Mechanism of photo-oxidation of heptafluorobutyric anhydride in the presence of NO2. Synthesis and characterization of heptafluoropropyl peroxynitrate, CF3CF2CF2OONO2

Article abstract of DOI:10.1021/jp307172g

The photolysis of heptafluorobutyric anhydride at 254 nm in the presence of NO₂ and O₂ has been studied. It leads to the formation of CF₃CF₂CF₂OONO₂, CF ₃CF₂OONO₂, and CF₂O as the only fluorine-containing carbonaceous products. The formation of the new heptafluoropropyl peroxynitrate (HFPN, CF₃CF₂CF ₂OONO₂), as one of the main products, is a consequence of the formation of CF₃CF₂CF₂OO radicals followed by the reaction with NO₂. To characterize HFPN, the UV absorption cross sections and their temperature dependence between 245 and 300 K have been measured over the wavelength range 200-300 nm as well as the infrared absorption cross sections. Kinetic parameters for its thermal decomposition are also presented in the temperature range between 281 and 300 K. The Rice-Ramsperger-Kassel-Marcus calculation reveals that the rate coefficient for the thermal decomposition at 285 K is almost independent of total pressure. The mechanism for the decomposition of CF₃CF₂CF ₂OONO₂ in the presence of NO was adjusted by a kinetic model, which enabled the calculation of important rate coefficients.

Full text of DOI:10.1021/jp307172g

Article
pubs.acs.org/JPCA
Mechanism of Photo-Oxidation of Heptafluorobutyric Anhydride
in the Presence of NO2. Synthesis and Characterization of
Heptafluoropropyl Peroxynitrate, CF3CF2CF2OONO2
Adriana G. Bossolasco, Fabio E. Malanca,* Maxi A. Burgos Paci, and Gustavo A. Arguello
̈
Instituto de Investigaciones en Físico Química de Cor
Facultad de Ciencias Químicas, Universidad Nacional de Cor
́
doba (INFIQC) CONICET-UNC, Departamento de Físico Química,
doba, Ciudad Universitaria, X5000HUA Cordoba, Argentina
́
́
S
* Supporting Information
ABSTRACT: The photolysis of heptafluorobutyric anhydride
at 254 nm in the presence of NO2 and O2 has been studied. It
leads to the formation of CF3CF2CF2OONO2, CF3CF2OONO2,
and CF2O as the only fluorine-containing carbonaceous prod-
ucts. The formation of the new heptafluoropropyl peroxynitrate
(HFPN, CF3CF2CF2OONO2), as one of the main products, is
a consequence of the formation of CF3CF2CF2OO• radicals
followed by the reaction with NO2. To characterize HFPN, the
UV absorption cross sections and their temperature dependence between 245 and 300 K have been measured over the wavelength
range 200−300 nm as well as the infrared absorption cross sections. Kinetic parameters for its thermal decomposition are also
presented in the temperature range between 281 and 300 K. The Rice−Ramsperger−Kassel−Marcus calculation reveals that the rate
coefficient for the thermal decomposition at 285 K is almost independent of total pressure. The mechanism for the decomposition of
CF3CF2CF2OONO2 in the presence of NO was adjusted by a kinetic model, which enabled the calculation of important rate
coefficients.
Thus, heptafluoropropyl peroxynitrate (HFPN, CF3CF2CF2-
OONO2) could act as a reservoir species in the atmosphere,
sequestering ROO• radicals and NO2 and converting them into
less reactive, electron paired, thermally stable molecules which at
the same time could be transported from the source production
site to remote places.
INTRODUCTION
■
Perfluoroperoxy radicals (CxF2x+1OO•) are formed in the degrada-
tion of chlorofluorocarbons (CFCs), hydrofluorocarbons
(HFCs), and hydrofluoroethers (HFEs), a series of compounds
used as refrigerants, blowing and cleaning agents, emulsifiers,
and solvents in general. In particular, perfluoropropyl peroxy
radicals could be formed in the atmospheric oxidation of mole-
cules containing CF3CF2CF2 fragments such as HFC-227ca
(CF3CF2CF2H)1 and n-perfluoropropyl formate (n-CF3CF2CF2-
OC(O)H).
Reaction 2 has been studied by Giessing et.al,1 who measured
the rate coefficient for the reaction between CF3CF2CF2OO•
and NO2 through pulse radiolysis and determined a value of
k2 = (7.6 2.4) × 10−12 cm3 molecule−1 s−1; however, no effort
was made to evaluate the products of the reaction. In this work
we went further and characterized the heptafluoropropyl
peroxynitrate (HFPN, CF3CF2CF2OONO2) from infrared and
UV spectra. Besides, we measured the temperature dependence
of the ultraviolet absorption cross sections and the thermal
stability as a function of both pressure and temperature. The
study of the formation mechanism from the photolysis of
heptafluorobutyl anhydride (HFBA) in the presence of NO2 and
O2 was also performed. HFBA was only used as a clean source to
provide the C3F7• radicals.
