Atmospheric lifetimes and global warming potentials of CF 3CH2CH2OH and CF3(CH 2)2CH2OH

Article abstract of DOI:10.1002/cphc.201000365

A comprehensive study of several atmospheric degradation routes for two hydrofluoroalcohols, CF₃(CH₂)_x=1,2CH ₂OH, is presented. The gas-phase kinetics of their reactions with hydroxyl radicals (OH) and chlorine (Cl) atoms are investigated by absolute and relative techniques, respectively. The room-temperature rate coefficients (±σ, in cm³ molecule^-1 s^-1) k _OH and k_Cl, are respectively (9.7±1.1)× 10^-13 and (1.60±0.45)× 10^-11 for CF ₃CH₂CH₂OH, and (2.62±0.32)× 10^-12 and (8.71±0.24)× 10^-11 for CF ₃(CH₂)₂CH₂OH. Average lifetimes of CF₃CH₂CH₂OH and CF₃(CH ₂)₂CH₂OH due to the OH and Cl reactions are estimated to be 12 and 4 days, and greater than 20 and 4 years, respectively. Also, the IR and UV absorption cross sections of CF₃(CH ₂)_x=1,2CH₂OH are determined in the spectral ranges of 500-4000 cm^-1 and 200-310 nm. Photolysis of CF ₃(CH₂)_x=1,2CH₂OH in the actinic region (I≥290 nm) is negligible compared to their homogeneous removal. Additionally, computational IR spectra are consistent with the experimental ones, thus giving high confidence in the obtained results. The lifetimes of CF₃(CH₂)_x=1,2CH₂OH and IR spectra reported herein allow the calculation of the direct global warming potential of these hydrofluoroalcohols. The contribution of CF₃(CH ₂)_xCH₂OH to radiative forcing of climate change will be negligible.

Full text of DOI:10.1002/cphc.201000365

DOI: 10.1002/cphc.201000365
Atmospheric Lifetimes and Global Warming Potentials of
CF CH CH OH and CF (CH ) CH OH
3
2
2
3
2 2
2
Elena Jimꢀnez, Marꢁa AntiÇolo, Bernabꢀ Ballesteros, Ernesto Martꢁnez, and
[a]
Josꢀ Albaladejo*
A comprehensive study of several atmospheric degradation
routes for two hydrofluoroalcohols, CF (CH ) CH OH, is pre-
tively. Also, the IR and UV absorption cross sections of
CF (CH ) CH OH are determined in the spectral ranges of
3
2 x=1,2
2
3
2 x=1,2
2
ꢀ
1
sented. The gas-phase kinetics of their reactions with hydroxyl
radicals (OH) and chlorine (Cl) atoms are investigated by abso-
lute and relative techniques, respectively. The room-tempera-
500–4000 cm and 200–310 nm. Photolysis of CF (CH )
-
3
2 x=1,2
CH OH in the actinic region (lꢂ290 nm) is negligible com-
2
pared to their homogeneous removal. Additionally, computa-
tional IR spectra are consistent with the experimental ones,
thus giving high confidence in the obtained results. The life-
3
ꢀ1 ꢀ1
ture rate coefficients (ꢁs, in cm molecule s ) k and k , are
OH
Cl
ꢀ11
ꢀ
13
respectively (9.7ꢁ1.1)ꢂ10
and (1.60ꢁ0.45)ꢂ10
for
ꢀ
12
ꢀ11
CF CH CH OH, and (2.62ꢁ0.32)ꢂ10
and (8.71ꢁ0.24)ꢂ10
times of CF (CH )
CH OH and IR spectra reported herein
3
2
2
3
2 x=1,2 2
for CF (CH ) CH OH. Average lifetimes of CF CH CH OH and
allow the calculation of the direct global warming potential of
these hydrofluoroalcohols. The contribution of CF (CH ) CH OH
3
2 2
2
3
2
2
CF (CH ) CH OH due to the OH and Cl reactions are estimated
3
2 2
2
3
2 x
2
to be 12 and 4 days, and greater than 20 and 4 years, respec-
to radiative forcing of climate change will be negligible.
1
. Introduction
The identification of suitable long-term alternatives to ozone-
depleting compounds remains a challenge due to the complex
combination of performance, safety, and environmental prop-
erties required. Some chlorofluorocarbon (CFC) replacements,
such as hydrofluorocarbons (HFCs) and hydrochlorofluorocar-
bons (HCFCs), were proposed in the Montreal Protocol and
Clean Air Act amendments. Also, hydrofluoroethers (HFEs) and
hydrofluoroalcohols (HFAs) have been proposed as a new gen-
eration of CFC alternatives and are applied in many fields, such
as cleaning of electronic components, refrigeration, and carrier
know, the gas-phase kinetics of OH and Cl with CF CH CH OH
3 2 2
[7–9]
have been studied by several authors.
data on CF (CH ) CH OH are available.
In contrast, no kinetic
3
2 2
2
Herein, we present the gas-phase kinetics for the OH and Cl
reactions with CF (CH ) CH OH at 298 K [reactions (1) and (2)]:
3
2 x
2
OH þ CF ðCH Þ CH OH ! Products
kOH
ð1Þ
ð2Þ
3
2
x
2
[
1]
compounds for lubricants. Substitution of chlorine (and/or
fluorine) atoms by hydrogen atoms in the molecule implies an
increase in the reactivity towards tropospheric hydroxyl (OH)
Cl þ CF ðCH Þ CH OH ! Products
kCl
3
2
x
2
[
2]
Also, a detailed study on the UV and IR absorption cross sec-
tions of CF (CH ) CH OH in the 200–310 nm and 500–
radicals with respect to that of CFCs. HFCs containing a CF
3
[
3]
group do not cause significant ozone depletion. Although
these HFCs and HFEs are ozone-friendly, they have other ad-
verse environmental effects, such as high global warming po-
tentials (GWPs) due to the strong absorption of infrared (IR) ra-
3
2 x
2
ꢀ
1
4
000 cm spectral regions is reported here. Additionally, ab
initio calculations were performed to simulate the IR spectra
ꢀ
1
between 0 and 4000 cm and to assign the features in the ob-
served IR spectra.
[
4]
diation by CꢀF bonds. Good candidates for CFC replacements
must not affect significantly the atmospheric radiative forcing.
In that sense, a complete understanding of their IR absorption
properties and their tropospheric lifetimes (t) would be
needed to better evaluate their environmental impact.
Based on our results, the atmospheric lifetimes of these
HFAs and their direct GWPs were calculated here. A discussion
of the suitable use of these HFAs as substitutes for ozone-de-
pleting substances is included.
Recently, some HFAs, namely CFH CH OH, CF HCH OH, and
2
2
2
2
[
a] Dr. E. Jimꢀnez, M. AntiÇolo, Dr. B. Ballesteros, Prof. E. Martꢁnez,
Prof. J. Albaladejo
Departamento de Quꢁmica Fꢁsica, Facultad de Ciencias Quꢁmicas
Universidad de Castilla-La Mancha
Avda. Camilo Josꢀ Cela, s/n. 13071 Ciudad Real (Spain)
Fax: (+34)926-29-53-18
E-mail: jose.albaladejo@uclm.es
CF CH OH, have been reported to be good candidates as CFC
3
2
[
5,6]
substitutes, since their direct GWP is insignificant.
It is ex-
pected that longer HFAs, CF (CH ) CH OH (x=1 and 2), have a
3
2 x
2
shorter t and thus lower GWP. A complete study of the gas-
phase kinetics of the tropospheric photooxidation of these
HFAs initiated by OH radicals and chlorine (Cl) atoms is needed
to better estimate t, and therefore their GWPs. As far as we
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cphc.201000365.
