Reactions of CF3O- with atmospheric trace gases

Article abstract of DOI:10.1021/jp951928u

The rate coefficients and product yields for the reactions of CF₃O^- with ClONO₂, HNO₃, HCl, N₂O₅, SO₂, HI, and H₂O were measured at 295 K and ~～.4 Torr using the flowing afterglow technique. The reactions of CF₃O^- with HO₂NO₂ and H₂SO₄ were also studied qualitatively. CF₃O^- reacts rapidly with ClONO₂, HO₂NO₂, SO₂, and HCl by only fluoride transfer and with HNO₃, HI, and H₂SO₄ by both fluoride and proton transfer. CF₃O^- also reacts with HI to form IF₂^- and with N₂O₅ to produce NO₃^-. CF₃O^- undergoes a slow clustering reaction with H₂O and is transformed within water clusters into F^-·HF and F^-·(HF)₂. CF₃O^- is unreactive with CH₃NO₃, CH₃C(O)O₂NO₂, O₃, NO₂, CO₂, and O₂. These results demonstrate that CF₃O^- is an excellent candidate as a reagent ion for the selective detection of ClONO₂, HCl, and HNO₃ in the upper troposphere and stratosphere with a chemical ionization mass spectrometer. The observed reaction of CF₃O^- with H₂O within water clusters indicates that CF₃O^- will hydrolyze in aqueous solution to form F^-, HF, and CO₂. This provides insight into the mechanism for the heterogeneous loss of CF₃OH in the atmosphere.

Full text of DOI:10.1021/jp951928u

190
J. Phys. Chem. 1996, 100, 190-194
Reactions of CF3O- with Atmospheric Trace Gases
L. Gregory Huey,*,† Peter W. Villalta,† Edward J. Dunlea,† David R. Hanson,† and
Carleton J. Howard
NOAA, Aeronomy Laboratory, Boulder, Colorado 80303
ReceiVed: July 12, 1995; In Final Form: September 21, 1995X
The rate coefficients and product yields for the reactions of CF3O- with ClONO2, HNO3, HCl, N2O5, SO2,
HI, and H2O were measured at 295 K and ∼0.4 Torr using the flowing afterglow technique. The reactions
of CF3O- with HO2NO2 and H2SO4 were also studied qualitatively. CF3O- reacts rapidly with ClONO2,
HO2NO2, SO2, and HCl by only fluoride transfer and with HNO3, HI, and H2SO4 by both fluoride and proton
transfer. CF3O- also reacts with HI to form IF2- and with N2O5 to produce NO3 . CF3O- undergoes a slow
-
clustering reaction with H2O and is transformed within water clusters into F-‚HF and F-‚(HF)2. CF3O- is
unreactive with CH3NO3, CH3C(O)O2NO2, O3, NO2, CO2, and O2. These results demonstrate that CF3O- is
an excellent candidate as a reagent ion for the selective detection of ClONO2, HCl, and HNO3 in the upper
troposphere and stratosphere with a chemical ionization mass spectrometer. The observed reaction of CF3O-
with H2O within water clusters indicates that CF3O- will hydrolyze in aqueous solution to form F-, HF, and
CO2. This provides insight into the mechanism for the heterogeneous loss of CF3OH in the atmosphere.
-
Introduction
fluoride transfer properties similar to SF6 that is unreactive
toward relatively abundant atmospheric species such as CO2 or
O3. An ion that potentially meets these requirements is CF3O-.
The fluoride affinities of CF2O and SF5 are 42.6 ( 28 and 38.2
( 49 kcal mol-1, respectively. This suggests that CF3O- and
ClONO2, HCl, and HNO3 are important trace species involved
in the heterogeneous chemistry that leads to the Antarctic polar
ozone hole.1 The relatively inert chlorine reservoir species,
ClONO2 and HCl, are converted into the photolabile species
HOCl and Cl2 by heterogeneous reactions 1 and 2.2
-
SF6 may fluoride transfer in a similar manner. CF3O has a
very large electron affinity (4.2 ( 0.2 eV 9), and consequently
it is unlikely that CF3O- will charge transfer to any atmospheric
species. However, CF3O- may undergo proton transfer with a
strong acid such as HNO3 to form NO3-. This could destroy
the selectivity of CF3O- as a reagent ion for the detection of
HNO3, because there are many species in the atmosphere such
as N2O5, ClONO2, and organic nitrates which might also react
with CF3O- to give NO3-. Therefore, the chemistry of CF3O-
was studied to determine whether it reacts selectively with
atmospheric species. The rate coefficients and product yields
were measured for the reactions of CF3O- with a series of
compounds. This series includes the primary candidates for
CIMS detection, HNO3 and ClONO2, and other atmospheric
constituents such as CH3NO3, HO2NO2, and H2O which could
possibly interfere with CF3O- detection schemes.
