State-specific low temperature reactions of DBr+2 i(v+) + (D2, DBr) → D2Br+

Article abstract of DOI:10.1016/S0009-2614(02)00564-X

State-specific reactions of DBr⁺(² _i, v⁺) with D₂ and DBr were studied in a low temperature free jet flow reactor. The selected spin-orbit and vibrational states of the DBr⁺ ion were prepared by resonance-enhanced multiphoton ionization. All of the reactant and product ions were monitored using time-of-flight mass spectrometry. Rate coefficients for the DBr⁺/DBr reaction are ～1.4 × 10^-9 cm³ s^-1 independent of the ion internal state, similar to those for the HBr⁺/HBr reaction. Rate coefficients of the DBr⁺/D₂ reaction rise from 0 (lower spin-orbit-vibrational states) to 2 × 10^-11 cm³ s^-1 (highest energy states) and are observed to follow a mass independent threshold function similar to the HBr⁺/H₂ reaction.

Full text of DOI:10.1016/S0009-2614(02)00564-X

24 June 2002
Chemical Physics Letters 358 (2002) 57–63
www.elsevier.com/locate/cplett
State-specific low temperature reactions of
þ
þ
DBr PiðvþÞ þ ðD2; DBrÞ ! D2Br
2
1
Andrey E. Belikov , Mark A. Smith
*
Department of Chemistry, University of Arizona, 1306 E. University Dr., Tucson, AZ 85721, USA
Received 20 November 2001; in final form 11 March2002
Abstract
þ
2
State-specific reactions of DBr ð Pi; vþÞ withD and DBr were studied in a low temperature free jet flow reactor.
2
The selected spin–orbit and vibrational states of the DBrþ ion were prepared by resonance-enhanced multiphoton
ionization. All ofþthe reactant and product ions were monitored using time-of-flight mass spectrometry. Rate coeffi-
cients for the DBr =DBr reaction are ꢀ1:4 Â 10À9 cm3 sÀ1 independent of the ion internal state, similar to those for the
HBrþ=HBr reaction. Rate coefficients of the DBrþ=D2 reaction rise from 0 (lower spin–orbit-vibrational states) to
2 Â 10À11 cm3 sÀ1 (highest energy states) and are observed to follow a mass independent threshold function similar to
the HBrþ=H2 reaction. Ó 2002 Published by Elsevier Science B.V.
1. Introduction
nance enhanced multiphoton ionization (REMPI)
of HCl and HBr have demonstrated fine internal
state selection of the HX ð Pi; vþÞ cation amena-
ble to reaction study [1]. We have recently ex-
plored fine structure (i) and vibrational (v) state
dependence for the reactions [2]
þ
2
Control of chemical reaction efficiency through
selection of reactant state remains a primary goal
of the field of reaction dynamics. It is anticipated
that state-specific behavior of chemical reaction
systems will reveal information concerning reac-
tive potential energy surfaces difficult to obtain by
other means and that state selective dependence of
branching between competitive collision pathways
(inelastic and reactive) will elucidate the nature of
coupling between multiple channels on these sur-
faces. In this regard, recent developments in reso-
HBrþ2PiðvþÞ þ HBr ! H2Brþ þ Br
HBrþ2PiðvþÞ þ H2 ! H2Brþ þ H
ð1aÞ
ð1bÞ
Reaction (1a) is an ion-dipole driven reaction
on a completely attractive potential energy surface
for which the model of reactive capture appears to
hold. In this case reaction efficiency appears to
remain fixed at the collision ion-dipole capture
limit for all temperatures and for all internal HBrþ
states probed. In contrast, Reaction (1b) is 0.35 eV
endothermic and occurs on a potential surface for
which little is directly known. Reaction (1b) is
* Corresponding author. Fax: +1-520-621-8407.
E-mail address: msmith@u.arizona.edu (M.A. Smith).
1
Permanent address: Institute of Thermophysics, Novosib-
irsk, 630090, Russia.
0009-2614/02/$ - see front matter Ó 2002 Published by Elsevier Science B.V.
