Intermediates in photolyses of 2,2′-diazidoazobenzene and 2-azidoazobenzene in low-temperature matrices

Article abstract of DOI:10.1016/S1386-1425(97)00231-X

The photolyses of 2,2′-diazidoazobenzene (1) and 2-azidoazobenzene (4) were investigated in argon and 2-methyl-tetrahydrofuran (MTHF) matrices at cryogenic temperature. Nitrene and dinitrene were observed in the photolysis of 1 in an MTHF matrix by electron spin resonance (ESR). The dinitrene in an argon matrix was also observed by ultraviolet-visible (UV-vis) and Fourier transform infrared (FT-IR) spectroscopies. UV absorption bands at 430 and 460 nm and IR peak at 1263 cm^-1 were assigned to the dinitrene. Furthermore, mononitrene as an intermediate of the photolysis of 4 was confirmed by UV-vis and FT-IR spectroscopies in an argon matrix. An IR band at 1225 cm^-1 of mononitrene was observed.

Full text of DOI:10.1016/S1386-1425(97)00231-X

Spectrochimica Acta Part A 54 (1998) 411–418
Intermediates in photolyses of 2,2%-diazidoazobenzene and
2-azidoazobenzene in low-temperature matrices
Tsuguyori Ohana *, Yoshinori Koga
National Institute of Materials and Chemical Research, Higashi 1-1, Tsukuba, Ibaraki 305, Japan
Received 17 July 1997; received in revised form 27 August 1997; accepted 19 September 1997
Abstract
The photolyses of 2,2%-diazidoazobenzene (1) and 2-azidoazobenzene (4) were investigated in argon and 2-methyl-
tetrahydrofuran (MTHF) matrices at cryogenic temperature. Nitrene and dinitrene were observed in the photolysis of
1 in an MTHF matrix by electron spin resonance (ESR). The dinitrene in an argon matrix was also observed by
ultraviolet-visible (UV-vis) and Fourier transform infrared (FT-IR) spectroscopies. UV absorption bands at 430 and
460 nm and IR peak at 1263 cm−1 were assigned to the dinitrene. Furthermore, mononitrene as an intermediate of
the photolysis of 4 was confirmed by UV-vis and FT-IR spectroscopies in an argon matrix. An IR band at 1225 cm−1
of mononitrene was observed. © 1998 Elsevier Science B.V. All rights reserved.
Keywords: Low-temperature matrix; Photolysis; Diazido compounds; Intermediates; Dinitrene
1. Introduction
concentration of intermediates in the solution.
The intermolecular reaction between intermedi-
ates, which were generated efficiently in pulsed-
laser photolysis, led to a polymer in good yield.
These intermediates in the photolysis of the di-
azides were identified as dinitrenes using matrix
isolation technique [11].
In this paper, we report the identification of
intermediates generated in the photolysis of 2,2%-
diazidoazobenzene (1) and 2-azidoazobenzene (4).
Previously, thermal and photochemical decom-
positions of 1 in a solution were reported by
Carboni and Castle [12], and Hall et al. [13].
They showed that decomposition of 1 proceeded
stepwise by a loss of nitrogen to give 5,6H-di-
benzo[b, f ]-1,3a,4,6a-tetraazopentalene (3) via
2-(2-azidophenyl)-2H-benzotriazole (2) (Scheme
Intermediates having two nitreno groups are a
subject of interest in connection with the synthe-
ses of high-spin organic compounds [1–3]. Many
investigations of the intramolecular interaction of
two nitreno groups have been reported on the
basis of electron spin resonance (ESR) spec-
troscopy [4–8].
In the case of the photolysis of aromatic com-
pounds substituted with two azido groups at p,p%-
position, a polymerization was observed in a
solution [9,10]. It was suggested that the degree of
polymerization of the polymers depended on the
* Corresponding Author. Tel.: +81 298 544572; fax: +81
298 544474; e-mail; ohana@ccmail.nimc.go.jp
1386-1425/98/$19.00 © 1998 Elsevier Science B.V. All rights reserved.
PII S1386-1425(97)00231-X

