Reexamination of the quenching of NO+ vibrations by O 2(a 1δg)

Article abstract of DOI:10.1021/jp1042987

The quenching of vibrationally excited NO⁺ by O₂(a ¹δ_g) has been examined using the monitor ion technique and chemical generation of O₂(a ¹δ _g). In contrast to previous results which showed that the rate constant was much larger than for ground state O₂, this study finds that the rate constant for quenching is below the detection limit (<10 ^-11 cm³ s^-1) of this experiment. The previous experiments produced O₂(a ¹δ_g) in a discharge, which would also produces O atoms. We found that the monitor ion CH₃I⁺ reacts with O atoms to produce CHIOH⁺. This is the likely cause of error in the previous experiments.

Full text of DOI:10.1021/jp1042987

7506
J. Phys. Chem. A 2010, 114, 7506–7508
1
Reexamination of the Quenching of NO+ Vibrations by O2(a ∆g)
Nicole Eyet†,‡,§ and A. A. Viggiano*,‡
Department of Chemistry, Saint Anselm College, 100 Saint Anselm DriVe, Manchester, New Hampshire 03102, USA,
Space Vehicles Directorate, Air Force Research Laboratory, 29 Randolph Road, Hanscom Air Force Base,
Massachusetts 01731-3010, USA, and Institute for Scientific Research, Boston College
ReceiVed: May 11, 2010; ReVised Manuscript ReceiVed: June 7, 2010
The quenching of vibrationally excited NO+ by O2(a 1∆g) has been examined using the monitor ion technique
and chemical generation of O2(a 1∆g). In contrast to previous results which showed that the rate constant was
much larger than for ground state O2, this study finds that the rate constant for quenching is below the detection
1
limit (<10-11 cm3 s-1) of this experiment. The previous experiments produced O2(a ∆g) in a discharge,
which would also produces O atoms. We found that the monitor ion CH3I+ reacts with O atoms to produce
CHIOH+. This is the likely cause of error in the previous experiments.
Introduction
O2(X 3Σg) (k ) 2 × 10-13 cm3 s-1).10 The O2(a 1∆g) was produced
in a microwave discharge, and its concentration was not measured.
The fraction of O2(a 1∆g) produced was estimated to be 5-15%
and it was assumed that no other reactive oxygen species (O, O2(V),
O3) were involved. For these reasons, we have decided to
re-examine the kinetics of this quenching reaction, and the results
are reported here. In the process, we have also identified a
potentially complicating reaction involved in the previous measure-
ments and report on that chemistry as well.
To resolve a discrepancy in the rate constants for O- and
O2- reacting with O2(a 1∆g), we have developed a new technique
for studying ion-molecule reactions with electronically excited
oxygen.1 The technique relies on (1) an emission cell calibrated
to the PSI standard for absolute intensity using a NIST traceable
blackbody source and (2) a chemical generator for producing
O2(a 1∆g).2 Both previous measurements of the reactions
produced O2(a 1∆g) in a microwave discharge.3,4 To our surprise,
we found that both previous measurements of the rate constants
for the reactions were incorrect and that the O- reaction also
proceeded through a previously undiscovered charge transfer
channel. The new results are important in modeling the ratio of
electrons to negative ions in the upper atmosphere.
