Rate constants for total relaxation from the rotational levels J = 7.5, 20.5, 31.5 and 40.5 in NO(X2Π1/2, v = 2) in collisions with He, Ar and N2: A comparison between experiment and theory

Article abstract of DOI:10.1016/S0009-2614(99)00402-9

Using the infrared-ultraviolet double resonance (IRUVDR) technique, rate coefficients have been measured for the relaxation of NO(X²Π_1/2, v = 2, J = 7.5, 20.5 and 31.5) in collisions with He, Ar, N₂ and NO itself. For relaxation from levels J = 31.5 and 40.5, rates were measured by a new method in which the IRUVDR method was implemented instantly on NO molecules formed by photodissociation of NO₂ at 355 nm. The rate coefficients for collisions with He, Ar and N₂ are found to decrease at high J, with the decrease least with He. The experimental results for NO-He and NO-Ar are shown to agree well with those from coupled states quantum scattering calculations on ab initio potential energy surfaces.

Full text of DOI:10.1016/S0009-2614(99)00402-9

2
8 May 1999
Chemical Physics Letters 305 Ž1999. 311–318
Rate constants for total relaxation from the rotational levels
2
ž
/
Js7.5, 20.5, 31.5 and 40.5 in NO X P , Õs2 in collisions
1
r2
with He, Ar and N : a comparison between experiment and theory
2
Meezanul Islam , Ian W.M. Smith a,), Millard H. Alexander b
a
a
School of Chemistry, UniÕersity of Birmingham, Edgbaston, Birmingham B15 2TT, UK
b
Department of Chemistry and Biochemistry, UniÕersity of Maryland, College Park, MD 20742-2021, USA
Received 24 March 1999; in final form 6 April 1999
Abstract
Using the infrared–ultraviolet double resonance ŽIRUVDR. technique, rate coefficients have been measured for the
2
relaxation of NOŽ X P , Õs2, Js7.5, 20.5 and 31.5. in collisions with He, Ar, N and NO itself. For relaxation from
1
r2
2
levels Js31.5 and 40.5, rates were measured by a new method in which the IRUVDR method was implemented instantly
on NO molecules formed by photodissociation of NO at 355 nm. The rate coefficients for collisions with He, Ar and N are
2
2
found to decrease at high J, with the decrease least with He. The experimental results for NO–He and NO–Ar are shown to
agree well with those from coupled states quantum scattering calculations on ab initio potential energy surfaces. q 1999
Elsevier Science B.V. All rights reserved.
1
. Introduction
low vibrational states ŽÕ(3. by direct infrared ab-
sorption w1–7,10x or to levels of higher vibrational
In recent years, optical–optical double resonance
states, e.g., Õs20, by stimulated emission pumping
techniques have been widely used to measure two
kinds of rate coefficient for collisionally induced
rotational energy transfer: Ži. those for total transfer
out of a specific rovibrational level, and Žii. those for
state-to-state transfer between specific initial and
final levels. Collisions involving NOŽX P . have
been the subject of a large number of experimental
investigations w1–11x, including several from this
laboratory w4–7,10x. In part, this emphasis on NO is
a result of its favourable spectroscopic properties.
Thus, it can be excited to specific rotational levels of
w8,9x. Moreover, the subsequent fate of these excited
molecules can then be followed by means of laser-
induced fluorescence ŽLIF. w4–10x or resonance-en-
hanced multiphoton ionization ŽREMPI. w1–3,11x
2
techniques. The increase in experimental information
on rotational energy transfer in NO has been paral-
leled by progress in theory w9–19x. Collisions involv-
V
2
ing NOŽX P ., particularly those with the rare-gas
V
atoms He and Ar, where accurate ab initio potential
surfaces are available w13x, have become an interest-
ing and challenging prototype for encounters involv-
2
ing molecular radicals in P states.
