Photochemistry of 5-allyloxy-tetrazoles: Steady-state and laser flash photolysis study

Article abstract of DOI:10.1039/b718104c

The photochemistry of three 5-allyloxy-tetrazoles, in methanol, acetonitrile and cyclohexane was studied by product analysis and laser flash photolysis. The exclusive primary photochemical process identified was molecular nitrogen elimination, with formation of 1,3-oxazines. These compounds were isolated in reasonable yields by column chromatography on silica gel and were fully characterized. DFT(B3LYP)/6-31G(d,p) calculations predict that these 1,3-oxazines can adopt two tautomeric forms (i) with the NH group acting as a bridge connecting the oxazine and phenyl rings and (ii) with the -N bridge and the proton shifted to the oxazine ring. Both tautomeric forms are relevant in the photolysis of oxazines in solution. Secondary reactions were observed, leading to the production of phenyl vinyl-hydrazines, enamines, aniline and phenyl-isocyanate. Transient absorption, detected by laser flash photolysis, is attributed to the formation of triplet 1,3-biradicals generated from the excited 5-allyloxy-tetrazoles. The 1,3-biradicals are converted to 1,6-biradicals by proton transfer, which, after intersystem crossing, decay to generate the products. Solvent effects on the photoproduct distribution and rate of decomposition are negligible.

Full text of DOI:10.1039/b718104c

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Photochemistry of 5-allyloxy-tetrazoles: steady-state and laser flash
photolysis study†
L. M. T. Frija,a,b I. V. Khmelinskii,c C. Serpa,b I. D. Reva,b R. Faustob and M. L. S. Cristiano*a
Received 22nd November 2007, Accepted 24th January 2008
First published as an Advance Article on the web 14th February 2008
DOI: 10.1039/b718104c
The photochemistry of three 5-allyloxy-tetrazoles, in methanol, acetonitrile and cyclohexane was
studied by product analysis and laser flash photolysis. The exclusive primary photochemical process
identified was molecular nitrogen elimination, with formation of 1,3-oxazines. These compounds were
isolated in reasonable yields by column chromatography on silica gel and were fully characterized.
DFT(B3LYP)/6-31G(d,p) calculations predict that these 1,3-oxazines can adopt two tautomeric forms
=
(i) with the NH group acting as a bridge connecting the oxazine and phenyl rings and (ii) with the –N
bridge and the proton shifted to the oxazine ring. Both tautomeric forms are relevant in the photolysis
of oxazines in solution. Secondary reactions were observed, leading to the production of phenyl
vinyl-hydrazines, enamines, aniline and phenyl-isocyanate. Transient absorption, detected by laser flash
photolysis, is attributed to the formation of triplet 1,3-biradicals generated from the excited
5-allyloxy-tetrazoles. The 1,3-biradicals are converted to 1,6-biradicals by proton transfer, which, after
intersystem crossing, decay to generate the products. Solvent effects on the photoproduct distribution
and rate of decomposition are negligible.
family of heterocycles. However, the nature of the ring substituents
Introduction
strongly determines the precise nature and relative amounts of the
final photoproducts, making this type of compound an ongoing
scientific challenge.12–14
Tetrazoles have received much attention due to their important
industrial and biological practical applications.1 Most of these
applications are related to the acid–base properties of the tetrazolic
ring. In fact, the tetrazolic acid fragment, –CN4H, has similar
acidity to the carboxylic acid group, –CO2H, and is almost isosteric
with it, being metabolically more stable at physiologic pH.2 This
metabolic stability of the tetrazole ring strongly influences applica-
tions of its derivatives in medicine and, then, this heterocycle comes
out in the structure of many highly efficient drugs.3–9 Tetrazoles
are also used as stabilizers in photography, gas-generating agents
in airbags in the automobile industry, and herbicides, fungicides
and plant growth regulators in agriculture.10,11
In previous contributions we have described the photochemistry
of some representative tetrazolyl compounds isolated in low-
temperature matrices.15–20 Recently we have also studied the
photochemistry of a series of 4-allyl-tetrazolones (1, Scheme 1)
in solution,21 and proved that photolysis in alcoholic solutions
results in exclusive formation of 3,4-dihydro-6-substituted-3-
phenylpyrimidin-2(1H)-ones (2, Scheme 1), which are versatile
intermediates in heterocyclic synthesis.22
In view of the widespread importance of many tetrazolyl deriva-
tives, we have held a continued interest in structure–reactivity
relationships of this class of compounds. Recently, photochemical
reactions of tetrazole-based compounds have been under scrutiny
in our group.
The photochemistry of tetrazole and some substituted tetrazoles
has been addressed elsewhere. According to these investigations,
tetrazole-ring cleavages leading either to azides or aziridines
represent the main characteristic photochemical reactions of this
Scheme 1 Photochemical conversion of 4-allyl-tetrazolones 1 into 3,4-di-
hydro-6-substituted-3-phenylpyrimidin-2(1H)-ones 2.
Mechanistically, we postulated that these reactions presumably
take place via formation of a triplet biradical intermediate 1(i).
Photoexcitation of 4-allyl-tetrazolones promoted by UV radiation
leads to elimination of N2 from the tetrazole ring, yielding the
biradical 1(i). In a second step, this intermediate rapidly undergoes
a 1,2-migration of hydrogen and closes the ring to form the
pyrimidinone 2, in an exothermic process.
