M., Ruedy, R., 1997. Radiative forcing and
climate response. Journal of Geophysical Research 102,
6831–6864.
Heicklen, J., 1966. The reactions of ozone with perfluoroolefins.
Journal of Physical Chemistry 70, 477–480 (correction sheet
added by author to page 480).
Hesstvedt, E., Hov, Ø., Isaksen, I.S.A., 1978. Quasi steady-state
approximation in air pollution modelling: comparison of
two numerical schemes for oxidant prediction. International
Journal of Chemical Kinetics 10, 971–994.
Highwood, E.J., Shine, K.P., 2000. Radiative forcing and
global warming potentials of 11 halogenated compounds.
Journal of Quantitative Spectroscopy and Radiative Transfer 66, 169–183.
Holtslag, A.A.M., DrBruijn, E.I.F., Pan, H.-L., 1990. A high
resolution air mass transformation model for short-range
weather forecasting. Monthly Weather Review 118, 1561–
1575.
Intergovernmental Panel on Climate Change (IPCC), 1990. In:
Houghton, J.T., Jenkins, G.J., Ephraums, J.J. (Eds.),
Climate Change: The IPCC Scientific Assessment. Cambridge University Press, Cambridge, UK.
Intergovernmental Panel on Climate Change (IPCC), 1994. In:
Houghton, J.T., Jenkins, G.J., Ephraums, J.J. (Eds.),
Climate Change: The IPCC Scientific Assessment. Cambridge University Press, Cambridge, UK.
Intergovernmental Panel on Climate Change (IPCC), 1996. In:
Houghton, J.T., Meira Filho, L.G., Callander, B.A., Harris,
N., Kattenberg, A., Maskell, K. (Eds.), Climate Change:
The Science of Climate Change. Cambridge University
Press, Cambridge, UK, 1995.
Jain, A.K., Briegleb, B.P., Minschwaner, K., Wuebbles, D.J.,
2000. Radiative forcing and global warming potentials of
thirty-nine greenhouse gases. Journal of Geophysical
Research 105, 20773–20790.
Jensen, N.R., Hjorth, J., Lohse, C., Skov, H., Restelli, G., 1991.
Products and mechanism of the reaction between NO3 and
dimethylsulfide. Atmospheric Environment 25A, 1897–
1904.
Jonson, J.E., Sundet, J.K., Tarrason, L., 2001. Model calculations of present and future levels of ozone and ozone
precursors with a global and regional model. Atmospheric
Environment 35, 525–537.
Kanakidou, M., dentener, F.J., Crutzen, P.J., 1995. A global
three-dimensional study of the fate of HCFCs and HFC134a in the troposphere. Journal of Geophysical Research
100, 18781–18801.
Lashof, D.A., 2000. The use of global warming potentials in the
Kyoto Protocol. Climatic Change 44, 423–425.
Logan, J.A., 1985. Tropospheric ozone: seasonal behavior,
trends and anthropogenic influence. Journal of Geophysical
Research 90, 10463–10482.
G. Acerboni et al. / Atmospheric Environment 35 (2001) 4113–4123
Marchionni, G., 2000. Ausimont (I), Private communications.
Mashino, M., Ninomiya, Y., Kawasaki, M., Wallington, T.J.,
Hurley, M.D., 2000. Atmospheric chemistry of CF3CFCF2:
kinetics and mechanism of its reactions with OH radicals, Cl
atoms, and ozone. Journal of Physical Chemistry, submitted
for publication.
McIlroy, A., Tully, P., 1993. Kinetic study of OH reactions with
perfluoropropene and perfluorobenzene. Journal of Physical
Chemistry 97, 610–614.
Mclinden, C.A., Olsen, S., Hannegan, B., Wild, O., Prather,
M.J., Sundet, J., 2000. Stratospheric ozone in 3-d models: a
simple chemistry and the cross-tropopause flux. Journal of
Geophysical Research 105, 14653–14665.
Monteol Protocol on Substances that Deplete the Ozone Layer,
1987. Final Act., UNEP, 1987; revised 1990, London
Amendment; revised 1992, Copenhagen Amendment.
.
Muller,
J., 1992. Geographical distribution and seasonal
variation of surface emissions and deposition velocities of
atmospheric trace gases. Journal of Geophysical Research.
97, 3787–3804.
Myhre, G., Stordal, F., 1997. Role of spatial and temporal
variations in the computation of radiative forcing and
GWP. Journal of Geophysical Research 102, 11181–11200.
Myhre, G., Highwood, E.J., Shine, K.P., Stordal, F., 1998. New
estimates of radiative forcing due to well mixed greenhouse
gases. Geophysical Research Letters 25, 2715–2718.
