Generation of NBr(a1Δ) by the reaction of N3 radicals with Br atoms: A flow reactor source for quenching rate constant measurements

Article abstract of DOI:10.1016/S0009-2614(98)01116-6

The reaction between azide radicals (N₃) and Br atoms is shown to produce electronically excited NBr(a¹Δ) molecules in a room temperature flow reactor. This chemical system provides adequate concentration of NBr(a¹Δ) so that this molecule can be systematically studied. The yield of NBr(b¹Σ⁺) is minor. The quenching reactions of NBr(a) with HCl, HBr, HI, NH₃, Br₂, CF₂Br₂, and O₂ were examined; the rate constants are (22 ± 5) × 10^-14, (280 ± 30) × 10^-14, (2300 ± 200) × 10^-14, (35 ± 3) × 10^-14, (2600 ± 300) × 10^-14, (37 ± 6) × 10^-14, and (230 ± 30) × 10^-14 cm³ molecule^-1 s^-1, respectively.

Full text of DOI:10.1016/S0009-2614(98)01116-6

27 November 1998
Ž
.
Chemical Physics Letters 297 1998 335–342
1
ž /
Generation of NBr a D by the reaction of N3 radicals with Br
atoms: a flow reactor source for quenching rate constant
measurements
)
Kevin B. Hewett, D.W. Setser
Department of Chemistry, Kansas State UniÕersity, Manhattan, KS 66506, USA
Received 3 August 1998; in final form 29 September 1998
Abstract
The reaction between azide radicals N3 and Br atoms is shown to produce electronically excited NBr a D molecules in
1
Ž
.
Ž
.
1
Ž
.
Ž .
a room temperature flow reactor. This chemical system provides adequate concentration of NBr a D so that this molecule
can be systematically studied. The yield of NBr b S is minor. The quenching reactions of NBr a with HCl, HBr, HI,
1
q
Ž
.
y
14
y14
Ž
.
Ž
.
Ž
.
NH3, Br2, CF2Br2, and O2 were examined; the rate constants are 22"5 =10 , 280"30 =10 , 2300"200 =
14
y
14
y
14
y
14
y14
10y , 35"3 =10 , 2600"300 =10 , 37"6 =10 , and 230"30 =10
cm3 moleculey1 sy1, respec-
Ž
.
Ž
.
Ž
.
Ž
.
tively. q 1998 Elsevier Science B.V. All rights reserved.
1. Introduction
tronic states of NX are X3 Sy, a1D and b1 Sq, with
the latter being 9281 and 14834 cmy1 above the
w
x
ground state of NBr, respectively 1,2 . The enthalpy
Electronically excited molecules can serve as en-
ergy storage systems, provided that the radiative
decay rate is small and that the generation of the
excited state is easily accomplished by chemical
of formation of NBr is y206 kJ moly1 3 . Due to
w x
the forbidden nature of the a1D™X3 Sy transition,
the a1D state is long-lived. The lifetimes are 5 s for
Ž . w
x
Ž . w
x
NF a 4,5 and 2–3 s for NCl a 6–8 . The lifetime
Ž
means. The nitrogen monohalides NX, XsF, Cl, or
Ž .
of NBr a seems to be 5–10 times smaller than that
.
Br have been proposed as candidates for such en-
Ž .
of NCl a based upon computations and observation
of decay times in an Ar matrix 9,10 . Chemical
generation of NX a D molecules is accomplished by
ergy storage systems. The three lowest energy elec-
w
x
1
Ž
.
Ž
.
the reaction of azide radicals N3 with halogen
atoms, and in this Letter we show that BrqN3 can
Ž .
be used as a source of NBr a in a flow reactor.
Azide radicals are produced by the reaction of
hydrazoic acid with F atoms with a rate constant of
)
Corresponding author. E-mail: setserdw@ksu.edu
0009-2614r98r$ - see front matter q 1998 Elsevier Science B.V. All rights reserved.
Ž
.
PII: S0009-2614 98 01116-6

(
)
336
K.B. Hewett, D.W. SetserrChemical Physics Letters 297 1998 335–342
1.1=10y10 cm3 moleculey1 sy1 11 . Recent mea-
surements of the branching fraction of
2. Experimental methods
w
x
Ž .
Ž .
The production of NBr a by reaction 2 and the
subsequent quenching reaction studies were carried
out in a flow reactor at room temperature. The
FqHN3 ™ HFqN3 , Dr H2o98 sy180 kJ moly1
.