412
T. Ohana, Y. Koga / Spectrochimica Acta Part A 54 (1998) 411–418
1). They suggested that benzotriazole 2 was
formed through a mononitrene generated in the
first step of nitrogen loss from 1. However, a
dinitrene as an intermediate was not observed in
the solution. Therefore, we investigated the pho-
tolysis of the diazides using the matrix-isolation
technique. The dinitrene was identified in low-
temperature matrices by ultraviolet-visible (UV-
vis), Fourier transform infrared (FT-IR), and
ESR measurements. The mononitrene which gen-
erates at the first step of the photolysis of 1 was
also confirmed by comparison with the spectro-
scopic data of the photolysis of 4 in cryogenic
matrices.
was carried out at 20–77 K with monochromatic
light at 365 nm (an USHIO 500 W high-pressure
mercury lamp equipped with a JASCO CT-10
monochromator) in an ESR cavity. During the
photolysis the ESR tube was rotated occasionally
in order to make the photolysis homogeneous. All
spectra were measured in the dark after the pho-
tolysis.
2.3. Photolyses of 1 in a solution
A hexane solution of 1 (ca. 10−4 M) was
irradiated at room temperature in a quartz cell
(optical path length, 1 cm) with monochromatic
light at 365 nm. The photolysis was followed by a
UV-vis spectrophotometer (Shimadzu, UV-300).
2. Experimental section
2.4. Photolysis of 1 and 4 in an MTHF matrix at
77 K
2.1. General data
2,2%-Diazidoazobenzene (1) was prepared from
2,2%-diaminoazobenzene through the reaction of
its diazonium salt and sodium azide by a conven-
tional procedure [14] and purified by recrystalliza-
tion from a hexane-ether solution. 2-Azidoazo-
benzene (4) was prepared by coupling of 2-azi-
doaniline with aniline and purified by recrystal-
lization from a hexane solution [15]. In IR,
UV-vis, and nuclear magnetic resonance (Brucker
AC200) spectra of 1 and 4, the impurity was not
observed.
An MTHF solution of 1 (ca. 10−4 M) was
placed in a quartz cell (optical path length, 1 cm),
and degassed by freez-pump-thaw cycles. The
photolysis of the sample was carried out in liquid
nitrogen with monochromatic light at 365 nm,
and followed by a UV-vis spectrophotometer. The
photolysis of 4 in an MTHF matrix was carried
out in a similar manner to the photolysis of 1 in
an MTHF matrix.
2.5. Photolyses of 1 and 4 in an argon matrix
2-Methyltetrahydrofuran
(MTHF,
Tokyo
Chemical Industry) was purified by distillation
from sodium. Spectroscopic grade n-hexane
(Kanto Chemical) was used without purification.
Experiments in an argon matrix [17] were car-
ried out at 18–20 K using a closed-cycle refrigera-
2.2. ESR measurement at cryogenic temperature
The ESR spectra were measured with a JEOL
JES-RE3X spectrometer equipped with an optical
transmission cavity and Air products LDT-3 liq-
uid helium transfer system. The experiment of
ESR measurement was carried out in a similar
manner to the reference [16].
Diazide 1 and azide 4 were dissolved in MTHF.
The solution (ca. 10−3 M) was placed in an ESR
sample quartz tube (4 mm o.d.) and degassed by
freeze-pump-thaw cycles. Photolysis of samples
Scheme 1.