As for other peroxy radicals (ROO•), the atmospheric fate of
CF3CF2CF2OO• radicals is the reactions with NO, NO2, HO2, as
1,2
•
́
well as other peroxy radicals (ROO ).
CFCF CF OO• + NO → CFCF CF O• + NO2
(1a)
(1b)
3
2
2
3
2
2
CFCF CF OO• + NO + M → CFCF CF ONO2 + M
3
2
2
3
2
2
CFCF CF OO• + NO2 + M → CFCF CF OONO2 + M
3
2
2
3
2
2
(2)
•
CFCF CF OO• + HO2 → products
(3)
3
2
2
EXPERIMENTAL SECTION
■
CFCF CF OO• + ROO• → products
(4)
Photolysis of (CF3CF2CF2C(O))2O in the Presence of
NO2/O2: Isolation and Recognition of CF3CF2CF2OONO2.
3
2
2
Among the preceding reactions, reaction 2 leads to the forma-
tion of CF3CF2CF2OONO2 that should, in turn, decompose via
reaction −2, as any known peroxynitrate
Received: July 19, 2012
Revised: September 4, 2012
Published: September 8, 2012
CFCF CF OONO2 → CFCF CF OO• + NO2
(−2)
3
2
2
3
2
2
© 2012 American Chemical Society
9904
dx.doi.org/10.1021/jp307172g | J. Phys. Chem. A 2012, 116, 9904−9910

The Journal of Physical Chemistry A
Article
Photolysis (λ= 254 nm) of mixtures of (C3F7C(O))2O/NO2/O2
(0.7−1.0 mbar/0.3−0.5 mbar/610 mbar) were performed at
room temperature (294 K) in a standard quartz infrared cell
(23.0 cm path length), located in the optical path of a FTIR
spectrophotometer which allowed us to monitor the appearance of
products and the disappearance of reactants as a function of time.
The procedure used to synthesize CF3CF2CF2OONO2 is
similar to that used to obtain the trifluoromethyl peroxynitrate by
Chiappero et al.3 The synthesis was conducted using a 10-L glass
flask and keeping the temperature at 283 K. Typical pressures
of (C3F7C(O))2O, NO2, and O2 used were: 4.0, 2.5, and 1000
mbar, respectively. The progress was monitored every 30 min
through infrared spectroscopy (transferring approximately 1% of
the mixture to the IR cell each time), and photolysis was stopped
when NO2 concentration had decreased to a quarter of its initial
value (approximately 2 h of irradiation). Scheme 1 shows the
photochemical route after HFBA photolysis. The final resultant
mixture was collected by slowly passing it through three traps
kept at 87 K in order to remove excess O2 and CO formed. It was
Thermal Decomposition of CF3CF2CF2OONO2. Thermal
stability was determined by monitoring, through FTIR, the
temporal variation of the band at 997 cm−1 as a function of
temperature and total pressure. Typical runs were carried out
using 0.6−1.5 mbar of HFPN, 2.0 mbar of NO (that assures the
capture of every peroxy radical formed), and N2 up to the total
pressure. Temperature dependence was studied between 281.0
and 300 K at 10 mbar of total pressure, while the pressure
dependence study was performed at 284.2 K in the range from
10.0 to 600 mbar. Data at each temperature and pressure were
analyzed according to a first-order rate law, maintaining a low
conversion percentage. For each temperature, the decay was
fitted by a linear regression and the value associated with the rate
coefficient was plotted in the Arrhenius form.
Computational Details. Density functional theory has been
used to evaluate the ground-state geometric parameters,
vibrational frequencies, and the relative conformers populations
of HFPN. The geometries, harmonic vibrational frequencies, and
zero-point energies of the conformers were calculated using the
hybrid density functional B3LYP with the 6-311+G* basis set.
Density functional theory (DFT) methods take into account
electron correlation energy to a partial extent,5 and the particular
method selected (though with a smaller basis set) has been
extensively applied to the determination of geometric parameters
of related oxygenated fluorocarbon compounds, yielding results
that have been contrasted with gas electron diffraction experi-
ments.6−8 B3LYP/6-311+G* was used also for the determination
of the geometric and vibrorotational parameters of the transition
state for the thermal decomposition of HFPN, needed in the
evaluation of the rate coefficients within Rice−Ramsperger−
Kassel−Marcus (RRKM) formalism. These calculations were
run using the G09 program package9 in conjunction with
GaussView 5.0.10
Scheme 1. Reaction Mechanism for HFBA Photolysis in the
a
Presence of NO2/O2
a
Products observed and quantified are highlighted in bold.