ChemPhysChem 2010, 11, 4079 – 4087
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4079

J. Albaladejo et al.
0
Experimental Section
or k ꢀk
0
¼ kOH½CF
3
ðCH
2
Þ
x
CH
2
OHꢃ
ð5Þ
Gas-Phase Kinetic Measurements: Experiments to measure the
where k is the rate coefficient determined in the absence of HFA.
0
rate coefficients of the OH and Cl reactions with CF (CH ) CH OH
3
2 x
2
[
10–14]
The bimolecular rate coefficient kOH at 298 K was determined from
the slope of k’ꢀk versus [CF (CH ) CH OH] plots.
were conducted with previously described setups.
A brief de-
0
3
2 x
2
scription is provided below. Prior to performing the kinetic experi-
ments, the analysis of CF (CH ) CH OH samples was carried out by
3
2 x
2
FTIR Smog-Chamber System: Room-temperature rate coefficients for
static headspace gas chromatography/mass spectrometry (GC–MS)
at room temperature to detect any impurity that could affect the
determined rate coefficients. The headspace gas was extracted
from a vial filled with the liquid sample and was injected into a
gas chromatograph (Thermo Electron Co., model Trace GC Ultra),
which separated the components of the sample based on polarity.
The separated components then went into a mass-selective detec-
tor (Thermo Electron Co., model DSQ II). The resulting mass spec-
trum allowed the identification of the components using a stan-
dard reference library.
the reaction of Cl atoms with CF (CH ) CH OH were obtained rela-
3
2 x
2
tive to propene and/or ethene using Cl photolysis as a source of
2
Cl atoms. All relative kinetic experiments were carried out in a
glass environmental chamber (L=0.90 m, V=16 L) that has been
[13,14]
described previously.
It was interfaced to a Fourier transform
IR spectrometer via a three-mirror silver-coated Hanst reflection
system, which provided an optical path length of 96 m. The reactor
was surrounded by four germicidal lamps (Philips TL-K 40 W),
which emitted radiation over the range 300–460 nm and were
used to photochemically initiate the experiments. Mixtures con-
taining Cl2 (8–15 ppm), the HFA (7–11 ppm), and the reference
compound (7–15 ppm) in synthetic air (740 Torr) were irradiated
for periods of between 60 and 90 min. After irradiation of the reac-
tion mixture, the IR spectrum was recorded every 10–15 min after
The purity of CF CH CH OH in the gas phase was found to be ap-
3
2
2
proximately 99% and the main impurity (<0.5%) was ethanol. For
gaseous CF (CH ) CH OH, the purity was found to be about 96%.
3
2 2
2
In this case, the impurities were identified as 3,3,3-trifluoro-1-pro-
ꢀ1
3
2 accumulations at a spectral resolution of 2 cm .
pene (CF CH=CH , 3%) and 2,2,2-trifluoroethyl iodide (CF CH I,
3
2
3
2
1
%). The influence of these impurities on the rate coefficients will
The relative disappearance rates of HFAs and the reference organic
compound were measured in the presence of Cl atoms [reac-
tions (2) and (6)]:
be discussed in Section 2.
Pulsed Laser Photolysis Coupled to Laser-Induced Fluorescence (LIF)
Detection: Flows of buffer gas (He), helium bubbled through an
aqueous solution of H O , and fluoroalcohol mixtures (containing
Cl þ CF ðCH Þ CH OH ! Products
kCl
ð2Þ
ð6Þ
3
2
x
2
2
2
0
.8–1.7% of CF CH CH OH and 0.2–0.9% of CF (CH ) CH OH diluted
3 2 2 3 2 2 2
Cl þ REF ! Products
kCl,ref
in He) were introduced into the reactor at a total flow rate of be-
3
ꢀ1
tween 140 and 540 cm min , STP (standard conditions of temper-
ature and pressure). Flow rates of helium and reactant/inert gas
mixtures were measured by calibrated mass flow meters. The con-
centrations of the gases in the reactor were determined by meas-
uring the mass flow rates and the total pressure with a capacitance
manometer. OH radicals were produced by the pulsed photolysis
where REF is a reference compound for which the Cl rate coeffi-
cient, kCl,ref, is reliably known. Provided that CF (CH ) CH OH and
3
2 x
2
the reference compound react only with Cl, the following expres-
sion can be derived [Eq. (7)]:
ꢀ
ꢁ
½
HFAꢃ
kCl
½REFꢃ
ln
0
0
(
repetition rate of 10 Hz) of gaseous H O at 248 nm from a KrF ex-
ln
ꢀ klosst ¼ k
ꢀ kloss;r
t
ð7Þ
2
2
½
HFAꢃ
Cl;ref
½REFꢃ
t
t
cimer laser [reaction (3)]:
where the subscripts 0 and t refer to the beginning of the reaction
and to a certain reaction time t, respectively; kloss and kloss,r are the
rate coefficients for the wall losses plus the photolysis process of
HFAs and the reference compounds, respectively, measured in ex-
H O þ hnðl ¼ 248 nmÞ ! 2 OH
ð3Þ
2
2
Hydrogen peroxide (initially 50 wt% in water) was concentrated by
bubbling helium through the solution for several days prior to use
to remove water. From the photolysis laser fluence (3.9–
periments performed in the absence of Cl precursor. The k was
loss
ꢀ5 ꢀ1
ꢀ5 ꢀ1
1
.1ꢂ10
(CH ) CH OH; k was 2.2ꢂ10
loss,r
s
for CF CH CH OH and 2.5ꢂ10
s
for CF -
3
2
2
3
ꢀ
1
ꢀ2
8
.5 mJpulse cm ), the absorption cross section of H O , and the
ꢀ5 ꢀ1
ꢀ5 ꢀ1
2
2
s
for propene and 1.8ꢂ10
s
2
2
2
[15]
quantum yield for OH production at 248 nm, upper limits of the
initial OH concentration were estimated to be lower than 9.0ꢂ
for ethene. For the kinetic analysis, Equation (7) can be expressed
in terms of the integrated areas (A ) of selected bands for CF -
int
3
1
0
ꢀ3
1
0 moleculecm . The OH temporal profiles were measured by
(
[
CH ) CH OH and the reference compound. A is defined as
2 x 2 int
Eq. (8)]:
LIF using a Nd:YAG (Continuum) pumped frequency-doubled dye
laser which was triggered at a variable time after the photolysis
Z
2
+
2
pulse. The probe pulse excited the (1,0) band of the A S –X P
transition of the OH radical at around 282 nm. LIF from excited OH
radicals was used to monitor the OH decay during the course of
the reaction. The OH LIF signal was averaged by a boxcar unit
from 50 to 100 times and recorded onto a computer. The radical
decay signal at each reactant concentration was analyzed as de-
Aint
¼
Að ~n Þd ~n
ð8Þ
band
where Að ~n Þ is the absorbance at a given wavenumber. For
CF CH CH OH and CF (CH ) CH OH, some bands were selected be-
3
2
2
3
2 2
ꢀ1
2
[10–12]
ꢀ1
scribed by Albaladejo and Jimꢀnez et al.
to obtain the pseudo-
tween 1415 and 1032 cm and 1427 and 999 cm , respectively.
For the reference compounds, the band centered at 912 cm for
propene and the band centered at 949 cm for ethene were used
in the analysis. Thus, k /k was obtained from the slope of the
ꢀ1
first-order decay rate coefficient (k’) due to the reaction under
study. k’ includes the disappearance of the OH radical in the reac-
tor due to reaction (1) and to its reaction with H O and impurities,
ꢀ1
2
2
Cl Cl,ref
as well as the diffusion loss process [Eqs. (4) and (5)]:
plot of Equation (7) by linear least-squares analysis.
