On the photochemistry of 4-azidoquinoline 1-oxide: Structural elucidation of primary photoproduct

Article abstract of DOI:10.1016/j.molstruc.2011.07.006

The primary product of the photolysis of 4-azidoquinoline 1-oxide (1) is believed to be the azo-dimer (4), consistent with the photochemistry of the pyridine analogue (2). This species has not been unambiguously identified, and previous workers have only isolated various isomers of the azoxy species (5). A combined IR, NMR, LC/MS and computational study confirms the structure 4 as the primary product of photolysis of 1 at 355 nm, suggesting that 5 arises from secondary photochemistry of 4.

Full text of DOI:10.1016/j.molstruc.2011.07.006

Journal of Molecular Structure 1003 (2011) 41–46
Contents lists available at ScienceDirect
Journal of Molecular Structure
journal homepage: www.elsevier.com/locate/molstruc
On the photochemistry of 4-azidoquinoline 1-oxide: Structural elucidation
of primary photoproduct
Larry Sallans a, James S. Poole b,
⇑
a Department of Chemistry, University of Cincinnati, Cincinnati, OH 45221, USA
b Department of Chemistry, Ball State University, Muncie, IN 47306, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
The primary product of the photolysis of 4-azidoquinoline 1-oxide (1) is believed to be the azo-dimer (4),
consistent with the photochemistry of the pyridine analogue (2). This species has not been unambigu-
ously identified, and previous workers have only isolated various isomers of the azoxy species (5). A com-
bined IR, NMR, LC/MS and computational study confirms the structure 4 as the primary product of
photolysis of 1 at 355 nm, suggesting that 5 arises from secondary photochemistry of 4.
Ó 2011 Elsevier B.V. All rights reserved.
Received 10 May 2011
Received in revised form 5 July 2011
Accepted 5 July 2011
Available online 14 July 2011
Keywords:
Triplet nitrenes
Azo-arenes
Density functional theory
1. Introduction
presumably via the triplet nitrene. Sawanishi et al. [7] have de-
scribed the chemistry of 1 in acidic media, which is presumed to
The chemistry of azidoheteroaryl N-oxides such as 4-azidoquin-
oline 1-oxide (1, C9H6N4O) and pyridine 1-oxide (2) has been stud-
ied due to the potential of these compounds to act as anti-tumor
agents. The chemistry of these species appears to largely dictated
by the chemistry of their respective nitrenes. It has been shown
[1,2] that the photolysis of 2 leads to the formation of triplet nit-
rene, which then dimerizes (or reacts with unreacted azide to gen-
erate 4,40-azobis(pyridine 1-oxide) 3. The structure of this
compound was studied spectroscopically by Muniz-Miranda et al.
[3]. Product distribution studies of the thermo- and photo-chemis-
try of 2 yielded complex mixtures [4]. This implies that formation
of secondary photoproducts (and possibly thermochemical prod-
ucts) derived from 3 is possible under the experimental conditions.
proceed via the nitrenium ion.
In terms of the elucidation of the chemistry of 1, structural evi-
dence for the formation of 4 is limited, in part due to its low solu-
bility in most organic solvents, and an apparent susceptibility to
oxidation. Kamiya reported that photolysis of 1 in an inert atmo-
sphere led to formation of a compound with mp. 235–236° (de-
comp.) and IR comparison was carried out against with an
authentic standard of 4 obtained by synthesis (no spectral data
was published). It was also noted that photolysis in oxygen gave
another product, believed to be 4,40-azoxybis(quinoline 1-oxide)
5 (C18H12N4O3); with mp. 235 °C (decomp.). The products of reac-
tion mixtures under various conditions exhibited decomposition
at temperatures as low as 225 °C, and as high as 241 °C. The later
work [6] isolated the azoxy compound 5 (on the basis of elemental
analysis and TLC) with a melting/decomposition range of 278–
280 °C. It is unclear whether the azoxy compound 5 was formed
by oxidation of 2 in situ by adventitious oxygen, or by the intercep-
tion of the triplet nitrene by molecular oxygen, as has been postu-
lated for alkyl and aryl nitrenes [8,9].
