An experimental and theoretical study of the reactions OIO + NO and OIO + OH

Article abstract of DOI:10.1021/jp055364y

The kinetics of the reaction OIO + NO were studied by pulsed laser photolysis/time-resolved 'cavity ring-down spectroscopy, yielding k(235-320 K) = 7.6_+4.0-3.1 × 10^-13 exp[(607±128)/T] cm ³ molecule^-1 s^-1. Quantum calculations on the OIO + NO potential-energy surface show that the reactants form a weakly bound OIONO intermediate, which then dissociates to the products IO + NO₂. Rice-Ramsberger-Kassel-Markus (RRKM) calculations on this surface are in good accord with the experimental result. The most stable potential product, IONO₂, cannot form because of the significant rearrangement of OIONO that would be required. The reaction OIO + OH was then investigated by quantum calculations of the relevant stationary points on its potential-energy surface. The very stable HOIO₂ molecule can form by direct recombination, but the bimolecular reaction channels to HO₂ + IO and HOI + O ₂ are closed because of significant energy barriers. RRKM calculations of the HOIO₂ recombination rate coefficient yield k _rec.0 = 1.5 × 10^-27 (T/300 K)^-3.93 cm⁶ molecule^-2 s^-1, k_rec.∞ = 5.5 × 10^-10 exp(46/T) cm³ molecule^-1 s^-1, and F_c = 0.30. The rate coefficients of both reactions are fast enough around 290 K and 1 atm pressure for these reactions to play a potentially important role in the gas phase and aerosol chemistry in the marine boundary layer of the atmosphere.

Full text of DOI:10.1021/jp055364y

J. Phys. Chem. A 2006, 110, 93-100
93
An Experimental and Theoretical Study of the Reactions OIO + NO and OIO + OH
J. M. C. Plane,*,† D. M. Joseph,† B. J. Allan,† S. H. Ashworth,‡ and J. S. Francisco§
School of EnVironmental Sciences, UniVersity of East Anglia, Norwich NR4 7TJ, U.K., School of Chemical
Sciences and Pharmacy, UniVersity of East Anglia, Norwich NR4 7TJ, U.K., and Department of Chemistry,
Purdue UniVersity, West Lafayette, Indiana 47907-2084
ReceiVed: September 21, 2005; In Final Form: NoVember 8, 2005
The kinetics of the reaction OIO + NO were studied by pulsed laser photolysis/time-resolved cavity ring-
down spectroscopy, yielding k(235-320 K) ) 7.6+4.0
× 10-13 exp[(607(128)/T] cm3 molecule-1 s-1.
-3.1
Quantum calculations on the OIO + NO potential-energy surface show that the reactants form a weakly
bound OIONO intermediate, which then dissociates to the products IO + NO2. Rice-Ramsberger-Kassel-
Markus (RRKM) calculations on this surface are in good accord with the experimental result. The most
stable potential product, IONO2, cannot form because of the significant rearrangement of OIONO that would
be required. The reaction OIO + OH was then investigated by quantum calculations of the relevant stationary
points on its potential-energy surface. The very stable HOIO2 molecule can form by direct recombination,
but the bimolecular reaction channels to HO2 + IO and HOI + O2 are closed because of significant energy
barriers. RRKM calculations of the HOIO2 recombination rate coefficient yield krec,0 ) 1.5 × 10-27 (T/300
K)-3.93 cm6 molecule-2 s-1, krec,∞ ) 5.5 × 10-10 exp(46/T) cm3 molecule-1 s-1, and Fc ) 0.30. The rate
coefficients of both reactions are fast enough around 290 K and 1 atm pressure for these reactions to play a
potentially important role in the gas phase and aerosol chemistry in the marine boundary layer of the atmosphere.
Introduction
reaction 1 indicates that the lifetime of OIO during the day must
be less than about 140 s.
Observations of the iodine oxide radicals IO1-3 and OIO3,4
in the marine boundary layer have established the importance
of iodine chemistry in the lower troposphere. Iodine can have
several major impacts, including the destruction of ozone,5-7
the activation of chlorine and bromine from sea salt,7,8 the
removal of nitrogen oxides in semi-polluted air masses,6 and
the formation of ultrafine iodine oxide aerosol.9-11
The major sources of iodine in the marine boundary layer
appears to be iodocarbons of biogenic origin such as CH3I and
CH2I2, which evade from the ocean,12 and I2, which is released
from macroalgae exposed at low tide.3 Following photolysis of
these species to yield atomic I, the iodine oxide radical (IO) is
produced by reaction with O3. The OIO radical can form either
through the self-reaction of IO
There appear to be two possible explanations for this absence
of OIO during daytime. First, it could photolyze following
absorption in the strong bands between about 470 and 610 nm.
Although there is some experimental evidence for photolysis,16
we have recently shown that absorption leads to internal
conversion into the highly vibrationally excited ground state
rather than dissociation, so the quantum yield for photolysis is
probably much less than 10% at 562 nm.14 Nevertheless, a small
probability of photolysis, integrated over the absorption bands,
could still account for the removal of OIO during daytime.17
A second possibility is that OIO is removed during daytime
by reaction with radicals that have marked diurnal cycles with
daytime maxima, such as NO, OH, and IO. We have shown in
a recent modeling study17 that the recombination of IO with
OIO to make I2O3 is probably a major pathway for the removal
of OIO in environments where there are significant daytime IO
concentrations, in particular, when there are large emissions of
I2 from macroalgae. In contrast, the reactions of OIO with NO
and OH may be important over the open ocean, where the source
flux of iodine precursors is lower. These reactions are the subject
of this paper.
