Temperature dependence of the Cl+HN3 reaction from 300 to 480 K

Article abstract of DOI:10.1016/S0009-2614(99)00726-5

The rate constant for Cl+HN₃ over the temperature range 300-480 K has been studied in a flow reactor. Based on the rate of loss of HN₃ and the rate of NCl(a¹Δ) generation, the temperature dependence of this reaction is des

Full text of DOI:10.1016/S0009-2614(99)00726-5

27 August 1999
Ž
.
Chemical Physics Letters 310 1999 111–120
www.elsevier.nlrlocatercplett
Temperature dependence of the ClqHN3 reaction from
300 to 480 K
)
Gerald C. Manke II , Thomas L. Henshaw, Timothy J. Madden, Gordon D. Hager
Air Force Research Laboratory, AFRLrDELC, 3550 Aberdeen AÕenue SE, Kirtland AFB, NM 87117-5776, USA
Received 10 May 1999; in final form 28 June 1999
Abstract
The rate constant for ClqHN3 over the temperature range 300–480 K has been studied in a flow reactor. Based on the
1
Ž
.
rate of loss of HN3 and the rate of NCl a D generation, the temperature dependence of this reaction is described by the
y
1
collision theory expression 1.2"0.3=10y11 T 0.5 exp y1514"93rT , with E0s3.0"0.2 kcal mol or an Arrhenius fit
Ž
.
y
10
y
Ž .
Ž
.
k T s2.0"1.0=10
exp y1452"150rT with Eas2.9"0.2 kcal mol 1. q 1999 Elsevier Science B.V. All rights
reserved.
1
Ž
.
1. Introduction
used for the generation of NCl a D is summarized
Ž .
Ž .
by reactions 1 and 2 :
The recent report of a direct measurement of gain
Ž2
P1r2 –2 P3r2 transition in the
.
I
ClqHN3 ™HClqN3
1
Ž .
w x
1
on the
1
Ž
.
NCl a D rI chemical system has increased interest
ClqN ™NCl a 1D, b 1 Sq, X 3 Sy qN
2
Ž .
y1
Ž
.
3
2
Ž
.
in the use of metastable nitrene halides NX as
The D H00 value for reaction 1 is y9.3 kcal mol
,
energy donors for iodine-based chemical lasers anal-
Ž .
Ž
.
ogous to the chemical oxygen iodine laser COIL
Ž .
and the enthalpy for reaction 2 is y39, y22, and
1
1
q
y65 kcal moly1 for generation of the a D , b S
,
w
x
2,3 . There are several important advantages and
disadvantages to the use of these molecules, which
Ž
. Ž
.
3
y
Ž
.
0Ž
.
and X S states, respectively, using D Hf NCl s
77.4 kcal moly1 14 . The rate determining step for
this process is reaction 1 , which has a room tem-
are isovalent with O2. The most obvious advantage
w x
1
Ž
.
is generation of the a D states in the gas phase
Ž .
y12
w
x
4–10 rather than the mixed gasraqueous chemistry
that is necessary for COIL operation. On the other
hand, the NX a D molecules are much more reac-
tive than O2 a D 11–13 , and the scaling issues for
w x
perature rate coefficient 15 of 1.1"0.3=10
Ž2 U Ž2
.
.
cm3 moleculey1 sy1. Gain on the I P3r2 –I P1r2
transition was achieved when a small flow of HI is
1
Ž
.
1
1
Ž
. w
x
Ž
.
added to a flow of NCl a D :
1
Ž
.
generating high concentrations of NCl a D have not
2
ClqHI™HClqI P
3
Ž .
yet been resolved. The ClrHN3 chemical system
ž /
3r2
NCl a 1D qI 2 P
™NCl X 3 Sy qIU 2 P
Ž
.
Ž
.
ž /
ž /
3r2
1r2
)
Corresponding author. Fax: q1-505-846-4807; e-mail:
mankeg@plk.af.mil
4
Ž .
0009-2614r99r$ - see front matter q 1999 Elsevier Science B.V. All rights reserved.
Ž
.