Experimental Section
The monitor ion11-13 and O2(a 1∆g) production techniques1,2
are thoroughly described in the literature. Here we describe them
briefly. The experiments described here were carried out in a
selected ion flow tube (SIFT)14 incorporating a monitor ion
inlet.13 NO+(V) ions were produced in a moderate pressure ion
source from NO. The ions were mass selected and injected into
the flow tube through a Venturi inlet. The extent of vibrational
excitation was controlled both by the conditions in the ion source
and by the injection energy. Higher injection energies resulted
in more excitation. In these experiments we found an injection
energy of 70 eV to be optimal because it produced significant
excitation but little NO+ fragmentation. A helium buffer carried
the ions down the tube and thermalized the translational and
rotational, but not vibrational, degrees of freedom. Ions were
sampled at the end of the flow tube through a pinhole orifice in
the nose cone, mass analyzed by a quadrupole mass filter, and
Since that time, we have exploited the technique to study the
general reactivity of ions with O2(a 1∆g).1,2,5-7 In these studies, it
was shown that the excitation energy often promoted otherwise
endothermic processes such as charge transfer and electron
detachment. A particularly unique example is that energy transfer
from O2(a 1∆g) to OH-(H2O)1,2 results in the dissociation of a H2O
ligand. An extensive study of the reactivity of O2(a 1∆g) with anions
has been carried out; good agreement was found for the six organic
ion reactions that had been previously studied.8 Positive ion
reactions have been more challenging since the method produces
about ∼80% O2(X 3Σg) and, unless extreme caution is taken, H2O
is also introduced into the flow tube. Both of these species are
much more reactive with cations than with anions, leading to
several competing reactions occurring simultaneously. We have
also succeeded in evaluating several charge transfer reactions with
1
detected by a discrete dynode multiplier. O2(a ∆g) was added
59 cm from the sampling orifice. Two centimeters before the
sampling orifice a monitor gas was added to convert vibra-
tionally excited NO+ into other ions. We used two monitors
that react with various excited states at different rates.15-17 The
simplest chemistry occurs with CH3I as the monitor since it
does not cluster to any vibrational state of NO+. For V ) 1-4,
the rate constants for the monitor reaction,
1
O2(a ∆g) and found that no concrete correlation emerges when
these rate constants are compared with the analogous measurements
with O2(X 3Σg).
The vibrational quenching of NO+(V) by O2(a ∆g) has not
1
been reexamined with this new technique. This reaction had
been studied elsewhere by the monitor ion technique,9 and in
that study it was found that O2(a ∆g) quenched NO+ vibra-
1
tions much more rapidly (k ) 3 ( 2 × 10-10 cm3 s-1) than
NO+(V) + CH3I f CH3I+ + NO
(1)
* To whom correspondence should be addressed. E-mail:
albert.viggiano@hanscom.af.mil.
are 2, 9, 13, and 14 × 10-10 cm3 s-1, respectively. In addition,
CH3I quenches NO+(V) vibrational excitation with rate constants
equal to 14 and 7 × 10-10 cm3 s-1 for V ) 1 and 2, respectively.
† Saint Anselm College.
‡ Air Force Research Laboratory.
§ Boston College.
10.1021/jp1042987  2010 American Chemical Society
Published on Web 06/28/2010

1
Quenching of NO+ Vibrations by O2(a ∆g)
J. Phys. Chem. A, Vol. 114, No. 28, 2010 7507
Therefore, NO+(V ) 1) was monitored but not as efficiently as
made determining very small slopes difficult, and this method
was abandoned in favor of the former. The absence of a change
in either monitor ion signal also indicates that the electronic
higher states. For that reason, we also used C2H5I as a monitor,
1
energy from O2(a ∆g) does not transfer to NO+ vibrational
NO+(V) + C2H5I f C2H5I+ + NO
(2)
excitation. In that case, an increase in the monitor ion signal
1
would have occurred with O2(a ∆g) addition. There is the
extremely unlikely possibility that the V-TR deexcitation exactly
balances the E-V excitation.
The rate constants for reaction 2 are 0.4, 7.3, 11.2, 16, and
16 × 10-10 cm3 s-1, for V ) 0-4, respectively.16 Quenching of
NO+(V) is less important than for CH3I, with rate constants of
7 and 8 × 10-10 cm3 s-1, for V ) 1 and 2. All rate constants
refer to measurements made at low pressures in an ICR. Under
the SIFT conditions, NO+ is observed to cluster to C2H5I but
not to CH3I. However, we do not know if that applies to all
states or just V ) 0. Thus, the monitor ion chemistry can be
complex with C2H5I.