Ž .
Ž .
1
Ž .
relative concentrations of NBr a and NF a were
measured by monitoring the chemiluminescence in-
tensity at 1077 and 874 nm of the a™X transitions,
respectively. The flow reactor and the emission mon-
itoring system, which are the same as previously
Ž .
Reaction 1 demonstrated conclusively that N3 is
the major product with a branching fraction of 0.97
w
x
Ž . w
x
12 . The resulting azide radicals react with Br atoms
used for the study of NCl a 20,21 , are described
below.
Ž .
to form NBr. Reaction 2 ,
2.1. Flow reactor
BrqN3
The flow reactor was a 150 cm long, 7.4 cm
diameter Pyrex tube. The surface of the flow reactor
™ N qNBr X3 Sy
,
Dr H2o98 sy206 kJ moly1
Dr H2o98 sy95 kJ moly1
Dr H2o98 sy29 kJ moly1
,
Ž
.
2
™ N qNBr a1D ,
Ž
was coated with halocarbon wax Halocarbon Prod-
Ž
.
2
.
ucts, series 600 to prevent the loss of F, Br, N3 and
™ N qNBr b1 Sq
,
Ž
.
2
other radical species by reaction with the reactor
walls. The flow reactor consisted of two distinct
regions; a pre-reactor where NBr was generated by
2
Ž .
Ž .
Ž .
reactions
1
and 2 , and the reactor where the
Ž .
has previously been observed to occur and produce
Ž . Ž . Ž . w
quenching reactions of NBr a occurred. The pre-re-
actor was 37 cm in length. The inlet tubes for the Ar
carrier gas, the F atom source CF4, the Br atom
source CF2 Br2 , and HN3 were located in an alu-
x
NBr a and possibly some NBr X and NBr b 2,13 .
w
x
Ž
A study by Liu et al. 13 reported rate constants for
reactions between azide radicals and halogen atoms,
however, recent work has shown that their reported
rate constant for the reaction between ClqN3 was
Ž
.
minum flange attached via ano-ring joint to the front
Ž
.
end of the flow reactor. The majority ;90% of the
Ar carrier gas was added to the pre-reactor through a
perforated ring; the F atoms and Br atoms were
added with the remainder of the Ar through separate
10 mm o.d. quartz tubes; and the HN3 was added
through another perforated tube located at the exit of
the F and Br inlets. The quenching reagent gases
were added to the flow reactor via a perforated ring,
which was located 37 cm downstream of the HN3
inlet and marked the starting point of the reactor.
w
x
w x
too large 6,14–17 . Their reported value 13 of
y10
cm3 moleculey1 sy1 can be consid-
Ž
.
3"2 =10
Ž .
ered an upper limit for reaction 2 and the true value
is probably near 5=10y11 cm3 moleculey1 sy1, the
w
x
accepted value for FqN3 13,18 .
Ž .
The quenching rate constants of NBr a are im-
portant for the development of energy storage sys-
tems and for comparison with the other nitrogen
1
1
Ž
. w
x
Ž
. w x
monohalides NF a D 18,19 and NCl a D 20 . In
this study we develop a method that is suitable for
systematic studies employing a microwave discharge
in CF2 Br2 as the Br atom source. As a test of the
method, we measured the magnitudes of the quench-
The reactor was pumped using
a mechanical
pumprblower combination which provided a linear
flow velocity of 1200 cm sy1 at a typical Ar pres-
sure of 0.4 Torr
Ž .
ing rate constants of NBr a with the molecules
The flow velocity in the reactor corresponds to a
minimum reaction time of 35 ms in the pre-reactor.
Under typical reaction conditions, this time was suf-
NH3, Br2 , CF2 Br2 , HCl, HBr, HI, and O2 and
compare them with those of NF a and NCl a . The
comparison with NCl a is especially interesting since
Ž .
Ž .
Ž .
Ž .
Ž .
ficient to allow reactions 1 and 2 to proceed to
completion before the addition of the quenching
the reported energies are 9280"10 and 9281"10
cmy1 for NCl a and NBr a , respectively 1,2 .
Ž .
