Mechanism of sulfone formation in the reaction of sulfides and singlet oxygen: Intermediacy of S-hydroperoxysulfonium ylide

Article abstract of DOI:10.1021/ja960323o

H-D exchange was observed in the methyl group during the formation of sulfones in the reaction of dimethyl sulfide-d₆ or thioanisole-α,α,α-d₃ with singlet oxygen in aprotic solvents. No exchange was observed in the sulfoxides obtained. The proton in the sulfones was shown to come from adventitious water, since the oxidation of C₆H₅SCH₃ in the presence of D₂O lead to the formation of sulfones with monodeuteriation. The ¹⁶O₂-¹⁸O₂ tracer study demonstrated no oxygen scrambling in the sulfones. All the results indicate that the sulfones are formed intramolecularly via an intermediate possessing one activated proton exchangeable with trace water, a suggested structure for which is S-hydroperoxysulfonium ylides (RS⁺(OOH)CH₂^-). Kinetic isotope effects (k(H)/k(D) = 2-4) observed for methyl protons in the sulfone formation suggest that the rate-determining step is the intramolecular proton abstraction in the persulfoxides (RS⁺(OO^-)CH₃) generating S-hydroperoxysulfonium ylides. The conversion of the ylide intermediates to sulfones is shown to be facile on the basis of PM3 theoretical calculations.

Full text of DOI:10.1021/ja960323o

J. Am. Chem. Soc. 1996, 118, 7265-7271
7265
Mechanism of Sulfone Formation in the Reaction of Sulfides
and Singlet Oxygen: Intermediacy of S-Hydroperoxysulfonium
Ylide
Katsuya Ishiguro, Masaki Hayashi, and Yasuhiko Sawaki*
Contribution from the Department of Applied Chemistry, Faculty of Engineering,
Nagoya UniVersity, Chikusa-ku, Nagoya 464-01, Japan
ReceiVed January 30, 1996X
Abstract: H-D exchange was observed in the methyl group during the formation of sulfones in the reaction of
dimethyl sulfide-d6 or thioanisole-R,R,R-d3 with singlet oxygen in aprotic solvents. No exchange was observed in
the sulfoxides obtained. The proton in the sulfones was shown to come from adventitious water, since the oxidation
of C6H5SCH3 in the presence of D2O lead to the formation of sulfones with monodeuteriation. The 16O2-18O2
tracer study demonstrated no oxygen scrambling in the sulfones. All the results indicate that the sulfones are formed
intramolecularly via an intermediate possessing one activated proton exchangeable with trace water, a suggested
structure for which is S-hydroperoxysulfonium ylides (RS+(OOH)CH2-). Kinetic isotope effects (kH/kD ) 2-4)
observed for methyl protons in the sulfone formation suggest that the rate-determining step is the intramolecular
proton abstraction in the persulfoxides (RS+(OO-)CH3) generating S-hydroperoxysulfonium ylides. The conversion
of the ylide intermediates to sulfones is shown to be facile on the basis of PM3 theoretical calculations.
Reactions of singlet oxygen with sulfides affording sulfoxides
and sulfones have been extensively studied for over two decades
because of mechanistic and biological interest.1 Three important
reactions are the quenching of singlet oxygen and sulfoxide and
sulfone formations. The pioneering work was done by Foote
et al.,2 demonstrating that zwitterionic persulfoxides are an
appropriate intermediate on the basis of trapping experiments.
The structure of the intermediates and the overall oxygenation
mechanism are, however, not yet clarified thoroughly in spite
Scheme 1
revealed that the photo-oxidation of sulfides with 18O2 yielded
sulfones with retention of the two oxygen atoms, indicating an
unimolecular rearrangement of 2 to sulfones (eq 2).6 The yields
1
of numerous studies on the O2 reaction of sulfides.1b
In the reaction of sulfides with singlet oxygen, widely
accepted is the intermediacy of zwitterionic persulfoxides (1)
as the primary intermediate (PI). Persulfoxide intermediates
of sulfones were not affected by the excess amounts of sulfides
or by the addition of diphenyl sulfide and sulfoxide, which
suggests that the sulfone-forming intermediate is different from
SI in Scheme 1. A detailed theoretical study by Jensen7 could
not give a definitive conclusion for the conversion of 1 to 2 or
of 2 to sulfones.
The third important reaction of sulfides and singlet oxygen
is C-S cleavage to afford fragmentation products as observed
for sulfides bearing active R-C-H bonds, e.g., benzyl,8 9-fluo-
renylethyl,9 and cyclic sulfides.10 In these reactions a third
peroxidic intermediate, S-hydroperoxysulfonium ylide (3), has
been believed to be formed by the intramolecular abstraction
of R-proton (Scheme 2). Ylide intermediates from cyclic
sulfides10 were also suggested as active species for epoxidations
of olefins.
in aprotic solvents oxidize sulfoxides more efficiently than
sulfides and behave as nucleophilic oxygen-atom transfer
agents.3 From competitive trapping experiments, Foote and co-
workers4 proposed a reaction mechanism involving the inter-
conversion of 1 into a secondary peroxidic intermediate (SI) as
shown in Scheme 1. Since the second species oxidizes sulfides
electrophilically,5 cyclic thiadioxiranes (2) were proposed for
this intermediate.4 On the other hand, 18O-tracer experiments
X Abstract published in AdVance ACS Abstracts, July 15, 1996.
(1) (a) Foote, C. S. In Singlet Oxygen; Wasserman, H. H., Murray, R.
W., Ed.; Academic Press: New York, 1979; p 139. (b) Ando, W.; Takata,
T. In Singlet O2; Frimer, A. A., Ed.; CRC Press: Boca Raton, FL, 1985;
Vol. 3, Chapter 1, p 1. (c) Organic Peroxide; Ando, W., Ed.; Wiley: New
York, 1992 and references cited therein.
