Measurement of the rate coefficient for the OH + NO2 reaction under the atmospheric pressure: Its humidity dependence

Article abstract of DOI:10.1016/j.cplett.2005.12.026

Humidity dependence of the rate coefficient of the OH + NO₂ reaction has been studied. The rate coefficients, measured at the H ₂O partial pressures (PH2O) of 29.1 and 3.7 hPa and at 298 K are (1.15 ± 0.02) × 10^-11 and (1.

Full text of DOI:10.1016/j.cplett.2005.12.026

Chemical Physics Letters 419 (2006) 474–478
www.elsevier.com/locate/cplett
Measurement of the rate coefficient for the OH + NO reaction
2
under the atmospheric pressure: Its humidity dependence
a,b,
c
c
a,
*
*
Y. Sadanaga
, S. Kondo , K. Hashimoto , Y. Kajii
a
Department of Applied Chemistry, Graduate School of Engineering, Tokyo Metropolitan University, 1-1 Minami-Osawa, Hachioji, Tokyo 192-0397, Japan
b
Japan Science and Technology Agency, 4-1-8, Honcho, Kawaguchi, Saitama 332-0012, Japan
Department of Chemistry, Graduate School of Science, Tokyo Metropolitan University, 1-1 Minami-Osawa, Hachioji, Tokyo 192-0397, Japan
c
Received 29 August 2005; in final form 2 December 2005
Available online 4 January 2006
Abstract
Humidity dependence of the rate coefficient of the OH + NO reaction has been studied. The rate coefficients, measured at the H O
2
2
ꢀ
11
ꢀ11
3
ꢀ1 ꢀ1
partial pressures ðPH OÞ of 29.1 and 3.7 hPa and at 298 K are (1.15 ± 0.02) · 10
and (1.40 ± 0.10) · 10
cm molecule s
(± 2r),
2
2 2
respectively. The equilibrium constant of the NO –H O system estimated using ab initio calculations is too small to explain the retar-
dation of the reaction under our experimental conditions. Thus, we conclude that the NO –H O cannot contribute significantly to the
2
2
OH + NO
2
reaction in the atmosphere near the ground surface.
ꢀ
2005 Elsevier B.V. All rights reserved.
1
. Introduction
Recently, a new mechanism has been proposed for this
reaction. Namely, a reversible reaction also proceeds to
produce a significant amount of pernitrous acid, HOONO
[7].
The reaction of OH with NO is one of the most impor-
2
tant processes in the troposphere for two reasons: First,
this reaction produces nitric acid (HNO ) and contributes
to acidification of the Earth’s environment.
3
OH þ NO þ M $ HOONO þ M
ð2Þ
2
A recent spectroscopic experiment has demonstrated the
existence of HOONO [8]. This branching reaction has been
speculated to be one of the reasons for the uncertainties in
the recommended rate coefficients. Another reason is that
only few reliable measurements have yet been made on
the rate coefficient of this reaction under atmospheric con-
ditions [6].
OH þ NO
2
þ M ! HNO
3
þ M
ð1Þ
Second, this reaction is a main chain-termination step of
the oxidation cycles involving HO (=OH and HO ) and
x
2
NO (=NO and NO ) radicals. This is because both HO
x
x
2
and NO are removed via this reaction from the atmo-
x
sphere. In spite of numerous investigations on the kinetics
of this reaction at various pressures and temperatures (see
e.g. [1–6]), its rate coefficients are still controversial, and the
recommended values have large uncertainties.
In the present study, we have measured the rate coeffi-
cient of the OH + NO reaction under ambient conditions,
2
especially the dependence on humidity. To the best of our
knowledge, this is the first measurement of the humidity
dependence under ambient conditions.
*
Corresponding authors. Present address: Department of Applied
2. Experimental
Chemistry, Graduate School Engineering, Osaka Prefecture University,
1
7
-1 Gakuen-cho, Sakai, Osaka 599-8531, Japan (Y. Sadanaga). Fax: +81
2 254 9326 (Y. Sadanaga), +81 426 77 2837 (Y. Kajii).
E-mail addresses: sadanaga@chem.osakafu-u.ac.jp (Y. Sadanaga),
The rate constant was measured by the use of a com-
bined technique of laser flash photolysis/laser-induced fluo-
rescence. Fig. 1 shows a schematic diagram of the
kajii@atmchem.apchem.metro-u.ac.jp (Y. Kajii).