Absorption Cross Sections of CF (CH ) CH OH: UV Absorption
Cross Sections, s : The gas-phase absorption spectroscopy tech-
l
3
2
x
2
k0 ¼ kOH½CF ðCH Þ CH OHꢃ þ k
ð4Þ
3
2
x
2
0
4
080
www.chemphyschem.org
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPhysChem 2010, 11, 4079 – 4087

Atmospheric Lifetimes of Hydrofluoroalcohols
nique was used to measure UV absorption cross sections of CF3-
(
CH ) CH OH between 200 and 310 nm at room temperature. The
2 x 2
experimental setup and the methodology, based on the Beer–Lam-
[16,17]
bert law, have been previously described elsewhere.
The ab-
sorption spectra of both CF (CH ) CH OH were recorded under
3
2 x
2
static conditions using pure HFA. The 107-cm-long absorption cell
was filled with HFA pressures that ranged from 1.0 to 5.3 Torr. UV
absorption cross sections (base e), s , at a given wavelength in
l
the 200–310 nm range were obtained from the slope of the ab-
sorbance versus HFA pressure.
In Figure 1S of the Supporting Information some examples of the
Beer–Lambert plots for CF CH CH OH and CF (CH ) CH OH are dis-
3
2
2
3
2 2
2
played at several wavelengths. These plots were linear at all wave-
lengths with intercepts close to zero.
Figure 1. Structural parameters for CF CH CH OH and CF (CH ) CH OH calcu-
3
2
2
3
2
2
2
lated by the MP2/6-31G(d) level of theory.
IR Absorption Cross Sections, s : IR absorption cross sections of CF -
n
3
(
4
CH ) CH OH were measured in the wavenumber range of 500–
000 cm at a spectral resolution of 1 cm . The experimental
2
x
2
ꢀ
1
ꢀ1
CF CH CH OH and CF (CH ) CH OH. During the optimization pro-
3
2
2
3
2 2
2
cess, the anti isomer of CF CH CH OH evolves to the syn one, and
setup employed in this study has been recently described by Anti-
3
2
2
[18]
the syn isomer of CF (CH ) CH OH evolves to the anti one.
Çolo et al. Typically, spectra were obtained from the co-addition
of 16 interferograms, which required about 1 min of acquisition
time. A White-type gas cell was used with a fixed optical path
length of 8 m. IR spectra were recorded under static conditions
from diluted mixtures of CF (CH ) CH OH in helium or air. Dilution
3
2 2
2
In this work the harmonic vibrational frequencies were calculated
using the gradient-corrected level of density functional theory
(DFT) utilizing Becke’s three-parameter exchange functional and
3
2 x
2
the Lee–Yang–Parr correlation functional (B3LYP). Scott and
factors ranged from 0.10 to 0.24%. The total pressure inside the
FTIR cell ranged from 5 to 164 Torr. The concentration of the HFA
was calculated from the measured partial pressure inside the FTIR
[20]
Radom reported that the B3LYP/6-31G* method yields a root-
ꢀ1
mean-square error of 34 cm for a set of 122 molecules and 1066
[21]
experimental vibrational frequencies. Hashikawa et al., for exam-
14
16
ꢀ3
cell ([CF (CH ) CH OH]=(1.9ꢂ10 –1.2ꢂ10 ) moleculecm ).
3
2 x
2
ple, used the same computational method for predicting the IR
The absorption cross section (base 10) at each wavenumber, sn,
spectra of fluorinated aldehydes, CxF2x+1CHO (x=1–4), which re-
produces with good accuracy the observed fundamental vibration-
al frequencies.
was obtained by applying the Beer–Lambert law as described in
AntiÇolo et al. by plotting the absorbance versus HFA concentra-
tion [Eq. (9)]:
[18]
A
ꢀ
¼ sn
ꢀ
‘½CF ðCH Þ CH OHꢃ
ð9Þ
n
3
2
x
2
2
. Results
Some examples of the plots of Equation (9) at several wavenum-
bers are displayed in Figure 1S of the Supporting Information for
CF CH CH OH and CF (CH ) CH OH. The absorption bands were ob-
2.1. Rate Coefficients (kOH and kCl) for CF (CH ) CH OH
3 2 x 2
3
2
2
3
2 2
2
An example of the pseudo-first-order plot at room tempera-
ture is shown in Figure 2 for reactions (1). As can be seen, no
pressure dependence of kOH was observed in the 45–210 Torr
range. From the slope of these plots, the absolute rate coeffi-
cients for the reactions of OH radicals with CF CH CH OH and
served to obey the Beer–Lambert law in the concentration range
investigated.
Reactants: Liquid samples of HFAs were from Apollo Scientific Lim-
ited and were degassed by freeze–pump–thaw cycles prior use.
Purities in the liquid samples were 98 and 97% for CF CH CH OH
3
2
2
3
2
2
CF (CH ) CH OH were obtained and are listed in Table 1. Also,
and CF (CH ) CH OH, respectively. An aqueous solution of H O
2
3
2 2
2
3
2 2
2
2
(
Sharlau, 50% w/v) was treated as previously described by Jimꢀnez
Gases from Praxair were employed as supplied: He
99.999%), synthetic air (99.999%), and Cl (99.8%). Reference com-
2
the employed experimental conditions (p range, number of
T
[
11]
et al.
(
experiments, and HFA concentrations) are listed in Table 1. Due
to the low vapor pressure of these HFAs and the presence of
small amounts of impurities, the main systematic error in kOH is
related to the measurement of the HFA concentration in the
reaction cell and the pseudo-first-order rate coefficients (k’).
Considering the presence of CH CH OH (0.5% in
pounds were ethene (Sigma–Aldrich, 99%) and propene (Fluka,
8%).
9
Computational Methods: The harmonic vibrational frequencies and
IR intensities of CF CH CH OH and CF (CH ) CH OH, as well as the
3
2
2
3
2 2
2
3
2
geometries of both fluorinated alcohols, were obtained with the
CF CH CH OH) and of CF CH=CH (3% in CF (CH ) CH OH), k’
3
2
2
3
2
3
2 2
2
[19]
Gaussian 03 package of programs running on the Supercomput-
ing Service of the University of Castilla-La Mancha. The geometry
optimization of fluorinated alcohols was performed by using the
second-order Møller–Plesset perturbation theory (MP2/6-31G*). The
optimized geometries of CF CH CH OH and CF (CH ) CH OH are
depicted in Figure 1, where the most important structural parame-
ters are shown. These geometries were verified to be true potential
energy minima by the absence of imaginary vibrational frequen-
cies. Two conformational isomers with respect to the position of
the OH group and F atom were found and optimized for
must be corrected to account for the possible OH loss due to
[22,23]
reactions (10) and (11):
OH þ CH CH OH ! Products
k ¼ 3:26 ꢄ 10 cm molecule
3
2
3
2
2
3
2 2
2
ð10Þ
ð11Þ
ꢀ
12
3
ꢀ1 ꢀ1
s
OH þ CF CH¼CH ! Products
3
2
ꢀ
12
3
ꢀ1 ꢀ1
k ¼ 1:53 ꢄ 10 cm molecule
s
ChemPhysChem 2010, 11, 4079 – 4087
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemphyschem.org
4081

J. Albaladejo et al.
Figure 3. Examples of the relative kinetic data for the reaction of Cl atoms
with CF
3 2 2 3 2 2 2
CH CH OH (black symbols) and CF (CH ) CH OH (white symbols) at
298 K and 750 Torr.
For reactions (2) some examples of the plot of Equation (7)
are shown in Figure 3 for different reference compounds. The
rate coefficient ratios, k /kCl,ref, and the extracted rate coeffi-
Cl
cients kCl are listed in Table 2, together with the experimental
conditions used. kCl,ref values for propene and ethene were
1
0
taken from the literature (k (ethene)=1.1ꢂ10 and kCl-
Cl
10
3
ꢀ1 ꢀ1 [24]
(
propene)=2.7ꢂ10 cm molecule s ). The uncertainty as-
Figure 2. Absolute kinetic data at 298 K as a function of total pressure for
the reaction of OH radicals with a) CF CH CH OH and b) CF (CH CH OH.
