Photolysis of 5-azido-3-phenylisoxazole at cryogenic temperature: Formation and direct detection of a nitrosoalkene

Article abstract of DOI:10.3390/molecules25030543

To enhance the versatility of organic azides in organic synthesis, a better understanding of their photochemistry is required. Herein, the photoreactivity of azidoisoxazole 1 was characterized in cryogenic matrices with IR and UV-Vis absorption spectrosco

Full text of DOI:10.3390/molecules25030543

molecules
Article
Photolysis of 5-Azido-3-Phenylisoxazole at Cryogenic
Temperature: Formation and Direct Detection of
a Nitrosoalkene
Upasana Banerjee 1 , William L. Karney 2, Bruce S. Ault 1 and Anna D. Gudmundsdottir 1,
*
1
Department of Chemistry, University of Cincinnati, PO Box 210172, Cincinnati, OH 45221-0172, USA;
upasanabanerjee05@gmail.com (U.B.); bruce.ault@uc.edu (B.S.A.)
Department of Chemistry, University of San Francisco, 2130 Fulton Street, San Francisco, CA 94117, USA;
karney@usfca.edu
2
*
Correspondence: anna.gudmundsdottir@uc.edu
ꢀꢁꢂꢀꢃꢄꢅꢆꢇ
ꢀꢁꢂꢃꢄꢅꢆ
Received: 23 December 2019; Accepted: 24 January 2020; Published: 27 January 2020
Abstract: To enhance the versatility of organic azides in organic synthesis, a better understanding of
their photochemistry is required. Herein, the photoreactivity of azidoisoxazole
in cryogenic matrices with IR and UV-Vis absorption spectroscopy. The irradiation (
azidoisoxazole in an argon matrix at 13 K and in glassy 2-methyltetrahydrofuran (mTHF) at 77 K
yielded nitrosoalkene . Density functional theory (DFT) and complete active space self-consistent
field (CASSCF) calculations were used to aid the characterization of nitrosoalkene and to support
the proposed mechanism for its formation. It is likely that nitrosoalkene is formed from the singlet
excited state of azidoisoxazole via a concerted mechanism or from cleavage of an intermediate
1
was characterized
λ
= 254 nm) of
1
3
3
3
1
singlet nitrene that does not undergo efficient intersystem crossing to its triplet configuration.
Keywords: photoreactivity; azidoisoxazole; cryogenic matrices; nitrosoalkene; nitrenes
1. Introduction
Organic azides are one of the most versatile functional groups in synthesis, as they can be used to
form new C–N bonds, either through electrophilic or nucleophilic substitution [1]. Additionally, their
dipolar character makes them useful for forming new C–N bonds through cycloaddition. For example,
the Huisgen 1,3-dipolar cycloaddition of organic azides to alkynes is commonly used to form triazole
compounds, which have utility in medicinal and pharmaceutical chemistry [2–4]. The pursuit of
sustainable chemistry has redirected efforts toward the use of photochemistry for synthetic applications
because sunlight, photocatalysts, or eco-friendly light-emitting diodes (LEDs) can be used to drive
photoreactions [5]. However, unlike the ground state reactivity of organic azides, their photochemistry
is complex and not well understood because the reactivity is affected by both their structure and the
environment. Better insight into the photoreaction mechanisms of organic azides will undoubtedly aid
in their use in synthetic applications [6].
The photochemistry of vinyl azide derivatives has been studied and their reactivity has been
exploited to synthesize various heterocyclic products [
another type of organic azide, the thermal decomposition behavior has been studied but the
photochemistry has not been reported extensively. Kumar et al. and later L’Abb and Godts studied
the thermal decomposition of 4-azidoisoxazole derivatives and demonstrated that they undergo facile
cleavage to form cyano products, as shown in Scheme 1 [ 10], presumably through a concerted
reaction mechanism. In comparison, the thermal activation of 3-azidoisoxazole derivatives results in
the formation of both isoxazoloisoxazole and hydroxypyrrole products (Scheme 2) [11 12]. L’Abb and
co-workers proposed that these products are formed by different reaction mechanisms, with the
7,8]. However, for azidoisoxazole derivatives,
é
9
,
,
é
Molecules 2020, 25, 543; doi:10.3390/molecules25030543
www.mdpi.com/journal/molecules

Molecules 2020, 25, 543
2 of 15
hydroxypyrroles formed through nitrosoalkene intermediates and the isoxazoloisoxazoles obtained
from nitrene intermediates [12].
Scheme 1. Thermal decomposition of 4-azidoisoxazole derivatives.
Scheme 2. Thermal activation of 3-azidoisoxazole derivatives.
The thermal decomposition of a 5-azidoisoxazole in the presence of 2,3-dimethyl buta-1,3-diene
yields the corresponding oxazine derivative (Scheme 3) [12]. Therefore, it was concluded that the
thermal activation of 5-azidoisoxazole yields the corresponding nitrosoalkene, which can be trapped
with dienes.
Scheme 3. Trapping of the thermolysis product from azidoisoxazole 1.
In this paper, we report the photoreactivity of azidoisoxazole
2-methyltetrahydrofuran (mTHF) matrices. We used IR and UV-Vis absorption spectroscopy to
verify that azidoisoxazole forms nitrosoalkene . The proposed mechanism for the formation
of nitrosoalkene was supported by density functional theory (DFT) and complete active space
1 in cryogenic argon and
1
3
3
self-consistent field (CASSCF) calculations.
2. Results
2.1. Matrix IR Spectroscopy
To determine the photoreactivity of azidoisoxazole
1 at cryogenic temperatures, it was deposited
into an argon matrix at 13 K and irradiated with a 254 nm UV pen. The reactivity was monitored by IR

