Gaseous nitryl azide N4O2: A joint theoretical and experimental study

Article abstract of DOI:10.1016/j.molstruc.2006.11.034

Gaseous nitryl azide N₄O₂ is generated by the heterogeneous reaction of gaseous ClNO₂ with freshly prepared AgN₃ at -50 °C. The geometric and electronic structure of the molecule in the gas phase has been characterized by in situ photoelectron spectroscopy (PES) and quantum chemical calculations. The experimental first vertical ionization energy of N₄O₂ is 11.39 eV, corresponding to the ionization of an electron on the highest occupied molecular orbital (HOMO) {4a″(π_nb(N4-N5-N6))}^-1. An apparent vibrational spacing of 1600 ± 60 cm^-1 (ν_asO1N2O3) on the second band at 12.52 eV (π_nb(O1-N2-O3)) further confirms the preference of energetically stable chain structure in the gas phase. To complement the experimental results, the potential-energy surface of this structurally novel transient molecule is discussed. Both calculations and spectroscopic results suggest that the molecule adopts a trans-planar chain structure, and a five-membered ring decomposition pathway is more favorable.

Full text of DOI:10.1016/j.molstruc.2006.11.034

Journal of Molecular Structure 840 (2007) 59–65
www.elsevier.com/locate/molstruc
Gaseous nitryl azide N4O2: A joint theoretical and experimental study
a,
a
c
a
Xiaoqing Zeng a,b, Maofa Ge *, Zheng Sun , Jiang Bian , Dianxun Wang
a
Beijing National Laboratory for Molecular Sciences (BNLMS), State Key Laboratory for Structural Chemistry of Unstable
and Stable Species, Institute of Chemistry, Chinese Academy of Sciences, Beijing 100080, China
b
Graduate School of Chinese Academy of Sciences, Beijing 100039, China
College of Chemistry and Molecular Engineering, Peking University, Beijing 100871, China
c
Received 13 September 2006; received in revised form 6 November 2006; accepted 10 November 2006
Available online 18 December 2006
Abstract
Gaseous nitryl azide N4O2 is generated by the heterogeneous reaction of gaseous ClNO2 with freshly prepared AgN3 at ꢀ50 ꢀC. The
geometric and electronic structure of the molecule in the gas phase has been characterized by in situ photoelectron spectroscopy (PES)
and quantum chemical calculations. The experimental first vertical ionization energy of N4O2 is 11.39 eV, corresponding to the ionization
of an electron on the highest occupied molecular orbital (HOMO) {4a00(pnb(N4–N5–N6))}ꢀ1. An apparent vibrational spacing of
1600 60 cmꢀ1 (masO1N2O3) on the second band at 12.52 eV (pnb(O1–N2–O3)) further confirms the preference of energetically stable chain
structure in the gas phase. To complement the experimental results, the potential-energy surface of this structurally novel transient mol-
ecule is discussed. Both calculations and spectroscopic results suggest that the molecule adopts a trans-planar chain structure, and a five-
membered ring decomposition pathway is more favorable.
ꢁ 2006 Elsevier B.V. All rights reserved.
Keywords: Nitryl azide; HeI photoelectron spectroscopy (PES); Ionization energy; IRC
1. Introduction
the isolation of pure states under normal conditions,
although there have been many theoretical discussions.
Similar to N5þ [8] and N4O [4], nitryl azide N4O2 decom-
poses completely to two moles of thermodynamically stable
N2O with an energy release of 58.9 kcal molꢀ1 (MP2/6-
31+G*) [5b]. The first attempt to prepare this novel com-
pound was carried out by Doyle et al., they approached
this unstable nitrogen oxide in CCl4 solution through the
‘‘reaction between azide and nitronium ions (LiN3 and
NO2BF4)’’, and pointed out that N4O2 is significantly more
stable than nitrosyl azide (N4O), which decomposes solely
to nitrous oxide [5a]. Then Klapo¨tke et al. achieved strong
evidence for the formation of nitryl azide through the reac-
tion of NO2SbF6 with NaN3 in liquid CO2 (ꢀ55 ꢀC 6
T 6 ꢀ35 ꢀC), and for the first time provided its experimen-
tal vibrational frequencies by means of low-temperature
Raman spectroscopy [5b]. However, no further experimen-
tal properties for this intrinsically unstable molecule have
been reported until now. As mentioned by Klapo¨tke
et al., N4O2 is highly unstable and reactive in the
Since the identification of N2O, NO and NO2 in the 18th
century, many other oxides of nitrogen have also been
known, such as N2O3, N2O5, NO3 and several p*–p* bound
binary complexes (N2O2, N2O4 (N2O)2 and (N2O)3) [1].