0
009-2614/$ - see front matter ꢀ 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.cplett.2005.12.026

Y. Sadanaga et al. / Chemical Physics Letters 419 (2006) 474–478
475
powermeter
photodiode
2
66-nm
cence detector (Model 42S, Thermo Electron). To study
to pump
flow tube
pulsed laser
photon counting
board
the humidity dependence of the OH + NO rate coefficient,
2
the humidity of zero-air was controlled. A part of the flow
was divided, bubbled through a trap of distilled water and
then returned to the main flow. The humidity of the NO2/
zero-air mixture, monitored by a sensor (Shinyei, THP-
CA9), was found to be constant within ± 3ꢀ RH (relative
zero air + NO2
A/D converter
PC
humidity) during the experiment. The H O partial pressure
2
during the experiment ranged from 4 to 29 hPa.
The NO /O /H O/zero-air mixture was introduced into
2
3
2
a flow tube. The pressure in the flow tube was approxi-
mately 990 hPa, as measured using a capacitance manom-
eter (Baratron model 626A, MKS). A fourth harmonic of
a Nd:YAG laser (Quanta-Ray INDI-40, Spectra-Physics)
with a low repetition rate (1 or 2 Hz) was irradiated to gen-
erate OH radicals.
3
08-nm
to vacuum pump
pulsed laser
photomultiplier tube
delay/pulse generator
fluorescence detection cell
photon counting unit
Fig. 1. Schematic drawing of the experimental apparatus.
1
O
3
þ hm ! O
1
2
þ Oð DÞ
ð5Þ
ð6Þ
experimental apparatus, described in detail in Refs. [9,10].
All the experiments were conducted at 298 K.
Oð DÞ þ H O ! 2OH
2
The sample of NO/N gas (5.16 ppmv, Nippon Sanso,
ppmv: parts per million by volume) was diluted by zero-
air, and the total NO/air flow was controlled to be
2
The OH radical reacted with NO in the flow tube, and the
concentration of OH declined after the irradiation of the
2
266-nm laser pulse. The decay of the OH concentration
5
00 sccm (standard cubic centimeters per minute). The
after the 266-nm laser pulse was measured by the time-re-
solved LIF technique. OH was excited at 308 nm using a
tunable frequency-doubled dye laser (Scanmate, Lambda
NO concentrations of the NO/air mixture were controlled
to range from 1 to 5 ppmv. The NO/air gas was mixed with
O /zero-air mixture (ꢁ250 ppmv of O ) to generate NO .
3
3
2
Physik) pumped by a frequency-doubled Nd:YVO laser
4
The flow rate of the mixture ranged up to 20 sccm, depend-
ing on the NO concentration in the NO/air mixture. Zero-
air was generated by passing compressed ambient air
through a hot Pt oven (623 K) and purafil-charcoal filters
to remove most of the OH reaction partners. In zero-air,
with a repetition rate of 10 kHz (YHP40-532Q, Spectra-
Physics). The resonant fluorescence was detected using a
dynode-gated photomultiplier tube (R2256P, Hamamatsu).
The interval of pulse trains of the 308-nm laser (corre-
sponding to 100 ls) was used as a clock for the measure-
ment of the OH decay rates.
the concentrations of CO, NO and hydrocarbons were less
x
than 10 ppbv (parts per billion by volume), 50 and 10 pptv
parts per trillion by volume), respectively. O was pro-
3
The NO concentration in the flow tube was measured
2
(
using the LIF technique [11,12]. Since NO2 sensitivity
decreases with humidity, due to the fast quenching of the
excited NO by H O, the NO measurement system was
duced in the photolysis of O in zero-air using a low-pres-
2
sure mercury lamp (SP-5-2H, Sen light).
2
2
2
The reaction time of NO with O was controlled to con-
3
calibrated, including consideration of H O quenching [11].
2
vert NO to NO entirely, and to avoid generations of NO
2
3
The first-order decay rate of the OH radical was mea-
and N O via subsequent reactions.
