Oxorhenium(V) dithiolates catalyze the oxidation by tert-butyl hydroperoxide of sulfoxides and sulfides, including 4,6-dimethyldibenzothiophene

Article abstract of DOI:10.1021/ic011091j

tert-Butyl hydroperoxide (TBHP) efficiently converts a wide variety of sulfides to sulfoxides and sulfones. The method offers the advantage that one product or the other can be obtained in high purity by a modest variation of conditions. The reactions occur smoothly at 25-50°C in chloroform and, to the extent studied, in toluene and methylene chloride. A catalyst is required; the most extensively studied was MeReO(mtp)PPh₃, 1, where mtpH₂ is 2-(mercaptomethyl)thiophenol. Other chelating dithiolate ligands can be used with comparable results. These oxidations were tested for dialkyl, alkyl-aryl, and diaryl sulfides; thiophenes; and thianthrene. Even the hard sulfide, 4,6-dimethyldibenzothiophene (DMDBT) was quantitatively oxidized to the dioxide with TBHP:DMDBT 3.0-3.5 and 0.05-3.8 mol % 1. The mechanism was explored in kinetics studies carried out only for methyl tolyl sulfide. The product buildup curve was complex, with an induction period followed by a rapid growth phase. The kinetic data could be modeled adequately but not perfectly by allowing five rate constants to refine. Their values are consistent with the chemical sense of the mechanism.

Full text of DOI:10.1021/ic011091j

Inorg. Chem. 2002, 41, 1272−1280
Oxorhenium(V) Dithiolates Catalyze the Oxidation by tert-Butyl
Hydroperoxide of Sulfoxides and Sulfides, Including
4,6-Dimethyldibenzothiophene
,†
Ying Wang,† Ga´bor Lente,‡ and James H. Espenson*
Ames Laboratory and Department of Chemistry, Iowa State UniVersity of Science and Technology,
Ames, Iowa 50011
Received October 22, 2001
tert-Butyl hydroperoxide (TBHP) efficiently converts a wide variety of sulfides to sulfoxides and sulfones. The method
offers the advantage that one product or the other can be obtained in high purity by a modest variation of conditions.
The reactions occur smoothly at 25−50 °C in chloroform and, to the extent studied, in toluene and methylene
chloride. A catalyst is required; the most extensively studied was MeReO(mtp)PPh3, 1, where mtpH is
2
2-(mercaptomethyl)thiophenol. Other chelating dithiolate ligands can be used with comparable results. These oxidations
were tested for dialkyl, alkyl−aryl, and diaryl sulfides; thiophenes; and thianthrene. Even the “hard” sulfide, 4,6-
dimethyldibenzothiophene (DMDBT) was quantitatively oxidized to the dioxide with TBHP:DMDBT 3.0−3.5 and
0.05−3.8 mol % 1. The mechanism was explored in kinetics studies carried out only for methyl tolyl sulfide. The
product buildup curve was complex, with an induction period followed by a rapid growth phase. The kinetic data
could be modeled adequately but not perfectly by allowing five rate constants to refine. Their values are consistent
with the chemical sense of the mechanism.
Introduction
kinetic efficiency. Their general formula is MeReO(dithi-
olate)PZ3, PZ3 representing a general alkyl, aryl phosphine,
as depicted in Chart 1. In this study, we have concentrated
on MeReO(mtp)PPh3, 1, where mtpH2 is 2-(mercaptometh-
yl)thiophenol. Many aspects of the reaction chemistry of 1
are now coming to be understood.10,11 Among such reactions
are bimolecular ligand displacements.10 In the case at hand,
the reactions can be referenced to Li ) PPh3 and Lj ) SRR′,
OSRR′, OPPh3, tert-butyl hydroperoxide (TBHP), or tmtu
(1,1,3,3-tetramethyl-2-thiourea).
The conversion of organic sulfides to sulfoxides and
sulfones holds interest from several points of view, to which
both laboratory and biologically related perspectives apply.1
Sulfide oxidation and sulfoxide reduction are the reverse of
one another in oxygen atom transfer (OAT) reactions. This
means that a given catalyst will be effective in either direction
depending on stoichiometric source or sink of the oxygen
atom. Sulfoxide reduction in nature is catalyzed by the
molybdenum-containing enzyme DMSO-reductase. It also
metabolizes aryl alkyl sulfoxides and other dialkyl sulf-
oxides.2-4
MeReO(mtp)Li + Lj f MeReO(mtp)Lj + Li
(1)
Certain oxorhenium(V) compounds5-9 such as MeReO-
(dithiolate)PPh3 catalyze OAT reactions with remarkable
Our studies of OAT have included catalyzed sulfoxide
reductions12 and, as reported here, sulfide oxidations with
* To whom correspondence should be addressed. E-mail: espenson@
ameslab.gov.
(6) Jacob, J.; Lente, G.; Guzei, I. A.; Espenson, J. H. Inorg. Chem. 1999,
38, 3762-3763.
† Iowa State University.
‡ Department of Inorganic and Analytical Chemistry, University of
Debrecen, Hungary.
(7) Jacob, J.; Guzei, I. A.; Espenson, J. H. Inorg. Chem. 1999, 38, 1040-
1041.
(1) Oae, S. In Organic Sulfur Chemistry: Structure and Mechanism; CRC
Press: Boca Raton, FL, 1991; pp 203-291.
(8) Lente, G.; Jacob, J.; Guzei, I. A.; Espenson, J. H. Inorg. React. Mech.
(Amsterdam) 2000, 2, 169-177.
(2) Hille, R. Chem. ReV. 1996, 96, 2757-2816.
(3) George, G. N.; Hilton, J.; Temple, C.; Prince, R. C.; Rajagopalan, K.
V. J. Am. Chem. Soc. 1999, 121, 1256-1266.
(9) Lente, G.; Espenson, J. H. Inorg. Chem. 2000, 39, 4809-4814.
(10) Lahti, D. W.; Espenson, J. H. J. Am. Chem. Soc. 2001, 123, 6014-
6024.
(4) Lippard, S. J.; Berg, J. M. Principles of Bioinorganic Chemistry;
University Science Books: Mill Valley, CA, 1994.
(5) Jacob, J.; Espenson, J. H. Chem. Commun. 1999, 1003-1004.
(11) Lente, G.; Guzei, I. A.; Espenson, J. H. Inorg. Chem. 2000, 39, 1311-
1319.
(12) Koshino, N.; Espenson, J. H. Manuscript in preparation.
1272 Inorganic Chemistry, Vol. 41, No. 5, 2002
10.1021/ic011091j CCC: $22.00 © 2002 American Chemical Society
Published on Web 02/14/2002