The research on them mainly focuses on structures and
bonding, also their effects on the atmospheric environment
(NOx) through the combustion of fossil fuels [2], and their
extensive applications as oxidizers in the rocket fuels [3].
Recently, large efforts have been made on understanding
the structures and bonding properties of binary nitrogen
oxides as well as isoelectronic high energetic homologues,
such as N4O [4], N4O2 [5], O(N3)2 [6], and ON–NCO,
O2N–NCO, N–NCO, ON–NC, O2N–NC, etc. [7]. These
novel nitrogen-containing compounds decompose at room
temperature with large release of energy, which precludes
*
Corresponding author. Tel.: +86 10 62554518; fax: +86 10 62559373.
E-mail address: gemaofa@iccas.ac.cn (M. Ge).
0022-2860/$ - see front matter ꢁ 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.molstruc.2006.11.034

60
X. Zeng et al. / Journal of Molecular Structure 840 (2007) 59–65
condensed phase at high temperature, so, similar to the
preparation of ONNCO [7c] and SO(N3)2 [9a], we studied
the gas–solid reaction of ClNO2 with finely divided AgN3
at different temperatures in vacuum (10ꢀ4 torr), the gas-
eous product was monitored in real time with in situ pho-
toelectron spectroscopy (PES). Only decomposition
product N2O was observed above ꢀ30 ꢀC, by lowering
the temperature to ꢀ50 ꢀC, the novel transient N4O2 was
detected without decomposition. Its electronic structure
was characterized for the first time by in situ photoelectron
spectroscopy (PES), in combination with theoretical calcu-
lations. To elucidate the decomposition mechanism from
the point of experimental results, the potential-energy sur-
face was also discussed.
effect of electron correlation and reorganization beyond
the Hartree-Fock approximation. The self-energy part
was expanded up to third-order, and contributions of high-
er-orders were estimated by means of a renormalization
procedure. On discussing the N4O2 potential-energy sur-
face, all the geometric structures and the harmonic vibra-
tional frequencies of two isomers of N4O2, three
transition states during the decomposition process were
calculated at B3LYP/6-311+G(d) level, the single point
of energy for all species were calculated with CCSD(T)/6-
311+G(d) [15] and MP2/6-311+G(3df) [16] levels of theo-
ry; the minimum-energy pathways for decomposition of
the N4O2 were confirmed using the intrinsic reaction coor-
dinate (IRC) method [17]. All above calculations were per-
formed using Gaussian 98 Program [18].
2. Experiment and quantum chemical calculations
3. Results and discussions
Caution: Pure dried AgN3 is extremely explosive, it
should be handled only on a small scale with appropriate
safety precautions (face shield, leather gloves, and protec-
tion clothing).
PES was recorded on a double-chamber UPS-II
machine [9] which was built specifically to detect transient
species at a resolution of about 30 meV indicated by the
standard Ar+(2P2/3) photoelectron band. Experimental ver-
tical ionization energies (IP in eV) are calibrated by simul-
taneous addition of a small amount of argon and methyl
iodide to the sample.
Recently, many azide-containing compounds have been
successfully prepared and characterized, and found that
AgN3 is an ideal precursor for in situ preparation of highly
explosive azides, through heterogeneous reactions with
reactive halogen-containing molecules [9]. The freshly pre-
pared AgN3 (0.5 g) [10] should be dried in vacuum
(10ꢀ4 torr) prior to the PES experiment, and the silver salt
was supported on glass wool and loosely packed into a
5 cm inlet column. By lowering the temperature of the reac-
tion tube to ꢀ50 ꢀC, the reaction went smoothly by passing
ClNO2 vapor over the solid AgN3 (Eq. (1)), new spectrum
was obtained in comparison to the known PE spectrum of
N2O [11]. While, when the temperature was slowly
increased to ꢀ30 ꢀC, spectrum of pure decomposition
product N2O appeared. Attempts have been made to iso-
late pure nitryl azide by condensing ClNO2 on the surface
of AgN3 at ꢀ195.8 ꢀC (liquid nitrogen), but failed in fierce
explosion when the temperature was increased to ꢀ55 ꢀC.
