Photooxidation of methyldithiepins into dithiepin carboxaldehydes in carbon tetrachloride

Article abstract of DOI:10.1021/ol990870p

(equation presented) Methyldihydrodithiepins are converted into the corresponding carboxaldehydes via a simple and efficient oxidation by oxygen in carbon tetrachloride. The reaction is not a radical chain process. The proposed mechanism involves photoind

Full text of DOI:10.1021/ol990870p

ORGANIC
LETTERS
1999
Vol. 1, No. 6
937-939
Photooxidation of Methyldithiepins into
Dithiepin Carboxaldehydes in Carbon
Tetrachloride
Yongqin Wan, Loren A. Barnhurst, and Andrei G. Kutateladze*
Department of Chemistry and Biochemistry, UniVersity of DenVer,
DenVer, Colorado 80208-2436
akutatel@du.edu
Received July 26, 1999
ABSTRACT
Methyldihydrodithiepins are converted into the corresponding carboxaldehydes via a simple and efficient oxidation by oxygen in carbon
tetrachloride. The reaction is not a radical chain process. The proposed mechanism involves photoinduced electron transfer from the starting
dithiepin to solvent (CCl4) followed by a series of events depicted in Scheme 1. The critical feature of the mechanism is that formation of
superoxide (O •-) is avoided, preventing direct oxidation of sulfur atoms.
2
Benzylic photooxidations in the presence of electron-transfer
(ET) sensitizers often provide a useful alternative to catalytic
oxidation of methyl-substituted aromatics into aldehydes.1
Potentially, the advantage of such photo ET-based reactions
over catalytic processes should be obvious for organosulfur
substrates that are capable of poisoning the catalyst. Most
commonly, however, photooxidation of sulfides and other
SII organic species with oxygen is directed at the sulfur atom,
leading to higher oxidation states of sulfur and/or C-S bond
cleavage.2 Although generation of cation-radicals derived
from an organosulfur compound is usually a simple and
efficient process, the limited choice of end results diminishes
the synthetic appeal of such reactions. Recently, we have
been focusing our research efforts on the development of
synthetically useful methods in which the generated sulfur-
centered cation-radical is facilitating chemical changes in the
molecule, besides S oxidation or C-S bond cleavage.
Previously, we reported a photo-ET-induced C-C bond
cleavage in hydroxyalkyl dithianes and its relevance to the
development of photoremovable protecting groups for car-
bonyls.3 We now report our findings on photo-ET-induced
oxidation of the methyl group in substituted dithiepins into
a carbonyl, which is a rare example of a sulfide moiety
surviving a photooxidation in the presence of molecular
oxygen.
We have found that irradiation (Pyrex filter) of 2-methyl-
3-phenyl-5,6-dihydro-1,4-dithiepin (1a) in CCl4 in the pres-
ence of oxygen furnished aldehyde 2a as the major product.4
Under nitrogen atmosphere no reaction occurred.
(1) Julliard, M.; Galadi, A.; Chanon, M. J. Photochem. Photobiol., A:
Chem. 1990 54, 79-90 and references therein. (b) Santamaria, J.; Jroundi,
R. Tetrahedron Lett. 1991 32, 4291-4.
We then synthesized a series of 3-aryl-substituted 2-meth-
yl-1,4-dithiepins 1 and studied their photooxidation into
(2) Most notably, such ET-initiated cleavages are used for photochemical
deprotection of dithioacetals: (a) Schmittel, M.; Levis, M. Synlett 1996,
315-316. (b) Tanemura, K.; Dohya, H.; Imamura, M.; Suzuki, T.;
Horaguchi, T. Chem. Lett. 1994, 965-8. (c) Kamata, M.; Murakami, Y.;
Tamagawa, Y.; Kato, M.; Hasegawa, E. Tetrahedron 1994, 50, 12821-8.
(d) Epling, G. A.; Wang, Q. Tetrahedron Lett 1992, 33, 5909-12. (e) Epling,
G. A.; Wang, Q. Synlett 1992, 335-6.
(3) McHale, W. A.; Kutateladze, A. G. J. Org. Chem. 1998, 63, 9924-
9931.
(4) All photooxidations were carried out with a medium-pressure Hg
UV source in Pyrex tubes with bubbling oxygen. A 10 mM amount of
1a-g in carbon tetrachloride was used, and irradiations were carried out
until complete conversion of the starting material was achieved.
10.1021/ol990870p CCC: $18.00 © 1999 American Chemical Society
Published on Web 08/25/1999

