Unexpected production of 2,4,6-triphenyl-1,3,5-triazine in the electroreduction of 3,4-diphenyl-1,2,5-thiadiazole 1-oxide. Theoretical estimation of reactive sites for 1-oxide and 1,1-dioxide 1,2,5-thiadiazoles

Article abstract of DOI:10.1016/S0040-4039(00)00449-4

3,4-Diphenyl-1,2,5-thiadiazole 1-oxide (1a) in acetonitrile solution is electroreduced to 2,4,6-triphenyl-1,3,5-triazine and 3,4-diphenyl-1,2,5- thiadiazole. This behavior is very different from that of similar compounds with other oxidation states of the

Full text of DOI:10.1016/S0040-4039(00)00449-4

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 41 (2000) 3531–3535
Unexpected production of 2,4,6-triphenyl-1,3,5-triazine in the
electroreduction of 3,4-diphenyl-1,2,5-thiadiazole 1-oxide.
Theoretical estimation of reactive sites for 1-oxide and
1,1-dioxide 1,2,5-thiadiazoles
a
a
a
b
Silvia L. Aimone, María V. Mirífico, José A. Caram, Daniel Glossman Mitnik,
c
d
a,∗
Oscar E. Piro, Eduardo E. Castellano and Enrique J. Vasini
a
INIFTA, Facultad de Ciencias Exactas, Departamento de Química, U.N.L.P., C.C. 16, Suc. 4, (1900) La Plata, Argentina
b
CIMAV, Departamento de Polímeros, Chihuahua, Mexico
Facultad de Ciencias Exactas, Departamento de Física, U.N.L.P., La Plata, Argentina
c
d
Instituto de Física de São Carlos, Universidade de São Paulo, C.P. 369, 13560 São Carlos (SP), Brazil
Received 23 February 2000; revised 15 March 2000; accepted 16 March 2000
Abstract
3,4-Diphenyl-1,2,5-thiadiazole 1-oxide (1a) in acetonitrile solution is electroreduced to 2,4,6-triphenyl-1,3,5-
triazine and 3,4-diphenyl-1,2,5-thiadiazole. This behavior is very different from that of similar compounds with
other oxidation states of the heterocyclic sulfur atom, such as the 1,1-dioxide derivative (2) and the aromatic 3,4-
diphenyl thiadiazole parent ring. The Fukui functions were calculated for 1a and 2 to estimate their reactivity,
compare their reactive site, and rationalize the divergent electrochemical properties of 1a. © 2000 Elsevier Science
Ltd. All rights reserved.
Keywords: thiadiazoles; electrochemistry; structure–activity; reduction.
1
,2,5-Thiadiazoles have been broadly applied in the areas of pharmaceutical, agricultural, industrial
1
,2
and polymer chemistry. In several of these applications, 1,1-dioxide and 1-oxide derivatives were em-
ployed. We have studied aspects of the physical-organic chemistry of several 1,1-dioxides, but similar
work with 1-oxides is almost non-existent. We are not aware of any publication on the electrochemistry of
3–9
1
-oxide derivatives. Theoretical molecular orbital calculations for 1 have been performed with a modified
1
0
STO-3G* basis set, but the thiadiazole ring was constrained to be planar. A single-point 6-31G* energy
was also calculated for the same molecule, using a fully optimized STO-3G structure. No theoretical
estimations of reactive sites and reactivity have been attempted for the 1-oxide derivatives.
*
11
∗
Corresponding author. Fax: +54-221-425-4642; e-mail: vasini@inifta.unlp.edu.ar (E. J. Vasini)
0
040-4039/00/$ - see front matter © 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(00)00449-4
tetl 16729