N3
N3
O
N
N
N
N
O
O
N
N
O
(1)
(2)
(3)
The thermal and photochemistry of 1 in neutral organic sol-
vents was described by Kamiya [5], and more recently by Shaforost
et al. [6]. Both authors postulated that like its pyridine congener, 1
reacts to generate 4,40-azobis(quinoline 1-oxide) 4 (C18H12N4O2),
O
N
N O
N
N
N O
O
N
N
O
N
⇑
Corresponding author. Tel.: +1 765 285 8071.
E-mail address: jspoole@bsu.edu (J.S. Poole).
(5)
(4)
0022-2860/$ - see front matter Ó 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.molstruc.2011.07.006

42
L. Sallans, J.S. Poole / Journal of Molecular Structure 1003 (2011) 41–46
As part of our continuing work on the photochemistry of azido-
vibrational spectra in the solution phase, geometries were re-opti-
mized and vibration frequencies calculated using a simple reaction
field (polarizable continuum) model for the solvent. In lieu of any
solvent based recommendations regarding correction of vibrational
frequencies, the gas-phase correction factors were used. NMR spec-
heteroarene N-oxides, we sought to confirm the structure of the
primary photoproduct, and to determine whether this species
was susceptible to oxidation. A combined spectroscopic and com-
putational study was carried out to elucidate these structures,
and we report our findings below.
tra were simulated using
a GIAO model [13] at the HF/6-
311+G(2d,p)//B3LYP/6-31G(d) level of theory, and chemical shifts
were calculated relative to tetramethylsilane (TMS).
2. Experimental
4-Azidoquinoline 1-oxide 1 was prepared from the 4-nitro com-
pound (purchased from Aldrich) via literature procedure [6], and
recrystallized from acetone as yellow crystals, that darkened to
red-brown upon exposure to light. NMR analysis suggests that
the material crystallized as a monohydrate. Compound 3 was gen-
erated by reduction of the nitro-precursor according to the method
of Muniz-Miranda et al. [3]. Deuterated dichloromethane and chlo-
roform (99% isotopic purity) were obtained from commercial
sources and used without further purification.
4. Results and discussion
The compound 4 may exist in one of two diastereomeric (E/Z)
forms, with other possible conformational variations within each
of the diastereomers (4a–c, see Fig. 1). Compound 5 shows a similar
degree of conformational variation. In the case of 4, two possible
rotamers of the E-isomer exist (4a, 4b), with 4a favored by
10.8 kJ/mol at the B3LYP/6-31G⁄ level of theory, and a rotational
barrier of approximately 40 kJ/mol. The formation of the Z-isomer
is thermally disfavored, and given the triplet nitrene has ample
time to thermally equilibrate (its half-life at 298 K is on the order
of microseconds), the formation of this species is likely to be minor.
A Boltzmann distribution of rotamers based on the relative energies
suggests that the C2h form 4a constitutes approximately 99% of the
distribution of 4 in vacuo. Similar calculations in dichloromethane
(simulated using a polarizable continuum model) indicate that
the structure 4a is slightly distorted from C2h symmetry, and the
energy gap between 4a and 4b in solution decreases to 9.4 kJ/mol.
In contrast, the Z-isomer of the azoxy compound 5 has three
potential rotamers, of which 5a has the lowest enthalpy (B3LYP/
6-31G⁄). However, the rotamer 5b is only 1.7 kJ/mol higher in
enthalpy at 298 K, with a rotational barrier between the two of
approximately 16 kJ/mol. The similarity in energy (which gives a
nearly 2:1 Boltzmann distribution at 298 K) is consistent with the
observations by Shaforost et al. of multiple isomers of 5 in their
experiments. The rotamer 5b0 and the E-isomer 5c are both ther-
mally disfavored species, by approximately 25 and 55 kJ/mol
respectively. It is useful to note that despite the way they appear
in Fig. 1, neither 5a nor 5b are coplanar: in both cases, the quinoline
ring attached to the unoxidized end of the azoxy group remains
approximately co-planar with the N@NAO moiety, but the other
ring does not remain so: the dihedral angle between the two ring
planes is approximately 145° in 5a, and approximately 30° in 5b.
2.1. Solution photolysis
A dilute solution of 1 in d2-dichloromethane or d-chloroform was
deoxygenated by a thin stream of bubbling argon, and subsequently
irradiated at 355 nm by a Continuum Surelite II Nd-YAG laser (oper-
ating on 3rd harmonic, 10 Hz rep rate, pulse energy 20 mJ/pulse).