The participation of the reactive triplet intermediate 1(i) could
not be unequivocally proved. However, this mechanism is sup-
ported by the effect of solvent viscosity on the photolysis quantum
yields, and the sensitizing effect of the dissolved oxygen upon the
photodegradation of tetrazolone 1, interpreted as a consequence
of the T → S conversion of triplet biradicals 1(i) opening the way
aDepartment of Chemistry, Biochemistry and Pharmacy, F.C.T. and CC-
MAR, University of Algarve, 8005-039 Faro, Portugal. E-mail: mcristi@
ualg.pt
bDepartment of Chemistry, F.C.T., University of Coimbra, R. Larga, P-
3004-535 Coimbra, Portugal
cDepartment of Chemistry, Biochemistry and Pharmacy, F.C.T and CIQA,
University of Algarve, 8005-039 Faro, Portugal
† Electronic supplementary information (ESI) available: 1H NMR for
compounds 4a–c, geometries and energies of compounds 4a–c, 5a–c, 6,
7a–c, 8, 9, 10 and 11a–c, and Cartesian coordinates of compounds 4a–c,
5a–c, 6, 7a–c, 8, 9, 10 and 11a–c. See DOI: 10.1039/b718104c
1046 | Org. Biomol. Chem., 2008, 6, 1046–1055
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
to the formation of the product 2. Both effects were explained by
the involvement of a caged triplet radical pair.21
Dunkin et al. described the photochemistry of 1,4-dihydro-
5H-tetrazole derivatives isolated in low-temperature matrices.23
In that work, the authors also interpreted the results obtained
upon matrix photolysis on the basis of a biradical intermediate,
invoking a mechanism similar to the one proposed by us for
the photo-conversion of 4-allyl-tetrazolones into pyrimidinones.
Earlier, Quast and co-workers had studied the photochemistry of a
series of tetrazole derivatives, concluding that the photochemically
induced opening of the tetrazole ring is a very common process.24–28
Herein we present results on the photochemistry of 5-allyloxy-
tetrazoles (3a–c, Scheme 2) in solution. Besides our own interest
in understanding the mechanism of photocleavage of these ethers,
the possibility of isolation and characterization of reaction photo-
products may be particularly important for synthetic applications.
Scheme 2 Structure of 5-allyloxy-tetrazoles 3a–c.
Results and discussion
Steady-state photolysis: photoreactivity and photoproduct
distribution
The 5-allyloxy-tetrazoles studied were synthesized from the cor-
responding allylic alcohols, initially converted into the alkox-
ides with sodium hydride, and then reacted with 5-chloro-1-
phenyltetrazole.29
UV spectra of ethers 3a and 3b show absorption maxima around
230 nm, while the ether 3c exhibits an intense absorption band at
about 250 nm, due to the bathochromic effect of the phenyl ring
(Fig. 1). The substrates dissolved in methanol, acetonitrile and
cyclohexane were irradiated with a low-pressure mercury lamp
(k = 254 nm). The progress of the photoreactions was monitored
by HPLC and GC-MS and the quantum yields for photocleavage
were determined in each one of the tested solvents.
Fig. 1 shows changes in the UV absorption spectra of com-
pounds 3a–c caused by the irradiation (k = 254 nm) of their air-
equilibrated methanolic solutions. For compound 3a, the absorp-
tion peak around 230 nm decreased with irradiation time, while a
new absorption band grew at 238 nm. These spectral changes are
very similar to those observed for derivative 3b, indicating that the
methyl group attached to the allylic chain exerts a negligible effect
on the photoreactivity. Moreover, in the spectra of compounds
3a and 3b (Fig. 1), the absorptivity at 254 nm registered for the
new band belonging to the photoproducts is higher than that of the
initial substrates at the same wavelength. This fact will significantly
reduce the apparent photodecomposition rates of the two ethers,
since the formed photoproducts absorb a large part of the incident
UV light.
Fig. 1 Changes in UV spectra of 5-allyloxy-tetrazoles 3a–c in methanol
(1 × 10−4 mol dm−3) induced by UV irradiation (k = 254 nm) at room
temperature, with intervals of 30 s (3a, 3b) and 15 s (3c). The vertical
arrows indicate the evolution of absorbance with the irradiation time.
we can observe that the absorptivity at 254 nm registered for the
new bands belonging to the photoproducts is lower than that of
the initial substrate at the same wavelength. For derivative 3c, the
competition for the UV light between the photoproducts and the
Interestingly, for ether 3c, the reaction proceeds faster than for
compounds 3a and 3b, as seen from their absorption spectral
changes, showing an isosbestic point at ≈275 nm. In Fig. 1 (3c)
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 1046–1055 | 1047
©

View Article Online
substrate is much less marked, due to the absorption of the large
part of the incident UV light by the substrate.
and 1H-NMR analysis of the isolated primary extracts confirmed
the structures of compounds 4a–c.