Molecules 2020, 25, 543
3 of 15
spectroscopy as a function of irradiation time. After 536 s irradiation, the bands due to azidoisoxazole
1
at 2144, 2115, 1611, 1608, 1598, 1581, 1577, 1448, 1415, 1404, 1314, 1240, 1127, 1031, 948, 915, 757, and 746
cm−1 were fully depleted. Concurrently, several new bands were observed at 2218, 1624, 1573, 1527,
1487, 1450, 964, 829, 764, 753, 696, and 668 cm−1 (Figure 1a). The intensity of the new bands increased
with irradiation time, and they formed at same rate as the bands due to azidoisoxazole
We assign these new IR bands to nitrosoalkene based on comparison to its calculated IR spectrum.
The calculated and scaled (0.9613 [13]) IR spectrum of nitrosoalkene has the most significant bands at
2218 (C
N stretch), 1605, 1559, 1518, 1478, 1427, 955, 827, 754, 750, 680, and 667 cm−1 (Figure 1b,c),
which fits with the observed spectrum. It should be noted that the band at 829 cm−1 was considerably
weaker than the other experimental bands, and thus, it is only tentatively assigned to nitrosoalkene
1 were depleted.
3
3
≡
3
.
In contrast, the computed IR spectrum of singlet nitrene 12 does not fit the experimental IR spectra
because the major calculated bands at 1462, 1441, 1409, 1157, 918, 762, 750, 703, and 633 cm−1 (Figure 1b)
could not be reasonably matched with the observed major bands. Similarly, the calculated IR spectrum
of triplet nitrene 32 does not fit well with the experimental spectra, as the major bands at 1495, 1442,
1435, 1386, 1170, 1157, 1147, 1075, 946, 919, 730, 707, 674, 670, and 630 cm−1 (Figure 1b) could not be
easily match with the observed major bands. Thus, we conclude that the irradiation of azidoisoxazole
1 in argon matrices yields nitrosoalkene 3.
Figure 1. Cont.

Molecules 2020, 25, 543
4 of 15
Figure 1.
in an argon matrix; (
B3LYP/6-31+G(d) [14] theory and matrix IR difference spectrum (536–0 s); (
(
a
) Experimental IR spectra as a function of irradiation time (
) IR spectra of nitrenes 32 and 12 and nitrosoalkene
) Comparison of matrix IR
λ
= 254 nm) of azidoisoxazole
1
b
3
calculated using
c
difference spectrum (536–0 s) with calculated scaled IR spectrum of nitrosoalkene 3.
2.2. Glassy Matrix UV-Vis Absorption Spectroscopy
To further characterize nitrosoalkene
3, its UV-Vis absorption was investigated in a glassy mTHF
matrix at 77 K. The absorption spectrum of azidoisoxazole
1
in the glassy mTHF matrix was recorded.
Then, the matrix was irradiated with a 254 nm UV pen and the absorption difference spectra were
obtained at periodic irradiation times (Figure 2a). Upon irradiation, negative absorption was observed
below 300 nm due to the depletion of azidoisoxazole
max at ~360 nm and a tail extending to 470 nm.
1, along with a new broad absorption band with
λ

Molecules 2020, 25, 543
5 of 15
Figure 2.
azidoisoxazole
density function theory (TD-DFT) absorption spectra of nitrosoalkene
CAM-B3LYP/6-31+G(d), and M06-2X/6-31+G(d,p) using tetrahydrofuran (THF) as an implicit solvent
with C-PCM and I-PCM solvation models; ( ) Calculated TD-DFT absorption maxima of nitrosoalkene
above 300 nm using B3LYP/6-31+G(d), CAM-B3LYP/6-31+G(d), and M062X/6-31+G(d,p) in the gas
(a
) Differential UV-Vis absorption spectra as a function of irradiation time (
in glassy 2-methyltetrahydrofuran (mTHF) at 77 K; ( ) Calculated time-dependent
using B3LYP/6-31+G(d),
λ = 254 nm) of
1
b
3
c
3
phase and using THF as an implicit solvent with C-PCM, I-PCM, SMD, and IEFPCM solvation models.
Time-dependent density function theory (TD-DFT) calculations of the absorption spectrum
of nitrosoalkene
3 using the B3LYP/6-31+G(d) level of theory and basis set in the gas phase gave
major electron transitions at 257 (f = 0.2486), 272 (f = 0.0781), 290 (f = 0.0632), 438 (f = 0.0077),

