Direct measurements of the rate constants of the reactions NCN + NO and NCN + NO2 behind shock waves

Article abstract of DOI:10.1021/jp208715c

The high-temperature rate constants of the reactions NCN + NO and NCN + NO2 have been directly measured behind shock waves under pseudo-first-order conditions. NCN has been generated by the pyrolysis of cyanogen azide (NCN ₃) and quantitatively detected by sensitive difference amplification laser absorption spectroscopy at a wavelength of 329.1302 nm. The NCN ₃ decomposition initially yields electronically excited ¹NCN radicals, which are subsequently transformed to the triplet ground state by collision-induced intersystem crossing (CIISC). CIISC efficiencies were found to increase in the order of Ar < NO₂ < NO as the collision gases. The rate constants of the NCN + NO/NO₂ reactions can be expressed as k_NCN+NO/(cm³ mol ^-1s^-1) = 1.9 × 10¹² exp[- 26.3 (kJ/mol)/RT] (±7%, ΔE_a = ± 1.6 kJ/mol, 764 K < T < 1944 K) and k_NCN+NO2/(cm³ mol^-1s ^-1) = 4.7 × 10¹² exp[- 38.0(kJ/mol)/RT] (±19%,ΔE_a = ± 3.8 kJ/mol, 704 K < T < 1659 K). In striking contrast to reported low-temperature measurements, which are dominated by recombination processes, both reaction rates show a positive temperature dependence and are independent of the total density (1.7 × 10^-6 mol/cm < ρ < 7.6 × 10 mol/cm³). For both reactions, the minima of the total rate constants occur at temperatures below 700 K, showing that, at combustion-relevant temperatures, the overall reactions are dominated by direct or indirect abstraction pathways according to NCN + NO f CN + N₂O and NCN + NO₂ → NCNO + NO. (Figure presented)

Full text of DOI:10.1021/jp208715c

ARTICLE
pubs.acs.org/JPCA
Direct Measurements of the Rate Constants of the Reactions NCN + NO
and NCN + NO Behind Shock Waves
2
J. Dammeier and G. Friedrichs*
Institut f €u r Physikalische Chemie, Christian-Albrechts-Universit €a t zu Kiel, Olshausenstrasse 40, D-24098 Kiel, Germany
S Supporting Information
b
ABSTRACT: The high-temperature rate constants of the reactions NCN + NO and
NCN + NO have been directly measured behind shock waves under pseudo-first-order
2
conditions. NCN has been generated by the pyrolysis of cyanogen azide (NCN ) and
3
quantitatively detected by sensitive difference amplification laser absorption spectroscopy
at a wavelength of 329.1302 nm. The NCN decomposition initially yields electronically
3
1
excited NCN radicals, which are subsequently transformed to the triplet ground state by
collision-induced intersystem crossing (CIISC). CIISC efficiencies were found to increase
in the order of Ar < NO < NO as the collision gases. The rate constants of the NCN + NO/
2
3
ꢀ1 ꢀ1
12
NO reactions can be expressed as k
/(cm mol
s
) = 1.9 ꢁ 10 exp[ꢀ26.3 (kJ/
2
NCN+NO
3
mol)/RT] ((7%,ΔE = ( 1.6 kJ/mol, 764 K < T < 1944 K) and k
/(cm
a
NCN+NO2
ꢀ
1 ꢀ1
12
mol
s
) = 4.7 ꢁ 10 exp[ꢀ38.0(kJ/mol)/RT] ((19%,ΔE = ( 3.8 kJ/mol, 704 K < T < 1659 K). In striking contrast to
a
reported low-temperature measurements, which are dominated by recombination processes, both reaction rates show a positive
ꢀ6
3
ꢀ6
3
temperature dependence and are independent of the total density (1.7 ꢁ 10 mol/cm < F < 7.6 ꢁ 10 mol/cm ). For both
reactions, the minima of the total rate constants occur at temperatures below 700 K, showing that, at combustion-relevant tem-
peratures, the overall reactions are dominated by direct or indirect abstraction pathways according to NCN + NO f CN + N O and
2
NCN + NO f NCNO + NO.
2
I. INTRODUCTION
N2O þ O f NO þ NO
ð5Þ
Encouraged by the demand for clean energy sources, the efforts
to reducethe emissionsof hazardous substancesfrom combustion
processes continue. Nitrogen oxides NO (NO and NO ) are
among the main pollutants, which are generated in the combus-
tion of fossil fuels as well as fuels made from regrowing resources.
Finally, especially under fuel-rich combustion conditions,
prompt-NO is generated by the reaction of small hydrocarbon
9
radicals with N
. Fenimore suggested that a key initiating step of
x
2
2
the reaction sequence is the spin-forbidden reaction
1
2
1
þ
4
1
þ
CHð ΠÞ þ N ð Σ Þ f Nð SÞ þ HCNð Σ Þ
ð6aÞ
2
The hazardous potential of NO lies in its contribution to photo-
x
chemical smog in the troposphere and to ozone depletion in the
In fact, it has long been recognized that the intersystem crossing
probability from the doublet to the quartet potential energy
surface necessary for reaction 6a is low, and thus theoretically
predicted rate constants were lower than experimentally mea-
2
stratosphere. Furthermore, nitrous oxide (N O), a secondary
2
product of NO reactions, is an active atmospheric trace gas with a
x
3
high global warming potential.
There are several formation mechanisms and two principal
10
sured. In the year 2000, the products of this textbook reaction
1
11,12
sources of NO in combustion processes. First, NO stems from
have been revised by Moskaleva et al.
according to
x
x
nitrogen impurities in the fuel, which are oxidized. This mecha-
nism is referred to as fuel-N-conversion. Second, NO can be
generated from atmospheric molecular nitrogen. At high tem-
peratures, the primary formation pathway takes place according
to the Zeldovich mechanism:
2
1
þ
2
3
ꢀ
4
CHð ΠÞ þ N2ð Σ Þ f Hð SÞ þ NCNð Σ Þ
ð6bÞ
x
g
By quantum chemical and statistical rate calculations they showed
that this spin-allowed reaction constitutes a relevant pathway and
yields rate constants that are consistent with the experimental
data. The pathway generates NCN radicals, a species which had
not been discussed in the context of combustion research before.
Later, it was experimentally proven by flame and shock tube mea-
5
,6
O þ N f NO þ N
ð1Þ
ð2Þ
ð3Þ
2
N þ O2 f NO þ O
N þ OH f NO þ H
surements that the NCN channel (reaction 6b) is in fact the
dominatingreaction channel ofreaction6.
13ꢀ16
Sincethen, several
Third, in low temperature combustion at high pressures, NO is
yielded via the reaction sequence
x
7
,8
Received: September 9, 2011
Revised:
November 5, 2011
N2 þ O þ M f N2O þ M
ð4Þ
Published: November 08, 2011
r 2011 American Chemical Society
14382
dx.doi.org/10.1021/jp208715c
|J. Phys. Chem. A 2011, 115, 14382–14390

The Journal of Physical Chemistry A
ARTICLE
1
1,17ꢀ19,21ꢀ23
16ꢀ18,24ꢀ27
theoretical
and experimental
studies have
measured rate constants in the temperature range of 254 K <
addressed NCN reactions and available rate constants have been
T < 353 K following the 193 nm photolysis of cyanogen azide
14,27ꢀ31
introduced into existing combustion models.