ClNO2 was synthesized according to the literature [12].
3.1. Calculated structure
There have been many theoretical investigations on the
cyclic and chain isomers of N4O2 [19]. All the results pre-
dicted the trans bent structure (chain isomer in Fig. 1) to
be the most stable, the preference for this trans planar mol-
ecule can be rationalized by a higher degree of p delocaliza-
tion in comparison to the cyclic isomer (Fig. 1) as explained
by Klapo¨tke et al., according to the experimental vibra-
tional frequencies (Raman) and theoretical predictions
(RHF/6-31+G*). They concluded that the cyclic isomer
does not represent a local minimum at correlated level
(RMP2/6-31+G*) [5b].
The geometrical features for two planar isomers are
summarized in Fig. 1 and Table 1, respectively, and corre-
sponding energies are listed in Table 2. As for the opti-
mized stable chain and cyclic isomers, the latter shows
11.5 kcal molꢀ1 less stable at B3LYP/6-311+G(d) level,
including ZPE corrections, and the other two levels of the-
ory (CCSD(T) and MP2) predict similar relative energies
(Table 2). However, previous detailed theoretical calcula-
tions on nitrosyl azide N4O predicted that the cyclic isomer
(O–NONNN) is more stable than the open chain trans iso-
mer by 11.3 kcal molꢀ1 (including ZPE), the reverse in sta-
bility between two isomers can be ascribed to the different
degree of p delocalization in N4O and N4O2 [5b,20]. The
N2–N4 bond length in the chain isomer is calculated to
˚
be 1.491 A, which is very close to that of calculated
˚
˚
˚
trans-ONN3 (1.501 A) [20], O2NNCO (1.451 A) [7a], and
trans-NNCO (1.498 A) [7d], this rather long bond (typical
ꢀ50ꢁC
10ꢀ4torr
˚
values: N–N single bond, 1.447 A in N2H4, N@N double
ClNO2ðgÞ þ AgN3ðsÞ
!
N4O2ðgÞ þ AgClðsÞ
ð1Þ
˚
bond, 1.252 A in N2H2) indicates a significant donor-ac-
ceptor interaction (hyperconjugation) [7a,20], of the lone-
pairs on both oxygen atoms with the unoccupied, anti-
bonding r* orbitals of the N–N bond (LP(O) fi r*
(N–N)). The O–N bond length in this weakly bound mole-
To assign the PE spectrum of nitryl azide, Outer valence
Green’s function (OVGF) [13] calculations with 6-
311+G(d) basis sets based on the B3LYP [14] optimized
geometry were performed for isomers of N4O2. The vertical
ionization energies (Ev) were calculated at the ab initio level
according to Cederbaum’s OVGF method, including the
˚
cule is predicted to be 1.212 and 1.199 A for O1–N2 and
N2–O3, respectively, a little longer than the undisturbed
˚
NO2 (1.197 A) [21].

X. Zeng et al. / Journal of Molecular Structure 840 (2007) 59–65
61
Fig. 1. Optimized geometries (B3LYP/6-311+G(d)) of the isomers and transition structures on the potential-energy surfaces of the decomposition of
˚
N4O2. Bond distances in A.
Table 1
Distances and angles for N4O2 isomers calculated at B3LYP/6-311+G(d) level (refer to Fig. 1)
˚
Molecule
Distances (A)
O1–N2
N2–O3
N2–N4
N4–N5
N5–N6
N5–O1
1.441
N6–O1
Chain
TS1
Cyclic
TS2
1.212
1.282
1.533
1.968
1.599
1.199
1.202
1.188
1.185
1.183
1.125
1.491
1.409
1.277
1.193
1.242
1.185
1.254
1.303
1.368
1.600
1.669
1.124
1.183
1.266
1.224
1.135
1.861
1.326
1.237
TS3
Diss
Molecule
Angles (deg)
O1–N2–O3
O1–N2–N4
N2–N4–N5
N4–N5–N6
N5–N6–O1
Chain
TS1
Cyclic
TS2
129.4
124.4
119.4
113.3
120.9
118.2
115.7
105.8
99.2
109.5
103.5
108.6
113.1
87.6
173.8
126.3
111.6
106.9
137.8
43.5
96.2
111.1
125.3
26.2
TS3
99.5
Diss
180.0
Table 2
for stable isomers, transition states (TS) and decomposi-
tion products 2N2O. Structures were fully optimized and
confirmed to be local minima or transition states on the
potential-energy surface. IRC studies were carried out at
the B3LYP/6-311+G(d) level of theory to connect minima
with the corresponding TS.