PII: S0009-2614 99 00726-5

(
)
112
G.C. Manke II et al.rChemical Physics Letters 310 1999 111–120
Ž .
To date, only reaction 4 has been characterized at a
pseudo-first-order conditions, the slopes of ln Inten-
3
w
x
Ž .
Ž
Ž
..
temperature other than 300 K 16 , and k4 T s1.1
sity NH A P versus Dt plots are equal to
=10y10 exp y519"143 KrT cm moleculey1 sy1
.
yk1 Cl . Under non-pseudo-first-order conditions or
in the presence of a modest F , the plots are ana-
3
Ž
.
w x
Ž . Ž .
w x
Due to the exothermicity of 1 – 3 and the mecha-
nism used to generate Cl, the temperature depen-
lyzed by a simulation process to get k1.
Ž .
dence of reaction
parameters in the development of the NCl a D r
1 is one of the most critical
1
Ž
.
Ž2
.
2. Experimental methods
I P3r2 chemical system as a chemical laser source.
Ž .
We have used two techniques to measure k T for
Ž
.
The high temperature flow reactor HTFR used
for this experiment has been described in detail
ClqHN3 in a flow reactor. The first simply moni-
1
Ž
.
tors the intensity of NCl a D at 1077 nm as a
w
x
previously 13 . A few minor modifications to the
HTFR have been made for this work. First, an
indicator inlet was installed 5 cm upstream of the
w
x
function of reaction time Dt for known Cl and
0
w
x0
HN3 and at a variety of temperatures. A simple
3
q
kinetic model and a least-squares fitting routine was
used to extract k1. Table summarizes the
Ž
.
observation zone to allow the addition of N2 A Su
1
or H2 S. Also, the quartz rods were removed and
glass windows were installed. Time resolution was
achieved by fixed point observation and addition of
reagents through one of two sliding injectors. For
this work, the inner injector was used to add HN3
FrClrHN3 kinetic model. The second method moni-
3
q
w
x
Ž
.
tored HN3 by adding N2 A Su and observing the
3
Ž
.
NH A P emission as a function of reaction time
3
q
Ž
.
and temperature. The reaction N2 A Su qHN3
N A 3 Sq qHN ™NH X 3 Sy q2 N X 1 Sq
Ž
.
and could extend from the observation zone Dts0
to 36 cm upstream. Chlorine atoms were generated
Ž . w
Ž
.
2 Ž
.
2 ž
/
u
3
g
5a
Ž
.
.
x
in the 150 cm long pre-reactor via reaction 6 20 .
FqHCl™HFqCl
In nearly all cases, excess HCl was added, and
Ž .
™NH A 3P q2 N X 1 Sq
6
Ž .
Ž
.
2 ž
/
g
5b
Ž
w
x
w x
reaction 6 goes to completion, giving Cl s F
0
0
has a rate constant 17,18 of 7.3"0.2=10y11
w
x
before HN3 was added. The initial F atom concentra-
cm3 moleculey1 sy1 at room temperature, and the
w x
tions, F , were determined by chemiluminescent
titration. A small flow of H2 S Matheson, CP grade
was added through the indicator inlet and the HF
DÕsy3 emission at 875 nm monitored as a
function of added Cl2 3% in He, Matheson or CF3I
0
3
Ž
x
.
w x
Ž
.
radiative lifetime of NH A P is 0.4 ms 19 . Even
though the branching 17 to NH A P is only 0.025
this combination permits
3
w
Ž
.
Ž .
5
Ž
.
to be a convenient
Ž
.
monitor of the relative concentration of HN3. Under
Table 1
Rate coefficients for FrClrHN3 reaction system
Reaction
k, model input
Ž
Ref.
3
cm moleculey1 sy1
.
w x
FqHN3 ™HFqN3 1
1.1"0.2=10y10
5"2=10y11
5
1
q
1
FqN3 ™NF a D, b S , X 3 Sy qN2 X Sg
4,5,10
Ž
.
.
Ž
.
w
x
1
1
1
1.5"0.6=10y11
f1.5"0.6=10y11
7.3"0.2=10y11
1.3=10y12
8,10,13
8,10,13
Ž
Ž
.