PII: S0009-2614(02)00564-X

58
A.E. Belikov, M.A. Smith / Chemical Physics Letters 358 (2002) 57–63
observed to increase in rate when internal energy
in excess of the thermodynamic barrier is placed in
the HBrþ reactant (for example in the i ¼ 1=2;
vþ ¼ 1 state) however, even in this case, the rate
coefficient is measured to be a small fraction of the
collision limit indicating either a low propensity
for reactivity or rapid inelastic relaxation of the
prepared state. In order to further elucidate the
nature of these observations and perhaps lend in-
sight to other important reactions such as the
isoelectronic radical reactions [3]
values were: zi ¼ 1 cm; zf ¼ 3 cm; and p0 ¼ 80–300
Torr. The low DBr fractions were used to better
isolate reaction (3b), which was much slower than
reaction (3a). An additional reason to use low
density and low DBr concentration flows was to
minimize clustering that may affect both gas-dy-
namic properties and chemistry. Appearance of
clustered ions was monitored and thereby con-
trolled through mass spectrometry.
þ
State selection of DBr ð Pi; vþÞ cations was
achieved via (2 + 1) REMPI of DBr through the f 3
D2ðv ¼ 0Þ; f 3D2ðv ¼ 1Þ; F 1D2ðv ¼ 0Þ and F 1D2ðv ¼
1Þ Rydberg intermediates, using the R(1) line in all
cases (the corresponding wavelengths were
k ¼ 269:18, 263.30, 259.44, 253.98 nm). According
to investigations by Xie and Zarþe [1], REMPI
2
OH þ HBr ! H2O þ Br
OH þ H2 ! H2O þ H
ð2aÞ
ð2bÞ
we have investigated the state selected reactions
2
through these states produces DBr ð Pi; vþÞ ions
DBrþ2PiðvþÞ þ DBr ! D2Brþ þ Br
DBrþ2PiðvþÞ þ D2 ! D2Brþ þ D
ð3aÞ
ð3bÞ
almost exclusively (greater than 98%) in a single
spin–orbit and vibrational state, namely, ði ¼
3=2; vþ ¼ 0Þ; ði ¼ 3=2; vþ ¼ 1Þ; ði ¼ 1=2; vþ ¼ 0Þ,
and ði ¼ 1=2; vþ ¼ 1Þ, respectively. In all cases, we
chose laser wavelengths, which led to equal effi-
ciency of D79Brþ and D81Brþ production.
at the low translational temperatures within a free
jet expansion [4]. This study has not only increased
the number of available internal energy states
presented to these reaction systems, but also ex-
plores the role of isotope effects [5–8] within both
the reactive and inelastic pathways.
Loss of DBrþ ions along the free jet axis is
caused by the reactions (3a) and (3b), as well as by
geometric expansion of the flow
Àd½DBrþŠ=dt ¼ ka½DBrŠ½DBrþŠ þ kb½D2Š½DBrþŠ
þ ½DBrþŠf ðtÞ:
ð4Þ
2. Experiment
Here ka and kb are the rate coefficients of reaction
(3a) and (3b) and the function f ðtÞ describes the
geometric expansion of the flow and is assumed to
be equal for charged and neutral particles [4]. In-
putting a as the fraction of DBr molecules in the
D2 þ DBr mixture, a ¼ ½DBrŠ=ð½DBrŠ þ ½D2ŠÞ; N1þ
as the fraction of DBrþ ions, N1þ ¼ ½DBrþŠ=
ð½DBrþŠ þ ½D2BrþŠÞ; the total gas density n ¼
½DBrŠ þ ½D2Š; and an effective rate coefficient
A detailed description of the free jet flow reac-
tor, including preparation of reagent ions using
REMPI and the monitoring of products by time-
of-flight mass spectrometer, can be found in [9]. A
pulsed axisymmetric nozzle withan orifice diam-
eter of 0.4 mm was used as a gas source employing
D2 buffers seeded with0.4–2.8% DBr. The REMPI
laser beam was focused onto the center streamline
of the subsequent jet at some distance, zi, from the
nozzle. The laser radiation in the range of interest,
k ¼ 250–270 nm (Dk K 0:01 nm), was produced
from the frequency doubled (BBO) output of a
Nd/YAG pumped dþye laser (Coumarin 503). The
kab ¼ kaa þ kbð1 À aÞ:
ð5Þ
Eq. (4) can be simply expressed in a first order
linear form:
ÀdN1þ=dt ¼ nkabN1þ:
ð6Þ
2
newly prepared DBr ð Pi; vþÞ cations contained in
Integration of Eq. (6) over the distance along the
free jet axis from the point of ionization, zi, to th e
point of extraction, zf , gives the final expression
that was used for data analysis
the field free core of the free jet flow moved and
interacted withfree jet core species until at some
distance, zf , they were extracted into the time-of-
flight mass spectrometer. Typical experimental

A.E. Belikov, M.A. Smith / Chemical Physics Letters 358 (2002) 57–63
59
(3a) are independent of ionic state and much
higher than rates of the reactions with H2 and D2,
which are sensitive to the internal energy of ionic
reactants. The differences between rates of reaction
(3b) and zero are statisticallyþinsignificant for all
ði ¼ 1=2; vþ ¼ 1Þ, and even the rate coefficient for
that highest ionic state is still much lower than the
Langevin value for ion-molecule capture.