Myhre, G., Nielsen, C.J., Powell, D.L., Stordal, F., 1999.
Infrared absorbtion cross section, radiative forcing, and
GWP of four hydrofluoro(poly)ethers. Atmospheric Environment 33, 4447–4458.
Niki, H., Maker, P.D., Savage, C.M., Breitenbach, L.P.,
Hurley, M.D., 1986. An FTIR spectroscopic study of the
kinetics and mechanism for the NO2-initiated oxidation of
tetramethyl ethylene at 298 K. International Journal of
Chemical Kinetics 18, 1235–1247.
Noxon, J.F., 1983. Nitrogen trioxide and nitrogen dioxide in
the mid-pacific troposphere. Journal of Geophysical Research 88, 11017–11021.
O’Neill, B.C., 2000. The jury is still out on global warming
potentials. Climatic Change 44, 427–443.
Orkin, V.L., Huie, R.E., Kurylo, M.J., 1997. Rate constant for
the reactions of OH with HFC-245cb (CH3CF2CF3) and
some fluoroalkenes (CH2CHCF3, CH2CFCF3, CF2CFCF3
and CF2CF2). Journal of Physical Chemistry 101, 9118–
9124.
Pinnock, S., Hurley, M.D., Shine, K.P., Wallington, T.J.,
Smyth, T.J., 1995. Radiative forcing of climate by hydroclorofluorocarbons and hydrofluorocarbons. Journal of
Geophysical Research 100, 23277–23288.
Platt, U., Heintz, F., 1994. Nitrate radicals in tropospheric
chemistry. Israel Journal of Chemistry 34, 289–300.
Prather, M.J., 1986. Numerical advection by conservation of
second-order moments. Journal of Geophysical Research
91, 6671–6681.
4123
Prinn, R.G., Weiss, R.F., Miller, B.R., Huang, J., Alyea, F.,
Cunnold, D., Fraser, P., Hartley, D., Simmonds, P., 1995.
Atmospheric trends and lifetime of CH3CCl3 and global OH
concentration. Science 269, 187–192.
Ravishankara, A.R., Solomon, S., Turnipseed, A.A., Warren,
R.F., 1993. Atmospheric lifetimes of long-lived halogenated
species. Science 259, 194–199.
Ravishankara, A.R., Turnipseed, A.A., Jensen, N.R., Barone,
S., Mills, M., Howard, C.J., Solomon, S., 1994. Do
hydrofluorocarbons destroy stratospheric ozone? Science
263, 71–75.
Sehested, J., Wallington, T.J., 1993. Atmospheric chemistry of
hydrofluorocarbon 134a. Fate of the alkoxy radical CF3O.
Environmental Science and Technology 27, 146–152.
Shine, K.P., Forster, P.M., 1999. The effects of human activity
on radiative forcing of climate change: a review of recent
developments. Global and Planetary Change 20, 205–225.
Shira, K., Hurley, M.D., Shine, K.P., Wallington, T.J., 2001.
Updated radiative forcing estimates of sixty-five halocarbons and non-methane hydrocarbons. Journal of Geophysical Research, in press.
Smith, S.J., Wigley, T.M.L., 2000. Global warming potentials: 2
accuracy. Climatic Change 44, 459–469.
Sundet, J.K., 1997. Model Studies with a 3-d Global CTM
using ECMWF data. Ph.D. Thesis, Department of Geophysics, University of Oslo, Norway.
Taylor, W.D., Allston, T.D., Moscato, M.J., Fazekas, G.B.,
Kozlowski, R., Takacs, G.A., 1980. Atmospheric photodissociation lifetimes for nitromethane, methyl nitrite and
methyl nitrate. International Journal of Chemical Kinetics
12, 231–240.
Tiedtke, M., 1989. A comprehensive mass flux scheme for
Cumulus parameterisation on large scale models. Monthly
Weather Review 117, 1779–1800.
Toby, F.S., Toby, S., 1976. Kinetics of the reaction of ozone
with tetrafluoroethene. Journal of Physical Chemistry 80,
2313–2316.
Wesley, M.L., 1989. Parameterization of surface resistances to
gaseous dry deposition in regional-scale numerical models.
Atmospheric Environment 23, 1293–1304.
Wild, O., Zhu, X., Prather, M.J., 2000. Fast-J: accurate
simulation of in- and below cloud photolysis in global
chemical models. Journal of the Atmospheric Chemistry
37, 245–282.
World Meteorological Organization (WMO), 2000. Scientific
Assessment of Ozone Depletion. Global Ozone Research
and Monitoring Project, World Meteorological Organization, Report No. 4.
Zurer, P.S., 1993. Ozone depletion’s recurring surprises
challenge atmospheric scientists. Chemical and Enginering
News 71, 8–18.