T. Ohana, Y. Koga / Spectrochimica Acta Part A 54 (1998) 411–418
413
Fig. 1. ESR spectra obtained of the photolysis of 1 and 4 in an
MTHF matrix upon 365 nm light irradiation: (a) the irradia-
tion of 1 for 160 min at 77 K; (b) the irradiation of 4 for 40
min at 20 K.
Fig. 2. Intensity changes of ESR resonance of the photolysis of
1 in an MTHF matrix at 77 K.
tor system (Air Products, Displex CS-202). The
photolyses were followed by UV-vis and FT-IR
spectrophotometers (Shimadzu, FT-IR 8000).
Sapphire and CsI plates were used for the UV-vis
and FT-IR measurements, respectively.
Diazide 1 and azide 4 were sublimed at 338 and
300 K with argon gas, respectively, and deposited
on cold plates. The isolated 1 in an argon matrix
at 18–20 K was photolyzed with a high-pressure
mercury lamp (USHIO 500 W) through a Toshiba
UV-35 cut-off filter (u\350 nm) and a water
filter (path length 20 cm), and the isolated 4 in an
argon matrix was irradiated with monochromatic
light at 365 nm.
275, and 370 mT as shown in Fig. 1(a). In addi-
tion, the measurement of ESR spectra in the
photolysis of 4 at 20 K was also carried out in an
MTHF rigid matrix. The irradiation of a matrix
containing 4 showed a resonance at 665 mT as
shown in Fig. 1(b).
A resonance line at around 330 mT has been
assigned to doublet radicals which are often ob-
served in the photolyses of azides in organic rigid
matrices [18]. The resonance lines at 665 mT, which
were observed in the photolysis of 1 and 4, are
triplet mononitrenes. It is known that the ESR
resonance of triplet mononitrenes appeared around
3. Results and discussion
3.1. The ESR measurement of intermediates from
1 and 4
The formation of intermediates from 1 and 4 in
low-temperature matrices was confirmed by an
ESR spectrometer (Fig. 1). Cryogenic ESR spec-
tra were measured in an MTHF rigid matrix. At
the initial stage (10 min) of photolysis of 1 at 77
K, the ESR spectrum showed a new resonance
line at 665 mT. With subsequent irradiation (40
min), three new resonance lines appeared at 150,
Fig. 3. UV-vis spectra obtained by irradiation (365 nm) of
2-azidoazobenzene (4) in an argon matrix at 18 K: (a) before
irradiation; (b) after irradiation for 16 min; and (c) after
irradiation for 206 min.

414
T. Ohana, Y. Koga / Spectrochimica Acta Part A 54 (1998) 411–418
Fig. 4. IR difference spectra of the photolysis of 4 in an argon matrix at 18 K between before (negative) and after 365 nm light
irradiation: (a) after irradiation for 5 min; (b) after irradiation for 20 min; (c) after irradiation for 100 min. The peaks marked with
“ and ꢀ are assigned to the bands due to 8 and 5, respectively.
670 mT [19,20]. The intensity of resonance line at
665 mT observed in the photolysis of 1 was
maximum at 120 min and decreased with further
irradiation as shown in Fig. 2. And the resonance
line at 150 mT appeared after the photolysis for 40
min. It is indicated that the photolysis of 1 in the
cryogenic matrix proceeds in two steps. The reso-
nance lines at 150, 275, and 370 mT in the photol-
ysis of 1 come from a sole intermediate and are
assigned to triplet dinitrene, 2,2%-dinitrenoazoben-
zene (7). According to the method by Wasserman
et al. [21], the zero filed splitting parameters for
triplet nitrene and dinitrene 7 were estimated to be
D=0.968 and 0.09 cm−1, respectively. Moreover,
it was reported that the D value of 1,4-dini-
trenobenzene (10) is 0.169 [22], and that of 4,4%-dini-
trenoazobenzene (11) is 0.058 cm−1 [23]. Therefore
the D value (0.09 cm−1) was consistent with the
assignment of dinitrene 7, which has intramolecular
interaction between two nitreno groups.
3.2. The photolysis of 4 in an argon matrix
UV-vis spectral changes of the photolysis of 4
in an argon matrix are shown in Fig. 3. Upon
irradiation, new absorption maxima at ca. 370,
385 and 415 nm appeared in a longer wavelength
region than those of azide 4. Upon further irra-
diation, the absorption maxima at 297, 304, 312,
319, and 328 nm of 2-phenyl-2H-benzotriazole
(5) [24] were observed, while absorption maxima
appearing at 370, 385 and 415 nm decreased with
the same behavior. These spectral changes were
Scheme 2.