Energies (hartrees), ZPE corrections (kcal molꢀ1) and energies relative to
the chain isomer (kcal molꢀ1) for N4O2 isomers, transition states, and
decomposition product, 2N2O
Molecule E(B3LYP)a
ZPEa E(CCSD(T))a,c E(MP2)b,c DEd
Chain
TS1
Cyclic
TS2
TS3
Diss
a
ꢀ369.363518 15.1
0.0
21.7
11.4
18.2
59.3
0.0
29.6
10.3
12.5
59.7
0.0
21.2
11.5
17.3
58.0
ꢀ369.327663 14.6
ꢀ369.340584 15.2
ꢀ369.334312 14.2
ꢀ369.267241 13.8
ꢀ369.434036 13.8
The first decomposition pathway for nitryl azide is:
chain O2N–N3 fi TS1 fi cyclic ONONNN fi TS2 fi
2N2O. Similar to the decomposition of N4O [20], a five-
membered cyclic isomer Cs–ONONNN forms properly in
this process. Based on IRC analysis, the cyclic isomer
formed via a transition state TS1, this can be mechanisti-
cally thought to be driven by a nucleophilic attack of O1
of the nitryl group on the N6 of the azido moiety, and sim-
ilar transition states have been discussed during the decom-
position of ON–NCO, O2N–NCO etc. [7a]. The transition
state TS1 is basically a distorted O1–N2–N4–N5–N6 ring,
ꢀ48.7
ꢀ54.9
ꢀ50.0
At 6-311+G* basis set.
At 6-311+G(3df) basis set.
Based on the B3LYP/6-311+G(d) optimized structure.
Including ZPE corrections.
b
c
d
3.2. Potential-energy surface
˚
To explore both isomerization and decomposition
mechanism, an exhaustive theoretical study was performed
with rather long O–N bond distance of 1.816 A, this
˚
distance resembles more the cyclic isomer O–N, 1.326 A,

62
X. Zeng et al. / Journal of Molecular Structure 840 (2007) 59–65
Fig. 2. Qualitative description of the N4O2 potential-energy surface.
this is in agreement with Hammond’s postulates that the
TS lies more towards the higher energy side [22]. Mean-
while, by vibrational analysis, an imaginary vibration
occurs for TS1 at ꢀ386.6 cmꢀ1, it embodies the concerted
in-plane stretching vibration of bonds O1–N6 and N2–
N4, the vibration shortens the distances between the nitryl
group and the azido moiety, which makes the cyclization
possible. Energetically, to form the cyclic isomer, a barrier
of 21.2 kcal molꢀ1 needs to be surmounted, indicating that
the chain isomer should be intrinsically more stable. The
decomposition of cyclic isomer to experimentally observed
2N2O experiences a transition state TS2 as well, during this
process, the bonds of O1–N2 and N4–N5 become longer to
3.3. Photoelectron spectrum
The PE spectra of the gaseous product obtained by the
reaction (1) at different temperatures are given in Fig. 3. It
can be clearly seen that there are three clear peaks at 12.89,
16.38, and 18.20 eV, respectively, as depicted in Fig. 3b,
which corresponds to the characteristic HeI photoelectron
˚
1.968 and 1.600 A, respectively, an imaginary vibrational
frequency at ꢀ472.1 cmꢀ1 of TS2 helps it further decom-
pose to two moles of thermally stable N2O, with an energy
release of 67.3 kcal molꢀ1. The dissociation barrier for the
cyclic isomer Cs–ONONNN is computed to be only
5.8 kcal molꢀ1 (including ZPE) at B3LYP/6-311+G(d)
level, similar to the low decomposition barrier of ring iso-
mer of N4O [20], a barrier of this size lies much lower than
its ZPE (15.2 kcal molꢀ1), throwing doubt on the existence
of the cyclic isomer of N4O2.