Molecules 2020, 25, 543
6 of 15
and 466 (f = 0.0647) nm. The calculated bands above 300 nm overestimate the near-UV absorption
band of nitrosoalkene by ~100 nm. The major calculated band at 466 nm is due to a mixed
electronic transition of the lone pair on the oxygen and the aromatic -system into an antibonding
-orbital. In an attempt to obtain a better fit between this band and the observed absorption spectra,
3
π
π
we preformed solvation calculations. TD-DFT calculations with tetrahydrofuran (THF) as an implicit
solvent and employing several solvation models, such as Integral Equation Formalism version of the
Polarizable Continuum Model (IEFPCM) [15], Universal Solvation Model based on Solute Electron
Density (SMD) [16], Isodensity Polarizable Continuum Model (I-PCM) [17], and Conductor-like
Polarizable Continuum Model (C-PCM) [18], located the major electric transition above 300 nm at 492
(f = 0.0818), 499 (f = 0.0834), 483 (f = 0.0666), and 493 (f = 0.0846) nm, respectively (Figure 2c). However,
these calculations also overestimated the absorption of nitrosoalkene (Figure 2a). Cramer and Truhlar
3
have explained that dispersion forces are important in estimating short-range dielectric polarization
effects in the solvation model of an electronic spectrum [19]. Furthermore, the B3LYP level of theory
has significant deficiencies when estimating weak but important noncovalent interactions such as van
der Waals forces [20,21].
To better model the absorption of nitrosoalkene
CAMB3LYP [22] level of theory, which has several long-range corrected functionals, with the 6-31+G(d)
basis set. The TD-DFT calculations of nitrosoalkene , in the gas phase and with THF as an implicit
3, we optimized its structure using the
3
solvent, employing IEFPCM [15], SMD [16], I-PCM [17], and C-PCM [18] solvation models, resulted in
the major electronic transition above 300 nm being located at 358 (f = 0.2017), 390 (f = 0.1463), 394 (f =
0.1484), 386 (f = 0.1197), and 390 (f = 0.1508) nm. Although most of these calculations also overestimated
the most significant electronic transition for nitrosoalkene 3, in comparison to the observed spectra,
they yielded significantly better results than those obtained using the B3LYP level of theory (Figure 2b).
Finally, we opted to use the hybrid functional from Zhao and Truhlar [23] and the M062X level
of theory with the 6-31+G(d,p) basis set to optimize nitrosoalkene
spectrum, as these hybrid functionals have shown promising results in predicting dispersion forces.
The TD-DFT/M062X calculated spectrum of nitrosoalkene in the gas phase and with THF as an implicit
3 and calculate its absorption
3
solvent with IEFPCM, SMD, I-PCM, and C-PCM solvation models exhibited major transitions above
300 nm at 356 (f = 0.2017), 369 (f = 0.1463), 372 (f = 0.1484), 362 (f = 0.1197), and 369 (f = 0.1508) nm,
which match well with the observed spectrum of nitrosoalkene 3. Thus, we conclude that irradiation
of azidoisoxazole
1
in mTHF matrices results in the formation of nitrosoalkene
3, as observed upon
irradiation in argon matrices.
Based on the fact that photolysis of azidoisoxazole
1 in cryogenic matrices results in the formation
of nitrosoalkene
, we theorize that azidoisoxazole
excited state (S1) of azidoisoxazole
of azidoisoxazole
3
and because we did not observe the formation of either singlet or triplet nitrene
forms nitrosoalkene in a concerted manner from the singlet
(pathway 1, Scheme 4). However, it is possible that the S1
forms singlet lnitrene 12, which subsequently rearranges into nitrosoalkene
2
1
3
1
1
3
(pathway 2, Scheme 4). Furthermore, if nitrene 12 is formed, we cannot rule out that it intersystem
crosses to its triplet configuration, nitrene 32, which could then rearrange and intersystem cross to
form nitrosoalkene 3 (pathway 3, Scheme 4).

Molecules 2020, 25, 543
7 of 15
Scheme 4. Photoreactivity of azidoisoxazole 1 in cryogenic matrices.
2.3. Quantum Chemical Calculations
To determine the feasibility of the various pathways displayed in Scheme 4, we calculated
the stationary points on the potential energy diagram using DFT calculations (Gaussian 16,
B3LYP/6-31+G(d) [14
minimal energy conformations,
chromophore (Figure 3). The calculated energies of azidoisoxazole 1A and 1B were within 1 kcal/mol of
each other, with being the more stable conformation. TD-DFT calculations were performed to locate
the vertical excitation energies of the first singlet and triplet excited states (S1 and T1) of azidoisoxazole
. The S1 of and the T1 of were located 86 and 64 kcal/mol, respectively, above the ground state
(S0). The open shell triplet and singlet configurations of nitrene were optimized, with the singlet
configuration being optimized using a broken symmetry calculation, as achieved using guess=mix as
a keyword in Gaussian 16 (Figure 3). Triplet nitrene was 11 kcal/mol more stable than the singlet
,
24,
25]). We optimized the structure of azidoisoxazole
1 and obtained two
A
and , which differed mainly in the orientation of the azido
B
A
1
1
1
2
2
configuration; however, it should be noted that due to spin contamination, DFT calculations do not
accurately estimate the energy gap between singlet and triplet nitrenes. Furthermore, we optimized
two minimal conformers of nitrosoalkene, 3A and 3B, with 3A being 2 kcal/more stable than 3B
(Figure 3).
Due to the high spin contamination in the DFT calculations, the singlet and triplet configurations
of nitrene
2 were also computed using the CASSCF and CASPT2 methods (Table 1). At the
CASPT2(14,13)/cc-pVTZ//CASSCF(14,13)/cc-pVDZ level of theory, 12 is predicted to lie 13.8 kcal/mol
above 32.