2
5
sured by varying the NO concentrations under the same
2
humidity conditions. Fig. 2 shows an example of the mea-
NO þ O ! NO þ O
ð3Þ
ð4Þ
2
3
3
2
sured OH decay profile by the reaction of OH with NO in
2
NO
3
þ NO
2
þ M $ N
2
O
5
þ M
logarithmic scale. The time series of OH signals shows two
decay components. The slower decay reflects the
Typically, the reaction time was set to 3 s. These concentra-
tions were estimated by box model calculations, which indi-
cated that the concentrations of N O and NO were 2 and
OH + NO reaction, OH diffusion and turbulence in the
2
flow tube. The fast decay can be attributed to rapid diffu-
sion of OH radicals by the ‘local’ 266-nm laser irradiation,
where ‘local’ means that the diameter of the 266-nm laser
beam, 10 mm, is smaller than the inner diameter of the flow
tube, 40 mm. In other words, the OH radical is first gener-
ated within the 10 mm center circuit of the flow tube by the
laser irradiation and then OH diffuses toward the inner
wall of the flow tube. When the 266 nm laser was irradiated
ꢁ15 mm upside of the radial center of the flow tube, no fast
decay was observed but the rise of the LIF signal slowed
down (corresponding to the timescale of the fast decay).
This indicates that no OH radical is first generated at the
2
5
3
5
orders of magnitude smaller than that of NO , respec-
2
tively. Therefore, the interferences of NO and N O are
3
2
5
negligible, in comparison with the rate constants for the
reactions of OH with these species. In addition, the actual
NO and N O concentrations should be even smaller than
3
2
5
these estimates, because the wall loss of NO was ignored
in these calculations. Similar results were obtained for the
other concentrations of initial NO (1–5 ppmv).
3
The NO /O mixture was diluted with a large flow
2
ꢀ
3
1
(
ꢁ25 L min ) of zero-air. Residual NO was found to be
negligible by a measurement using an O chemilumines-
3

476
Y. Sadanaga et al. / Chemical Physics Letters 419 (2006) 474–478
1000
40
3
2
1
0
0
0
1
00
1
0
0
1
0
.0
0.5
1.0
1.5
2.0
2.5 3.0x1012
0.00
0.10
0.20
0.30
-
3
[
NO ] / molecules cm
Time / s
2
0
Fig. 2. Examples of the measured OH decay profile by the reaction of OH
Fig. 3. Plots of measured first-order decay rate ðkNO Þ versus NO
2
2
11
with NO
2
3
in logarithmic scale. Black line: [NO
2
] = 5.61 · 10 mole-
concentrations at 745 Torr total air pressure and at 298 K. Solid line
shows linear regression lines. Its slope gives the second-order decay rate
ꢀ
12
ꢀ3
cules cm . Gray line: [NO
2
] = 1.68 · 10 molecules cm
.
ꢀ11
3
ꢀ1 ꢀ1
constant of (1.21 ± 0.03) · 10
cm molecule s . The error bars,
representing ± 2r, for the number densities fall within the circles except
center of the flow tube (i.e., inlet of the fluorescence detec-
tion cell) and then OH diffuses toward the inlet of the fluo-
rescence detection cell. A previous measurement of the rate
coefficient of the OH + CO reaction demonstrated the
accuracy of this instrument [9]. The decay profiles were
analyzed by the following fitting equation, in which the
slower decay component was taken into account.
in the extreme end.
1
.6x10-11
1.2
½
OHꢂ ¼ ½OHꢂ expðꢀkN0 O2 tÞ
ð7Þ
0
0.8
0.4
0
The measured decay rates in the presence of NO ðk
Þ
2
NO
2
and in zero-air (k ) were measured alternately to confirm
zero
the zero point of the decay rate.
k
0.0
3
. Results and discussion
0
5
10
15
20
25
30
3
.1. Humidity dependence of the rate coefficient
Kinetic measurements were conducted under pseudo-
H2O partial pressure / hPa
Fig. 4. The measured second-order rate coefficients for the OH + NO
2
reaction as a function of the partial pressure of water vapor at 298 K. The
total pressure in the reaction tube was measured to be 745 Torr in air
condition. Error bars show ± 2r. A solid line is a just guide to the eyes.
first-order conditions. The initial OH concentrations were
9
ꢀ3
ꢁ
10 radicals cm . The concentrations of NO ranged
2
from 3 · 10 to 3 · 10 molecules cm . The contribution
1
1
12
ꢀ3
of N O via the following equilibrium can be neglected
because of its small equilibrium constant (2.8 · 10
2
4
ꢀ
19
mol-
where r stands for standard error, arises from the measure-
ꢀ
1
3
ecules cm at 298 K [13]).
ments of the OH decay rate (± 15ꢀ) [9] and the NO con-
2
centration (± 5ꢀ). Considering these uncertainties, we
conclude that the observed decrease in the rate coefficient
is significant.