Our experimental work in Birmingham 4–7,10 ,
including that reported in the present Letter, has
V
)
w
x
Corresponding author. Fax: q44 121 414 4426; e-mail:
i.w.m.smith@bham.ac.uk
0
009-2614r99r$ - see front matter q 1999 Elsevier Science B.V. All rights reserved.
PII: S0009-2614Ž99.00402-9

3
12
M. Islam et al.rChemical Physics Letters 305 (1999) 311–318
employed the infrared–ultraviolet double resonance
ŽIRUVDR. technique to study rotational energy
transfer in NO–rare-gas collisions. A pulsed, tun-
and all these results are compared with the results of
quantum scattering calculations performed on ab ini-
tio potential energy surfaces.
able, infrared ‘pump’ laser is used to excite NO
As well as fundamental interest, in particular the
investigation of whether the theoretical results con-
tinue to match experiment for high values of J,
where the spacing between neighbouring rotational
molecules to a specific rotational level in either the
2
Õs2 or Õs3 vibrational level of the X P elec-
V
tronic ground state. A tunable dye laser then excites
LIF from the rovibrationally excited molecules via
levels approaches k T, the investigation of rota-
B
2
q
2
transitions in the A S –X P system. Rate coeffi-
cients for total relaxation are determined by tuning
this ‘probe’ laser to a transition from the rovibra-
tional level directly populated by IR absorption, and
observing the variation of the LIF signal intensity as
the time delay between the pulses from the pump
and probe lasers is varied. Alternatively, the delay
between the two laser pulses can be fixed to corre-
spond to a small fraction of the time between colli-
sions and the frequency of the probe laser scanned to
record a LIF spectrum. The line intensities in this
spectrum reflect the concentrations in levels popu-
lated in collisions of the NO molecules in the level
originally excited and hence state-to-state rate coef-
ficients can be determined.
tional energy transfer at high J is important in
respect of various aspects of atmospheric chemistry.
Some evidence for a decrease in the total rate for
rotational relaxation with increasing rotational quan-
tum number comes from the observation of emission
from high rotational levels of NO in the thermo-
sphere w24,25x. The absence of a thermalised distri-
bution over rotational levels points to the possibility
of a decrease in the rate of rotational relaxation with
increasing rotational quantum number. Further indi-
rect evidence for the dependence of the total rate of
rotational relaxation on the initial rotational state
comes from line broadening measurements on transi-
tions in the Ž2, 0. overtone band in the ground state
of NO w26,27x. With increasing rotational quantum
The IRUVDR method as just described has one
major drawback. The excited states that can be ini-
tially selected are limited by the thermal populations
number, a slight decrease was observed in the pa-
rameter which quantifies line-broadening and which
can often be correlated with the rate coefficients for
rotational relaxation.
2
of rotational levels in NOŽX P , Õs0. and the
V
optical selection rules for the IR transitions. This
restriction is particularly apparent in the extensive
measurements that we have made on rotational en-
ergy transfer at temperatures down to 7 K w7,10x. In
2. Experimental
the experiments reported in the present Letter, the
range of data on rotational energy transfer in
The experiments were performed using the IRU-
VDR technique. Pulses of tunable IR radiation were
generated by difference frequency mixing the output
of a Nd:YAG pumped dye laser with the fundamen-
tal wavelength of the pump laser ŽContinuum
2
NOŽX P . has been extended to rotational levels
V
up to Js40.5. This has been achieved by perform-
ing IRUVDR experiments on NO immediately after
its formation in the photodissociation of NO at 355
ND8010r6000. in a LiNbO crystal. To excite
2
3
2
nm. This photolytic process creates NOŽX P . in
molecules to ŽÕs2, Vs1r2, J. the frequency of
V
rotational levels in Õs0 and Õs1 up to the ener-
this pump laser was tuned into resonance with a
selected line in the R branch of the ŽVs1r2§1r2.