Ž
w
x
x
ClqN3 ™NCl a D qN2 X Sg
1
q
1
1
ClqN3 ™NCl b S , X 3 Sy qN2 X Sg
.
Ž
.
w
w
w
w
3
q
3
1
1
Ž
Ž
.
.
Ž
.
Ž
.
x
N2 A Su qHN3 ™NH A P q2N2 X Sg
17,18
20
13
3
q
1
1
Ž
.
x
N2 A Su qHCl™N2 X Sg qHqCl
7.2"0.9=10y12
1
1
Ž
Ž
Ž
Ž
.
.
.
.
Ž
.
x
NCl a D qNCl a D ™products
NCl a D qF™FqNCl X 3 Sy
2.2"0.7=10y11
1
Ž
.
w x
8
1
q1.0
NCl a D qCl™ClqNCl X 3 Sy
1.0
=10y12
Ž
.
w x
8
y0.5
1.5"0.4=10y14
1.2=10y11
1
w
12
x
x
NCl a D qHCl™products
w
18
FqHCl™HFqCl

(
)
G.C. Manke II et al.rChemical Physics Letters 310 1999 111–120
113
Ž
.
Ž
.
Lancaster, 97% . The smoothly extrapolated, zero-
lute capacitance manometers Baratron, model 622 ,
measured the reactor pressure, and all species except
HCl were metered with calibrated MKS mass flow
Ž
.
w
x
intensity intercept from plots of I HF versus titrant
w x
gives F .
0
Ž
.
The mechanical pumprblower combination pro-
duced linear flow velocities of 3400–1150 m sy1 at
0.5–1.5 torr when throttled with a gate valve. Abso-
meters Model 1179A . Several freeze pump–thaw–
cycles were performed while preparing 20–30%
mixtures of HCl Scott Specialty Gas, 99.999% , and
Ž
.
Ž
.
Fig. 1. Upper panel: A representative spectrum from the FrClrHN3 reaction system after a reaction time of 0.013 s. The HF DÕsy3
12
emission was insignificant 0.025 s downstream of the HN3 inlet. The experimental conditions were F s4.4=10 , Cl s3.8=1012 , and
w x
w x
1
1
HN3 s1.0=10 cmy3. The intense NF b S and NCl b S emissions, not shown, were observed at 530 and 665 nm, respectively.
Lower panel: The emission intensity of NH A P is plotted as a function of HN3 . The intensity at 335 nm see inset spectrum is linear
13
q
q
w
x
Ž
.
.
Ž
.
3
Ž
w
x
Ž
.
over a wide range of initial HN3 and is easily observable for HN3 s1.0=10 moleculecmy3
.
11
w
x
w
x

(
)
114
G.C. Manke II et al.rChemical Physics Letters 310 1999 111–120
3
Ž
Ž
..
Ž
.
Fig. 2. Pseudo-first-order plots of ln IrI0 NH A P versus reaction time. The experimental conditions and results are as follows: 300 K
Cl s1.2=10 , HN3 s2.0=10 cmy3, ks1.4"0.1=10y12 cm3 moleculey1 sy1
;
373 K Cl s8.0=10 , HN3 s3.0=10
13
12
12
12
w
x
w
x
Ž
. w
x
w
x
12
12
cmy3
,
ks4.5"0.3=10y12 cm3 moleculey1 sy1
;
,
HN3 s2.8=10
cmy3 ks6.5"0.4=10y12
,
Ž
413
.
K
w x
Cl s6.0=10
12
w
x
cm3 moleculey1 sy1; 458 K Cl s6.0=10 , HN3 s1.7=10 cmy3, ks8.5"0.3=10y12 cm3 moleculey1 sy1
.
12
Ž
. w
x
w
x
CF3I in 12 L bulbs. The flow rate of HCl was
determined by diverting the stream to a vessel of
known volume and measuring the rate of pressure
stored in a stainless steel vessel. HN3 was added to
the reactor via one of two sliding Pyrex injectors.
