Matrix isolation infrared spectroscopic and theoretical study of the BNNO and AlNNO molecules

Article abstract of DOI:10.1016/j.chemphys.2007.09.024

The BNNO and AlNNO molecules were prepared by the reactions of laser-evaporated boron and aluminum atoms with nitrous oxide molecules in solid argon, and were identified on the basis of isotopically substituted infrared absorptions as well as theoretical calculations. Both molecules were predicted to have a ²A′ ground state with a zigzag geometry. The N₂O subunit in BNNO is highly activated with a quite long N{single bond}N bond length. The AlNNO molecule is an aluminum nitrous oxide complex with a less activated N₂O subunit.

Full text of DOI:10.1016/j.chemphys.2007.09.024

Available online at www.sciencedirect.com
Chemical Physics 342 (2007) 90–94
www.elsevier.com/locate/chemphys
Matrix isolation infrared spectroscopic and theoretical study
of the BNNO and AlNNO molecules
Guanjun Wang, Mingfei Zhou *
Department of Chemistry, Shanghai Key Laboratory of Molecular Catalysts and Innovative Materials, Advanced Materials Laboratory,
Fudan University, Shanghai 200433, People’s Republic of China
Received 11 May 2007; accepted 18 September 2007
Available online 22 September 2007
Abstract
The BNNO and AlNNO molecules were prepared by the reactions of laser-evaporated boron and aluminum atoms with nitrous oxide
molecules in solid argon, and were identified on the basis of isotopically substituted infrared absorptions as well as theoretical calcula-
2
0
tions. Both molecules were predicted to have a A ground state with a zigzag geometry. The N O subunit in BNNO is highly activated
2
with a quite long NAN bond length. The AlNNO molecule is an aluminum nitrous oxide complex with a less activated N
Ó 2007 Elsevier B.V. All rights reserved.
2
O subunit.
Keywords: Nitrous oxide; Boron; Aluminium; Matrix isolation; Infrared absorption; Density functional calculation
1
. Introduction
plexes Ni(NNO) , Pd(NNO) (x = 1, 2) and PtNNO have
x x
been produced and trapped in solid argon [22,23]. These
complexes were characterized to have linear structures with
Nitrous oxide is an important molecule in atmospheric
and industrial chemistry. It is a potent green house gas
and bears responsibility for the depletion of stratospheric
the terminal N atom of N O being bound to the metal
centers.
2
ozone. N O also acts as an important oxidation agent in
chemical industry and the nitrogen and/or oxygen source
of chemical vapor deposition in the manufacture of micro-
In this paper, we report a combined matrix isolation
infrared spectroscopic and theoretical study of the BNNO
and AlNNO molecules formed via the reactions of laser-
2
electronic materials. The reactions of N O with metal cen-
evaporated B and Al atoms with N O in solid argon. We
2
2
ters have received considerable attention [1–17].
will show that the bonding in BNNO and AlNNO is dis-
tinctly different because of the intrinsic differences between
B and Al.
Previous investigations have shown that the reactions of
transition metal atoms with N O lead to the NAO bond
2
activation to form the metal oxide and N . The reactions
2
were predicted to proceed via the initial formation of a
weakly bound complex. Nitrous oxide is a good oxygen
atom source but a poor ligand. Only very few transition
metal complexes containing the nitrous oxide ligand
have been characterized [18–21]. Recently, copper and
silver chloride–nitrous oxide complexes ClCuNNO and
ClAgNNO, and binary transition metal nitrous oxide com-
2. Experimental and computational methods
The experimental setup for pulsed laser evaporation and
matrix isolation infrared spectroscopic investigation has
been described previously [24]. Briefly, the 1064 nm
Nd:YAG laser fundamental (Spectra Physics, DCR 150,
20 Hz repetition rate and 8 ns pulse width) was focused
onto the rotating metal targets through a hole in a CsI win-
dow, which was mounted on a cold tip of a closed-cycle
helium refrigerator (Air Products, Model CSW202). The
*
Corresponding author. Tel./fax: +86 21 6564 3532.