(2) Foote, C. S.; Peters, J. W. J. Am. Chem. Soc. 1971, 93, 3795.
(3) Sawaki, Y.; Ogata, Y. J. Am. Chem. Soc. 1981, 103, 5947.
(4) Liang, J.-J.; Gu, C.-L.; Kacher, M. L.; Foote, C. S. J. Am. Chem.
Soc. 1983, 105, 4717.
(5) Ando, W.; Kabe, Y.; Miyazaki, H. Photochem. Photobiol. 1980, 31,
191.
(6) Watanabe, Y.; Kuriki, N.; Ishiguro, K.; Sawaki, Y. J. Am. Chem.
Soc. 1991, 113, 2677.
(7) Jensen, F. J. Org. Chem. 1992, 57, 6478.
(8) Corey, E. J.; Quanne`s, C. Tetrahedron Lett. 1976, 4263.
(9) Ando, W.; Nagashima, T.; Saito, K.; Kohmoto, S. J. Chem. Soc.,
Chem. Commun. 1979, 154.
(10) Akasaka, T.; Sakurai, A.; Ando, W. J. Am. Chem. Soc. 1991, 113,
2696.
S0002-7863(96)00323-X CCC: $12.00 © 1996 American Chemical Society

7266 J. Am. Chem. Soc., Vol. 118, No. 31, 1996
Ishiguro et al.
1
a
Table 2. Effect of H2O on the O2 Oxidation of Sulfide 5b-d3
Scheme 2
yields, %
isotope ratios in 7bb
added
[H2O], mM
6b
7b
7b-d2
7b-d3
0
27.8
55.6
278
1.1c
1.4c
1.7c
3.3c
0.47
0.41
0.39
0.43
100
100
100
100
18.0 ( 0.2
18.5 ( 0.2
16.3 ( 0.2
31.1 ( 0.2
a Irradiation of 0.05 M PhSCD3 (5b-d3) and 0.1 mM TPP in benzene
under O2 atomosphere for 20 h. b Isotope ratios as determined by GC-
MS. c 6b-d2:6b-d3 ) <1:>99.
1
Table 1. Products at the Initial Stage of the O2 Oxidation of
Sulfide 5a-d6
a
yields, %
isotope ratios in 7ab
irradiation
time, min
6a
7a
7a-d5
7a-d6
The effect of the addition of water in the photo-oxidation of
5b-d3 was examined. As shown in Table 2, the addition of
H2O resulted in the acceleration of sulfoxide formation but did
not affect the sulfone formation; sulfone 7b-d3 without H-D
exchange was increased only in the presence of large amount
of water. No effect of water on the sulfone yields and on the
H-D exchange suggests that the H-D exchange itself is not
so kinetically important.
20
40
60
80
0.3c
0.9c
2.5c
5.2c
0.13
0.24
0.34
0.43
100
100
100
100
11.4 ( 0.3
19.1 ( 0.3
35.2 ( 0.2
52.2 ( 0.2
a Irradiation (>400 nm) of 0.1 M (CD3)2S and 0.1 mM TPP in
benzene under O2. b Isotope ratios as determined by GC-MS. Means
of three to five determinations. c 6a-d5:6a-d6 ) <0.5:>99.5.
The photo-oxidation of undeuteriated thioanisole (5b-d0) in
the presence of 0.05% D2O (27.6 mM) afforded the monodeu-
teriated sulfone (eq 4), the ratio of which was 7b-d0:7b-d1 )
We became interested in the intervention of persulfoxides
and/or thiadioxiranes for the formation of sulfones and C-S
cleavages. Deuterium and 18O-tracer studies revealed that at
low conversions the sulfone formation accompanies, unexpect-
edly, an exchange of one R-hydrogen atom. Herein we
summarize our mechanistic study on the 1O2 reaction of sulfides,
suggesting that sulfonium ylides (3) are intermediates for C-S
cleavage and sulfone formation at early stages of the reaction.
Results
100:48.2 ( 0.1. It is interesting to note that only one proton in
7b was exchanged with D2O and no deuterium was incorporated
in the sulfoxide (6b-d0, C6H5SOCH3) and the recovered sulfide.
These results clearly indicate that the sulfone formation proceeds
via an intermediate with only one activated proton exchangeable
with water.
Singlet Oxygenation of Deuteriated Sulfides. Irradiation
of oxygen-saturated benzene solution of 0.1 M perdeuteriated
dimethyl sulfide (5a-d6) and 0.1 mM meso-tetraphenylporphyrin
(TPP) at wavelengths longer than 400 nm at room temperature
yielded dimethyl sulfoxide (6a) and dimethyl sulfone (7a). The
16O2-18O2 Tracer Study. The formations of sulfoxide either
by unimolecular reaction or by secondary oxidation of sulfoxide
may be distinguished by following the isotope scrambling of
oxygen gas.6 Consequently, the photo-oxidation of 5a-d6 with
a mixture of 16O2 and 18O2 (16O2:16O18O:18O2 ) 100:0.79:11.9)
was examined in the absence or presence of added water. The
sulfones with six different isotope contents, 7a-d5 and 7a-d6
with 16O2, 16O18O, and 18O2 were analyzed by GC-MS, and
the results are summarized in Table 3. The isotope ratios of
oxygen atoms in 7a-d5 were always close to those of the starting
oxygen gas, indicating that the d5 sulfone with the H-D
exchange was formed with incorporation of one oxygen
molecule (i.e., retention).
deuterium contents of the products were determined by GC-
MS spectroscopy. As shown in Table 1, a mixture of 7a-d5
and 7a-d6 was obtained. The former d5 sulfone was the major
component at the initial stage of reaction, and no sulfones
carrying more than two hydrogens were obtained. It is
interesting to note that such an H-D exchange was not observed
for sulfoxide 6a or for the recovered sulfide 5a. After the
irradiation, no deuterium exchange occurred after standing in
the dark for 12 h even in the presence of 0.1 M H2O, suggesting
that only one R-deuterium was replaced during the photo-
oxidation. The amount of nonexchanged sulfone (7a-d6) was
increased linearly with progress of the photo-oxidation and was
almost zero at initial time. This fact indicates that only 7a-d5
was the primary product while 7a-d6 was formed by the
secondary oxidation of sulfoxide 6a-d6. This was confirmed
by 16O2-18O2 tracer experiments as described later.