HPLC and GC-MS analysis of irradiated methanolic solutions
of ethers 3a–c led to the identification of one primary photoprod-
uct for each of the three derivatives. These primary photoproducts
result from N2 elimination from the tetrazolyl ring system through
photo-induced cleavage of the two weakest N–N formally single
bonds of the heterocycle, and were characterized as N-phenyl-1,3-
oxazines 4a (m/z = 174), 4b (m/z = 188) and 4c (m/z = 251)
(see Fig. 2). Preparative-scale irradiations of the three tetrazolyl
ethers in methanol were performed, aiming at the isolation of
the primary photoproducts. These irradiations were carried out in
cylindrical quartz cells with a 50 mm light path, using 50 mg of the
starting material (see experimental section) and were controlled
by TLC and HPLC. Irradiation of the sample continued until
the first detection of a secondary photoproduct. The remaining
starting material was separated from the primary photoproduct
by column chromatography on silica gel. Mass spectrometry (CI)
Oxazines 4a–c can adopt two tautomeric forms, depending on
the position of the amino function on the molecule. The NH
group can act as a bridge, connecting the oxazine and phenyl
rings, or this group can be alternatively included in the oxazine
ring. DFT calculations performed at the B3LYP/6-31G(d,p)
level of theory carried out for the two tautomers of oxazines
4a and 4b, led to the identification, for both compounds, of
two low-energy local minima when the NH group is connected
to the two rings (structure 4(i) in Fig. 2), and of three low-
energy local minima when the NH function is included in the
oxazine ring (structure 4(ii) in Fig. 2). Calculations performed
at the same level of theory for oxazine 4c revealed nine low-
energy local minima for the two tautomers of this compound:
four minima with the NH group linking the phenyl and oxazine
rings, and five minima with the NH included on the oxazine
ring. All possible conformers and tautomers of oxazines 4a–c
have been theoretically investigated and the calculated Hessian
matrices showed no imaginary frequencies, confirming that all the
structures are true energy minima. For simplicity, only the data
for the most stable structure of each compound are included in
the ESI (pages S4–S6).†
In order to estimate the conformational and tautomeric com-
position of compounds 4a–c in solution, the thermochemical
◦
calculations were carried out for all compounds at 25 C. Based
on the calculated Gibbs free energies, the equilibrium Boltzmann
populations were estimated for all tautomers/conformers and the
selected results are presented in Table 1.
Because of the significant energy differences between the
tautomers 4(i) and 4(ii), it can be expected that the population
of oxazines at 25 ◦C will be dominated by structures 4(i). Indeed,
for all compounds 4a–c, predicted populations of forms 4(i) exceed
96% (see Table 1). The contribution of the minor tautomer 4(ii)
to the equilibrium mixture was predicted to range from about 2
to 4%, for isolated molecules in vacuum. It is important to call
attention to the fact that the total dipole moments of tautomers
4(ii) are systematically higher than those of the tautomers 4(i)
(Table 1). It may be expected that, in the polar media, forms 4(ii)
will undergo additional stabilization with respect to forms 4(i),
and the relative population of the minor conformer will increase.
Based on this reasoning, we may conclude that both the tautomers
4(i) and 4(ii) will be relevant for further photolysis of oxazines 4a–c
in solution. Therefore, the secondary photoproducts of 5-allyloxy-
tetrazoles 3a–c will be formed via photodecomposition of forms
4(i) and 4(ii), as presented in Fig. 2.
Fig. 2 Proposed photodegradation pathways for 5-allyloxy-tetrazoles
3a–c, in solution.
Table 1 B3LYP/6-31G(d,p) calculated relative energies, populations and dipole moments for tautomers (i) and (ii) of oxazines 4a–ca
Relative energy/kJ mol−1
Population (%)
Dipole moment (Debye)
Substrate
(i)
(ii)
(i)
(ii)
(i)
(ii)
4a
4b
4c
0
0
0
9.8
7.6
8.2
98.2
96.0
98.3
1.8
4.0
1.7
1.97
1.63
1.88
3.80
4.21
4.33
a Relative energies and dipole moments correspond to the most stable conformations of tautomers (i) and (ii); populations correspond to the sum of
populations of all conformers within a given tautomer, and were estimated based on the calculated relative Gibbs energies at 25 ◦C.
1048 | Org. Biomol. Chem., 2008, 6, 1046–1055
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
We propose four different photochemical channels for the
photodegradation of oxazines 4a–c in methanol (routes A–D,
Fig. 2).
similar to those used for 5-allyloxy-tetrazoles 3a–c. Phenyl-
isocyanate remained photostable upon prolonged irradiation at
254 nm, and no aniline formation was ever detected.
In the first channel (route A), photoproduct 5 (a phenyl vinyl-
hydrazine) is formed by extrusion of carbon monoxide from the
oxazine ring. In the second photochemical channel (route B), the
oxazine is directly converted, via a concerted mechanism, into
aniline 6 and photoproduct 7, also with CO extrusion.
Another photoproduct was identified upon irradiation of ethers
3a–c in methanolic solutions, namely benzyl carbamate (10, m/z =
151) (route D, Fig. 2). One possible explanation for the formation
of compound 10 could be the rearrangement of methyl phenyl-
urethane (9, Fig. 2), produced from the reaction of phenyl-
isocyanate, formed via route C, with a solvent molecule. To check
this hypothesis, we have used an authentic sample of carbamate 9
and irradiated it in a methanolic solution under conditions similar
to those used for the tetrazolyl ethers. However, chromatographic
analysis demonstrated that benzyl carbamate is not a product of
photodegradation of carbamate 9. Therefore, at the present stage
the pathway leading to formation of benzyl carbamate 10 upon
photolysis of tetrazolyl ethers 3a–c remains unclear.
The kinetics of photodecomposition of 5-allyloxy-tetrazoles
3a–c in methanolic solution are shown in Fig. 3, along with
the kinetics of photoproducts formation. After 120 minutes of
UV-irradiation, about 90% of the reagents 3a–c were consumed.
The photodecomposition of ether 3c occurs faster. After 30
minutes, 63% of compound 3c was consumed, against 38% for
compounds 3a,b. After formation of the primary photoproducts
4a–c, we found that, among the secondary photoproducts, the
transformation of oxazines into phenyl-isocyanate (route C, Fig. 2)
proceeds with the highest efficiency, as can be observed in Fig. 3.
This observation indicates that, in solution, the population of
tautomer with the higher dipole moment, 4(ii), might be higher
than that predicted for the gas phase on the basis of molecular
orbital calculations. But more important is the fact that the
population of 4(ii) is constantly maintained as “small-but-present”
due to dynamic equilibrium with 4(i) in solution. Then the
photoreaction may be derived from the constantly re-populated
4(ii), if 4(i) is not reactive (or not reactive enough) and serves
mainly as a source of 4(ii).