T. Ohana, Y. Koga / Spectrochimica Acta Part A 54 (1998) 411–418
415
triplet phenylnitrene in an argon and a nitrogen
matrix [26]. They showed major IR bands at 1552,
1524, 1287, 1286, 1148, 1079, 746 and 654 cm−1
of phenylnitrene without the assignment of these
bands. The bands at 1287 and 1286 cm−1 may
correspond to the band at 1225 cm−1 observed in
the photolysis of 4 though the wavenumber shift
of ca. 60 cm−1 was observed. Moreover, Dunkin
and Thomson reported the IR peak at 1284 cm−1
of pentafluorophenyl nitrene [27]. Therefore, the
photolysis of 4 in low-temperature matrix is sum-
marized in Scheme 2.
3.3. The photolysis of 1 in an argon matrix
The photochemical changes of 1 irradiated in
an argon matrix were followed by UV-vis spec-
troscopy, as shown in Fig. 5. The similar spectral
changes were observed in UV-vis spectra in the
MTHF and the argon matrices. In the first step of
the photolysis, UV-vis spectra showed the de-
crease of 1 and the appearance of new absorption
maximum at 440 nm after 10 s (Fig. 5b). In the
next step, maxima at 430 and 460 nm appeared
confining 440 nm band (Fig. 5c, d). Subsequently,
Fig. 5. UV-vis spectra obtained by irradiation (u\350 nm) of
2,2%-diazidoazobenzene (1) in an argon matrix at 18 K: (a)
before irradiation; (b) after irradiation for 10 s; (c) after
irradiation for 1 min; (d) after irradiation for 2 min; (e) after
irradiation for 10 min; and (f) after irradiation for 30 min.
also observed in the photolysis of 4 in an MTHF
matrix. These results indicated that the same in-
termediate was formed in the argon and MTHF
matrices. Therefore the absorption maxima at
370, 385 and 415 nm corresponds with 2-nitrenoa-
zobenzene (8), which showed the resonance line at
665 mT in cryogenic ESR spectrum.
The FT-IR spectral changes of the photolysis of
4 in an argon matrix showed two steps as shown
in Fig. 4. The peaks at 683, 770, and 1225 cm−1
appeared upon irradiation, and then these peak
intensities decreased in the same behavior upon
further irradiation. Finally, the peaks at 1502,
1345, 1293, 964, 815, 759, 747, and 688 cm−1
appeared. These eight peaks were ascribed to
2-phenyl-2H-benzotriazole (5). Therefore, the
peaks at 683, 770, and 1225 cm−1 correspond
with nitrene 8. The peak position of 1225 cm−1
would be ascribed to C–N stretching mode [25].
Hayes and Sheridan reported the IR bands of
Fig. 6. UV-vis spectra of 2,2%-diazidoazobenzene (1) in a
hexane solution at room temperature: (a) before irradiation;
(b) after irradiation for 2 min; and (c) after irradiation for 20
min.

416
T. Ohana, Y. Koga / Spectrochimica Acta Part A 54 (1998) 411–418
Fig. 7. IR spectra of the photolysis of 1 in an argon matrix at 18 K upon the irradiation (u\350 nm): (a) before irradiation; (b)
after irradiation for 1 min; (c) after irradiation for 2 min; (d) after irradiation for 4 min; (e) after irradiation for 9 min; (f) after
irradiation for 19 min; (g) after irradiation for 59 min; (h) after irradiation for 89 min. The peaks marked with ꢁ, ꢂ and ꢃ, are
assigned to the bands due to 6, 7 and 3, respectively.
spectra are shown in Fig. 7. The photolysis of 1
proceeded in three steps as follows; (1) bands at
680, 744, 759, and 1225 cm−1 appeared upon
irradiation (Fig. 7b, c); (2) these peak intensities
decreased and the intensities at 739, 1096, and
1263 cm−1 increased upon further irradiation;
and (3) the bands (1389, 1155, 1103, 812, 748, and
733 cm−1) ascribed to 3 were observed with de-
crease of intensities of other peaks (Fig. 7e–h).
The bands (1501, 1312, 1292, 1276, 1217, 963,
761, and 745 cm−1) ascribed to 2 were not ob-
served during the photolysis of 1 in an argon
matrix. Therefore the reaction path from 6 to 2
would be suppressed in a cryogenic matrix
(Scheme 3).
The intermediate observed in the first step on
the IR spectra can be assigned 2-azido-2%-ni-
trenoazobenzene (6), and the intermediate ob-
served in the second step can be assigned
2,2%-dinitrenoazobenzene (7) on the basis of the
UV-vis spectroscopic analysis. As mentioned
above, it is indicated that the C–N stretching
band of nitrene 8 appeared at 1225 cm−1. There-
fore the IR peak at 1225 cm−1 of nitrene 6 would
new absorption maxima appeared at 252, 381,
and 402 nm, while the maxima at 430 and 460 nm
decreased (Fig. 5e, f). The maxima at 252, 381,
and 402 nm were the band of 3 [28], which also
formed in the photolysis of 1 in the solution (Fig.
6). On the contrary, the maxima at 440, 430, and
460 nm were not observed in the photolysis of 1
in solution at room temperature. The photolysis
of 1 in a solution proceeded via 2, and then 3 was
formed. However, the formation of 2 was not
observed in the photolysis of 1 in cryogenic ma-
trices. The UV-vis absorption bands observed in
the photolysis of 1 in cryogenic matrices were
compared with the results of ESR experiments. In
the case of ESR measurement, it should be noted
that the generation rates of 6 and 7 were very
slow because of low intensity of light leading to
the ESR tube (Figs. 1 and 2). Therefore, the
absorption maximum at 440 nm was assigned to
mononitrene 6, and the maxima at 430 and 460
nm were ascribed to dinitrene 7.
The FT-IR spectral changes on the photolysis
of 1 measured in an argon matrix were similar to
those obtained in the UV-vis spectra. The FT-IR