Objects of previous theoretical work were the reactions NCN +
(NCN ) as a source of NCN. In the examined temperature
3
range, both studies support the expected negative temperature
2
2
21
23
25
O (7), NCN + O (8), NCN + OH (9), NCN +
dependence. Welz extended the temperature range to 254 K <
2
1
7,19,20
18
11
NO (10),
NCN + NO (11), and NCN + H (12).
T < 485 K and characterized the pressure dependence at press-
2
Rate constants and product branching ratios have been theore-
tically predicted over a wide range of temperatures and pressures.
In contrast, available experimental results are limited to a few
reactions and mainly to temperatures below T < 600 K. An
exception is the reaction NCN + H, which has been investigated
behind shock waves in the temperature range 2378 K < T < 2492
ures ranging from p = 30 mbar to p = 50 bar. Again, NCN was
detected by LIF following the photodissociation of NCN at λ =
3
193 nm and λ = 248 nm.
Reaction 11, NCN + NO , has been the subject of two com-
2
1
8
bined experimental and theoretical studies. Yang et al. mea-
sured rate constants in the temperature range 260 K < T < 296 K
at pressures between p = 133 mbar and p = 666 mbar. NCN was
16
K by Vasudevan et al. NCN was obtained by the pyrolysis of
ethane (C H ), which yields CH radicals that subsequently
generated by 193 nm photolysis of NCN and detected by LIF.
2
6
3
reacted with the bath gas N to form NCN (reaction 6b).
Additionally, rate constants have been calculated at temperatures
of 200 K < T < 2000 K using a combination of quantum mechan-
ical (G2M//B3LYP/6-311+G(d)) and statistical (RRKM/ME)
methods. On the basis of these calculations, an association re-
action is expected for temperatures T < 700 K to take place
exhibiting a negative temperature dependence, whereas at higher
temperatures a positively temperature-dependent abstraction
channel is predicted to yield NCNO + NO. In similar experi-
2
Observed NCN decays were attributed to the reaction NCN +
H (12). An indirect determination of the rate constant of
2
7
reaction 12 has also been reported by Lamoureux et al. In a
combined modeling and flame diagnostics study, they attrib-
uted deviations of the simulated from the measured CH, NCN
and NO concentrations to reaction 12. Furthermore, high-
temperature reaction rates of the unimolecular NCN decom-
26
position NCN + M f C + N + M (13) have been directly
ments, Kappler extended the temperature range to 255 K < T <
349 K and measured falloff curves in the pressure range 155 mbar
< p < 38 bar. In agreement with Yang et al., a negative temperature
and a positive pressure dependence has been noted, which was ex-
plained by an associationꢀisomerization mechanism finally yield-
2
measured (C-ARAS) behind shock waves at temperatures of
3
2,33
1
800ꢀ2950 K.
Recently, in order to provide much needed high temperature rate
constant data for bimolecular NCN reactions, we have thoroughly
investigated the thermal decomposition of NCN as a quantitative
ing NCO and N O as the products.
3
2
34
high-temperature NCN radical source. NCN formation rates and
high temperature absorption cross section have been directly mea-
sured showing that NCN can be sensitively detected in the parts per
million (ppm) range behind shock waves. At such low concentra-
tions, subsequent NCN secondary chemistry is efficiently suppres-
So far, only the low-temperature recombination processes of
reactions 10 and 11 could be experimentally confirmed. High-
temperature experiments suitable to assess the dominating N O +
2
CN channel for reaction 10 and NCNO + NO for reaction 11 are
not available. In this spirit, the high-temperature measurements
presented in this paper provide an opportunity to bring together
theory and experiment. Assessing and characterizing the tem-
perature and pressure dependencies of bimolecular reactions of
small radicals, i.e., the preference of an association complex at low
temperatures and the opening of an activation energy-controlled
direct or indirect abstraction reaction channel at high tempera-
tures, remains challenging. Recent examples from our shock tube
lab addressing similar problems are the measurements of the rate
sed turning the thermal NCN decomposition into an ideal NCN
3
radical source. Nevertheless, as a requirement for accurate rate con-
stant measurements, a consistent NCN background mechanism
has been derived from pyrolysis experiments reporting rate
constants for the reactions NCN + M (13) at 2010 K < T <
35
3
250 K and NCN + NCN (14) at 965 K < T < 1900 K.
Furthermore, by adding O atoms generated from the N O de-
2
compositionto the reaction mixtures, we wereable to measure the
rate constant of the reaction NCN + O (7) for the first time (1825
K < T < 2785 K).
36
37
constants of the reactions HCO + NO/NO2 and HCO + O2.
In this paper, first high-temperature measurements on the re-
actions
Regarding reaction 10, NCN + NO, the theoretical investiga-
tions agree that the reaction proceeds via an NCNNO inter-
1
7,19,20,25
17
NCN þ NO f N2O þ CN
and
NCN þ NO2 f NCNO þ NO
ð10Þ
mediate.
Huang et al. performed quantum chemical
(
G2M/CC5) and statistical rate (VTST/RRKM/ME) calcula-
tions. It was found that the overall reaction exhibits negative tem-
perature dependence at T < 500 K and a pronounced positive
temperature dependence toward higher temperatures. At low
temperatures, a recombination reaction according to NCN +
NO + M h NCNNO + M takes place, whereas at high tem-
ð11Þ
will be presented to provide rate constant data at combustion
relevant temperatures and to verify the theoretically predicted
dominance of the direct or indirect abstraction reaction
channels.
peratures an indirect abstraction channel yields N O + CN. One
2
of the first experimental works on elementary NCN gas phase
reactions was concerned with reaction 10 and was performed by
24
II. EXPERIMENTAL SECTION
Baren and Hershberger in 2002. NCN was generated by 193 nm
photolysis ofdiazomethane(CH N ) in the presence of cyanogen
All experiments have been performed using a shock tube
2
2
3
7,38
(
C N ). Rate constants could be determined from NCN con-
setup
equipped with a narrow bandwidth laser absorption
2
2
34
centrationꢀtime profiles measured by means of laser-induced
spectrometer to measure NCN concentrationꢀtime profiles.
fluorescence (LIF) in the temperature range 298 K < T < 573 K.
Briefly, the 4.4 m long test section of the shock tube with an inner
ꢀ
7
Both NCN (by LIF) and the possible products N O and
diameter of 81 mm could be evacuated down to pressures of 10
mbar by a combination of oil-free diaphragm and turbomolecular
2
1
7
CO (by IR absorption) were detected. Similarly, Huang et al.