The derived quenching rate constants are much smaller than
those reported by Dotan et al., who found k ) 3 ( 2 × 10-10
cm3 s-1, using CH3I+ as the monitor.9 A rate as large as the
lower limit (10-10 cm3 s-1) would be easy to measure in our
apparatus. The two experiments were performed in a similar
1
manner except that in the previous experiments the O2(a ∆g)
was generated by discharging O2 rather than through a chemical
reaction. The O2(a 1∆g) concentration in the Dotan et al.
experiments was not measured but estimated based on previous
experiments with O2(a 1∆g). Our experience is that the estimated
The O2(a 1∆g) generator and detector have been described in
detail elsewhere.6 Cl2 was bubbled through a basic solution of
hydrogen peroxide to produce both ground and excited states
3
1
3
1
5-15% conversion of O2(X Σg) to O2(a ∆g) in the discharge
was reasonable;1 however, discharging O2 produces other species
including O atoms. In previous experiments in our laboratory,
when O2(a 1∆g) was made in a discharge with glass wool added
to reduce the O atom concentration, we found that the fractions
of O2, that is, O2(X Σg) and O2(a ∆g), as shown by eq 3.
H O + Cl + 2KOH f O X 3Σ and a 1∆
+
(
2
)
2
2
2
g
g
2 KCl + 2 H2O (3)
3
1
of O2(X Σg) converted to O2(a ∆g) and O were 9 and 1%,
respectively.1 Without the glass wool, the concentration of O
atoms introduced into the flow tube would be larger. Therefore,
O atoms may be expected to affect the chemistry. Comparison
to quenching rates by other atoms shows that quenching of
NO+(V) by O atoms is probably negligible.12
100% of the Cl2 reacted to form one of the states of O2. This
1
reaction is a well-known source of O2(a ∆g)18,19 and has been
used to create a chemical O2/I2 laser (COIL).20 Water was
collected in a trap submerged in a methanol-liquid nitrogen
1
slush maintained at -60 to -70 °C. O2(a ∆g) emissions at
An alternative explanation for the apparent fast quenching
rate in the Dotan et al. experiment is that O atoms could have
reacted with the CH3I+ monitor. We have tested this possibility
and have been able to confirm that a reaction occurs, as shown
in eq 4, where the exothermicity was calculated using Gaussian
03 at the B3LYP/6-311++G(d,p) level of theory.22-24
1270 nm were monitored in a calibrated cell to determine its
concentration.1 Using this technique, mainly He, O2(X 3Σg), and
O2(a 1∆g) entered the SIFT. The fraction of O2 as O2(a 1∆g) in
the current experiments varied from 12-16%.
We also performed two other experiments to look for
interferences in the monitor technique. In the first, the reaction
of CH3I+ with O was studied since the experiment performed
by Dotan et al.9 relied on a discharge method to generate O2(a
1∆g). O atoms are a known product in such a discharge. For
that experiment, the O2(a 1∆g) inlet was replaced with an inlet
incorporating a microwave discharge. We produced O atoms
in the conventional manner.21 N2 was discharged to produce N
atoms, and NO was added downstream in the side arm. The
reaction of N atoms with NO produced O and N2. In a second
experiment, the reactivity of CH3I+, C2H5I+, and NO+(C2H5I)
CH3I+ + O f CHIOH+ + H + ∼220 kJ mol-1 (4)
Unfortunately, the oxygen atoms affected our sampling ef-
ficiency and prevented rate constants from being derived. In
the experiments of Dotan et al.9 a decline in the CH3I+ signal
was observed when the discharge was turned on and was
interpreted as being due to quenching. However, based on our
observations, we believe the decline was actually due to reac-
tion 4.
1
with O2(a ∆g) was determined. For that measurement, we
moved the alkyl iodide addition 30 cm upstream of the O2(a
1∆g) inlet and monitored signal intensity as O2(a 1∆g) was added.
A similar error would exist if the monitor ions reacted with
1
O2(a ∆g). In order to investigate this possibility, we injected
Results and Discussion
NO+ and added either CH3I or C2H5I 30 cm upstream of the
1
Reaction rate constants were measured by monitoring the
signal intensity of CH3I+ or C2H5I+ monitor ions as the flow of
O2(a ∆g) inlet to produce CH3I+ or C2H5I+ and NO+(C2H5I).