Thermal rate coefficients and their pressure dependence for
the unimolecular decomposition of HFPN were evaluated using
standard RRKM theory implemented in the UNIMOL program.11
The KINTECUS program package12 was used to run kinetics
simulations. This program allows the fitting of the time variations
in the concentration of reagents and products while optimizing
selected rate coefficients according to the mechanism proposed.
then distilled between 188 and 153 K to eliminate the CF2O and
CO2 formed. Further distillation between 213 and 173 K allows
the separation of the HFBA that remains in the less volatile
fraction. The more volatile fraction contains pure peroxynitrate.
From this sample, reference IR spectra were taken.
To avoid HFPN decomposition, syntheses were carried out in
the presence of sufficient NO2 concentration to favor shifting the
equilibrium
RESULTS AND DISCUSSION
■
Photolysis of HFBA in the Presence of NO2 and O2.
Figure 1 shows a set of infrared spectra from the photolysis of
mixtures of (C3F7C(O))2O, NO2 and O2 together with selected
reference spectra. The first trace corresponds to the starting
mixture. Appropriate subtraction of reactants to the trace
obtained at 24 min of photolysis leads to the spectrum of the
products (trace “A”), which shows the appearance of peaks
corresponding to carbonyl fluoride and others (around 1760 and
790 cm−1) compatible with fluorinated peroxynitrates
(ROONO2). The subtraction of carbonyl fluoride leads to a
trace, which is the sum of the contribution of various pero-
xynitrates, mainly heptafluoropropyl peroxynitrate (HFPN,
CF3CF2CF2OONO2) as can be seen by comparison of trace
“C” with the spectrum of the pure compound “D”. Further
subtraction reveals the presence of another perfluorinated
peroxynitrate, CF3CF2OONO2, whose reference spectrum13 is
shown in trace “F”.
CFCF CF OONO2 ⇆ CFCF CF OO• + NO2
(−2, 2)
3
2
2
3
2
2
toward the reactant and keeping the temperature at about 10 °C
since its stability increases.
UV and IR Spectroscopy of CF3CF2CF2OONO2. Ultra-
violet absorption spectrum of HFPN has been measured over the
wavelength range 200−340 nm from 245 to 300 K using the
experimental design described elsewhere.4 Infrared spectra were
recorded at room temperature with a resolution of 2 cm−1
averaging 16 scans in the range of 4000−400 cm−1. The pressures
ranged from 2.0 to 6.0 mbar for the UV and from 0.7 to 1.5 mbar
for the IR spectra, respectively.
Absorption cross sections (σ) were determined according to
σ (cm2 molecule−1
= 31.79 × 10−20(mbar cm3 molecule−1 K−1)TA(pd)−1
)
These products could be explained taking into account that the
rupture of the precursor molecule leads to the formation of CO,
(5)
•
where T is the temperature (K), A the absorbance, p the pressure
(mbar), and d the optical path (cm).
CO2, and perfluorinated radicals (CF3CF2CF2 ) in a similar way
to trifluoroacetic (TFAA)14 and pentafluoro propionic anhydride
9905
dx.doi.org/10.1021/jp307172g | J. Phys. Chem. A 2012, 116, 9904−9910

The Journal of Physical Chemistry A
Article
of the skeleton FCF2CF2CF2OONO2, which
can be reduced to four because φ0(FCF2CF2CF2) always
adopts the anti configuration. These four dihedrals are:
φ1(F3CCF2CF2O), the relative position of the CF3
group with respect to the CO bond; φ2(F3CCF2CF2
OO), the relative position of the C2F5 fragment with respect to
the peroxy bond; φ3(C3F7OON), the relative position of
the carbonaceous chain with respect to the ONO2 bond and
φ4(C3F7OONOO), the relative position of the NO2
group with respect to the peroxy bond. Enantiomers are not
taken into account because there is no energy difference between
them. Dihedrals φ1 and φ2 could adopt sin (0°), anti (180°), or
gauche (∼60 or ∼120°) configurations. Experimental and
theoretical studies of similar peroxynitrates have shown that
the XOOY and OONO dihedrals (φ3 and φ4 in
our case) adopt gauche and syn configurations, respectively.16,17
Within this analysis the minima of the configurational space can
be reduced to the different possibilities of φ1 and φ2. A relaxed
potential energy surface (PES) scan for the joint variation of φ1
and φ2 from 0 to 180° was run, and the result is shown as a
contour map in Figure 3. The contour map was constructed by
Figure 1. Photooxidation of PFBA in the presence of NO2.