The second dissociation channel is: chain O2N–
N3 fi TS3 fi 2N2O. The barrier for the formation of this
four-membered ring transition state TS3 is calculated to
be 58.0 kcal molꢀ1 at the B3LYP/6-311+G(d) level of the-
ory. According to the previously NBO analysis [5b], the
negative oxygen atom O1 of nitryl group can also take
nucleophilic attack on the positive nitrogen atom N5 of
the azido moiety to form this four-membered ring TS3,
similar four-membered transition states have also been
proposed during the decomposition of ON–NCN [23],
and N4 [24], this assumption is confirmed by IRC analysis
(Fig. 2). However in contrast to the five-membered ring
transition state TS1, the latter pathway is considered more
favorable due to lower decomposition barrier.
Fig. 3. HeI photoelectron spectrum of the gaseous products of the
reaction (1) at different temperatures (a, ꢀ50 ꢀC; b, ꢀ30 ꢀC).

X. Zeng et al. / Journal of Molecular Structure 840 (2007) 59–65
63
Table 3
stretching (masO1N2O3) of the nitro group (Table 4), which
confirms the dominant existence of energetically favorable
chain isomer in the gas phase, as reported by previous the-
oretical predictions [5].
Since it is a planar open-chain molecule, O2N–N3 can be
described quite adequately as a loosely bound combination
of a reorganized O2N and a N3 radical. More specifically,
the molecular orbitals (MOs) of O2N–NNN can be treated
as MOs localized on either the O2N or the N3 fragment of
the neutral molecule. So, in comparison to the known PE
spectra of HN3 [25], SO(N3)2 [9a], CH3NO2 [11], the
assignments of the PE bands in Fig. 3 will be relatively
straightforward.
Experimental (IP in eV) and calculated (Ev in eV) vertical ionization
energies, and molecular orbital characters for the chain isomer of N4O2
IP (eV)
Ev (eV)a
Chainb
MO
Character
Cyclic
11.39
12.52
13.32
11.19 (0.90)
12.12 (0.91)
12.37 (0.88)
12.49 (0.89)
14.80 (0.89)
11.36 (0.90)
12.29 (0.90)
13.53 (0.89)
13.64 (0.89)
14.05 (0.89)
4a00 (22)
3a00 (21)
18a0 (20)
17a0 (19)
16a0 (18)
pnb(N4–N5–N6)
pnb(O1–N2–O3)
nO1, nN4
p0nbðO1–N2–O3Þ
14.52
16.65
a
pb(O1–N2–O3),
pb(N4–N5–N6)
r(O1–N2–O3)
p(N4–N5–N6)
17.66 (0.87)
15.60 (0.85)
15a0 (17)
,
OVGF calculations with 6-311+G(d) basis set.
Polestrength in parentheses.
b
The first band at 11.39 eV (Fig. 3) exhibits a narrow
Frank-Condon profile in agreement with its assignment
as a non-bonding orbital high symmetry molecule, and
the ionization potential is higher than those of HN3
(10.72 eV) [25] and SO(N3)2 (10.18 eV) [9a]. As we know,
the first ionization energies of HN3 (10.72 eV) [25] and
CH3NO2 (11.29 eV) [11] are assigned to the non-bonding
orbitals with pnb(N–N–N) and pnb(O–N–O) characters, respec-
tively. Since the bonding orbitals (pb) in both moieties are
delocalized all over the three atoms, whereas the non-bond-
ing orbitals (pnb) have a nodal plane at the central atom,
perpendicular to the frame of the moiety. Then, the pnb
orbital is primarily localized on the two outer atoms, which
for ONO is the two more electronegative oxygen atoms
leading to a higher ionization potential in comparison to
the orbital embodying pnb(N–N–N) character. So, this band
with high intensity can be easily assigned to the ionization
of the HOMO (4a00) electron on the azido group with pri-
mary pnb(N4–N5–N6) character, as is shown in Fig. 4.