2
1
4383
dx.doi.org/10.1021/jp208715c |J. Phys. Chem. A 2011, 115, 14382–14390

The Journal of Physical Chemistry A
ARTICLE
Table 1. Selected Reactions of the Mechanism Used in the Numerical Simulations to Extract the Rate Constants of the Reaction
a
NCN + NO
no.
reaction
A
E
a
ref.
this work
44
3
12
9
1
(10)
NCN þ NO f CN þ N2O
1.9 ꢁ 10
3.3 ꢁ 10
26.3
71.2
71.2
set equal for NCN, see text
1
F = 1.8 ꢁ 10ꢀ6 mol/cm3
F = 3.5 ꢁ 10ꢀ6 mol/cm3
1.0% NO, F = 1.8 ꢁ 10ꢀ6 mol/cm3
1.0% NO, F = 3.5 ꢁ 10ꢀ6 mol/cm3
(
(
15)
NCN3 f NCN þ N2
9
5
.0 ꢁ 10
44
1
3
6
6
16)
NCN f NCN
1.1 ꢁ 10
.3 ꢁ 10
17.4
17.4
33.5
this work
this work
11
2
3
3
14
13
(
19)
20)
NCN þ CN h C2N2 þ N
NCN þ N h CN þ N2
1.3 ꢁ 10
1.0 ꢁ 10
(
48
a
40
Possible secondary chemistry has been taken into account by the GRI-Mech. 3.0 and a subset of NCN and CN reactions as outlined in our previous
paper. Rate constants are given as k
35
n
i
= AT exp[ꢀE
a
/RT]. Units are cm, mol, s, K and kJ.
Table 2. Selected Reactions of the Mechanism Used in the Numerical Simulations to Extract the Rate Constants of the Reaction
a
NCN + NO2
no.
reaction
A
E
a
ref.
35
3
3
13
12
9
1
(7)
NCN þ O h CN þ NO
9.6 ꢁ 10
4.7 ꢁ 10
5.0 ꢁ 10
5.8
set equal for NCN, see text
1
(
(
11)
15)
NCN þ NO2 f NCNO þ NO
38.0
71.2
71.2
30.1
24.8
this work
44
set equal for NCN, see text
1
F = 3.5 ꢁ 10ꢀ6 mol/cm3
F = 7.0 ꢁ 10ꢀ6 mol/cm3
NCN3 f NCN þ N2
9
7
.7 ꢁ 10
44
1
3
6
, F = 3.5 ꢁ 10ꢀ6 mol/cm3
, F = 7.0 ꢁ 10ꢀ6 mol/cm3
(16)
NCN f NCN
2.3 ꢁ 10
.7 ꢁ 10
this work
this work
0.7% NO
2
6
1
0.7% NO
2
1
1
1
1
1
1
5
2
3
2
7
3
(17)
(21)
(22)
(23)
(24)
(25)
NO2 þ M h NO þ O þ M
NO2 þ NO2 h 2NO þ O2
NO2 þ NO2 h NO3 þ NO
NO2 þ O h NO þ O2
4.0 ꢁ 10
2.0 ꢁ 10
1.0 ꢁ 10
3.9 ꢁ 10
4.0 ꢁ 10
1.0 ꢁ 10
251
105
108
ꢀ1
51
51
51
52
51
52
NO3 þ M h NO2 þ O þ M
NO3 þ O h NO2 þ O2
200
a
40
Possible secondary chemistry has been taken into account by the GRI-Mech. 3.0 and a subset of NCN and CN reactions as outlined in our previous
paper. Rate constants are given in terms of k = A exp[ꢀE /RT]. Units are cm, s, mol and kJ.
35
a
pump (Pfeiffer Vacuum). Hydrogen, nitrogen or mixtures of
hydrogen and nitrogen were used as the driver gases; diaphragms
were made from 30 or 80 μm thick aluminum foil. Pressures and
temperatures behind shock waves were calculated based on pre-
shock conditions, and the shock wave velocity was measured by a
fast count unit wired to four piezo-electric sensors (PCB Piezo-
tronics M113A21). Real gas effects and the shock wave damping
on the order of 1% were taken into account.
of the fundamental has been measured by an interferometric type
ꢀ
1
wavemeter (MetroLux WL200) with an accuracy of (0.0015 cm .
Detection and reference beam were divided by a 50:50 beam split-
ter plate, the detection beam was passed through the shock tube
and the reference beam could be attenuated by a variable neutral
density filter. Both beams were band-pass filtered and coupled
into two optical fibers (Thorlabs BF H22-550), which were con-
nected to a balanced photo detector and amplifier (Thorlabs
PDB150A-EC). The resulting difference signal was further am-
plified (Ortec Fast Preamp 9305) low pass filtered by 1.4 MHz
and stored by an analog input board (PCI-DAS4020/12, 20 MHz,
12 bit). The absorption signals were transformed into concen-
trations using the previously determined NCN absorption cross
NCN has been detected in its triplet ground state at a wave-
ꢀ
1
3
length of λ = 329.1302 cm (superposition of the Π sub-band
1
3
3
~
~
of the A Π (000) ꢀ X Σ (000) transition and the Q band head
u
g
1
3
+
3
of the Σ (010) ꢀ Π(010) vibrationally excited Renner-Teller
split transition) by a difference amplification laser absorption
scheme. Laser light (1.5 mW) of the respective wavelength was
generated by internal frequency doubling of a stabilized continu-
ous wave ring-dye laser (Coherent 899) with DCM Special
2
ꢀ4
34
section (log (σ/cm /mol)) = 8.9ꢀ8.3 ꢁ 10 ꢁ T/K).
10
The measurements were evaluated by fitting numerical simu-
lations to the measured concentrationꢀtime profiles using an ap-
propriate reaction mechanism including reaction rate constants
for NCN forming and consuming reactions (see Tables 1 and 2).
(
Radiant Dyes) as a dye, which was pumped by a Nd:YVO solid
4
state laser (Coherent Verdi V10) at λ = 532 nm. The wavelength
1
4384
dx.doi.org/10.1021/jp208715c |J. Phys. Chem. A 2011, 115, 14382–14390

The Journal of Physical Chemistry A
ARTICLE
Figure 2. Arrhenius plot of measured first order CIISC rate constants
2
at several total densities with NO or NO added to the gas mixtures.
Arrhenius expressions of rate constants in pure Ar have been determined
1
44
from NCN concentrationꢀtimeprofiles and are shownas dashed lines.
after preparation. Before all shock tube experiments, the shock
tube was flushed with the respective test gas mixtures at a pres-
sure of p ≈ 50 mbar. No dependence of measured signals on the
flushing time or the time between sample preparation and
experiment could be noticed revealing that wall absorption ef-
fects or room temperature reactions of NCN with NO or NO
3
2
did not perturb the measurements.