Ž .
w
x
Ž .
reagent. The reactions of NBr a with added quench-

(
)
K.B. Hewett, D.W. SetserrChemical Physics Letters 297 1998 335–342
337
ing reagents then occurred during the 47 ms spent
for travel down the flow reactor. The actual reaction
time was determined by measuring the quenching
of a 0.5 m monochromator. The PMT was an S-1
Ž
.
type Hamamatsu R1767 cooled to y808C. The
monochromator has a 600 linesrmm grating blazed
at 1000 nm. Typical dark current was -15 pA. The
Ž .
rate of NF a with NH3, which has an established
rate constant of 3.6=10y12 cm3 moleculey1 sy1
current was monitored using an electrometer Keith-
Ž
w
x
.
18 . The flow velocity could be reduced, if desired,
ley, model 614 and recorded by a computer. The slit
by a throttling valve located before the pump.
The argon carrier gas was purified by passage
width was typically 0.5 mm.
The monochromator was attached to a movable
platform, allowing it to travel along the length of the
flow reactor and thus changing the reaction time at
which observations were made. The monochromator
could be scanned to produce continuous spectra or it
was set to a fixed wavelength to monitor the maxi-
Ž
through cooled 196 and 77 K for high and low
.
pressure, respectively molecular sieve filled traps.
Approximately 10% of the Ar was used to dilute the
CF4 and ArrCF2 Br2 flows before they passed
through the microwave discharges.The F atoms were
Ž .
Ž .
produced by passing the resulting ArrCF mixture
mum of the NBr a or NF a emission. For the fixed
wavelength measurements, the computer read the
output from the electrometer at ;0.5 s intervals.
The collection time was typically 20 s for each data
point.
4
through a microwave discharge in an alumina tube.
Under our experimental conditions, the CF4 is nearly
completely dissociated into 2F q CF2 . The
ArrCF2 Br2 mixture was passed through a second
microwave discharge in an alumina tube to produce
the Br atoms. The dissociation efficiency of CF2 Br2
is not known; however, it was assumed to be similar
to that of CF2Cl2 whose dissociation efficiency has
been measured in our laboratory to be 1.5 Cl atoms
3. Proof of method
w
x
w x
per CF2Cl2 molecule 21 .
Previous work by Pritt et al. 2 determined that
The gases CF2 Br2 , HCl, HBr, and NH3 were
obtained from commercial vendors, purified via
freeze–pump–thaw cycles, and stored as pure gases
or as dilute mixtures in Ar. The O2 and CF4 were
obtained from commercial vendors and used without
purification. The HI was synthesized by the reaction
the formation of excited NBr molecules occurs upon
reaction of azide radicals with Br atoms. They de-
Ž .
tected chemiluminescence from both the NBr a and
NBr b electronic states using F atoms, Br2 and HN3
as reactants. We set out to determine experimental
conditions under which emission from NBr a was
strong enough to measure and relatively constant
along the length of our flow reactor, so that quench-
ing rate constants could be measured.
Ž .
Ž .
w
x
of iodine with boiling tetrahydronaphthalene 22 ,
distilled and stored as a mixture in Ar. The HN3 was
prepared by the reaction of excess stearic acid with
Ž
.
sodium azide NaN3 heated to 363 K under vacuum
1
w
x
18 . The product, which was collected in a 12 l
Pyrex reservoir, was diluted to 10% with argon. The
flow rates of Ar, CF4, and CF2 Br2 were controlled
by needle valves and measured by Hastings mass
flow meters. The flow rate of HN3 and the quench-
ing gases were controlled by needle valves and
measured by the rate of pressure rise in a calibrated
volume.
2.2. Detection system
Ž .
The emission from the NBr a was monitored by
a photomultiplier tube PMT attached to the exit slit
w
Ž
.

(
)
338
K.B. Hewett, D.W. SetserrChemical Physics Letters 297 1998 335–342
12
w x
initial reactant concentrations of F f2.0=10 ,
11
12
w
x
w
x
Br f4.5=10 , and HN3 f3.0=10 molecule
cmy3. The atom concentrations are based upon F s
w x
w
x
w
x
w
x
2 CF4 and Br s1.5 CF2 Br2 . Using these condi-
Ž .
tions, the emission intensity profiles of NBr a ,
Ž .
Ž .
NBr b and NF a as a function of time were mea-
sured and they are shown in Fig. 1. The emission
Ž .
intensity from NBr a decreased by ;35% from
35–100 ms which is satisfactory for quenching stud-
Fig. 2. The emission spectrum, corrected for detector response,
from the flow reactor after a reaction time of 40 ms. The reactant
16
12
w
x
w
x
concentrations were Ar s 1.4= 10
,
CF4 s 1.1= 10
,
.