Our fluorescence quenching experiments show that CCl4
does quench the fluorescence of compounds 1 in degassed
acetonitrile solutions.
Table 1
Although the precise mechanism of this reaction is still
under investigation, our experimental observations led us to
the following mechanistic rationale (see Scheme 1). Carbon
Scheme 1
carboxaldehydes 2 in carbon tetrachloride. As is evident from
Table 1, most of the substrates we studied gave the
corresponding aldehydes. Although the isolated yields of the
aldehydes were moderate, the experimental simplicity and
the mere fact that after irradiation the CCl4 solution contains
>90% (by NMR) of the product greatly enhances the
preparative value of the method. Dithiepins 1 are easily
available from 2-methyl-1,3-dithiane 3 via the Corey-
Seebach dithiane-carbonyl adducts 45 with subsequent
dehydration accompanied by ring expansion.
tetrachloride does not absorb light above the Pyrex filter
cutoff, and therefore, it is the substrate 1 that is the primary
light absorbing species, with the λmax ranging from 305 to
350 nm. After the initial excitation, single electron transfer
occurs from the excited methyldithiepin moiety to a molecule
of solvent producing a cation-radical/anion-radical pair, {1•+/
Mechanistically significant results were obtained with
p-methoxycarbonyl (1f) and p-cyano (1g) species. These
dithiepins carrying an electron-withdrawing substituent were
virtually unreactive.
To rule out any radical chain process we ran a cross-
experiment by irradiating a mixture of approximately 30%
of (reactive) compound 1a and 70% of (unreactive) cyano-
substituted 1g. The production of 2a had slowed, and no 2g
was observed. If anything, abstraction of a hydrogen atom
from the methyl group in 1g is expected to be faster than it
is in the unsubstituted 1a. However, we did not detect any
2g in the reaction mixture.
•-
CCl4•-}. It is well established that CCl4 is unstable and
falls apart into chloride anion and trichloromethyl radical,6
which is then intercepted by O2. In a detailed kinetic study,
Alfassi and Mosseri7 had demonstrated that in carbon
•
tetrachloride CCl3 reacts with molecular oxygen with k2 ≈
1.12 × 105 M-1 s-1. The rate is not diffusion controlled, but
a relatively high estimated oxygen concentration (approxi-
mately 12 mM in CCl4 with partial oxygen pressure 1 atm8)
makes this pathway quite plausible in our case. 1•+ is then
•
deprotonated by Cl-, and 1• radical recombines with CCl3O2 ,
undergoes O-O bond cleavage, and produces the final
aldehyde 2. Judging by the experimental and theoretical data
accumulated in the literature, cation-radicals (e.g., of toluenes
or methyl-substituted naphthalenes) are generally found to
be superacids. For example, Arnold reported the pKa of the
Finally, a thermal reaction of 1a-g in the presence of
radical initiators (AIBN or dibenzoyl peroxide) did not
produce any amount of 2.
We also ran the oxidative photolysis in the presence of
methylene blue and detected no aldehyde formation. This
ruled out the involvement of singlet oxygen in the photo-
oxidation.
(6) Chateauneuf, J.; Lusztyk, J.; Ingold, K. U. J. Org. Chem. 1990, 55,
1061-1065 and references therein.
(7) Alfassi, Z. B.; Mosseri S. J. Phys. Chem. 1984, 88, 3296-3300.
(8) Battino, R. Solubility Data Series: Vol. 7, Oxygen and Ozone;
Pergamon: Oxford, England, 1981.
(5) For a review, see: Gro¨bel, B.-T.; Seebach, D. Synthesis 1977, 357-
402.
938
Org. Lett., Vol. 1, No. 6, 1999