sub-band of the Ž2, 0. first vibrational overtone band
getically accessible limits w20–23x. In our experi-
ments, IR pulses from a difference-frequency laser
source have excited molecules to Õs2, Js31.5
and Õs2, Js40.5 and kinetic experiments have
been performed to measure rate coefficients for total
removal from these levels in collisions with He, Ar
at ca. 2.6 mm. The pulse energy of the IR radiation
was in the range 1.5–2.0 mJ and the bandwidth of
the laser output was ca. 0.1 cm . The frequency of
the IR laser was fine-tuned to the appropriate transi-
tion with the aid of a spectrophone containing 50
Torr of NO. Tunable UV probe radiation, with a
y1
and N . These results are compared with new, con-
2
ventional IRUVDR measurements on the rotational
relaxation of NO from levels Js7.5, 20.5 and 31.5
y1
bandwidth of ca. 0.4 cm , was produced by fre-

M. Islam et al.rChemical Physics Letters 305 (1999) 311–318
313
quency doubling the output of an excimer-pumped
dye laser ŽLambda-Physik, FL2002.. LIF was ex-
Js37.5 level, the frequency of the pump laser was
then scanned to wavelengths where the Ž2, 0. R-
cited by tuning this UV probe radiation to an appro-
branch IR transition to Js40.5 was expected, again
at a delay corresponding to ca. 1 collision. A LIF
signal was detected when the IR laser was in reso-
nance with the line promoting molecules to the
Js40.5 level. This was confirmed by subsequently
2
q
priate transition in the Ž0, 2. band of the NO A S –
2
X P system at ca. 247 nm.
Experiments on the highest rotational levels Ž Js
3
1.5 and 40.5. required the initial photolysis of NO
2
2
q
to generate rotationally hot NO molecules in the
ground vibrational state. For this purpose, laser radi-
ation was generated at 355 nm by sum frequency
mixing the fundamental and doubled output of a
Nd:YAG laser ŽSpectron, SL 803.. Typical pulse
tuning the UV probe laser to the A S –
2
X P1r2Ž0, 2. transition from this level at short time
delays. The kinetic experiments in which the time
delay between the pulses from the pump and probe
lasers was varied were then performed. In the three
laser experiments, the time delay between the pulse
energies of ca. 20 mJ and a bandwidth of 0.1 cmy1
were obtained.
from the laser photolysing NO and that from the IR
2
The beams from the IR and UV lasers counter-
propagated along the axis of a cylindrical Pyrex cell
with a CaF2 window mounted on each end. The
photomultiplier, interference filter and a collecting
lens were mounted in a central housing which was
clamped to this cell. The equipment for controlling
the firing of the lasers and for recording, accumulat-
ing and analysing the LIF signals was the same as
that described by Frost and Smith w28x.
pump laser was -100 ns.
In all experiments, the time delay between the
pump and probe laser pulses was varied with a delay
generator ŽStanford DG535. and the resulting LIF
signal was detected. To discriminate against scat-
tered light from the probe laser, the fluorescence was
observed using a solar blind photomultiplier tube
ŽHamamatsu R801., through a quartz window and an
interference filter ŽCorion, FWHM 10 nm. centred at
228 nm, which isolated the fluorescence in the Ž0, 0.
Conventional IRUVDR experiments were carried
out to measure the total rates of rotational relaxation
2
q
2
band of the A S –X P system at ca. 226 nm.
The dependence of the observed rate coefficients
on the relative polarisations of the lasers was exam-
ined by rotating the plane of polarisation of the
probe laser through 908 from its usual orientation
using a Fresnel rhomb. The measured rate coeffi-
cients were found to be independent of this change
within the error limits of the measurements. This
confirmed that our measurements yield rates of rota-
tional energy transfer and do not contain a signifi-
cant component due to re-orienting collisions with-
out change of rotational level.
