A mild, efficient and selective oxidation of sulfides to sulfoxides

Article abstract of DOI:10.1039/a804180f

Several titanium derivatives supported on silica have been investigated as catalysts in the sulfide→sulfoxide oxidation; the supported titanium/tartaric acid catalyst is highly suited to reactions carried out with H₂O₂, leading to high yields and high sulfoxide/sulfone selectivities, while small asymmetric inductions (up to 13%) are observed.

Full text of DOI:10.1039/a804180f

View Article Online / Journal Homepage / Table of Contents for this issue
A mild, efficient and selective oxidation of sulfides to sulfoxides
José M. Fraile, José I. García, Beatriz Lázaro and José A. Mayoral*†
Departamento de Química Orgánica y Química Física, Instituto de Ciencia de Materiales de Aragón, Universidad de
Zaragoza-C.S.I.C., E-50009 Zaragoza, Spain
Several titanium derivatives supported on silica have been
investigated as catalysts in the sulfide?sulfoxide oxidation;
the supported titanium/tartaric acid catalyst is highly suited
to reactions carried out with H2O2, leading to high yields and
high sulfoxide/sulfone selectivities, while small asymmetric
inductions (up to 13%) are observed.
sulfone selectivity is obtained with H2O2. The recovered
catalysts show a similar activity, but the product selectivity
increases, mainly in the reaction carried out with H2O2, which
indicates that the recovered catalyst does not have the same
structure. It seems logical that in the presence of a large excess
of water (we use 30% H2O2), the isopropoxy groups are
hydrolysed to hydroxy groups, and that this modification of the
Ti environment influences the course of the reaction.
In view of these results we decided to assess the effect of the
introduction of polar groups near the titanium and we treated the
original catalyst with diethyl (R)-tartrate (Scheme 2) to obtain
cat-2 (1.06 mmol Ti g21). The modification of the catalyst was
monitored by IR spectroscopy, which shows the appearance of
a broad carbonyl band. The width of this band indicated the
presence of several kinds of carbonyl groups, and Scheme 2
shows only one of the possibilities. The catalytic tests (Table 1,
entries 4–7) show that the introduction of diethyl tartrate
reduces the catalytic activity and increases the sulfoxide/sulfone
The selective oxidation of sulfides to sulfoxides has been a
challenge for many years, partly due to the importance of
sulfoxides as intermediates in organic synthesis.1 In recent
years selective oxidation of sulfides to sulfoxides has been
carried out with a large number of supported reagents.2
Recently, the use of H2O2 has taken on a new importance due to
environmental implications, however, a catalyst is necessary to
activate this oxidant. TS-13,4 and TS-2,5 promote sulfide
oxidation, but the use of bulky sulfides is precluded by their
limited access to the Ti active sites. In order to overcome this
limitation other molecular sieves have been used.3–6 Ti-MCM-
41 allows the oxidation of bulky sulfides,6 but with low
sulfoxide/sulfone selectivity.
O
O
O
Following a methodology closely related to that used by Shell
to prepare TiCl4-modified silica,7 we have treated silica with
Ti(OPri)4 and shown8 that this solid is an efficient catalyst in the
epoxidation of alkenes of any size with hydroperoxides. We
considered it interesting to test these silica-supported titanium
compounds as catalysts in the selective oxidation of sulfides to
sulfoxides (Scheme 1).
catalyst
S
S
R1
R2
S
+
R1
R2
R1
R2 H2O2 or
TBHP
2
3
1
a R1 = p-Tol, R2 = CH3
b R1 = R2 = Bun
f
R1 = R2 = Ph
g R1 = Ph, R2 = CH CH2
c R1 = CH3, R2 = But
d R1–R2 = (CH2)4
e R1 = Ph, R2 = Bn
O
Firstly we tested the silica treated with Ti(OPri)4 (cat-1, Ti
content: 1.02 mmol g21) as a catalyst in the oxidation of methyl
p-tolyl sulfide with TBHP and with H2O2 (Table 1, entries 1–4).
Although the oxidation is faster with TBHP, a better sulfoxide/
h R1–R2
Scheme 1
Table 1 Results obtained in the oxidation of methyl p-tolyl sulfide 1a promoted by different supported titanium catalystsa
Yield (%)
Entry
Catalyst
Sulfide
Oxidant
t/h
0.5
2
3
3
4
3
3.5
4
4
4.5
24
24
4
24
4
24
24
24
24
Total
sulfoxide 2
sulfone 3
2:3
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
cat-1
cat-1b
cat-1
cat-1b
cat-2
cat-2
cat-2b
cat-3
cat-3
cat-3b
—
cat-3
cat-3
cat-3
cat-3
cat-3
cat-3
cat-3
cat-3
1a
1a
1a
1a
1a
1a
1a
1a
1a
1a
1a
1a
1b
1c
1d
1e
1f
TBHP
TBHP
H2O2
H2O2
TBHP
H2O2
H2O2
TBHP
H2O2
H2O2
H2O2
H2O2
H2O2
H2O2
H2O2
H2O2
H2O2
H2O2
H2O2
79
85
87
88
90
89
85
91
96
96
49
97
100
97
98
94
97
97
95
59
71
75
20
14
12
4
7
12
1
9
1
1
2
3
0
3
2
1
2
3
3
75:25
83:17
86:14
95:5
92:8
87:13
99:1
90:10
99:1
99:1
96:4
97:3
—
97:3
98:2
99:1
98:2
97:3
97:3
84
83c
77d
84
82e
95f
95
47
94
100
94
96
93
95
94
92
1g
1h
a At 25 °C with 187 mg cat per mmol sulfide and 1 equiv. oxidant, in CH2Cl2 (with TBHP) or MeOH (with H2O2). The reactions were monitored by GC
and the final results determined by 1H NMR spectroscopy after filtering and washing the catalyst. b Catalyst recovered from a reaction with the same oxidant.
c 7% ee. d 2% ee. e 8% ee. f 13% ee.
Chem. Commun., 1998
1807