Figure 1. Upper: NCN Concentrationꢀtime profiles for experiments
with and without NO added to the reaction mixtures. With NO: T =
III. RESULTS AND DISCUSSION
ꢀ
6
3
9
96 K, p = 149 mbar, F = 1.80 ꢁ 10 mol/cm , x(NCN ) = 13.4 ppm;
3
ꢀ6 3
without NO: T = 1002 K, p = 150 mbar, F = 1.81 ꢁ 10 mol/cm ,
A. NCN + NO. NCN concentrationꢀtime profiles of mixtures
x(NCN ) = 14.0 ppm. Below: Sensitivity analysis of the experiment
3
of NCN and NO behind incident shock waves have been
3
with NO.
recorded in the temperature range 764 K < T < 1944 K at two
ꢀ6
3
different total densities of F ≈ 1.8 ꢁ 10 mol/cm and F ≈
39
Numerical simulations were performed using the Chemkin-II
ꢀ6
3
3
.5 ꢁ 10 mol/cm (123 mbar < p < 690 mbar). Mole fractions
program package. For sensitivity analysis, the sensitivity coeffi-
cient σ(i,j,t) for reaction i of species j at time t was normalized
with respect to the maximum concentration cmax of the species j
of NCN were between 5 and 15 ppm and much lower than the
3
mole fractions of the excess component NO, which were 5000 or
1
0000 ppm. A typical concentrationꢀtime profile together with
over the time history, σ(i,j,t) = 1/cmax ꢁ (∂c(j,t)/∂ ln k ). The
i
the respective numerical simulation and a sensitivity analysis is
shown in Figure 1. To illustrate the influence of added NO, an ex-
periment without NO is shown as well. Two strong Schlieren
signals accompany the incident and reflected shock wave pas-
sages and thus define t = 0 of the corresponding temperature and
pressure jumps.
40
GRI-Mech 3.0 natural flame mechanism has been used as a
background mechanism and an additional subset of reactions
taking into account NCN and CN secondary chemistry has been
35
adopted from our previous paper. Rate constants of reverse re-
actions have been taken into account based on thermodynamic
41
data from Konnov, except for NCN, for which the data were
Without the addition of NO, NCN consuming reactions are
negligible. Up to temperatures of T = 2000 K, stable NCN plateaus
were found. At higher temperatures, the decomposition of NCN
set in according to NCN + M f C + N (13). The initial increase
of the NCN absorption signal in Figure 1 reflects two processes:
42
taken from Goos et al.
NCN was generated by the thermal decomposition of NCN3.
NCN was synthesized from gaseous BrCN and solid NaN using
3
3
2
3
4
43
a variant of the method described by Milligan et al. NCN is an
3
extremely explosive substance, hence, no attempt was made to
further purify the initially generated gas. Its purity was checked by
means of FTIR spectroscopy. Remaining BrCN impurities were
The decomposition of the precursor molecule NCN yielding
3
1
NCN and the subsequent relaxation of the electronically excited
NCN to its triplet ground state ( NCN) via collision-induced
1
3
intersystem crossing (CIISC):
usually <1% and never exceeded 3%. CO was on the order of
2
0
.01% and water was not detectable. Freshly prepared NCN was
1
3
NCN3ð þ MÞf NCN þ N2ð þ MÞ
ð15Þ
ð16Þ
immediately diluted by argon down to mole fractions of 0.1%,
nevertheless, a slow decomposition of ∼5% per day was found to
take place.
Gases and chemicals used were argon (Air Liquide, 99.99%),
hydrogen and nitrogen as driver gases (Air Liquide, 99%), NO
1
3
NCNð þ MÞf NCNð þ MÞ
Rate constants for both processes have been reported in our pre-
3
4
vious publications. Updated rate constant expressions as used in
1
2
and NO (Air Liquide, 99.95%), BrCN (Acros, 97%), and NaN3
this work rely on recent NCN concentrationꢀtime profile mea-
44
(Merck, 99%). Gas mixtures were prepared using the partial
surements, which will be reported elsewhere. Upon the addition
3
pressure method; the NO /N O equilibrium has been taken
of 0.5 ꢀ 1% NO, the initial increase of the NCN signal gets much
2
2 4
into account. NO and NO were purified by repeated freezeꢀ
faster. Since the unimolecular decomposition process can be
2
pumpꢀthaw cycles. All gas mixtures were used within 3 days
expected to be only slightly affected by such small amounts of a
1
4385
dx.doi.org/10.1021/jp208715c |J. Phys. Chem. A 2011, 115, 14382–14390

The Journal of Physical Chemistry A
ARTICLE
different bath gas, this increase is attributed to a much more
efficient CIISC process of NO compared to argon. To support
this assumption, test experiments have been performed with the
1
detection laser frequency tuned to a strong NCN transition ( ~v =
ꢀ
1
1
3
0045.46 cm ). In the measured NCN concentrationꢀtime
profiles, the addition of NO did not alter the formation rate of
NCN, but had a strong influence on the NCN decays.
1
1
With 1% NO added to the mixture, rise times are shortened by
a factor of about 5 showing that NO exhibits a CIISC efficiency
higher by a factor of 400 compared to argon. Such a strong de-
pendence of the CIISC efficiency on the nature and spin state of
1
the collision gas is not unusual. For example, in the case of NH,
collisions with Xe lead to CIISC rate constants 3 orders of
45
magnitude higher than collisions with Ar. Rate constants of the
relaxation reaction 16 were extracted from the measured profiles
44
and are shown together with corresponding data for argon and
NO (see Section III.B) in Figure 2. Similar values of the ap-
parent activation energies on the order of 20 kJ/mol are found for
Figure 3. Comparison of measured rate constants of reaction 10,
2
1
7,24,25
NCN + NO, with selected literature data.
of Baren and Hershberger and Welz have been scaled to 133 mbar
based on the Troe parameters given in the work of Welz.
The literature values
2
4
25
NO, NO and Ar. The apparent activation energies as well as the
2
observed density dependences (n ≈ 1 for NO, n ≈ 0.5 for NO ,
2
n
n ≈ 0.6 for Ar; according to k = ((F )/(F )) k ) reflect the com-
b
b
a
a
1
the reaction NCN + NO. However, as the observed absolute
plicated underlying physical processes that govern the CIISC
3
4
6,47
NCN mole fractions were found to be in quantitative agreement
probability and are difficult to predict.
with the initial NCN precursor mole fractions within error
As it is discernible in the sensitivity analysis shown in the lower
frame of Figure 1, the CIISC process is completed within 50 μs
and thus does not critically interfere with the rate constant mea-
surement of the reaction NCN + NO. In fact, reaction 10 is the
only sensitive NCN consuming reaction showing that the mea-
surements essentially follow pseudo-first-order kinetics. Possible
secondary chemistry is successfully suppressed by using very low
3
1
limits, a major influence of the reaction NCN + NO can be ruled
out anyway. Given that no literature data are available, for the
1
3
sake of a complete reaction mechanism, the NCN and NCN ra-
dical reactions have been assumed to proceed with the same rate.