Molecules 2020, 25, 543
8 of 15
Figure 3. Optimized structures of azidoisoxazole 1 3 using the
, nitrene 12, nitrene 32, and nitrosoalkene
B3LYP/6-31+G(d) level of theory. The numbers in parentheses are the energies in kcal/mol relative to
the energy of conformer A of 1. The calculated spin densities of nitrene 32 are displayed in red.
Table 1. Computed singlet–triplet energy gaps (kcal/mol) for Nitrene 2.
Method
Singlet Energy
CASSCF(14,13)/cc-pVDZ
CASPT2(14,13)/cc-pVDZ
CASPT2(14,13)/ANO-S
CASPT2(14,13)/cc-pVTZ
13.9
15.2
14.3
13.8
Energy of 1A” singlet state relative to 3A” triplet state. All energies were obtained at the CASSCF(14,13)/cc-pVDZ
geometries and are corrected for differences in zero-point energies.
Finally, we calculated stationary points on the singlet and triplet excited state surfaces of
azidoisoxazole
not use DFT to optimize the transition state of S1 of
12 for the formation of nitrosoalkene
1
for the formation of nitrosoalkene
3
1
. However, due to spin contamination, we could
for the formation of nitrene 12 or that of nitrene
. In contrast, we calculated the transition state barrier for nitrene
3
32 to form nitrosoalkene 33, which is located 10 kcal/mol above nitrene 32. The calculated stationary
points on the energy surface of azidoisoxazole 1 are displayed in Figure 4.

Molecules 2020, 25, 543
9 of 15
Figure 4. Calculated stationary points on the singlet and triplet energy surfaces of azidoisoxazole
1
using the B3LYP/6-31+G(d) level of theory. The numbers in parentheses are the energies in kcal/mol.
3. Discussion
Because the energy gap between the S1 and T1 of azidoisoxazole
intersystem crossing between these excited states was not expected to be efficient. Thus, it is reasonable
that the photoreactivity of azidoisoxazole occurred from its singlet excited state. The calculated
energy gap between nitrenes 12 and 32 (13.8 kcal/mol) was slightly less than the singlet–triplet energy
gap measured for phenylnitrene (14.9 kcal/mol) [26 27]. At cryogenic temperatures, irradiation of
phenyl azide resulted in the formation of triplet phenylnitrene via intersystem crossing from singlet
phenylnitrene. Thus, the reactivity of azidoisoxazole at cryogenic temperatures was different from
1
was large (22 kcal/mol),
1
,
1
that of phenyl azide, as a stable triplet nitrene was not formed [28,29].
Based on computational studies of radical stabilization in furanylnitrenes using the spin-flip
coupled-cluster method [30], Wenthold reported that the singlet–triplet energy gap of 2-furanylnitrene is
only 10.9 kcal/mol (Scheme 5). He attributed the decreased singlet–triplet energy gap of 2-furanylnitrene
to radical stabilization of its singlet state resulting from increased delocalization and decreased electron
pair repulsion. In comparison, the calculated singlet–triplet energy gap of 3-furanylnitrene was larger
(17.3 kcal/mol) due to a decrease in delocalization.

Molecules 2020, 25, 543
10 of 15
Scheme 5. Singlet–triplet energy gaps for furanylnitrenes and phenylnitrene.
Additionally, 3-furanylnitrene has been reported to be formed by direct irradiation of
3-furanylisocyanate in cryogenic matrices, presumably because intersystem crossing from the singlet
to the triplet nitrene is efficient (Scheme 6) [31]. In contrast, photolysis of 2-furanylisocyanate gives
cyanoacrolein rather than 2-furanylnitrene because 2-furanylisocyanate reacted in a concerted manner
to form cyanoacrolein or through singlet 2-furanylnitrene, which underwent cleavage faster than it
intersystem crossed to its triplet configuration. Thus, the photoreactivity of azidoisoxazole
that observed for 2-furanylisocyanate.
1 resembled
Scheme 6. Photoreactivity of furanylisocyanates in cryogenic matrices.
Finally, the photoreactivity of azidoisoxazole
1 was compared with that of isoxazoles in cryogenic
matrices. The irradiation of isoxazole derivatives at cryogenic temperatures resulted in the formation
of the corresponding ketenimine. It has been theorized that the ketenimines are formed by intersystem
crossing from triplet vinylnitrenes that are not stable (Scheme 7) [32
however, be detected directly by laser flash photolysis of their isoxazole precursors in solution at
ambient temperature [32 35]. Furthermore, they can be rendered stable at cryogenic temperatures by
–34]. Triplet vinylnitrenes can,
,
incorporating the vinyl bond into a cyclic structure (Scheme 8). The CASPT2 calculated singlet–triplet
energy gaps of cyclic vinylnitrenes, reported to be between 10 and 13 kcal/mol, were similar to or
slightly smaller than those calculated for nitrene 2 [36–39]. Thus, azidoisoxazole 1 reacts differently
from isoxazole derivatives because it does not undergo cleavage of its C–O bond, as its azido moiety is
more reactive.