E-mail address: mfzhou@fudan.edu.cn (M. Zhou).
0
301-0104/$ - see front matter Ó 2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.chemphys.2007.09.024

G. Wang, M. Zhou / Chemical Physics 342 (2007) 90–94
91
laser-evaporated boron and aluminum atoms were co-
BO were also observed [28,29]. The experiment was
repeated with the natural abundance boron target. As
shown in Figs. 2 and 3, trace b, additional absorptions
were observed at 1795.7, 1500.3, 836.5 and 626.9 cm with
their IR intensities approximately four times stronger than
deposited with N O in excess argon onto the 12 K CsI win-
2
dow for 1–2 h at a rate of approximately 5 mmol/h. Typi-
cally, 5–10 mJ/pulse laser energy was used. Natural
ꢀ
1
1
0
11
abundance boron ( B: 19.8%, B: 80.2%, Merck), isoto-
1
0
pic-enrich B, and aluminum targets were used in the
experiments. N O and N O (99%, Isotec) were used to
those of the corresponding 1837.0, 1502.3, 838.2 and
1
5
ꢀ1
633.8 cm
absorptions, respectively. This observation
2
2
prepare the N O/Ar mixtures. After sample deposition,
indicates that the new absorber involves one boron atom.
2
1
5
infrared spectra were recorded on a Bruker IFS113v spec-
Similar experiments were performed by using the N O
2
ꢀ
1
14
15
10
trometer at 0.5 cm resolution with a DTGS detector in
and N O + N O samples with the B-enriched target
2 2
ꢀ
1
the spectral range of 4000–400 cm . Matrix samples were
annealed at different temperatures, and selected samples
were subjected to broad band irradiation using a tungsten
lamp or a high-pressure mercury arc lamp with glass filters.
Density functional theory calculations (DFT) were per-
formed using the Gaussian 03 program [25]. The most pop-
ular Becke’s three-parameter hybrid functional, with
additional correlation corrections due to Lee, Yang, and
(Figs. 2 and 3, traces c and d). The 1837.0, 1502.3, 838.2
ꢀ
1
and 633.8 cm
absorptions shifted to 1807.8, 1476.4,
ꢀ1
15
815.4 and 624.7 cm in the experiment with the N O
2
sample. The experiment with equal molar mixture of
1
4
15
N O and N O revealed that the absorptions for the
2
2
new product are the superposition of the absorptions
1
4
15
observed in the experiments using N O and N O sepa-
2
2
rately. Thus, it follows that the new product contains one
N O subunit. Thus, the new absorber has the BN O stoi-
*
Parr was utilized (B3LYP) [26,27]. The 6-311+G basis
2
2
ꢀ
1
set was used for the B, O, Al, and N atoms. Geometries
were fully optimized and vibrational frequencies were cal-
culated with analytical second derivatives, and zero point
vibrational energies were derived.
chiometry. The 1837.0 cm absorption exhibits isotopic
B/ B and N/ N ratios of 1.0230 and 1.0162, respec-
tively, and is due to a BAN stretching mode. The
1
0
11
14
15
ꢀ
1
1502.3 cm absorption shows very small boron isotopic
shift, but quite large N-15 shift. The band position and
1
4
15
3
. Results and discussion
N/ N isotopic ratio of 1.0176 imply that this absorption
ꢀ
1
is due to a terminal NAO stretching mode. The 838.2 cm
absorption exhibits very small boron isotopic shift
(1.7 cm ), but quite large nitrogen isotopic shift
(22.8 cm ). The N/ N isotopic frequency ratio of
1.0276 suggests that this absorption is largely due to a
NAN stretching mode. The 633.8 cm absorption is an
Fig. 1 shows the spectra in selected regions from co-
ꢀ
1
deposition of laser-evaporated boron atoms with 0.5%
N O in argon using an isotopic-enriched B target. Besides
the strong N O absorptions, a group of new absorptions at
1
0
ꢀ1
14
15
2
2
ꢀ
1
ꢀ1
1
837.0, 1502.3, 838.2 and 633.8 cm were observed after
one hour of sample deposition (Fig. 1, trace a). These
new absorptions increased on sample annealing to 25 and
in-plane bending mode. Accordingly, we assign the
1837.0, 1502.3, 838.2 and 633.8 cm absorptions to the
ꢀ1
3
0 K, but decreased upon broad band irradiation with
the output of high-pressure mercury lamp
250 < k < 580 nm). Absorptions due to BOB, BNB and
BAN stretching, NAO stretching, NAN stretching and
a
(
ꢀ
1
Fig. 2. Infrared spectra in the 1850–1780 and 1520–1450 cm regions
from co-deposition of laser-evaporated boron atoms with isotopic-labeled
ꢀ
1
Fig. 1. Infrared spectra in the 1850–1400 and 900–600 cm regions from
co-deposition of laser-evaporated boron-10 atoms with 0.5% N O in
1
0
14
2
N
2
O in excess argon. (a) B + 0.5%
2
N O; (b) natural abundance
1
4
14
15
argon: (a) 1 h of sample deposition at 12 K; (b) after annealing to 25 K; (c)
after annealing to 30 K and (d) after 30 min of broad band irradiation.