View Article Online
CO2R
CO2R
HO
HO
Table 2 Recovery of the catalyst in the oxidation of 1b with H2O2 promoted
by cat-3
CO2R
CO2R
OPri
OPri
O
O
O
O
O
O
+ 2 PriOH
Ti
Ti
Yield (%)
cat-1
Run
t/h
Total
Sulfoxide 2b Sulfone 3b
Ti/mmol g21a
cat-2 R = Et
cat-3 R = H
1
2
3
4
5
6
4
5
8
8
9
—
100
98
96
90
91
—
100
96
92
82
82
—
0
2
4
8
9
—
0.82
—
—
—
—
Scheme 2
selectivity in the reaction carried out with TBHP, while the
catalytic performance is almost unchanged when H2O2 is used.
The recovered catalyst gives an exceptional sulfoxide/sulfone
selectivity (entry 7), which points to a new structure of the
catalytic sites after recovery, as shown by the modification of
the IR carbonyl band, which suggests the hydrolysis of the ester
groups.
0.22
a Determined by plasma emission spectroscopy.
In view of this we decided to treat the initial catalyst with
(R)-tartaric acid and to test the catalytic activity of this solid
(cat-3, 0.82 mmol Ti g21). The results obtained with TBHP
(entry 8) are very similar to those obtained using the catalyst
containing diethyl (R)-tartrate (cat-2). However, the reactions
carried out with H2O2 (entries 9 and 10) lead to an excellent
sulfoxide/sulfone ratio and a very high selectivity (97%) with
respect to the H2O2.
(R)-Tartaric acid is strongly adsorbed onto silica, thus it is
important to assess the modification of the titanium centres and
the catalytic role of these centres. The modification of the
environment of titanium can be estimated from the amount of
PriOH lost during the preparation of the catalyst. Although it is
not possible to obtain quantitative conclusions, given that some
PriOH may remain adsorbed on the silica surface, GC analysis
shows that 65 ± 5% of the PriO groups present in the original
solid are lost during treatment. This result agrees with the
modification of most of the titanium atoms. The catalytic role of
the titanium was confirmed by adsorbing onto silica gel the
same amount of (R)-tartaric acid used to modify the catalyst and
carrying out the reaction in the presence of this solid. The results
obtained are the same as those reached in the absence of
catalyst, showing that the catalytic activity is related to the
presence of titanium.
part, in the homogeneous phase. In order to clarify this point we
treated cat-3 with H2O2 for 4 h. The solid was separated by
filtration and the filtrate used in the oxidation of a correspond-
ing amount of sulfide 1a. The filtrate promotes the reaction with
a 83:17 sulfoxide/sulfone ratio. In another experiment we
carried out the reaction of 1a until 85% conversion (15 min),
then the catalyst was separated by filtration and both the solid
and the filtrate were used in a new reaction. The sulfide was
consumed in both reactions at almost the same rate.
In our opinion these results indicate the co-existence of the
homogeneous and heterogeneous reactions and that a very
active homogeneous catalyst is obtained under these conditions.
Therefore, it seems that the identification of the homogeneous
species and the improvement of the behaviour of the heteroge-
neous system against leaching are important aims.
These results clearly show that modification of the environ-
ment of titanium via introduction of organic molecules can
modify the performance of the catalysts. This strategy opens the
way to the preparation of new families of supported chiral
catalysts in which grafting is carried out through the metal