Molecules 2020, 25, 543
11 of 15
Scheme 7. Photoreactivity of isoxazoles in cryogenic matrices.
Scheme 8. Calculated singlet–triplet energy gaps of cyclic vinylnitrenes.
Because we did not observe nitrene 32 in cryogenic matrices, not even at shorter irradiation times, it
seems likely that the reactivity occurs on the singlet surface of azidoisoxazole 1 in a concerted manner or
that nitrene 12 cleaves more efficiently to form nitrosoalkene 3 than it can intersystem cross to nitrene 32.
4. Materials and Methods
4.1. Preparation of Azidoisoxazole 1
Azidoisoxazole
1 was synthesized by following the previously reported procedure [12].
Sodium hydride (0.041 g, 1.7 mmol) was placed a 25 mL round bottom flask, washed three times with
hexane (2 mL) and once with dichloromethane (2 mL), and then vacuum dried for 1 h. Dry THF (3 mL)
was added to the round bottom flask and the reaction was cooled to
30 ◦C using a xylene–dry ice bath.
−
A solution of 5-amino-3-phenyl isoxazole (0.16 g, 0.99 mmol) and tosyl azide (1.5 mL, 7.5 mmol) in dry
THF (3 mL) was added dropwise to the cooled suspension over 10 min under an argon atmosphere.
The xylene bath temperature was raised to room temperature and the black reaction mixture was
stirred overnight, diluted with water (25 mL), extracted with diethyl ether (25 mL), and dried under
reduced pressure. Azidoisoxazole
1 was purified by silica gel column chromatography eluted with 5%
ethyl acetate in hexane. Recrystallized azidoisoxazole
1 crystals (diethyl ether; 50 mg, 17% yield) were
characterized using 1H-NMR and IR spectroscopy, and the obtained spectra corresponded well with
1
the reported data [12]. H-NMR (400 MHz, CDCl3): 7.77–7.74 (m, 2H), 7.48–7.45 (m, 3H), 6.01 (s, 1H)
ppm; IR (neat): 3129, 3059, 2121, 1598, 1570, 1469, 1368, 1168 cm−1
.
4.2. Matrix IR Spectroscopy
Apparatus description: Matrix isolation of azidoisoxazole
1
in argon matrices was performed on
conventional equipment [40]. This experiment was conducted in a completely stainless-steel home built
vacuum system, with Nupro Teflon-seat valves (Swagelok Company, Colon, Ohio, USA). Pumping was
provided by a Model 1402B Welch vacuum pump (Sargent Wlech Corporation, Buffalo Grove, Illinois,
USA), and a Var
í
an M-2 diffusion pump (Varian Corporation, Palo Alto, California, USA), with liquid
an) were generally recorded at
nitrogen trap. Pressures near 10−7 mm at the gauge (cold cathode, Var
í
the start of an experiment. Cryogenics were supplied by a Model 21 closed cycle refrigerator (CTi,
Inc., Waltham, Massachusetts, USA), operating down to 13 K. Argon was deposited on a CsI window.
The sample was deposited by heating for 3.5 h at 313 K (40 ◦C), followed by 30 min at 315 K (42 ◦C)
under vacuum. The vaporized sample was entrained in flowing argon (2 mmol/h), carried to the
cryogenic surface and deposited.

Molecules 2020, 25, 543
12 of 15
The matrix was irradiated through a KBr window with a UV pen (
λ
= 254 nm). IR spectra at
1 cm−1 resolution were collected periodically as a function of irradiation time with a Perkin Elmer
100 Fourier transform infrared spectrometer (Perkin Elmer Corportation, Waltham, Massachusetts,
USA). We used IgorPro8 software to plot and process the data.
4.3. Glassy Matrix UV-Vis Absorption Spectroscopy
Apparatus description: Glassy matrix UV-Vis spectroscopy was performed using the following
instrumentation setup. The cryostat unit is a commercial Optistat DN2, a Variable Temperature Liquid
Nitrogen Cryostat obtained from Oxford Instruments, Concord, Massachusetts, USA. (Temperature
can be varied between 77 K and 500 K). Mercury iTC cryogenic environment controller was obtained
from the same commercial source and it was used as a temperature controller. Vacuum chambers were
pumped to high vacuum using KJLC RV series Rotary Vane pump to create thermal isolation. We used
JASCO V-750 spectrophotometer (from JASCO Easton, Maryland, USA) for recording UV-Vis spectra.
IgorPro8 software was used for plotting and processing the data.
A stock solution of azidoisoxazole
1 (1.2 mg, 0.0001 mmol) in mTHF (0.5 mL) was prepared.
Subsequently, 0.01 mL of this stock solution was added to mTHF (2 mL) and the resulting solution was
placed in a quartz cuvette. After degassing the solution by bubbling with argon for 3 min, the cuvette
was capped and sealed with Teflon tape. The UV-Vis absorption spectrum of
1 at ambient temperature
showed a broad absorption band up to 300 nm (see Figure S1) and the maximum absorbance of the
solution was below 0.6. This solution was cooled to 77 K using liquid nitrogen to form a glassy mTHF
matrix. The absorption spectrum of
1 before irradiation (0 min) was recorded and then set as the
baseline. Then, the glassy mTHF matrix was irradiated with a 254 nm UV pen and the absorption
difference spectrum was collected periodically.
4.4. Quantum Chemical Calculations
Geometries of all species were optimized at the B3LYP, CAM-B3LYP, and M06-2X levels of theory
with the 6-31+G(d) basis set, as implemented in the Gaussian 16 programs at the Ohio Supercomputer
Center [14,23,24,41]. The absorption spectra were calculated using TD-DFT calculations [42,43] in
the gas phase and with THF as an implicit solvent employing IEFPCM [15], SMD [16], I-PCM [17],
and C-PCM [18] solvation models. The calculated IR spectra of the intermediates and products were
obtained by frequency calculations using the parameters mentioned above, and the frequencies were
scaled by a factor of 0.9613 [13]. The transition states were confirmed to have one imaginary vibrational
frequency by analytical determination of the second derivative of the energy with respect to the internal
coordinates. Intrinsic reaction coordinate (IRC) calculations were used to verify that the transition
states were correlated with the products and the precursors [44,45].
CASSCF and CASPT2 calculations on nitrene
2 employed a 14-electron, 13-orbital active space
consisting of all * MOs plus the in-plane nonbonding MO on the nitrene nitrogen. The CASSCF
π/π
and CASPT2 calculations were performed with Molcas 8.2 [46]. The molecular orbitals (MOs) were
visualized with Molden [47].
5. Conclusions
IR and UV-Vis absorption spectroscopy verified that irradiation of azidoisoxazole
matrices results in the formation of nitrosoalkene . Thus, both thermal and light activation of
azidoisoxazole
yield the same product. Because the formation of nitrene 32 was not observed in
cryogenic matrices, it was concluded that nitrosoalkene is formed in a concerted manner from the
singlet excited state of azidoisoxazole
or from nitrene 12, which does not intersystem cross to its
1 in cryogenic
3
1
3
1
triplet configuration as efficiently as it cleaves to form nitrosoalkene
an efficient precursor of nitrosoalkene 3.
3. Thus, azidoisoxazole 1 is