The reaction mixture was monitored by 1H NMR (JEOL 400 MHz
Multinuclear FT-NMR) to ensure that a detectable amount of the
primary photoproduct was formed. Complete conversion of 1 was
found to be difficult, as photoproduct begins to precipitate at rela-
tively modest concentrations, and for this reason 13C NMR charac-
terization of the photoproduct was not achieved. Chemical shifts
are reported relative to the peak of the residual monodeuterated sol-
vent (CHDCl2, d = 5.32 ppm, CHCl3, d = 7.26 ppm). Infra-red spectra
were obtained in a NaCl solution cell (0.5 mm path length) mounted
in a Perkin Elmer Spectrum 100 FT-IR. Differential spectra were cal-
culated from the same solution pre- and post-photolysis.
2.2. Solid-state photolysis
A sample of 1 was prepared by compression into a KBr disk, and
irradiated using a medium pressure mercury lamp (max. emission
at 254 nm), with IR spectra obtained over intervals of irradiation
time. Attempts to photolyze the sample with the Nd-YAG laser
were unsuccessful due to shattering of the KBr plate by the laser
pulse. Differential spectra were calculated from the same sample
pre- and post-photolysis.
4.1. NMR analysis
Laser photolysis of dilute deoxygenated solutions of 1 in CDCl3
and CD2Cl2 both yield a single major photoproduct, indicated by
the 1H NMR difference spectra in Fig. 2. The spectra were assigned
to the azo-dimer 4, based on the fact that the signals for protons on
the A ring are clear doublets, suggesting that either the primary
photoproduct (a) contains a single quinoline ring, (b) is symmetric,
or (c) is subject to a high degree of chemical shift equivalence. Par-
tial connectivity assignments were possible by COSY spectroscopy,
but the primary photoproduct has limited solubility in organic sol-
vents (higher in chloroform than dichloromethane), which does
not allow collection of a 13C NMR spectrum of this compound. In-
deed, under the high dilution conditions used in these experi-
ments, limited by the low solubility of the photoproduct, it is
unlikely that relatively minor conformers such as 4b will be use-
fully observable in the 1H NMR spectrum.
2.3. LC/MS
The chloroform solution generated for NMR studies was
subjected to LC/MS analysis on a Thermo-Scientific (Finnigan)
LTQ-FTTM, a hybrid linear ion trap and FT-ICR mass spectrometer.
Chromatographic separation was achieved by a Symmetry C18
column (5
gradient.
l
, 2.1 ꢀ 150 mm) using a 2% formic acid/methanol:water
3. Computational analysis
All MO calculations were performed using the GAUSSIAN 03 pro-
gram suite [10]. Geometry optimizations and vibrational frequency
calculations were performed at the B3LYP/6-31G(d) level of theory
[11], and each optimized structure was ascertained to have zero
imaginary frequencies. Calculated vibrational frequencies were
corrected using the single correction factor recommended by Scott
and Radom [12] for this level of theory. In an effort to simulate
Assignment and confirmation of identity was attempted by the
calculation of GIAO NMR chemical shifts. Direct calculation of pro-
ton chemical shifts by such methods does not provide quantita-
tively accurate data, and so we attempted to identify correlations
between calculated and experimentally observed chemical shifts

L. Sallans, J.S. Poole / Journal of Molecular Structure 1003 (2011) 41–46
43
N N
N
N O
O N
N
N
N O
N
O
N
O
O N
N
4c (C2)
4b (C1)
4a (C2h
)
[+39.5]
Rel. Energya
(kJ/mol)
0.0
(defined)
+10.8
+67.8
O
N
N
N O
O N
5b (C1)
O
N N
+1.7
[+16.2]
O
N
N O
0.0
(defined)
O N
N
N
O
N
O
+24.5
[+30.3]
5c (C1)
5a (C1)
+55.5
O
N
N
N O
O N
5b' (C1)
Fig. 1. Structural forms of 4 and 5, including relative enthalpies at 298 K, estimated at the B3LYP/6-31G⁄ level of theory, with appropriately scaled DHvib,298 enthalpic
corrections.
for structurally related systems. To this end, a series of calculations
were performed on pyridine 1-oxide and quinoline 1-oxide species,
substituted at the 4-position with nitrogenous functional groups at
various levels of oxidation. These included the 4-nitropyridine and
quinoline 1-oxides, azides 1 and 2, and azo-dimer 3, all of which
were obtained commercially, or prepared by literature methods.
