Photoinduced molecular rearrangements. Some comments on the ring-photoisomerization of 1,2,4-oxadiazoles into 1,3,4-oxadiazoles

Article abstract of DOI:10.1002/jhet.5570380340

The ring-photoisomerization of 3-amino- and 3-methylamino-5-phenyl-1,2,4-oxadiazoles into the corresponding 2-amino- and 2-methylamino-5-phenyl-1,3,4-oxadiazoles has been reinvestigated by examining the effect of a base on the photoreaction. On irradiatin

Full text of DOI:10.1002/jhet.5570380340

May-Jun 2001
Photoinduced Molecular Rearrangements. Some Comments on the
Ring-Photoisomerization of 1,2,4-Oxadiazoles into 1,3,4-Oxadiazoles
Silvestre Buscemi [a], Andrea Pace[a], Nicolò Vivona*[a] and Tullio Caronna [b]
777
[a] Dipartimento di Chimica Organica "E. Paternò", Università degli Studi di Palermo,
Viale delle Scienze - Parco D’Orleans II, 90128 Palermo, Italy
[b] Dipartimento di Chimica, Politecnico di Milano, Via Mancinelli 7, 20131 Milano, Italy
Received October 24, 2000
The ring-photoisomerization of 3-amino- and 3-methylamino-5-phenyl-1,2,4-oxadiazoles into the corres-
ponding 2-amino- and 2-methylamino-5-phenyl-1,3,4-oxadiazoles has been reinvestigated by examining
the effect of a base on the photoreaction. On irradiating at λ = 254 nm in methanol, yields of the ring-
photoisomers were found to be significantly enhanced by the addition of triethylamine (TEA) in the
photoreaction medium. By contrast, irradiation of the 3-amino-5-phenyloxadiazole in acetonitrile
containing TEA gave an almost complete photoreduction into benzoylguanidine, while few percent of the
ring photoisomer were detected. Furthermore, the pyrene-sensitized photolysis of 3-amino-5-phenyloxadiazole
in acetonitrile containing triethylamine also gave benzoylguanidine but no traces of the ring photoisomer.
J. Heterocyclic Chem., 38, 777 (2001).
The photochemistry of the 1,2,4-oxadiazole heterocycle
Besides the above photochemistry, ring-photoisomer-
ization of the 1,2,4-oxadiazoles into the corresponding
1,3,4-oxadiazoles has been also reported [3,7] (Scheme 1)
and claimed to occur (in low yields, however) only for
some structurally defined substrates. In fact, this photo-
reaction, which has been related to the isoxazole to
oxazole photorearrangement and interpreted according to
the ring contraction-ring expansion route [8], was found
to be dependent on the presence of an XH moiety such as
suggests that photolytic intermediates arising from the
ring O-N bond breaking will develop into different final
products depending on their structure and/or experimental
conditions [1-6]. Thus, they can stabilize into open-chain
compounds in a solvolytic [2,3] or reduction [4] pattern,
or they can undergo subsequent heterocyclization reac-
tions with or without involvement of an added reagent
[
4,5]. In this context, we recently reported [6] that irradia-
tion of some 3-substituted 5-aryl-1,2,4-oxadiazoles 1 in
methanol in the presence of different sensitizers or elec-
tron-donors such as triethylamine (TEA) gave both
heterocyclization to yield 2 by involving the aryl moiety
at C-5 of the oxadiazole nucleus and open-chain reduction
products 3 in different ratios depending on the structure of
intermediates and photoreaction conditions (Scheme 1).
This photoreactivity has been explained by assuming a
photoinduced electron transfer (either from the sensitizer
in its excited state to the oxadiazole substrate or from the
electron donor reagent to the excited oxadiazole) produc-
ing an oxadiazole radical anion from which breaking of
the ring O-N bond takes place.
NH , NHMe or OH at C(3) of the ring. This restriction
2
suggested that the ring-photoisomerization reaction
should involve tautomeric or deprotonated forms in the
ground or excited state of the oxadiazole substrate, or in
the outcoming three-membered ring intermediate [3].
To verify this hypothesis and get more insight into this
peculiar photoreactivity of the 1,2,4-oxadiazole nucleus,
we have now reinvestigated the ring-photoisomerization
of 3-amino- (4a) and 3-methylamino-5-phenyloxadiazole
(
4b) by examining the effect of an added base (such as
TEA) to the photoreaction medium [9]. Moreover, in view
of results concerning photolysis of 3-substituted 5-aryl-
oxadiazoles [6], we became also interested in checking
the possibility of an electron-transfer mechanism in the
ring-photoisomerization reaction.
Results and Discussion.
Irradiations of compounds 4a or 4b at λ = 254 nm in
methanol and in the presence of triethylamine (TEA) or
methylamine showed that both bases significantly
enhance their photoconversion into the ring isomers 5a or
5
b, respectively, by a comparable extent [10]. A typical
experiment carried out on both oxadiazoles 4a or 4b
showed that, after 1 hour of irradiation yields of the ring
isomer (5a or 5b) increase from a few percent in the
absence of the base to about 20% when irradiations were
carried out in the presence of TEA or methylamine.
Analysis of photolysates after 1 hour of irradiation of