IO + IO f OIO + I
(1)
which has a yield of around 30-50%,13,14 or through the
reaction15
BrO + IO f OIO + Br
(2)
Observations of OIO by the technique of differential optical
absorption spectroscopy3,4 showed that during daytime the OIO
mixing ratio was consistently below the instrumental detection
limit of ∼3 parts per trillion (ppt, where 1 ppt ) 2.5 × 107
molecule cm-3 at 290 K and 1 atm pressure), using a new
absolute cross section for OIO.14 Since the daytime IO mixing
ratio tends to be around 5 ppt,1-3 a steady-state calculation using
The reaction between OIO and NO could proceed via several
channels
OIO + NO f IO + NO2
(+M) f IONO2
f I + NO3
∆H0 ) -70 kJ mol-1 (3a)
∆H0 ) -166 kJ mol-1
∆H0 ) -50 kJ mol-1
(3b)
(3c)
* To whom correspondence should be addressed. E-mail: j.plane@
uea.ac.uk.
† School of Environmental Sciences, University of East Anglia.
‡ School of Chemical Sciences and Pharmacy, University of East Anglia.
§ Purdue University.
f INO + O2(1∆g)
∆H0 ) -10 kJ mol-1
(3d)
10.1021/jp055364y CCC: $33.50 © 2006 American Chemical Society
Published on Web 12/13/2005

94 J. Phys. Chem. A, Vol. 110, No. 1, 2006
Plane et al.
where the reaction enthalpy changes are calculated from ab initio
quantum theory described later in the paper. A fifth channel,
OIO + NO f INO2 + O(1D), is highly endothermic (∆H0 )
191 kJ mol-1) and so is excluded from further consideration.
Note that if the reaction proceeds via channel 3a, this leads to
a null O3-depleting cycle with reaction 1, because the NO2
produced will photolyze to generate O and hence O3, thereby
canceling out the O3 removed by the I atom produced in reaction
1. However, k3a would have to be significantly faster than the
rate coefficients for the analogous reactions of BrO and ClO18,19
in order to account for the observed upper limit to the lifetime
of OIO. For instance, in a very clean marine environment with
[NO] less than 30 ppt, k3a would need to be greater than ∼1 ×
10-11 cm3 molecule-1 s-1. Although recombination to produce
IONO2 is the most exothermic channel of reaction 3, this
pathway involves significant rearrangement.
to protect the cavity mirrors (RMI, Boulder: reflectivity
nominally greater than 0.999 at 565 nm). The probe beam was
a Nd:YAG-pumped dye laser (Sirah Cobra) operating with
rhodamine 6G dye, which provided a spectral resolution better
than 0.2 cm-1 (∼0.006 nm) in the 570-590-nm region. This
beam was steered into the CRD cell, after using a wedged quartz
beam splitter to pick off portions that were directed into an I2
fluorescence cell and a Coherent Wavemaster for wavelength
calibration. In the cell there was a 30-cm overlap between the
excimer and dye laser beams. The laser radiation leaking out
of the ring-down cavity was passed through a 485-nm cut-on
filter and detected with a photomultiplier tube. The data were
displayed on a LeCroy 9361C digital oscilloscope (300 MHz
bandwidth, 2.5 GS/s) and captured on computer by fitting each
ring-down signal to an exponential decay. The fluorescence from
the I2 cell and the wavelength measured by the Wavemaster
were recorded simultaneously. The ring-down time of the empty
cavity was also recorded so that the mirror reflectivities could
be used to normalize the raw spectra. The ring-down time for
the empty cavity was typically τ0 ) 750 ns, indicating that the
mirror reflectivity was actually about 0.994%. This was routinely
recorded at the start of a set of experiments, and the standard
deviation of τ0 for an average of 10 laser shots was consistently
less than 1%.
The reaction between OIO and OH could also have several
channels
OIO + OH (+M) f HOIO2
f IO + HO2
∆H0 ) -204 kJ mol-1
(4a)
∆H0 ) -25 kJ mol-1
(4b)
f HOI + O2(1∆g)
The absorption spectra of OIO were contaminated with I2
absorption to some degree, even though the gas residence time
was kept short (0.7-1.4 s at 20 and 70 Torr, respectively), and
the excimer laser was triggered at only 3 Hz to avoid the build-
up of I2 from the fast recombination of I atoms in this
photochemical system.14 OIO was also monitored at 567.808
nm, close to the peak in one of the OIO vibrational bands14 but
at a trough in the I2 spectrum, to further minimize interference
from I2. The time-resolved variation of the OIO absorption was
recorded using a scanned boxcar gate to trigger the dye laser.
The baseline was recorded at long times after the excimer pulse
(up to 3 ms) and then the absorption monitored as a function
of time delay.
∆H0 ) -128 kJ mol-1 (4c)
where again the reaction enthalpy changes are calculated from
ab initio quantum theory described below. Unlike reaction 3,
the recombination pathway in reaction 4 involves simple
attachment of the oxygen atom of OH to the iodine atom of
OIO, to yield the most stable isomer of HOIO2, whereas a recent
theoretical study of the IO + HO2 and HOI + O2 potential
surfaces has shown that these contain significant energy
barriers.20 Channel 4a is particularly interesting because this
reaction oxidizes the iodine atom to the +5 oxidation state,
forming the iodate ion. This ion is often the major form of iodine
in marine aerosols (where iodine is heavily enriched compared
to its abundance in seawater), but its source is unexplained.21
Formation of HOIO2 in the gas phase, followed by uptake on
sea-salt aerosol, might provide an important route.
In this paper we will describe an experimental study of
reaction 3 using a laser flash photolysis/time-resolved cavity
ring-down (CRD) technique,16 where k3 was measured over a
range of temperature and pressure. The experimental results are
then modeled using ab initio calculations on the relevant
potential energy surfaces, combined with Rice-Ramsberger-
Kassel-Markus (RRKM) theory. An attempt to study reaction
4 was unsuccessful because the photochemical generation of
OIO also produced significant concentrations of I2 in the reactor,
and the I2 + OH reaction then dominated the removal of OH.
Nevertheless, theory is used to predict the reaction pathways
and to calculate a rate coefficient for reaction 4. Finally, the
atmospheric relevance of both reactions is considered.