(PFPA).15 As it is shown in Scheme 1, in general terms, per-
fluorinated radicals react with oxygen leading to the formation of
peroxy radicals (CF3CF2CF2OO• or CF3CF2OO•), which in turn
react with NO2 leading to the formation of the corresponding
peroxynitrates (CF3CF2CF2OONO2 or CF3CF2OONO2, respec-
tively). HFPN decomposes thermally to reform CF3CF2CF2OO•
radicals, which could recombine or react with NO (formed by the
photolysis of NO2) to form alkoxy radicals that ultimately lead to
formation of CF2O.
Temporal variation of reactants and products are shown in
Figure 2a. As expected, the concentration of CF2O and per-
oxynitrates increases as time elapses. According to the mechanism
proposed (Scheme 1) and the carbon balance derived from it, the
rupture of the HFBA molecule leads to the following relationship
2Δp(CF CF CF C(O)) O = pCF CF CF OONO + (pCF O = pCF CF OONO )
3
2
2
2
3
2
2
2
2
3
2
2
(6)
where p corresponds to the partial pressure (mbar) of each reactant
or product.
Figure 2b shows the concentration of the products formed
as a function of HFBA disappearance. The relative percentage
values for the formation of CF3CF2CF2OONO2, CF2O, and
CF3CF2OONO2 give 130 9, 68 5, and 65 5, respectively.
Taking into account the preceding relationship, the percentage of
products observed explain about 98% of the disappearance of
HFBA, in agreement with the non-observation of other
fluorinated products (e.g., CF3OONO2).
Figure 3. Relaxed PES scan for the joint variation of φ1 and φ2 from 0 to
360°.
appropriate reflections of the scanned points to show variations
of 360° for each angle. Four different minima are found in the
PES being the one with φ1 = 180° and φ2 = 56° the global
minimum. These four minima are designated from 1 to 4 in
ascending order of their relative energy. A qualitative inspection
of the PES shows that interconversion between the minima could
DFT Calculations. The conformational space of HFPN can
be described taking into account the five main dihedral angles
Figure 2. (a) Temporal variation of reactants and products in the photolysis of a mixture containing HFBA, NO2, and O2. Symbols correspond to
■
□
●
○
(C3F7C(O))2O ( ), NO2 ( ), CF3CF2CF2OONO2 (Δ), CF2O ( ), and C2F5OONO2 ( ). (b) Formation of products CF3CF2CF2OONO2 (Δ),
□
○
CF2O ( ), and C2F5OONO2 ( ) as a function of PFBA disappearance.
9906
dx.doi.org/10.1021/jp307172g | J. Phys. Chem. A 2012, 116, 9904−9910

The Journal of Physical Chemistry A
Article
involve trajectories needing less than 4 kcal mol−1 and that the
relative energies are less than 1.8 kcal mol−1 with respect to 1
(Figure I in the Supporting Information depicts the structure of1).
Table 1 in the Supporting Information shows the B3LYP/
DFT value of 1.560 Å is really long and it involves an uncertainty
of ∼2%, it is very interesting that it anticipates an even longer
N−O distance for HFPN than for CF3OONO2. An experimental
GED study for this molecule is really desirable.
UV and IR Absorption Cross Sections. Figure 4 shows the
experimental infrared absorption cross sections in the range of
Table 1. Calculated Geometric Parameters of the Most Stable
Conformer
geometric parameters
C3F7OONO2
CF3OONO2
GED
1.322
distance (Å)
1.343
CF
CF (mean)
CC (mean)
C4O5
1.564
1.386
1.411
1.560
1.182
CO
OO
ON
NO
1.378
1.414
1.523
1.187
O5O6
O6N7
N7O(8.9)
angles (deg)
FCF
108.9
116.5
107.4
109.4
108.9
116.2
108.3
135.5
−56.7
179.9
104.7
178.1
0.9
FCF
108.8
C2C3C4
C5C8O11
C4O5O6
COO
OON
107.7
108.4
Figure 4. Infrared absorption cross sections for CxF2x+1OONO2, x = 1−3.
The bottom trace (gray line) corresponds to the simulated
CF3CF2CF2OONO2 spectrum.
O5O6N7
O6N7O9
O6N7O8
O8N7O9
ONO
135.2
2000 to 600 cm−1, for CF3OONO2,18 CF3CF2OONO2,13 and
CF3CF2CF2OONO2 (first, second, and third traces) and the
theoretical unscaled spectrum calculated for CF3CF2CF2OONO2
(bottom trace). As can be seen, the HFPN spectrum is in
agreement with the general trend shown by many peroxynitrates.