The entire reactor was encased by resistive heat-
ing units, which consisted of Nichrome wire heli-
cally wound inside a ceramic jacket. The temperature
was measured by inserting several type K thermo-
couples into the gas stream at various points along
the reactor. A flexible thermocouple was inserted
Ž
rise. The bulk of the flow typically 2.5 SLPM, 1850
mm sy1 consisted of Ar Airgas, UHP grade . Two
.
Ž
.
Ž
.
microwave discharges on a CF4 Airgas, 99.5% rAr
mixture produced up to 1.5=1013 cmy3 F atoms at
1.5 torr. Pre-prepared mixtures of HN3 and He were
Table 2
Ž
.
Summary of the k T measurements via loss of HN3
Ž
.
Ž
k T
.
T
k T
Ž
T
y12
y12
=10
cm3 moleculey1 sy1
=10
cm3 moleculey1 sy1
.
Ž
K
.
.
Ž
K
.
Ž
300
333
348
368
373
378
1.3"0.4
2.0"0.2
2.2"0.5
3.4"0.9
4.4"0.5
4.5"1.5
388
413
438
458
463
468
5.3"1.5
6.5"1.5
7.2"1.9
7.5"0.7
11"1
10"4

(
)
G.C. Manke II et al.rChemical Physics Letters 310 1999 111–120
115
Ž
.
into the one of the movable injectors to provide a
temperature measurement at the center of the tube.
The heaters were regulated with Omega controllers
A 0.3 m monochromator Instruments S.A. dis-
persed the chemiluminescence collected by a short
Ž
.
focal length lens. Two different gratings 500 or 900
Ž
.
Model CN76000 with an accuracy of "18 at room
nm blaze, both with 1200 grooves mmy1 were used,
temperature and "78 at 480 K. All inner surfaces
were coated with PTFE, which limited the tempera-
ture to 500 K.
depending on the emission of interest. The emissions
1
1
1
q
Ž
.
Ž
.
Ž
.
Ž
.
of HF DÕsy3 , NF a D , NCl a D , NF b S
,
1
q
Ž
.
were monitored with a cooled
and NCl b S
3
Ž
Ž
.
Fig. 3. Non-pseudo-first-order plots of I NH A P versus reaction time. A least-squares fitting routine was used to extract k1 from the
data. The experimental conditions are as follows. Upper panel: Ts300 K, Cl s1.1=10 , F s2.0=10 , HN3 s2.0=10 cmy3
.
13
12
12
w
x
w x
w
x
12
12
12
HN3 s1.5=10 cmy3. Note that the data can be satisfactorily fit by
w x w x
F s1.5=10 ;
w
x
Lower panel: Ts388 K, Cl s6.5=10
;
several values for k1, and the relative error for this method is large.

(
)
116
G.C. Manke II et al.rChemical Physics Letters 310 1999 111–120
Ž
.
Ž
.
.
y808C R1767 PMT Hamamatsu while the UV
PMT from the reaction of F and Cl atoms with HN3
3
Ž
.
emission from NH A P was detected with a R374
is shown in the upper panel of Fig. 1.
The relative concentration of HN3 was monitored
by adding N2 A Su via the indicator inlet and
monitoring the resultant NH A P signal at 335 nm.
The N2 A Su molecules were generated by energy
transfer from metastable Ar P2 , P0 atoms using a
Ž
.
PMT Hamamatsu cooled to y508C. When neces-
sary, band pass or long pass filters were used to
isolate signals of interest from unwanted background
or second-order emissions. A representative spec-
3
q
Ž
.
3
Ž
.
3
q
Ž
.
3
Ž
Ž3
.
trum uncorrected for the relative response of the S-1
1
1
Ž
.
w
Ž
.x
13
Fig. 4. Measurement of k1 via generation of NCl a D . A least-squares model fit to NCl a D versus Dt was used to determine
k ClqHN3 . The experimental conditions are as follows. Upper panel: Ts300 K, Cl s1.8=10 , HN3 s2.0=10 cmy3. Lower
13
Ž
.
w
x
w
x
12
13
panel: Ts433 K, Cl s8.0=10 , HN3 s1.4=10 cmy3. Note that a wide range of values for k1 can satisfactorily fit the data, and
w
x
w
x
Ž
.
the relative error for this method is large "30% .