For the study of reaction 4, H2 or H2O was added to the CF3I/
N2O mix flowing into the CRD cell. OH was then produced by
the rapid reaction of O(1D), produced from the 193 nm
photolysis of N2O, with H2 or H2O, and also by the direct
photolysis of H2O at 193 nm. OH was then probed at 281.904
nm [OH(A2Σ - X2Π), (1,0)] using a second dye laser
orthogonal to the excimer and CRD laser beams. The time-
resolved nonresonant laser induced fluorescence (LIF) at 307-
312 nm [OH(A2Σ - X2Π), (0,0)] was detected by a photomul-
tiplier after passing through an interference filter at 307 ( 5
nm and captured by a gated integrator.
The central section of the CRD cell, containing the reaction
volume where the excimer and CRD lasers overlapped, was
enclosed in an insulated jacket through which ethanol was
pumped through a temperature-controlled bath. This enabled
the temperature of the gases in the reaction volume to be varied
between 235 and 320 K, as measured by a chromel-alumel
thermocouple that could be inserted into the center of the reactor.
Experimental Section
Materials. CF3I (99% pure, Fluorochem Ltd), N2 (99.9999%
pure, Air Products), H2 (99.9999% pure, Air Products), and N2O
(99.99% pure, Air Products) were used without further purifica-
tion. NO (99.9% pure, Air Products) was distilled from 77 K
to remove traces of NO2. MilliQ H2O was freeze-pump-
thawed for several cycles in a glass handling line and H2O vapor
then made up to a known ratio in about 760 Torr of bath gas in
a glass bulb.
The experimental system for producing OIO and monitoring
its concentration by CRD spectroscopy has been described
previously.14,16 IO was formed by reacting CF3I with O atoms
produced by the photolysis of N2O at 193 nm using an ArF
excimer laser (fluence in the cell ) (0.8-2.4) × 1016 photon
cm-2). The reagents were diluted with N2 gas, which flowed
into the ends of the 1.45 m long CRD cell and acted as curtains

Study of the Reactions OIO + NO and OIO + OH
J. Phys. Chem. A, Vol. 110, No. 1, 2006 95
Figure 1. Typical OIO time profiles measured by CRD during the
study of the reaction of OIO + NO. Experimental conditions: (a) [NO]
) 2.78 × 1014 molecule cm-3 and (b) [NO] ) 8.94 × 1014 molecule
cm-3; [CF3I] ) 1.34 × 1016 cm-3; [N2O] ) 9.45 × 1016 cm-3; [N2] )
5.34 × 1017 cm-3; T ) 300 K. The solid lines are model calculations,
which include a value for k3 of 6.0 × 10-12 cm3 molecule-1 s-1 obtained
by fitting a single exponential to the OIO decay an appropriate interval
after the peak (see text). The concentration of OIO is derived from the
model assuming a branching ratio for OIO formation from the IO self-
reaction of 0.32.
Experimental Results
Figure 1 illustrates time-resolved profiles of the OIO con-
centration at low and high concentrations of NO. The OIO
absorbance measured by CRD was converted into absolute OIO
concentration using our recently determined absorption cross
section of (1.5 ( 0.18) × 10-17 cm2 at 567.93 nm.14 Examples
of OIO spectral scans between 558 and 578 nm using this CRD
system are given in two previous papers.14,16 Sufficiently high
levels of IO were produced to ensure that OIO was formed
essentially within 100 µs. In all experiments where NO was
added, the removal of OIO was then dominated by reaction 3
and could be described at sufficiently long times after the peak
by the pseudo-first-order decay coefficient
Figure 2. Plots of k′ vs [NO]: (a) measurements at 300 K and pressures
of 20, 40, and 72 Torr, where the data-sets at 40 and 72 Torr have
been offset for clarity; (b) measurements at 40 Torr and 235, 252, and
323 K. The solid lines are linear least-squares fits through each set of
data.
k′ ) kdiff + k3 [NO]
(I)
where kdiff includes diffusion of OIO out of the volume sampled
by the dye laser and other removal processes (including reaction
of OIO with the small amount of NO generated by the excimer
laser through the reaction O(1D) + N2O f 2NO). In these
experiments, kdiff varied between 1800 and 5000 s-1, depending
on the excimer laser fluence and reagent concentrations. This
ensured that k3[NO] . kdiff. As expected, the decays were well
fitted, after an appropriate time delay (20-50 µs after the peak
of the OIO profile), by a single exponential form yielding k′.
Figure 2 shows plots of k′ vs [NO], along with linear regression
fits to the data. The slopes of these plots yield k3. Figure 2a
shows measurements of k3 at 300 K made at three different
pressures (20, 40, and 72 Torr), which demonstrate that k3 is
independent of pressure within the experimental uncertainty.
Figure 3b shows a selection of plots over the range of
temperatures studied, illustrating that k3 has a negative temper-
ature dependence. Table 1 lists the measurements of k3 as a
function of temperature and pressure.
Figure 3. Arrhenius plot for reaction 3. The points are measurements
of k3 with 2σ error bars. The solid line is a linear regression through
these points. The broken line is an RRKM fit to the data, using the
potential-energy surface in Figure 4.
profiles very satisfactorily. The modeled absolute [OIO] is
obtained by assuming a yield of 32% for OIO formation in
reaction 1.14 The use of the model confirms that fitting the
decays after an appropriate delay to obtain k′ is consistent with
the removal of OIO at long reaction times being dominated by
the large excess of NO in the reactor.
The model can also be used to examine the possible
interference of CF3 radicals in the measurement of k3. CF3 is
produced by the reaction of O with CF3I and, to a minor extent,
by photolysis of CF3I at 193 nm. The predicted [CF3] reaches
The solid lines fitted through the time profiles of [OIO] in
Figure 1 are the predictions of a model that includes 38 reactions
describing the chemistry in the reactor.14 The model inputs are
the temperature, pressure, [CF3I], [N2O], [NO], and the excimer
laser fluence measured in the reactor. By use of k3 derived
above, the modeled OIO profile describes the experimental

96 J. Phys. Chem. A, Vol. 110, No. 1, 2006
Plane et al.