sociated with kCl is the standard deviation of the fit to Equa-
tion (7). As can be seen, no difference outside the error limits
was observed in kCl for CF (CH ) CH OH using different refer-
3
2
2
3
2
)
2
2
3
2 2
2
ence compounds. Thus, a weighted average value of kCl using
The contribution of these reactions to kOH was estimated to
be around 1%. Therefore, the uncertainty stated in Table 1 is
the statistical standard deviation plus an error of ꢁ10%. The
averaged rate coefficients obtained in this work at 298 K are as
follows:
both reference compounds is recommended at 298 K:
ꢀ
11
3
ꢀ1 ꢀ1
k ðCF CH CH OHÞ ¼ ð1:60 ꢁ 0:45Þ ꢄ 10 cm molecule
s
Cl
3
2
2
ꢀ
11
3
ꢀ1 ꢀ1
k ðCF ðCH Þ CH OHÞ ¼ ð8:71 ꢁ 0:24Þ ꢄ 10 cm molecule
s
Cl
3
2
2
2
ꢀ
13
3
ꢀ1 ꢀ1
kOHðCF CH CH OHÞ ¼ ð9:7 ꢁ 1:1Þ ꢄ 10 cm molecule
s
A summary of these results is presented in Tables 1 and 2.
The fact that kOH and kCl values for CF (CH ) CH OH are higher
3
2
2
3
2 2
2
ꢀ
12
3
ꢀ1 ꢀ1
kOHðCF ðCH Þ CH OHÞ ¼ ð2:62 ꢁ 0:32Þ ꢄ 10 cm molecule
s
than those for CF CH CH OH can be explained by the increase
3 2 2
3
2
2
2
of methylene groups in the mol-
ecule that are susceptible to
being abstracted by OH and Cl
Table 1. Absolute rate coefficients for the reactions of OH radicals with CF
98 K.
3
CH
2
CH
2
OH and CF
3
(CH
2
)
2
CH
2
OH at
2
atoms in the C alcohol. The ob-
4
served reactivity enhancement
is more pronounced for the Cl
reactions than for the corre-
sponding reactions with OH rad-
icals. Thus, the rate ratio k(CF3-
1
2
Compound
p
T
[Torr]
No. of experiments
[CF
3
(CH
2
)
x
CH
2
OH]
k
OH ꢂ10
1
4
ꢀ3
3
ꢀ1 ꢀ1
[
10 moleculecm
]
[cm molecule
s
]
CF
CF
3
CH
(CH
2
CH
2
OH
OH
47–211
46–203
8
8
1.1–13.0
0.26–8.16
0.97ꢁ0.11
2.62ꢁ0.32
3
2
)
2
CH
2
(
CH ) CH OH)/k(CF CH CH OH)
2 2 2 3 2 2
was observed to be 3 for the
OH reaction and 5 for the Cl re-
action. A comparison between
Table 2. Relative rate coefficients for the reactions of Cl atoms with CF
98ꢁ1 K and 740 Torr of air.
3
CH
2
CH
2
OH and CF
3
(CH
2
)
2
CH
2
OH at
2
kOH
and
kCl
values
for
[
a]
11
Compound
No. of experiments
Reference
kCl/kref
kCl ꢂ10 /
CF CH CH OH obtained in this
3
ꢀ1 ꢀ1
3
2
2
compound
[cm molecule
s
]
work with those found in the lit-
erature is shown in Tables 3 and
CF
CF
3
CH
(CH
2
CH
2
OH
OH
4
4
4
ethene
ethene
propene
0.15ꢁ0.04
0.80ꢁ0.03
0.31ꢁ0.02
1.60ꢁ0.45
8.80ꢁ0.35
8.40ꢁ0.65
3
2
)
2
CH
2
4, respectively. Within the error
limits, these rate coefficients are
in excellent agreement with
1
0
10
3
ꢀ1 ꢀ1 [24]
[a] kCl(ethene)=1.1ꢂ10 and kCl(propene)=2.7ꢂ10 cm molecule s .
4
082
www.chemphyschem.org
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPhysChem 2010, 11, 4079 – 4087

Atmospheric Lifetimes of Hydrofluoroalcohols
tions. A similar mechanism is ex-
pected for the reaction of OH
and Cl with CF (CH ) CH OH, to
Table 3. Comparison of the rate coefficients for the reactions of OH with CF
from the literature.
3
(CH
2
)
x
CH
2
OH at 298 K with those
3
2 2
2
1
2
[a]
Compound
p
T
[Torr]
k
[
OH ꢂ10
Technique
Reference
produce CF (CH ) CHO.
3
2 2
3
ꢀ1 ꢀ1
m molecule
s
]
[
39]
CF
3
3
CH
2
2
OH
700
0.102
0.123
0.10
0.97ꢁ0.11
0.691
1.06
0.89
2.62ꢁ0.32
RR/FTIR
RR/FTIR
DF-PLP-FP/LIF
PLP/LIF
RR/FTIR
RR/GC-FID/FTIR
PLP/LIF
Hurley et al.
[6]
7
4
60
–70
Sellevꢄg et al.
2.2. Absorption Cross Sections
[
40]
Tokuhashi et al.
This work
of CF (CH ) CH OH
CF
CH
CH
2
OH
47–211
3
2 x
2
[
7]
700
760
100
Hurley et al.
[
8]
2.2.1. UV Region (200–310 nm)
Kelly et al.
Kelly et al.
This work
[
8]
The UV absorption spectra of
CF
3
(CH
2
)
2
CH
2
OH
46–203
PLP/LIF
CF CH CH OH
and
CF3-
3
2
2
[
a] RR, relative rate; FTIR, Fourier transform infrared; DF, discharge flow; PLP, pulsed laser photolysis; LIF, laser-
(
CH ) CH OH between 200 and
2
2
2
induced fluorescence; GC, gas chromatography; FID, flame ionization detection.
310 nm are shown in Figure 4.
As can be seen in the inset, the
absorption of HFAs drastically
decreases below 230 nm (s<5ꢂ
ꢀ
22
2
ꢀ1
Table 4. Comparison of the rate coefficients for the reactions of Cl with CF
from the literature.
3
(CH
2
)
x
CH
2
OH at 298 K with those
Reference
10 cm molecule ). At wave-
lengths greater than 230 nm,
the absorption is too low to
1
1
[a]
Compound
p
T
[Torr]
k
[
Cl ꢂ10
Technique
3
ꢀ1 ꢀ1
cm molecule
s
]
properly characterize s . There-
l
[
26]
fore, averaged sl (in base e)
values between 200 and 230 nm
are listed in Table 1S of the Sup-
porting Information. In the ac-
tinic region (above 290 nm), the
photolysis of CF CH CH OH and
CF
3
CF
3
CF
3
CH
2
2
OH
50–210
0.071
0.065
0.0742
0.063
1.6ꢁ0.45
1.59
2.24
1.90
8.71ꢁ0.24
PLP/RF
RR/FTIR
RR/FTIR
DF/MS
RR/FTIR
RR/FTIR
RR/GC-FID/FTIR
DF/MS
RR/FTIR
Garzꢅn et al.
Hurley et al.
[
39]
7
7
2
00
60
ꢂ10
[
6]
Sellevꢄg et al.
ꢀ
3
[25]
Papadimitriou et al.
This work
CH
CH
2
OH
740
[
7]
7
7
2
00
60
ꢂ10
Hurley et al.
Kelly et al.
[
8]
3
2
2
ꢀ
3
[9]
Papadimitriou et al.