instead of through the chiral auxiliary.
This work was made possible by the generous financial
support of the C.I.C.Y.T. (Project MAT96-1053).
Although it was not the aim of the present work, the
introduction of chiral centers may produce an asymmetric
reaction, and so this possibility was considered. The ee of the
sulfoxides was determined as previously described,9 and the
absolute configuration was determined by polarimetry.10 The
low ee obtained for the (R)-sulfoxide provides additional
evidence for the modification of the catalytic centers.
Given that the catalyst obtained by treatment with (R)-tartaric
acid (cat-3) leads to the best selectivity, we tested the effect of
using a smaller amount of catalyst. Very good results were
obtained, with a sulfide/titanium ratio = 140, by simply
increasing the reaction time. In order to test the general scope of
this method it was applied to several other sulfides (Scheme 1,
Table 1, entries 12–19).
This solid promotes the reaction with a very small amount of
catalyst and leads to high yields and selectivities. The
methodology is applicable to bulky sulfides and is compatible
with the presence of other oxidisable functions, as shown in the
last two examples where epoxidation of the double bond (1g) or
Baeyer–Villiger oxidation of the ketone (1h) were not ob-
served.
As far as the mechanism is concerned, either the H2O2 may be
the oxidation agent, via a Ti–OOH intermediate, or a supported
peracid intermediate may be formed.
We tested the recovery of the catalyst in the oxidation of
dibutyl sulfide 1b (Table 2). A gradual leaching of titanium
takes place. However, this leaching is very slow given that after
five reactions each mmol of titanium has been treated with
about 5000 mmol of water. Furthermore, even the fifth reaction
takes place with reasonable yield and selectivity. The existence
of leaching suggests that the reaction can take place, at least in
Notes and References
† E-mail: mayoral@posta.unizar.es
1 J. Drabowski, P. Kielbasinski and M. Mikolajczyk, Synthesis of
Sulfoxides, Wiley, New York, 1994; M. Madesclaire, Tetrahedron,
1986, 42, 549.
2 G. W. Breton, J. D. Fields and P. J. Kropp, Tetrahedron Lett., 1995, 36,
3825; P. Kannan, R. Sevvel, S. Rajagopal and K. Pitchumani,
Tetrahedron, 1997, 53, 7635; R. S. Varna, R. K. Saini and H. M.
Meshram, Tetrahedron Lett., 1997, 38, 6525; M. H. Ali and W. C.
Stevens, Synthesis, 1997, 764; M. Hirano, S. Yakabe, S. Itoh, J. H. Clark
and T. Morimotoa, Synthesis, 1997, 1161; V. Hulea, P. Moreau and F.
di Renzo, J. Mol. Catal. A, 1996, 111, 355.
3 R. S. Reddy, J. S. Reddy, R. Kumar and P. Kumar, J. Chem. Soc., Chem.
Commun., 1992, 84.
4 P. S. Raglavan, V. Ramaswamy, T. T. Upadhya, A. Sudalai, A. V.
Ramaswamy and S. Sivasankar, J. Mol. Catal. A, 1997, 122, 75.
5 S. V. N. Raju, T. T. Upadhya, S. Ponatnam, T. Daniel and A. Sudalai,
Chem. Commun., 1996, 1969.
6 A. Corma, M. Iglesias and F. Sánchez, Catal. Lett., 1996, 39, 153.
7 R. A. Sheldon, in Aspects of Homogeneous Catalysis, ed. R. Ugo,
Reidel, Dordrecht, 1981, vol. 4, pp. 1–70.
8 C. Cativiela, J. M. Fraile, J. I. García and J. A. Mayoral, J. Mol. Catal.
A, 1996, 112, 259.
9 M. Deshmukh, E. Duñach, S. Juge and H. B. Kagan, Tetrahedron Lett.,
1984, 25, 3467.
10 K. Mislow, M. M. Green, P. Laur, J. T. Melillo, T. Simmons and A. L.
Ternay Jr., J. Am. Chem. Soc., 1965, 87, 1958.
Received in Cambridge, UK, 3rd June 1998; 8/04180F
1808
Chem. Commun., 1998