2
from NOŽX P , Õs2, Js7.5, 20.5 and 31.5. in
1
r2
collisions with He, Ar, N and NO. In these experi-
2
ments, the IR pump laser was first tuned to the
appropriate line in the Ž2, 0. vibrational overtone
band using the spectrophone. Then a short delay was
set between the pump and probe pulses and the UV
laser was scanned to bring its frequency into reso-
2
q
nance with the appropriate line in the A S –
2
X PŽ0, 2. band.
IRUVDR experiments on the NO produced from
photolysis of NO were carried out to measure the
2
2
total rates of rotational relaxation from NOŽX P
,
1
r2
Õs2, Js31.5 and 40.5. in collisions with He, Ar,
N . Tuning the lasers was now rather more difficult.
3. Experimental results
2
As m2entioned above, for measurements on
NOŽX P , Õs2, Js31.5., the correct IR pump
The decay in the LIF signals as a function of time
for a given gas mixture was fitted to an exponential
function from which a pseudo-first-order rate coeffi-
cient was obtained. Examples of such fittings for
1
r2
frequency could be located using the spectrophone.
2
In addition, a LIF spectrum from NOŽX P , Õs2.
1
r2
could be taken at a time delay corresponding to ca. 1
2
q
2
2
collision in order to locate the A S –X P1r2Ž0, 2.
transitions from higher, collisionally populated, rota-
tional levels up to Js37.5. With the frequency of
the UV probe laser fixed on the line from the
NOŽX P , Õs2, Js40.5. in mixtures contain-
1r2
ing different partial pressures of helium are shown in
Fig. 1. A series of such measurements was carried
out and the first-order rate coefficients were plotted

3
14
M. Islam et al.rChemical Physics Letters 305 (1999) 311–318
2
Fig. 1. Traces of LIF signal from NOŽX P , Õs2, Js40.5. versus time recorded from NO produced in the photolysis of NO at 355
1
r2
2
nm. All mixtures contained 20 mTorr of NO2 in Ža. 221, Žb. 301, Žc. 400, and Žd. 507 mTorr He.
against the concentration of collision partner to yield,
from the gradient, the second-order rate coefficient
for total rotational relaxation from that particular
level in collisions with the selected collision partner.
Fig. 2 shows one such plot, and the measured sec-
ond-order rate constants are listed in Table 1. Strictly,
because the population in the observed level decays
to its equilibrium value, the measurements yield the
sum of the rate coefficients for relaxation out of and
into the observed level. However, only for the lower
levels studied here was it necessary to correct for the
reverse process, as described in earlier papers w4–
7,10x.
The derived second-order rate coefficients for re-
laxation by He, Ar and N2 from different initial
rotational levels are presented graphically in Fig. 3.
Two general features are apparent. First, the rate
coefficients show a significant, if small decrease
with increasing rotational quantum number. Second,
this change is least for helium as collision partner.
Further, we note that the 17% decrease in the rate
coefficient for relaxation by Ar in going from Js7.5
to Js20.5 is very similar to the 14% decrease in
the collision-broadening parameters measured by
Pine w27x in going from Js7.5 to Js17.5.
Table 1 lists two values of the rate coefficients for
relaxation from Js31.5, one from conventional
IRUVDR experiments the other from experiments in
which NO was generated in a non-thermal distribu-
tion by photolysis of NO at 355 nm. In the latter
2
case, the NO molecules will be created with a non-
thermal translational distribution, their average trans-
y1
lational energy being equivalent to ca. 460 cm
,
Fig. 2. Plot of the first-order rate coefficients obtained from the
2
about twice the thermal value at 298 K. The agree-
ment between the two derived values of the rate
coefficients suggest that this makes little difference
LIF decay traces associated with relaxation of NOŽX P1r2
,
Õs2, Js40.5. plotted against the concentration of helium added
to 20 mTorr of NO .