Molecules 2020, 25, 543
13 of 15
Supplementary Materials: The following are available online at http://www.mdpi.com/1420-3049/25/3/543/s1,
Figure S1: UV-Vis absorption spectra of azidoisoxazole 1. Optimized structures of 1–3; CASSCF and CASPT2
absolute energies of nitrenes 12 and 32.
Author Contributions: A.D.G. guided the research, and U.B. and B.S.A. performed the experiments. U.B. and
W.L.K. carried out the computational study. All authors contributed to writing and proofreading the manuscript
and have read and approved the final manuscript.
Funding: This research was funded by the National Science Foundation (NSF) (grant numbers CHE-1800140 and
CHE-1565793), and the Ohio Supercomputer Center (OSC) (grant number PES-0597).
Acknowledgments: U.B. is grateful for Ann P. Villalobos/Lange and Laws Fellowships from the UC
Chemistry Department.
Conflicts of Interest: The authors declare no conflict of interest.
References
1.
Bräse, S.; Gil, C.; Knepper, K.; Zimmermann, V. Organic azides: An exploding diversity of a unique class of
compounds. Angew. Chem. Int. Ed. 2005, 44, 5188–5240. [CrossRef] [PubMed]
Kacprzak, K.; Skiera, I.; Piasecka, M.; Paryzek, Z. Alkaloids and isoprenoids modification by
copper(I)-catalyzed Huisgen 1,3-dipolar cycloaddition (click chemistry): Toward new functions and molecular
architectures. Chem. Rev. 2016, 116, 5689–5743. [CrossRef] [PubMed]
2.
3.
4.
5.
6.
7.
8.
9.
Meldal, M.; Tornøe, C.W. Cu-catalyzed azide–alkyne cycloaddition. Chem. Rev. 2008, 108, 2952–3015.
[CrossRef] [PubMed]
Mandoli, A. Recent advances in recoverable systems for the copper-catalyzed azide-alkyne cycloaddition
reaction (CuAAC). Molecules 2016, 21, 1174. [CrossRef] [PubMed]
Schultz, D.M.; Yoon, T.P. Solar synthesis: Prospects in visible light photocatalysis. Science 2014, 343, 1239176.
[CrossRef] [PubMed]
Banerjee, U.; Thenna-Hewa, K.; Gudmundsdottir, A.D. Triplet vinylnitrenes. In Patai’s Chemistry of Functional
Group; Rappoport, Z., Ed.; Wiley: Hoboken, NJ, USA, 2019; pp. 1–37.
Fu, J.; Zanoni, G.; Anderson, E.A.; Bi, X. α-Substituted vinyl azides: An emerging functionalized alkene.
Chem. Soc. Rev. 2017, 46, 7208–7228. [CrossRef]
Hayashi, H.; Kaga, A.; Chiba, S. Application of vinyl azides in chemical synthesis: A recent update. J. Org.
Chem. 2017, 82, 11981–11989. [CrossRef]
L’Abb
é, G.; Godts, F. Synthesis and thermal decomposition of 4-azidoisoxazoles. Bull. Soc. Chim. Belg. 1987,
96, 229–236. [CrossRef]
10. Anderson, D.J.; Muchmore, C.R. Rearrangements of 5-azidoisoxazoles. J. Heterocycl. Chem. 1995, 32,
1189–1196. [CrossRef]
11. Kumar, G.; Rajagopalan, K.; Swaminathan, S.; Balasubramanian, K.K. An interesting fragmentation during
the thermolysis of azido isoxazoles. Tetrahedron Lett. 1979, 20, 4685–4686. [CrossRef]
12. L’Abbé, G.; Dyall, L.; Meersman, K.; Dehaen, W. Rates and products from thermolysis of 5-azidoisoxazoles
in solution. J. Chem. Soc. Perkin Trans. 2 1996, 2111–2118. [CrossRef]
13. Foresman, J.B.; Frisch, Æ. Exploring Chemistry with Electronic Structure Methods; Gaussian, Inc.: Pittsburgh,
PA, USA, 1996.
14. Lee, C.; Yang, W.; Parr, R.G. Development of the Colle-Salvetti correlation-energy formula into a functional
of the electron density. Phys. Rev. B 1988, 37, 785–789. [CrossRef] [PubMed]
15. Tomasi, J.; Mennucci, B.; Cancès, E. The IEF version of the PCM solvation method: An overview of a new
method addressed to study molecular solutes at the QM ab initio level. J. Mol. Struct.: THEOCHEM 1999
464, 211–226. [CrossRef]
,
16. Marenich, A.V.; Cramer, C.J.; Truhlar, D.G. Universal solvation model based on solute electron density and
on a continuum model of the solvent defined by the bulk dielectric constant and atomic surface tensions.
J. Phys. Chem. B 2009, 113, 6378–6396. [CrossRef]
17. Foresman, J.B.; Keith, T.A.; Wiberg, K.B.; Snoonian, J.; Frisch, M.J. Solvent effects. 5. Influence of cavity
shape, truncation of electrostatics, and electron correlation on ab initio reaction field calculations. J. Phys.
Chem. 1996, 100, 16098–16104. [CrossRef]

