Investigation of the reaction of O3+ with N2 and O2 from 100 to 298 K

Article abstract of DOI:10.1021/jp020311r

The kinetics of the reaction of O₃⁺ with N₂ and O₂ have been studied at 100 and 298 K in a variable temperature-selected ion flow tube (VT-SIFT). The rate constants for O₃⁺ reacting with N₂ measured in the VT-SIFT are slow, proceeding at <5 × 10^-13 cm³ s^-1 at both temperatures, in disaggrement with recent measurements by Cacace et al.¹ of the total rate constant of ca. 1.7 × 10^-10 cm³ s^-1 at 298 K. However, a N₂O₃⁺ intermediate postulated to be involved in the reaction has been observed at 100 K, in agreement with the previous experimental and theoretical results. Rate constants for the reaction of O₃⁺ with O₂ at 100 and 298 K have also been measured in the VT-SIFT. The O₂ reaction rate constants are 3.1 × 10^-10 and 2.9 × 10^-10 cm³ s^-1 at 100 and 298 K, respectively, which are ca. half of the Langevin collision rate constant. Discrepancies between the current results and those of Cacace et al. in regard to the rate constants measured and the N₂O⁺ product ions observed are discussed in light of newer data for the O₃⁺ chemistry and the contributions of excited-state species.

Full text of DOI:10.1021/jp020311r

J. Phys. Chem. A 2002, 106, 11739-11742
11739
+
Investigation of the Reaction of O3 with N2 and O2 from 100 to 298 K
,†
†
‡
Anthony J. Midey,* Skip Williams, Thomas M. Miller, Patrick T. Larsen, and
A. A. Viggiano
Air Force Research Laboratory, Space Vehicles Directorate, 29 Randolph Road, Hanscom Air Force Base,
Massachusetts 01731-3010
ReceiVed: January 31, 2002; In Final Form: August 1, 2002
+
The kinetics of the reaction of O
3
with N
2
and O
2
have been studied at 100 and 298 K in a variable temperature-
+
selected ion flow tube (VT-SIFT). The rate constants for O
3
reacting with N
slow, proceeding at <5 × 10 cm s at both temperatures, in disagreement with recent measurements by
Cacace et al. of the total rate constant of ca. 1.7 × 10 cm s at 298 K. However, a N intermediate
postulated to be involved in the reaction has been observed at 100 K, in agreement with the previous
2
measured in the VT-SIFT are
-
13
3
-1
1
-10
3
-1
+
2 3
O
+
experimental and theoretical results. Rate constants for the reaction of O
also been measured in the VT-SIFT. The O
reaction rate constants are 3.1 × 10 and 2.9 × 10 cm s
at 100 and 298 K, respectively, which are ca. half of the Langevin collision rate constant. Discrepancies
between the current results and those of Cacace et al. in regard to the rate constants measured and the N O
2
product ions observed are discussed in light of newer data for the O
3
2
with O at 100 and 298 K have
-
10
-10
3
-1
2
+
+
3
chemistry and the contributions of
excited-state species. [Cacace, F.; et al. Angew. Chem., Int. Ed. Engl. 2001, 40, 1938.]
+
Introduction
be attributed to [N2O3] . Also, the CAD spectrum for the peak
at 76 amu agrees with the CAD spectrum from a CI experiment
1
+
Cacace et al. have recently investigated the reaction of O3
with N2 as a possible source of N2O, a major atmospheric
precursor of NOx pollutants and a pernicious greenhouse gas.
They have found that this reaction produced either N2O or N2O
+
+
for O2 reacting with N2O, suggesting that the N2O3 formed
has an [N-N-O‚‚‚O-O] arrangement. Furthermore, reion-
ization of the neutrals produced in the CAD experiment from
N2O3 shows a major peak at m/z ) 44 that indicated that some
of the CAD events generate N2O intact.
+
+
+
-
10
3
with a total rate constant of approximately 1.7 × 10
cm
1
-1
s . At this rate, they speculate that this reaction may produce
a significant amount of N2O in the atmosphere, most notably
near power lines, in corona discharges and during thunderstorms.
This source is not presently accounted for in current atmospheric
models; thus, it could have a substantial impact on simulations.