This work
CF (CH ) CH OH is then expect-
3 2 2 2
(CH
2
)
2
CH
2
OH
740
ed to be negligible.
[a] PLP, pulsed laser photolysis; RF, resonance fluorescence; RR, relative rate; FTIR, Fourier transform infrared;
DF, discharge flow; MS, mass spectrometry; GC, gas chromatography; FID, flame ionization detection.
ꢀ
1
2
.2.2. IR Region (0–4000 cm )
Experimental
IR
absorption
ꢀ
1
those previously reported by absolute and relative methods,
thus giving us confidence in the rate coefficients reported here
for CF (CH ) CH OH.
spectra of CF (CH ) CH OH between 500 and 4000 cm are dis-
3 2 x 2
played in Figures 5a and 6a, respectively. For each HFA, s (in
n
3
2 2
2
base e) values are listed as a function of wavenumber in
Table 2S of the Supporting Information. There was no discerna-
Also, Tables 3 and 4 show a comparison between the OH
and Cl reactivity towards CF CH OH, CF CH CH OH, and CF -
3
2
3
2
2
3
ble effect of diluent (He or synthetic air) on s or on the inte-
n
ꢀ
1
(
CH ) CH OH. The deactivating effect of the CF group in k
grated absorption cross section, S , over the 500–4000 cm
2
2
2
3
OH
int
and kCl is stronger for CF CH OH. For instance, the OH and Cl
range for both HFAs. Sint is defined as follows [Eq. (12)]:
3
2
rate coefficients for CF CH OH are much lower (on the order of
3
2
ꢀ
13
3
ꢀ1 ꢀ1
Z
n2
10
cm molecule s ) than those for the HFAs studied in
Sint
¼
sð ~n Þd ~n
ð12Þ
this work. Rate ratios k(CF (CH ) CH OH)/k(CF (CH ) CH OH)
3
2 x
2
3
2 xꢀ1
2
n1
range from about 3 (x=2) to 10 (x=1) for the OH reactions,
whereas the ratio ranges from 5 to 25 for the Cl reactions. This
increase in kOH and kCl with the length of the hydrocarbon
chain has also been observed for the OH and Cl rate coeffi-
where sð ~n Þ is the absorption cross section at a wavenumber ~n .
In this work,
S
(base e) values were (1.66ꢁ0.04)ꢂ
int
ꢀ1
[18]
ꢀ16
2
ꢀ1
cients for partially fluorinated aldehydes CF (CH ) CHO.
10 cm molecule cm
(for x=1) and (1.06ꢁ0.02)ꢂ
3
2 x
ꢀ
16
2
ꢀ1
ꢀ1
Regarding the degradation products of reactions (1) and (2),
mechanistic studies carried out both in the absence and pres-
10 cm molecule cm (for x=2). Stated uncertainties are
[27]
ꢁ2s (only statistical error). Sellevꢄg et al. reported Sint for
[
6,25,26]
ꢀ1
ence of NO on the OH and Cl reactions with CF CH OH
seven bands between 515 and 3760 cm , which resulted in a
x
3
2
[
7–9]
ꢀ16
2
ꢀ1
ꢀ1
and CF CH CH OH
confirm that the main reaction pathway
total Sint of 1.5ꢂ10 cm molecule cm , in excellent agree-
ment with our data. These authors stated that the reported ab-
sorption cross sections of CF CH CH OH were 5–10% lower
3
2
2
proceeds via hydrogen atom abstraction from the CH group
2
adjacent to the OH group, since the main primary product was
the corresponding aldehyde. Theoretical studies recently per-
formed by our group on the Cl-initiated reaction with
3
2
2
[
28]
than those previously reported by Waterland et al. As shown
in Figure 5a, the band positions and shapes for CF CH CH OH
3
2
2
[
26]
CF CH OH
are consistent with the experimental observa-
are in excellent agreement with the high-resolution spectrum
3
2
ChemPhysChem 2010, 11, 4079 – 4087
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemphyschem.org
4083

J. Albaladejo et al.
Figure 4. UV absorption spectra of CF
*) between 200 and 310 nm at 298 K. Inset: absorption of the HFAs below
30 nm.
3
CH
2
CH
2
OH (
&
3 2 2 2
) and CF (CH ) CH OH
(
2
3 2 2 2
Figure 6. a) Experimental and b) calculated IR spectra of CF (CH ) CH OH ob-
tained in this work at room temperature.
ber was observed between the experimental and theoretical
spectra of CF (CH ) CH OH. The averaged shift is 27 and
3
2 x
2
ꢀ
1
3
7 cm for CF CH CH OH and CF (CH ) CH OH, respectively.
3 2 2 3 2 2 2
In Tables 5 and 6, some of the assigned vibrational modes
for these HFAs are described. The most intense calculated
ꢀ
1
bands were those centered at 1141 and 1130 cm
for
CF CH CH OH and CF (CH ) CH OH, respectively. These bands
3
2
2
3
2 2
2
correspond to a complex normal mode that includes the Hꢀ
OꢀC bending, CꢀF stretching, and ꢀCH ꢀ wagging modes.
2
This complex vibrational mode was also described by Water-
land et al. for a series of fluorinated alcohols, including
[
28]
CF CH CH OH. The second largest peak appears at 1356 and
3
2
2
1
ꢀ
1
244 cm for CF CH CH OH and CF (CH ) CH OH, respectively.
3 2 2 3 2 2 2
This normal mode is again a complex motion in both mole-
cules that affects the H-O-C bending, (F) CꢀCH stretching, and
3
2
ꢀ
CH ꢀ wagging modes. In the simulated spectra of
2
ꢀ1
Figure 5. a) Experimental and b) calculated IR spectra of CF
tained in this work at 298 K. Inset in (a): Comparison of the IR spectrum of
3
CH
2
CH
2
OH ob-
CF CH CH OH, the band centered at 1229 cm is the sum of
3 2 2
ꢀ
1
the peaks at 1229 and 1226 cm , which result from the artifi-
cial broadening with Lorentzian lineshape functions. As Fig-
ures 5 and 6 show, the location and the relative intensity of
the major spectral features in the experimental and theoretical
spectra are in reasonably good agreement.
ꢀ1
CF
3 2 2
CH CH OH with previously reported data between 800 and 1700 cm by
[
28]
Waterland et al.
[
28]
reported by Waterland et al. Therefore, s is not resolution
n
ꢀ
1
dependent between 0.004 and 1 cm .
Theoretical spectra allow the assignment of the bands and
ꢀ
1
the simulation of the spectrum in the 0–500 cm region. The
ꢀ1
calculated IR spectra for both HFAs in the region 0–4000 cm
are shown in Figures 5b and 6b. The calculated frequencies 3. Discussion
were scaled by a factor of 0.9614 as proposed by Scott and
3
.1. Estimation of Atmospheric Lifetimes of Hydro-
[
20]
Radom. The theoretical spectral peaks have been artificially
broadened with Lorentzian lineshape functions of half width at
fluoroalcohols, tHFA
ꢀ
1
half maximum of 10 cm to allow comparison with the corre-
sponding experimental spectra. A slight shift in the wavenum-
The tropospheric removal rate of CF (CH ) CH OH, kloss(HFA),
3
2 x
2
can be defined as [Eq. (13)]:
4
084
www.chemphyschem.org
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPhysChem 2010, 11, 4079 – 4087

Atmospheric Lifetimes of Hydrofluoroalcohols
1
1
klossðHFAÞ ¼ t ¼ t
þ
Table 5. Spectral assignment of the IR features of CF
trum.