TABLE 1: Measured Rate Coefficients (2σ Uncertainties)
for the Reaction OIO + NO as a Function of Temperature
temperature
(K)
rate coefficient k3
pressure
(Torr)
(10-12 cm3 molecule s-1
)
235
252
258
262
295
296
296
300
300
300
323
10.2 ( 1.4
8.0 ( 1.3
7.4 ( 0.9
6.8 ( 1.3
6.9 ( 1.8
5.3 ( 0.3
4.6 ( 0.7
6.0 ( 0.6
6.4 ( 0.9
6.6 ( 0.9
4.0 ( 0.3
40
40
40
40
40
40
40
20
40
72
40
a maximum of ∼8 × 1013 cm-3 about 20 µs after the excimer
pulse. We have shown previously14 that the exothermic channels
of a possible reaction between CF3 and OIO would produce
CF3O + IO or CF3OI + O. In both cases, IO would result and
recycle back to OIO, thus not causing the rapid removal of OIO
that is observed. Furthermore, the recombination of CF3 and
NO is quite rapid (7 × 10-12 cm3 molecule-1 s-1 at 300 K and
40 torr N222). Thus, at the highest NO concentrations employed
here, CF3 would essentially be removed within 150 µs, which
would have distorted the single exponential decays of OIO if
CF3 had a significant impact on the temporal behavior of OIO.
The other major byproduct in this system of OIO production
is atomic I. We have shown recently14 that in the absence of a
reagent such as NO, the recombination of I with OIO appears
to control the rate of OIO removal. i.e., kdiff in equation I.
However, the model indicates that [I] should have increased
by only ∼25% at the maximum [NO] used in these experiments,
so this would have had a minor effect on k′ and the determi-
nation of k3. A final point is that the predicted [NO2] produced
in the reactor would have been about 1.2 × 1013 cm-3 at the
maximum added [NO] and thus much too small to cause any
measurable absorbance in the CRD signal at 567.8 nm.
Figure 3 is an Arrhenius plot for reaction 3. An unweighted
least-squares fit through the experimental data points yields (2σ
errors)
Figure 4. Geometries of stationary points on the OIO + NO potential-
energy surface. The geometries are listed in Z matrix format in Table
2, and their relative energies are illustrated in Figure 5.
consistent with the observed removal rate of OH, since k(OH
+ I2) ) 2.1 × 10-10 cm3 molecule-1 s-1 23
.
Quantum Calculations and Application of RRKM Theory.
Ab initio quantum calculations were carried out to explore the
reaction pathways of reactions 3 and 4. The hybrid density
functional/Hartree-Fock B3LYP method was employed from
within the Gaussian 03 suite of programs,24 combined with a
recently published basis set for I 25 and the standard 6-311+g-
(2d,p) triple-ú basis set for O, N, and H. Following geometry
optimizations of the relevant points on the potential-energy
surfaces and the determination of their corresponding vibrational
frequencies and (harmonic) zero-point energies, energies relative
to the reactants were obtained. Spin-orbit corrections of -17
and -5 kJ mol-1 were applied to the B3LYP energies of I and
IO, respectively. These were estimated by comparing the ab
initio and the experimental bond energies (at 0 K) of I2 and IO,
which are taken as 148.826 and 236.0 kJ mol-1 27
, respectively.
Figure 4 illustrates the geometries of the stationary points
on the resulting potential-energy surface of singlet spin multi-
plicity for the reaction OIO + NO; the geometries are described
in Z matrix notation in Table 2. Figure 5 shows the potential-
energy surface, and the ab initio rotational constants and
vibrational frequencies for the relevant stationary points and
asymptotes are listed in Table 3. For the reaction between OIO
and OH, the stationary point geometries and potential-energy
surface are shown in Figures 6 and 7, respectively. The Z matrix
geometries are listed in Table 4 and rotational constants and
vibrational frequencies in Table 5.
k3(235 - 320 K) )
7.6+4.0-3.1 × 10-13 exp[(607(128)/T] cm3 molecule-1 s-1
(II)
The average rate coefficient at room temperature is k3(298 K)
) (6.0 ( 1.1) × 10-12 cm3 molecule-1 s-1, where the quoted
uncertainty of (18% is estimated by combining in quadrature
a 10% systematic error (comprising the sum of the uncertainties
in the flow rates and total pressure) with a 15% standard
deviation from the average of 6 room-temperature measure-
ments.
When studying reaction 4, very rapid decays of OH (with
first-order decay rates between 8 000 and 16 000 s-1) were
observed when CF3I was added to the N2O/H2 mixture. Since
the maximum concentration of OIO was only ∼6 × 1012 cm-3
(Figure 1), the removal of OH was clearly not dominated by
reaction 4. In fact, we have shown elsewhere that I2 is produced
by the recombination of I atoms in this system, perhaps
facilitated by the iodine oxides acting as chaperones.14 A spectral
scan of the CRD laser in the 560-575-nm region, at a delay of
more than 1 ms after the excimer when all the OIO has reacted
away (Figure 1), enables the I2 concentration to be measured.