B + 0.4%
N
2
O; (c) natural abundance B + 0.4%
N
2
O + 0.4%
2
N O
1
5
and (d) natural abundance B + 0.2%
2
N O.

92
G. Wang, M. Zhou / Chemical Physics 342 (2007) 90–94
ꢀ1
Fig. 5. Infrared spectra in the 1640–1560 and 760–700 cm regions from
co-deposition of laser-evaporated alumimum atoms with isotopic-labeled
ꢀ
1
Fig. 3. Infrared spectra in the 850–800 and 650–610 cm regions from
co-deposition of laser-evaporated boron atoms with isotopic-labeled N
1
4
14
15
N
2
O in excess argon. (a) 0.2%
N
2
O; (b) 0.2%
N
2
O + 0.2%
2
N O, and
2
O
1
5
1
0
14
(
c) 0.2%
2
N O.
in excess argon. (a) B + 0.5%
2
N O; (b) natural abundance B + 0.4%
1
4
14
15
N
2
O; (c) natural abundance B + 0.4%
N
2
O + 0.4%
2
N O and (d)
15
natural abundance B + 0.2%
2
N O.
ꢀ
15
1
ꢀ1
15
1
625.5 cm shifted to 1575.5 cm with the N O sample.
2
1
4
The N/ N isotopic frequency ratio of 1.0323 is very close
1
0
in-plane bending modes of the BNNO molecule, and the
795.7, 1500.3, 836.5 and 626.9 cm absorptions to the
BNNO molecule.
to that of the NAN stretching mode of N O, suggesting
2
ꢀ
1
ꢀ
1
1
that the 1625.5 cm absorption is due to a NAN stretch-
1
1
ꢀ1
ing mode. The 736.4 cm
absorption shifted to
ꢀ
1
15
14
15
Similar experiments were conducted by using the alumi-
7
17.0 cm with N O, corresponding to a N/ N isoto-
2
num metal target. Fig. 4 shows the spectra from co-deposi-
tion of laser-evaporated aluminum atoms with 0.2% N O
pic frequency ratio of 1.0271. This ratio is comparable to
that of the bending mode of N O observed at 588.6 cm
2
ꢀ
1
2
in argon. Besides the strong N O absorptions, new product
2
in solid argon. The spectrum shown in Fig. 5 (trace b) with
ꢀ
1
1
4
15
absorptions at 1625.5 and 736.4 cm were observed after
sample deposition. These two absorptions increased
slightly on annealing, but almost disappeared upon broad
band irradiation using a tungsten lamp with a 500 nm
long-wavelength pass filter. Similar experiments were done
the mixed N O + N O sample confirms the involve-
2
2
ment of one N O subunit. Hence, we assign the 1625.5
2
ꢀ1
and 736.4 cm absorptions to the NAN stretching and
N O bending modes of the AlNNO molecule.