2

M. Islam et al.rChemical Physics Letters 305 (1999) 311–318
315
Table 1
Rate coefficients Žkr10
He, Ar, N and NO
y10
cm3 molecule
y1 y1
2
s
. for relaxation from NOŽX P , Õs2, Js7.5, 20.5, 31.5 and 40.5. in collisions with
r2
1
2
a
b
Collision partner
He
Js6.5
Js7.5
Ž3.0"0.3.c
Js20.5
Js31.5
Js40.5
Ž3.1 "0.3.
Ž2.3 "0.2.
w8x
Ä34d
Ä24
Ž2.4 "0.3.
Ä34
Ä24
Ž2.1"0.3.
Ä24
Ž2.2"0.3.
Ä24
Ä34
Ä24
6
5
e
e
e
w3Žb.x
2.77
2.43
2.12
5
e
2
.30
Ar
Ž1.9"0.2.
w3Žb.x
Ž3.4"0.2.
Ž2.8"0.3.
Ž1.7 "0.2.
Ž2.0"0.3.
Ä34
Ä24
Ž1.3"0.2.
5
e
e
e
2
.74
2.25
1.23
e
1
.60
N2
NO
Ž3.5"0.5.
Ž3.8"0.2.
w3Žb.x
w3Žb.x
Ž4.6"0.5.
Ž4.2"0.5.
Ä34
Ä24
Ž4.4"0.5.
Ž4.4"0.3.
Ä44
Ä24
Ž4.0"0.4.
Ž3.5"0.4.
Ä24
Ä24
Ž2.5"0.3.
–
a
The rate coefficients in this column are from earlier work and are included for comparison.
The upper values were obtained from conventional IRUVDR experiments, the lower values from experiments in which NO was generated
b
by 355 nm photodissociation of NO .
2
c
Cited errors correspond to two standard deviations.
Number of individual series of experiments.
Theoretical calculations Žpresent Letter., exclusive of rate for e™f transfer within the initially populated rotational level.
d
e
to the efficiency of rotational energy transfer. In the
case of NO produced in Õs0, Js40.5, the transla-
tional energy will be about half the thermal value.
We have assumed that the resultant non-thermal
distribution of collision energies makes no signifi-
cant difference to the rate coefficients for relaxation.
molecular electronic–rotational wavefunctions, mul-
tiplied by functions which describe the orbital mo-
tion of the noble-gas collision partner with respect to
the diatomic molecule. Collision-induced transitions
between any pair of levels, at least one of which is
described in this intermediate coupling regime, will
be induced by both the average and difference poten-
tials w12,14,32x. The extent of the cross-coupling of
4
. Theoretical calculations
these two potentials will be governed by the degree
of mixing of the pure Hund’s case Ža. rotational
basis states of the NO molecules w14,18,19,32x.
The theoretical rate scattering calculations follow
the formalism we have presented in detail earlier
w10,12–14x. To avoid repetition we shall repeat here
only those details which bear directly on our new
results.
We have used our earlier CEPA w29–31x calcula-
tions w18,19x of the two potential energy surfaces
To calculate the rotationally inelastic cross sec-
tions based on these PESs we used the Hibridon4
1
package. In the present calculations, the coupled
states ŽCS. approximation w12x, which has been
shown previously w18,19x to be accurate for colli-
2
sions of NOŽX P. with both He and Ar, was used.
ŽPES., V
X
and VA
Y
, which arise when a noble-gas
As before, the NO molecule was assumed to be a
rigid rotor with a bond distance equal to the average
A
2
atom interacts with an NO molecule in its P ground
state w13x. In the scattering calculations it is most
convenient to use the average and the half-difference
of the two PES, namely w13x
1
HIBRIDON is a package of programs for the time-indepen-
Vsum
Ž
R,u
.
s0.5 VA
Y
Ž
R,u
.
qVA
X
Ž
R,u
.