The concentration was typically (4-8) × 1013 cm-3, which is
Rate coefficients for the relevant channels of reactions 3 and
4 were then estimated from RRKM theory, employing a Master
Equation (ME) formalism with inverse Laplace transformation
(ILT).28 Reaction 3 is assumed to proceed via the formation of
an excited adduct (OIONO*), which can either dissociate back
to OIO + NO, be stabilized by collision with the third body
(N2) to form the OIONO intermediate, or dissociate to the
products IO + NO2. In the case of reaction 4, the excited adduct
(HOIO2*) can either dissociate back to OIO + OH or be
stabilized to HOIO2. For both reactions, the adduct energy was
divided into a contiguous set of grains (width 30 cm-1), each
containing a bundle of rovibrational states. Each grain was then
assigned a set of microcanonical rate coefficients for dissocia-
tion, which were determined using ILT to link them directly to
k
rec,∞, the high-pressure limiting recombination coefficient.28 ILT
is particularly useful here because krec,∞ can be estimated using

Study of the Reactions OIO + NO and OIO + OH
J. Phys. Chem. A, Vol. 110, No. 1, 2006 97
TABLE 2: Geometries of the Stationary Points on the OIO
+ NO Potential Surface, Expressed in Z Matrix Notation
bond
dihedral
molecule atom no.a N1 N2 N3 lengthb anglec angled
OIONO
TS1
I
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
O
O
N
O
I
O
O
N
O
I
O
O
N
O
N
O
O
O
I
1
1
3
4
1.83
2.01
1.51
1.15
2
1
3
105.5
111.9
109.4 -177.4
2
1
85.7
1
1
3
4
1.83
2.04
1.51
1.15
2
1
3
107.6
116.7
108.7
2
1
0.0
180.0
TS2
1
1
2
4
1.92
1.93
3.11
1.14
2
1
1
63.1
54.1
123.9
3
2
163.5
24.1
IONO2
1
1
1
4
1.20
1.20
1.45
2.03
2
3
1
131.3
110.5
117.3
2
2
180.0
0.0
a Atom number, see Figure 6. b Between atom and N1, in angstroms.
c Between atom, N1 and N2. d Between atom, N1, N2, and N3.
Figure 6. Geometries of stationary points on the OIO + OH potential-
energy surface. The geometries are listed in Z matrix format in Table
4, and their relative energies are illustrated in Figure 7.
Figure 7. Potential-energy surface for the OIO + OH reaction. The
geometries of the stationary points are illustrated in Figure 6.
Figure 5. Potential-energy surface for the OIO + NO reaction. The
geometries of the stationary points are illustrated in Figure 4.
to a typical value for N2 at 300 K of 250 cm-1 32
. The collision
long-range capture theory, assuming there are no energy barriers
for these radical-radical reactions. For reaction 3, this is
dominated by the dipole of OIO (3.89 D at the B3LYP/6-
311+g(2d,p) level of theory) inducing a dipole in NO (polar-
izability ) 1.7 × 10-24 cm3).29 The dipole-induced dipole
capture rate30 is then krec,∞ ) 2.7 × 10-10 exp(-46/T) cm3
molecule-1 s-1. For reaction 4, the dominant long-range force
is between the dipole of OIO and that of OH (1.77 D29). This
frequency between the adduct and N2 was calculated using an
intermolecular potential described by the parameters σ ) 3.5
Å and ꢀ/k ) 350 K. To use the ME to simulate irreversible
stabilization of the adduct, an absorbing boundary was set 50
kJ mol-1 below the energy of the reactants, so that collisional
energization from the boundary to the threshold was highly
improbable. The absorbing boundary is used to limit the size
of the matrixes involved in the RRKM calculation; for both
reactions, it was verified that the resulting rate coefficients were
essentially independent of the depth of the boundary when this
results in krec,∞ ) 5.5 × 10-10 exp(46/T) cm3 molecule-1 s-1
,
calculated using a recent implementation of long-range varia-
tional transition state theory.31
was raised by 20 kJ mol-1
.
The density of states of the adduct was calculated using a
combination of the Beyer-Swinehart algorithm for the vibra-
tional modes (without making a correction for anharmonicity)
and a classical densities of states treatment for the rotational
modes.32 The ME describes the evolution with time of the adduct
grain populations. The probability of collisional transfer between
grains was estimated using the exponential down model,32 where
the average energy for downward transitions, ∆E down, was set
Discussion
Reaction of OIO with NO. The OIO + NO potential-energy
surface is illustrated in Figure 5. It should be noted that
theoretical calculations on some sections of this surface have
been published recently,33,34 with which the present study is in
good agreement. The most exothermic potential product is

98 J. Phys. Chem. A, Vol. 110, No. 1, 2006
Plane et al.
TABLE 3: Calculated Vibrational Frequencies, Rotational Constants, and Energies of the Stationary Points and Asymptotes on
the OIO + NO Singlet Potential-Energy Surface
species
vibrational frequenciesa
rotational constantsb
potential energyc
OIO + NO
OIONO
OI-O-NO TS1
IO + NO2
IONO2
273, 809, 831 and 1955
18.31, 7.05, 5.09 and 51.41
9.36, 1.42, 1.29
8.72, 1.45, 1.24
9.84 and 241.3, 13.05, 12.38
12.36, 1.305, 1.180
13.85, 13.82, 6.918
16.89, 2.20, 1.94
0
-55
-28
-70
-166
-50
101
73, 152, 188, 235, 388, 477, 800, 848, 1811
-126, 117, 147, 226, 347, 457, 787, 823, 1787
648 and 762, 1382, 1672
114, 177, 366, 588, 728, 753, 826, 1311, 1703
254, 267, 805, 1103, 1112, 1126
170, 385, 407, 784, 853, 1649
570i, 54, 101, 150, 181,411,579, 716, 1870
240, 517, 1857 and 1629
I + NO3
IONO + O
ON-I-O2 TS2
INO + O2(1∆g)
7.14, 1.84, 1.45
86.77, 2.80, 2.71 and 43.28
254
-10
a In cm-1 b In GHz. c In kJ mol-1, including zero-point energy and spin-orbit coupling of I and IO (see text).
.
10.5 kJ mol-1 to be 17.5 kJ mol-1 below the reactant energy.
The RRKM calculation now reproduces both the absolute
magnitude of k3 and the small negative temperature dependence.
If the barrier height is not raised, then the only way to match
k3 is to reduce krec,∞. For instance, the experimental k3(240 K)
can be calculated by reducing krec,∞ by a factor of ∼20 from
the dipole-induced dipole capture frequency. However, there
are no obvious reasons for doing this; furthermore, the RRKM
calculation then predicts that k3 has almost no temperature
dependence, so that the rate constant is 70% higher than the
experimental value at 320 K. With the higher barrier, the
OIONO* energized adduct most likely dissociates back to OIO
and NO at these pressures, rather than being quenched to
OIONO or dissociating over the barrier to IO + NO2.