Cacace et al. have proposed the mechanism given by eqs 1
Consistent with their experimental results, density functional
theory (DFT) calculations performed by Cacace et al. indicate
+
that two stable [N2O3] structures exist that are connected by a
-
1
transition state with a 10.6 kcal mol barrier. These structures
require that the N2 moiety be bound to a terminal O atom. In
addition, previous ab initio calculations by Snyder and Sapse
have shown that ion-dipole bonding of N2 to the central O
atom with O3 in a C2V symmetry does not produce a stable
1
+
and 2 for O3 reacting with N2. The reaction proceeds through
+
+
a N2O3 (m/z ) 76) reaction intermediate, where channels (1)
2
structure.
k1
O3+ + N T [N O ] 98 N2O + O
+
+
(1)
(2)
Due to the potential importance of this reaction in atmospheric
chemistry, the kinetics of the reaction of O3 with N2 have been
studied further in the AFRL variable temperature-selected ion
flow tube (VT-SIFT) under thermalized conditions. This distinc-
tion is important as Cacace et al. have noted that excited states
may be generated in their experiments. Different sources of O3
have been used in the VT-SIFT, and temperature-dependent
studies of the kinetics have been made. The reaction of O3
2
2
3
2
+
k2
+
98O2 + N O
2
and (2) are 4.2 and 22.9 kcal mol-1 exothermic, respectively.
Rate constants for reactions 1 and 2 were estimated to be ca.
+
-
11
-10
3 -1
1
.4 × 10 and 1.6 × 10 cm s , respectively, as measured
+
in a Fourier transform ion cyclotron resonance (FT-ICR) mass
spectrometer. In a separate experiment using a ZAB multisector
mass spectrometer, ions at m/z ) 32, 44, and 76 were generated
in their chemical ionization (CI) source using a 5% mixture of
with O2 has also been studied because of its relative importance
to the N2 reaction. The results of this study are reported at
temperatures of 100 and 298 K.
+
O3 in O2 with N2. These ions have been postulated to be O2 ,
Experimental Section
+
+
N2O , and [N2O3] , respectively. The collisionally activated
dissociation (CAD) mass spectra of these ions are consistent
with this assignment, demonstrating that the ion at mass 76 can
3
The VT-SIFT apparatus is briefly described as it pertains to
these experiments. Electron impact ionization on a supersonic
expansion of O2 (99.999% pure, AGA) with 40 psig backing
pressure behind a 0.2 mm nozzle produces O3 ions, presumably
through the clustering reaction of O with O2. Identical results
+
*
To whom correspondence should be addressed:. E-mail:
Anthony.Midey@hanscom.af.mil.
+
†
Under contract to Visidyne, Inc., Burlington, MA.
4
‡
have been obtained using a 5% O3 in O2 mixture in the
stagnation chamber where O3 can be ionized directly. For
Current address: United States Air Force Academy, HQ USAFA,
Colorado Springs, CO 80840.
1
0.1021/jp020311r CCC: $22.00 © 2002 American Chemical Society
Published on Web 11/12/2002

11740 J. Phys. Chem. A, Vol. 106, No. 48, 2002
Midey et al.
simplicity, pure O2 expansion has been used for most of the
experiments discussed here. The ions produced are mass selected
with a quadrupole mass spectrometer and injected into a fast
flow of helium (99.997% pure, AGA) buffer gas that is
introduced through a Venturi inlet. Breakup of the O3+ into O2+
occurs upon injection into the flow tube so the source conditions
have been adjusted to minimize this dissociation during injec-
tion. Unfortunately, the process cannot be eliminated entirely.
The least amount of breakup results in approximately equal
+
+
amounts of O3 and O2 .
The buffer gas is passed through a liquid nitrogen cooled
sieve trap to remove water vapor. Passing liquid nitrogen
through copper lines surrounding the flow tube cools the flow
tube to 100 K for the low temperature study. Nitrogen (99.999%
pure, AGA) is introduced at one of two inlets downstream from
the ion injection point and is allowed to react for a previously
measured reaction time of approximately 3-7 ms to obtain the
rate constants under pseudo-first-order kinetics conditions.
Residual reactant ions and any product ions are sampled by an
aperture in a blunt nose cone, then mass analyzed with a second
quadrupole mass spectrometer, and detected with an electron
multiplier. The instrument is interfaced with a desktop computer
that controls the data acquisition. Rate constants measured in
this manner have relative errors of (15% and absolute errors
of (25%.3
Results and Discussion
Parts a and b of Figure 1 show mass spectra measured at
2
98 K in the VT-SIFT with and without N2 added. In the
+
+
absence of N2 gas, O2 and O3 are the only ions present. As
mentioned above, O2 arises from collisions of O3+ with the
helium buffer gas upon injection. Both internal energy and
translational energy can lead to the dissociation process. In the
spectra in Figure 1, comparable amounts of O2 and O3 are
found. This is the best ratio of these two ions attained in our
apparatus because using lower injection energies leads to a
complete loss of signal. In the presence of N2, a small signal
+
+
+
+
Figure 1. Mass spectra for the reaction of O
3
2
with N at 298 K as
measured in a variable temperature-selected ion flow tube (VT-SIFT).