7
78
S. Buscemi, A. Pace, N. Vivona, and T. Caronna
Vol. 38
solutions of 3-aminooxadiazole 4a containing different
molar ratios of TEA showed that the photoconversion was
markedly accelerated even by the presence of small
amounts of the added base (0.1 TEA/oxadiazole molar
ratio) and that the concentration of the ring-photoisomer
while only a few percent of the ring-isomer 5a could be
detected (Scheme 2). The highly efficient photoreduction
observed upon photolysis of 4a in MeCN/TEA clearly
suggests an electron transfer between the electron-donor
reagent (TEA) and the oxadiazole in its excited state. The
main fate of the resulting oxadiazole radical anion, likely
through its protonated form arising from the involvement
of the coupled TEA radical-cation, will be the breaking of
the ring O-N bond followed by reduction of open-chain
intermediate species into the final product 8. In protic
solvent, one could suppose that hydrogen bond formation
and solvation effects may change the mode of excited
state collapse resulting in its interaction with TEA as a
base rather than as a reducing agent.
To search a clear-cut electron transfer mechanism we
then examined the pyrene-sensitized photolysis of 4a in
acetonitrile containing TEA. In this case, the pyrene
radical-anion (which is formed by an electron transfer
from TEA to the excited pyrene) [12] will be the electron-
donor species toward the oxadiazole substrate in its
ground state [13] (Scheme 3). As expected, this photo-
reaction gave an almost complete photoreduction to 8,
while no significant amounts of the ring-photoisomer 5a
were detected. Again, the resulting oxadiazole radical-
anion will be the precursor of a ring-cleaved intermediate
such as 9 which will be converted essentially to 8 in a
reduction rather than in a heterocyclization reaction.
5
a reaches a maximun value of 20% at a TEA/oxadiazole
molar ratio of about 1/2, and remains practically
unchanged at higher ratios. Furthermore, upon more
prolonged irradiation times, the yield of the ring-photo-
isomer increased. In fact, a preparative scale photolysis of
the oxadiazole 4a in the presence of methylamine (used
instead of TEA to minimize photoreduction processes)
gave 5a in 40% yield. On the other hand, prolonged
irradiations of compounds 4a,b in methanol without
added base did not give acceptable preparative results,
since the photoreaction was complicated by the formation
of side-products.
To explain the above results it is noteworthy to con-
sider the role of the NHR group linked at C(3) of the
ring, from which base-catalized tautomeric equilibrium
may take place. It is well known [11] that amino-azoles
exist predominantly in the amino-form tautomer (at least
in the ground state). If one considers the excitation
process, we may suppose that the NH moiety of 3-
amino- (4a) or 3-methylamino-oxadiazole (4b) (in their
amino or tautomeric imino forms) will show a higher
acidity in the excited state than in the ground state; a
proton transfer may then occur which would be
markedly effected by the presence of a base. Our opin-
ion is that the excited anion 6 could be the key-species
collapsing into a diaziridine-type intermediate 7 from
which the final ring-isomer would take place (Scheme
2
). Involvement of an excited anion agrees with the
observation that a high yield of photoconversion also
takes place even at low molar ratio of the base. In the
case of irradiations in the absence of base, methanol (as
amphoteric solvent) can promote the photorearrange-
ment reaction, but to a lesser extent.
Interestingly, it is worthy to note that the effect of the
added base was found to be markedly dependent on the
solvent used in the photoreaction. Thus, irradiation of 4a
in acetonitrile (MeCN) containing TEA gave an almost
complete photoreduction into the benzoylguanidine (8),