Figure 8 illustrates the calculated rate coefficients for the
OIONO association and IO + NO2 atom abstraction channels,
as a function of pressure at 300 K. This shows that even at
atmospheric pressure (760 Torr), OIONO formation is almost
2 orders of magnitude slower. In any case, the barrier to IO +
NO2 is only 40.5 kJ mol-1 above the OIONO minimum, so
that even if the adduct forms it will dissociate to IO + NO2
very rapidly. Figure 8 also shows that the direct atom abstraction
channel has a very small pressure dependence, in accord with
the experimental measurements. For the purpose of atmospheric
modeling, the recommended rate coefficient at 1 atm is
TABLE 4: Geometries of the Stationary Points on the OIO
+ OH Potential Surface, Expressed in Z Matrix Notation
bond
dihedral
molecule atom no.a N1 N2 N3 lengthb anglec angled
HOIO2
TS1
H
O
I
O
O
H
O
I
O
O
H
O
O
I
O
O
H
O
I
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3
3
0.97
1.96
1.79
1.79
1
2
2
109.3
101.9
101.9
1
1
55.8
-55.8
1
2
3
4
0.97
2.03
1.93
2.05
1
2
2
110.2
145.4
56.5
1
1
59.4
-43.8
TS2
1
2
3
4
0.97
2.03
1.83
1.81
1
2
2
121.5
82.3 -130.6
109.6
1
1
90.4
TS3
1
1
3
3
1.16
1.34
2.29
2.70
2
1
1
102.1
101.9
60.3
2
2
-23.9
-9.7
O
H
O
O
I
HOOIO
1
2
3
3
0.97
1.44
2.03
3.08
1
2
2
101.3
111.8
101.4
1
1
-98.1
-63.1
O
a Atom number, see Figure 6. b Between atom and N1, in angstroms.
c Between atom, N1 and N2. d Between atom, N1, N2 and N3.
k3(240 - 320 K) )
(1.1 ( 0.4) × 10-12 exp((542(130)/T) cm3 molecule-1 s-1
IONO2. However, this cannot form from the addition of OIO
to NO, because a cyclic transition state of very high energy
would be required. The formation of I + NO3 is prevented for
the same reason. The abstraction by NO of the iodine atom from
OIO is exothermic, even if spin is conserved and INO + O2-
(1∆g) is formed. However, this channel also has a substantial
energy barrier. The only reaction pathway at the temperature
range of the present study is therefore association of OIO and
NO to form a weakly bound OIONO adduct, which then
dissociates over a barrier to IO + NO2 (reaction 3a). This picture
is consistent with the observed kinetics: k3 is independent of
pressure over more than a 3-fold change in pressure, indicating
that recombination to OIONO or IONO2 is not important. The
small negative temperature dependence of k3 (eq II) is also
characteristic of a reaction that proceeds via a shallow complex.
When the RRKM theory described above was applied to
reaction 3, the single adjustable parameter was the height of
the barrier between the OIONO adduct and the products IO +
NO2 (Figure 4). The quantum calculations predict this to be 28
kJ mol-1 below the energy of the reactants OIO + NO, with
an uncertainty of (16 kJ mol-1 at the level of theory
employed.35 A satisfactory fit to the experimental data, as shown
in Figure 3, is obtained when the barrier height is raised by
(III)
where the uncertainties are consistent with the experimental
uncertainty (eq II).
Finally, it is interesting to compare the rate coefficient of
(6.0 ( 1.1) × 10-12 cm3 molecule-1 s-1 measured here for OIO
+ NO at 300 K, with rate coefficients of 3.4 × 10-13 and 1.8
× 10-12 cm3 molecule-1 s-1 at 300 K for the analogous reactions
for OClO19 and OBrO,18 respectively. Thus, the halogen dioxide
+ NO reactions become significantly faster down Group VII,
in contrast to the reactions of the halogen monoxides with NO,
which all have rate coefficients of about 2 × 10-11 cm3
molecule-1 s-1 at 300 K.36
Reaction of OIO with OH. The potential-energy surface for
this reaction (Figure 7) shows that the formation of HOIO2 can
occur by the simple addition of OH to OIO (reaction 4a),
promoted by the substantial long-range attraction between the
large dipole moments of OH and OIO (see above). The predicted
geometry of HOIO2, and its stability with respect to these
products, are in good accord with another recent theoretical
study which focused on the reaction pathways of the HO2 +
IO reaction.20 Experiments37,38 have shown that the rate coef-
ficient for this reaction has a negative activation energy similar

Study of the Reactions OIO + NO and OIO + OH
J. Phys. Chem. A, Vol. 110, No. 1, 2006 99
TABLE 5: Calculated Vibrational Frequencies, Rotational Constants, and Energies of the Stationary Points and Asymptotes on
the OIO + OH Singlet Potential-Energy Surface
species
vibrational frequenciesa
rotational constantsb
potential energyc
OIO + OH
273, 809, 831 and 3707
18.31, 7.05, 5.09 and 558.7
5.69, 5.36, 3.63
9.14, 2.77, 2.23
7.59, 3.32, 2.90
9.84 and 619.1, 33.68, 31.94
9.30, 4.02, 2.92
0
-204
-68
46
-25
89
HOIO2
74, 258, 274, 303, 589, 875, 905, 977, 3734
77, 212, 264, 378, 519, 796, 866, 1387, 3736
436i,113, 264, 280, 327, 762, 798, 856, 3720
648 and 1167, 1436, 3604
675i, 142, 231, 282, 522, 611, 737, 935, 3800
603, 1084, 3803 and 1629
HOOIO
HO-O-IO transition state
IO + HO2
HO-I-O2 transition state
HOI + O2(1∆g)
623.9, 8.18, 8.08 and 43.28
-129
a In cm-1 b In GHz. c In kJ mol-1, including zero-point energy and spin-orbit coupling of I and IO (see text).