2
To validate the experimental assignment and to obtain
insight into the structure and bonding of BNNO and
AlNNO, we performed density functional theory calcula-
tions. The optimized geometries are shown in Fig. 6. The
1
5
14
15
with the N O and N O + N O samples, and the IR
2
2
2
spectra in selected regions are illustrated in Fig. 5. The
2
0
BNNO molecule was predicted to have a A ground state
with a planar zigzag geometry. Geometry optimization on
the cis-structure converged to the zigzag geometry. This
four-atom nonlinear molecule should have six vibrational
fundamentals. As listed in Table 1, the four experimentally
1
0
observed modes for BNNO were calculated at 1900.9,
1
N
.125 1.185
N
O
N2O 1Σ
N
N
Al
N
B
N
1
.864
1
.258
O
O
ꢀ1
AlNNO C 2A'
Fig. 4. Infrared spectra in the 1650–1600 and 760–700 cm regions from
co-deposition of laser-evaporated aluminum atoms with 0.2% N O in
BNNO Cs 2A'
s
2
argon. (a) 1 h of sample deposition at 12 K; (b) after annealing to 30 K; (c)
after 30 min of irradiation with k > 500 nm.
Fig. 6. Optimized structures (bond lengths in angstrom and bond angles
in degree) of the N O, BNNO and AlNNO molecules.
2

G. Wang, M. Zhou / Chemical Physics 342 (2007) 90–94
93
Table 1
and is mainly B 2 s in character. The population analysis
also shows that the spin is predominantly located on the
terminal B atom. Therefore, doublet BNNO is a univalent
ꢀ
1
Experimentally observed and calculated vibrational frequencies (cm ),
intensities (km/mol, in parentheses) and isotopic frequency ratios of the
1
0
BNNO molecules
Mode Frequency (intensity)
Cal. Obs.
BAN stretch 1900.9 (96) 1837.0 1.0226 1.0230 1.0169 1.0162
NAO stretch 1587.6 (447) 1502.3 1.0012 1.0013 1.0175 1.0176
radical. The Lewis structure can be drawn as
1
0
11
14
15
B/
B
N/
N
Å
B ꢁ NAN@O. The N O subunit in BNNO is highly
2
Cal.
Obs.
Cal.
Obs.
activated.
The observed infrared absorptions suggest that AlNNO
can be regarded as an aluminum–nitrous oxide complex.
The ground state of AlNNO was predicted to have AlAN,
NAN stretch 862.1 (127)
Bending 671.0 (38)
838.2 1.0026 1.0017 1.0297 1.0276
633.8 1.0122 1.0115 1.0151 1.0151
˚
NAN and NAO bond lengths of 1.864, 1.213, and 1.213 A,
1
1
60.2 (7)
45.4 (8)
respectively. The NAN and NAO bonds of N O elongate
2
˚
by 0.088 and 0.028 A upon aluminum coordination. The
population analysis shows that the spin is mainly located
ꢀ
1
1
587.6, 862.1 and 671.0 cm , respectively. These four
on the N O subunit (N: 0.16, N: 0.52, O: 0.26).
2
modes were predicted to have appreciable IR intensities.
The other two modes were predicted to absorb at far infra-
red frequency region with very low IR intensities (Table 1).
The BNNO and AlNNO molecules were formed by the
reactions of B and Al atoms with N O in solid argon
2
0
2
1
2
1
0
11
14
15
B ( P) + N O ( R) ! BNNO ( A ) DE = ꢀ42.4 kcal/mol
2
The calculated B/ B and N/ N isotopic frequency
ratios for the four experimentally observed modes are in
good agreement with the experimental values (Table 1).
ð1Þ
0
2
1
2
Al ( P) + N
2
O ( R) ! AlNNO ( A ) DE = ꢀ24.4 kcal/mol
2
0
The AlNNO molecule was predicted to have a
A
ð2Þ
ground state with a planar zigzag geometry (Fig. 6). Geom-
etry optimization on the cis-structure converged to the zig-
These reactions were predicted to be exothermic by 42.4
and 24.4 kcal/mol, respectively. Both the BNNO and
AlNNO absorptions increased upon sample annealing,
indicating that reactions (1) and (2) require negligible acti-
vation energy. Both the BNNO and AlNNO absorptions
were destroyed upon broad band irradiation. As the
NAN bond in BNNO is highly activated, the formation
of BN and NO is expected. However, no BN and NO
absorptions were observed in the present experiments. Pre-
zag geometry. The NAN stretching and N O bending
2
ꢀ1
modes were computed at 1727.8 and 730.5 cm , respec-
1
4
15
tively, with the calculated N/ N isotopic frequency
ratios in excellent agreement with the observed values
(
Table 2).