(
)
G.C. Manke II et al.rChemical Physics Letters 310 1999 111–120
117
w
x
rolled tantalum foil discharge design 21 . The
described above and the error bars for k1 are signifi-
3
Ž
.
Ž .
NH A P emission intensity increased linearly with
cantly larger. Table 2 summarizes the k T data for
Ž
.
Ž .
w
x
the HN3 flow rate see the lower panel of Fig. 1 .
reaction 1 based on the loss of HN3 .
With optimization of the light collection and
3
q
N2 A Su generator, HN3 as low as 1=1011
1
Ž
.
w
x
(
)
3.2. Generation of NCl a D
moleculecmy3 could be routinely detected even
Ž
.
though the branching ratio to 5b is low.
The rate determining step for the generation of
1
Ž
.
Ž .
F
NCl a D is reaction 1 . For nearly any set of
w x
conditions where no
is present, the rate of
3. Results and discussion
1
Ž
.
NCl a D growth is dependent upon k1. Fig. 4 shows
1
Ž
.
several examples of the growth of NCl a D versus
Dt at a variety of temperatures. In each case the
linear least-squares fit to the data is indicated by the
solid line, while alternate values for k1 are shown by
broken lines. As expected, the established value for
k1 is reproduced at room temperature. However, the
precision of this method is quite poor. As the plot
shows, several values, up to "30%, also give rea-
sonable fits to the data. This method is also suscepti-
ble to error in other ways. For example, as k1
[
]
3.1. Monitoring the loss of HN3
Žw
..
x
Under pseudo-first-order conditions
HN3 the slope of a given ln I NH A P versus
Dt plot is equal to k1 Cl , and the rate constant is
given by simply dividing by the known Cl . Measur-
able decomposition of HN3 does not occur below
560 K and reactive loss should be the only removal
process 22 . Fig. 2 shows several such measure-
ments taken at a variety of temperatures. In each
Cl 4
3
w
x.
Ž Ž
Ž
w
x
w
x
w
x
Ž . Ž
increases the value of k2 T which was fixed to the
w
x
case, the initial Cl was determined by first measur-
ing F and then adding a slight excess of HCl. The
intensity decreased with time, and a regression fit to
0
.
room temperature value becomes more important
w x
0
w
x
because N3 and HN3 compete for reaction with Cl .
Most importantly, it is difficult to generate observ-
able I NCl a D without high Cl and HN3
w
x
the data gives the product k1 Cl . At room tempera-
ture, the data is within the combined error bars of the
established value for k1, 1.1 " 0.3 = 10y12
cm3 moleculey1 sy1. Several measurements were
performed at each temperature and Table 2 lists the
average and 2 standard deviations at each tempera-
ture. The value of k1 increases up to 1.1"0.1=
1
Ž
Ž
..
w
x
w
x0
,
0
which may cause additional problems with quench-
ing. Although this method is less satisfactory for the
determination of k1, the results agree quite well with
3
Ž
Ž
.. Ž
those obtained by monitoring I NH A P
see
.
Table 3 .
Ž .
Ž .
All of the k T data for reaction 1 are summa-
10y11 cm3 moleculey1 sy1 for Ts460 K.
3
rized in Tables 2 and 3 and in Fig. 5. A least-squares
Ž
q. Ž
kQ s1.3=
Because HCl quenches N2 A Su
10y12 cm3 moleculey1 sy1 23 , it was not always
possible to add excess HCl and a few measurements
of k1 were performed with a mix of F and Cl atoms
present with HN3. A known flow of HCl is added to
. w
x
Table 3
1
Ž
.
Ž
.
Summary of the k T measurements via generation of NCl a D
Y
X
w x
w x
w x
a measured F and F and Cl calculated from
the bimolecular rate law. The temporal dependence
of HN3 , as monitored by I NH A P , is fit by the
kinetic model listed in Table 1. In cases such as
these, accurate measurement of k1 requires Cl 4
F , and F ( HN3 0. Otherwise, F removes nearly
all of the HN3 and it becomes impossible to accu-
rately extract k1 5 . Several examples are shown in
0
Ž
k T
.
T
y12
cm3 moleculey1 sy1
.