.
modeling, the predicted rate coefficient at 1 atm pressure is
k4(240 - 320 K) )
2.2 × 10-10 exp(243/T) cm3 molecule-1 s-1 (V)
We have considered three sources of uncertainty in this RRKM
prediction: an uncertainty of (16 kJ mol-1 35 in the calculated
binding energy of OIO and OH, a (20% uncertainty in the
lowest vibrational frequency of HOIO2 (Table 4), and an
uncertainty of (100 cm-1 in ∆E down. By combination in
quadrature the effects of these uncertainties on the RRKM
calculation, the estimated uncertainty of k4 in eq V varies from
(11% at 240 K to (22% at 320 K.
The analogous reaction OH + OClO has been studied in a
fast-flow tube.39 A rate coefficient of 4.5 × 10-13 exp(804/T)
cm3 molecule-1 s-1 was measured, and it was shown that the
major product channel (at about 1 Torr pressure) was HOCl +
O2. A theoretical study40 of the potential-energy surface of this
reaction shows that the relevant barrier (analogous to TS1 in
Figure 7) is below the energy of the reactants, allowing the
bimolecular reaction to proceed. This is an interesting contrast
with reaction 4.
Figure 8. RRKM calculations at 300 K of the pressure dependences
of the atom abstraction and recombination channels of reaction 3 and
the recombination channel of reaction 4.
to the present result for reaction 3 (eq II). The kinetic behavior
of the HO2 + IO reaction is interpreted as proceeding through
the HOOIO (and possibly HOOOI) intermediate and then
forming HOIO2 or dissociating to HOI + O2.20
In the case of reaction 4, however, there are significant
barriers to forming the possible exothermic products, HOI +
O2(1∆g) and HO2 + IO (Figure 7). Thus, recombination to form
HOIO2 is likely to be the only major channel at atmospheric
temperatures. The RRKM model described above was then used
to calculate k4, the recombination rate coefficient, over a range
of temperature (200-400 K) and pressure (0-1000 Torr). The
results were fitted to the Lindemann expression modified by a
broadening factor Fc
Atmospheric Implications. In the Introduction, we demon-
strated that field observations of IO and OIO indicate that the
OIO lifetime during the day in the clean coastal marine boundary
layer is less than 140 s. To gauge an upper limit to the
importance of reactions 3 and 4, first ignore photolysis of OIO
and recombination with IO to form I2O3. Taking our measured
value for k3 at 290 K, then the absence of detectable levels of
OIO (<3 ppt) could be due to the presence of more than just
50 ppt of NO. Although there is not sufficient OH in the marine
boundary layer to control the OIO concentration through reaction
4, a typical summertime average OH concentration during the
day of 1 × 106 cm-3 would cause ∼5% of OIO to form HOIO2.
If HOIO2 was then removed only by uptake on aerosol, this
would provide an important route to aerosol iodate.
k
rec,0[M]
k4 )
FcK
rec,0[M]
k
1 +
krec,∞
Conclusions
where
In this paper, we have explored the reactions of OIO with
NO and OH. The rate coefficient for OIO + NO was determined
experimentally and that for OIO + OH by RRKM theory using
quantum calculations of the relevant points on the potential-
energy surface. The rate coefficients of both reactions are fast
enough around 290 K and 1 atm pressure for these reactions to
play a potentially important role in the gas phase and aerosol
chemistry in the marine boundary layer of the atmosphere.
In addition, these radical-radical reactions make an interest-
ing contrast. OIO + NO can form a weakly bound OIONO
intermediate, which then dissociates to the products IO + NO2.
However, the most stable potential product, IONO2, cannot form
because of the significant rearrangement of OIONO that would
be required. In contrast, OIO + OH can form directly the very
stable HOIO2 molecule, but the bimolecular reaction channels
1
K )
(IV)
2
krec,0[M]
1 + log 10
{
(
(
)) }
krec,∞
where the low-pressure limiting rate coefficient, krec,0 ) 1.5 ×
10-27 (T/300 K)-3.93 cm6 molecule-2 s-1, and the high-pressure
limiting rate coefficient, krec,∞ ) 5.5 × 10-10 exp(46/T) cm3
molecule-1 s-1, are obtained from the RRKM calculation. The
best fit yields Fc ) 0.30, which is somewhat smaller than the
value of 0.6 often assumed.36
The falloff curve for reaction 4a at 300 K is shown in Figure
8; the reaction is predicted to be close to the high-pressure limit
at pressures above 20 Torr. For the purpose of atmospheric

100 J. Phys. Chem. A, Vol. 110, No. 1, 2006
Plane et al.
(22) Ley, L.; Masanet, J.; Caralp, F.; Lesclaux, R. J. Phys. Chem. 1995,
99, 1953.
(23) Gilles, M. K.; Burkholder, J. B.; Ravishankara, A. R. Int. J. Chem.
to HO2 + IO and HOI + O2 are closed because of significant
energy barriers.
Kinet. 1999, 31, 417.
Acknowledgment. This work was supported by the UK
Natural Environment Research Council under the COSMAS
initiative and by the THALOZ project funded by the European
Commission.
(24) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,
M. A.; Cheeseman, J. R.; J. A. Montgomery, J.; 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, A. 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 03,
Revision B.03; Gaussian Inc.: Pittsburgh, PA, 2003.
(25) Glukhovtsev, M. N.; Pross, A.; McGrath, M. P.; Radom, L. J. Chem.
Phys. 1995, 103, 1878.
(26) Chase, M. W.; Davies, C. A.; Downey, J. R.; Frurip, D. J.; A., M.
R.; Syverud, A. N. J. Phys. Chem. Ref. Data 1985, 14.
(27) Bedjanian, Y.; Le Bras, G.; Poulet, G. J. Phys. Chem. A 1997, 101,
4088.
(28) De Avillez Pereira, R.; Baulch, D. L.; Pilling, M. J.; Robertson, S.