CF2 Br2 s3.0=1011, and HN3 s3.0=10 molecule cmy3
Emission was observed from NBr a D , NBr b S , NF a D and
12
w
x
w
x
1
1
1
q
Ž
.
Ž
.
Ž
.
3
Ž
.
.
N2 B Pg
Ž .
ies. The fast rise of the NF a emission intensity
w
x
w x
results from the excess HN3 , which removes the F
Ž .
early in the pre-reactor. The slow rise of NBr a
emission indicates that the magnitude of the rate
Ž .
constant for reaction 2 is comparable to that for the
FqN3 reaction, 5=10y11 cm3 moleculey1 sy1
.
x
w
We cannot assign a better estimate because the Br
was not measured. It was observed that higher
Ž .
CF2 Br2 flows actually reduced the NBr a emission
intensity, this is believed to be the result of quench-
Ž .
ing of NBr a by Br atoms, since CF2 Br2 has a
negligible quenching rate at this concentration.
Fig. 2 shows a spectrum, corrected for detector
response, recorded after a reaction time of 40 ms.
Ž .
The ratio of emission intensity from NBr a and
Ž .
NBr b is about 24:1, which corresponds to a con-
centration ratio of 28000:1 using the Einstein A
coefficients 6.57 and 7651.4 sy1 for NBr a and
Ž .
Ž .
w
x
NBr b , respectively 9,10 . Even after a reaction
time of only 5 ms, when the emission intensity ratio
Ž .
Ž .
of NBr a and NBr b is approximately unity, the
concentration ratio is 1165:1. The A coefficient for
1
1
Ž
. w
x
Ž
q . w
x
Fig. 1. The emission intensity of NBr a D ' , NBr b S
v ,
1
Ž .
w
Ž .x
Ž
. w
B
x
NBr b is large enough that NBr b reaches a
steady-state value and does not accumulate in the
reactor. As a result, the NBr b r NBr a ratio is
always small, much less than 1%. The main source
and NF a D
as a function of time. The initial reactant
16
12
w
x
w
x
concentrations were Ar s 1.4= 10
,
CF4 s 1.0= 10
,
.
CF2 Br2 s2.3=1011, and HN3 s3.5=10 molecule cmy3
12
w
x
w
x
w
Ž .x w
Ž .x
Ž .
The NBr a emission reaches its maximum intensity in 35–40 ms
Ž .
Ž .
and then begins to slowly decay. NBr b and NF a reach their
maximum intensity in 15–20 ms. The quenching agents were
added 35 ms after the start of the BrqN3 reaction.
Ž .
of NBr b at each time is believed to be from the
Ž .
excitation of NBr a by vibrationally excited HF

(
)
K.B. Hewett, D.W. SetserrChemical Physics Letters 297 1998 335–342
339
Ž .
Ž .
molecules formed by reaction 1 ; this NBr b excita-
tion mechanism is analogous to the mechanism for
Ž .
Ž . w x
Ž .
production of NF b and NCl b 6 . The N2 B
emission seen in Fig. 2 is from nitrogen impurity
passing through the microwave discharges.
4. Rate constants
After determining the optimum conditions for
Ž .
producing NBr a , we set out to determine some
quenching rate constants. The optimum conditions
Ž .
for NBr a production gave well-behaved, pseudo-
first-order decay kinetics for the addition of a reagent.
The molecule NH3 was used as a standard reference
Ž .
Ž . w
x
in previous work on NF a and NCl a 6,18,20 , and
we utilized it to calibrate the reaction time based
1
Ž
. w
x
Fig. 3. Stern–Volmer plot for the quenching of NBr a D v and
Ž .
upon the residual NF a quenching. Quenching by
1
Ž
. w
w
x
w
x
NF a D B by NH3. The reactant concentrations were Ar s1.4
12
11
=1016, CF4 s1.1=10 , CF2 Br2 s2.7=10 , HN3 s2.5=
Br2 was examined since FqBr2 has previously been
x
w
x
w
x
1012 , and NH3 F3=10 molecule cmy3. The emission was
13
w x
used as a Br atom source 2 . We also examined
w
x
monitored 45 ms after the addition of NH3. The time was
CF2 Br2 , our Br atom source. We chose the hydrogen
Ž .
calculated from the NF a quenching plot and the known rate
Ž .
halides to give a systematic comparison with NF a
constant 360=10y14 cm3 moleculey1 sy1 see text . The NBr a
Ž
.