In contrast, the oxygen isotope ratios in 7a-d6 exhibited the
scrambling of oxygen atoms; that is, the doubly labeled 18O-
18O was incorporated predominantly as scrambled 16O-18O.
When a small amount of H2O was added to complete the H-D
exchange, the oxygen isotope ratios in 7a-d6 became close to
those calculated for the scrambling mechanism. Thus, in the
presence of sufficient water, the two types of sulfones could be
characterized clearly as depicted in Scheme 3. That is, all the
7a-d6 is yielded by the secondary oxidation of sulfoxide 6a-
11
d6 while the sulfone 7a-d5 is formed intramolecularly with
incorporation of the two oxygen atoms of oxygen molecule.
Singlet Oxygenation of Di-tert-butyl Sulfide. To see
whether or not the H-D exchange is essential for the intramo-
lecular sulfone formation, the 1O2 oxidation of sulfides without
R-hydrogen has been examined. Although diaryl sulfides and
phenyl tert-butyl sulfide were unreactive toward singlet oxygen,
di-tert-butyl sulfide (8) was found to be oxidized slowly.6
Similarly, the photo-oxidation of 0.05 M thioanisole-d3 (5b-
d3) in benzene for 20 h yielded the corresponding sulfoxide
without H-D exchange (i.e., 6b-d3) and predominantly sulfone
with H-D exchange (i.e., 7b-d2:7b-d3 ) 100:18). The photo-
oxidation of 5a-d6 in acetonitrile sensitized with 0.1 mM
Methylene Blue (MB) gave a similar result.
The Effect of Water. Adventitious H2O in the solvents is a
conceivable source of the hydrogen incorporated in the sulfone.
(11) Clennan, E. L.; Zhang, H. J. Org. Chem. 1994, 59, 7952.

1
Sulfone Formation in the Reaction of R2S and O2
J. Am. Chem. Soc., Vol. 118, No. 31, 1996 7267
1
a
Table 3. 16O2-18O2 Tracer Study on the Sulfone Formation in the O2 Oxidation of 5a-d6
isotope ratios in 7ac
yields, %
7a-d5
16O-18O
7a-d6
16O-18O
added
solvent [H2O],b mM sulfoxide 6a sulfone 7a 16O-16O
18O-18O 16O-16O
18O-18O
d
C6H6
0
2.3
2.5
3.2
0.13
0.12
0.17
100
100
100
: 0.92 ( 0.11 : 10.82 ( 0.22 :
: 0.93 ( 0.24 : 10.62 ( 0.21 :
: 0.76 ( 0.04 : 10.02 ( 0.08 :
23.95 ( 0.34 : 2.64 ( 0.04 : 0.97 ( 0.03
(100 : 11.0 ( 0.2
: 4.05 ( 0.14)e
16.58 ( 0.27 : 3.00 ( 0.45 : 0.33 ( 0.04
(100 : 18.1 ( 2.7
: 1.99 ( 0.24)e
21.32 ( 0.03 : 4.02 ( 0.05 : 0.18 ( 0.01
(100 : 18.9 ( 0.2
: 0.84 ( 0.05)e
9.3
18.5
CH3CNf
0
1.4
1.7
2.3
0.22
0.19
0.18
100
100
100
100
: 1.27 ( 0.26 : 11.01 ( 0.24 :
: 1.07 ( 0.15 : 10.18 ( 0.15 :
: 1.29 ( 0.16 : 11.50 ( 0.33 :
: 0.79 ( 0.16 : 11.88 ( 0.37
36.17 ( 0.99 : 6.87 ( 0.16 : 0.71 ( 0.06
(100 : 19.0 ( 0.7
: 1.96 ( 0.17)e
40.30 ( 0.25 : 8.83 ( 0.22 : 0.47 ( 0.01
(100 : 21.9 ( 0.6
: 1.17 ( 0.03)e
9.3
18.5
31.86 ( 0.48 : 6.93 ( 0.15 : 0.35 ( 0.02
(100
: 21.8 ( 0.6
: 1.10 ( 0.06)e
calcd for retention
calcd for scrambling
100
: 23.8 ( 0.9
: 1.42 ( 0.06
a Irradiation (>400 nm) of 0.1 M (CD3)2S under oxygen gas; 16O2:16O18O:18O2 ) 100:0.79 ( 0.16:11.88 ( 0.37. b Amounts of added water.
c Isotope ratios for 7a-d1 and -d6 with 16O-16O, 16O-18O, and 18O-18O as determined by GC-MS. d Irradiation in the presence of TPP for 70 min.
e Normalized ratios in 7a-d6. f Irradiation in the presence of 0.1 mM MB for 10 min.
Scheme 3
unimolecularly as stated above, the estimation of isotope effect
is not straightforward since 7a-d0 can be formed by both intra-
and intermolecular pathways. Provided that the secondary
oxidation of sulfoxides by peroxidic intermediate 1 has also no
isotope effect, the kinetic isotope effect in the unimolecular
sulfone formation, kH/kD, can be calculated by [(7a-d0) - (7a-
d6)]/(7a-d5). As shown in Table 4, however, the values of kH/
kD for the sulfone formation were increased gradually with the
irradiation time, which is probably due to the inclusion of isotope
effect on the secondary reactions. Thus, the kinetic isotope
effect for the sulfone formation was obtained as 3.0 ( 0.4 by
the extrapolation of these values to zero time.