lnðN1þÞ ¼ kabð1=zi À 1=zf Þn0z02=u;
ð7Þ
where n0 is the stagnation density, z0 ¼ aD, where
D is the nozzle diameter, a is a constant, which is
dependent on the heat capacity ratio c ¼ Cp=Cv
and is equal to 0.401 or 0.298 for c ¼ 5=3 and 7/5,
respectively. Here we used a model of spherical
expansion [10]: n ¼ n0ðz=z0ÞÀ2, which works very
well in a hypersonic region of the flow and is
confirmed for measurements along the center
streamlines of the jet flow.
The ratio between DBrþ and D2Brþ concen-
trations was measured by their corresponding
time-of-flight signals while varying the stagnation
gas density, n0, for a number of DBr concentra-
tions in the mixture, a. The effective rate coefficient
kab was derived via a best fit to Eq. (7) for each
particular value of a. Finally, dependence of kab on
a allowed for determination of the rate coefficients
ka and kb, according to Eq. (5).
2
studied states of the ions DBr ð Pi; vþÞ, excluding
An energy diagram for reaction (3b) is presented
in Fig. 1 for a number of low rotational levels of
DBrþ and D2 restricted by temperature and
photodynamics. The energy of every level is refer-
enced to the lowest spin–orbit, vibrational and
rotational state of both reactants. The dashed lines
display the boundary of exothermicity with the
uncertainty calculated using heats of formation for
all the species of reaction (3b) at T ¼ 0 K [12]. The
rotational levels of the DBrþ ions are restricted in
accordance withthe results of Xie and Zare [1],
who observed a measurable population of only a
few levels surrounding J, the rotational level of the
intermediate state selecteþd by the laser. According
3. Results
2
to Fig. 1, only the DBr ð P1=2; vþ ¼ 1Þ and per-
2
The rate coefficients of the reactions (1a), (1b)
and (3a), (3b) are presented in Table 1. The results
for bothisotopes are very similar in all cases.
There is no apparent kinetic isotope effect for ei-
ther reaction. The rates of the reactions (1a) and
haps a small fraction ð P1=2; vþ ¼ 0Þ collisions are
able to surmount the reaction barrier with D2.
The rate coefficients k1a andk3a are presented in
Fig. 2 as a function of excess energy that is defined
as
Table 1
Rate coefficientsa of the state-specific ion-molecule reactions
ðH2; HBrÞ þ HBrþ2PiðvþÞ ! H2Brþ þ ðH; BrÞ
and
ðD2; DBrÞ þ DBrþ2PiðvþÞD2Brþ þ ðD; BrÞ
Ionic state
2P3=2; vþ ¼ 0
2P3=2; vþ ¼ 1
2P1=2; vþ ¼ 0
2P1=2; vþ ¼ 1
For HBrþ
b
Neutral reactant
HBr
H2
13:8 Æ 1:6c
15:0 Æ 1:3
16:6 Æ 0:9
12:8 Æ 1:6
0:016 Æ 0:021
0:047 Æ 0:017
0:084 Æ 0:010
0:207 Æ 0:019
For DBrþ
d
DBr
D2
15 Æ 3
11 Æ 3
13 Æ 3
12 Æ 3
0:016 Æ 0:03
0:002 Æ 0:03
0:015 Æ 0:03
0:20 Æ 0:03
a Units of 10À10 cm3
s
À1. The error limits represent 2r values. The Langevin collision limits are 1:5 Â 10À9 and 1:1 Â 10À9 cm3 sÀ1 for
the HBrþ þ H2 and DBrþ þ D2 systems, respectively.
b Data from [2].
c For thermal conditions [11]: kfHBrþði ¼ 3=2; vþ ¼ 0Þ þ HBrg ¼ ð6:7 Æ 1:6Þ Â 10À10 cm3 sÀ1
.
d Present work.