May-Jun 2001
Photoinduced Molecular Rearrangements
779
As a conclusive comment, this investigation confirms
that the ring-photoisomerization of the 1,2,4-oxadiazole
nucleus into the 1,3,4-oxadiazole system can not be
considered a general photoreaction since it appears
strictly determined by structural features of substituent at
C(3) of the 1,2,4-oxadiazole heterocycle as well as by
experimental conditions. Further spectroscopic and
theoretical investigations are in progress to provide
greater understanding of different excited states involved
in these reactions and how the electronic transitions
depend on the structure of the oxadiazole substrate.
zole from 0.1 to 2. The samples were simultaneously irradiated
at λ = 254 nm for 1 hour, and photolysates were analysed by
HPLC. The concentration (%) of compound 5a reaches the
maximum value (20%) at the molar ratio TEA/oxadiazole of
0.5, and then remains practically invariated.
Preparative Scale Irradiation of the Oxadiazole (4a) in Methanol
containing Methylamine.
A solution of compound 4a (0.2 g) in methanol (200 ml) con-
taining methylamine (0.2 ml) was apportioned in five quartz
tubes and then irradiated at λ = 254 nm for 4 hours. After
removal of the solvent under reduced pressure, chromatography
of the residue (performed with light petroleum:ethyl acetate 1:1)
returned starting material (80 mg, 40%), and gave 5a (80 mg,
EXPERIMENTAL
40%), mp 242 °C, lit [14] mp 242 °C and 8 (10 mg, 5%), mp
1
60 °C lit [17] mp 160 °C.
For instruments and general procedures see our previous
papers [3] [4] [6]. HPLC analyses were performed by using a
C-18 SIL X-10 Perkin-Elmer column (25 cm x 4.6 mm
diameter) eluting with water/acetonitrile (1/1 mixtures).
Triethylamine and pyrene were obtained from Aldrich Chemical
Co. Ethanolic (33%) methylamine was obtained from Fluka;
anhydrous methanol and acetonitrile from Romil Co. Flash
chromatography was performed on Macherey-Nagel silica gel
Irradiation of the Oxadiazole (4a) in Acetonitrile Containing
TEA.
A sample of the oxadiazole 4a (10 mg, 0.06 mmol) in
acetonitrile (10 ml) containing TEA (0.13 mmol) was irradiated
at λ = 254 nm for 1 hour. HPLC analysis of the photolysate gave
starting material 4a (10%), benzoylguanidine (8) (85%) and the
oxadiazole 5a (5%).
60 (230-400 mesh).
Irradiation of the Oxadiazole (4a) in the Presence of Pyrene.
Photochemical reactions were carried out in anhydrous
methanol or in acetonitrile (the solutions were purged by
nitrogen bubbling before irradiation) by using a Rayonet
RPR-100 photoreactor equipped with 16 Hg lamps irradiating,
respectively, at λ = 254 nm (RPR-2537Å) (in quartz vessels)
and at λ = 365 nm (RPR-3500Å) (in pyrex vessels), and a
merry-go-round apparatus. Experimental conditions (irradiation
time and concentration of solutions to be irradiated) have been
chosen as to minimize the formation of side-products. However,
in each series of experiments, all the parameters involved have
been kept homogeneous. Since oxadiazole substrates do not
absorb at λ = 365 nm, in the pyrene-sensitized photolysis the
sensitizer was the only absorbing species. Quantitative analyses
of photolisates were accomplished by HPLC.
Irradiation of the oxadiazole 4a (10 mg, 0.062 mmol) in
acetonitrile (10 ml) containing pyrene (25 mg, 0.12 mmol) and
TEA (0.12 mmol) at λ = 365 nm for 1 hour gave 8 in higher
than 95% yields.
Acknowledgements.
The Authors are indebted to Professor James W. Pavlik
(Worcester Politechnic Institute - Massachusetts, USA) for
useful suggestions. Financial support from Italian CNR and
MURST is gratefully acknowledged.
REFERENCES AND NOTES
Oxadiazoles 4a [14] and 4b [7] (which were used for irradia-
tions), and compounds 5a [15], 5b [16] and 8 [17] which were
used as authentic samples for comparison and HPLC determina-
tions were obtained as reported in the above references.
[
[
[
1] N. Vivona and S. Buscemi, Heterocycles, 41, 2095 (1995).
2] H. Newman, Tetrahedron Lett., 2421 (1968).
3] S. Buscemi, M. G. Cicero, N. Vivona and T. Caronna,
J. Heterocyclic Chem., 25, 931 (1988).
4a] S. Buscemi, N. Vivona and T. Caronna, J. Org. Chem., 61,
397 (1996); [b] N. Vivona, S. Buscemi, S. Asta and T. Caronna,
[
Irradiations of 3-Amino- (4a) and 3-Methylamino-5-phenyl-
8
1,2,4-oxadiazole (4b) in Methanol in the Presence of Bases.
Tetrahedron, 53, 12629 (1997).
A solution of the oxadiazole 4a or 4b (0.19 mmol) in
[5a] S. Buscemi and N. Vivona, J. Heterocyclic Chem., 25, 1551
(
1988); [b] S. Buscemi and N. Vivona, Heterocycles, 29, 73 (1989); [c]
S. Buscemi, G. Macaluso and N. Vivona, Heterocycles, 29, 1301
1989); [d] S. Buscemi, G. Cusmano and M. Gruttadauria, J.
Heterocyclic Chem., 27, 861 (1990).
6] S. Buscemi, A. Pace, N. Vivona, T. Caronna and A. Galia, J.
methanol (30 ml) was apportioned into three quartz tubes (10 ml
each). In two of these samples the appropriate base (methyl-
amine or TEA) (0.13 mmol) was added and all three samples for
each oxadiazole substrate were simultaneously irradiated at
λ = 254 nm for 1 hour. HPLC analysis of samples irradiated
without the base showed the unreacted starting material (about
(
[
Org. Chem., 64, 7028 (1999).
[7] S. Buscemi, M.G. Cicero, N. Vivona and T. Caronna,
J. Chem. Soc., Perkin Trans. 1, 1313 (1988).
9
5%) and few percent (3-5%) of the 1,3,4-oxadiazole 5a or 5b,
respectively. HPLC analysis of samples irradiated in the
presence of bases showed comparable results for both irradia-
tions of 4a or 4b (recovered in 80 and 78 % yields, respectively)
determining compounds 5a (20%) or 5b (22%), respectively.
A solution of compound 4a (60 mg) in methanol (60 ml) was
apportioned into six quartz tubes to which variable amounts of
TEA were added as to have increasing molar ratios TEA/oxadia-
[8] A. Padwa, Rearrangements in Ground and Excited States, ed.
by P. De Mayo, Academic Press, Inc., New York, 1980 vol. 3, p 501.
[9] Examples of TEA-assisted ring-phototransposition have been
skillfully described in the isothiazole series. See: J. W. Pavlik, P.
Tongcharoensirikul, and K. M. French, J. Org. Chem., 63, 5592 (1998);
J. W. Pavlik and P. Tongcharoensirikul, J. Org. Chem., 65, 3626 (2000).
[10] Yields of the ring-photoisomers were found to be signifi-
t
cantly enhanced also by addition of anionic bases (such as BuOK) in