The reactivity of CF3O- is also of fundamental interest
because of the formation of CF3O and CF3OH in the atmosphere.
CF3O radicals are a key intermediate in the degradation of many
hydrofluorocarbon and hydrochlorofluorocarbon compounds
which are currently being used as replacements for chlorofluo-
rocarbons.10 The primary loss for CF3O radicals in the lower
atmosphere is reaction with hydrocarbons11 and possibly water12
to form CF3OH. The only efficient loss process for CF3OH in
the atmosphere that has been identified is reactive uptake onto
water droplets.13-16 It is likely that CF3OH dissociates in
aqueous solution because it is a relatively strong acid.9,17
ClONO2 + HCl f Cl2 + HNO3
ClONO2 + H2O f HOCl + HNO3
(1)
(2)
These reactions are essential for the occurrence of the dramatic
ozone loss during the Antarctic spring. In addition, the
partitioning of the reservoir chlorine compounds (i.e., [ClONO2]/
[HCl]) at midlatitudes is not well understood.3 Despite the
importance of these compounds, there are few specific in situ
measurements of ClONO2 and HNO3 in the stratosphere. These
species can be detected using chemical ionization mass spec-
trometry (CIMS) as has been demonstrated by Hanson and
Ravishankara,4 Arnold and co-workers,5 and Viggiano and co-
workers.6 We have undertaken the development of a chemical
ionization scheme that can be used for the simultaneous in situ
measurement of ClONO2, HNO3, and HCl in the stratosphere.
HNO3, ClONO2, HCl, and many other species have been
selectively detected in laboratory studies using CIMS and
- 4,7
fluoride transfer reactions of SF6
.
For example, HNO3 and
ClONO2 are detected using the following reactions.
-
SF6- + HNO3 f NO3 ‚HF + SF5
(3)
(4)
-
SF6- + ClONO2 f NO3 ‚FCl + SF5
These reactions allow specific detection of these compounds
with a mass spectrometer at the parent mass plus 19 amu.
However, the fast charge transfer reaction of SF6 with O3
CF3OH(aq) + H2O(1) T H3O+(aq) + CF3O-(aq) (5)
-
makes this ion unsuitable for the in situ detection of trace species
Lovejoy et al.13 have postulated that CF3OH may be lost in
aqueous solution due to the reaction of CF3O- with water. For
this reason, another goal of this work is to examine the reactivity
of CF3O- with water in the gas-phase to provide insight into
the reactivity of CF3O- in aqueous solution.
in the atmosphere.7 Thus, it is desirable to identify an ion with
†Also affiliated with Cooperative Institute for Research in Environmental
Sciences, University of Colorado, Boulder, CO.
X Abstract published in AdVance ACS Abstracts, December 1, 1995.
0022-3654/96/20100-0190$12.00/0 © 1996 American Chemical Society
+
+

Reactions of CF3O- with Atmospheric Trace Gases
J. Phys. Chem., Vol. 100, No. 1, 1996 191
TABLE 1: Reactions of CF3O-
yielda kmeasb (10-9 cm3
Experimental Section
k
coll (10-9 cm3
(%) molecule-1 s-1) molecule-1 s-1
)
The reaction rate coefficients for CF3O- were measured using
a flowing-afterglow apparatus. Both the experimental technique
and apparatus have been described thoroughly7,18 and are only
briefly discussed here. The ion-molecule reactions were carried
out in a 130 cm long × 7.3 cm i.d. flow tube which was sampled
into a quadrupole mass filter. Helium was used as the carrier
gas at a flow rate of 100 STP cm3 s-1 (STP ) 0 °C, 1 atm) and
a pressure of ∼0.4 Torr. This yielded a flow velocity of ∼5000
cm s-1 and reaction times on the order of 10 ms. The ion source
was a heated thoriated-iridium filament biased at -50 V and
regulated to give a constant emission current of 5-10 µA.