60
A.E. Belikov, M.A. Smith / Chemical Physics Letters 358 (2002) 57–63
respectively. Here X ¼ H for reaction (1b) or
0.6
0.4
0.2
0
X ¼ D for reaction (3b). The rotational distribu-
tion of the DBrþ ions was assumed to be that of
the HBrþ ions as measured by Wales et al. [13]
DBr+(i=1/2,v=1)
DBr+(i=1/2,v=0)
under the same ionization scheme: N1=2:N3=2:N5=2
:
DBr+(i=3/2,v=1)
DBr+(i=3/2,v=0)
N7=2 ¼ 0:49: 0.34:0.14:0.03, neglecting rotational
relaxation. The terminal rotational distribution of
the H2 in the free jet under our typical conditions
were estimated in [2], based on the data of Latimer
[14], as N0:N1:N2:N3 ¼ 0:16:0.71:0.09:0.04 (withan
accuracy ꢀ 0:04 for all numbers). We assume that
the rotational relaxation times in H2 and D2 are
the same. Taking into account the different nuclear
statistics of the D2 molecules, we estimate N0:N1:
N2:N3 ¼ 0:43:0.30:0.24:0.03 for D2 in the jet.
The heats of formation at T ¼ 0 K for H2;
HBrþ; H; H2Brþ; D2, D were taken from [12] as 0,
262, 51.6, 218, 0, 52.5 kcal/mol, respectively. The
heat of formation for DBrþ was estimated as 261.9
kcal/mol using the heats of formation for the pairs
(D, Brþ) and (H, Brþ) along withthe known zero-
point energies of DBrþ and HBrþ. The heat of
formation for D2Brþ was estimated as
+
J
}
J{D2} = 0
1
2
3
0
1
2
3
þ
Fig. 1. Energy diagram for the reaction DBr ð Pi; vþ; JþÞ þ
D2ðv ¼ 0; JÞD2Brþ þ D. The energy E ¼ 0 is referenced to the
lowest internal energy level of the reactants. The threshold of
exothermicity based on the data from [12] is shown by dashed
lines (witherror limits).
2
0.25
H2 + HBr+
D2 + DBr+
0.2
0.15
0.1
0.05
0
Df H0fD2Brþg ¼ Df H0fH2Brþg þ Df H0fD2Oþg
À Df H0fH2Oþg
¼ 217 kcal=mol:
ð10Þ
Accuracy of the values of Df H0fHBrþg; Df H0
fDBrþg; Df H0fH2Brþg and Df H0fD2Brþg is about
Æ0:4; Æ0:4; Æ0:7 and Æ1:2 kcal/mol, respectively.
These errors dominate the error limits shown in
Figs. 1 and 2.
-0.4
-0.2
0
0.2
0.4
Excess of energy [eV]
þ
Fig. 2. Rate coefficients of the reaction XBr ð Pi; vþÞ þ X2ðv ¼
0Þ ! X2Brþ þ X, (X ¼ H, D) as a function of excess internal
energy (see text, Eqs. (8) and (9)). The horizontal error bars are
dominated by uncertainty in heats of formation for the H2Brþ
and D2Brþ ions.
2
The apparent energy dependent rate coefficients
of reactions (1b) and (3b) appear to matchto each
other surprisingly well. The major point of concern
regards the small absolute value of the rate coef-
ficients for the ions in the highest internal states for
which the exoergicity is in excess of 0.1 eV. In fact,
the values of the rate coefficients measured in the
present work should be strictly considered only as
lower limits, if relaxation processes (spin–orbit or/
and vibrational) are collisionally competitive, since
relaxation has been left out of the current analysis.