Fig. 4 shows the changes in the UV absorption spectra of
ethers 3a–c upon UV-irradiation (k = 254 nm) in air-equilibrated
acetonitrile and cyclohexane solutions. The evolution of the UV
absorption spectra reveals that the photoreactivity of 3a–c should
be very similar in acetonitrile and cyclohexane. The absorption
band around 230 nm present in the spectra of compounds 3a
and 3b decreased with irradiation time, while a new band grew at
238 nm. Fig. 5 shows the kinetics for the photodecomposition of
5-allyloxy-tetrazoles (3a–c) in acetonitrile. As in methanol so-
lutions, the photoreactivities of compounds 3a and 3b are very
similar, whereas the photodecomposition of 3c is faster. Both the
reaction rates and the photoproduct distributions are very similar
in the three solvents, with the extrusion of the N2 from the tetrazole
ring remaining the sole primary process of photodegradation of
tetrazoles 3a–c, leading to the formation of oxazines 4a–c.
Analysis of the kinetic results obtained for the photodegrada-
tion of ethers 3a–c in the three tested solvents clearly shows that the
effect of the nature of the solvent on the photoproduct distribution
is very weak, i.e., the relative percentage of photoproducts is
maintained for all cases. In methanolic solution, formation of
benzyl-carbamate is observed, due to nucleophilic attack by
methanol.
The proposal of the independent photochemical channel (route
A) leading to formation of aniline and products 5 and 7, is
fully justified by the fact that product 5 and aniline start to be
detected at the same reaction time. Upon continued exposure to
UV radiation, even after complete decomposition of substrates
3a–c into the primary photoproducts, we noted that the concen-
tration of product 5 was decreasing, accompanied by a similar
growth of aniline concentration. Hence, aniline and product
7 cannot result exclusively from oxazine photodecomposition
through a concerted mechanism (route B), being also a product of
photodecomposition of the photoproduct 5 (route A). Probably,
both routes A and B operate via the same transient species formed
after absorption of a photon by the oxazine tautomers 4(i) (Fig. 2).
The production of aniline, directly from oxazine tautomers
4(i) or by photodegradation of compound 5, should involve
the formation of the phenylnitrene intermediate [Ph–N:], which
subsequently abstracts two hydrogen atoms from the environment,
forming product 6 (see Fig. 2). The proposed source of the two
hydrogen atoms required to form aniline will be addressed further
in this work, together with a discussion of the kinetics of the
photoreactions.
The photoproducts 7a (R = H; m/z = 55) and 7b (R = CH3;
m/z = 69) have never been detected chromatographically in the
course of the irradiation experiments. This could be explained
by their high volatility, fast photodecomposition into molecular
products of lower weight, or very rapid elution in the chromato-
graphic conditions used. On the other hand, the photoproduct
7c (R = Ph; m/z = 131) was clearly identified by GC-MS. The
attribution of chemical structures to products 5 and 7 was based
on their mass spectra, through the analysis of the fragmentation
patterns exhibited. Aniline was easily identified, comparing the
respective chromatogram with that of an authentic aniline sample.
A single photochemical channel is proposed for the formation
of product 8 (phenyl-isocyanate). This species is formed via
degradation of the minor oxazine tautomer 4(ii) (route C, Fig. 2).
As for aniline, phenyl-isocyanate was identified using an authentic
sample for comparison. Formation of phenyl-isocyanate in the
photochemical channel C implies formation of a by-product also
produced in routes A and B, namely product 7.
Apparently, phenyl-isocyanate, also obtained as a stable sec-
ondary photoproduct upon photolysis of 4-allyl-tetrazolones21
could undergo photocleavage to phenylnitrene [Ph–N:], with
elimination of CO, which could then abstract hydrogen to form
aniline. However, in view of the experimental observations, we
consider that aniline and phenyl-isocyanate are produced via two
independent photodegradation channels of oxazines (routes A,
B and C, Fig. 2). In the course of the irradiation experiments,
simultaneous formation of photoproducts 6 and 8 was detected,
indicating that aniline does not derive from phenyl-isocyanate.
This hypothesis was confirmed by HPLC analysis of irradiated
solutions of phenyl-isocyanate in methanol, under conditions
The analysis of the kinetics of photoreactions led us to propose
a possible mechanism for aniline formation from phenyl-nitrene
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 1046–1055 | 1049
©

View Article Online
(the most labile H’s) are abstracted from positions 3 and 6 of
the oxazine ring of 4a–c photoproducts in their 4(ii) tautomeric
form. The suggested mechanism for hydrogen abstraction is
shown in Scheme 3. The optimized structures of derivatives
11a–c are included as ESI (pages S14–S15).† An analogous
photochemically induced hydrogen abstraction was reported for
1,4-cyclohexadiene in pentane solution30 and the possibility of
the occurrence of such a reaction was also shown theoretically.31
Photolysis quantum yields
Quantum yields for the photodecomposition of 5-allyloxy-
tetrazoles 3a–c were determined in the three solvents. They are
collected in Table 2 and show that the photoreactivity is very
similar in acetonitrile and cyclohexane and is only slightly higher
in methanol. Thus, the solvent effect on reactivity is indeed very
weak. However, the enhancement of the photolysis quantum yield
caused by the introduction of a phenyl on the allylic chain is much
more pronounced.
Laser flash photolysis and mechanistic considerations
Laser flash photolysis of 5-allyloxy-tetrazoles 3a–c in acetonitrile
at 266 nm revealed formation of short-lived transient species,
as detailed in Table 3. The transient decay for ether 3a is
apparently biexponential, with the contribution of the slower
transient increasing at higher concentrations of the substrate. The
faster transient for ether 3a is spectrally similar to that of ether 3b,
presented in Fig. 6 (top), with the lifetimes only slightly affected by
the presence of dissolved oxygen. We identify these two transients
as reactive radical species.