Ž
K
.
Ž
=10
3
w
x
Ž
Ž
..
300
338
368
383
388
403
433
448
468
0.89"0.27
2.6"0.8
3.4"1.0
4.8"1.4
5.8"1.7
7.7"2.3
8.4"2.5
9.3"2.8
11"3
w
x
w x
w x
w
x
w x
Fig. 3. In each case, the least-squares analysis pro-
duces a satisfactory fit to the data. This method,
however, is inherently less sensitive than the method

(
)
118
G.C. Manke II et al.rChemical Physics Letters 310 1999 111–120
Ž
.
Ž .
data. Upper panel: The k1 T data is fit by the Arrhenius expression. The best fit is
Fig. 5. Arrhenius and collision theory fits to the k1
T
achieved for As2.0=10y10 and Ea s2.9 kcal moly1. Lower panel: A collision theory expression is used and As1.2=10y11 and
1
E0 s3.0 kcal moly1. In both panels data from both the loss of HN3 and generation of NCl a D techniques are shown.
Ž
.
w x
analysis of the data using the collision theory expres-
from the data reported in Ref. 1 , and excellent
agreement is achieved.
sion, AT 0.5 exp yE0rRT , results in a satisfactory
fit with As1.2"0.5=10y11 and E0 s3.0"0.3
Ž
.
kcal moly1. The Arrhenius expression was also used
4. Discussion
y10
Ž .
Ž
to fit the data: k T s2.0"1.0=10
exp y1452
"150rT with Ea s2.9"0.3 kcal moly1. The data
in Fig. 5 also includes an estimate for k1 calculated
While several azideratom F, Cl, and H reac-
tions have been characterized at room temperature,
.
Ž
.

(
)
G.C. Manke II et al.rChemical Physics Letters 310 1999 111–120
119
there have been few measurements for T)300 K.
and requires some comment. Hydrogen atom abstrac-
Ž .
Ž
For example, reaction 7
tion reactions such as FqHCl ks4.4"1.5=
3
y1
0y11 exp y 400 " 100rT cm molecule
s y1
.
Ž
.
2 FqHN ™HFqN ™HFqNF a 1D qN
Ž
.
y11
3
3
2
x
Ž
Ž
.
28 and ClqHBr ks4.8=10
exp y454rT
cm3 moleculey1 sy1 29 have A factors which are
;2–3 smaller than the present result. Halogen atom
addition reactions, on the other hand, typically have
much larger pre-exponential factors. For example, a
relevant comparison may be made with the fast
reaction F q HN3, A 0 k300 s 1.1 " 0.2 = 10y10
cm3 moleculey1 sy1. Chlorine atoms are known to
7a
7b
Ž
.
.
. w
x
™HNFqN ™HFqNF a 1D qN
Ž
.
2
2
Ž
is known to be an efficient source of the N3 and
NF a D radicals. The room temperature rate con-
1
Ž
.
y10
w x
Ž .
stant 5 for 7a is established as 1.1"0.2=10
Ž
w
x
w x
add to olefins e.g., C2 H4 30 , C2 H3Br 31 , and
C2 H3Cl 32 and these reactions are very fast at
cm3 moleculey1 sy1 and the nascent HF vibrational
w
x.
Ž
.
distribution from reaction 7a is 0.36:0.36:0.22:0.06
for Õs1–4. Microscopic branching between direct
abstraction and addition–elimination channels may
account for the rather flat vibrational distribution.
y10
Ž
.
room temperature, k 300 s 1.5 " 0.4 = 10
cm3 moleculey1 sy1. The Clqolefin reactions also
have very small activation energies, probably on the
order of a few hundred cal moly1. For example,
Ž
.
w x
The branching to HNF 7b has been established 24
assuming Ea s0.3 kcal moly1, and using k298 Clq
Ž
as 0.03– 0.001
q0.02 . The rate coefficient for the reaction of
y10
.
C2 H3Br s 1.43 " 0.29 = 10
cm3 moleculey1
w
x
H atoms with HN3 has been measured 25 from 300
sy1, the A factor is 2.4=10y10 cm3 moleculey1 sy1
.
y11
Ž .