It should be noted however, that a first order
treatment of the kinetics including relaxation
suggest that the experimentally determined rate
coefficients for bothreaction (1b) and (3b) are
DEi;v ¼ hEii;v À DðDf H0Þ;
ð8Þ
where DðDf H0Þ is the difference between the T ¼ 0
K heats of formation of products and reactants;
X X
þ
þ
hEii;v
¼
ei;vðJ; J ÞNJ NJ ;
ð9Þ
Jþ
J
where ei;vðJ; JþÞ is the total internal energy of the
þ
reactants, fXBr ð Pi; vþ; JþÞ; X2ðJÞg above the
2
þ
lowest level fXBr ð P3=2; vþ ¼ 0; Jþ ¼ 0Þ; X2ðJ ¼
0Þg and NJ and NJþ are the relative rotational level
populations of the X2 molecule and XBrþ ion,
2

A.E. Belikov, M.A. Smith / Chemical Physics Letters 358 (2002) 57–63
61
accurate within the reported error limits, suggest-
ing an inherent dynamic bottleneck to reactivity
within the states currently probed. Such bottle-
necks may represent restrictions on open product
channels, angular momentum barriers, or restric-
tions not currently recognized.
possible channels for energy transfer are strictly
limited in our low temperature collisions. Trans-
lational and rotational degrees of freedom cannot
readily absorb these large transition energies ac-
cording to momentum and angular momentum
conservation. If the energyþtransfer mechanism is
þ
2
to vibration of the ion DBr (or HBr ) ½ð P1=2; vþ
2
¼ 1Þ ! ð P3=2; vþ > 1ފ, it would still be available
4. Analysis of possible relaxation processes
for driving reactions (1a) and (3a) and should not
necessarily lead to a decrease in the rate coeffi-
cients observed. So, there is primarily only one
channel for collisional quenching of the DBrþ (or
Direct data on spin–orbit or vibrational relax-
þ
2
ation that may lead to decay of the HBr ð P1=2
;
þ
vþ ¼ 1Þ and DBr ð P1=2; vþ ¼ 1Þ states is not
present in the literature. Imajo et al. [15] measured
the thermal rate of the reaction
þ
2
2
HBr Þ ð P1=2; vþ ¼ 1) states that can directly in-
fluence the interpretation of the results displayed
in Fig. 2: ion relaxation via vibrational excitation
of the DBr (HBr) or D2 ðH2Þ molecules. The rate
coefficients for the reactions (1b) and (3b) were
measured in the limit of a very low fraction of
heavy molecules in the mixture to exclude inter-
ference with the reactions (1a) and (3a). Thus, the
effect of collisional relaxation between the ions and
parent molecules may be significant only if the rate
coefficient is more than the rate of the reactions
(1a) and (3a), that is 1:5 Â 10À9 cm3 sÀ1. Xie and
Zare [1] used REMPI in the mixture of DBr + HBr
(2:1) and monitored internal states of ions by laser
induced fluorescence (LIF). They observed only
one state present for the reactant ion distribution
throughout the reaction time (i.e., over many col-
lisions): ½2P3=2; vþ ¼ 0Š, or ½2P3=2; vþ ¼ 1Š, or
½2P1=2; vþ ¼ 0Š, or ½2P1=2; vþ ¼ 1Š, depending on
intermediates chosen in the REMPI scheme. In
particular, they observed that > 98% of the DBrþ
COþðA2P1=2; vþ ¼ 1Þ þ He
¢ COþðA2P3=2; vþ ¼ 1Þ þ He
ð11Þ
and found the rate coefficient of 1:7 Â 10À10 cm3
sÀ1 for the forward and 1:3 Â 10À10 cm3 sÀ1 for the
reverse transition. Sudbo and Loy [16,17] studied
the spin–orbit relaxation for neutral NO molecule
in specific rotational and vibrational states
2
NOð P3=2; v; JÞ þ ðNO; Ar; He; . . .Þ
¢ NOð P1=2; v; JÞ
2
ð12Þ
and found for v ¼ 2; J ¼ 21=2; kfNOg ¼ 5Â
10À11 cm3
s
À1; kfArg ¼ 6 Â 10À12 cm3 sÀ1; kfHeg
¼ 3 Â 10À12 cm3 sÀ1, which differ significantly. On
the other hand, the rate coefficient for the reverse
2
2
reaction (from the lower P1=2 to the higher P3=2
state), kꢁfNOg ¼ 7 Â 10À12 cm3 sÀ1, is very close
to the value measured by Joswig et al. [18] for the
rotationally specific spin–orbit excitation NO
þ
2
(HBr ) ions were in the ð P1=2; vþ ¼ 1Þ when they
used REMPI through the F 1D2½2P1=2Š Rydberg
state. If collisional relaxation of this spin–orbit-
vibrational state had a rate coefficient ꢀ1:5 Â
10À9 cm3 sÀ1 in collisions withthe HBr/DBr mol-
ecules, relaxation could not have remained unde-
tected in their high HBr/DBr concentrations.