The experimental NMR data and correlation plots are included in
the Supplementary data.
CDCl3
The species used for correlation purposes represent a limited
set, but the structural motifs and substitutions provide useful ana-
logs for species 4 and 5. Observed correlations for 1H NMR chemi-
cal shifts are relatively strong, with R2 values of 0.96 and 0.97 in
CD2Cl2 and CDCl3 respectively. Correlations for 13C nuclei are
somewhat weaker, with R2 values of 0.93 and 0.92 in CD2Cl2 and
CDCl3 respectively. The equations of the lines of best fit for 1H
chemical shifts are as follows:
dHexp;CDCl ¼ 0:578dcHalc þ 3:02ðppmÞ
ð1Þ
3
dHexp;CD
¼ 0:587dHcalc þ 2:98ðppmÞ
ð2Þ
2Cl3
With Eqs. (1) and (2) in hand, we may estimate the chemical
shifts for species 4 and 5, and compare against experimental data
(Table 1).
Comparison between the geometrical isomers of the azo com-
pound, 4a and 4c, shows significant differences in A-ring proton
environments, particularly with respect to H3 resonances. The Z-
isomer 4c is calculated to have significant upfield shift in H3 rela-
tive to the E-isomer 4a. The optimized geometry for 4c exhibits C2
symmetry, with the quinoline rings in a non-coplanar configura-
tion. Each H3 proton sits about 2.7 Å above C4 of the opposing ring,
placing it out of the shielding zone arising from ring current effects.
There is a reasonable degree of agreement between the observed
spectra, and that calculated for 4a, with the exception of the H8
CD2Cl2
9.5 9.3 9.1 8.9 8.7 8.5 8.3 8.1 7.9 7.7 7.5 7.3 7.1 6.9 6.7 6.5
δ
Chemical shift (ppm)
Fig. 2. Difference 1H NMR spectra following laser photolysis of 1 (negative signals)
at 355 nm in d2-dichloromethane (bottom) and d-chloroform (top).

44
L. Sallans, J.S. Poole / Journal of Molecular Structure 1003 (2011) 41–46
Table 1
Comparison of calculated and experimental NMR chemical shifts.a
Experimental NMR data for primary photoproduct
Estimated chemical shifts (ppm)
Assgt
d2-dichloromethane
d-chloroform
d
Mult.
J
d
Mult.
J (Hz) 4a
4c
5a
5b
(ppm)
(Hz) (ppm)
CD2Cl2 CDCl3 CD2Cl2 CDCl3 CD2Cl2 CDCl3 CD2Cl2 CDCl3
8.55
7.94
8.67
7.88
7.87
9.06
d
7.0
7.0
NA
8.60
7.92
8.76
7.88
7.87
9.03
d
7.0
7.0
8.50
8.14
8.52
7.81
7.82
8.69
8.45
8.09
8.47
7.77
7.78
8.64
8.19
6.71
8.32
7.81
7.86
8.68
8.14
6.69
8.28
7.77
7.82
8.63
8.46
8.50
8.06
8.90
8.71
8.33
7.81
7.75
7.82
7.78
8.71
8.69
8.42
8.45
8.02
8.85
8.65
8.28
7.77
7.71
7.78
7.74
8.66
8.64
8.50
8.51
8.34
8.91
8.58
8.36
7.83
7.72
7.81
7.77
8.77
8.70
8.45
8.46
8.30
8.85
8.53
8.31
7.79
7.69
7.77
7.74
8.72
8.65
H2
H3
H5
H6
H7
H8
d
d
Mult.
dd
ꢂ7.0,
1.8
Overlapping pair of
distorted signals
Overlapping pair of
distorted signals
Mult.
NA
dd
ꢂ7.5,
2.4
a
Chemical shifts estimated using Eqs. (1) and (2), above.
signal, which differs by approximately 0.4 ppm, although this is
true for all of the potential photoproducts.