T. Ohana, Y. Koga / Spectrochimica Acta Part A 54 (1998) 411–418
417
Thus, a small D value indicates that the interac-
tion of two nitreno groups was relatively small.
As mentioned above, in the case of dinitrene 7,
the D value is smaller than that of dinitrene 10. It
is indicated that the double bond character of
C–N bond of 10 would be stronger than that of
7.
The conversion of 7 into 3 proceeded photo-
chemically. As mentioned above, 3 was observed
in the matrix with further irradiation of 7. How-
ever, in the photolysis of 1 in the MTHF matrix,
3 was not observed when the matrix containing 7
was melted to room temperature. In conclusion,
the photolysis of 1 in low-temperature matrices
can be summarized in Scheme 3.
be assigned to the C–N stretching band in the
same manner. On the other hand, the IR peaks at
739, 1096 and 1263 cm−1 were assigned to the
dinitrene 7. The IR peaks at 1263 cm−1 would be
assigned to C–N or CꢀN stretching band of
dinitrene 7. Previously, we reported that the IR
peaks at 1759 and 1779 cm−1 were ascribed to the
CꢀN stretching band of 1,4-dinitrenobenzene (10)
in an argon matrix [11]. These peaks appeared at
higher wavenumber positions compared with the
corresponding band of 7 (1263 cm−1). The reason
for the shift of IR peaks could be explained in
terms of the double bond character of nitrene.
Because of strong interaction of two nitreno
groups in dinitrene 10, a quinonoid structure of
dinitrene 10 was confirmed. Therefore, it is con-
sidered that the double bond character in dini-
trene 7 is relatively week than dinitrene 10. These
assignments are consistent with ESR observation.
We have reported on the relationship between
singlet-triplet energy gaps and the D values for
some quinonoidal dinitrenes [29]. The D values
are related to the magnitude of polarized y spin.
4. Conclusion
The photolyses of 2,2%-diazidoazobenzene 1 and
2-azidoazobenzene 4 were investigated in low-
temperature matrices. In the photolysis of 4 in
low-temperature matrices, mononitrene 8 as an
intermediate was observed in ESR, IR and UV-vis
spectra. The IR band at 1225 cm−1 of C–N
stretching mode of 8 was assigned. In the case of
the photolysis of 1, the generated intermediates of
mononitrene 6 and dinitrene 7 were confirmed.
The photolyses of 1 in low-temperature matrices
proceeded by stepwise loss of nitrogen to give 6
and 7. The IR bands at 1225 and 1265 cm−1 were
ascribed to the C–N stretching band of 6 and 7,
respectively. The peak position of the C–N
stretching mode of 7 and the D value estimated by
ESR spectrum suggested that the interaction of
two nitreno groups of 7 was relatively small com-
pared with 1,4-dinitrenobenzene 10.
References
[1] T. Matsumoto, T. Ishida, N. Koga, H. Iwamura, J. Am.
Chem. Soc. 114 (1992) 9952.
[2] T. Doi, A.S. Ichimura, N. Koga, H. Iwamura, J. Am.
Chem. Soc. 115 (1993) 8928.
[3] C. Ling, M. Minato, P.M. Lathi, H. van Willigen, J.
Am. Chem. Soc. 114 (1992) 9959.
[4] T. Harder, J. Bendig, G. Scholz, R. Stro¨sser, J. Am.
Chem. Soc. 118 (1996) 2497.
Scheme 3.