Molecules 2020, 25, 543
14 of 15
18. Takano, Y.; Houk, K.N. Benchmarking the conductor-like polarizable continuum model (CPCM) for aqueous
solvation free energies of neutral and ionic organic molecules. J. Chem. Theory Comput. 2005, 1, 70–77.
[CrossRef]
19. Cramer, C.J.; Truhlar, D.G. Implicit solvation models: Equilibria, structure, spectra, and dynamics. Chem. Rev.
1999, 99, 2161–2200. [CrossRef]
20. Tirado-Rives, J.; Jorgensen, W.L. Performance of B3LYP density functional methods for a large set of organic
molecules. J. Chem. Theory Comput. 2008, 4, 297–306. [CrossRef]
21. Kruse, H.; Goerigk, L.; Grimme, S. Why the standard B3LYP/6-31G* model chemistry should not be used in
DFT calculations of molecular thermochemistry: Understanding and correcting the problem. J. Org. Chem.
2012, 77, 10824–10834. [CrossRef]
22. Yanai, T.; Tew, D.P.; Handy, N.C. A new hybrid exchange–correlation functional using the
Coulomb-attenuating method (CAM-B3LYP). Chem. Phys. Lett. 2004, 393, 51–57. [CrossRef]
23. Zhao, Y.; Truhlar, D.G. The M06 suite of density functionals for main group thermochemistry, thermochemical
kinetics, noncovalent interactions, excited states, and transition elements: Two new functionals and systematic
testing of four M06-class functionals and 12 other functionals. Theor. Chem. Acc. 2008, 120, 215–241.
24. Becke, A.D. Density-functional thermochemistry. III. The role of exact exchange. J. Chem. Phys. 1993, 98,
5648–5652. [CrossRef]
25. Frisch, M.J.; Trucks, G.W.; Schlegel, H.B.; Scuseria, G.E.; Robb, M.A.; Cheeseman, J.R.; Scalmani, G.; Barone, V.;
Petersson, G.A.; Nakatsuji, H.; et al. Gaussian 16, Revision, C.01; Gaussian, Inc.: Wallingford, CT, USA, 2016.
26. Travers, M.J.; Cowles, D.C.; Clifford, E.P.; Ellison, G.B. Photoelectron spectroscopy of the phenylnitrene
anion. J. Am. Chem. Soc. 1992, 114, 8699–8701. [CrossRef]
27. Wijeratne, N.R.; Da Fonte, M.; Ronemus, A.; Wyss, P.J.; Tahmassebi, D.; Wenthold, P.G. Photoelectron
spectroscopy of chloro-substituted phenylnitrene anions. J. Phys. Chem. A 2009, 113, 9467–9473. [CrossRef]
28. Hayes, J.C.; Sheridan, R.S. The IR spectrum of triplet phenylnitrene. On the origin of didehydroazepine in
low temperature matrices. J. Am. Chem. Soc. 1990, 112, 5879–5881. [CrossRef]
29. Leyva, E.; Platz, M.S.; Persy, G.; Wirz, J. Photochemistry of phenyl azide: The role of singlet and triplet
phenylnitrene as transient intermediates. J. Am. Chem. Soc. 1986, 108, 3783–3790. [CrossRef]
30. Wenthold, P.G. Spin-state dependent radical stabilization in nitrenes: The unusually small singlet–triplet
splitting in 2-furanylnitrene. J. Org. Chem. 2012, 77, 208–214. [CrossRef]
31. Feng, R.; Lu, Y.; Deng, G.; Xu, J.; Wu, Z.; Li, H.; Liu, Q.; Kadowaki, N.; Abe, M.; Zeng, X. Magnetically
bistable nitrenes: Matrix isolation of furoylnitrenes in both singlet and triplet states and triplet 3-furylnitrene.
J. Am. Chem. Soc. 2018, 140, 10–13. [CrossRef]
32. Rajam, S.; Jadhav, A.V.; Li, Q.; Sarkar, S.K.; Singh, P.N.D.; Rohr, A.; Pace, T.C.S.; Li, R.; Krause, J.A.; Bohne, C.;
et al. Triplet sensitized photolysis of a vinyl azide: Direct detection of a triplet vinyl azide and nitrene. J. Org.
Chem. 2014, 79, 9325–9334. [CrossRef]
33. Zhang, X.; Sarkar, S.K.; Weragoda, G.K.; Rajam, S.; Ault, B.S.; Gudmundsdottir, A.D. Comparison of the
photochemistry of 3-methyl-2-phenyl-2H-azirine and 2-methyl-3-phenyl-2H-azirine. J. Org. Chem. 2014, 79,
653–663. [CrossRef]
34. Lopes, S.; Nunes, C.M.; Gómez-Zavaglia, A.; Pinho e Melo, T.M.V.D.; Fausto, R. Structure and photochemical
behavior of 3-azido-acrylophenones: A matrix isolation infrared spectroscopy study. Tetrahedron 2011, 67,
7794–7804. [CrossRef]
35. Gamage, D.W.; Li, Q.; Ranaweera, R.A.A.U.; Sarkar, S.K.; Weragoda, G.K.; Carr, P.L.; Gudmundsdottir, A.D.
Vinylnitrene formation from 3,5-diphenyl-isoxazole and 3-benzoyl-2-phenylazirine. J. Org. Chem. 2013, 78,
11349–11356. [CrossRef] [PubMed]
36. Gatlin, D.M.; Karney, W.L.; Abe, M.; Ault, B.S.; Gudmundsdottir, A.D. Formation and reactivity of triplet
vinylnitrenes as a function of ring size. J. Org. Chem. 2019, 84, 9215–9225. [CrossRef] [PubMed]
37. Sarkar, S.K.; Osisioma, O.; Karney, W.L.; Abe, M.; Gudmundsdottir, A.D. Using molecular architecture to
control the reactivity of a triplet vinylnitrene. J. Am. Chem. Soc. 2016, 138, 14905–14914. [CrossRef]
38. Sarkar, S.K.; Sawai, A.; Kanahara, K.; Wentrup, C.; Abe, M.; Gudmundsdottir, A.D. Direct detection of
a triplet vinylnitrene, 1,4-naphthoquinone-2-ylnitrene, in solution and cryogenic matrices. J. Am. Chem. Soc.
2015, 137, 4207–4214. [CrossRef]
39. Osisioma, O.; Chakraborty, M.; Ault, B.S.; Gudmundsdottir, A.D. Wavelength-dependent photochemistry of
2-azidovinylbenzene and 2-phenyl-2H-azirine. J. Mol. Struct. 2018, 1174, 94–101. [CrossRef]