Enough neutral reagent was introduced such that the chemistry
was complete before O2(a 1∆g) addition, and this simulated the
monitor situation more realistically than if the ions were injected
from the source. None of the alkyl iodide ions were observed
1
O2(a ∆g) was turned on and off on an electronic strip chart.
This technique has been used previously with good success.5
The CH3I+ signal remained essentially constant as the O2(a 1∆g)
was turned on or off (by adding Cl2), indicating little quenching
occurred. The intensity of the C2H5I+ signal showed a very small
1
to react with O2(a ∆g). Therefore, secondary reactions with
1
O2(a ∆g) could not cause an artifact.
3
decrease, <5%. Quenching by O2(X Σg) is slow and had little
The general description of quenching of ions by neutrals has
been laid out by Ferguson.12 Breifly, at thermal energies, the
quenching generally involves formation of a long-lived complex
followed by vibrational predissociation. Good, but not perfect,
correlations between three-body association rate constants for
V ) 0 and vibrational quenching rate constants have been found.
Calculations indicated that the predissociative lifetime was 10-9
to 10-10 s for many systems. These are based on assuming the
effect on the NO+ vibrational distribution.10 Upper limits of the
rate constants for the disappearance of the monitor ions with
1
O2(a ∆g) are <2 × 10-11 cm3 s-1 for C2H5I+ and <8 × 10-12
cm3 s-1 for CH3I+. These are essentially at our limit of detection
for O2(a 1∆g) reactions. Data were also obtained in the normal
1
manner, as a function of added O2(a ∆g). They are consistent
with the strip chart method, but instabilities in the ion signals

7508 J. Phys. Chem. A, Vol. 114, No. 28, 2010
Eyet and Viggiano
complex for ground and excited state complexes form at the
collisional rate and that the lifetime of the complex can be
equated with the efficiency of the reactions.
References and Notes
(1) Midey, A. J.; Dotan, I.; Lee, S.; Rawlins, W. T.; Johnson, M. A.;
Viggiano, A. A. J. Phys. Chem. A. 2007, 111, 5218.
(2) Midey, A. J.; Dotan, I.; Viggiano, A. A. J. Phys. Chem. A 2008,
112, 3040.
(3) Upschulte, B. L.; Marinelli, P. J.; Green, B. D. J. Phys. Chem.
1994, 98, 837.
(4) Fehsenfeld, F. C.; Albrittion, D. L.; Burt, J. A.; Schiff, H. I. Can.
J. Chem. 1969, 47, 1793.
(5) Eyet, N.; Midey, A.; Bierbaum, V. M.; Viggiano, A. A. J. Phys.
Chem. 2010, 114, 1270.
(6) Midey, A. J.; Dotan, I.; Seeley, J. V.; Viggiano, A. A. Int. J. Mass
Spectrom. 2009, 280, 6.
(7) Midey, A. J.; Dotan, I.; Viggiano, A. A. Int. J. Mass Spectrom.
2008, 273, 7.
(8) Bierbaum, V. M.; Schmitt, R. J.; DePuy, C. H. EnViron. Health
Perspect. 1980, 36, 119.
(9) Dotan, I.; Barlow, S. E.; Ferguson, E. E. Chem. Phys. Lett. 1985,
121, 38.
3
O2(X Σg) does not quench NO+(V) effectively because the
two molecules do not interact strongly, in part, because spin
effects lead to a repulsive potential at long-range, instead of
the more common attractive potential. This was confirmed in a
previous study that showed that NO+(O2) is not formed, even
at 90 K,10 indicating a very weak bond. It was hypothesized
that quenching by O2(a 1∆g), because it eliminates the spin effect
problems, may result in an attractive interaction. The original
results supported this idea. That is, O2(a 1∆g) quenched NO+(V)
rapidly. However, when comparing the quenching rate constant
to that found for other diatomics, one finds that the previously
1
measured O2(a ∆g) value is inconsistent. The rate constants
for NO+(V) quenching by the similar molecules, N2 and CO,
are 0.7 and 1.0 × 10-11 cm3 s-1 10,12
1∆g) and on the order of the upper limit reported here.