418
T. Ohana, Y. Koga / Spectrochimica Acta Part A 54 (1998) 411–418
[5] C. Ling, P.M. Lathi, Chem. Lett. (1994) 1357.
[18] O.L. Chapman, R.S. Sheridan, J.P. LeRoux, J. Am.
Chem. Soc. 100 (1978) 6245.
[19] E. Wasserman, Prog. Phys. Org. Chem. 8 (1971) 319.
[20] G. Solinsky, E. Wasserman, W.A. Yager, J. Am. Chem.
Soc. 84 (1962) 3220.
[21] E. Wasserman, R.W. Murray, W.A. Yager, A.M. Troz-
zolo, G. Smolinsky, J. Am. Chem. Soc. 89 (1967) 5079.
[22] S. Nimura, O. Kikuchi, T. Ohana, A.Yabe, M. Kaise,
Chem. Lett. (1996) 125.
[23] B. Singh, J.S. Brinen, J. Am. Chem. Soc. 93 (1971) 540.
[24] J.H. Hall, J. Org. Chem. 33 (1968) 2954.
[25] G. Varsa´nyi, Vibrational Spectra of Benzene Deriva-
tives, Ch. 3, Academic Press, New York, 1964, p. 380.
[26] J.C. Hayes, R.S. Sheridan, J. Am. Chem. Soc. 112 (1990)
5879.
[6] C. Ling, P.M. Lathi, Chem. Lett. (1993) 769.
[7] T. Ohana, M. Kaise, M. Nimura, O. Kikuchi, A. Yabe,
Chem. Lett. (1993) 765.
[8] S. Sasaki, H. Iwamura, Chem. Lett. (1992) 1759.
[9] H.H. Bo¨ssler, R.C. Schulz, Makromol. Chem. 158
(1972) 113.
[10] T. Ohana, M. Shimoyama, H. Niino, A. Yabe, J. Pho-
tochem. Photobiol. A. Chem. 87 (1993) 61.
[11] T. Ohana, A. Ouchi, H. Moriyama, A. Yabe, J. Pho-
tochem. Photobiol. A. Chem. 72 (1993) 83.
[12] R.A. Carboni, J.E. Castle, J. Am. Chem. Soc. 84 (1962)
2453.
[13] J.H. Hall, J.G. Stephanie, D.K. Nordstrom, J. Org.
Chem. 33 (1968) 2951.
[14] R.A. Carboni, J.E. Castle, J. Am. Chem. Soc. 89 (1967)
2618.
[15] J.H. Hall, F.W. Dolan, J. Org. Chem. 43 (1978) 4608.
[16] S. Nimura, O. Kikuchi, T. Ohana, A. Yabe, S. Kondo,
M. Kaise, J. Phys. Chem. A 101 (1997) 2083.
[17] R.J. McMahon, O.L. Chapman, R.A. Hayes, T.C. Hess,
H.P. Krimmer, J. Am. Chem. Soc. 107 (1985) 7597.
[27] I.R. Dunkin, P.C.P. Thomson, J. Chem. Soc. Chem.
Commun. (1982) 1192.
[28] R.A. Carboni, J.C. Kauer, J.E. Castle, H.E. Simmons, J.
Am. Chem. Soc. 89 (1967) 2618.
[29] S. Nimura, O. Kikuchi, T. Ohana, A. Yabe, M. Kaise,
Chem. Lett. (1996) 125.
.