Molecules 2020, 25, 543
15 of 15
40. Ault, B.S. Infrared spectra of argon matrix-isolated alkali halide salt/water complexes. J. Am. Chem. Soc.
1978, 100, 2426–2433. [CrossRef]
41. Stratmann, R.E.; Scuseria, G.E.; Frisch, M.J. An efficient implementation of time-dependent density-functional
theory for the calculation of excitation energies of large molecules. J. Chem. Phys. 1998, 109, 8218–8224.
[CrossRef]
42. Bauernschmitt, R.; Ahlrichs, R. Treatment of electronic excitations within the adiabatic approximation of
time dependent density functional theory. Chem. Phys. Lett. 1996, 256, 454–464. [CrossRef]
43. Foresman, J.B.; Head-Gordon, M.; Pople, J.A.; Frisch, M.J. Toward a systematic molecular orbital theory for
excited states. J. Phys. Chem. 1992, 96, 135–149. [CrossRef]
44. Gonzalez, C.; Schlegel, H.B. An improved algorithm for reaction path following. J. Chem. Phys. 1989, 90,
2154–2161. [CrossRef]
45. Gonzalez, C.; Schlegel, H.B. Reaction path following in mass-weighted internal coordinates. J. Phys. Chem.
1990, 94, 5523–5527. [CrossRef]
46. Aquilante, F.; Autschbach, J.; Carlson, R.K.; Chibotaru, L.F.; Delcey, M.G.; De Vico, L.; Fern
ández Galván, I.;
Ferr , N.; Frutos, L.M.; Gagliardi, L.; et al. MOLCAS 8: New capabilities for multiconfigurational quantum
é
chemical calculations across the periodic table. J. Comput. Chem. 2016, 37, 506–541. [CrossRef] [PubMed]
47. Schaftenaar, G.; Noordik, J.H. Molden: A pre- and post-processing program for molecular and electronic
structures. J. Comput. Aided Mol. Des. 2000, 14, 123–134. [CrossRef] [PubMed]
Sample Availability: Samples of azidoisoxazole 1 is available from the authors.
©
2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).