Laser flash photolysis of a deaerated 10−4 mol dm−3 solution
of tetrazole 3b in acetonitrile yielded a wide transient absorption,
with a maximum at 300 nm (see Fig. 6). The lifetime of this species
was around 800 ns under our experimental conditions.
Ether 3c is once again a special case. The corresponding
transient appears at shorter wavelengths, with the decay kinetics
clearly accelerated by the dissolved oxygen (see Table 3). We
identify this transient as a triplet excited state. Compared with
Table 2 Quantum yields (Uobs) for the photodegradation of 5-allyloxy-
tetrazoles 3a–c, in the range of solvents used (k = 254 nm, 25 ◦C)
Fig. 3 Time evolution of the amount of 5-allyloxy-tetrazoles 3a–c and of
photoproducts resulting from UV-irradiation (k = 254 nm) in methanolic
solution (1 × 10−4 mol dm−3). The amount of reagent before irradiation
was assumed to be 100%. The yields of different photoproducts were
monitored by gaseous chromatography. Note that the ordinate scale is
logarithmic. Note the assignment of geometrical symbols (rhombs—3a–c,
squares—oxazine, triangles—aniline, circles—phenylisocyanate, etc.) to
particular photoproducts on top of each frame.
Solvent (air-equilibrated)
3a, Uobs
3b, Uobs
3c, Uobs
CH3CN
Cyclohexane
CH3OH
0.023
0.025
0.037
0.021
0.020
0.035
0.106
0.112
0.144
Table 3 Selected spectral and kinetic parameters for the transient species
detected by laser flash-photolysisa
in photochannels A and B. All kinetics resemble each other
very closely, and occur in solvents that are very different from
the viewpoint of proton-affinities: methanol, cyclohexane and
acetonitrile. It is very unlikely that the two additional hydrogen
atoms required for formation of aniline are provided by the
solvent. Most probably, the compound itself undergoes an
additional photochemical transformation resulting in abstraction
of two H atoms. This abstraction does not involve the three
R-substituents (hydrogen, methyl, phenyl), since aniline formation
is observed in all cases. We suggest that the two hydrogen atoms
Transient 1
Transient 2
Ether
Lifetime/ls
k
max/nm
Lifetime/ls
kmax/nm
3a (N2, sat.)
3a (O2, sat.)
3b (N2, sat.)
3b (O2, sat.)
3c (N2, sat.)
3c (O2, sat.)
2.2
1.3
0.8
0.7
2.3
0.5
350
330
300
320
290
290
13.8
10.0
—
—
—
310
290
—
—
—
—
—
a Results obtained in acetonitrile with excitation at 266 nm.
1050 | Org. Biomol. Chem., 2008, 6, 1046–1055
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
Fig. 4 Changes in UV spectra of 5-allyloxy-tetrazoles 3a–c in acetonitrile (left) and cyclohexane (right) (1 × 10−4 mol dm−3) induced by UV irradiation
(k = 254 nm) at room temperature, with intervals of 30 s (3a, 3b) and 15 s (3c). The vertical arrows indicate the evolution of absorbance with the irradiation
time.
ethers 3a–b, the excited-state lifetime of the substrate 3c is
increased due to the presence of a larger “p”-electronic system,
whereas the hydrogen transfer reaction is further accelerated by
the presence of the R = Ph substituent and does not limit the
overall reaction kinetics any more. This justifies the identification
of the transients observed for 3a–b as radical species, and that for
3c as the triplet excited state. This assumption is further supported
by the considerations that follow.
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 1046–1055 | 1051
©

View Article Online
Fig. 6 Absorption spectra observed for the transient species formed
upon laser excitation of 3b (top) and 3c (bottom) in acetonitrile (1 ×
10−4 mol dm−3), under nitrogen, immediately after the laser (266 nm) pulse.
The inserts show decays, as monitored at 310 nm (3b, top) and 290 nm (3c,
bottom).
Indeed, the photoinduced loss of nitrogen from tetrazoles 3a–c
may involve a cycloelimination leading to a diazirine that reacts
further to give the observed oxazines 4a–c, or may occur through a
biradical intermediate that subsequently cyclises to give the same
compounds 4a–c. Usually, the transition state associated with a
concerted mechanism (cleavage of two N–N formally single bonds
and formation of one new N–N formally single bond and the
Fig. 5 Time evolution of the amount of 5-allyloxy-tetrazoles 3a–c and of
photoproducts resulting from UV-irradiation (k = 254 nm) in acetonitrile
solution (1 × 10−4 mol dm−3). The amount of reagent before irradiation
was assumed to be 100%. The yields of different photoproducts were
monitored by gaseous chromatography. Note that the ordinate scale is
logarithmic. Note the assignment of geometrical symbols (rhombs—3a–c,
squares—oxazine, triangles—aniline, circles—phenylisocyanate, etc.) to
particular photoproducts on top of each frame.
=
transformation of N N double bond into N≡N triple bond) has a
higher polarity than the initial substrate. Considering the variation
in polarity associated with the three solvents used in this work, it
appears that the solvent polarity effects are weak or absent during
tetrazolone photolysis, contrasting with what could be expected
for a concerted process occurring via a polar transition state.
Scheme 3 Proposed mechanism for the hydrogen abstraction from oxazines 4a–c in their 4(ii) tautomer modification.