CF3O- was generated by electron attachment to CF3OOCF3,
which was added to the flow tube just downstream of the
filament.19,20 CF3O- was the only ion identified as a product
of electron attachment to CF3OOCF3. CF3O-‚HF was the only
secondary ion generated by this method and was typically
<0.1% of CF3O-. Neutral reactant was added through either
of two inlets which were spaced along the flow tube axis to
give reaction distances of 55 and 80 cm. The CF3O- signal
was monitored as a function of the neutral reactant concentration
reactant
product(s)
ClONO2 NO3-‚FCl,
100
95
5
1.1
1.2
CF2O
HNO3
NO3-‚HF,
CF2O
2.2
1.9
NO3-, CF3OH
d
N2O5
HCl
SO2
NO3-, CF3ONO2 100
0.9 ( 0.4
Cl-‚HF, CF2O
100
100
100?
1.4
0.6
?
1.3
1.5
?
FSO2-, CF2O
HO2NO2 NO3-‚HOF,
CF2Od
HI
I-, CF3OH
I-‚HF, CF2O
IF2-, HF, COd
HSO4-, CF3OH
HSO4-‚HF,
CF2O
75
20
5
87
13
1.2 ( 0.6
1.2c
0.8
H2SO4
H2O
PAN
CH3NO3 N.R.
O3
NO2
O2
CO2
CF3O-‚H2O
N.R.
100
3 × 10-5 c
<0.05
<0.001
<0.001
<0.001
<0.0001
<0.0001
N.R.
N.R.
N.R.
N.R.
for both reaction distances and the rate coefficients were
18
calculated using the standard method.7,
The use of two
a The estimated uncertainty in the product yields is 20% for the major
channels and is a factor of 2 for the minor channels. b The estimated
uncertainty is (30% unless otherwise noted. c The estimated uncertainty
is +100%, -50%. d See discussion in text.
reaction distances provided a test of the calculated reaction time.
The SO2 (99.9%), HCl (99.95%), H2SO4 (96%), O2 (99.99%),
HI (99.5%), and CO2 (99.99%) were all obtained commercially.
The ClONO2, HNO3, N2O5, O3, and NO2 were synthesized and
purified by procedures which have been described previously.7
The CH3C(O)O2NO2 (PAN),21 CH3NO3,22 and HO2NO223 were
synthesized by standard methods. The concentrations of the
SO2, HCl, O3, N2O5, ClONO2, NO2, and HNO3 were measured
by UV absorption as described previously.7 The concentrations
of PAN, CH3NO3, CO2, and O2 were determined from the
pressure rate of change in a calibrated volume due to a flow of
the pure compound. HI was prepared as a dilute mixture
(∼0.1% partial pressure) in UHP He and stored in a glass bulb.
The concentration of the bulb was checked by UV absorption
on a diode array spectrometer and was found to be ap-
proximately a factor of 2 lower than determined manometrically.
In general, HI was found to be sticky and difficult to handle.
For these reasons, the HI rate constant was calculated using
the average of the concentrations derived from the absorption
and pressure measurements.
uncertainty is assigned to this determination primarily due to
mass discrimination effects.
The CF3O- + H2O reaction rate coefficient was measured
on another apparatus designed to operate at flow tube pressures
of 10-50 Torr using air or N2 carrier gas. This apparatus is
identical to the first apparatus7 except it is equipped with a
radioactive ion source (polonium-210) and has a 3.5 cm i.d.
flow tube. H2O was introduced to the flow tube by passing a
measured flow of UHP nitrogen (0.3-2 STP cm3 s-1) through
a bubbler filled with distilled water at atmospheric pressure
(∼620 Torr). The [H2O] was calculated assuming saturation
of the nitrogen flow. The saturated stream entered the flow
tube through a heated inlet to prevent condensation of the water.
The inlet was located 70 cm from the ion sampling orifice. The
CF3O- signal was monitored as a function of [H2O], and a
pseudo-second-order rate coefficient was derived from this decay
using the assumption that the reaction time was given by the
reaction distance (70 cm) divided by the average flow velocity.
The standard correction to the flow velocity was not made in
this case due to the large distance (∼50 cm) needed to establish
laminar flow for the reaction conditions. For this reason, the
rate coefficient is reported with a factor of 2 uncertainty.