,
respectively. These are at
(10) Viggiano, A. A.; Morris, R. A.; Dale, F.; Paulson, J. F.; Ferguson,
E. E. J. Chem. Phys. 1989, 90, 1648.
(11) Bohringer, H.; Durup-Ferguson, M.; Ferguson, E. E.; Fahey, D. W.
Planet. Space Sci. 1983, 31, 483.
(12) Ferguson, E. E. J. Phys. Chem. 1986, 90, 731.
least a factor of 30 slower than the previous value for O2(a
The previous work on collisional quenching discussed above12
finds a distinct correlation between the quenching rate and that
for three-body association to NO+(V ) 0). With the molecular
parameters of N2, O2(a 1∆g), and CO being similar (vibrational
frequencies are high enough to be negligible) the association
rate should be roughly dependent on the bond strength of
(13) Morris, R. A.; Viggiano, A. A.; Dale, F.; Paulson, J. F. J. Chem.
Phys. 1988, 88, 4772.
(14) Viggiano, A. A.; Morris, R. A.; Dale, F.; Paulson, J. F.; Giles, K.;
Smith, D.; Su, T. J. Chem. Phys. 1990, 93, 1149.
(15) Wyttenbach, T.; Bowers, M. T. J. Phys. Chem. 1993, 97, 9573.
(16) Wyttenbach, T.; Bowers, M. T. J. Phys. Chem. 1992, 96, 8920.
(17) Viggiano, A. A.; Morris, R. A.; Paulson, J. F.; Brown, E. R.; Sutton,
E. A. J. Chem. Phys. 1993, 99, 6579.
(18) Khan, A. A.; Kasha, M. J. Chem. Phys. 1963, 39, 2105.
(19) Seliger, H. Anal. Biochem. 1960, 1, 60.
(20) McDermott, W. E.; Pchelkin, N. R.; Benard, D. J.; Bousek, R. R.
1
NO+-(X), where X ) N2, O2(a ∆g), CO. We calculated the
bond strengths for X with G3 theory using the Gaussian 0322
suite of programs and found them to be 0.22, 0.30, and 0.21
1
eV for N2, O2(a ∆g), and CO, respectively. These values are
Appl. Phys. Lett. 1978, 32, 469.
the same within the error of the calculation and should indicate
that the rates should be comparable. The quenching rate
constants for N2 and CO, are similar to detection limit for O2(a
1∆g) quenching. This may indicate that the O2(a 1∆g) quenching
rate may be just below our limit and that it may still be
(21) Viggiano, A. A.; Howorka, F.; Albritton, D. L.; Fehsenfeld, F. C.;
Adams, N. G.; Smith, D. Astrophys. J. 1980, 236, 492.
(22) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,
M. A.; Cheeseman, J. R.; Montgomery, J. J. A.; Vreven, T.; Kudin, K. N.;
Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.;
Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.;
Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.;
Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M. ; Li,
X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Adamo, C.; Jaramillo, J.;
Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, Z. J.; Cammi, R.;
Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.;
Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels,
A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.;
Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.;
Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz,
P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.;
Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson,
B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian 03W,
ReVision C.02; Gaussian, Inc.: Pittsburgh PA, 2003.
3
substantially greater than that for O2(X Σg).
In summary, we have found that quenching of NO+(V) by
O2(a 1∆g) is much slower than previously reported and
potentially more consistent with what is expected based on
polarizibilities. We tested several of the potential problems that
may have occurred and found that the reaction of CH3I+ with
O atoms probably affected the previous results.
Acknowledgment. We are grateful for the support of the Air
Force Office of Scientific Research for this work. N.E.
acknowledges funding from the Institute for Scientific Research
of Boston College (FA8718-04-C-0055) and the Air Force
Summer Faculty Fellowship Program. We thank Eldon Ferguson
for many helpful discussions over the years.
(23) Becke, A. D. J. Chem. Phys. 1993, 98, 5648.
(24) Lee, C.; Yang, W. T.; Parr, R. G. Phys. ReV. B 1998, 37, 785.
JP1042987