However, there is a distinguishable signature present in the
HFPN at 996 cm−1 that allows its differentiation from the other
fluorinated peroxynitrates, and that was used for its identifica-
tion and quantification. The infrared bands (in units of cm−1),
the corresponding absorbance cross sections (σ × 1018, cm2
molecule−1), and their assignment for the main peaks are 1764
(3.1 0.1) νas (NO2), 1302 (1.58 0.06) νs (NO2), 1246 (5.7
0.1) νas (CF3), 1125 (1.28 0.05) νas (C−F), 996 (1.43 0.05)
νs (O−O), and 790 (1.19 0.05) def. δNO2. As stated before,
four different rotamers should be in equilibrium in a gas-phase
sample of HFPN at room temperature, according to the DFT
calculation. Because of this, the theoretical spectrum shown in the
bottom of Figure 4 was synthesized from the linear combination
of the theoretical spectrum of each rotamer multiplied by the
corresponding population. Every vibrational transition was
modeled with a Lorentzian function with 4 cm−1 of fwhh. The
comparison between the experimental and theoretical spectra
shows a good correlation for the featured peaks.
Figure 5 shows the UV spectrum of CF3CF2CF2OONO2
between 200 and 300 nm and its temperature dependence in
the range from 245 to 300 K; meanwhile Table 2 of Supporting
Information summarizes the absorption cross sections at
different temperatures. As can be seen the UV absorption cross
sections increase with the temperature. This increment is con-
sistent with an increased population of the vibrational and
rotational levels of the ground electronic state of the molecule at
higher temperatures.20
The temperature dependence is more pronounced toward the
longer wavelength tail as shown in the representation of log(σ) vs
T, which leads to linear fits according to log(σ) = BT + log(σ0),
where σ and σ0 are the cross sections (cm2 molecule−1) at T and
0 K, respectively. The intercepts log(σ0) and slopes (B) obtained
C2C3C4O5
C3C4O5O6
C4O5O6N7
O5O6N7O8
O5O6N7O9
φ(COON)
φ(OONO3)
105.1
178.3
6-311+G* absolute energies together with the relative energies
and room temperature populations for the rotamers. Relative
populations were calculated according to the formula
Nx/N = Q /Q exp(−ΔE/kT)
(7)
1
1
x
where Qx is the total partition function, taken from G09, of
rotamer x. It is noteworthy that the calculation anticipates a blend
of the four rotamers at room temperature.
Table 1 shows the geometrical parameters for this rotamer
together with the experimental GED results of the similar
peroxynitrate, CF3OONO2.18 The agreement between the
calculated parameters with those in similar compounds is
remarkable for the level of theory used. The most interesting
parameters in peroxynitrates are the XO−O−N dihedral and
the O−N distance. Previous studies on peroxy compounds
show that the dihedral is around 120° or 90° for two sp3 or sp2
substitutes, respectively, and 105° for one sp3/sp2 substitute, as is
the case for peroxynitrates.16,18 The calculated value of this
parameter is 104.7° for HFPN, and it agrees well with this
tendency. The N−O distance is extremely long in peroxynitrates,
and it has been observed that it depends a lot on the electro-
negativity of the group attached to the −ONO2 fragment. Values
of 1.507 and 1.523 Å are reported for FONO2 and CF3OONO2,
respectively,18,19 suggesting that the CF3O group is more
electronegative than F. In HFPN a value of 1.560 Å is found,
and it can be thought that the C3F7O group is still more electro-
negative than CF3O. A DFT calculation for CF3CF2OONO2
with the same basis set gives 1.560 Å for the O−N distance,
showing that addition of more CF3 groups to the carbon chain
does not influence the O−N distance. Although the absolute
9907
dx.doi.org/10.1021/jp307172g | J. Phys. Chem. A 2012, 116, 9904−9910

The Journal of Physical Chemistry A
Article
Figure 6. Dependence of the rate coefficient with pressure for
CxF2x+1OONO2 obtained from RRKM calculations (x = 1, dashed
line; x = 2 dash−dotted line; x = 3, solid line) and from experimental
data (x = 2, triangles; x = 3, circles).The insert shows the Arrhenius plot
for the thermal decomposition of CF3CF2CF2OONO2 at 10.0 mbar of
total pressure between 284 and 300 K.