The reactions of CF3O- with HO2NO2 and H2SO4 were
studied qualitatively due to the difficulty in determining the
concentration of these species. The HO2NO2 could not be
delivered to the flow tube without a significant HNO3 impurity,
and consequently several potential products of the CF3O-/HO2-
NO2 reaction were possibly obscured. The rate coefficient for
the reaction of CF3O- with H2SO4 was estimated relative to
the Cl-/H2SO4 reaction rate coefficient (k ) 2.7 × 10-9 cm3
molecule-1 s-1 24
) in the following manner. H2SO4 was added
Results
to the flow tube in low enough concentrations so that the reactant
ion concentration was essentially unchanged. For these condi-
tions, the ratio of the sum of the product ion signals (P-) to the
reactant ion signal (R-) is given by
The measured rate coefficients and product yields for the
reactions of CF3O- are listed in Table 1. The product yields
were determined at low concentrations of the neutral reactants
and at low resolution of the quadrupole mass filter to minimize
the effects of secondary reactions and mass discrimination,
respectively.7 Also in Table 1 are the collisional rate coefficients
which were calculated by the method of Su and Chesnavich.25
The dipole moment data for these calculations were taken from
ref 26, except for ClONO2.27 The polarizability data were taken
(P-)/(R-) ) k[H2SO4]t
(6)
where k is the rate coefficient for reaction of R- with H2SO4,
[H2SO4] is the sulfuric acid concentration, and t is the reaction
time. The relative rate coefficients were extracted from
measurements of the ion ratio defined in eq 6 for the reactions
of Cl- and CF3O- with H2SO4 for the same experimental
conditions, i.e., constant t and [H2SO4]. A factor of 2
29
from ref 28, except for ClONO2 and HNO3 which was
estimated to be 4.5 Å3. The thermochemical data were taken
from the NIST Negative Ion Database9 for ions and from ref
30 for neutrals. The thermochemistry of reactions of CF3O-
+
+

192 J. Phys. Chem., Vol. 100, No. 1, 1996
Huey et al.
is based on the fluoride affinity of CF2O, except where noted,
because this is the most reliable thermodynamic quantity
available for the CF3O- ion.9
is trifluoromethyl nitrate (CF3ONO2) which is thought to be
the product of the association reaction of CF3O with NO2.31
This gives a lower limit for the CF3O-NO2 bond strength of
>25 kcal mol-1 based upon a lower limit of >4.0 eV for the
As expected, CF3O- fluoride transfers rapidly to HNO3, H2-
SO4, ClONO2, HCl, and SO2. The products of the fluoride
electron affinity of CF3O.9
-
-
transfer reactions are NO3 ‚HF, HSO4-‚HF, NO3-‚FCl, Cl-‚HF,
CF3O- reacts rapidly with HI and forms I-, I-‚HF, and IF2
.
and FSO2-, respectively. All of these product ions except for
FSO2- are believed to have the structure of a diatomic molecule
clustered to a stable core ion (X-‚AB). The assigned structures
of the HNO3 and ClONO2 fluoride transfer product ions are
supported by the observed fast ligand switching reactions with
excess neutral reactant to form NO3-‚HNO3 and NO3-‚ClONO2,
respectively. The assigned structure of the H2SO4 fluoride
transfer product ion is supported by the observed ligand
switching reaction with HNO3 to form HSO4-‚HNO3. Ligand
switching reactions with H2O were also attempted, but none of
the fluoride transfer product ions were found to react (k < 10-13
cm3 molecule-1 s-1). In order to further test the stability of
the product ions, laboratory air (∼20% of the total flow) was
added to the flow tube. The cluster product ions did not react
with any of the major constituents of the laboratory air and were
found to cluster slowly with water (k < 10-14 cm3 molecule-1
s-1 at total pressures of ∼25 Torr).
The first two products are formed by proton and fluoride
transfer, respectively. The third product, IF2-, is formed by a
complex rearrangement reaction. HF and CO are assigned as
the neutral products of the IF2- channel because this gives the
most exothermic reaction.
CF3O- + HI f IF2- + HF + CO
(9)
Reaction 9 places a limit of <-154 kcal mol-1 on the heat of
-
formation of IF2
.
CF3O- reacted slowly with water with a pseudo-second order
rate coefficient of ∼3 × 10-14 cm3 molecule-1 s-1 over the
pressure range 10-25 Torr of N2. The primary product of the
reaction was identified as CF3O-‚H2O. Larger water clusters
(CF3O-‚(H2O)n) were also observed (up to n ) 6). The signals
for the n g 4 clusters were always much less than for the n e
3 clusters for all [H2O]. The product ions F-‚HF and F-‚(HF)2
were also observed in the mass spectrum when water clusters
were present. This is consistent with CF3O- undergoing the
following reactions in water clusters:
CF3O- reacts with the strong acids HI, H2SO4, and HNO3
by both proton and fluoride transfer and with HCl by only
fluoride transfer. The proton transfer channel (reaction 7) was
identified by the appearance of X- in the mass spectrum of the
CF3O- + HX reaction.