(4a)
The alternative azoxy photoproducts 5a and the near degener-
ate rotamer 5b, show similar chemical shifts to 4a in many re-
spects, but there is significant differentiation between pairs of
protons H3 and H30, and H5 and H50 (where Hn0 refers to protons
on the ring not attached to the NAO side of the azoxy group) in
5a. In particular, significant downfield shifts are noted for H30
and to a lesser extent H5, presumably due to the shielding anisot-
ropy of the proximal azoxy group (2.11 and 2.19 Å distance be-
tween H and O respectively). The effect is considerably greater
0
(1)
0.15
0.10
0.05
0.00
for the co-planar H30 (
DdH3 –H3 = 0.85 ppm), than the elevated H5
0
(D
dH5 –H5 = ꢁ0.38 ppm). A similar, but smaller, effect may be ob-
served in the calculated chemical shifts for H3 and H30 in 5b.
CDCl3
4.2. Infra-red spectroscopy
solution
FTIR spectra were obtained following laser photolysis of 1 under
the same conditions as the NMR experiments, and difference spec-
tra, along with a simulated IR spectra for azide 1 and azo com-
pound 4a (calculated at the B3LYP/6-31G⁄ level of theory and
corrected by a factor of 0.9614) are shown in Fig. 3. A similar dif-
ference spectrum obtained from photolysis of a KBr plate sample
of 1 at 254 nm is shown in the same figure.
CD2Cl2
solution
All experiments appear to yield the same photoproduct, and the
observed spectral features are summarized in Table 2.
Vibrational spectroscopy alone does not provide unambiguous
proof of structure: the observed IR spectra are consistent with
the calculated spectra for both 4a and 5a. In the C2h form of 4a,
The vibrations of the N-oxide moieties are strongly coupled, with
a strong coupled asymmetric stretch around 1313 cmꢁ1, and a
symmetric AN@NA stretching frequency around 1440 cmꢁ1 with
zero oscillator strength. In 5a, the symmetry is broken, the NAO
stretching of the quinoline 1-oxide rings becomes uncoupled, with
frequencies at 1317 and 1385 cmꢁ1, and the coupled stretching of
the azoxy NAO with the AN@NA moiety yields a relatively intense
absorbance at 1460 cmꢁ1. This latter band is not apparent in the
difference spectrum.
KBr disk
-0.05
1000
1200
1400
1600
1800
2000
2200
Frequency (cm-1)
Fig. 3. Top: simulated gas phase spectra for 1 and 4a calculated at B3LYP/6-31G⁄
level of theory (frequencies were scaled according to Scott and Radom. Difference
FT-IR spectra following 355 nm photolysis of 1 (negative signals) at 355 nm in
CDCl3, CD2Cl2; and 254 nm photolysis of 1 in a KBr disk. Shaded regions correspond
to regions of significant absorption by the solvent.
4.3. Mass spectrometry
Direct infusion of the chloroform solution obtained from laser
photolysis, diluted with methanol (0.2% formic acid) yielded the
mass spectrum shown in Fig. 4, obtained in the ion-trap (LTQ)
detector. Accurate masses (m/z) and high resolution spectra were
obtained by FT-ICR from the same solution. Each of the major
peaks in the spectrum may be classified as being from one of
two sources: 1 or 4 (Fig. 5). There is evidence of minor amounts
of 5 in the mixture.

L. Sallans, J.S. Poole / Journal of Molecular Structure 1003 (2011) 41–46
45
Table 2
Comparison of calculated and experimental vibrational frequencies.a
Vibrational frequencies (cmꢁ1
)
KBr
CD2Cl2
CDCl3
calc.
KBr
CD2Cl2
CDCl3
calc. 4a
calc. 5a
Azide
2114
1615
1569
1453
1432
1391
1369
1317
1287
1253
1217
1142
1066
1021
2118
2118
2178
1581
1541
1437
1426
1384
1356
1338
1303
Primary photoproduct
1602
1580
1606
1581
1602
1582
1607
1576
1607
1576
1547
1512
1562
1452
1435
1392b
1564
1454
1435
1392
1363
1322
1289
1509
1511
1511
1510
1441
1460
1437, 1431
1422, 1417
1385, 1375
1421
1392
1360
1313
1264
1237
1320
1290
1410
1402b
1402
1299
1260
1240
1297
1261
1243
1233
1296
1260
1244
1232
1317
1288
1256
1238
1214
1181
1163
1146, 1139
1121
1217
1143
1066
1214
1143
1072
1232
1156
1053
1013
1202
1182
1212
1161
1176
1156
1043
1176
1156
1043
1156
1080
1043
1140
1117
1036
1009
1044
1013
a
Calculated vibrational frequencies are scaled by a factor of 0.9614, recommended by Scott and Radom [10].