The second broad band at 12.52 eV with apparent vibra-
tion spacing of 1600 60 cmꢀ1 (Fig. 3a) is designated to
the result of ionizing an electron on the second HOMO
(SHMO) 3a00(21) of the N4O2 molecule, not only owing
to the fact that its value is close to computed value of
spectrum of N2O [11]. That is, the gaseous product of the
reaction at ꢀ30 ꢀC leads completely to N2O, derived from
the dissociation of N4O2 (Eq. (1)). This is in agreement
with the previous report by Klapo¨tke et al. that N4O2
decomposes to yield N2O at room temperature [5b]. By
lowering the temperature to ꢀ50 ꢀC, five different bands
occur in the region of 10.0–17.0 eV, which are considered
to be the PE spectrum of another sole gaseous product of
the reaction (1), N4O2. To make assignment of the
observed PE bands to nitryl azide N4O2, OVGF calcula-
tions in combination with molecular orbital analysis based
on the optimized stable chain and cyclic isomers at the
6-311+G(d) basis set were performed. Table 3 presents
experimental (IP in eV) and calculated (Ev in eV) vertical
ionization energies, and corresponding ionized molecular
orbital characters. It can be seen clearly that the experi-
mental results match well with the ionization potentials cal-
culated for both the chain and the cyclic isomers, we could
not exclude the existence of the latter simply from the ion-
ization energies. However, an apparent vibrational spacing
of 1600 60 cmꢀ1 is observed on the band centered at
12.52 eV (Fig. 3), it is assigned to the typical asymmetric
Table 4
Computed vibrational frequencies (cmꢀ1) of the chain and cyclic isomers of N4O2, respectively
Mode
Chain
Cyclic
Calcd.a,b
Exptlc
Assignment
Calcd.a
Assignment
m1
m2
m3
m4
m5
m6
m7
m8
m9
m10
m11
m12
a
2237 (320)
1689 (428)
1320 (211)
1173 (313)
837 (238)
773 (0.8)
736 (11)
530 (6)
512 (16)
440 (49)
189 (0.9)
100 (0.01)
2107 (8)
1556 (2)
1151 (3)
1060 (10)
758 (8)
masN4N5N6
masO1N2O3
msO1N2O3
msN4N5N6
dN4O2
dN4O2
cN4O2
dN4O2
dN4O2
1738 (512)
1378 (6)
1194 (60)
1054 (16)
1033 (40)
980 (10)
809 (24)
687 (0.8)
646 (2)
mO3N2
msN5N6
dN4O2
dsO1N6N5
msN4N5
dN4O2
dN4O2
cN4O2
cN4O2
dN4O2
605 (3)
485 (7)
398 (2)
cN4O2
543 (7)
308 (2)
270 (25)
Calculated at B3LYP/6-311+G(d) level.
Intensities in parentheses.
Raman data taken from Ref. [5b].
b
c

64
X. Zeng et al. / Journal of Molecular Structure 840 (2007) 59–65
Fig. 4. Selected molecular orbitals of N4O2.
12.12 eV, but also the vibration spacing 1600 60 cmꢀ1 on
this band is in well agreement with the calculated asymmet-
ric stretching (masO1N2O3) vibrational frequency
1689 cmꢀ1 (B3LYP/6-311+G(d)) of nitro group, this value
is very close to the experimentally determined 1556 cmꢀ1
(Raman), reported by Klapo¨tke et al. [5b]. With the aid
of molecular orbital analysis (Fig. 4), this band is the ion-
ization result of a p non-bonding electron mainly localized
on the NO2 group, similar to the first ionization process of
CH3NO2 (11.29 eV) [11].
The third band, ranging from 12.90 to 13.70 eV can be
tentatively assigned to the ionization of electrons of both
18a0 and 17a0 orbitals according to the OVGF calculations
and molecular orbital analysis (Fig. 4); the remaining
bands at 14.52 and 16.65 eV should be the results of ioniza-
tion of electrons of deeper orbitals in N4O2, which could
hardly be accurately assigned.
products 2N2O, which precludes the attempt to isolate
the cyclic isomer. Meanwhile, a five-membered ring decom-
position pathway is more favorable, and it is accompanied
by an energy release of 50.0 kcal molꢀ1
.
Acknowledgments
This project was supported by Chinese Academy of Sci-
ences (Hundred talents fund) and the National Natural Sci-
ence Foundation of China (Contract Nos. 20577052,
50372071, 20673123, 20473094, 20503035). Xiaoqing Zeng
thanks Chinese Academy of Sciences for a scholarship dur-
ing the period of this work.