The spectra have been taken with (a) no reactant gas in the flow tube,
(
b) with N
2
2
added to the flow tube, and (c) with O added to the flow
tube.
+
due to N2 is found. This signal comes from a small fraction
+
5
6
(1%) of electronically excited O2 reacting with N2. Thus, in
constants in the FT-ICR. The current kinetics measurements
in the VT-SIFT, on the other hand, reflect the thermal reaction
rate constant. The participation of excited electronic states as
well as vibrational excitation in the FT-ICR experiments is
reasonable considering that the four lowest doublet states of
the present experiments, addition of N2 comprising about 1%
of the total gas flow to the flow tube changes the spectra to an
almost a negligible extent. Particularly relevant is the fact that
neither a peak for N2O3 (m/z ) 76) nor N2O (m/z ) 44) is
observed, despite higher operating pressures than in the CI
+
+
+
the O3 cation lie in an energy range of 2 eV and the two lowest
+
7,8
experiments of Cacace et al. The authors do note that no N2O3
states are nearly degenerate. Consequently, the discrepancy
1
signal was observed in the FT-ICR experiments.
is strictly because of the different conditions under which the
rate constants have been measured in the FT-ICR vs the VT-
SIFT; i.e., the large rate constants observed in the FT-ICR
+
The total rate constant for the O3 reaction with N2 measured
cm s at 298 K. It cannot be
ruled out that a small amount of O2 is produced via reaction
-
13
3
-1
in the VT-SIFT is <5 × 10
+
1
+
experiments may be due to the presence O3 excited-state
species.
2
, which would be masked by the signal from dissociation of
+
O3 upon injection. However, the kinetics occur at the limit of
the capabilities of the VT-SIFT; thus, the reported rate constant
represents the upper limit that can be detected. The current value
is in serious disagreement with the rate constant measured by
The ion selected at 48 amu in the VT-SIFT source region is
+
+
most likely O3 because of its breakup into O2 upon injection.
However, to rule out the remote possibility that the remaining
+
ion at 48 amu was not SO (possibly formed from trace sulfur
-
10
3
-1 1
Cacace et al. in their FT-ICR, i.e., 1.7 × 10
cm s . In
impurities in the source region from experiments performed
months earlier), O2 has been added to the flow tube. Charge
response to the present results, the authors have stated that the
FT-ICR rate constant should be viewed as phenomenological,
+
+
transfer of O3 with O2 is exothermic but SO charge transfer
+
9,10
+
strictly providing an explanation for the formation of the N2O3
is endothermic,
so that SO will be unreactive with O2.
+
6
and N2O ions created in their CI source. The authors further
noted that the phenomenological rate constant decreases when
the ionizing electron energy is lowered and approaches zero
Adding O2 to the flow tube confirms that the ion at m/z ) 48
+
+
is O3 because it is entirely consumed by O2, producing O2
as seen in Figure 1c at 298 K. The rate constants for the reaction
+
+
-10
-10
3 -1
when the O3 ions are cooled according to a thermalizing
of O3 with O2 are 3.1 × 10 and 2.9 × 10 cm s at 100
and 298 K, respectively, which are almost half of the Langevin
collision rate constant.
technique based on a short, high-pressure Ar pulse invariably
utilized in previous works when measuring thermal rate

Reaction of O3+ with N2 and O2
J. Phys. Chem. A, Vol. 106, No. 48, 2002 11741
+
Figure 3. Kinetics data for the reaction of O
3
2
with N at 100 K as
measured in a variable temperature-selected ion flow tube (VT-SIFT).
The ion signal is plotted vs the concentration of N where the symbols
2
+ + +
+
4
are defined as follows: (b) O , (9) O , (O) N O , (0) O , and
3 2 2 3
+
2 2
]) N O .
(
1
Cacace et al. Figure 3 shows typical kinetics data measured at
+
1
00 K. The N2O3 signal increases with increasing N2 flow
+
with a concomitant decrease in the O3 signal. Even at 100 K
where the complex is stable, the total rate constant for the
+
-13
3 -1
reaction of O3 with N2 is approximately 1.1 × 10 cm s .