CF3O-‚(H2O)n f F-‚HF + HF + CO2 + (n - 1)H2O
CF3O- + HX f X- + CF3OH
(7)
(10)
f F-‚(HF)2 + CO2 + (n - 1)H2O (11)
This determination is possibly ambiguous as X- can be formed
from the collisional dissociation of X-HF in the ion sampling
process or directly by a fluoride transfer reaction.
The marked decrease in the intensity of the n ) 4 cluster
compared to the n e 3 clusters possibly suggests that reactions
10 and 11 are faster for water clusters of n g 4. This is not
due to a thermodynamic threshold because ∆rG°298 < -10 kcal
mol-1 for reactions 10 and 11 even for n ) 1. This suggests
that there may be a kinetic barrier to reactions 10 and 11 for
the smaller water clusters. The sharp decrease in abundance
from the n ) 3 to the n ) 4 water cluster could also be due to
weak binding of the fourth water molecule. This might be
expected if the first three water molecules form a full solvent
shell around CF3O-.
CF3O- + HX f X- + HF + CF2O
(8)
However, the absence of Cl- in the mass spectrum of the
CF3O-/HCl reaction ([Cl-] < 0.001[Cl-‚HF]) indicates that
X-‚HF clusters are stable for the sampling conditions used for
these experiments. We also found that the X- to X-‚HF ratio
for the HNO3 reaction was insensitive to the magnitude of the
ion extraction voltages. This indicates that the observed X-
channels are not formed from the collisional dissociation of the
fluoride transfer products (X-‚HF). Reaction 8 is energetically
Discussion
favored when HX is a strong acid. For example, ∆rG°298
)
-3.1 ( 2.0 kcal mol-1 for reaction 8 with X ) Cl. However,
our results indicate that the forward rate coefficient must be
less than 1.4 × 10-12 cm3 molecule-1 s-1. This suggests that
there is a significant barrier to reaction 8. Therefore, the X-
channels from HI, H2SO4, and HNO3 are tentatively assigned
as due to proton transfer (reaction 7).
CF3O- fluoride transfers to ClONO2, HCl, and HNO3 at
essentially the collision rate. The products of these reactions
are very stable cluster ions that provide selective identification
of these compounds at the parent mass plus 19 amu. The
product ions do not react with H2O or other relatively abundant
atmospheric species such as CO2 and O3. This allows for
sensitive and selective in situ detection of these species using
CIMS with CF3O- as the reactant ion.
A theoretical detection limit for HNO3 for typical operating
conditions of our CIMS can be estimated in the following
manner. HNO3 is detected utilizing reaction 12 (k12 ) 2.1 ×
10-9 cm3 molecule-1 s-1).
CF3O- reacts with HO2NO2 by fluoride transfer to form
HFNO4-. There were no other product channels, however,
minor NO3- and HF‚NO3- products cannot be ruled out due to
the presence of HNO3. Two likely structures for the observed
fluoride transfer product are NO4-‚HF and NO3-‚HOF. It was
difficult to obtain structural information from these experiments,
but we propose NO3-‚HOF as the most stable structure. This
product channel is exothermic if the NO3--HOF bond strength
-
CF3O- + HNO3 f NO3 ‚HF + CF2O
(12)
-
is >18 ( 3 kcal mol-1. There was no evidence for NO4
formation in the mass spectrum. For example, NO4-‚HF might
react with HNO3 to form NO4-‚HNO3.
The rate of production of the product ion is given by
-
d[NO3 ‚HF]/dt ) k12[HNO3][CF3O-]
(13)
CF3O- reacts rapidly with N2O5 to form NO3-. For this
reaction to be exothermic, CF3O and NO2 must bond to offset
the breaking of the NO2-NO3 bond in N2O5. A likely product
For sampling air with our CIMS, the ratio of the concentration
+
+

Reactions of CF3O- with Atmospheric Trace Gases
J. Phys. Chem., Vol. 100, No. 1, 1996 193
of a species in the ambient air to its concentration in the CIMS
is approximately 100.32 This ensures that the product of the
ion-molecule rate coefficient (k), the ion-molecule reaction
time (t), and [HNO3] is less than 0.1. For these conditions, the
CF3O- signal is changed by less than 10% due to the presence
of HNO3, and the ratio of the product to the reagent ion signal
at reaction time t is given by
The reaction of CF3O- and HO2NO2 could not be studied
quantitatively because HO2NO2 could not be prepared without
a large HNO3 impurity. However, the results indicate that HO2-
NO2 can also be detected in the atmosphere as a fluoride transfer
product of CF3O-. We could not rule out HO2NO2 as an
interference to the detection of HNO3 as NO3-‚HF. This
product could be formed by reaction 15.