Ž
.
to 460 K and k T s2.5=10
exp y2315rT
Clearly, the large pre-exponential factor for ClqHN3
is consistent i.e., within a factor of 2–3 with other
Cl atom reactions and is suggestive of a Cl atom
addition reaction rather than hydrogen abstraction.
cm3 moleculey 1 sy 1
Ž2
. w
.
The reaction produces
Ž
.
x
NH2 A1, n2 (20 26 and minor amounts of
NH A P and NH b S .
3
1
Ž
.
Ž
.
Ž .
The main products for reaction 1 are assumed to
HqHN3 ™NH2 qN2
8a
Ž
.
.
be HCl and N3 even though vibrationally excited
HCl has not been observed only HCl Õs1 is
energetically possible and the generation of HNCl
cannot be ruled out. Independent confirmation of this
could be accomplished in one of several ways:
laser-induced fluorescence detection of N3 or in-
frared absorption of HCl Õs0, 1 generated by reac-
tion 1 are the most easily implemented. Virtually
3
1
Ž
.
Ž
The mechanisms for NH A P and NH b S
production are not well established, but they are
presumed to be generated by secondary reactions
Ž
Ž
.
.
w
x
such as 26 :
HqHN3 ™N3 qH2
8b
9
Ž
Ž
.
Ž
.
HqN ™NH b 1 S qN
Ž
.
Ž .
Ž .
3
2
nothing is known about the species HNCl. If one
2 NHU b 1 S k a 1D ™NH A 3P
10
Ž
.
Ž
.
.
assumes that ClqHNCl is sufficiently exothermic to
1
Ž
.
Prior to this report, the only halogenrazide reaction
that had been characterized T)300 K was Clq
ClN3:
give NCl a D , by analogy to the FqHN3 system,
both N3 and HNCl should react rapidly with Cl
1
Ž
.
atoms to give NCl a D .
ClqClN3 ™N3 qCl2
11
Ž
.
Combourieu et al. used mass spectrometric detection
to measure the temperature-dependent rate constant
5. Conclusions
w
x
between 300 and 657 K 27 . They reported a moder-
Ž .
ate temperature dependence of k T s 2.3 =
The temperature dependence of the reaction Clq
3
10y11 exp y554rT cm moleculey1 sy1
.
HN3 has been measured in a heated flow reactor.
Ž
.
The present results indicate a moderate tempera-
ture dependence for the reaction of ClqHN3, with
Ea f3 kcal moly1. The pre-exponential factor is large
The rate coefficient was measured by monitoring the
1
Ž
.
loss of HN3 and the rate of generation of NCl a D
versus time for known starting concentrations. The

(
)
120
G.C. Manke II et al.rChemical Physics Letters 310 1999 111–120
w
x
Ž
.
10 K.B. Hewett, D.W. Setser, J. Phys. Chem. A 102 1998
rate constant data was fit by an Arrhenius expression
y10
6274.
Ž .
Ž
.
of k T s2.0"1.0=10
exp y1452"150rT
w
w
x
x
Ž
.
11 K.Y. Du, D.W. Setser, J. Phys. Chem. 94 1990 2425.
12 K.B. Hewett, G.C. Manke II, D.W. Setser, G. Brewood, J.
with Ea s2.9"0.3 kcal moly1. A collision theory
fit to the data gave the expression 1.2"0.5=
Ž
.
Phys. Chem. A 1999 , submitted.
10y11 T 0.5 exp y1514"150rT , with E0 s3.0"
w
w
w
w
x
Ž
.
.
13 T.L. Henshaw, S.D. Herrerra, L.A. Schlie, J. Phys. Chem. A
0.3 kcal moly1
Ž
.
101 1997 4048.
x
14 S.S. Xantheas, T.H. Dunning Jr., A. Mavridis, J. Chem.
Ž
.
Phys. 106 1997 3280.
x
Ž
.
15 G.C. Manke II, D.W. Setser, J. Phys. Chem. A 102 1998
Acknowledgements
153.
x
16 T.L. Henshaw, S.D. Herrera, L.A. Schlie, J. Phys. Chem. A
Ž
.