+
Although it is clear that thermal O3 does not react rapidly
+
with N2, N2O is made in the CI source in the experiments of
Cacace et al. A likely source of this ion is excited-state
1
reactions. In further tests by Cacace and co-workers, the intensity
+
of the N2O3 peak drops significantly as the pressure of O3 in
the CI source is increased. Decreasing the energy of the ionizing
Figure 2. Mass spectra for the reaction of O +
with N at 100 K as
2
measured in a variable temperature-selected ion flow tube (VT-SIFT).
The spectra have been taken with (a) no reactant gas in the flow tube
3
+
electrons in the CI produces a similar decrease in the N2O3
6
+
signal. In addition, the observation of N3 formed in the CI
2
and (b) with N added to the flow tube.
1
+
mass spectrum is known to involve the reaction of N2 in a
long-lived electronically excited state with N2, further indicating
that excited species are formed in the CI source. Consequently,
it is clear that excited-state ions are generated in the CI source
of Cacace and co-workers.
To further verify that the differences between the FT-ICR
1
measurements and the current VT-SIFT measurements are not
+
due to the reaction of different isomers of O3 , Gaussian 98W
calculations have been performed.11 Several optimizations have
been performed at the MP2(Full)/6-31G(d) level of theory with
different starting points such as the geometry of neutral O3 with
C2V symmetry, as well as a linear structure. Only one stable
To test for the possibility of excited species using the VT-
+
SIFT, O3 has been injected directly into an N2 buffer. In this
+
experiment, the first collisions of O3 with N2 involve electroni-
+
+
cally, vibrationally, and translationally hot O . Most of the
isomer of O3 has been found having C2V symmetry, which
3
+
either electronically or vibrationally excited O3 decomposes
agrees with the previous calculations of the energy levels of
+
+
8,12
This result indicates that the O3+ ions studied in the
into O2 upon injection. However, a small amount of the ions
O3 .
(
around 1-2% of the total ions observed) is converted into
two experiments are the same isomer, even though the source
chemistry is different.
+
N2O . This amount may be somewhat higher as the N2 buffer
+
2
contains 0.001% O , as discussed earlier, and the reaction of
To increase the likelihood of observing the N2O3 complex
+
1,5
+
N O with O2 is fast. Nevertheless, this observation is
in the VT-SIFT, the reaction of O3 with N2 has also been
2
+
consistent with the Cacace measurements showing N2O is
produced in their ion source with O , N , and O present. The
studied at 100 K. Parts a and b of Figure 2 show mass spectra
1
+
+
taken at 100 K both with and without N2 added. The O3 /O2
3
2
2
results also support the assertion that the reactions of excited
ratio is smaller than at room temperature because the longer
+
reaction time allows for more O3+ to react with the small O2
species contribute to the N O ion signal observed previously
2
by Cacace et al.
impurity in the He buffer at 100 K. A peak is observed at mass
7
6, verifying that N2O3+ is stable. This complex must be weakly
The proposed scheme for N2O formation in the atmosphere
+
bound because it is only observed in small abundance (∼25%
will be important only under certain conditions. First, O3 ions
+
of the original O3 signal with a maximum amount of N2 added)
must be produced. Reaction 1 must be faster than the competing
+
+
and only observed at low temperature. No N2O at mass 44 is
reaction of O3 with O2. However, the present VT-SIFT
+
observed. The NO ions arise from the reaction of N2 with
experiments establish that reaction 1 is slow and that the charge-
+
5,13
+
+
O
that is created by the dissociation of O3 during injection.
O4 is seen in Figure 2b with only N2 added because of the
.001% of O2 impurity present in the N2 supply. This ion
transfer reaction of O3 with O2 is fast over a wide temperature
+
+
range. These results indicate that any O3 that is formed will
0
react rapidly with the O2 present in air rather than with N2. In
+
+
+
presumably arises from the reactions of O2 , N2O2 , and maybe
N2O3 with the trace amount of O2. Nevertheless, it is clear
addition, because O3 reacts at close to the collision rate with
+
5
+
O2, the phenomenological rate constant corresponding to O3 *
that the ion with m/z ) 76 arises from N2 in the presence of
O3 ; thus, the stability of the weakly bound N2O3 complex
has been confirmed at 100 K in agreement with the results of
reacting with N2 relevant to conditions prevailing in an
atmospheric discharge plasma, where excited ions are generated,
would have to be close to the collision rate to compete with
+
+

11742 J. Phys. Chem. A, Vol. 106, No. 48, 2002
Midey et al.
the O2 reaction and produce N2O in substantial amounts. This
mechanism would require substantial excited-state production
of O3 , which may be difficult in these highly energetic
(3) Viggiano, A. A.; Morris, R. A.; Dale, F.; Paulson, J. F.; Giles, K.;
Smith, D.; Su, T. J. Chem. Phys. 1990, 93, 1149-1157.