toluene cation-radical in acetonitrile to be -12.9 This would
argue that chloride anion is an adequate base for deproto-
nating 1•+. We cannot entirely rule out a mechanism in which
1• radical reacts with O2, as long as the resulting hydro-
peroxide radical is not involved in chain propagation (as a
As a result, at no time does the reaction mixture contain
an oxygen nucleophile capable of attacking the S-atom in
•
D•+ directly. In our view, CCl3O2 is too electrophilic to react
with the cation-radical.
In addition to an easy synthetic approach to thiohetero-
cyclic carboxaldehydes, one of the possible synthetic ap-
plications of this reaction could be simple and efficient
synthesis of cis R,â-unsaturated aldehydes, which we il-
lustrate here using 2a as a starting material.11
•
possibility, it can be “capped” by CCl3 radical). To reiterate,
our cross-irradiation experiment with 1a and 1g seems to
have ruled out radical hydrogen abstraction from the methyl
group in the starting dithiepin.
In contrast to the data reported in the literature on ET-
sensitized oxidations of methyl-substituted aromatic com-
pounds into corresponding aldehydes,1 our attempt to convert
1a into 2a upon ET-sensitization with 1,4-dicyanonaphtha-
lene (DCN) in acetonitrile with bubbling oxygen failed.
Although 1a was slowly consumed, no trace of aldehyde 2a
was detected in the reaction mixture by NMR or GC-MS
analysis. Failure of DCN-sensitized oxidation to produce the
aldehyde in the dithiepin case can be explained in terms of
Foote’s classic scheme for photosensitized oxygenation of
sulfides via a non-singlet-oxygen mechanism:10
In conclusion, we have uncovered a novel photooxidation
of methyldithiepins into carboxaldehydes that features an
extremely simple experimental procedure. The unique aspect
of the mechanism is that after the initial electron transfer
there is no species of appreciable reducing potential present
•-
in the reaction mixture. The partition between CCl4
reducing molecular oxygen or decomposing into trichloro-
methyl radical and chloride anion favors the latter, thus
preventing superoxide formation, which in turn precludes
the undesirable S-directed oxidation.
Acknowledgment. Support for this research from the
National Science Foundation (CHE-9876389 Career Award)
is gratefully acknowledged.
A critical feature of this mechanistic scheme is that the
initially formed anion-radical of the ET-sensitizer im-
mediately reduces triplet oxygen into superoxide. The
reactions where D•+ is an organosulfur cation radical then
Supporting Information Available: 1H and 13C NMR
spectra for compounds 1a-g, 2a-e, and 4a-g and mass
spectra for compounds 2a-e. This material is available free
of charge via the Internet at http://pubs.acs.org
•-
always result in O2 nucleophilic attack at sulfur, leading
to S-oxidation. On the contrary, our scheme utilizes an
excited donor and a unique acceptor, which falls apart before
it has a chance to reduce O2:
OL990870P
(9) Nicholas, A. M. d. P.; Arnold, D. R. Can. J. Chem. 1982, 60, 2165-
2179
(10) Eriksen J.; Foote C. S.; Parker T. L. J. Am. Chem. Soc. 1977, 99,
6455-6456.
(11) Desulfurization in a similar dithiine system has recently been carried
out with Raney-Ni deactivated with acetic acid: Caputo, R.; Guaragna,
A.; Palumbo, G.; Pedatella, S. J. Org. Chem. 1997, 62, 9369-9371. In our
case, this technique did not work and we utilized the Raney-Ni/acetone
approach described in: Williams, J. R.; Tran, P. B. Synthesis 1988, 705-
706.
Org. Lett., Vol. 1, No. 6, 1999
939