2
NO
2
$ N
2
O
4
ð8Þ
Fig. 3 shows an example of the measured first-order rate
coefficients for the reaction of OH with NO as a function
2
of the NO number density. The solid line in Fig. 2 shows
3.2. Possibility of OH–H O and NO –H O complexes
2
2
2
2
the weighted linear least-squares fit to the data. The slope
of the regression line gives the second-order rate coefficient.
Fig. 4 represents the humidity dependence of the second-
formation as a cause of the slowdown of the reaction
Our results indicate that the rate coefficient for the
OH + NO2 reaction decreases in the presence of water
vapor. One may suspect that molecular complexes, such
as OH–H O and NO –H O play some role in the slow-
order rate coefficient. At the H O partial pressure ðPH O
Þ
2
2
of 3.7 hPa, the rate coefficient was found to be
ꢀ11
3
ꢀ1 ꢀ1
(
1.40 ± 0.10) · 10
cm molecule
s , which is in excel-
2
2
2
lent agreement with the IUPAC1997 recommended value
down of the reaction. The structures of four relevant
complexes, shown in Fig. 5, were optimized at the B3LYP/
aug-cc-pVTZ level using the GAUSSIAN03 program [15]. In
[
14]. However, this coefficient decreased with the H O con-
2
ꢀ11
3
ꢀ1 ꢀ1
centration to (1.15 ± 0.02) · 10
2
cm molecule
s
at
9.1 hPa. The main uncertainty, ± 2r, in this coefficient,
the most stable structure of OH–H O, the OH radical acts
2

Y. Sadanaga et al. / Chemical Physics Letters 419 (2006) 474–478
477
2 2 2 3 3 2
Fig. 5. Optimized geometries of: (a) OH–H O, (b) NO –H O, and (c) HNO and HNO (H O) at the B3LYP/aug-cc-pVTZ level. Geometric parameters
are given in angstroms and degrees.
as a proton donor, in agreement with a recent experimental
observation [16]. The calculated binding energy is
temperature, can be evaluated using the partition functions
of the complex and monomers within the rigid rotator
approximation. The rotational constants, harmonic fre-
quencies and binding energies at the B3LYP/aug-cc-pVTZ
level were used for the calculations of the partition func-
tions. The translational and rotational partition functions
were computed using textbook formulas [18]. For calcula-
tion of the vibrational partition function for the complex,
the levels below the dissociation energy were included
straightforwardly, and thus, a total of 29 states were used
[19]. Total pressure, P, and temperature, T, were set to
ꢀ
3.2 kcal/mol with both counterpoise correction (CPC)
17] for the basis set superposition error (BSSE) and
[
zero-point correction (ZPC). On the other hand, the bind-
ing energy of NO –H O, where the H O molecule is bound
2
2
2
to N by the O atom, is ꢀ0.50 kcal/mol with CPC and ZPC.
Neither reaction, OH–H O + NO ! HNO (H O) or
2
2
3
2
OH + NO –H O !HNO (H O), indicates a potential bar-
2
2
3
2
rier, suggesting that the appearance of the barrier is not
responsible for the slow reaction in the presence of the
water. So, the water may interfere with the collision
990 hPa and 298 K, respectively. The calculated K was
P
ꢀ
5
ꢀ1
between OH and NO resulting in the reduced reaction
1.52 · 10 bar , while the ratio P
/PNO2 was
values were
2
NO –H2O
2
ꢀ8
ꢀ7
cross section, though further effort at the molecular level
is necessary to unveil the reason for the retardation of
the reaction by the water.
5.62 · 10 and 4.42 · 10 when the P
H O
2
3.7 and 29.1 hPa, respectively. The values of P, T and
/P employed are close to the standard values for the
P
H O
2
The equilibrium constant, K , or the abundance of the
mid-latitudes [20]. Since the intermolecular harmonic fre-
quencies at the present level are known to be usually over-
estimated, giving rise to lower density of states, the
P
complexes, is informative to assess their contribution to
the reaction under our experimental conditions. It is diffi-
cult at present to evaluate the K for OH + H O M OH–
calculated K may be slightly below the actual value [19].
P
2
P
H O, mainly because there is a low-lying electronic excited
2
Nevertheless, the order of the magnitude of the PNO2–
state [16]. Thus, we focused on the abundance of the NO2–
H O complex relative to the free NO molecule. The ratio
H2O/PNO2 based on our K is deemed to be so small that
P
the NO –H O cannot contribute significantly to the
2
2
2
2
of the partial pressure of NO –H O, P
, over that of
OH + NO reaction under our conditions, namely, in the
2
2
NO –H2O
2
2
free NO , PNO2, was estimated using the following formula,
atmosphere near the ground surface.