Figure 5. UV spectra of CF3CF2CF2OONO2 at different temperatures:
245 K, dashed black; 253 K, dashed gray; 268 K, dotted; 283 K, solid
gray; 300 K, solid black.
by linear regression analysis for the wavelength range 200−
285 nm are given in Table 3 of the Supporting Information As
can be seen, B increases with wavelength as a consequence of the
larger temperature dependence at longer wavelengths.
Thermal Stability of CF3CF2CF2OONO2: Kinetic Model-
ing. The thermal stability was studied between 281 and 300 K at
10.0 mbar of total pressure. The rate coefficient values for the
peroxynitrate decomposition (reaction −2) were corrected con-
sidering that, as time elapses, the NO added leads to an increase
in NO2 formation, slightly shifting the equilibrium toward the
formation of CF3CF2CF2OONO2 via reaction 2. The equation
used to correct the rate coefficient has been reported elsewhere.21
ranges between 2.1 and 2.5 × 10−3 s−1. For x = 2, the pressure
dependence is stronger and for x = 1 even stronger. A good
agreement between the experimental and calculated data can be
observed for pressures higher than 1 mbar, while for pressures
lower than 1 mbar, the dependence with the length of the carbon
chain is reversed.
Arrhenius parameters derived from a plot of ln k vs (1/T)
(inset of Figure 6), at 10.0 mbar total pressure, are: activation
energy (Ea = 84.9 kJ/mol) and pre-exponential factor (A = 1.0 ×
1013). The comparison (Ea and A) for the series CxF2x+1OONO2
(x = 1, 2, 3) (90.8 and 1.05 × 1014),24 (87.7 and 3.8 × 1013),13
(84.9 and 1.0 × 1013) kJ/mol, respectively, shows that Ea
decreases as the length of the perfluorinated alkyl group
increases, thus, the longer the chain, the shorter the lifetime,
consistent with the trends of the calculated and experimental
values (see Figure 6) for pressures higher than 1 mbar.
Table 2 shows the mechanism proposed for the thermal
decomposition, according to the products observed. The
values for the rate coefficients were obtained either from the
literature or adjusted with the KINTECUS model. The most
relevant to note is that (a) the value for kIV is in agreement
with the value suggested by Giessing et. al (>1.5 × 105 s−1)
and (b) the values for kVI, which resulted from the fit of the
kinetic model to the experimental points and are reported for
the first time in this work, are consistent with the reported
values for similar reactions.25 The good fit obtained is presented
in Figure 7.
⎛
⎞
k2[NO2]
k−2 = kobs 1 +
⎜
⎟
⎠
k [NO]
⎝
(8)
1
where kobs corresponds to the uncorrected experimental rate
of disappearance of CF3CF2CF2OONO2. The ratio k2/k1 was
determined using the values for the rate coefficients determined
by Giessing et al.1
Master equation calculations within the RRKM formalism
were run to evaluate the pressure dependence of the rate
coefficient. The model used involves the following approx-
imations. Intermolecular collisions of the peroxynitrate with the
buffer gas (N2 in all cases) take place under a Lennard-Jones
potential. The critical parameters were estimated by the methods
in chapter 6 of Gilbert and Smith.22 The values obtained are σLJ =
6.45 Å and σLJ = 3.9 Å for HFPN and N2, respectively. All internal
modes and moment of inertia of the reagent and TS were taken
from the DFT calculations. All internal modes were treated as
active vibrations and multiplied by the scale factor 1.0013 as
suggested by Scott and Radom.23 The external rotation of lowest
moment of inertia was treated as fully active and the two
remaining as adiabatic modes. Rate coefficients are very sensitive
to the critical energy, and there is a difference of more than
1 kcal mol−1 between the calculated and experimental values.
Although this is acceptable for the evaluation of a bond fission
reaction at the B3LYP/6-311+G* level, for modeling kinetics
parameters more precision is needed, and because of this we
decided to use the experimental value of the activation energy. A
Morse potential was fitted to evaluate the ROO···NO2
interaction using the experimental activation energy and the
proper stretching frequency.
Atmospheric Implications. Photochemistry of HFBA in
the presence of NO2 and O2 leads to the formation of a new
peroxynitrate, HFPN, a compound whose stability is similar to
other fluorinated alkyl peroxynitrates CxF2x+1OONO2 (x = 1, 2)
and that follows the expected trends.
The loss of peroxynitrates in the atmosphere occurs
through different processes, namely thermal decomposition,
photolysis, and reaction with OH radicals, whose relative
importance depend on the region of the atmosphere and the
very nature of the peroxynitrate.26 In particular, HFPN is
unstable at temperatures of the free troposphere, and its
thermal lifetime should be of the order of minutes. At altitudes
higher than 5 km, where the temperatures are lower than
250 K, its stability increases to achieve a thermal lifetime of
one year at the tropopause; beyond 20 km it should start
decreasing. However, between 10 and 30 km, the thermal
lifetime should be higher than 3 months and considering the
atmospheric horizontal mixing at these altitudes HFPN would
be transported for long distances.