1052 | Org. Biomol. Chem., 2008, 6, 1046–1055 This journal is The Royal Society of Chemistry 2008
©

View Article Online
It is well known that biradicals are among the most important
intermediates in solution-phase organic photochemistry. However,
and in agreement with reports in the literature, detection of these
intermediates is not easy, and could not be achieved in this work.
In some related investigations, such species were detectable only
in a limited number of cases.23 On the other hand, based on the
experimental results obtained, namely on the weak solvent effects,
it is our conviction that a radicalar mechanism operates in this
case.
Hence, we attribute the observed transients of 5-allyloxy-
tetrazoles 3a–c to the triplet biradicals 3(i) obtained through
homolytic cleavage of two N–N single bonds, and the observed
kinetics to the respective proton transfer reaction 3(i) → 3(ii)
(see Fig. 2). After conversion of triplet 1,6-biradicals 3(ii) into
singlet biradicals via intersystem crossing, these intermediates
decay to the ground state yielding final products. Presumably,
faster intersystem crossing occurs when the approach of both
radical termini becomes easier. In the case of 3c, the triplet excited
state lives longer, whereas the proton transfer reaction 3(i) → 3(ii)
operates faster, by influence of the phenyl substituent group, and
therefore does not limit the overall reaction kinetics, as happens
for the substrates 3a and 3b.
of ca. 10−6 s, which subsequently isomerizes to form a triplet 1,6-
biradical. After intersystem crossing, the 1,6-biradical decays to
the ground state to form the products.
Experimental section
Materials
All chemicals were used as purchased from Aldrich. Solvents for
extraction and chromatography were of technical grade. When
required, solvents were freshly distilled from appropriate drying
agents before use. Compounds 3a–c were synthesized as described
elsewhere.29
Equipment and experimental conditions
Analytical TLC was performed with Merck silica gel 60 F254
plates. Melting points were recorded on a Stuart Scientific SMP3
melting point apparatus and are uncorrected. UV absorption
spectra were recorded on a Varian CARY 50 Bio UV–Visible
Spectrophotometer. Mass spectra were obtained on a VG 7070E
mass spectrometer by electron ionization (EI) at 70 eV. 1H
NMR (400 MHz) spectra were obtained on a Bruker AM-
400 spectrometer using TMS as the internal reference (d =
0.0 ppm). Elemental analyses were performed on an EA1108-
Elemental Analyzer (Carlo Erba Instruments). Infrared spectra
were obtained as neat oils, or solids in NaCl on a Bruker FTIR-
TENSOR 27 spectrometer.
HPLC analyses were performed using an Agilent 1100 Series
chromatograph with a 655A-22 UV detector and Shimadzu SPD-
M6A Photodiode Array. A LiChroCART 125 column (RP-18,
5 lm, Merck) was used and the runs were performed using a
mixture of water and acetonitrile (40 : 60) as the eluent.
GC-MS analyses were carried out using a 6890-N Network
GC System gas chromatograph with a 5973 inert Mass Selective
Detector (E.I. 70 eV) of Agilent Technologies, using a DB-
35MS capillary column with 30 m length and 0.25 mm I.D.
(J & W Scientific, Agilent). The initial temperature of 50 ◦C was
maintained during 3 min and then a heating rate of 10 ◦C min−1 was
applied, until a final temperature of 250 ◦C was reached. Analyses
were conducted on irradiated samples and on control solutions,
kept in darkness. Controls showed no sign of degradation.
Conclusions
The results reported in this paper answer several important mech-
anistic questions concerning the photochemistry of 5-allyloxy-
tetrazoles 3a–c. Photolysis of these compounds in the steady-state
has been investigated in methanol, acetonitrile and cyclohexane
solutions. Through this study we have shown that the exposure
of tetrazoles 3a–c to UV light (k = 254 nm) induces a primary
photochemical process in which molecular nitrogen is elimi-
nated, producing oxazines 4a–c. In preparative-scale experiments,
compounds 4a–c were isolated from the reaction medium and
fully characterized. DFT(B3LYP)/6-31G(d,p) calculations pre-
dict that, in the gaseous phase, oxazines 4a–c should mostly exist in
the tautomeric form where the NH group is the bridge connecting
the oxazine and phenyl rings (the lowest-energy structures). The
tautomers of 4a–c with the NH function included in the oxazine
ring are higher in energy, the differences exceeding 7.5 kJ mol−1.
However, it is expected that, in the polar media, the higher-
energy forms of oxazines will undergo additional stabilization
with respect to the lowest-energy structures, and the relative
population of the minor conformer will increase. In fact, secondary
photoproduct analysis supports this interpretation. It is clear from
the experiments that the presence of both oxazine tautomers
is relevant in solution, increasing the number of secondary
photodegradation channels in the three solvents. From these,
phenyl vinyl-hydrazines 5a–c, enamines 7a–c, aniline and phenyl-
isocyanate were obtained. The pathway leading to production of
phenyl-isocyanate predominates, this resulting from photocleav-
age of oxazine 4(ii). Kinetic results for the photodegradation
of ethers 3a–c in different solvents show that the effect of the
nature of the solvent on the photoproduct distribution is very
weak and the relative yields of photoproducts are maintained in
all cases. However, there is a structural effect on the photolysis
quantum yields upon introduction of a phenyl on the allylic chain.