Potentially obscured by solvent absorption. All values in bold correspond to relatively intense absorptions.
b
159.17
100
95
90
85
80
75
70
65
60
55
50
45
40
35
30
25
20
15
10
5
317.17
142.17
187.17
191.17
301.25
131.25
174.17
209.08
333.17
289.25
0
120 140 160 180 200 220 240 260 280 300 320 340 360 380 400 420 440 460 480 500
m/z
Fig. 4. Low resolution (LTQ) mass spectrum of a sample of 1 photolyzed at 355 nm in d-chloroform (see Figs. 2 and 3). d-Chloroform solution was diluted in methanol (2%
formic acid) prior to infusion. High-resolution accurate masses (obtained by FT-ICR) are given in Fig. 5.
As would be expected, the most labile moieties in the system
are the azido- and N-oxide moieties: loss of nitrogen is the primary
fragmentation pathway for protonated 1, and loss of hydroxyl the
primary fragmentation pathway for protonated 4. The formation of
the species with m/z = 191 may be shown to be associated with the
azide 1. Extracted ion chromatograms from LC/MS studies indicate
that detection of m/z 187 and 191 are correspond to the elution of
the azide. The intermediate species with m/z = 159 (formed by loss
of N2 from protonated 1) may have nitrenium characteristics, and
reaction of nitrenium species with alcohols is known in the litera-
ture [14]. Collision induced dissociation (CID) experiments show
the ion m/z = 189 readily loses N2, and thus the species with m/
z = 159 may be formed by collisions with m/z = 189 in the electro-
spray process. In addition, CID experiments indicate that the ion m/
z = 159 from the CID of m/z = 189, and that observed in the overall
mass spectrum have identical masses and identical fragmentation
patterns. The peak with m/z = 317 may be shown to correspond to
compound 4: extracted ion chromatograms from LC/MS studies
indicate that the species responsible for this ion elutes at a differ-
ent retention time than the azide, precluding alternative species
such as a proton-bound dimer of m/z = 159.
It is worth noting that the same solution generated for NMR
analysis was used for LC/MS analysis, and had been stored at room
temperature, protected from light, for a period of 6 weeks between

46
L. Sallans, J.S. Poole / Journal of Molecular Structure 1003 (2011) 41–46
chemistry of 4a. It should be noted that 4a in this study was gen-
erated by laser photolysis, which in turn has been postulated to
arise from one of two paths: dimerization of triplet nitrenes, or
by reaction of triplet nitrene with the parent azide. Therefore, this
study does not preclude the possibility of oxygen trapping of the
triplet nitrene intermediate (to form azoxy- or nitro-species)
successfully competing with the formation of 4a under dilute con-
ditions. Further study of the kinetics of nitrene reactivity is cur-
rently underway.
Acknowledgements
JSP acknowledges financial support from the ACS Petroleum Re-
search Fund and the Lilly Foundation. Calculations were carried out
on the BSU College of Science and Humanities Beowulf Cluster. The
authors thank Dr. Anna Gudmundsdottir for helpful suggestions
and comments.
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.molstruc.2011.07.006.
Fig. 5. High resolution (FT-ICR) mass/charge ratios and formula assignments for
peaks observed in the infused sample and LC/MS of photolyzed d-chloroform
solution of 1, diluted in methanol (2% formic acid). Fragmentation patterns were
established by collision induced dissociation of individual fragment ions. Figures in
parentheses correspond to relative abundances, as determined by LTQ, rather than
ICR methods.
References
[1] K.N. Crabtree, K.J. Hostetler, T.E. Munsch, P. Neuhaus, P.L. Lahti, W. Sander, J.S.
Poole, J. Org. Chem. 73 (2008) 3441.
[2] K.J. Hostetler, K.N. Crabtree, J.S. Poole, J. Org. Chem. 71 (2006) 9023.