,
Ž
1a
.
dent quantum treatment of inelastic collisions and photodissocia-
tion written by M.H. Alexander, D.E. Manolopoulos, H.-J. Werner,
B. Follmeg, with contributions by P.F. Vohralik, D. Lemoine, G.
Corey, B. Johnson, T. Orlikowski, W. Kearney, A. Berning, A.
Degli-Esposti, C. Rist, P. Dagdigian. More information andror a
copy of the code can be obtained from the website http:rrwww-
mha.umd.edur;mharhibridon.
and
Vdif
Ž
R,u
.
s0.5 VA
Y
Ž
R,u
.
yVA
X
Ž
R,u
.
.
Ž
1b
.
The total wavefunction of the NO–noble-gas sys-
tem is then expanded as a sum of products of the

3
16
M. Islam et al.rChemical Physics Letters 305 (1999) 311–318
over all final states to recover rate coefficients for
total removal from the initially selected rotational
level.
The theoretical results are compared with the
experimental rate coefficients in Fig. 3.
5
. Discussion
For lower rotational levels, the present experimen-
tal results can be compared with the rate coefficients
measured previously for NOŽÕs2, Js6.5. by Is-
lam et al. w5x and for NOŽÕs3, Js6.5. by James et
al. w10x Žsee Table 1.. James et al. only reported a
room temperature rate coefficient for relaxation in-
duced by collisions with He. It is in excellent agree-
ment with our result for Õs2, Js7.5. In most
cases, agreement with the rate coefficients obtained
by Islam et al. is quite good, although their rate
coefficients are generally slightly smaller than the
present values with a significant disagreement in the
case of NO–Ar. Table 1, as well as Fig. 3, compares
the theoretical and experimental rate coefficients for
total removal by Ar and He. The agreement is excel-
lent, both in absolute magnitude and in variation
with the initially pumped level.
The rate coefficients measured in the present work
do demonstrate that at higher rotational levels the
rates of total rotational energy transfer do diminish,
albeit slightly, for the range of levels up to Js40.5.
In the highest rotational level studied Ž Js40.5., the
Fig. 3. Second-order rate coefficients for relaxation of
2
NOŽX P , Õs2. from selected rotational levels J plotted
1
r2
against J. The filled symbols represent the experimental data, the
open symbols the results of the quantum scattering calculations
described in the text.
rotational period of an isolated NO molecule is 0.24
ps. During collisions between NO and Ar and be-
tween NO and N at the mean relative speed corre-
2
sponding to 298 K, the intermolecular separation will
change by only about 1.5 A˚ during this rotational
value in the Õs2 vibrational state. Scattering calcu-
lations were carried out on a fixed grid of )120
total energies, where the total energy is the sum of
the energy of the initial internal state and the colli-
sion energy. Subtraction of the internal energy then
transformed these results to cross-sections as a func-
tion of collision energy w10x. Thermal state-to-state
rate coefficients were obtained from the inelastic
integral cross-sections by multiplication of the
cross-sections by the relative velocity and subse-
quent integration over a Maxwellian velocity distri-
bution at 300 K. Finally, the results were summed
period. Consequently, during its interaction with
these collision partners, the NO molecule will un-
dergo a substantial fraction of a full rotation with the
result that the intermolecular potential will be sub-
stantially averaged over orientations which is likely
to lower the probability of rotational energy transfer.
With He as the collision partner, the mean relative
speed is greater, by a factor )2, and less averaging
will occur. This may explain why the rate coeffi-
cients for He show less sensitivity to the initial value
of the NO rotational quantum number.

M. Islam et al.rChemical Physics Letters 305 (1999) 311–318
317
Table 2
tory investigations, where it is far preferable to
employ an effectively instantaneous unimolecular
photodissociation to produce rotationally ‘hot’ NO in
a time much shorter than the average collision time.