(4) Williams, S.; Campos, M. F.; Midey, A. J.; Arnold, S. T.; Morris,
+
R. A.; Viggiano, A. A. J. Phys. Chem. A 2002, 106, 997-1003.
(5) Ikezoe, Y.; Matsuoka, S.; Takebe, M.; Viggiano, A. A. Gas-Phase
Ion-Molecule Reaction Rate Constants Through 1986; Maruzen Co.,
Ltd.: Tokyo, 1987.
environments considering the small 0.6 eV bond energy of
+
7,14
O3 .
Thus, the current results do not support the contention
+
that N2O may be formed in the atmosphere from O3 and N2
and that the atmospheric impact of the reaction of O3+ with N2
would be minimal under thermalized conditions.
(
6) Cacace, F. Personal communication, 2001.
(7) Muller, H.; Koppel, H.; Cederbaum, L. S. J. Chem. Phys. 1994,
101, 10263-10273.
8) Schmelz, T.; Chambaud, G.; Rosmus, P.; Koppel, H.; Cederbaum,
L. S.; Werner, H.-J. Chem Phys. Lett. 1991, 183, 209-217.
9) Lias, S. G.; Bartmess, J. E.; Liebman, J. F.; Holmes, J. L.; Levin,
R. D.; Mallard, W. G. J. Phys. Chem. Ref. Data 1988, 17, Supplement 1,
-861.
10) Lias, S. G.; Bartmess, J. E.; Liebman, J. F.; Holmes, J. L.; Levin,
(
However, the conditions in atmospheric plasmas and coronas
will have localized nonequilibrated regions relative to the bulk
atmosphere, where a high local concentration of O3 may arise
under conditions involving energetic ionizing electrons. At this
point, electronically and vibrationally excited species should be
abundant. In these concentrated areas, the conditions prevailing
in the CI source of the multisector mass spectrometer of Cacace
(
1
(
R. D.; Mallard, W. G. Ion Energetics Data. In NIST Chemistry WebBook,
NIST Standard Reference Database Number 69; Mallard, W. G., Linstrom,
P. J., Eds.; NIST: Gaithersburg, 1998; pp (http://webbook.nist.gov).
(11) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,
M. A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. 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.; Cioslowski, 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.; Al-Laham, M. A.;
Peng, C. Y.; Nanayakkara, A.; Gonzalez, C.; Challacombe, M.; Gill, P. M.
W.; Johnson, B.; Chen, W.; Wong, M. W.; Andres, J. L.; Gonzalez, C.;
Head-Gordon, M.; Replogle, E. S.; Pople, J. A. Gaussian 98, version A.9;
Gaussian, Inc.: Pittsburgh, PA, 1998.
1
et al. may be more representative of these nonequilibrium
systems.
Acknowledgment. We thank Fulvio Cacace for his helpful
discussions regarding our mutual experiments. We also thank
John Williamson and Paul Mundis for technical support. This
research has been supported by the Air Force Office of Scientific
Research under Project No. 2303EP4. A.J.M. and T.M.M. have
been supported through Visidyne contract number F19628-99-
C-0069.
(
12) Muller, H.; Koppel, H.; Cederbaum, L. S.; Schmelz, T.; Chambaud,
G.; Rosmus, P. Chem. Phys. Lett. 1992, 197, 599.
13) Hierl, P. M.; Dotan, I.; Seeley, J. V.; Van Doren, J. M.; Morris, R.
References and Notes
(
(
1) Cacace, F.; Petris, G. d.; Rosi, M.; Troiani, A. Angew. Chem., Int.
Ed. Engl. 2001, 40, 1938.
2) Snyder, G.; Sapse, D. Chem. Phys. Lett. 1994, 218, 372.
A.; Viggiano, A. A. J. Chem. Phys. 1997, 106, 3540-3544.
(14) Moseley, J. T.; Ozenne, J.-B.; Cosby, P. C. J. Chem. Phys 1981,
74, 337-341.
(