Such a complexity could be eliminated for the 1O2 oxidation
of PhSCH3 and PhSCD3, since the isotope effects obtained were
almost constant throughout the reaction as shown in Table 5.
The resulting kinetic isotope effects were kH/kD ) 2.1 ( 0.2,
2.8 ( 0.1, and 4.1 ( 0.2 in C6H6, C6H6-Et2O (1:1), and
CH3CN, respectively.
Thus, all the results indicate that the rate-determining
R-proton abstraction is involved in the intramolecular sulfone
formation. The apparent constant isotope effects probably
reflect the unimportance of the secondary oxidation of sulfoxides
for the case of thioanisole.
Therefore, the sulfone formation was investigated by tracer
experiments. The TPP-sensitized photo-oxidation of 100 mM
di-tert-butyl sulfide with 16O2-18O2 (16O2:16O18O:18O2 ) 100:
0.79:11.9) in benzene for 6 h afforded the corresponding
sulfoxide 9 and sulfone 10 in 2.3% and 0.24% yields,
respectively. The isotope ratios in the sulfone (10-16O2:10-
16O18O:10-18O2 ) 100:18.2 ( 0.2:1.2 ( 0.1) showed the
complete scrambling of oxygen atoms, indicating that the sulfone
was formed via the secondary reaction. Thus, it is apparent
that sulfides carrying no R hydrogens are not converted to
sulfones directly.
Discussion
Persulfoxide intermediates (1) in the 1O2 reaction of sulfides
have been characterized by their trapping experiments.2-5 The
intermediacy of thiadioxiranes (2) is not clarified adequately
although they have been sometimes written as the secondary
intermediate as shown in Scheme 1. Previously, we proposed
that in the 1O2 reaction of sulfides sulfones were formed by an
unimolecular rearrangement of 2 (eq 2).6 The present results,
however, clearly indicate that the sulfone formation necessarily
accompanies the exchange of an R proton with adventitious
water. This unexpected result denies such a simple mechanism
of formation and rearrangement of thiadioxiranes (eq 2).
A direct exchange of protons between water and an R-proton
of persulfoxides (1′) could be excluded since no proton exchange
was observed in resulting sulfoxides and recovered sulfides. The
selective one proton exchanged in the sulfone formation suggests
an intervention of ylide-type intermediates (3) possessing the
carbanionic structure of S-CH2-. As stated in the introduction,
S-hydroperoxysulfonium ylide (3) has been written as an
intermediate for C-S cleavage in the 1O2 oxidation of sulfides.8,9
Kinetic Isotope Effect. The kinetic isotope effect of
deuterium has been investigated to clarify whether the abstrac-
tion of R protons leading to the sulfone formation is the rate-
determining step or not. Competitive photo-oxygenation of a
1:1 mixture of (CH3)2S and (CD3)2S yielded three different
sulfones, 7a-d0, -d5, and -d6, together with two sulfoxides, 6a-
d0 and -d6 (eq 5). While the product ratios of the sulfoxides
were ∼1:1, showing no isotope effect, the formation of sulfone
7a-d0 was clearly faster that of 7a-d5. While 7a-d5 is formed

7268 J. Am. Chem. Soc., Vol. 118, No. 31, 1996
Ishiguro et al.
1
Table 4. Kinetic Isotope Effects on the Competitive O2 Oxidation of Sulfides 5a-d0 and 5a-d6 in Benzenea
isotope ratiosb
yields, %
sulfoxide
sulfone
7a-d6
irradiation
time, min
c
d
sulfoxide 6a
sulfone 7a
6a-d0
6a-d6
kH/kD
7a-d0
7a-d5
kH/kD
20
40
60
80
0.39
0.96
2.50
4.40
0.18
0.34
0.48
0.60
e
e
100
100
e
e
e
e
100
100
100
100
30.5 ( 1.6
28.0 ( 0.6
25.4 ( 0.1
23.0 ( 0.4
1.9 ( 0.1
3.9 ( 0.1
7.2 ( 0.1
11.5 ( 0.3
average:
3.22 ( 0.17
3.43 ( 0.07
3.65 ( 0.01
3.85 ( 0.07
3.0 ( 0.4f
107 ( 3
104 ( 2
average:
0.93 ( 0.03
0.96 ( 0.02
0.95 ( 0.04
a Irradiation of (CH3)2S and (CD3)2S (50 mM each) and 0.1 mM TPP in benzene under O2. b Isotope ratios as determined by GC-MS. c Kinetic
isotope effects defined as (6a-d0)/(6a-d6). d Kinetic isotope effects defined as [(7a-d0) - (7a-d6)]/(7a-d5). e Not determined. f The kinetic isotope
effect obtained by the extrapolation at zero time.