H.; Zeng, G. J. Phys. Chem. 1997, 101, 9681.
(29) Lide, D. R. Handbook of Physics and Chemistry, 72rd ed.; CRC
Press: Boca Raton, FL, 1992.
(30) Maitland, G. C.; Rigby, M.; Smith, E. B.; Wakeham, W. A.
Intermolecular Forces Their Origin and Determination; Oxford University
Press: Oxford, 1981.
References and Notes
(1) Alicke, B.; Hebestreit, K.; Stutz, J.; Platt, U. Nature 1999, 397,
572.
(2) Allan, B. J.; McFiggans, G.; Plane, J. M. C.; Coe, H. J. Geophys.
Res.-Atmos. 2000, 105, 14363.
(3) Saiz-Lopez, A.; Plane, J. M. C. Geophys. Res. Lett. 2004, 31, art.
no.
(4) Allan, B. J.; Plane, J. M. C.; McFiggans, G. Geophys. Res. Lett.
2001, 28, 1945.
(5) Davis, D. J.; Crawford, J.; Liu, S.; McKeen, S.; Bandy, A.;
Thornton, D.; Rowland, F.; Blake, D. J. Geophys Res. 1996, 101, 2135.
(6) McFiggans, G.; Plane, J. M. C.; Allan, B. J.; Carpenter, L. J.; Coe,
H.; O’Dowd, C. J. Geophys. Res.-Atmos. 2000, 105, 14371.
(7) Vogt, R.; Sander, R.; Von Glasow, R.; Crutzen, P. J. J. Atmos.
Chem. 1999, 32, 375.
(8) McFiggans, G.; Cox, R. A.; Mossinger, J. C.; Allan, B. J.; Plane,
J. M. C. J. Geophys. Res.-Atmos. 2002, 107, art. no. 4271.
(9) Jimenez, J. L.; Bahreini, R.; Cocker, D. R.; Zhuang, H.; Varut-
bangkul, V.; Flagan, R. C.; Seinfeld, J. H.; O’Dowd, C. D.; Hoffmann, T.
J. Geophys. Res.-Atmos. 2003, 108, art. no. 4318.
(10) O’Dowd, C. D.; Jimenez, J. L.; Bahreini, R.; Flagan, R. C.; Seinfeld,
J. H.; Hameri, K.; Pirjola, L.; Kulmala, M.; Jennings, S. G.; Hoffmann, T.
Nature 2002, 632.
(11) McFiggans, G.; Coe, H.; Burgess, R.; Allan, J.; Cubison, M.;
Alfarra, M. R.; Saunders: R.; Saiz-Lopez, A.; Plane, J. M. C.; Wevill, D.
J.; Carpenter, L. J.; Rickard, A. R.; Monks, P. S. Atmos. Chem. Phys. 2004,
4, 701.
(12) Carpenter, L. J.; Malin, G.; Liss, P. S.; Kupper, F. C. Glob.
Biogeochem. Cycle 2000, 14, 1191.
(13) Cox, R. A.; Bloss, W. J.; Jones, R. L.; Rowley, D. M. Geophys.
Res. Lett. 1999, 26, 1857.
(14) Joseph, D. M.; Ashworth, S. H.; Plane, J. M. C. J. Photochem.
Photobiol., A 2005, 176, 68.
(15) Rowley, D. M.; Bloss, W. J.; Cox, R. A.; Jones, R. L. J. Phys.
Chem. A 2001, 105, 7855.
(16) Ashworth, S. H.; Allan, B. J.; Plane, J. M. C. Geophys. Res. Lett.
2002, 29.
(17) Saiz-Lopez, A.; Plane, J. M. C.; McFiggans, G.; Ball, S. M.; Bitter,
M.; Jones, R. L. Atmos. Chem. Phys. Discuss. 2005, 5, 5405.
(18) Li, Z. J.; Tao, Z. N. Chem. Phys. Lett. 1999, 306, 117.
(19) Li, Z. J.; Wuebbles, R. D.; Pylawka, N. J. Chem. Phys. Lett. 2002,
354, 491.
(20) Drougas, E.; Kosmas, A. M. J. Phys. Chem. A 2005, 109, 3887.
(21) Baker, A. R.; Tunnicliffe, C.; Jickells, T. D. J. Geophys. Res.-Atmos.
2001, 106, 28743.
(31) Georgievskii, Y.; Klippenstein, S. J. J. Chem. Phys. 2005, 122,
art. no. 194103.
(32) Gilbert, R. G.; Smith, S. C. Theory of Unimolecular and Recom-
bination Reactions; Blackwell: Oxford, 1990.
(33) Papayannis, D. K.; Kosmas, A. M. Chem. Phys. Lett. 2004, 398,
75.
(34) Papayannis, D. K.; Kosmas, A. M. Chem. Phys. 2005, 315, 251.
(35) Foresman, J. B.; Frisch, A. Exploring chemistry with electronic
structure methods; Gaussian, Inc.: Pittsburgh, PA, 1996.
(36) Sander, S. P.; Friedl, R. R.; Golden, D. M.; Kurylo, M. J.; Huie,
R. E.; Orkin, V. L.; Moortgat, G. K.; Ravishankara, A. R.; Kolb, C. E.;
Molina, M. J.; Finlayson-Pitts, B. J. Chemical Kinetics and Photochemical
Data for Use in Stratospheric Modelling: EValuation 14; JPL Publication
02-25: Pasadena, 2002.
(37) Cronkhite, J. M.; Stickel, R. E.; Nicovich, J. M.; Wine, P. H. J.
Phys. Chem. A 1999, 103, 3228.
(38) Knight, G. P.; Crowley, J. N. Phys. Chem. Chem. Phys. 2001, 3,
393.
(39) Poulet, G.; Zagogianni, H.; Lebras, G. Int. J. Chem. Kinet. 1986,
18, 847.
(40) Xu, Z.-F.; Zhu, R.; Lin, M. C. J. Phys. Chem. A 2003, 107, 1040.