3
From detailed simulations of the absolute NCN yield, at least an
1
upper limit for the NCN reaction could be estimated. According
to these simulations, a factor of 20 higher reaction rate constant for
(
∼10 ppm) mole fractions of NCN . Rate constants were
3
1
3
NCN + NO should have caused a detectable decrease of NCN
extracted in the temperature range of 764 K < T 1944 K. Toward
yield. On one hand, using such a high value in the numerical
simulations would have resulted in a merely 20% smaller rate
constants for the CIISC process, but would not have altered our
general conclusion that NO is a much more efficient collider than
Ar. On the other hand, the same rate constants would have been
low temperatures, the accessible temperature range was limited
by the rate of the NCN decomposition. It is slower than the
3
1
NCN relaxation below T = 750 K. At high temperatures, NCN
absorption cross sections are too low, and higher NCN concen-
3
trations used to compensate for the lower sensitivity resulted in
significant secondary chemistry. Selected rate constants of the
reaction used in the mechanism can be found in Table 1.
An Arrhenius plot of the rate constant data for k including
3
obtained for the reaction NCN + NO. These were extracted from
3
the pseudo-first-order decays ofthe NCN profiles atlong reaction
1
times, whereas any influence of NCN is restricted to the first few
1
0
17,24,25
microseconds.
selected literature results
is shown in Figure 3. Data of the
In contrast to all previous experimental studies, which have
been performed at temperatures below T = 573 K, a positive tem-
perature dependence has been found at temperatures greater
than 764 K. This positive temperature dependence has been pre-
dicted by Huang et al. in their theoretical study. In Figure 3,
theoretically derived rate constant expressions for two different
total pressures are shown. At low temperatures, according to a re-
combination process, the reaction proceeds close to the low
pressure limit at the investigated pressures p < 1 bar. The cor-
experimental conditions are given in the Supporting Information.
No dependence of the measured rate constants on the total den-
sity is discernible within the scatter of the data. All data points fall
within a narrow margin and can be very well represented by the
Arrhenius expression:
17
3
ꢀ1
ꢀ1Þ ¼ 1:89 ꢁ 1012 exp½ ꢀ 26:3ðkJ=molÞ=RTꢂð ( 7%Þ
k10=ðcm mol
s
The stated error estimate reflects a combination of the statistical
error of the Arrhenius fit ((1%, 2σ), uncertainties of the NO
responding density dependence is also supported by the experi-
3
17,24,25
mole fraction ((5%), and uncertainties in the NCN formation
mental studies.
In contrast, at high temperatures we find
mechanism ((1%). Due to the pseudo-first-order reaction con-
ditions, uncertainties of the NCN absorption cross section and
thus NCN concentration did not result in an additional error. The
uncertainty of the Arrhenius activation energy of (1.6 kJ/mol
corresponds to the 2σ error of the slope of the Arrhenius fit.
reaction rates that are independent of the total density. Rate con-
stant values are approximately a factor of 2 lower than the Huang
17
et al. prediction. Although no direct evidence of the postulated
minimum of the rate constants is provided by our measurements,
all experimental data are consistent with a minimum at a tem-
perature around T ≈ 670 K.
1
In additional measurements, NCN concentrationꢀtime profiles
have been recorded at a detection wavelength of λ = 332.8290 nm
A qualitative explanation of the complex temperature and
pressure dependences can be illustrated on the basis of the
characteristics of the potential energy surface. A more detailed
assessment, however, would have to rely on a full master equation
1
to assess potential differences of the reactivities of NCN and
3
NCN. Unfortunately, because of the fast CIISC process, the
evaluation of these signals did not yield reliable rate constants for
1
4386
dx.doi.org/10.1021/jp208715c |J. Phys. Chem. A 2011, 115, 14382–14390

The Journal of Physical Chemistry A
ARTICLE
Figure 4. Part of the potential energy surface of the reaction NCN +
NO 10 (G2M/CC5//B3LYP/6-311+G(d) level of theory), adopted
17
from Huang et al.
analysis. A simplified scheme of the potential energy surface,
17
adopted from the calculations of Huang et al., is outlined in
Figure 4. Other possible reaction channels forming CNO + N or
2
NCO + N do not play a role because the barriers on the cor-
2
responding minimum energy paths are too high. Two different
electronic states are involved. A simple association of NCN and
2
00
NO leads to the A state (dashed lines) of NCNNO with two
2
00
2
00
different isomers denoted cis- A and trans- A . The reaction
2
0
continues by crossing to the A surface through two conical in-
2
0
2
0
tersections (CI) yielding the local minima trans- A and cis- A
Figure 5. Upper: Absorbance-time profile of an NO2 experiment
NCNNO. Further dissociation to the products CN + N O in-
2
behind an incident shock wave. Both NO
absorbance are shown. Due to the interfering NO
2
absorbance and NCN
background absorp-
volves barriers that are significantly higher in energy than the
entrance energy of the initial NCN + NO fragments. At low tem-
peratures, these barriers prevent the formation of the products
2
tion, the baseline is not zero. Lower: NCN sensitivity analysis indicating
pseudo-first-order conditions after 100 μs.
CN + N O. Instead, the trans- and cis-NCNNO intermediates
2
can be stabilized by collisions. At not too high pressures, the
typical absorbance-time profile. Absorbance of NCN has been
used instead of NCN mole fractions to illustrate the effect of in-
2
0
17
formation of the A isomers dominates. Consequently, the
reaction exhibits the typical behavior of a recombination reaction,
i.e., a negative temperature dependence (faster redissociation to the
reactants at higher temperature) and a positive pressure depen-
dence (collisional stabilization of the intermediates). Toward high
temperatures, however, enough thermal energy is involved to over-
terfering NO background absorption stemming from its broad
2
2
2
49
A~ B ꢀ X~ A transition. Already before the arrival of the shock
2
1
wave, significant NO absorption is discernible. Due to the higher
2
density, higher background absorption values are observed behind
the incident shock wave. Superimposed to this background absorp-
2
0
come the exit barriers of the cis- and trans- A intermediates. More-
over, due to the increasingly fast forward reaction, collisional stabi-
lization becomes less significant. This isthe reason why virtually no
pressure dependence has been observed in our experiments.
Hence, the forward reaction resembles an indirect abstraction
reaction with a positive temperature dependence stemming from
tion, the NCN absorbance profile indicates the thermal decom-
3
position of NCN followed by CIISC and NCN formation. The
3
decay at longer times can be attributed to reaction 11, NCN + NO2.
In order to take into account NO background absorption,
2
temperature-dependent NO absorption cross sections at ~v =
2
ꢀ
1
30383.11 cm had been measured beforehand in experiments
the exit barrier. The measured activation energy of E = 26 kJ/mol
without NCN . NO absorption coefficients are best represented
a
3
2
is between the barriers of the trans- (61 kJ/mol with respect to the
by the expression
2
0
reactants) and the cis- A (14 kJ/mol) pathways. The much closer
agreement with the cis value reveals that the overall reaction at
high temperatures is dominated by the cis pathway.