In turn, Figure 6 shows the experimental dependence of the
rate coefficient with pressure (12 to 220 mbar at 284 K) and
its comparison with the results obtained from RRKM for
CxF2x+1OONO2 (x = 1, 2, 3) using the UNIMOL program
suite.11 As can be seen, the rate coefficient for x = 3 (i.e., HFPN)
shows a slight dependence with the pressure, and its value
9908
dx.doi.org/10.1021/jp307172g | J. Phys. Chem. A 2012, 116, 9904−9910

The Journal of Physical Chemistry A
Article
Table 2. Reaction Mechanism of HFPN Decomposition in the Presence of NO and Rate Coefficients Used in the Kinetic Model
(KINTECUS)
reaction
k
reference
this work
CFCF CF OONO2 → CFCF CF OO• + NO2
I
2.1 × 10−3
3
2
2
3
2
2
CFCF CF OO• + NO2 → CFCF CF OONO2
II
7.6 × 10−12
9.0 × 10−12
1.0 × 106
Wallington, 1996
Wallington, 1996
this work
3
2
2
3
2
2
CFCF CF OO• + NO → CFCF CF O• + NO2
III
IV
V
3
2
2
3
2
2
CFCF CF O• → CFCF • + CF O
3
2
2
3
2
2
1.5 × 10−12
1.8 × 10−11
1.0 × 10−1
this work
CFCF • + NO → CFCF NO
3
2
3
2
CFCF • + NO2 → CFC(O)F + FNO
VI
VII
this work
3
2
3
FNO + wall → HF + NO2
REFERENCES
■
(1) Giessing, A. M. B.; Feilberg, A.; Møgelberg, T. E.; Sehested, J.;
Bilde, M.; Wallington, T. J.; Nielsen, O. J. J. Phys. Chem. 1996, 100,
6572−6579.
(2) Wallington, T. J.; Schneider, W. F.; Worsnop, D. R.; Nielsen, O. J.;
Sehested, J.; Debruyn, W. J.; Shorter, J. A. Environ. Sci. Technol. 1994, 28,
320−326.
(3) Chiappero, M. S.; Burgos Paci, M. A.; Arguello, G. A.; Wallington,
T. J. Inorg. Chem. 2004, 43, 2714−2716.
(4) Malanca, F. E.; Chiappero, M. S.; Arguello, G. A.; Wallington, T. J.
̈
̈
Atmos. Environ. 2005, 39, 5051−5057.
Figure 7. Temporal variation of reactants and products in the thermal
(5) Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785−789.
■
⧫
▲
▽
)
decomposition of HFPN: ( ) CF3CF2CF2OONO2, ( ) CF2O, (
́
(6) Hnyk, D.; Machacek, J.; Arguello, G. A.; Willner, H.; Oberhammer,
̈
●
CF3COF, ( ) NO2, and ( ) CF3CF2NO. Lines represent the best fit
H. J. Phys. Chem. A 2003, 107, 847−851.
given by the KINTECUS model.
(7) Mayer, F.; Oberhammer, H.; Berkei, M.; Pernice, H.; Willner, H.;
Bierbrauer, K.; Burgos Paci, M.; Arguello, G. A. Inorg. Chem. 2004, 43,
̈
8162−8168.
The UV absorption cross sections at wavelengths shorter than
300 nm are relevant for the photochemical lifetime estimation.
Taking into account the altitude profile of UV wavelengths and
that HFPN starts absorbing mainly at wavelengths shorter than
300 nm, its photochemical rupture begins to take importance
from 35 km up, reaching values about minutes at altitudes higher
than 40 km. Therefore the thermal decomposition prevails over
the photochemical rupture from the surface until about 40 km.
Comparison with the lifetime of the most abundant per-
oxynitrate measured in the atmosphere, PAN −CH3C(O)-
OONO2- (from minutes in free troposphere to 20 days at
tropopause)27 reveals that this new peroxynitrate could be
formed in the atmosphere and act as a reservoir of NO2 and
peroxy radicals.
(8) Della Vedova, C. O.; Boese, R.; Willner, H.; Oberhammer, H. J.
Phys. Chem. A 2004, 108, 861−865.
(9) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,
M. A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A.;
Stratmann, R. E.; Burant, J. C.; Dapprich, S.; Millam, J. M.; Daniels, A.