Thus, the only possibly significant channel of re-
laxation left in our experiments is through collision
withlight molecules.
2
2
ð P1=2; J ¼ 1=2 ! P3=2; J ¼ 5=2Þ in collisions
withHe ð8 Â 10À12 cm3 sÀ1Þ and Ar ð7 Â 10À12
cm3 sÀ1Þ. The same order of magnitude ðꢀ10À11
cm3 sÀ1Þ was measured for vibrational relaxation
of neutrals: NO(v ¼ 3)/NO [19], CH(v ¼ 1)/(CO,
N2; H2) [20], while the vibrational quenching of
NOþ (v ¼ 1) by Ar, N2 and He were found to be
muchfaster under ultra-low temperatures ð9 Â
10À10; 1 Â 10À9; 1 Â 10À10, respectively) [21].
In spite of some inconsistency in the cited ex-
perimental data, we must conclude that the rates
of spin–orbit relaxation could be high enough to
affect our measurements of the rate coefficients for
reactions (1a), (1b) and (3a), (3b). However,
Consideration of kinetics for the relative con-
centration of the D2Brþ ions, including DBrþ=D2
relaxation withthe rate coefficient
kꢁ, gives the
following expression for the point of ion extrac-
tion, zf ,

62
A.E. Belikov, M.A. Smith / Chemical Physics Letters 358 (2002) 57–63
½D2BrþŠ
cients, measured in the present work, and that the
low range of the relaxation rates ½kꢁ ¼ ð0 À 1Þ Â
10À10 cm3 sÀ1Š and consequent reaction rate coef-
ficients kb ¼ ð2 À 4Þ Â 10À11 cm3 sÀ1 provide the
least deviation to our experimental results.
½D2BrþŠ þ ½DBrþŠ
ꢀ
ꢁ
kab
kꢁꢁ
¼
1 À expð À kꢁꢁð1=zi À 1=zf Þn0z20=uÞ ;
ð13Þ
where kꢁꢁ ¼ kab þ kꢁð1 À aÞ and all other values are
5. Conclusion
specified in Section 2.
Unfortunately, it is impossible to derive all rate
coefficients ka, kb; kꢁ independently from our ex-
perimental data. But if we begin withan assumed
rate coefficient, kꢁ for the relaxation process, we
can find the rate coefficient, kb, through best fit of
the right side of Eq. (13) to experimental data.
This function kb (kꢁ) is shown by line 1 in Fig. 3
(the rate coefficients, ka, are taken as 1:3 Â
10À9 cm3 sÀ1). The confidence intervals presented
by lines 2 answer to the random experimental er-
ror. Finally, line 3 shows the normalized sum of
the squared deviations between the left-hand side
(experimental data) and right-hand side (model
result) of Eq. (13). From this comparison we can
conclude that the relaxation processes do not
dramatically change the determined rate coeffi-
Rateþcoefficients for a variety of internal states
2
of DBr ð Pi; vþÞ withDBr were measured to be
near collision limited values as seen previously for
HBrþ withHBr and significantly greater tahn
values observed at 300 K [11]. The rate coefficients
for reaction of the same states of DBrþ withD
2
were found to be insignificant unless the total in-
ternal energy was in excess of the reaction barrier.
In this case, as with HBrþ þ H2, reaction proceeds
at a value of only a few percent of the collision
frequency. Simple analysis suggests this is not an
artifact of ignoring relaxation in our kinetic anal-
ysis while more detailed consideration will be
presented in future work. The rate coefficients of
the DBrþ=D2 and HBrþ=H2 reactions indicate a
common dependence on total energy. It is cur-
rently suggested that the slow rate for clearly ex-
oergic channels may suggest a dynamic barrier not
accounted for in the simple thermodynamic anal-
ysis.