3
CH
2
CH
2
OH and comparison with the experimental spec-
HFA
OH
1
1
þ t ¼ kOH½OHꢃ þ k ½Clꢃ þ k
ꢀ1
ꢀ1
Cl
wet
~n calc [cm
]
~n exp [cm
]
Intensity
5
1
127
1
6
30
3
67
8
105
162
74
72
118
36
20
13
3
45
44
7
12
17
Description of vibrational mode
rocking
twisting CH + CF + rocking CH
2 2 2
tCl
wet
6
9
3
9
4
24
ð13Þ
HꢀOꢀC out-of-plane bending
CF + CH rocking
where t is the lifetime of HFA
due to each loss process in the
troposphere (i.e., tOH and tCl,
due to reactions (1) and (2), re-
spectively, and twet, due to the
removal by wet deposition). As
stated in Section 2, photolysis of
509
2
2
527
635
786
988
552.1
659.6
CF
3
umbrella
rocking
twisting + rocking
1011
1008
027
141
152
HꢀOꢀC in-plane bending + C –C stretching
1
2
1
1052
1139.8
CꢀOH stretching
see text
1
1
rocking
rocking + twisting
HꢀOꢀC in-plane bending + wagging
twisting
HꢀOꢀC in-plane bending + wagging
CF CH CH OH
and
CF3-
3
2
2
1226
1229
1224
1259.4
(CH ) CH OH in the actinic
2
2
2
1292
1430
1453
1500
region (above 290 nm) is ex-
pected to be negligible com-
pared to their homogeneous re-
moval.
1375.1
1440.2
HꢀC
HꢀC
2
ꢀH scissors bending
1
ꢀH scissors bending
2918
957
2985
2906.5
2952.8
2960
2984.6
3674.6
C ꢀH sym. stretching
1
2
C
C
C
1
ꢀH asym. stretching
2
Averaged (24 h) oxidant con-
6
ꢀ3[29]
ꢀH sym. stretching
ꢀH asym. stretching
centrations of 1ꢂ10 cm
for
3
3
034
581
2
2
OH radicals and <10 atom
OꢀH stretching
ꢀ
3[30]
cm
for Cl were used to esti-
mate global tOH and tCl values.
The values of kOH and kCl mea-
sured at room temperature can
be used to provide a local esti-
mate of the lifetime of
Table 6. Spectral assignment of the IR features of CF CH CH OH and comparison with the experimental spec-
trum.
3
2
2
ꢀ
1
ꢀ1
~
[cm
]
~n exp [cm
]
Intensity
119
14
0.1
10
2
Description of vibrational mode
calc
3
623
746
8
8
1
1
31
HꢀOꢀC out-of-plane bending
CF CH CH OH
and
CF3-
3
2
2
656.2
3
CF umbrella
(
CH ) CH OH. As shown in
rocking
2
2
2
19
47
840
901
CꢀF sym. stretching + CꢀCF
rocking and wagging
HꢀOꢀC in-plane bending + C
twisting + rocking
3
stretching
ꢀC stretching
Table 7, tHFA estimated in this
work with respect to the homo-
geneous reaction with OH radi-
cals is short (12 days for
CF CH CH OH and 4 days for
001
020
4
1
2
96
46
45
228
69
2
163
14
124
49
50
4
7
5
4
60
65
4
24
1
1025
046
130
CꢀOH + C
2
ꢀC
3
stretching
1
1051
1157.7
1208
3
2
2
1
see text
twisting + rocking
HꢀOꢀC in-plane bending + wagging
CF (CH ) CH OH) compared with
3
2 2
2
1
1
1244
1
1
1
1
212
217
the tropospheric vertical mixing
time (1–2 months). The removal
of CF CH CH OH and CF3-
1259.9
1336.1
1392.9
275
298
307
387
twisting
3
2
2
HꢀOꢀC in-plane bending + wagging
twisting
(
CH ) CH OH by the Cl reaction
2 2 2
is globally negligible with re-
spect to that by OH radicals,
since tHFA is >20 and 4 years, re-
spectively. Under certain condi-
tions, when Cl concentrations
1453.7
HꢀOꢀC in-plane bending + wagging
HꢀOꢀC in-plane bending + wagging
1428
1
1
1
2
2911
2960
2
2
3022
3578
452
474
496
883
scissoring C
scissoring C
scissoring C
2
1
1
and C
, C , and C
and C
3
2
3
2
2891.5
2960
C
C
1
ꢀH sym. stretching
1
reach
a
peak
value
of
ꢀH asym. stretching
5
ꢀ3 [31]
CꢀH sym. (C and C ) stretching
2
2
2
2
3
3
3
3
1
7
1
0 cm ,
tHFA would be
972
999
CꢀH sym. (C
CꢀH sym. (C
CꢀH sym. (C
and C
and C
and C
) stretching
) stretching
) stretching
days for CF CH CH OH and
3
2
2
day for CF (CH ) CH OH.
3
2 2
2
32
13
Another removal process to
3676
OꢀH stretching
take into account is the possible
uptake and hydrolysis of these
HFAs in clouds. Estimation of twet was performed at a height in
and CF (CH ) CH OH to that for CF CH OH determined by
3
2 2
32]
2
3
2
[
ꢀ1
the troposphere of 500 m and assuming a global average pre-
Chen et al.
as a function of temperature (255matm at
ꢀ
1
ꢀ1
cipitation rate of 500 mmyear . Henry’s law coefficients, k
276 K and 81.7matm at 291 K). Using the procedure de-
H,cp
ꢀ
1
[12,33]
(
in matm ), were not found in the literature for these fluori-
scribed by Jimꢀnez et al.,
twet would range between ap-
nated alcohols. We assumed a similar kH,cp for CF CH CH OH
proximately 2 and 6 months. Contrary to CF CH OH, the deple-
3
2
2
3
2
ChemPhysChem 2010, 11, 4079 – 4087
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemphyschem.org
4085

J. Albaladejo et al.
genated compounds with lifetimes shorter than 0.5 year to es-
Table 7. Tropospheric lifetimes (t) for CF
used in the calculation of GWPs.
3
CH
2
CH
2
OH and CF
3
(CH
2
)
2
2
CH
2
OH
OH
[5,6,37,38]
timate their radiative forcing.
As radiative forcing is an
intermediate parameter in GWP calculation and is not affected
by t values, it would be worthwhile to make a comparison of
the radiative forcing for CF (CH ) CH OH with CFC-11 or CO ,
Loss process
t [days]
CF CH CH
t [days]
CF
3
2
2
OH
3
(CH
2
)
CH
2
3
2 x
2
2
[
a]
[a]
OH reaction
Cl reaction
wet deposition
total
12
7233
59–184
4
together with GWP. The short-, medium-, and long-term effect
of the emission of a certain amount of CF (CH ) CH OH on radi-
[
a]
[a]
1329
59–184
3.7–4
[
b]
[b]
3
2 x
2
[
a]
[a]
ative forcing is usually calculated for a typical time horizon of
0, 100, and 500 years. The radiative efficiency (radiative forc-
10–12
2
ꢀ
1
[
8
a] This work. [b] Based on kH,cp for CF
1.7matm at 291 K) reported by Chen et al.
3
CH
2
OH (255matm at 276 K and
ꢀ
1
[32]
ing per ppbv) and GWP (relative to CO ) for a typical time hori-
2
zon (TH) were calculated and are listed in Table 8. As can be
seen, our simple calculations provide a radiative efficiency for
ꢀ
2
ꢀ1
tion of CF CH CH OH and CF (CH ) CH OH through the uptake
CF CH CH OH (0.2 Wm ppbv ) similar to that of CFC-11
3
2
2
3
2 2
2
3
2
2
ꢀ
2
ꢀ1
in cloud water or rain is not of significance compared with
their homogeneous removal. However, further studies on their
air/water partitioning and hydrolysis process would be needed
to corroborate the earlier statement.
(0.25 Wm ppbv ); however, due to the shorter lifetime of
the HFA the GWP dramatically decreases with horizon time.