References
[1] (a) N.N. Greenwood, A. Earnshaw, Chemistry of the Elements,
Pergamon, Oxford, 1984;
(b) N. Burford, J. Passmore, J.C.P. Sanders, Mol. Struct. Energy 11
(1989) 53;
4. Conclusion
(c) G. Chaban, M.S. Gordon, K.A. Nguyen, J. Phys. Chem. A 101
(1997) 4283;
In summary, novel unstable compound N4O2 has been
prepared by heterogeneous reaction of ClNO2 with AgN3
at ꢀ50 ꢀC at low pressure (10ꢀ4 torr). The geometric and
electronic structure has been characterized by in situ photo-
electron spectroscopy (PES) and quantum chemical calcu-
lations. It is a planar chain molecule with Cs symmetry in
the gas phase, the first vertical ionization energy is experi-
mentally determined to be 11.39 eV, in well agreement with
the OVGF calculated 11.19 eV, corresponding to the
ionization of the electron on the highest occupied molecu-
lar orbital (HOMO) 4a00, with dominant character of
pnb(N4–N5–N6) on the N3 moiety. A apparent vibrational
spacing of 1600 60 cmꢀ1 (masO1N2O3) on the second
band at 12.52 eV (pnb(O1–N2–O3)) further confirms the pref-
erence of chain structure. At ꢀ30 ꢀC this novel oxide of
nitrogen decomposes completely into N2O. By potential-
energy surface analysis, the cyclic isomer Cs–ONONNN
is less stable than the experimentally observed Cs-nitryl
azide: O2N–N3, the former is just 5.8 kcal molꢀ1 lower in
energy than the transition state leading to decomposition
(d) R.E. Miller, L. Pedersen, J. Chem. Phys. 108 (1998) 436;
(e) G. de Petris, F. Cacace, A. Troiani, Chem. Commun. (2004) 326.
[2] J.A. Miller, C.T. Bowman, Prog. Energy Combust. Sci. 15 (1989) 287.
[3] (a) E.A. Fletcher, Chem. Eng. News. 71 (1993) 2;
(b) T.M. Klapo¨tke, G. Holl, Green Chem. 2 (2000) G75.
[4] (a) H.W. Lucien, J. Am. Chem. Soc. 80 (1958) 4458;
(b) A. Schultz, I.C. Tornieporth-Oetting, T.M. Klapo¨tke, Angew.
Chem. Int. Ed. 30 (1993) 1610;
(c) S. Stinson, Chem. Eng. News. 71 (1993) 51.
[5] (a) M.P. Doyle, J.J. Maciejko, S.C. Busman, J. Am. Chem. Soc. 95
(1973) 952;
(b) T.M. Klapo¨tke, A. Schulz, I.C. Tornieporth-Oetting, Chem. Ber.
127 (1994) 2181.
[6] (a) M.-J. Crawford, T.M. Klapo¨tke, Inorg. Chem. 38 (1999) 3006;
(b) R. Harcourt, F. Wang, T.M. Klapo¨tke, J. Mol. Model. 7 (2001)
271.
[7] (a) T.M. Klapo¨tke, A. Schulz, Inorg. Chem. 35 (1996) 7897;
(b) S.J. Klippenstein, D.L. Yang, T. Yu, S. Kristyan, M.C. Lin, S.H.
Robertson, J. Phys. Chem. A 102 (1998) 6973;
(c) X.Q. Zeng, M.F. Ge, Z. Sun, D.X. Wang, Inorg. Chem. 44 (2005)
9283;
(d) G. de Petris, F. Cacace, R. Cipollini, A. Cartoni, M. Rosi, A.
Troiani, Angew. Chem. Int. Ed. 44 (2005) 462;

X. Zeng et al. / Journal of Molecular Structure 840 (2007) 59–65
65
(e) H. Brand, P. Mayer, A. Schulz, J.J. Weigand, Angew. Chem. Int.
Ed. 44 (2005) 3929.
[17] (a) C. Gonzalez, H.B. Schlegel, J. Chem. Phys. 90 (1989) 2154;
(b) C. Gonzalez, H.B. Schlegel, J. Phys. Chem. 94 (1990) 5523.
[8] (a) K.O. Christe, W.W. Wilson, J.A. Sheehy, J.A. Boatz, Angew.