Ž .
Ž .
Ž .
and NCl a reactions. The quenching of NBr a by
oxygen was also measured because this should be an
excitation–transfer reaction.
quenching rate constant from the plot was determined to be
37=10y14 cm3 moleculey1 sy1. The average value obtained
y14
from two experiments was 35"3 =10
cm3 moleculey1 sy1
.
Ž
.
4.1. Ammonia
w x
Pritt et al. 2 and we used the microwave discharge
of CF2 Br2. Erroneous results can be obtained if the
residual precursor gases cause significant quenching
Ž .
The reaction between NBr a and NH3 was exam-
13
ined by adding NH3 F4=10 molecule cmy3 to
w
x
Ž .
the flow reactor and monitoring the NBr a emission
Ž .
w
w
x
of NBr a . Reagent concentrations of Br2 F2=
Ž .
intensity after 45 ms. The NF a emission was also
1012 or CF2 Br2 F7=10 molecule cmy3 were
12
x
monitored under the same conditions. Fig. 3 shows
Ž .
added to the reactor. Measuring the NBr a emission
Ž .
the Stern–Volmer plot for the quenching of NF a
after 47 ms for various Br2 and CF2 Br2 concentra-
Ž .
and NBr a by NH3; both reactions follow pseudo-
Ž
tions gave quenching rate constants of kBr s 2600
2
first-order kinetics. The reaction time was set by
y14
y14
cm3
.
Ž
.
"300 =10
and kCF s 37"6 =10
2 Br2
Ž .
using the accepted value for NF a quenching rate
moleculey1 sy1
.
constant of 360=10y14 cm3 moleculey1 sy1 18 .
w
x
Ž .
The rate constant for NBr a was determined to be
y14
kNH s 35"3 =10
cm3 moleculey1 sy1 based
Ž
.
4.3. Hydrogen halides
3
upon two separate measurements.
Ž .
Quenching of NBr a upon addition of hydrogen
4.2. Bromine and CF2 Br2
halides into the reactor was also studied. Reagent
14
w
x
w
x
concentrations of HCl F3=10 , HBr F8=
12
w
x
Two sources of Br atoms in flow reactors have
been utilized in studies on NBr; FqBr2 was used by
1013, and HI F5=10
molecule cmy3 were
added to the reactor. The quenching rate constants

(
)
340
K.B. Hewett, D.W. SetserrChemical Physics Letters 297 1998 335–342
culey1 sy1 for NF a , NCl a , and NBr a , respec-
Ž .
Ž .
Ž .
Ž .
Ž .
tively. The rate constants for NF a and NCl a are
within the uncertainty of the previously reported
w
x
Ž .
values 20 . The NBr a rate constants with the hy-
drogen halides are kHCl s 22"5 =10y14, kHBr
s
Ž
.
y14
Ž
.
Ž .
, and kHI s 2300 " 300 =
280 " 30 = 10
10y14 cm3 moleculey1 sy1. In each case, the decay
Ž .
also was measured for NF a to ensure that condi-
Ž .
tions were satisfactory. For HI, the NF a quenching
rate constant has not been reported and we deter-
y14
Ž
.
mined it to be 380"50 =10
cm3 moleculey1
sy1. The fast FqHBr reaction could be used as an
alternative source of Br atoms for reaction 2 in a
Ž .
w x
carefully designed flow reactor 6 .
4.4. Oxygen
1
1
Ž
. w .
x
Ž
Fig. 4. Stern–Volmer plot for quenching of NF a D v , NCl a D
The final quenching study was between O2 and
1
w
w
w
x
Ž
. w x
16
B , and NBr a D ' by HBr. The reactant concentrations were
12
Ž .
w
x
12
11
NBr a . Concentrations of O2 F6=10 molecule
x
w
x
w x
CF2 Br2 s 2.7=10 ,
13
Ar s 1.4=10
,
CF4 s 1.1=10
,
cmy3 were added to the reactor and the NBr a
HN3 s2.5=1012 , and HBr F7=10 molecule cmy3. The
Ž .
x
w
x
reaction time was 47 ms. The quenching rate constants obtained
emission monitored after 45 ms. The rate constant
from this plot are 10.2=10y14, 187=10y14, and 368=10y14
Ž
.
from a good first-order plot was kO s 230"30 =
cm3 moleculey1 sy1 for NF a , NCl a and NBr a , respectively.