[3] M. Muniz-Miranda, B. Pergolese, G. Sbrana, A. Bigotto, J. Mol. Struct. 339 (2005)
744.
the two experiments, with no significant decomposition or oxida-
tion of 4. Similarly, re-oxygenation of an NMR sample, by bubbling
a thin stream oxygen through the solution for 30 min, and leaving
the sample overnight (again, protecting the solution from light),
did not appreciably change the NMR spectrum (at least in terms
of new resonance signals). These observations suggest that 4a is
not intrinsically unstable, or susceptible to thermal oxidation (at
least over relatively short periods of time). The formation of spe-
cies such as 5a reported by previous workers appears to be due
to secondary photolysis of 4, rather than direct trapping of the nit-
rene intermediate with oxygen, at least at moderate to high con-
centrations of 1. It is possible that at very low concentrations of
1 in oxygenated solvents, oxygen may successfully compete to trap
the nitrene. Unfortunately, the concentrations involved are too low
to allow for characterization of any formed products by methods
such as NMR and IR spectroscopy.
[4] R.A. Abramovitch, B. Bachowska, P. Tomasik, Pol. J. Chem. 58 (1984) 805.
[5] S. Kamiya, Chem. Pharm. Bull. 6 (1962) 471.
[6] K.V. Shaforost, A.V. Ryzhakov, L.L. Rodina, Russ. J. Org. Chem. 44 (2008)
736.
[7] H. Sawanishi, T. Hirai, T. Tcuchiya, Heterocycles 19 (1982) 1043.
[8] P.N.D. Singh, S.M. Mandel, R.M. Robinson, Z. Zhu, R. Franz, B.S. Ault, A.D.
Gudmundsdóttir, J. Org. Chem. 68 (2003) 7951.
[9] (a) T.-Y. Liang, G.B. Schuster, J. Am. Chem. Soc. 109 (1987) 7803;
(b) E.A. Pritchina, N.P. Gritsan, J. Photochem. Photobiol. A 43 (1988) 165;
(c) J. Liu, C.M. Hadad, M.S. Platz, Org. Lett. 7 (2005) 549.
[10] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman,
J.A. Montgomery Jr., T. Vreven, K.N. Kudin, J.C. Burant, J.M. Millam, S.S. Iyengar,
J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega, G.A.
Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa,
M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J.E. Knox,
H.P. Hratchian, J.B. Cross, C. Adamo, J. Jaramillo, R. Gomperts, R.E. Stratmann, O.
Yazyev, A.J. Austin, R. Cammi, C. Pomelli, J.W. Ochterski, P.Y. Ayala, K.
Morokuma, G.A. Voth, P. Salvador, J.J. Dannenberg, V.G. Zakrzewski, S.
Dapprich, A.D. Daniels, M.C. Strain, O. Farkas, D.K. Malick, A.D. Rabuck, K.
Raghavachari, J.B. Foresman, J.V. Ortiz, Q. Cui, A.G. Baboul, S. Clifford, J.
Cioslowski, B.B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R.L.
Martin, D.J. Fox, T. Keith, M.A. Al-Laham, C.Y. Peng, A. Nanayakkara, M.
Challacombe, P.M.W. Gill, B. Johnson, W. Chen, M.W. Wong, C. Gonzalez, J.A.
Pople, Gaussian 03, Revision C.02, Gaussian Inc., Wallingford CT, 2004.
[11] (a) A.D. Becke, Phys. Rev. A 38 (1988) 3098;
5. Conclusions
(b) A.D. Becke, J. Chem. Phys. 98 (1993) 5648;
(c) C. Lee, W. Yang, R.G. Parr, Phys. Rev. B 37 (1988) 785.
[12] A.P. Scott, L. Radom, J. Phys. Chem. 100 (1996) 16502.
[13] K. Wolinski, J.F. Hilton, P. Pulay, J. Am. Chem. Soc. 112 (1990) 8251.
[14] V. Voskresenska, R.M. Wilson, M. Panov, A.N. Tarnovsky, J.A. Krause, S. Vyas,
A.H. Winter, C.M. Hadad, J. Am. Chem. Soc. 131 (2009) 11535.
It has been shown by a combination of computational and spec-
troscopic techniques, that the primary photoproduct of photolysis
of the azidoquinoline 1-oxide 1 is the azo-dimer 4a, which has
been partially characterized herein. The azoxy species identified
by previous workers are most likely to arise from secondary photo-