3
532
†
We report here voltammetric and bulk electrolysis electroreduction experiments of 1a (Scheme 1),
at ca. −1.5 V, in acetonitrile (ACN) solution. 2,4,6-Triphenyl-1,3,5-triazine (4) and 3,4-diphenyl-1,2,5-
thiadiazole (3) were produced with ca. 30 and 50% yield, respectively. 4 Formation implies the opening
of the heterocycle ring. Thus, the electroreduction of 1a differs from that of 24 which, at more positive
,6
12
potentials (ca. −0.8 V) is reversibly reduced to a relatively stable radical anion. Compound 3 is also
electroreduced to a radical anion at more negative potentials (ca. −2.4 V).
Scheme 1.
A typical cyclic voltammogram of 1a (Fig. 1, full line), presented several cathodic peaks (at −1.19,
−
1.63, −2.29 (shoulder), and −2.35 V). If the scan was reversed right after traversing the peak at
−
1.19 V, a corresponding anodic signal could be observed. The ratio of the anodic to the cathodic
current intensities of the peaks (R=Ipa/Ipc) increased with the increase of the sweep rate, ν, and
the switching potential, E . The voltammetric current function for a one-electron reversible charge
λ
transfer, applied in the range of ν and E for which R>0.7 allowed the estimation of a diffusion
λ
−5
2
−1
coefficient (D=2.8±0.1×10 cm seg ). It can be reasonably assumed that this monoelectronic,
diffusion controlled, voltammetric couple at −1.19 V corresponds to the electroreduction of the substrate
to the radical anion (Eq. (1)):
−
·−
1
a + e z 1a
(1)
Subsequent reaction(s) must take place to account for the dependence of R with ν and E .
λ
12
The peak at −2.35 V was assigned to the electroreduction of 3 (E
1
2
=−2.39 ) by comparative cyclic
voltammetry (a cyclic voltammogram of a sample of 3 is included in Fig. 1, broken line). Furthermore,
was also a product of the bulk electroreduction of 1a (see below). From the relative current intensities
3
of the voltammetric peaks of 1a and 3 (Fig. 1), and the experimental concentration of the respective
subtrates, it can be estimated that ca. 50% of the voltammetrically generated 1a. transforms into 3.
Bulk electrolysis experiments, at −1.50 V, of a solution of 1a in ACN were performed. The electrolysis
products, identified as follows, were: (a) 4. Compound 4 is practically insoluble in the electrolytic
−
†
Voltammetric experiments were performed in an undivided, gas-tight, glass cell swept by purified nitrogen. The cell was kept
+
and manipulated within a dry box. The reference electrode (r.e.) was Ag (0.1 M ACN)/Ag separated from the cell solution by a
porous-glass plug. A 5 cm Pt foil was the counter electrode (c.e.) and the working electrode (w.e.) was a Teflon encapsulated
vitreous carbon disk of 0.074 cm . Bulk electrolysis experiments were carried out in a divided cell. A sinterized glass plate
2
®
2
separated the anodic and the cathodic compartments. The w.e. (cathode) was a vitreous carbon rod (0.3 cm diameter and
2
1
.5 cm length), the c.e. was an 8 cm Pt foil. The w.e. described above for voltammetry was also included into the cathodic
compartment. The electrolysis was interrupted at selected time intervals and cyclic voltammograms, using this w.e., were run
to measure the concentration of substrate.

3
533
Fig. 1. Cyclic voltammograms, current function vs potential for ACN solutions of 1a, 2, and 3. Supporting electrolyte: 0.20 M
−
1
_______
TEAP. v=0.2 V seg . (
): [1a]=10.8 mM; (...........): [2]=3.41 mM; ( — — — ): [3]=3.36 mM
solution and precipitated as a white solid. It was identified by IR, UV–visible spectroscopies14 and
13
14
GC–MS. Its mp was 233.5–234.5°C (lit.: 235.0–235.4°C ). The yield was ca. 30%. (b) Compound 3,
identified by GC–MS in a chloroformic extract of the electrolyzed solution. The yield of 3, as measured
by the the current intensity of its voltammetric peak after exhaustive electrolysis, was in the 40–48%
range. (c) Benzonitrile, identified by GC. (d) Benzil, identified by GC–MS and (e) a minor unidentified
product (GC–MS).
The concentration of 1a was measured at selected times during the electrolysis using the current
intensity of its voltammetric peak at −1.19 V. The calculated number of electrons per mole of 1a
electrolyzed was 0.4.
The dotted line in Fig. 1 corresponds to a typical cyclic voltammogram of 2 in ACN solution.
Compound 2 is reduced to a radical anion and to a dianion in two successive, reversible voltammetric
4
,6
peaks. The bulk electrolysis of 2 in ACN solution, at the potential of the first charge transfer, produces
a strongly colored solution containing, presumably, the radical anion. The solution is stable under the dry
N atmosphere of the cell.
2
When the electroreduction of 1a was carried out in a 0.60 M water solution in ACN, 3 was the main
product. The formation of a precipitate of 4 was not observed, although 4 was chromatographically
detected as a minor component. Conversely, when the concentration of 1a was increased (decreasing
.−
the relative amount of residual water available to 1a ), the relative yield of 3 decreased. Thus, 3 was
probably formed (Scheme 2) by reaction of the radical anion with residual water. The proposed structure
.
−
for 1a and the mechanism of Scheme 2 are in accordance with the calculated values of the Fukui
functions (see below).
It might be recalled that the 1-oxide thiadiazole 1c (Scheme 1) has been reported to deoxygenate smo-
othly to the thiadiazole by thermolysis at 250°C, while a 1,1-dioxide compound, otherwise structurally
1
5
identical to 1c, was stable at 280°C.
.−
The unexpected formation of the triazine from 1a involves the loss of the >SO group and the opening
of the ring at the heterocyclic C–C bond. At this stage, the mechanism of 4 formation can only be a matter
.−
of speculation, although a cleavage of the N–S bond of 1a , followed by an attack of the resulting
nucleophilic radical anion on a molecule of 1a appear as reasonable initial steps. A detailed study of the