2
Ž .
Ž .
Ž .
10y14 cm3 moleculey1 sy1
.
were determined by monitoring the emission inten-
5. Conclusions
Ž .
sity of NBr a after a reaction time of 45 ms. Fig. 4
Ž .
Ž .
Ž .
compares the quenching of NF a , NCl a and NBr a
by HBr. The rate constants obtained are 10.2=
10y14, 187=10y14, and 368=10y14 cm3 mole-
The chemiluminescence of NBr has been detected
in a flow reactor from the reaction of azide radicals
Table 1
Ž . Ž .
Ž .
Comparison of quenching rate constants of NBr a with NF a and NCl a
Ž . c
a
b
Ž .
NF a
Ž .
NCl a
Reagent
NBr a
cm3 moleculey1 sy1
cm3 moleculey1 sy1
cm3 moleculey1 sy1
.
y14
y14
y14
Ž
10
.
Ž
10
.
Ž
10
NH3
Br2
CF2 Br2
O2
HCl
HBr
HI
360"20
11"2
1600"200
35"3d
3800"600e
2600"300f
37"6d
0.70"0.07
0.16"0.03
9.0"2.0b
380"50c
300"20
4.0"0.4
180"30
230"30d
22"5d
280"30d
2300"300f
2900"500
a
w
x
Ref. 18 .
b
w
x
Ref. 20 .
c This Letter, see text.
d Uncertainty is the standard deviation of multiple runs.
e
w
x
Ref. 19 .
f Uncertainty is from a least-squares fit to a pseudo-first-order plot.

(
)
K.B. Hewett, D.W. SetserrChemical Physics Letters 297 1998 335–342
341
Ž .
Ž .
Ž .
with Br atoms. The Br atoms were generated by a
microwave discharge through the precursor gas
CF2 Br2. The emission spectra correspond to the
constants between NF a vs. NCl a or NBr a . A
systematic study of the crossing of the potential
Ž .
energy surfaces for reaction 3 would be interesting
1
1
Ž
.
Ž
q. w
1,2 ;
x
w
x
previously detected NBr a D and NBr b S
Ž . Ž .
23 .
the yield of NBr b is -1% of the NBr a yield. For
NX a1D qO X3 S ™ NX X3 Sy qO a1D .
w x w
x
w
x
the same F , HN3 , and CF2 X2 concentrations, the
Ž
.
2 Ž
.
Ž
.
2 Ž
.
Ž .
Ž .
ratio of the NBr a rNCl a emission intensity was
3
Ž .
approximately 1:4. For optimum experimental condi-
Ž .
Ž .
tions of each case, the NBr a rNCl a ratio was
approximately 1:2. The present data strongly suggest
Ž .
that the branching fraction for NBr a formation is
Ž .
substantially smaller than for NCl a in the ClqN3
Acknowledgements
w x
reactions; the latter has been assigned as G0.5 6 .
The NBr a emission is very sensitive to the CF2 Br2
Ž .
This work was supported by the US Air Force
under Grant F49620-96-1-0110 from the Air Force
Office of Scientific Research.
flow rate, as increasing flow drastically decrease the
emission intensity, probably because of quenching
by Br atoms. Conditions could be optimized to pro-
Ž .
duce a NBr a concentration that slowly decreased
over 50 ms. Using these optimized conditions,
quenching rate constants were measured for several
References
Ž
.
molecules. The values obtained are kNH s 35"3
3
=10y14, kO s 230"30 =10y14, kHCl s 22"
Ž
.
Ž
2
5 =10y14, kHBr s 280"30 =10y14, kHI s 2300
M. Bielefeld, G. Elfers, E.H. Fink, H. Kruse, J. Wildt, R.
Winter, F. Zabel, J. Photochem. 25 1984 419.
.
Ž
.
Ž
w x
1
y14
"300 =10y14, kBr s 2600"300 =10
and
Ž
.
.
Ž
.
2
w x
2
y14
A.T. Pritt Jr., D. Patel, R.D. Coombe, J. Mol. Spectrosc. 87
kCF s 37"6 =10
cm3 moleculey1 sy1
.
Comparing the magnitude of the quenching rate
Ž
.
2 Br2
Ž
.
1981 401.
w x
3
A. Kalemos, A. Mavridis, S.S. Xantheas, J. Phys. Chem. A
Ž .