-
[NO3 ‚HF]/[CF3O-] ) kt[HNO3]
(14)
-
CF3O- + HO2NO2 f NO3 ‚HF + CO2 + F2 (15)
However, reaction 15 is endothermic if the NO3--HF bond
strength is less than ∼33 kcal mol-1. The NO3--HNO3 bond39
is <29 kcal mol-1, and it is likely that the NO3--HF bond is
weaker because HNO3 readily displaces HF from NO3-‚HF.
Therefore, although it is improbable that HO2NO2 would
interfere with HNO3 detection, it cannot be ruled out, and further
study of CF3O- + HO2NO2 is necessary.
A conservative estimate of the minimum ion ratio we can
measure for a 1 s integration period is 1 × 10-4 and a typical
reaction time is 0.2 s. This corresponds to a detection limit for
HNO3 in ambient air of 2.4 × 107 molecules cm-3 with a
sampling dilution of a factor of 100. In comparison, typical
atmospheric concentrations of HNO3 are on the order of 3 ×
109 molecules cm-3 in the altitude range 10-25 km.30 Theo-
retical detection limits for ClONO2, HCl, SO2, and HO2NO2
can be derived in an analogous manner to HNO3 and indicate
that these species can be simultaneously measured using CIMS
with CF3O- as a reagent ion.
The destruction of CF3O- in water clusters suggests that
CF3O- is unstable in aqueous solution. This assists the
interpretation of the measurements of the loss of CF3OH on
liquid water and sulfuric acid solutions by Lovejoy et al.13 They
found that the reaction probability (γ) of CF3OH on a liquid
The reaction of CF3O- in water clusters limits its use to
regions of the atmosphere where [H2O] is relatively low. The
core product ion (F-) formed by the CF3O-/H2O reaction is
not suitable for atmospheric use because it does not react
surface increases strongly with water activity (γ (aH O)
2.5).
2
The strong dependence of γ on water activity is consistent with
CF3OH(aq) dissociating to form CF3O-(aq) and H3O+(aq)
(reaction 5) followed by the hydrolysis of CF3O-. As the water
activity of a solution increases, the acidity decreases and the
amount of CF3OH(aq) that dissociates increases. CF3O-(aq)
will probably rapidly hydrolyze and directly convert CF3OH
into HF and CO2.
selectively with atmospheric species. For example, F- reacts
33
rapidly with HNO3, N2O5,
ClONO2,34 and many alkyl
nitrates35 to produce NO3-. To preserve the CF3O- reagent
ion, we found that the [H2O] in our CIMS must be <∼1 ×
1014 molecules cm-3. This is equivalent to ∼0.3 Torr of H2O
in the sampled air for a dilution factor of 100 and corresponds
to altitudes of greater than roughly 5 km for typical midlatitude
conditions.36 This includes the upper troposphere and strato-
sphere where it is most important to measure ClONO2, HNO3,
and HCl. It should be noted that water clustering to both the
reagent and product ions may be more efficient at the lower
temperatures of the upper atmosphere. However, this potential
problem can most likely be overcome by heating the ion flow
tube. Therefore, CF3O- appears to be an appropriate reagent
ion for the simultaneous measurement of HCl, ClONO2, and
HNO3 at altitudes above 5 km.
Acknowledgment. This work was supported in part by the
NOAA Climate and Global Change Program. We thank
AFEAS and Prof. Darryl Desmarteau for providing the CF3-
OOCF3. We also thank F. L. Eisele and R. L. Mauldin III for
sharing their results on CF3O- and for suggesting that CF3O-
is reactive within water clusters. We gratefully acknowledge
J. B. Burkholder and R. K. Talukdar for their help with the
synthesis of several of the compounds used in this study. We
also thank E. R. Lovejoy and the reviewers for several useful
comments.