Average loss of rotational angular momentum in inelastic colli-
sions of NOŽX P , Õs2, Js20.5, 31.5 and 40.5. in colli-
sions with He and Ar at 300 K
2
1
r2
a
Collision partner
He
Js20.5
y2.1
Js31.5
Js40.5
An alternative photochemical source to NO is pos-
2
y2.5
3.2
y2.6
3.1
y2.6
3.0
y2.1
2.4
sibly NOCl w34x. However, at longer photolysis
wavelengths Ž)300 nm. NO is not produced in any
3
.2
y3.0
.7
Ar
higher J states than from NO . Photolysis at shorter
3
2
wavelengths is accompanied by the formation of
substantial concentrations of vibrationally excited NO
molecules which may lead to complications in the
analysis of such experiments.
a
The first entry refers to ²D J: and the second to ²<D J <:; see Eqs.
Ž2a. and Ž2b.. In both cases the calculation include transfer from
the initially populated level to all rotational levels in both spin–
orbit manifolds.
An attractive alternative to such experiments is to
explore the relaxation of NO in the range of J levels
explored in the present experiments, but at much
reduced translational temperatures. Such experiments
should be possible using the CRESU ŽCin e´ tique de
R e´ action en Ecoulement Supersonique Uniforme.
apparatus which has been used already w7,10x to
determine rate coefficients for rotational relaxation
from low J levels in NO at temperatures down to 7
K. Such experiments would make it possible to
examine rotational relaxation under conditions where
the energy differences between neighbouring rota-
tional levels far exceed the average collision energy
and such experiments will be pursued in the future.
It is worthwhile to determine the average loss of
rotational angular momentum per collision ŽTable 2.,
which can be defined as
X
²
D J: s
Ž
JyJ
.
k
X
k ,
X
Ž
2a
.
J
Ý
J ™ J
Ý
J ™ J
X
X
J
J
and
<D J <: s
X
²
<
Ž
JyJ
.
<kJ ™ J
X
k
X
X
.
Ž
2b
.
J
Ý
Ý
J ™ J
X
J
J
We observe that for collisions with Ar, the average
changes in rotational quantum number decrease as J
increases. This reflects the increasing size in the
rotational energy spacing with increasing J. Colli-
sions with the faster moving He atom are more
impulsive, and hence less sensitive to the energy
loss. Consequently, for collisions with He, the aver-
age changes in rotational quantum number show
little variation with J, even up to Js40.5. Experi-
mental rate coefficients for state-to-state rotational
energy transfer will be reported in a future publica-
tion.
Acknowledgements
The Birmingham group are grateful to EPSRC for
a research grant in support of this work. The Contin-
uum laser system providing the IR pump beam was
obtained on loan from the EPSRC Laser Loan Pool
at the Rutherford–Appleton Laboratory, for which
we express our thanks. MA wishes to thank the US
National Science Foundation for support under Grant
CHE-9629385 and CHE-9971810.
The present experimental results, obtained using a
novel variant of the IRUVDR technique, substan-
tially extend the range of rotational levels in
2
NOŽX P . for which rate coefficients for relaxation
V
are available. Nevertheless, the upper limit of the
range, Js40.5, falls some way short of the rota-
tional states Žup to Jf100. which are populated in
References
w1x P. Esherick, R.J.M. Anderson, Chem. Phys. Lett. 70 Ž1980.
the thermosphere and observed via emission mea-
6
21.
surements w24,25x. The mechanism proposed w33x for
w2x Aa.S. Sudbo, M.M.T. Loy, Chem. Phys. Lett. 82 Ž1981. 135.
the formation of these molecules, namely the reac-
w3x Aa.S. Sudbo, M.M.T. Loy, J. Chem. Phys. 76 Ž1982. 3646.
4
tion of translationally excited NŽ S. atoms with O ,
2
4
w x
M.J. Frost, M. Islam, I.W.M. Smith, Can. J. Chem. 72 1994
606.