Although both the BNNO and AlNNO molecules have
2
0
a A ground state with zigzag geometry, the bonding in
BNNO and AlNNO is distinctly different because of the
intrinsic differences between B and Al. The BAN bond
vious calculations predicted that the reaction
B
2
length in the ground state of BNNO was computed to be
(
[
P) + N O ! BN + NO is endothermic by 49.2 kcal/mol
2
˚
1
.258 A. This value is very close to the terminal BAN bond
32]. Upon irradiation, the BO absorption increased at
˚
length of BNBO (1.254 A) calculated at the same level of
the expense of the BNNO absorptions, suggesting that
the BNNO molecule decomposed to form the BO and N2
upon excitation. No new product absorptions were pro-
duced upon irradiation of AlNNO. The AlNNO complex
theory, which is a formal BAN triple bond [30]. The
˚
NAN bond length was predicted to be 1.363 A, far longer
˚
than that in N O (1.125 A), and even much longer than the
2
˚
NAN double bond length in diazene (1.252 A), but is
most likely decomposed to Al and N O upon visible light
2
shorter than the NAN single bond in hydrazine
irradiation.
2
0
˚
(
H NANH , 1.460 A) [31]. The A ground state BNNO
2
2
8 2 2 2
0
0
0
has an electron configuration of (core) (5a ) (6a ) (7a )
8a ) (1a ) (9a ) (2a ) (10a ) (11a ) (12a ) . The singly
occupied 12a molecular orbital is a nonbonding orbital
0
2
00
2
0
2
00
2
0
2
0
2
0
1
(
4. Conclusions
0
The BNNO and AlNNO molecules were prepared by
the reactions of laser-evaporated boron and aluminum
Table 2
ꢀ
1
atoms with N O in solid argon, and their infrared spectra
2
Observed and calculated vibrational frequencies (cm ), intensities
km/mol, in parentheses) and isotopic frequency ratios of the AlNNO
(
were measured in solid argon. The absorptions at 1837.0,
ꢀ
1
molecule
1502.3, 838.2 and 633.8 cm are assigned to the BAN
stretching, NAO stretching, NAN stretching and in-plane
1
4
15
Mode
Frequency (intensity)
N/
N
1
0
bending modes of the BNNO molecule, and the absorp-
Cal.
Obs.
Cal.
Obs.
ꢀ
1
tions at 1625.5 and 736.4 cm are assigned to the NAN
stretching and NNO bending modes of the AlNNO mole-
cule. Although both molecules were predicted to have a
NAN stretch
NAO stretch
NNO bending
AlAN stetch
1727.8 (977)
1369.0 (19)
730.5 (107)
437.3 (130)
1625.5
1.0323
1.0277
1.0317
736.4
1.0271
2
0
A ground state with a zigzag geometry, the bonding prop-
8
7
8.4 (1)
0.1 (3)
erties are distinctly different because of the intrinsic differ-
ences between B and Al. The BNNO molecule is strongly

94
G. Wang, M. Zhou / Chemical Physics 342 (2007) 90–94
bonded having a formal BAN triple bond with the N O
subunit being highly activated with a quite long NAN
bond length. The AlNNO molecule is an aluminum–
[13] O. Tishchenko, C. Vinckier, M.T. Nguyen, J. Phys. Chem. A 108
2004) 1268.
2
(
[14] O. Tishchenko, C. Vinckier, A. Ceulemans, M.T. Nguyen, J. Phys.
Chem. A 109 (2005) 6099.