Mechanistically, primary photoexcitation of 5-allyloxy-tetrazoles
should involve formation of a triplet 1,3-biradical with a lifetime
Optical bench irradiation system
Steady-state photolysis experiments were carried out in acetoni-
trile, cyclohexane and methanol using 1 × 1 cm fused-silica cells
(Hellma) on a standard optical bench system. Photolyses were
conducted with the cells at a distance of 10 to 40 cm from the
Hg lamp. Generally, 10−4 mol dm−3 solutions of substrates were
used when the analysis was carried out by UV–Visible absorption
spectrophotometry. More concentrated solutions (10−2 mol dm−3)
were irradiated under the same conditions for product identifica-
tion using GC-MS and HPLC. The 254 nm UV radiation was
performed using a 16 W low-pressure mercury lamp (Applied
Photophysics). A suitable interference filter was used to isolate
the 254 nm line of the Hg lamp. Photolysis quantum yields were
measured using ferrioxalate actinometer.32–34
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 1046–1055 | 1053
©

View Article Online
Laser flash-photolysis
gel (hexane–AcOEt 6 : 1), to give 15 mg (33%) of 4c as a yellow oil.
IR (NaCl disks), mmax: 3255 (N–H), 1678, 1597, 1543, 1498, 1444,
Flash photolysis experiments were carried out on an Applied
Photophysics LKS.60 laser-flash-photolysis spectrometer, with a
Spectra-Physics Quanta-Ray GCR-130 Nd/YAG laser (fourth
harmonic, 266 nm) for excitation and a Hewlett–Packard Infinium
Oscilloscope for transient decay capture. Sample solutions were
pumped through a quartz cell at a 0.8 mL min−1 flow rate (SSI
chromatographic pump).
1
1314, 1250, 1223 and 752 cm−1; H NMR (400 MHz, CDCl3):
d 4.79–4.81 (2H, d, –CH2–), 5.42 (1H, s, N–H), 6.72–6.79 (1H,
=
t, CH), 7.28–7.33 (3H, m), 7.43–7.48 (4H, m), 7.55–7.60 (3H,
m) ppm; MS (EI), m/z 251 [M + H]+. Acc. Mass (CI): found =
251.2312, calcd for C16H15N2O: 251.2312.
Acknowledgements
Computational methods
The authors are grateful to Fundac¸a˜o para a Cieˆncia e a
Tecnologia (FCT) and FEDER for financial support, through
projects POCI/QUI/59019/2004, POCI/QUI/58937/2004 and
PTDC/QUI/67674/2006 and grant SFRH/BD/17945/2004
(L.M.T.F).
The equilibrium geometries for compounds 4a–c, 5a–c, 6,
7a–c, 8, 9, 10 and 11a–c were fully optimized at the DFT level
of theory with the standard 6-31G(d,p) basis set, using the
Gaussian 03 program.35 DFT calculations were carried out with
the B3LYP three-parameter density functional, which includes
Becke’s gradient exchange correction36 and the Lee, Yang and Parr
correlation functional.37 No symmetry restrictions were imposed
on the initial structures. Geometries, energies and Cartesian
coordinates for the most stable conformers of molecules studied
are included as ESI (pages S4–S26).†
References
1 G. Sandmann, C. Schneider and P. Boger, Z. Naturforsch., C, 1996, 51,
534.
2 S. C. S. H. Singh, A. Chawla, V. Kapoor, D. Paul and R. Malhotra,
Prog. Med. Chem., 1980, 17, 151.
General procedure for the preparation of compounds 4a–c
(preparative scale irradiations)
3 J. H. Toney, P. M. D. Fitzgerald, N. Grover-Sharma, S. H. Olson, W. J.
May, J. G. Sundelof, D. E. Vanderwall, K. A. Cleary, S. K. Grant, J. K.
Wu, J. W. Kozarich, D. L. Pompliano and G. G. Hammond, Chem.
Biol., 1998, 5, 185.
Irradiations at k = 254 nm were carried out using a 10 mL
cylindrical quartz cell with a 50 mm light path (Hellma). Solutions
of ethers 3a–c (50 mg in 10 mL of acetonitrile) were irradiated
with continuous stirring, with the cell at a distance of 10 cm from
the Hg lamp. The sample irradiation continued until detection
of a secondary photoproduct, corresponding to the substrate
conversion of about 30%, as monitored by HPLC analysis. At this
point the solutions were removed and evaporated to concentrate
the photoproducts. The photoproducts 4a–c were isolated from the
crude by column chromatography on silica gel, using a mixture
of hexane and ethyl acetate as eluent. 1,3-Oxazines 4a–c were
obtained as light-yellow oils.
4 R. J. Herr, Bioorg. Med. Chem., 2002, 10, 3379.
5 T. Mavromoustakos, A. Kolocouris, M. Zervou, P. Roumelioti, J.
Matsoukas and R. Weisemann, J. Med. Chem., 1999, 42, 1714.
6 Y. Hashimoto, R. Ohashi, Y. Kurosawa, K. Minami, H. Kaji, K.
Hayashida, H. Narita and S. Murata, J. Cardiovasc. Pharmacol., 1998,
31, 568.
7 A. Desarro, D. Ammendola, M. Zappala, S. Grasso and G. B. Desarro,
Antimicrob. Agents Chemother., 1995, 39, 232.
8 Y. Tamura, F. Watanabe, T. Nakatani, K. Yasui, M. Fuji, T. Ko-
murasaki, H. Tsuzuki, R. Maekawa, T. Yoshioka, K. Kawada, K.
Sugita and M. Ohtani, J. Med. Chem., 1998, 41, 640.
9 A. D. Abell and G. J. Foulds, J. Chem. Soc., Perkin Trans. 1, 1997, 2475.
10 G. I. Koldobskii, V. A. Ostrovskii and V. S. Poplavskii, Khim.
Geterotsikl. Soedin., 1981, 1299.
11 C. Zhao-Xu and X. Heming, Int. J. Quantum Chem., 2000, 79, 350.
12 Y. B. Chae, K. S. Chang and S. S. Kim, Daehan Hwahak Hwoejee, 1967,
11, 85.