D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.; Barone, V.; Cossi,
M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo, C.; Clifford, S.;
Ochterski, J.; Petersson, G. A.; Ayala, P. Y.; Cui, Q.; Morokuma, K.;
Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.;
Cioslowsk, J.; Ortiz, J. V.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz,
P.; Komaromi, I.; Gomperts, R.; Martin, R. L.; Fox, D. J.; Keith, T.;
Laham, Al M. A.; Peng, C. Y.; Nanayakkara, A.; Gonzalez, C.;
Challacombe, M.; Gill, P. M. W.; Johnson, B. G.; Chen, W.; Wong,
M. W.; Andres, J. L.; Gordon, Head M.; Replogle, E. S.; Pople, J. A.
Gaussian 98, revision A.7; Gaussian, Inc.: Pittsburgh, PA, 1998.
(10) Dennington, R., Keith, T. Millam, J. GaussView, version 5;
Semichem Inc.: Shawnee Mission, KS, 2009.
ASSOCIATED CONTENT
* Supporting Information
■
S
(11) Gilbert, R. G.; Smith, S. C.; Jordan, M. J. T. UNIMOL program
suite; School of Chemistry, Sydney University, New South Wales,
Australia, 2006.
The calculated structure for the most stable conformer of
C3F7OONO2 and a table describing the energies and populations
of the different conformers as well as two other tables dealing
with the UV absorption cross sections. This information is
available free of charge via the Internet at http://pubs.acs.org.
(12) Ianni, J. C. KINTECUS, Windows version 2004, 3.7.
(13) Bossolasco, A. G.; Malanca, F. E; Arguello, G. A. J. Photochem.
̈
Photobiol. A: Chem. 2012, 231, 45−50.
(14) Chamberlain, G. A.; Whittle, E. J. Chem. Soc., Faraday Trans 1
1972, 68, 88−95.
AUTHOR INFORMATION
Corresponding Author
*E-mail: fmalanca@fcq.unc.edu.ar.
■
(15) Chamberlain, G. A.; Whittle, E. J. Chem. Soc., Faraday Trans 1
1972, 68, 96−103.
(16) Kopitzky, R.; Beuleke, M.; Balzer, G.; Willner, H. Inorg. Chem.
1997, 36, 1994−1997.
Notes
The authors declare no competing financial interest.
(17) Scheffler, D.; Schaper, I.; Willner, H.; Mack, H. G.; Oberhammer,
H. Inorg. Chem. 1997, 36, 339−344.
ACKNOWLEDGMENTS
■
(18) Kopitzky, R.; Willner, H.; Mack, H. G.; Pfeiffer, A.; Oberhammer,
H. Inorg. Chem. 1998, 37, 6208−6213.
Financial support from SECYT-UNC, ANPCyT, and CONI-
CET is gratefully acknowledged. A.B. thanks CONICET for a
PhD fellowship.
(19) Casper, B.; Dixon, D. A.; Mack, H.-G.; Ulic, S. E.; Willner, H.;
Oberhammer, H. J. Am. Chem. Soc. 1994, 116, 8317−8321.
9909
dx.doi.org/10.1021/jp307172g | J. Phys. Chem. A 2012, 116, 9904−9910

The Journal of Physical Chemistry A
Article
(20) Okabe, H. Photochemistry of small molecules; Wiley: New York,
1978.
(21) Christensen, L. K.; Wallington, T. J.; Guschin, A.; Hurley, M. D. J.
Phys. Chem. A 1999, 103, 4202−4208.
(22) Gilbert, G. R.; Smith, S. C. Theory of Unimolecular and
Recombination Reactions; Blackwell Scientific Publications, Oxford,
UK, 1990.
(23) Scott, A. P.; Radom, L. J. Phys. Chem. 1996, 100, 16502−16513.
(24) Mayer-Figge, A.; Zabel, F.; Becker, K. H. J. Phys. Chem. 1996, 100,
6587−6593.
(25) Pagsberg, P.; Jodkowski, J. T.; Ratajczak, E.; Sillesen, A. Chem.
Phys. Lett. 1998, 286, 138−144.
(26) Roberts, J. M. PAN and Related Compounds. In Volatile Organic
Compounds in the Atmosphere; Koppmann, R., Ed.; Blackwell Publishing
Ltd, Oxford, UK., 2007.
(27) Talukdar, R. K.; Burkholder, J. B.; Schmoltner, A. M.; Roberts, J.
M.; Wilson, R. R.; Ravishankara, A. R. J. Geophys. Res. 1995, 100, 14163−
14173.
9910
dx.doi.org/10.1021/jp307172g | J. Phys. Chem. A 2012, 116, 9904−9910