2
ꢀ1
ꢀ4
log ðσ=cm mol Þ ¼ 5:28 ꢀ 1:21 ꢁ 10 ꢁ T=K
10
An extrapolation to room temperature yields a value of 1.8 ꢁ
B. NCN + NO . The kinetics of the reaction NCN + NO f
5
2
2
2
1
1
0 cm /mol, thus in very good agreement with a value of 1.9 ꢁ
NCNO + NO 11 has been investigated in the temperature range
5
2
50
0 cm /mol reported by Bogumil et al. A plot of the measured
7
3
04 K < T < 1659 K and at two different total densities of F ≈
cross sections can be found in the Supporting Information. NO2
ꢀ6 3 ꢀ6 3
.5 ꢁ 10 mol/cm and F ≈ 7.0 ꢁ 10 mol/cm . Correspond-
concentrationꢀtime profiles were simulated based on the initial
ing pressures were 182 mbar < p < 654 mbar. Initial mole fractions
NO concentrations and the reaction mechanism summarized in
2
of NCN in Ar were between 3 and 14 ppm, mole fractions of the
Table 2. Corresponding absorbance values are denoted “NO ab-
sorption” in Figures 5 and 6. Remaining absorbances have been at-
3
2
excess component NO were 7000 ppm or, for a validation ex-
2
periment, 19000 ppm. Detailed data of the experimental condi-
tions are given in the Supporting Information. Figure 5 shows a
tributed to NCN absorption, which could be extracted from the
profiles with high precision. As for the reaction NCN + NO, no
1
1
4387
dx.doi.org/10.1021/jp208715c |J. Phys. Chem. A 2011, 115, 14382–14390

The Journal of Physical Chemistry A
ARTICLE
Figure 7. Arrhenius plot of the measured rate constants for the reaction
NCN + NO
2
at two different total densities. For comparison, low-
18,24,26
temperature experimental data
as well as theoretical predictions of
Yang et al. are also shown. The original 4 mbar value of Baren and
1
8
2
4
Hershberger has been scaled to a pressure of 133 mbar based on the
2
6
Troe parameters reported by Kappler.
sults from a 5% uncertainty of the NO concentration, a 7%
2
statistical error (2σ), and 2% uncertainty attributable to the
precursor molecule decomposition and the singlet cyanonitrene
relaxation mechanism. Although uncertainties resulting from
possible inaccuracies of the NO decomposition mechanism are
2
only relevant at the high-temperature limit, based on simulations
with varied k , allowance was made for an additional 5% error for
17
Figure 6. High-temperature NO
sensitivity analysis.
2
experiment and corresponding NCN
k . The Arrhenius activation energy is accurate to ΔE = ( 3.8
1
1
a
kJ/mol. This estimate already includes the potential uncertainty
of the high temperature data points resulting from the NO de-
2
indication was found for a significant influence of the reaction
composition mechanism.
Selected results from the literature
Figure 7. At room temperature, all results are in very good agree-
ment, but toward higher temperatures, differences become
1
18,24,26
NCN + NO . Nevertheless, for the sake of a completeness,
are also included in
2
we included this reaction in the mechanism by assuming the same
rate constant for the singlet and triplet reactions.
CIISC rate constants extracted from the resulting NCN
26
apparent. The rate constant data by Kappler exhibit a signifi-
concentrationꢀtime profiles are included in Figure 2. The relaxa-
cantly stronger temperature dependence than those measured by
18
tion process mediated by collisions with NO is less efficient than
Yang et al. In both studies, rate constants have been determined
assuming pseudo-first-order conditions, thus excluding second-
ary chemistry. However, as Kappler already discussed in her
work, the used excimer laser photolysis at λ = 248 nm next to
2
with NO, but still a factor of approximately 40 more efficient than
3
with argon. After 100 μs, the NCN formation is completed, and
a decay of the absorption can be observed in Figure 5. The sen-
sitivity analysis reveals that reaction 11 is the only important re-
action for the consumption of NCN. In this regard, the ex-
perimental conditions are close to pseudo-first-order conditions
and thus enable an accurate determination of k . However, at
NCN radicals also generates NO and O atoms from NO
photo-
2
hν
lysis, NO
s NO + O (18). The possible overestimation of
f
2
the determined rate constants due to the very fast subsequent re-
action7, NCN+O,wasestimatedbyKapplertobe<5%.Yangetal.
used 193 nm instead of 248 nm photolysis. At this wavelength, NO
11
temperatures of >1400 K, more reactions gain importance. Such
an experiment is shown in Figure 6. The unimolecular decompo-
sition of NO , NO + M f NO + O + M (17) sets in, and thus
2
absorption cross sections are much larger, and O atom yields can
be expected to be higher. Although Yang et al. argued that O atom
formation did not interfere in their experiments due to subtle dif-
fusion effects, the neglected O atom chemistry via reaction 7 may
account for the differences and the data of Kappler can therefore
be assumed to be more reliable. The dashed curves in Figure 7 re-
present the theoretical fit of Yang et al. Clearly, a strong pressure
dependence emerges. However, due to the mentioned effect, the
low temperature rate constants may be somewhat overpredicted.
Our data represent the first high-temperature measurements
and can only be compared with the theoretical prediction of Yang
2
2
the consecutive reaction NCN + O (7) plays a role. This in-
terfering secondary chemistry constitutes a high temperature
limit for k11 measurements of about T ≈ 1700 K. Again, the low
temperature limit was set by the decomposition rate of NCN .
3
Measured rate constant values are compared to selected liter-
ature data in Figure 7. All data points can be nicely represented by
a straight Arrhenius fit, thus supporting the reliability of the
applied evaluation procedure.
3
ꢀ1
ꢀ1Þ ¼ 4:7 ꢁ 1012 exp½ ꢀ 38:0ðkJ=molÞ=RTꢂð ( 19%Þ
18
k11=ðcm mol
s
et al. In agreement with their work and similar to the reaction
NCN + NO, a positive temperature dependence is found at high
temperatures. Note that for the temperature and pressure range
of this work (182 mbar < p < 654 mbar), the overall reaction is
At temperatures 704 K < T < 1659 K, the rate constants were
found to be independent of the total density. The stated error re-
1
4388
dx.doi.org/10.1021/jp208715c |J. Phys. Chem. A 2011, 115, 14382–14390

The Journal of Physical Chemistry A
ARTICLE
NCNO + NO (11) have been measured. For reaction 10,
rate constants are represented by the Arrhenius expression
3
ꢀ1 ꢀ1
12
k /(cm mol
s
) = 1.9 ꢁ 10 exp[ꢀ26.3(kJ/mol)/RT] in
1
0
the temperature range 764 K < T < 1944 K. The experimentally
accessible temperature range was limited by the decomposition rate
of the NCN precursor NCN at the lower bound and too small
3
NCN absorption cross sections at high temperatures. Regarding re-
3
ꢀ1 ꢀ1
12
action 11, the expression k /(cm mol s ) = 4.7 ꢁ 10
11
exp[ꢀ38.0(kJ/mol)/RT] accounts for the measured data points
in the temperature range 691 K < T < 1659 K. Here, the tem-
perature range was limited by the onset of the NO decomposition
2
andinterferingsecondarychemistry. Instrikingcontrasttoprevious
low temperature studies, at high temperatures no pressure but a
pronounced positive temperature dependence has been deter-
Figure 8. Part of the potential energy surface of the reaction NCN +
NO (11) (G2M//B3LYP/6-311+G(d) level of theory), adopted from
mined. For reaction 10, E = (26.3 ( 1.6) kJ/mol and for reaction
2
a
18
Yang et al.