1
a
Table 5. Kinetic Isotope Effects on the Competitive O2 Oxidation of Sulfide 5b-d0 and 5b-d3
isotope ratiosb
yields, %
sulfoxide
sulfone
7b-d3
irradiation
time, h
c
d
solvent
sulfoxide 6b sulfone 7b 6b-d0
6b-d3
kH/kD
7b-d0
7b-d2
kH/kD
e
C6H6
8
12
16
18
20
0.28
0.62
0.77
1.20
1.40
0.20
0.30
0.34
0.42
0.46
100
100
100
100
100
103 ( 2 0.97 ( 0.02
104 ( 1 0.96 ( 0.01
106 ( 1 0.94 ( 0.01
103 ( 1 0.97 ( 0.01
105 ( 2 0.95 ( 0.02
average: 0.96 ( 0.04
104 ( 1 0.96 ( 0.01
105 ( 2 0.95 ( 0.02
average: 0.96 ( 0.02
103 ( 1 0.97 ( 0.01
98 ( 1 1.02 ( 0.01
106 ( 1 0.94 ( 0.01
105 ( 1 0.95 ( 0.01
average: 0.97 ( 0.03
100
100
100
100
100
46.7 ( 1.1 2.4 ( 0.1 2.09 ( 0.05
46.1 ( 0.6 2.9 ( 0.2 2.11 ( 0.03
46.5 ( 0.5 3.8 ( 0.1 2.07 ( 0.02
44.7 ( 0.2 6.3 ( 0.2 2.10 ( 0.01
43.1 ( 0.2 7.2 ( 0.1 2.15 ( 0.01
average: 2.1 ( 0.02
33.1 ( 0.6 4.0 ( 0.4 2.90 ( 0.05
34.8 ( 1.5 6.0 ( 0.9 2.70 ( 0.12
average: 2.8 ( 0.1
25.0 ( 0.2 1.4 ( 0.3 3.94 ( 0.03
22.8 ( 0.6 2.7 ( 0.5 4.27 ( 0.11
23.4 ( 0.5 4.2 ( 0.2 4.09 ( 0.09
23.0 ( 0.2 6.8 ( 0.3 4.05 ( 0.04
average: 4.1 ( 0.2
C6H6-Et2O (1:1)e
8
12
0.39
0.57
0.25
0.35
100
100
100
100
CH3CNf
2
4
8
0.70
1.50
4.10
7.10
0.10
0.19
0.31
0.47
100
100
100
100
100
100
100
100
12
a Irradiation of PhSCH3 and PhSCD3 (50 mM each) under O2. b Isotope ratios as determined by GC-MS. c Kinetic isotope effects defined as
(6b-d0)/(6b-d3). d Kinetic isotope effects defined as [(7b-d0) - (7b-d3)]/(7b-d2). e Irradiation in the presence of 0.1 mM TPP. f Irradiation in the
presence of 0.1 mM MB.
The importance of R-proton abstraction for the sulfone
formation is reflected in the kinetic isotope effect of kH/kD
)
2-4 (Tables 4 and 5). The isotope effect is understandable by
assuming the competition between the hydrogen abstraction (eq
6a) and the regeneration of sulfides (eq 6b). The latter
The detailed mechanism of these transformations may be
followed by theoretical calculations. Although ab initio calcula-
tions on sulfide-O2 adducts were shown to be highly sensitive
to the level of theory and frequently mislead to erroneous
results,6,14 a semiempirical RHF-PM3 method15 was found to
give relatively good results for the qualitative purpose. This
method affords the thiadioxirane (2) as a stable unsymmetrical
structure almost isoenergetic to persulfoxide 1, and the geometry
and energy of transition states for the inversion of 2 (TS1, ∆Hq
) 2.7 kcal/mol) or for the cyclization of 1 (TS2, ∆Hq ) 17.6
kcal/mol) are calculated in good agreement with those (∆Hq )
2.7 and 19.7 kcal/mol, respectively) by the MP4/6-311G(2d)/
/MP2/6-31G* calculations reported by Jensen.7
dissociation of persulfoxides to yield sulfides and oxygen is
well-known,12 and hence, the observation of intermolecular
kinetic isotope effect is explicable in terms of competitive
reactions of (6a) and (6b). The tracer experiments using doubly
labeled oxygen 18O2 in Table 3 demonstrate that the sulfone
formation with H-D exchange is produced solely by the
retention mechanism of the two 18O atoms. Hence, it is apparent
that S-hydroperoxysulfonium ylides (3) once formed are selec-
tively converted unimolecularly to sulfones according to eq 6a.
PM3 Calculation. Deuterium and 18O-tracer experiments
supported the intermediacy of sulfonium ylide 3 for the sulfone
formation as stated above. The conversion of ylides 3 to
sulfones involves an 1,2-shift of an OH group from the O to S
atom (eq 7), affording an S-ylide of sulfinic acid (11), i.e., a
tautomer of sulfone. A similar 1,2-shift of alkyl groups is
known in the rearrangement of sulfinates to sulfones.13 The
observed H-D exchange occurs either in the HOO group in 3
or HO group in 11.
The structures and energies (heats of formation, ∆H) calcu-
lated by the RHF-PM3 method for the suggested intermediates
and transition states are summarized in Figure 1. The S-
hydroperoxysulfonium ylide (3) was calculated slightly lower
in energy than dimethyl persulfoxide (1). The activation energy
for the intramolecular proton transfer (1 to 3, TS3) was 18.9
kcal/mol, which was comparable to that for the isomerization
of 1 to 2 (TS2). The OH migration from O to S in 3 affording
11 could be traced smoothly and was found to be highly
exothermic (-59.7 kcal/mol), the activation energy for TS4 from
3 being 25.3 kcal/mol. Thus, the present PM3 calculations
(12) Gu, C.-L.; Foote, C. S. J. Am. Chem. Soc. 1982, 104, 6060.
(13) Tillet, J. G. Chem. ReV. 1976, 76, 747.
(14) Jensen, F.; Foote, C. S. J. Am. Chem. Soc. 1988, 110, 2368.
(15) Stewart, J. J. P. J. Comput. Chem. 1989, 10, 209, 221.

1
Sulfone Formation in the Reaction of R2S and O2
J. Am. Chem. Soc., Vol. 118, No. 31, 1996 7269
Figure 1. Structures and heats of formation for the intermediates and transition states on the potential energy surface of Me2S + O2 as calculated
by RHF-PM3.
suggest that the three transition states TS2-4 are of ap-
proximately equal energy and the conversion of 3 to 11 is highly
exothermic.