Chem. Int. Ed. 38 (1999) 2004;
[18] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb,
J.R. Cheeseman, V.G. Zakrzewski, J.A. Montgomery Jr., R.E.
Stratmann, J.C. Burant, S. Dapprich, J.M. Millam, A.D. Daniels,
K.N. Kudin, M.C. Strain, O. Farkas, J. Tomasi, V. Barone, M. Cossi,
R. Cammi, B. Mennucci, C. Pomelli, C. Adamo, S. Clifford, J.
Ochterski, G.A. Petersson, P.Y. Ayala, Q. Cui, K. Morokuma, D.K.
Malick, A.D. Rabuck, K. Raghavachari, J.B. Foresman, J. Cioslow-
ski, J.V. Ortiz, B.B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I.
Komaromi, R. Gomperts, R.L. Martin, D.J. Fox, T. Keith, M.A. Al-
Laham, C.Y. Peng, A. Nanayakkara, C. Gonzalez, M. Challacombe,
P.M.W. Gill, B.G. Johnson, W. Chen, M.W. Wong, J.L. Andres, M.
Head-Gordon, E.S. Replogle, J.A. Pople, Gaussian 98, revision A.5,
Gaussian, Inc., Pittsburgh, PA, 1998.
(b) A. Vij, W.W. Wilson, V. Vij, F.S. Tham, J.A. Sheehy, K.O.
Christe, J. Am. Chem. Soc. 123 (2001) 6308.
[9] (a) X.Q. Zeng, F.Y. Liu, Q. Sun, M.F. Ge, J.P. Zhang, X.C. Ai, L.P.
Meng, S.J. Zheng, D.X. Wang, Inorg. Chem. 43 (2004) 4799;
(b) X.Q. Zeng, M.F. Ge, Z. Sun, D.X. Wang, J. Phys. Chem. A 110
(2006) 5685;
(c) X.Q. Zeng, W.G. Wang, F.Y. Liu, M.F. Ge, Z. Sun, D.X. Wang,
Eur. J. Inorg. Chem. (2006) 416;
(d) X.Q. Zeng, L. Yao, M.F. Ge, D.X. Wang, J. Mol. Struct. 789
(2006) 92.
[10] T.M. Klapo¨tke, B. Krumm, K. Polborn, J. Am. Chem. Soc. 126
(2004) 710.
[19] (a) W.H. Jones, J. Phys. Chem. 96 (1992) 5184;
[11] K. Kimura, S. Katsumata, Y. Achiba, T. Yamazaki, S. Iwata,
Handbook of HeI Photoelectron Spectra of Fundamental Organic
Molecules, Halsted Press, New York, 1981.
(b) M.R. Manaa, C.F. Chabalowski, J. Phys. Chem. 100 (1996) 611;
(c) K.J. Wilson, S.A. Perera, R.J. Bartlett, J. Phys. Chem. A 105
(2001) 7693;
[12] (a) H.H. Batey, H.H. Sisler, J. Am. Chem. Soc. 74 (1952) 3408;
(b) D.C. Frost, S.T. Lee, C.A. McDowell, N.P.C. Westwood, J.
Electron Spectrosc. Relat. Phenom. 7 (1975) 331.
[13] J.V. Ortiz, J. Chem. Phys. 108 (1998) 1008.
[14] (a) A.D. Becke, J. Chem. Phys. 98 (1993) 1372;
(b) C. Lee, W. Yang, R.G. Parr, Phys. Rev. B 37 (1988) 785.
[15] J. Cizek, Adv. Chem. Phys. 14 (1969) 35.
(d) Q.S. Li, X.D. Hong, Mol. Phys. 103 (2005) 249.
[20] J.M. Galbraith, H.F. Schaefer III, J. Am. Chem. Soc. 118 (1996)
4860.
[21] I. Love, J. Phys. Chem. A 110 (2006) 10507.
[22] G.S. Hammond, J. Am. Chem. Soc. 77 (1995) 334.
[23] H.T. Chen, J.J. Ho, J. Phys. Chem. A 109 (2005) 2564.
[24] F. Cacace, Chem. Eur. J. 8 (2002) 3838.
[16] C. Møller, M.S. Plesset, Phys. Rev. 46 (1943) 618.
[25] T. Cvitas, L. Klasinc, J. Chem. Soc. Faraday Trans. 72 (1976) 1240.