1
1 E = (38.0 ( 3.8) kJ/mol. These values are in agreement with
a
theoretical predictions and prove the postulated change of the
overall reaction from a recombination to an abstraction-like path-
way. Together with the low-temperature data taken from the
literature, the position of the minimum of the overall reaction rate
constant has been estimated to be around T ≈ 670 K for reaction
10 and T ≈ 500 K for reaction 11. Additionally, the influence of
predicted to be almost independent of the experimental pressure.
The predicted absolute values of k and the absence of pressure
11
dependence is in very good agreement with our data. Only at the
lowest experimental temperatures are deviations of about a factor
of 2 observed. These remaining discrepancies can be easily explained
by a slightly overpredicted temperature dependence. Unfortu-
nately, the expected inversion of the temperature dependence
could not be observed in the experiments. Obviously, at pres-
sures below p < 1 bar, the rate constant minimum arise at
temperatures T < 700 K. Combining the low-temperature data of
Kappler and the high-temperature rate constants determined in
this work, a rough estimate of T ≈ 500 K can be made for p = 133
mbar, hence ≈150 K lower than calculated by Yang et al.
The switch from a negative to a positive temperature depen-
dence can be explained by the competition of an association and
an abstraction reaction mechanism. The dominating reaction
channels of the potential energy surface as it has been explored
NO and NO on the CIISC process, which transfers electronically
2
1
3
excited NCN to the NCN ground state, has been investigated.
With respect to CIISC probability, NO and NO are 400 and 40
2
times more efficient collision partners than argon, respectively.
’
ASSOCIATED CONTENT
S
Supporting Information. Tables of the experimental data
b
and conditions as well as temperature-dependent absorption cross
sections of NO can be found in the Supporting Information. This
2
information is available free of charge via the Internet at http://
pubs.acs.org.
1
8
by Yang et al. are reproduced in Figure 8. Additional reaction
channels leading to CNO + N O or CN + N O do not play a role
2
2 2
’
AUTHOR INFORMATION
due to very high associated barriers.
Similar to the reaction NCN + NO, at low temperatures the re-
action is a typical example of an association reaction forming the
Corresponding Author
*Electronic address: friedrichs@phc.uni-kiel.de.
collisionally stabilized recombination product NCNNO . The
2
sequential transition state lies energetically too high, (116 kJ/mol
above the minimum, 34 kJ/mol above the entrance energy) such
’ ACKNOWLEDGMENT
that the thermodynamically most stable products NCO and N O
are not formed. At high temperatures, with a threshold energy of
E0 = 41 kJ/mol, a direct abstraction reaction channel becomes ac-
2
We would like to thank the German Science Foundation (FR
529/4), the cluster of excellence “The Future Ocean” (DFG -
EC 80), and Friedrich Temps for financial and scientific support.
Thanks to Anna Busch, Claudia Kappler, Matthias Olzmann, and
Oliver Welz (Karlsruher Institut f €u r Technologie, KIT) for
sharing data prior to publication.
1
cessible. Here, an O atom is abstracted from NO yielding the
2
high-temperature reaction products NCNO and NO. According
to Yang et al., even at high temperatures the contribution of the
reaction channel forming NCO + N O remains negligible. Cor-
2
responding rate constants are predicted to be several orders of
magnitude lower than for the NCNO + NO channel.
’ REFERENCES
The positive temperature dependence of the overall reaction,
which has been determined in this work, supports the proposed
reaction mechanism. The measured Arrhenius activation energy
(
1) Warnatz, J.; Maas, U.; Dibble, R. W. Combustion, 3rd ed. ;
Springer: Berlin/Heidelberg, 2001.
(2) Harrison, R. M., Ed. Principles of Environmental Chemistry; RSC
Publishing: Cambridge, 2007.
of E = 38.0 kJ/mol is in close agreement with the calculated
a
(3) Forster, P.; Ramaswamy, V.; Artaxo, P.; Berntsen, T.; Betts, R.;
threshold energy of E = 41 kJ/mol corresponding to the direct
0
Fahey, D.; Haywood, J.; Lean, J.; Lowe, D.; Myhre, G.; Nganga, J.; Prinn,
R.; Raga, G.; Schulz, M.; Dorland, R. V. Climate Change 2007: The
Physical Science Basis. Contribution of Working Group I to the Fourth
Assessment Report of the Intergovernmental Panel on Climate Change;
Solomon, S., Qin, D., Manning, M., Chen, Z., Averyt, M. M. K.,
Tignor, M., Miller, H., Eds.; Cambridge University Press: Cambridge,
United Kingdom, 2007.
NCNO + NO abstraction channel.
IV. CONCLUSION
For the first time, high-temperature rate constants of the
reactions NCN + NO f N O + CN (10) and NCN + NO f
2
2
1
4389
dx.doi.org/10.1021/jp208715c |J. Phys. Chem. A 2011, 115, 14382–14390

The Journal of Physical Chemistry A
ARTICLE
(
4) Glarborg, P.; Alzueta, M. U.; Dam-Johansen, K.; Miller, J. A.
Elementarreaktionen bei hohen Temperaturen., Ph.D. Thesis, Universit €a t
G €o ttingen, 1991.
Combust. Flame 1986, 65, 177–202.
(
(
5) Zeldovich, Y. B. Acta Physicochim. USSR 1946, 21, 577–628.
6) Baulch, D. L.; Cobos, C. J.; Cox, A. M.; Frank, P.; Hayman, J.;
(39) Kee, R. J.; Ruply, F. M.; Miller, J. A. Chemkin-II: A Fortran
Chemical Kinetics Package for the Analysis of Gas Phase Chemical Kinetics;
Sandia report sand89-8009; Sandia National Laboratories: Livermore,
CA, 1989.
Kerr, J. A.; Murrels, T.; Pilling, M. J.; Troe, J.; Walker, R. W.; Warnatz, J.
J. Phys. Chem. Ref. Data 1994, 23, 847–1033.
(7) Wolfrum, J. Chemieingenieurtechnik 1972, 44, 656–659.
(8) Malte, P. C.; Pratt, D. T. Proc. Combust. Inst. 1974, 15, 1061–1070.
(9) Fenimore, C. P. Proc. Combust. Inst. 1971, 13, 373–380.
(10) Cui, Q.; Morokuma, K.; Bowman, J. M.; Klippenstein, S. J.
(40) Smith, G. P.; Golden, D. M.; Frenklach, M.; Moriarty, N. W.;
Eiteneer, B.; Goldenberg, M.; Bowman, C. T.; Hanson, R. K.; Song, S.;
Jr., W. C. G.; Lissanski, V. V.; Qin, Z. GRI-MECH 3.0, http://www.me.
berkeley.edu/gri_mech/.