Formation of 3 from 1. We have reported that since the
sulfone formation was not affected by an efficient trapping agent
for 1, the possible formation of 2 was not from 1 but via a
HO vs HOO Migration in 3. The S-hydroperoxysulfonium
ylides (3) have been proposed as the intermediate leading to
C-S fission as shown in Scheme 2. In the present cases of
dimethyl sulfide and thioanisole, however, the overall yields
concerted cycloaddition of 1O2 to sulfides.6 On the other hand,
the present results show the rate-determining generation of
S-hydroperoxysulfonium ylide (3) as the sulfone-forming in-
termediate, which is explicable by the intramolecular proton
of sulfoxides and sulfones are always more than 90%, the
transfer in persulfoxide 1. The present PM3 calculations
supported the conversion of 1 to 11 as a reasonable pathway to
yield sulfones and the intermediate sulfonium ylide 3 is
S to C atom affording 12 (a dashed curve in Figure 1) was less
calculated to be more stable than persulfoxide 1. It is not
fragmentation being only a minor side reaction. According to
the PM3 calculation, the migration of the OOH group from the
exothermic (-30.6 kcal/mol) but the transition state (TS5) was
surprising that we previously reported that the sulfone-forming
∼10 kcal/mol lower in energy than that of TS4. The TS
pathway was insensitive to trapping agents.6 Even in the
calculations at this level may have only qualitative meaning,
presence of 0.1 M Ph2SO, only a 10% change in sulfone yield
since the prediction of the predominant OOH migration is
apparently inconsistent with the experimental results.
is anticipated.16
An alternative mechanism is a concerted proton abstraction/
addition reaction of 1O2 with sulfides, as in the well-known ene
reaction with olefins affording allylic hydroperoxides. Such a
ene-type transition state (see 13 in eq 8) could not be found on
the potential energy surface as calculated by PM3. However,
since the reliability of the calculation at this level is not so high,
we cannot eliminate such a mechanism. More detailed ab initio
calculations with large basis sets including electron correlation
are desired.
The C-S fragmentation is known to become a major pathway
in the singlet oxygenation of benzyl or cyclic sulfides.8-10 Thus,
the reaction pathway for either HO or HOO migration seems
to depend on the substituents, which may be explained in terms
of the stability and conformation of 3. The PM3-calculated
conformations of 3 and 11 show that the H2C- group is almost
orthogonal to S-OOH and S-OH bonds, respectively, indicat-
ing that the ylide carbanion interacts with the electron-deficient
S-O bonds. The calculation also shows the twisting of the
H2C- group during the transformation from 3 to 11 (see 14 in
eq 9), which may be the driving force on the HO migration.
(16) By the steady-state treatment of Scheme 3 in a manner similar to
that by Foote and co-workers,4 the ratio of sulfone yields in the presence
vs absence of Ph2SO can be formulated as 1/(1 + R[Ph2SO]), where R )
k
SO/(kx + kq + kPT). When we apply the relative rate constants for diethyl
sulfide, kq/kx ) 45 (in acetone)11 and kx/kSO ) 0.023 M (in benzene),4 and
the ratio of sulfone/sulfoxide yields in the present study, [R2SO2]/[R2SO]
) kPT/2kx ) 0.4, R ) 1.06 is obtained. Thus, the presence of the trapping
agent results in only 10% decrease, even in the case of [Ph2SO] ) 0.1 M,
in the sulfone yields.

7270 J. Am. Chem. Soc., Vol. 118, No. 31, 1996
Ishiguro et al.
Chromosorb WAW or a 0.2 mm × 25 m capillary column of Silicon
OV-1 (J & W Scientific, DB-1).
For the cases of the ylides from benzyl or 9-fluorenyl sulfides,
the π conjugation of the R-carbanion with aryl groups may result
in the decreased interaction with S-O bonds, and the twisting
of the ylide is evidently restricted in cyclic sulfides. For these
cases, the sulfone formation via 3 is practically inhibited.
The kinetic isotope effect observed on the sulfone formation
may be compared with that of C-S cleavage since a common
intermediate 3 is involved. We have obtained a similar isotope
effect (i.e., kH/kD ) 4.3) in the benzaldehyde formation from
1O2 and PhSCHDPh in acetonitrile,17 which is comparable to
that of the sulfone formation from 5b in the same solvent, kH/
kD ) 4.1. The observation of a similar kinetic isotope effect is
quite interesting and supports the intervention of common
intermediate 3.
The Secondary Intermediate (SI) in the Sulfoxide Forma-
tion. Thiadioxiranes (2) are shown not to be the intermediate
leading to sulfones but may participate in the sulfoxide formation
as the secondary intermediate (SI) as depicted in Scheme 1.4
The oxidation via SI becomes predominant pathway at low
temperature,12 indicating that the activation energy for the
conversion of 1 to SI is lower than those of other competing
processes, e.g., the intramolecular proton transfer leading to the
sulfone formation. However, as already pointed out by Jensen,7
the activation energy for the conversion of 1 to 2 may not be
so low. Then, the hypothetical intermediacy of 2 as SI seems
not to be appropriate at present.
Materials. Benzene was dried by distillation over sodium benzo-
phenone. Acetonitrile was distilled from phosphorus pentaoxide.
Dimethyl sulfide (5a, Tokyo Kasei) and dimethyl sulfide-d6 (5a-d6,
Aldrich, 99.9% atom D) were employed as received. Thioanisole (5b,
Tokyo Kasei) and di-tert-butyl sulfide (8, Tokyo Kasei) were distilled
before use. Methylene Blue (MB, Tokyo Kasei) and tetraphenylpor-
phyrin (TPP, Tokyo Kasei) were used without further purification. 18O2
(99% pure) gas from CEA was diluted with O2 gas.