[15] V.V. Lavrov, V. Blagojevic, G.K. Koyanagi, G. Orlova, D.K. B o¨ hme,
J. Phys. Chem. A 108 (2004) 5610.
nitrous oxide complex with a less activated N O subunit.
2
Acknowledgements
[16] V. Blagojevic, G. Orlova, D.K. B o¨ hme, J. Am. Chem. Soc. 127 (2005)
3
545.
17] M.D. Michelini, N. Russo, M.E. Alikhani, B. Silvi, J. Comput. Chem.
6 (2005) 1284.
18] J.N. Armor, H. Taube, J. Am. Chem. Soc. 91 (1969) 6874.
[
We gratefully acknowledge financial support from Na-
tional Natural Science Fundation of China (Grant No.
2
[
2
0433080) and National Basic Research Program of China
[19] J.T. Groves, J.S. Roman, J. Am. Chem. Soc. 117 (1995) 5594.
[20] C.B. Pamplin, E.S.F. Ma, N. Safari, S.J. Rettig, B.R. James, J. Am.
Chem. Soc. 123 (2001) 8596.
(
2007CB815203).
[21] F. Paulat, T. Kuschel, C. Nather, V.K.K. Praneeth, O. Sander, N.
Lehnert, Inorg. Chem. 43 (2004) 6979.
References
[
22] G.J. Wang, X. Jin, M.H. Chen, M.F. Zhou, Chem. Phys. Lett. 420
[
[
1] D. Ritter, J.C. Weisshaar, J. Phys. Chem. 94 (1990) 4907.
2] J.M.C. Plane, R.J. Rollason, J. Chem. Soc. Faraday Trans. 92 (1996)
4
(
2006) 130.
[
[
23] X. Jin, G.J. Wang, M.F. Zhou, J. Phys. Chem. A 110 (2006) 8017.
24] M.F. Zhou, L. Andrews, C.W. Bauschlicher Jr., Chem. Rev. 101
371.
3] D.E. Clemmer, K. Honma, I. Koyano, J. Phys. Chem. 97 (1993)
1480.
[
(
2000) 1931.
1
[25] Gaussian 03, Revision B.05, M.J. Frisch, et al. Gaussian, Inc.,
[
[
[
[
4] R. Matsui, K. Senba, K. Honma, J. Phys. Chem. A 101 (1997) 179.
5] M.L. Campbell, R.E. McClean, J. Phys. Chem. 97 (1993) 942.
6] M.L. Campbell, J. Chem. Phys. 104 (1996) 7515.
Pittsburgh PA.
[
[
[
26] A.D. Becke, J. Chem. Phys. 98 (1993) 5648.
27] C. Lee, W. Yang, R.G. Parr, Phys. Rev. B 37 (1988) 785.
28] L. Andrews, P. Hassanzadeh, T.R. Burkholder, J.M.L. Martin, J.
Chem. Phys. 98 (1993) 922.
29] T.R. Burkholder, L. Andrews, J. Chem. Phys. 95 (1991) 8697.
30] M.F. Zhou, N. Tsumori, Q. Xu, G.P. Kushto, L. Andrews, J. Am.
Chem. Soc. 125 (2003) 1371.
31] B.A. MacKay, M.D. Fryzuk, Chem. Rev. 104 (2004) 385.
32] Z. Wang, S.F. Li, L.M. Zhang, Z.Y. Sheng, S.Q. Yu, Propell. Explos.
Pyrotech. 29 (2004) 160.
7] M.L. Campbell, E.J. Kolsch, K.L. Hooper, J. Phys. Chem. A 104
(2000) 11147.
[
[
8] M.L. Campbell, J. Phys. Chem. A 107 (2003) 3048.
9] P.B. Armentrout, L.F. Halle, J.L. Beauchamp, J. Chem. Phys. 76
[
[
(1982) 2449.
[
[
10] P.M. Futerko, A. Fontijn, J. Chem. Phys. 95 (1991) 8065.
11] A. Delabie, C. Vinckier, M. Flock, K. Pierloot, J. Phys. Chem. A 105
[
[
(2001) 5479.
[
12] A. Stirling, J. Am. Chem. Soc. 124 (2002) 4058.