J. Chem. Phys. 1999, 110, 9469–9482.
(41) Konnov, A. A. Detailed reaction mechanisms for small hydro-
carbon combustion, http://homepages.vub.ac.be/akonnov/science/
mechanism/version0_5.html, 2007, release 0.5.
(11) Moskaleva,L.V.;Lin,M.C.Proc. Combust. Inst. 2000,28,2393–2401.
12) Moskaleva, L. V.; Xia, W. S.; Lin, M. C. Chem. Phys. Lett. 2000,
(
3
31, 269–277.
(42) Goos, E.; Burcat, A.; Ruscic, B. Ideal Gas Thermochemical
Database with updates from Active Thermochemical Tables, ftp://ftp.
technion.ac.il/pub/supported/aetdd/thermodynamics, 2010.
(43) Milligan, D. E.; Jacox, M. E.; Bass, A. M. J. Chem. Phys. 1965, 43,
3149–3160.
(44) Dammeier, J.; Oden, B.; Friedrichs, G. A consistent model for
3
the thermal decomposition of NCN and the singlet-triplet relaxation of
(13) Smith, G. P. Chem. Phys. Lett. 2003, 367, 541–548.
(14) Sutton, J. A.; Williams, B. A.; Fleming, J. W. Combust. Flame
2
008, 153, 465–478.
15) Klein-Douwel, R. J. H.; Dam, N. J.; ter Meulen, J. J. Opt. Lett.
008, 33, 2620–2622.
16) Vasudevan, V.; Hanson, R. K.; Bowman, C. T.; Golden, D. M.;
Davidson, D. F. J. Phys. Chem. A 2007, 111, 11818–11830.
17) Huang, C.-L.; Tseng, S. Y.; Wang, T. Y.; Wang, N. S.; Xu, Z. F.;
Lin, M. C. J. Chem. Phys. 2005, 122, 184321/1–9.
18) Yang, T.-J.; Wang, N. S.; Lee, L. C.; Xu, Z. F.; Lin, M. C. J. Phys.
Chem. A 2008, 112, 10185–10192.
(
2
(
NCN. Int. J. Chem. Kinet., submitted for publication.
(
(45) Hack, W.; Wilms, A. J. Phys. Chem. 1989, 93, 3540–3546.
(46) Freed, K. F. Adv. Chem. Phys. 1981, 47, 291–336.
(47) Bley, U.; Temps, F. J. Chem. Phys. 1993, 98, 1058–1072.
(48) Slack, M. W. J. Chem. Phys. 1975, 64, 228–236.
(49) Delon, A.; Georges, R.; Kirmse, B.; Jost, R. Faraday Discuss.
1995, 102, 117–128.
(
(19) Chen, H.-T.; Ho, J.-J. J. Phys. Chem. A 2005, 109, 2564–2571.
(20) Wei, Z.-G.; Li, Q.-S.; Zhang, S.-W.; Sun, Y.-B.; Sun, C.-C. J. Mol.
Struc. - THEOCHEM 2005, 722, 139–146.
(50) Bogumil, K.; Orphal, J.; Homann, T.; Voigt, S.; Spietz, P.;
Fleischmann, O.; Vogel, A.; Hartmann, M.; Kromminga, H.; Bovensmann, H.;
Frerick, J.; Burrows, J. J. Photochem. Photobiol. A 2003, 157, 167–184.
(51) R €o hrig, M.; Petersen, E. L.; Davidson, D. F.; Hanson, R. K. Int.
J. Chem. Kinet. 1996, 28, 599–608.
(21) Zhu, R. S.; Lin, M. C. Int. J. Chem. Kinet. 2005, 37, 593–598.
(22) Zhu, R. S.; Lin, M. C. J. Phys. Chem. A 2007, 111, 6766–6771.
(23) Zhu, R. S.; Nguyen, M. T.; Lin, M. C. J. Phys. Chem. A 2009,
1
13, 298–304.
24) Baren, R. E.; Hershberger, J. F. J. Phys. Chem. A 2002, 106,
1093–11097.
25) Welz, O. Laserspektroskopische Untersuchungen und Mole-
(
(52) Atkinson, R.; Baulch, D. L.; Cox, R. A.; Hampson, R. F.; Kerr,
J. A.; Troe, J. J. Phys. Chem. Ref. Data 1989, 18, 881–1108.
1
(
kularkinetische Modellierung der Kinetik von Radikalreaktionen in der
Gasphase, Ph.D. Thesis, Universit €a t Karlsruhe (TH), Universit €a tsverlag
Karlsruhe, Karlsruhe, 2009, ISBN 978ꢀ3-86644ꢀ356ꢀ3.
(26) Kappler, C. Untersuchungen komplexbildender bimolekularer
Reaktionen in der Gasphase mit laserspektroskopischen Methoden und
statistischer Reaktionstheorie, Ph.D. Thesis, Karlsruher Institut f €u r
Technologie, KIT Scientific Publishing, Karlsruhe, 2010, ISBN 978ꢀ
3
-86644ꢀ567ꢀ3.
(
27) Lamoureux, N.; Desgroux, P.; Bakali, A. E.; Pauwels, J. F.
Combust. Flame 2010, 157, 1923–1941.
28) El Bakali, A.; Pillier, L.; Desgroux, P.; Lefort, B.; Gasnot, L.;
Pauwels, J. F.; da Costa, I. Fuel 2006, 85, 896–909.
29) Gersen, S.; Mokhov, A. V.; Levinsky, H. B. Combust. Flame
008, 155, 267–276.
(
(
2
(30) Konnov, A. A. Combust. Flame 2009, 156, 2093–2105.
(31) Sepman, A. V.; Mokhov, A. V.; Levinsky, H. B. Int. J. Hydrogen
Energy 2011, 36, 4474–4481.
(32) Busch, A.; Olzmann, M. Shock-Tube Study of the Thermal
Decomposition of NCN, Vienna, paper p810138, Proc. European Com-
bust. Meeting, 2009.
(
33) Busch, A. Stoßwellenuntersuchungen zum Zerfall stickstoffhal-
tiger Verbindungen mit spektroskopischen Methoden, Ph.D. Thesis,
Karlsruher Institut f €u r Technologie, KIT Scientific Publishing, Karlsruhe,
2
010, ISBN 978ꢀ3-86644ꢀ586ꢀ4.
34) Dammeier, J.; Friedrichs, G. J. Phys. Chem. A 2010, 114,
2963–12971.
35) Dammeier, J.; Faßheber, N.; Friedrichs, G. Phys. Chem. Chem.
Phys. DOI: 10.1039/c1cp22123j.
36) Dammeier, J.; Colberg, M.; Friedrichs, G. Phys. Chem. Chem.
Phys. 2007, 9, 4177–4188.
(
1
(
(
(
(
37) Colberg, M.; Friedrichs, G. J. Phys. Chem A 2006, 110, 160–170.
38) Klatt, M. Quantitative Untersuchung der Bildung und des
Verbrauchs von H- und O-Atomen sowie OH-Radikalen in verschiedenen
1
4390
dx.doi.org/10.1021/jp208715c |J. Phys. Chem. A 2011, 115, 14382–14390