102 1998 6239.
The US Air Force Office of Scientific Research
supported this work. G.C.M. also wishes to acknowl-
edge the National Research Council for support. The
authors are grateful for helpful discussions with Prof.
R.D. Coombe, Prof. M.C. Heaven, and Prof. D.W.
Setser.
w
w
w
x
Ž
.
17 D.Z. Cao, D.W. Setser, J. Phys. Chem. 92 1988 1169.
18 T.L. Henshaw, Ph.D. Thesis, Denver University, 1987.
19 K.P. Huber, G. Herzberg, Molecular Spectra and Molecular
x
x
Structure IV, Constants of Diatomic Molecules, Van Nos-
trand Reinhold, New York, 1979.
w
w
x
x
Ž
.
.
20 M.A.A. Clyne, W.S. Nip, Int. J. Chem. Kinet. 10 1978 367.
Ž
21 J.H. Kolts, D.W. Setser, in: D.W. Setser Ed. , Reactive
Intermediates in the Gas Phase, Academic Press, New York,
1979.
References
w
x
22 M.C. Sneed, R.C. Brasted, in: Comprehensive Inorganic
Chemistry, vol. 5, Van Nostrad, Princeton, NJ, 1956, p. 42.
w x
1
J.M. Herbelin, T.L. Henshaw, B.D. Rafferty, B.T. Anderson,
R.F. Tate, T.J. Madden, G.C. Manke II, G.D. Hager, Chem.
w
w
x
Ž
.
23 M.F. Golde, Int. J. Chem. Kinet. 20 1988 75.
x
Ž
.
24 K.B. Hewett, D.W. Setser, J. Phys. Chem. A 101 1997
Ž
.
Phys. Lett. 299 1999 583.
9125.
w x
2
W.E. McDermott, N.R. Pchelkin, D.J. Benard, R.R. Bousek,
w
w
x
Ž
.
25 G. LeBras, J. Combourieu, Int. J. Chem. Kinet. 5 1973 559.
Ž
.
Appl. Phys. Lett. 32 1978 469.
x
26 O. Kajimoto, T. Kawajiri, T. Fueno, Chem. Phys. Lett. 76
w x
3
D.J. Benard, W.E. McDermott, N.R. Pchelkin, R.R. Bousek,
Ž
.
1980 315.
Ž
.
Appl. Phys. Lett. 34 1979 40.
w
x
27 J. Combourieu, G. Le Bras, G. Poulet, J.L. Jourdain, Proc.
16th Int. Combustion Symposium, The Combustion Institute,
Pittsburgh, PA, 1976.
w x
4
X. Liu, M.A. MacDonald, R.D. Coombe, J. Phys. Chem. 96
Ž
.
1992 4907.
w x
5
J. Habdas, S. Wategaonkar, D.W. Setser, J. Phys. Chem. 91
w
x
28 D.L. Baulch, J. Duxbury, S.J. Grant, D.C. Montague, J.
Ž . Ž
Ž
.
1987 451.
.
Phys. Chem. Ref. Data 10 1 1981 548.
w x
Ž
.
6
D.J. Benard, J. Phys. Chem. 100 1996 8316.
w
x
x
Ž
.
29 R. Rubin, A. Persky, J. Chem. Phys. 79 1983 4310.
w x
Ž
.
Ž
7
J.M. Herbelin, N. Cohen, Chem. Phys. Lett. 20 1973 605.
w
30 R.S. Iyer, P.J. Rogers, F.S. Rowland, J. Phys. Chem. 87
w x
.
8
G.C. Manke II, D.W. Setser, J. Phys. Chem. A 102 1998
Ž
.
1983 3799.
7257.
w
w
x
Ž
.
31 J.Y. Park, I.R. Slagle, D. Gutman, J. Phys. Chem. 87 1983
w x
9
D.J. Benard, M.A. Chowdhury, B.K. Winker, T.A. Seder,
1812.
Ž
.
H.H. Michels, J. Phys. Chem. 94 1990 7507.
x
Ž
.
32 R.S. Iyer, F.S. Rowland, J. Phys. Chem. 89 1985 3730.