The results of the reaction of CF3O- with HCl places a limit
on the gas phase acidity of CF3OH. The only reported
measurement of the gas phase acidity of CF3OH is ∆G°acid(CF3-
OH) ) 347.6 ( 2 kcal mol-1 by Taft et al.37 Taft et al. noted
that their measurement was possibly complicated by fluoride
transfer reactions (reaction 8) that give the same ion product as
proton transfer. The absence of the proton transfer product Cl-
in the reaction of HCl with CF3O- is strong evidence that CF3-
OH is a stronger gas phase acid than HCl (∆G°acid(CF3OH) <
∆G°acid (HCl) ) 328 kcal mol-1).9 The formation of X- from
the reaction of CF3O- with HX is not definitive evidence that
CF3OH is a weaker acid than HX as the X- product could be
due to fluoride transfer. Therefore, only a limit of <328 kcal
mol-1 can be derived for ∆G°acid(CF3OH) from this work.
However, the gas phase acidity of CF3OH was bracketed in a
related study of the reactions of CF3OH with a series of
conjugate bases (X-) of strong acids (HX).17 CF3OH was
determined to be a stronger acid than HCl and a weaker acid
than HNO3. This result supports the assignment of the X-
product from HNO3, HI, and H2SO4 as a proton transfer channel.
It should be noted that the branching between the fluoride and
proton transfer for the reaction of CF3O- with HNO3 was found
to be insensitive to pressure up to 30 Torr. Mauldin and Eisele38
have found that the fluoride transfer channel of this reaction
dominates at ∼620 Torr.
References and Notes
(1) Solomon, S. ReV. Geophys. 1988, 26, 131.
(2) Solomon, S.; Garcia, R. R.; Rowland, F. S.; Wuebbles, D. J. Nature
1986, 321, 755.
(3) Kawa, S. R.; et al. J. Geophys. Res. 1992, 97, 7905. Webster, C.
R.; et al. Geophys. Res. Lett. 1994, 21, 2575.
(4) For example: Hanson, D. R.; Ravishankara, A. R. J. Phys. Chem.
1992, 96, 2682. Hanson, D. R.; Ravishankara, A. R. J. Geophys. Res. 1991,
96, 17307.
(5) For example: Arnold, F.; Knopp, G. Int. J. Mass. Spectrom. Ion
Processes 1987, 81, 33. Mohler, O.; Arnold, F. J. Atmos. Chem. 1991, 13,
33. For a recent review of this work see: Viggiano, A. A. Mass Spectrom.
ReV. 1993, 12, 115.
(6) Viggiano, A. A.; Morris, R. A.; Van Doren, J. M. J. Geophys. Res.
1994, 99, 8221.
(7) Huey, L. G.; Hanson, D. R.; Howard, C. J. J. Phys. Chem. 1995,
99, 5001.
(8) Larson, J. W.; McMahon, T. B. J. Am. Chem. Soc. 1983, 105, 2944.
(9) Bartmess, J. E. NIST NegatiVe Ion Energetics Database, Version
3.0; Standard Reference Database 19B; National Institute of Standards and
Technology, 1993.
(10) Atkinson, R.; Cox, R. A.; Lesclaux, R.; Niki, H.; Zellner, R. An
Assessment of Potential Degradation Products in the Gas-Phase Reaction
of the AlternatiVe Fluorocarbons in the Troposphere; WMO, Global Ozone
Research and Monitoring Project, AFEAS Report 20, 1991; Vol. II.
(11) Bevilacqua, T. J.; Hanson, D. R.; Howard, C. J. J. Phys. Chem.
1993, 97, 350. Chen, J.; Zhu, T.; Niki, H.; Mains, G. J. Geophys. Res. Lett.
1992, 19, 2215. Jensen, N. R.; Hanson, D. R.; Howard, C. J. J. Phys. Chem.
1994, 98, 8574. Barone, S. B.; Turnipseed, A. A.; Ravishankara, A. R. J.
Phys. Chem. 1994, 98, 4602.
+
+

194 J. Phys. Chem., Vol. 100, No. 1, 1996
Huey et al.
(12) Wallington, T. J.; Hurley, M. D.; Schneider, W. F.; Sehested, J.;
Nielsen, O. J. J. Phys. Chem. 1993, 97, 7606.
(13) Lovejoy, E. R.; Huey, L. G.; Hanson, D. R. J. Geophys. Res. 1995,
100, 18775.