3
534
Scheme 2.
distribution of the products as a function of the conditions of the electrolysis is underway to clarify the
mechanism.
‡
Reactivity and reactive sites for 1a and 2 were estimated through the calculation of the Fukui
1
7
functions in an attempt to rationalize the different electroreduction behavior.
Bond lengths and bond angles for 1a were in good agreement with those calculated, or measured
11
10
15
by X-ray diffraction for similar molecules (1b, 1c ). Table 1 shows the net atomic charges and the
+
−
o
condensed Fukui functions f , f and f (for a nucleophilic or an electrophilic attack, or for a radical
reaction, respectively), for the heterocyclic atoms of 1a and 2. Positive values of the condensed Fukui
functions, and adequate charges (i.e. positive for a nucleophilic attack) indicates preferential reactivity
for the type of reaction in question. The results predicted, as expected, the N atoms as preferred sites for
electrophilic attacks. Less obviously, the preferred site for a nucleophilic attack or for the formation of a
+
o
radical anion, as judged by the f and the f values of the positively charged atoms, were the heterocyclic
C atoms for 2 and the S atom for 1a.
Table 1
‡
Ab-initio HF/3-21G** and HF/6-31G** Hartree–Fock LCAO molecular orbital (MO) calculations were performed for the
1
6
neutral molecule, the radical anion and the radical cation of 1a and 2 with the Gaussian 94W series of programs. RHF single
determinantal wavefunctions were used for closed-shell systems and UHF for the open shell systems, with orbitals being the
solutions of the Roothaan equations. The equilibrium conformation of the neutral molecule was calculated by the gradient
method.

3
535
Acknowledgements
We thank CONICET, CIC Pcia Bs.As. and the UNLP for financial support.
References
1
2
. Weinstock, L. M.; Shinkai, I. In Comprehensive Heterocyclic Chemistry; Katrisky, A. R.; Reeds, W. C., Eds. The 1,2,5-
thiadiazoles. Pergamon Press: Oxford, 1984; Vol. 6, pp. 107–163.
. Arán, V. J.; Goya, P.; Ochoa, C. In Advances in Heterocyclic Chemistry; Katrisky, A. R.; Boulton, A. J., Eds. Heterocycles
containing the sulfamide moiety. Academic Press: New York and London, 1988; Vol. 44, pp. 81–197.
. Mirífico, M. V.; Vasini, E. J.; Sicre, J. E. Int. J. Chem. Kinet. 1991, 23, 197–202.
. Mirífico, M. V.; Caram, J. A.; Vasini, E. J. Electrochim. Acta 1991, 36, 167–171.
. Mirífico, M. V.; Caram, J. A.; Vasini, E. J.; Sicre, J. E. J. Phys. Org. Chem. 1993, 6, 341–346.
. Caram, J. A.; Mirífico, M. V.; Vasini, E. J. Electrochim. Acta 1994, 39, 939–945.
. Mirífico, M. V.; Vasini, E. J. An. Quim. 1995, 91, 557–560.
3
4
5
6
7
8
9
. Caram, J. A.; Mirífico, M. V.; Aimone, S. L.; Vasini, E. J. Can. J. Chem. 1996, 74, 1564–1571.
. Castellano, E. E.; Piro, O. E.; Caram, J. A.; Mirífico, M. V.; Aimone, S. L.; Vasini, E. J.; Glossman, M. D. J. Phys. Org.
Chem. 1998, 11, 91–100.
10. Amato, J. S.; Karady, S.; Reamer, R. A.; Schlegel, H. B.; Sprigel, J. P.; Weinstock, L. M. J. Am. Chem. Soc. 1982, 104,
1
375.
1
1
1
1
1
1
1. Rozas, I. J. Phys. Org. Chem. 1992, 5, 74–82.
2. Sherman Jr., E. O.; Lambert, S. M.; Pilgram, K. J. Heterocycl. Chem. 1974, 11, 763.
3. Spencer, R. D. Spectrochim. Acta 1965, 21, 1543.
4. Spencer, R. D.; Beggs, B. H. Anal. Chem. 1963, 35, 1633.
5. Dunn, P. J.; Rees, C. W.; Slawin, A. M. Z.; Williams, D. J. J. Chem. Soc., Chem. Commun. 1989, 1134.
6. Gaussian 94 (Revision D.1), Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Gill, P. M. W.; Johnson, B. G.; Robb, M. A.;
Cheeseman, J. R.; Keith, T. A.; Petersson, G. A.; Montgomery, J. A.; Raghavachari, K.; Al-Laham, M. A.; Zakrzewski,
V. G.; Ortiz, J. V.; Foresman, J. B.; Peng, C. Y.; Ayala, P. Y.; Wong, M. W.; Andres, J. L.; Replogle, E. S.; Gomperts, R.;
Martin, R. L.; Fox, D. J.; Binkley, J. S.; Defrees, D. J.; Baker, J.; Stewart, J. P.; Head-Gordon, M.; Gonzalez, C.; Pople, J.
A. Gaussian, Inc., Pittsburgh PA, 1995.
17. Parr, R. G.; Yang, W. J. Am. Chem. Soc. 1984, 106, 4049.