1
1
0.8
2
3
0.6
Acknowledgements
0.4
0.2
0
The authors gratefully acknowledge financial
support of this work by the National Science
Foundation through Grant No. CHE-9984613.
References
0
0.5
1
1.5
2
Rate coefficient, k*, [10-10 cm3/s]
[1] J. Xie, R. Zare, J. Chem. Phys. 96 (1992) 4293.
[2] A.E. Belikov, C. Mullen, M.A. Smith, J. Chem. Phys. 114
(2001) 6625.
Fig. 3.
2
Rate
coefficient
of
the
reaction,
DBrþ
ð P1=2; vþ ¼ 1Þ þ D2 ! D2Brþ þ D, fit to experimental data in
the framework of the relaxation model Eq. (13) including decay
[3] N.I Bukovskaya, D.W. Setser, J. Phys. Chem. 1996 (1996)
4853.
of the ð P1=2; vþ ¼ 1Þ state by collisions withD molecules.
2
2
Lines 1 and 2 show the value and error limits of the reaction
rate coefficient, kb, at eachparticular relaxation rate coefficient
value, kꢁ. Deviation of the model from experimental data (in
arbitrary units) is shown by line 3.
[4] D.R. Miller, in: G. Scoles (Ed.), Atom and Molecular
Beam Methods, Oxford University Press, New York, 1987.
[5] A.A. Viggiano, R.A. Morris, J. Chem. Phys. 100 (1994)
2748.

A.E. Belikov, M.A. Smith / Chemical Physics Letters 358 (2002) 57–63
63
[6] J. Bedjanian, V. Riffault, G. Le Bras, G. Poulet, J.
Photochem. Photobiol. A: Chem. 128 (1999) 15.
[7] V.I. Jaramillo, M.A. Smith, J. Phys. Chem. A 105 (2001)
5854.
[13] N.P.L. Wales, W.J. Buma, C.A. de Lange, H. Lefebvre-
Brion, K. Wang, V. McKoy, J. Chem. Phys. 104 (1996)
4911.
[14] D.R. Latimer, Ph.D. Thesis, University of Arizona, Tuc-
son, 1994.
[8] J.M. Nicovich, C.A. van Dijk, K.D. Kreutter, P.H. Wine,
J. Phys. Chem. 95 (1991) 9890.
[15] T. Imajo, T. Imamura, I. Koyano, Chem. Phys. Lett. 223
(1994) 99.
[9] M.A. Smith, M. Hawley, in: N.G. Adams, L.M. Babcock
(Eds.), Advances in Gas Phase Ion Chemistry, JAI Press,
Greenwich, 1992, p. 167.
[16] A.S. Sudbo, M.M.T. Loy, Chem. Phys. Lett. 82 (1981) 135.
[17] A.S. Sudbo, M.M.T. Loy, J. Chem. Phys. 76 (1982) 3646.
[18] H. Joswig, P. Andresen, R. Schinke, J. Chem. Phys. 85
(1986) 1904.
[10] H. Ashkenas, F. Sherman, in: Rarefied Gas Dynamics,
Proceedings of the 4th International Symposium, vol. 2,
Academic Press, New York, 1966, p. 84.
[19] P.L. James, I.R. Sims, I.W.M. Smith, Chem. Phys. Lett.
276 (1997) 423.
[11] R.J. Green, J. Xie, R.N. Zare, A.A. Viggiano, R.A.
Morris, Chem. Phys. Lett. 277 (1997) 1.
[20] L.B. Herbert, I.R. Sims, I.W.M. Smith, D.W.A. Stewart,
A.C. Symonds, A. Canosa, B.R. Rowe, J. Phys. Chem. 100
(1996) 14928.
[12] S.G. Lias, J.E. Bartmess, J.F. Liebman, J.L. Holmes, R.D.
Levin, W.G. Mallard, J. Phys. Chem. Ref. Data 17 (Suppl.
1) (1988) 1.
[21] M. Hawley, M.A. Smith, J. Chem. Phys. 95 (1991) 8662.