CF CH CH OH and CF (CH ) CH OH, which are quickly removed
3
2
2
3
2 2
2
from the atmosphere, may initially have a large effect, but for
longer time periods it becomes less important. For example,
the 100-year GWP of CF CH CH OH is about 2.6, which means
Thus, atmospheric lifetimes of the studied HFAs can range
between 10 and 12 days for CF CH CH OH and between 3.7
3
2
2
3
2
2
and 4 days for CF (CH ) CH OH. As stated above, these life-
that the instantaneous release of a single kilogram of this spe-
cies would have the same effect on the radiative forcing over
the subsequent 100 years as the instantaneous release of
2.6 kg of CO . The estimated radiative efficiency for CF -
3
2 2
2
times are shorter than the average tropospheric transport
timescale and these species are not uniformly mixed through
the troposphere. However, in Equation (13) the opposite is as-
sumed and, thus, the lifetimes given in Table 7 provide only an
approximate estimate of the global mean lifetimes. The region-
al variations in the OH radical concentration as well as the spa-
tial and seasonal distributions of CF CH CH OH and CF -
2
3
ꢀ
2
ꢀ1
(CH ) CH OH (0.11 Wm ppbv ) is half that of CFC-11
2
2
2
ꢀ
2
ꢀ1
(0.25 Wm ppbv ) and its GWP (relative to CO ) drastically de-
2
creases from 1.8 (at TH=20 years) to 0.3 (at TH=500 years).
These results, together with the low abundances of these fluo-
rinated alcohols in the atmosphere, show that direct radiative
effects are expected to be negligible and these HFAs may not
contribute to the greenhouse effect. Also, the indirect radiative
effects due to the atmospheric degradation products of these
fluorinated alcohols (mainly hydrofluoroaldehydes, CF3-
(CH ) CHO) are expected to be of minor importance.
3
2
2
3
(
CH ) CH OH sources may be taken into account.
2 2 2
3.2. Global Warming Potential Calculation
Direct GWP is a measure of how an emission of one trace gas
affects future global warming relative to an emission of an
2
x
equal mass of a reference greenhouse gas, commonly CO or
2
CFCl (CFC-11). GWP depends upon the strength and spectral
3
4
. Conclusions
position of a compound’s IR absorption bands and quantifies
the future integrated radiative forcing of a unit mass pulse
Gas-phase rate coefficients of OH radicals and Cl atoms with
CF CH CH OH and CF (CH ) CH OH at 298 K are reported in
[
34]
emission of a greenhouse gas i. In the simple model of Pin-
3
2
2
3
2 2
2
[
35]
ꢀ1
nock et al.,
Sint values for a given 10 cm segment are
and
500 cm . In this work, S was obtained from the linear least-
this work by using two different experimental setups and tech-
niques. The values of kOH and kCl determined here are the first
measurements for CF (CH ) CH OH. Moreover, the IR absorp-
needed to calculate radiative forcing between
0
ꢀ
1
2
int
3
2 2
2
squares analysis of the plots of the integrated absorbance Aint
tion cross sections in the atmospheric window region have
also been measured to quantify the potential of these HFAs to
change the radiative forcing of the atmosphere.
versus the HFA concentration [Eq. (14)]:
Aint ¼ S ‘½CF ðCH Þ CH OHꢃ
ð14Þ
The major conclusion that can be drawn from our results is
that CF CH CH OH and CF (CH ) CH OH seem to be a good al-
int
3
2
x
2
3
2
2
3
2 2
2
where ‘ is the optical path
length (800 cm, in our study).
2
Table 8. Comparison of the GWPs for some trifluoro compounds relative to CO .
Calculation of GWP for a very
short lived species has not been
considered appropriate by the
World Meteorological Organiza-
tion (WMO) 2006, since a de-
tailed global modeling and
emission scenarios would be
Compound
Radiative efficiency
GWPTH=20 yr
GWPTH=100 yr
GWPTH=500 yr
ꢀ
2
ꢀ1
[Wm ppbv
]
[
a]
ꢀ5
CO2
CFCl
1.41ꢂ10
1
1
4750
70
2.6
0.5
1
1620
22
1.5
0.3
[
a]
3
(CFC-11)
0.25
0.13
0.20
0.11
6730
234
9.5
1.8
[
41]
CF
3
CF
3
CF
3
CH
CH
(CH
2
CH
CH
3
[
b]
2
2
OH
OH
[b]
2
)
2
CH
2
[
36]
needed. However, GWPs have
been calculated for some halo-
[
4]
[36]
2
[a] t=114 years for CO and 50 years for CFC-11; radiative forcing from WMO. [b] This work.
4
086
www.chemphyschem.org
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPhysChem 2010, 11, 4079 – 4087

Atmospheric Lifetimes of Hydrofluoroalcohols
ternative to CFCs, since they are short-lived species with low
GWPs. Therefore, they cannot be considered as greenhouse
gases. Based on the evaluation of their tropospheric lifetimes,
these HFAs are not expected to be well mixed in the tropo-
sphere and, thus, the atmospheric impact of CF CH CH OH and
[18] M. AntiÇolo, E. Jimꢀnez, A. Notario, E. Martꢁnez, J. Albaladejo, Atmos.
Chem. Phys. 2010, 10, 1911–1922.
[
19] Gaussian 03 (RevisionC.02), M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E.
Scuseria, M. A. Robb, J. R. Cheeseman, J. A. Montgomery, Jr., T. Vreven,
K. N. Kudin, J. C. Burant, J. M. Millam, S. S. Iyengar, J. Tomasi, V. Barone,
B. Mennucci, M. Cossi, G. Scalmani, N. Rega, G. A. Petersson, H. Nakatsu-
ji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Na-
kajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P.
Hratchian, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts,
R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Och-
terski, P. Y. Ayala, K. Morokuma, G. A. Voth, P. Salvador, J. J. Dannenberg,
V. G. Zakrzewski, S. Dapprich, A. D. Daniels, M. C. Strain, O. Farkas, D. K.
Malick, A. D. Rabuck, K. Raghavachari, J. B. Foresman, J. V. Ortiz, Q. Cui,
A. G. Baboul, S. Clifford, J. Cioslowski, B. B. Stefanov, G. Liu, A. Liashenko,
P. Piskorz, I. Komaromi, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham,
C. Y. Peng, A. Nanayakkara, M. Challacombe, P. M. W. Gill, B. Johnson, W.
Chen, M. W. Wong, C. Gonzalez, J. A. Pople, Gaussian, Inc., Wallingford,
CT, 2004.
3
2
2
CF (CH ) CH OH will depend on both the emission location
3
2 2
2
and the local atmospheric conditions and abundances. Anoth-
er conclusion is that indirect radiative effects due to the at-
mospheric degradation products (mainly hydrofluoroalde-
hydes) are expected to be negligible.
Acknowledgements
The authors would like to thank the Spanish Ministerio de Cien-
cia y Tecnologꢁa (CGL2004-03355/CLI) and the Consejerꢁa de Edu-
caciꢂn y Ciencia de la Junta de Comunidades de Castilla-La
Mancha (PCI08-0123-0381) for supporting this project. Also, M.A.
wishes to thank the Spanish Ministerio de Ciencia e Innovaciꢂn
for providing a grant. We thank A. Notario for helpful discus-
sions.
[
[
20] A. P. Scott, L. Radom, J. Phys. Chem. 1996, 100, 16502–16513.
21] Y. Hashikawa, M. Kawasaki, R. L. Waterland, M. D. Hurley, J. C. Ball, T. J.
Wallington, M. P. Sulbaek Andersen, O. J. Nielsen, J. Fluorine Chem.
2
004, 125, 1925–1932.
[22] E. Jimꢀnez, M. K. Gilles, A. R. Ravishankara, J. Photochem. Photobiol. A
003, 157, 237–245.