Thioanisole-r,r,r-d3 (5b-d3). To 3.5 g of chlorosulfonic acid (30
mmol), was added 1.0 mL of methanol-d4 (CEA, 99.8 atom % D, 24.6
mmol) dropwise at 0 °C. After 5 min of stirring, a solution of 3.0 mL
of thiophenol (29.2 mmol) and 10 g of sodium hydroxide (25 mmol)
in 50 mL of water was added at this temperature. The mixture was
refluxed for 3 h, and the product was extracted with ether. The extract
was washed with brine and dried over anhydrous sodium sulfate.
Evaporation of the solvent and chromatographic separation on silica
gel (haxane) gave thioanisole-R,R,R-d3 (1.2 g) as a colorless liquid in
38% yield. The purity of deuterium as determined by mass spectros-
copy was 99.5 atom % D.
Typical Procedure of the Photo-oxidation. A 10-mL benzene
solution of 0.05 M dimethyl sulfide-d6 (5a-d6) and 0.1 mM tetra-
phenylporphyrin (TPP) in a 25-mL Pyrex test tube with a septum rubber
cap was purged with oxygen and was irradiated at 20 ( 2 °C with a
300-W medium-pressure Hg lamp through a 5% NaNO2 filter solution
(i.e., >400 nm). The formation of sulfones and sulfoxides was
determined at appropriate intervals of time by GLC.
Determination of Deuterium Contents in the Photoproducts. The
resulting solution was concentrated and analyzed by GC-MS. For 6a
and 7a, the mass spectral data around the parent peak (M) were easily
obtained by an EI method with 70 eV ionization voltage. Since their
retention times on GC were slightly different depending on the number
of deuteriums, the mass intensities should be, for the quantitative
analyses, integrated by mass chromatography throughout the GC peaks.
A large number of mass data (up to ∼500 points per peak) were
collected by sampling of 0.05 s intervals. The differences in mass
intensities of parent peak (M) between perdeuteriated and nondeuteriated
dimethyl sulfoxides or sulfones were not negligible. Then, the relative
mass peak intensities were determined from a 1:1 mixture of 6a-d0
and 6a-d6 and that of 7a-d0 and 7a-d6, i.e., (6a-d0/6a-d6) ) 0.883 and
(7a-d0/7a-d6) ) 0.819. As the pure sample of 7a-d5 was not available,
the relative intensity for 7a-d5 could not be determined directly but
was calculated from that for 7a-d6 by assuming that the mass intensity
The sulfoxide formation in aprotic solvents was found to be
significantly affected by a trace amount of adventitious water.
Thus, a possible structure for SI may be an adduct of water
with persulfoxide 1.14,18 A previous trapping study, however,
has indicated that the reactivity of the intermediate in protic
solvents is different from those of 1 or SI in aprotic solvents,
and the quite complex solvent effects have been suggested by
Clennan.19 A further detailed study on the effect of protic
solvents may clarify the structure of SI.
Conclusion
The sulfone formation at earlier stage accompanied an H-D
exchange of the R-methyl group in the reaction of sulfides with
singlet oxygen in aprotic solvents. In contrast, no exchange
was observed in the sulfoxides obtained. The exchanged proton
in sulfone was shown to come from adventitious water, and
the doubly labeled 18O-tracer study indicated no scrambling of
two oxygen atoms. A significant kinetic isotope effect was
observed for the R-deuteriated sulfides, suggesting the intra-
molecular proton abstraction in the persulfoxides (e.g.,
RS+(OO-)CH3) being the rate-determining step. These results
clearly indicate that the sulfones are formed intramolecularly
via S-hydroperoxysulfonium ylides (RS+(OOH)CH2-). The
conversion of the sulfonium ylides to sulfones was shown to
be facile on the basis of PM3 calculations.
5
might depend on the number of deuteriums, [1 - (7a-d0/7a-d5)] ) /6
of [1 - (7a-d0/7a-d6)], i.e., (7a-d0/7a-d5) ) 0.849.
The mass spectrum of resulting sulfoxide 6a was identical to that
of authentic 6a-d6 obtained by the m-chloroperbenzoic acid oxidation
of 5a-d6 within experimental error. The sulfone 7a was shown to be
formed as a mixture of 7a-d5 and 7a-d6; the integrated peak areas were
obtained by mass chromatography monitored at m/e ) 99 and 100,
corresponding to the parent ions of C2D5HSO2 and C2D6SO2, respec-
tively, the latter data being corrected by abstracting the M + 1 value
of natural abundance in C2SO2. After correction of the relative mass
intensity, 7a-d5 vs 7a-d6, the isotope ratios in the sulfone 7a were
obtained. Three to five determinations were averaged, and the results
are summarized in Table 1.
The products from the photo-oxidation of 5b-d3 in the absence or
presence of 0.05-0.5% H2O were analyzed similarly from integrated
mass peak areas at the parent ions (M). Since the mass intensities of
parent peaks were almost independent of the number of deuteriums in
the sulfoxides 6b and the sulfones 7b, no correction was made on the
relative mass intensity. The mass data were corrected by abstracting
the M + 1 value of natural abundance, and the results averaged on
three determinations are listed in Table 2.
Isotope ratios of the products on the photo-oxidation of PhSCH3
(5b-d0) in the presence of D2O were analyzed similarly. The irradiation
of 0.05 M PhSCH3 (5b-d0) and 0.1 mM TPP in 0.05% D2O (27.6 mM)/
benzene under O2 atmosphere for 20 h afforded 6b and 7b in 1.5%
and 0.58% yields, respectively. The isotope ratio of 7b as determined
by GC-MS was PhSO2CH3 (7b-d0):PhSO2CH2D (7b-d1) ) 100:48.2
( 0.1. For 6b, the ratio of the integrated peak areas monitored at m/e
Experimental Section
General Aspects. 1H NMR spectra were recorded with a Varian
GEMINI-200 (200 MHz) NMR spectrometer. GLC analyses were
performed with Yanagimoto G180 and Shimadzu GC-14A gas chro-
matographs, using 2.5 mm × 1 m column of Carbowax 300 M (2%)
on Chromosorb WAW. A Shimadzu Chromatopac C-R3A integrator
was used for quantitative analyses. GC-MS analyses were carried
out with a JEOL D300 or a Shimadzu QP-5000 mass spectrometer
using a 2.5 mm × 1 m column of Carbowax 300 M (2%) on
(17) Sulfone was not obtained: Ishiguro, K.; Hatta, A.; Hayashi, M.;
Sawaki, Y. Unpublished result.