13 G. Maier, J. Eckwert, A. Bothur, H. P. Reisenauer and C. Schmidt,
Liebigs Ann., 1996, 1041.
14 A. Awadallah, K. Kowski and P. Rademacher, J. Heterocycl. Chem.,
1997, 34, 113.
15 L. M. T. Frija, I. D. Reva, A. Go´mez-Zavaglia, M. L. S. Cristiano and
R. Fausto, Photochem. Photobiol. Sci., 2007, 6, 1170.
16 L. M. T. Frija, I. D. Reva, A. Go´mez-Zavaglia, M. L. S. Cristiano and
R. Fausto, J. Phys. Chem. A, 2007, 111, 2879.
N-Phenyl-6H-1,3-oxazin-2-amine (4a). Ether 3a was irradi-
ated during 10 h. The crude oil was chromatographed on silica gel
(hexane–AcOEt 10 : 1), to give 15 mg (34%) of 4a as a yellow oil.
IR (NaCl disks), mmax: 3270 (N–H), 1672, 1642, 1595, 1560, 1498,
1
1448, 1228, 1073, 1021, 992 and 759 cm−1; H NMR (400 MHz,
CDCl3): d 4.22 (1H, s, N–H), 5.06–5.10 (2H, d, –CH2–), 5.33–5.38
=
(1H, d, CH–N), 6.05–6.08 (1H, m), 7.36–7.45 (1H, d), 7.46–7.52
(2H, t), 7.64–7.69 (2H, d) ppm; MS (EI), m/z 175 [M + H]+; Acc.
Mass (CI): found = 175.1312, calcd for C10H11N2O: 175.1344.
17 A. Go´mez-Zavaglia, I. D. Reva, L. Frija, M. L. Cristiano and R. Fausto,
J. Photochem. Photobiol., A, 2006, 180, 175.
18 A. Go´mez-Zavaglia, I. D. Reva, L. Frija, M. L. Cristiano and R. Fausto,
J. Photochem. Photobiol., A, 2006, 179, 243.
19 A. Go´mez-Zavaglia, I. D. Reva, L. Frija, M. L. Cristiano and R. Fausto,
J. Mol. Struct., 2006, 786, 182.
20 A. Go´mez-Zavaglia, I. D. Reva, L. Frija, M. L. Cristiano and R. Fausto,
J. Phys. Chem. A, 2005, 109, 7967.
21 L. M. T. Frija, I. V. Khmelinskii and M. L. S. Cristiano, J. Org. Chem.,
2006, 71, 3583.
22 L. M. T. Frija, I. V. Khmelinskii and M. L. S. Cristiano, Tetrahedron
Lett., 2005, 46, 6757.
23 I. R. Dunkin, C. J. Shields and H. Quast, Tetrahedron, 1989, 45, 259.
24 H. Quast, Heterocycles, 1980, 14, 1677.
4-Methyl-N-phenyl-6H-1,3-oxazin-2-amine (4b). Ether 3b was
irradiated during 10 h. The crude oil was chromatographed on
silica gel (hexane–AcOEt 10 : 1), to give 13 mg (30%) of 4b as a
yellow oil. IR (NaCl disks), mmax: 3264 (N–H), 2929, 1673, 1638,
1595, 1559, 1498, 1448, 1368 and 757 cm−1; 1H NMR (400 MHz,
CDCl3): d 0.92 (3H, s, –CH3), 4.20–4.24 (2H, t, –CH2–), 5.02–5.08
=
(1H, t, CH), 5.34 (1H, s, N–H), 7.45–7.54 (3H, m), 7.69–7.75
(2H, d) ppm; MS (EI), m/z 189 [M + H]+; Acc. Mass (CI): found =
189.2161, calcd for C11H13N2O: 189.2171.
25 H. Quast and L. Bieber, Angew. Chem., Int. Ed. Engl., 1975, 14, 428.
26 H. Quast and L. Bieber, Chem. Ber./Recl., 1981, 114, 3253.
27 H. Quast, A. Fuss and U. Nahr, Chem. Ber./Recl., 1985, 118, 2164.
N,4-Diphenyl-6H-1,3-oxazin-2-amine (4c). Ether 3c was irra-
diated during 6 h. The crude oil was chromatographed on silica
1054 | Org. Biomol. Chem., 2008, 6, 1046–1055
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
28 H. Quast and U. Nahr, Chem. Ber./Recl., 1985, 118, 526.
29 M. L. S. Cristiano and R. A. W. Johnstone, J. Chem. Soc., Perkin Trans.
2, 1997, 489.
30 S. De Feyter, E. W.-G. Diau and A. H. Zewail, Phys. Chem. Chem.
Phys., 2000, 2, 877.
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, V. Bakken, C. Adamo, J. Jaramillo,
R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi,
C. Pomelli, J. 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. G. Johnson,
W. Chen, M. W. Wong, C. Gonzalez and J. A. Pople, GAUSSIAN 03
(Revision C.02), Gaussian, Inc., Wallingford, CT, 2004.
31 S. Zimbros and Y. Haas, Phys. Chem. Chem. Phys., 2002, 4, 34.
32 C. A. Parker, Proc. R. Soc. London, Ser. A, 1953, 220, 104.
33 C. G. Hatchard and C. A. Parker, Proc. R. Soc. London, Ser. A, 1956,
235, 518.
34 J. G. Calvert and J. N. J. Pitts, Photochemistry, John Wiley and Sons,
New York, 1967.
35 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.
36 A. D. Becke, Phys. Rev. A, 1988, 38, 3098.
37 C. T. Lee, W. T. Yang and R. G. Parr, Phys. Rev. B, 1988, 37, 785.
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 1046–1055 | 1055
©