(14) Huey, L. G.; Hanson, D. R.; Lovejoy, E. R. J. Geophys. Res. 1995,
100, 18771.
(15) Wallington, T. J.; Schneider, W. F. EnViron. Sci. Technol. 1994,
28, 1198.
(16) Schneider, W. F.; Wallington, T. J.; Minschwaner, K.; Stahlberg,
E. A. EnViron. Sci. Technol. 1995, 29, 247.
(17) Huey, L. G.; Dunlea, E. R.; Howard, C. J. J. Phys. Chem.,
submitted.
of Electric Dipole Moments for Molecules in the Gas Phase, 1967.
(27) Suenram, R. D.; Johnson, D. R. J. Mol. Spectrosc. 1977, 65, 239.
(28) Miller, T. M. In CRC Handbook of Chemistry and Physics; Weast,
R. C., Ed., CRC Press: Boca Raton, FL, 1990; p E70-79.
(29) Haas, B. M.; Crellin, K. C.; Kuwata, K. T.; Okumura, M. J. Phys.
Chem. 1994, 98, 6740.
(30) DeMore, W. B.; Sander, S. P.; Golden, D. M.; Hampson, R. F.;
Kurylo, M. J.; Howard, C. J.; Ravishankara, A. R.; Kolb, C. E.; Molina,
M. J. Chemical Kinetics and Photochemical Data for Use in Stratospheric
Modeling, EValuation No. 11, JPL Publication 94-26, Jet Propulsion
Laboratory, California Institute of Technology: Pasadena, CA, 1994.
(31) Chen, J.; Young, T.; Niki, H. J. Phys. Chem. 1993, 97, 11696.
(18) Ferguson, E. E.; Fehsenfeld, F. C.; Schmeltekopf, A. L. AdV. At.
Mol. Phys. 1969, 5, 1.
(32) The concentration of a trace species in the CIMS is less than the
ambient concentration for two reasons. The sampled gas undergoes an
expansion from ambient pressure to ∼25 Torr, the operating pressure of
the CIMS. The ambient air is also further diluted by the addition of a flow
of dry nitrogen to the CIMS which is typically 2-4 times greater than the
flow of the sampled air.
(33) Ikezoe, Y.; Matsuoka, S.; Takebe, M.; Viggiano, A. A. Gas Phase
Ion-Molecule Reaction Rate Constants Through 1986; Maruzen: Tokyo,
1987.
(34) Huey, L. G. Unpublished results.
(35) Eberhard, J. E.; Villalta, P. W. Private communication.
(36) U.S. Standard Atmosphere, 1976.
(37) Taft, R. W.; Koppel, I. J.; Topsom, R. D.; Anvia, F. J. Am. Chem.
Soc. 1990, 112, 2047.
(38) Mauldin, R. L.; Eisele, F. L. Private communication.
(19) The CF3OOCF3 was synthesized by Prof. Darryl D. DesMarteau,
who prepared the peroxide in high yield and purity by a modification of
procedures given in ref 20. Trifluoromethyl hypofluorite and COF2 were
allowed to react in a Monel bomb containing coarse copper gauze coated
with AgF2. Using a 45% excess of CF3OF at 550 K, the yield of CF3-
OOCF3 based on COF2 was 95-98%.
(20) Porter, R. S.; Cady, G. H. J. Am. Chem. Soc. 1957, 79, 5628.
Roberts, H. L. J. Chem. Soc. 1964, 86, 4538.
(21) Gaffney, J. S.; Fajer, R.; Senum, G. I. Atmos. Eniron. 1984, 18,
215.
(22) Blatt, A. H., Ed. Organic Synthesis; Wiley: New York, 1943;
Collect. Vol. II.
(23) Molina, L. T.; Molina, M. J. J. Photochem. 1981, 15, 97.
(24) Dotan, I.; Albritton, D. L.; Fehsenfeld, F. C.; Streit, G. E.; Ferguson,
E. E. J. Chem. Phys. 1978, 68, 15.
(25) Su, T.; Chesnavich, W. J. J. Chem. Phys. 1982, 76, 5183. Su, T. J.
Chem. Phys. 1985, 82, 2164.
(26) Nelson, R. D.; Lide, D. R.; Maryott, A. A. National Standard
Reference Data Series-National Bureau of Standards-10, Selected Values
(39) Davidson, J. A.; Fehsenfeld, F. C.; Howard, C. J. Int. J. Chem.
Kinet. 1977, 9, 17.
JP951928U
+
+