(18) Clennan, E. L.; Yang, K. J. Am. Chem. Soc. 1990, 112, 4044.
(19) Clennan, E. L.; Yang, K. Tetrahedron Lett.1993, 34, 1697.

1
Sulfone Formation in the Reaction of R2S and O2
J. Am. Chem. Soc., Vol. 118, No. 31, 1996 7271
) 140 (M) and 141 (M + 1) of 100:8.57 ( 0.04 was almost identical
to that of authentic 6b-d0 (100:8.43 ( 0.03), i.e., 6b-d0:6b-d1 ) >99:
<1.
and the products, three sulfones, 7a-d0, -d5, and -d6, and two sulfoxides,
6a-d0 and -d6, were analyzed by GC-MS as described above. The
kinetic isotope effects, kH/kD, for sulfoxide 6a at each time were obtained
by the product ratios for the sulfoxides 6a-d0 and -d6, and those in the
unimolecular sulfone formation were calculated from [(7a-d0) - (7a-
d6)]/(7a-d5), assuming that equal amounts of 7a-d0 and -d6 were formed
by secondary oxidation. The actual kinetic isotope effect for the sulfone
formation, 3.0 ( 0.4, was obtained by the extrapolation of these values
to zero time. The results are shown in Table 4.
16O2-18O2 Tracer Experiment. A 1-mL benzene or acetonitrile
solution of 0.1 M dimethyl sulfide-d6 (5a-d6) and 0.1 mM TPP or MB,
respectively, was placed in a 5-mL test tube with a septum rubber cap.
After the solution was purged with argon, oxygen gas (16O2:16O18O:
18O2 ) 100:0.79:11.9) was introduced into the test tube through a
syringe. After irradiation of an appropriate time at >400 nm, the
resulting reaction mixture was analyzed by GC-MS. The mass
intensities of parent peaks for 7a-d5 and 7a-d6 with 16O2, 16O18O, and
18O2, monitored at m/e ) 99, 100, 101, 102, 103, and 104, were
integrated by mass chromatography throughout the GC peaks. After
the corrections of the relative intensities for 7a-d5/7a-d6 and those of
the M + 1 and M + 2 values of natural abundance, the isotope ratios
were obtained as summarized in Table 3.
Kinetic isotope effects on the oxidation of PhSCH3 and PhSCD3 in
C6H6, C6H6-Et2O (1:1), and CH3CN were obtained similarly, the results
being summarized in Table 5.
Theoretical Calculations. Semiempirical RHF/PM3 molecular
orbital calculations15 were carried out on Apple Macintosh computers
using the MOPAC version 6.02.20 All of the geometries of intermedi-
ates and transition states were optimized with the keyword PRECISE
to increase the criteria for the termination of optimization by a factor
of 100. Transition states were obtained with the keyword TS and were
confirmed by frequency calculation and intrinsic reaction coordination
(IRC) procedures. The energies and principal parameters in the
optimized structures are shown in Figure 1, and the full geometries in
the Z matrix are given in the supporting information.
The 16O2-18O2 tracer experiment of di-tert-butyl sulfide (8) was
carried out similarly. The sulfone 10 showed no parent peak (M )
178), but the ratio of oxygen isotopes could be obtained from the major
fragment peaks, M - 56 (-CH2dC(CH3)2), since the mass spectrum
of authentic 10 showed fragment peaks of t-BuS16O2H monitored at
m/e ) 121, 122 (M), 123, 124, and 125, in the ratio 3.13 ( 0.07:100:
5.78 ( 0.23:5.47 ( 0.12:0.58 ( 0.01, which was close to the theoretical
ratio for C4H10SO2, M:M + 1:M + 2 ) 100:5.34:4.88. The photo-
oxidation of 100 mM di-tert-butyl sulfide (8) and 0.1 mM TPP with
16O2-18O2 (16O2:16O18O:18O2 ) 100:0.79:11.9) in benzene for 6 h
afforded the corresponding sulfoxide 9 and sulfone 10 in 2.3% and
0.24% yields, respectively. The sulfone was analyzed by GC-MS,
and the integrated peak ratios monitored at m/e ) 122 (M of
t-BuS16O2H), 123, 124 (M of t-BuS16O18OH), 125, and 126 (M of
t-BuS18O2H) were corrected with the M - 1, M + 1, and M + 2 values
observed at the authentic sample. The isotope ratios in the sulfone 10
obtained by three determinations were 10-16O2:10-16O18O:10-18O2 )
100:18.2 ( 0.2:1.2 ( 0.1.
Acknowledgment. This work was supported in part by a
Grant-in-Aid for Scientific Research from the Ministry of
Education, Science and Culture of Japan.
Supporting Information Available: Tables of the PM3-
calculated geometries and energies for the intermediates and
transition states (4 pages). See any current masthead page for
ordering and Internet access instructions.
JA960323O
(20) Stewart, J. J. P. MOPAC ver. 6.00 (QCPE No. 445). QCPE Bull.
1990, 10, 86. Hirano, T. MOPAC ver. 6.02. JCPE Newsletter 1991, 2 (4),
39. Toyota, J. Macintosh version, 1991.
Kinetic Isotope Effect. A benzene solution of 50 mM (CH3)2S, 50
mM (CD3)2S, and 0.1 mM TPP was irradiated for appropriate time,