2
ꢀ
ꢁ
PNO2 ꢀH2O
ꢂ
ꢃ
P
P
PH2O
NO2ꢀH2O
PNO2
Acknowledgments
¼
ꢀ
ꢁ
¼ K
P
P
;
ð9Þ
PNO2
P
P
The authors are grateful to our colleagues (A. Yoshino,
K. Watanabe, A. Yoshioka, N. Akiyama, A. Nishiyama
and J. Matsumoto) for their assistance in the kNO2 mea-
surements. This work was supported financially by Core
where K is the equilibrium constant of NO –H O, while
and P are the partial pressures of water and the atmo-
spheric pressure, respectively. K , which is a function of
P
2
2
P
H O
2
P

478
Y. Sadanaga et al. / Chemical Physics Letters 419 (2006) 474–478
Research for Evolutional Science and Technology
CREST) from Japan Science and Technology Agency,
and by the Ministry of Education, Science, Sports and Cul-
ture of Japan (Grant-in-Aid No. 15201004).
[9] Y. Sadanaga, A. Yoshino, K. Watanabe, A. Yoshioka, Y. Wakazono,
Y. Kanaya, Y. Kajii, Rev. Sci. Instrum. 75 (2004) 2648.
10] Y. Sadanaga, A. Yoshino, S. Kato, A. Yoshioka, K. Watanabe, Y.
Miyakawa, I. Hayashi, M. Ichikawa, J. Matsumoto, A. Nishiyama,
N. Akiyama, Y. Kanaya, Y. Kajii, Geophys. Res. Lett. 31 (2004)
L08102, doi:10.1029/2004GL019661.
(
[
References
[11] J. Matsumoto, J. Hirokawa, H. Akimoto, Y. Kajii, Atmos. Environ.
5 (2001) 2803.
3
[
[
[
12] J. Matsumoto, Y. Kajii, Atmos. Environ. 37 (2003) 4847.
13] S.P. Sander et al., JPL Publ. 02–25 2002.
14] R. Atkinson, D.L. Baulch, R.A. Cox, R.F. Hampson Jr., J.A. Kerr,
M.J. Rossi, J. Troe, J. Phys. Chem. Ref. Data 26 (1997) 1329.
15] M.J. Frisch et al., Gaussian Suite GAUSSIAN 03, Revision B.03,
Gaussian Inc., Pittsburgh, PA, 2003.
[
[
[
[
1] R. Forster, M. Frost, D. Fulle, H.F. Hamann, H. Hippler, A.
Schlepegrell, J. Troe, J. Chem. Phys. 103 (1995) 2949.
2] N.M. Donahue, M.K. Dubey, R. Mohrschladt, K.L. Demerjian, J.G.
Anderson, J. Geophys. Res. 102 (1997) 6159.
[
[
3] D. Fulle, H.F. Hamann, H. Hippler, J. Troe, J. Chem. Phys. 108
(1998) 5391.
16] Y. Ohshima, K. Sato, Y. Sumiyoshi, Y. Endo, J. Am. Chem. Soc. 127
4] T.J. Dransfield, K.K. Perkins, N.M. Donahue, J.G. Anderson,
M.M. Sprengnether, K.L. Demerjian, Geophys. Res. Lett. 26
(
(
2005) 1108.
[
[
17] S.F. Boyds, F. Bernardi, Mol. Phys. 19 (1970) 553.
18] D.A. McQuarrie, J.D. Simon, Physical Chemistry. A Molecular
Approach, University Science Books, Sausalito, CA, 1997.
19] A. Sabu, S. Kondo, R. Saito, Y. Kasai, K. Hashimoto, J. Phys.
Chem. A 109 (2005) 1836.
1999) 687.
5] S.S. Brown, R.K. Talukdar, A.R. Ravishankara, Chem. Phys. Lett.
99 (1999) 277.
6] L. D’Ottone, P. Campuzano-Jost, D. Bauer, A.J. Hynes, J. Phys.
Chem. A 105 (2001) 10538.
[
[
2
[
[
20] G. Brasseur, S. Solomon, Aeronomy of the Middle Atmosphere,
second edn., D. Reidel, Dordrecht, The Netherlands, 1986.
[
[
7] D.M. Golden, G.P. Smith, J. Phys. Chem. A 104 (2000) 3991.
8] S.A. Nizkorodov, P.O. Wennberg, J. Phys. Chem. A 106 (2002) 855.