2
[
23] M. P. Sulbaek Andersen, O. J. Nielsen, A. Toft, T. Nakayama, Y. Matsumi,
R. L. Waterland, R. C. Buck, M. D. Hurley, T. J. Wallington, J. Photochem.
Photobiol. A 2005, 176, 124–128.
[
24] R. Atkinson, D. L. Baulch, R. A. Cox, J. N. Crowley, R. F. Hampson, R. G.
Hynes, M. E. Jenkin, M. J. Rossi, J. Troe, T. J. Wallington, Atmos. Chem.
Phys. 2006, 6, 3625–4055.
Keywords: atmospheric chemistry · gas-phase reactions ·
global warming potential · hydrofluoroalcohols · infrared
spectroscopy
[25] V. C. Papadimitriou, A. V. Prosmitis, Y. G. Lazarou, P. Papagiannakopoulos,
J. Phys. Chem. A 2003, 107, 3733–3740.
[
[
1] Environmental Protection Agency, http://www.epa.gov.
2] R. Atkinson, D. L. Baulch, R. A. Cox, J. N. Crowley, R. F. Hampson, R. G.
Hynes, M. E. Jenkin, M. J. Rossi, J. Troe, T. J. Wallington, Atmos. Chem.
Phys. 2008, 8, 4141–4496.
3] A. R. Ravishankara, A. A. Turnipseed, N. R. Jensen, S. Barone, M. Mills,
C. J. Howard, S. Solomon, Science 1994, 263, 71–75.
[26] A. Garzꢅn, M. Moral, A. Notario, A. A. Ceacero-Vega, M. Fernꢆndez-
Gꢅmez, J. Albaladejo, ChemPhysChem 2010, 11, 442–451.
[27] S. R. Sellevꢄg, B. D’Anna, C. J. Nielsen, Asian Chem. Lett. 2007, 11, 1–8.
[28] R. L. Waterland, M. D. Hurley, J. A. Misner, T. J. Wallington, S. M. L. Melo,
K. Strong, R. Dumoulin, L. Castera, N. L. Stock, S. A. Mabury, J. Fluorine
Chem. 2005, 126, 1288–1296.
[
[
4] S. A. Montzka, P. J. Fraser (lead authors) in Scientific Assessment of Ozone
Depletion: 2002, World Meteorological Organization, Geneva, 2003,
chap. 1.
5] B. Rajakumar, J. B. Burkholder, A. R. Ravishankara, Phys. Chem. Chem.
Phys. 2005, 7, 2498–2505.
[29] M. Krol, P. J. van Leeuwen, J. Lelieveld, J. Geophys. Res. [Atmos.] 1998,
103, 10697–10711.
[30] H. B. Singh, A. N. Thakur, Y. E. Chen, M. Kanakidou, Geophys. Res. Lett.
1996, 23, 1529–1532.
[31] J. Albaladejo, A. Notario, C. A. Cuevas, E. Jimꢀnez, B. CabaÇas, E. Martꢁ-
nez, Atmos. Environ. 2003, 37, 455–463.
[32] L. Chen, N. Takenaka, H. Bandow, Y. Maeda, Atmos. Environ. 2003, 37,
4817–4822.
[
[
[
6] S. R. Sellevꢄg, C. J. Nielsen, O. A. Søvde, G. Myhre, J. K. Sundet, F. Stordal,
I. S. A. Isaken, Atmos. Environ. 2004, 38, 6725–6735.
7] M. D. Hurley, J. A. Misner, J. C. Ball, T. J. Wallington, D. A. Ellis, J. W.
Martin, S. A. Mabury, M. P. Sulbaek Andersen, J. Phys. Chem. A 2005, 109,
[33] E. Jimꢀnez, B. Lanza, M. AntiÇolo, J. Albaladejo, Environ. Sci. Technol.
2009, 43, 1831–1837.
9816–9826.
[
8] T. Kelly, V. Bossoutrot, I. Magneron, K. Wirtz, J. Treacy, A. Mellouki, H.
Sidebottom, G. Le Bras, J. Phys. Chem. A 2005, 109, 347–355.
9] V. C. Papadimitriou, D. K. Papanastasiou, V. G. Stefanopoulos, A. M.
Zaras, Y. G. Lazarou, P. Papagiannakopoulos, J. Phys. Chem. A 2007, 111,
[34] M. J. Elrod, J. Chem. Educ. 1999, 76, 1702–1705.
[35] S. Pinnock, M. D. Hurley, K. P. Shine, T. J. Wallington; T. J. Smyth, J. Geo-
phys. Res. [Atmos.] 1995, 100, 23227–23238; T. J. Smyth, J. Geophys. Res.
[Atmos.] 1995, 100, 23227–23238.
[
1
1608–11617.
[36] K. S. Law, W. T. Sturges (lead authors) in Scientific Assessment of Ozone
Depletion: 2006, World Meteorological Organization, Geneva, 2007,
chap. 2.
[37] E. J. K. Nilsson, O. J. Nielsen, M. S. Johnson, M. D. Hurley, T. J. Wallington,
Chem. Phys. Lett. 2009, 473, 233–237.
[38] M. Baasandorj, G. Knight, V. C. Papadimitriou, R. K. Talukdar, A. R. Ravish-
ankara, J. B. Burkholder, J. Phys. Chem. A 2010, 114, 4619–4633.
[39] M. D. Hurley, T. J. Wallington, M. P. Sulbaek Andersen, D. A. Ellis, J. W.
Martin, S. A. Mabury, J. Phys. Chem. A 2004, 108, 1973–1979.
[40] K. Tokuhashi, H. Nagai, A. Takahashi, M. Kaise, S. Kondo, A. Sekiya, M. Ta-
kahashi, Y. Gotoh, A. Suga, J. Phys. Chem. A 1999, 103, 2664–2672.
[41] B. Rajakumar, R. W. Portmann, J. B. Burkholder, A. R. Ravishankara, J.
Phys. Chem. A 2006, 110, 6724–6731.
[
[
[
[
[
[
10] J. Albaladejo, B. Ballesteros, E. Jimꢀnez, P. Martꢁn, E. Martꢁnez, Atmos. En-
viron. 2002, 36, 3231–3239.
11] E. Jimꢀnez, B. Ballesteros, E. Martꢁnez, J. Albaladejo, Environ. Sci. Technol.
2
005, 39, 814–820.
12] E. Jimꢀnez, B. Lanza, M. AntiÇolo, J. Albaladejo, Atmos. Environ. 2009,
3, 4043–4049.
4
13] B. Ballesteros, A. Garzꢅn, E. Jimꢀnez, A. Notario, J. Albaladejo, Phys.
Chem. Chem. Phys. 2007, 9, 1210–1218.
14] B. Ballesteros, A. A. Ceacero-Vega, A. Garzꢅn, E. Jimꢀnez, J. Albaladejo, J.
Photochem. Photobiol. A 2009, 208, 186–194.
15] R. Atkinson, D. L. Baulch, R. A. Cox, J. N. Crowley, R. F. Hampson, R. G.
Hynes, M. E. Jenkin, M. J. Rossi, J. Troe, T. J. Wallington, Atmos. Chem.
Phys. 2004, 4, 1461–1738.
[16] E. Jimꢀnez, B. Lanza, A. Garzꢅn, B. Ballesteros, J. Albaladejo, J. Phys.
Chem. A 2005, 109, 10903–10909.
Received: May 7, 2010
[17] B. Lanza, E. Jimꢀnez, B. Ballesteros, J. Albaladejo, Chem. Phys. Lett. 2008,
Revised: June 21, 2010
454, 184–189.
Published online on September 17, 2010
ChemPhysChem 2010, 11, 4079 – 4087
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemphyschem.org
4087