780
S. Buscemi, A. Pace, N. Vivona, and T. Caronna
Vol. 38
the photoreaction medium, but the formation of some decomposition
products alters analytical determinations.
tions. Emission spectra of the oxadiazole 4a, both in methanol and in
acetonitrile, did not show significant phosphorescence, but allowed
determination of energy values of 356 kJ/mol for the singlet excited state
(umpublished results from our laboratories).
[11] J. Elguero, C. Marzin, A. R. Katritzky and P. Linda, The
Tautomerism of Heterocycles, Suppl. 1 of Adv. Heterocycl. Chem., ed.
by A. R. Katritzky and J. Boulton, Academic Press, Inc., New York
[14] G. Westphal and R. Schmidt, Z. Chem., 14, 94 (1974); Chem.
Abstr., 81, 13443 (1974).
[15] T. M. Lambe, N. R. Butler and F. L. Scott, Chem. Ind.
(London), 996 (1971).
[16a] G. Werber and F. Maggio, Ann. Chim. (Rome), 52, 747
(1962); Chem. Abstr., 58, 3417 (1963); [b] J. F. Giudicelli, J. Menin and
H. Najer, Bull. Soc. Chim. Fr., 153 (1966).
1
976.
12] S. Murata, R. Nakatsuji and H. Tomioka, J. Chem. Soc.,
Perkin Trans. 2, 793 (1995).
13] A sensitization through an energy-transfer mechanism from
excited pyrene (E = 321 kJ/mol; E = 202 kJ/mol; S. L. Murov,
[
[
s
t
Handbook of Photochemistry, Marcel Dekker, New York, 1973) to the
oxadiazole substrate can be excluded on the basis of energetic considera-
[17] W. Traube, Chem. Ber., 43, 3586 (1910).