Ž
.
seems unpromising from the point-of-view of labora-

3
18
M. Islam et al.rChemical Physics Letters 305 (1999) 311–318
w5x M. Islam, I.W.M. Smith, J.W. Wiebrecht, J. Phys. Chem. 98
Ž1994. 9285.
w20x H. Zacharias, M. Geilhaupt, K. Meier, K.H. Welge, J. Chem.
Phys. 74 Ž1981. 218.
w6x M. Islam, I.W.M. Smith, J.W. Wiebrecht, J. Chem. Phys.
w21x H. Zacharias, K. Meier, K.H. Welge, in: J. Hinze ŽEd..,
1
03 Ž1995. 9676.
Energy Storage and Redistribution in Molecules, Plenum
w7x P.L. James, I.R. Sims, I.W.M. Smith, Chem. Phys. Lett. 272
Press, New York, 1983.
Ž1997. 412.
w22x M. Mons, I. Dimicoli, Chem. Phys. 130 Ž1989. 307.
w23x I.W.M. Smith, R.P. Tuckett, C.J. Whitham, Chem. Phys.
Lett. 200 Ž1992. 615.
w24x S.J. Lipson, P.S. Armstrong, J.A. Dodd, J.R. Lowell, W.A.M.
Blumberg, R.M. Nadile, Geophys. Res. Lett. 21 Ž1994. 2421.
w25x P.S. Armstrong, S.J. Lipson, J.A. Dodd, J.R. Lowell, W.A.M.
Blumberg, R.M. Nadile, Geophys. Res. Lett. 21 Ž1994. 2425.
w26x A.S. Pine, A.G. Maki, N.Y. Chou, J. Mol. Spectrosc. 114
Ž1985. 132.
w27x A.S. Pine, J. Chem. Phys. 91 Ž1989. 2002.
w28x M.J. Frost, I.W.M. Smith, Chem. Phys. Lett. 191 Ž1992. 574.
w29x W. Meyer, Int. J. Quantum Chem. Sympos. 5 Ž1971. 341.
w30x W. Meyer, J. Chem. Phys. 58 Ž1973. 1017.
w31x W. Meyer, Theoret. Chim. Acta 35 Ž1974. 277.
w32x M.H. Alexander, W. Kearney, A.F. Wagner, J. Chem. Phys.
100 Ž1994. 1338.
w8x X. Yang, A.M. Wodtke, J. Chem. Phys. 96 Ž1992. 5123.
w9x M. Drabbels, A.M. Wodtke, M. Yang, M.H. Alexander, J.
Phys. Chem. 101 Ž1997. 6463.
w10x P.L. James, I.R. Sims, I.W.M. Smith, M.H. Alexander, M.
Yang, J. Chem. Phys. 109 Ž1998. 3882.
w11x A. Lin, S. Antonova, A.P. Tsakotellis, G.C. McBane, J.
Phys. Chem. 109 Ž1999. 1198.
w12x M.H. Alexander, J. Chem. Phys. 76 Ž1982. 5974.
w13x M.H. Alexander, Chem. Phys. 92 Ž1985. 337.
w14x G.C. Corey, M.H. Alexander, J. Chem. Phys. 85 Ž1986.
5
652.
w15x T. Orlikowski, M.H. Alexander, J. Chem. Phys. 79 Ž1983.
006.
w16x T. Orlikowski, M.H. Alexander, J. Chem. Phys. 80 Ž1984.
506.
w17x T. Orlikowski, M.H. Alexander, J. Chem. Phys. 80 Ž1984.
133.
6
1
w33x R.D. Sharma, Y. Sun, A. Dalgarno, Geophys. Res. Lett. 20
4
Ž1993. 2043.
w18x M.H. Alexander, J. Chem. Phys. 99 Ž1993. 7725.
w19x M. Yang, M.H. Alexander, J. Chem. Phys. 103 Ž1995. 6973.
w34x Y.Y. Bai, A. Ogai, C.X.W. Qian, L. Iwata, G.A. Segal, H.
Reisler, J. Chem. Phys. 90 Ž1989. 3903.