Ž .
Ž .
constants of NF a , NCl a and NBr a with NH3,
shown in Table 1, one notices that they lie in the
Ž
.
102 1998 submitted.
w x
Ž .
4
w x
5
R.J. Malins, D.W. Setser, J. Phys. Chem. 85 1981 1342.
Ž .
G.R. Bradburn, H.J. Lilefeld, J. Phys. Chem. 95 1991 555.
G. Manke II, D.W. Setser, J. Phys. Chem. A 102 1998 in
press.
Ž .
Ž .
Ž .
ing value for NCl a . However, the magnitude of the
order NF a )NBr a )NCl a . The rate constant
w x
6
Ž
.
Ž .
for NBr a is three times larger than the correspond-
Ž .
w x
Ž .
A.C. Becker, U. Schurath, Chem. Phys. Lett. 160 1989 586.
D.R. Yarkony, J. Chem. Phys. 86 1987 1642.
K. Bhanuprakash, P. Chandra, G. Hirsch, R.J. Buenker,
Chem. Phys. 133 1989 345.
7
w x
8
Ž .
rate constant for NF a is the largest because of its
stronger base property 18 . The magnitudes of the
rate constants for the hydrogen halides follow a
Ž
.
w
x
w x
9
Ž
.
w
w
w
w
w
x
10 A.C. Becker, K.-P. Lodemann, U. Schurath, J. Chem. Phys.
Ž .
Ž .
Ž .
systematic pattern, NF a -NCl a -NBr a , in-
Ž
.
87 1987 6266.
creasing as the halogen in the NX molecule increases
x
11 J. Habdas, S. Wategaonkar, D.W. Setser, J. Phys. Chem. 91
Ž .
in mass, with the exception of HI for which NCl a
Ž
.
1987 451.
Ž .
x
Ž
.
and NBr a have nearly equal rate constants.
12 K.B. Hewett, D.W. Setser, J. Phys. Chem. A 101 1997
9125.
The magnitudes of the rate constants with oxygen
x
13 X. Liu, M.A. MacDonald, R.D. Coombe, J. Phys. Chem. 96
Ž .
Ž .
Ž .
are ordered NF a <NBr a fNCl a . The excita-
Ž
.
1992 4907.
Ž .
tion–transfer reaction between NX a and O2 pro-
Ž .
duces excited oxygen molecules. The NX a
x
14 J. Combourieu, G. Le Bras, G. Poulet, J.L. Jourdain, Reac-
tion ClqN3Cl and reactivity of N3, NCl and NCl2 radicals
in relation to the explosive decomposition of N3Cl, 16th
molecules all have more energy than required to
Ž
.
Symp. Int. on Combustion, 1976, Mass. Inst. Technol.,
Cambridge, MA.
excite oxygen. The excitation–transfer reaction be-
tween NF a and O2 has 3553 cmy1 of excess
Ž .
w
w
x
15 J.L. Jourdain, G. Le Bras, G. Poulet, J. Combourieu, Com-
Ž .
Ž .
energy, while the reaction between NCl a or NBr a
and O2 only has 1398 cmy1 of excess energy. The
larger energy defect is the cause of the smaller rate
Ž
.
bust. Flame 34 1979 13.
x
16 T.L. Henshaw, S.D. Herrera, L.A.V. Schlie, J. Phys. Chem.
Ž
.
A 102 1998 6239.

(
)
342
K.B. Hewett, D.W. SetserrChemical Physics Letters 297 1998 335–342
w
x
Ž
.
w
x
Ž
.
17 K.B. Hewett, D.W. Setser, J. Phys. Chem. A 102 1998
21 G.C. Manke II, D.W. Setser, J. Phys. Chem. A 102 1998
6274.
153.
w
w
w
x
Ž
.
.
w
x Ž .
22 C.J. Hoffman, Inorg. Synth. 7 1963 180.
18 K.Y. Du, D.W. Setser, J. Phys. Chem. 94 1990 2425.
x
Ž
w
x
23 A.L. Kaledin, M.C. Heaven, K. Morokuma, Chem. Phys.
19 K.Y. Du, D.W. Setser, J. Phys. Chem. 96 1992 2553.
x
Ž .
Lett. 289 1998 110.
20 K.B. Hewett, G.C. Manke II, D.W. Setser, J. Phys. Chem. A
Ž
.
1998 , in preparation.