Oxorhenium(V) Dithiolate Catalysts
Chart 1
Sulfide oxidation provides the usual laboratory route to
sulfoxides. A useful reaction must proceed with 100%
selectivity, avoiding overoxidation to sulfones which has
sometimes been reported.24,25 Sulfones are also important
intermediates in organic reactions; again, quantitative con-
versions and 100% sulfone selectivity are needed.1,26 The
methods described here offer the further advantage that one
product or the other can be obtained at will. The method is
also applicable to thiophenes, which are inherently more
difficult to oxidize.27 Note, for example, the ease with which
RSR′ is oxidized by H2O2 with MeReO3 (MTO) as the
catalyst,28 as compared to thiophenes.29 To be specific,
MeSPh and dibenzothiophene are oxidized by MeRe(O)2-
(κ2-O2); the respective rate constants are 2.65 × 103 and 10.2
L mol-1 s-1. 2,5-Dimethyl thiophene reacts more slowly yet;
an estimate gives k ∼ 0.2 L mol-1 s-1.29
We wish particularly to point out that these procedures
are applicable even to 4,6-dimethyldibenzothiophene (we
designate this particular isomer as DMDBT; its monoxide
and dioxide are, respectively, DMDBTO and DMDBTO2).
It is a so-called “hard” sulfide that cannot be removed from
petroleum by current hydrodesulfurization (HDS) technology.
Perhaps DMDBT cannot make close contact with the surface
of the cobalt-molybdate HDS catalyst.30 The same reason
was invoked for the failure of bacterial bio-oxidation
processes, which work for other dimethylated-DBT isomers.31
DMDBT may now become the focus of more stringent
regulatory requirements because its sulfur content eventually
fouls catalytic converters. The value of oxidative chemistry
in this regard has been noted.32-34
TBHP as the stoichiometric oxidizing agent. The mechanisms
show considerable similarities between the rhenium com-
pounds and the molybdenum and tungsten enzymes and their
synthetic models.4,13-22 Rhenium, on the other hand, has been
relatively little investigated, but recent work with the catalyst
[ReO(hoz)2]+ (hozH ) 2-(2′-hydroxyphenyl)2-oxazoline) has
pointed the way.23
Experimental Section
Materials. Chloroform was used as the solvent in most cases.
Methylene chloride and toluene were also used for the oxidation
of DMDBT. The sulfides were used as received from commercial
sources because their 1H NMR spectra showed no impurities. TBHP
was obtained as a 5-6 M solution in nonane. The dimer {MeReO-
(mtp)}2, 2, was synthesized from MeReO335 and mtpH2.5 Reaction
of 2 overnight with a ligand L (PPh3, tmtu, or P(C6H4-4-CF3)3)
The rhenium-catalyzed reactions hold particular interest
because they are rapid and yield clean sulfoxides; under
minimally different conditions, sulfones are instead formed
cleanly and quantitatively. Both procedures are conveniently
implemented, as will be described. The sequential net
reactions are (TBA is tert-butyl alcohol)
(24) Madesclaire, M. Tetrahedron 1986, 42, 5459-5495.
(25) Patai, S.; Rappoport, Z. E. In Synthesis of Sulphones, Sulphoxides and
Cyclic Sulfides; John Wiley & Sons: New York, 1994.
(26) Trost, B. M.; Fleming, L. ComprehensiVe Organic Synthesis; Perga-
mon: Oxford, UK, 1991; Vol. 8, p 390.
(27) Galpern, G. D. In The Chemistry of Heterocyclic Compounds-
Thiophene and its DeriVatiVes; Gronowitz, S., Ed.; John Wiley &
Sons: New York, 1985; p 44.
(28) Vassell, K. A.; Espenson, J. H. Inorg. Chem. 1994, 33, 5491-5498.
(29) Brown, K. N.; Espenson, J. H. Inorg. Chem. 1996, 35, 7211-7216.
(30) We thank Dr. R. J. Angelici for pointing out to us the difficulty with
which highly hindered thiophenes are treated by current HDS
processes.
RSR′ + TBHP f RS(O)R′ + TBA
(2)
(3)
RS(O)R′ + TBHP f RS(O)2R′ + TBA
(13) Berg, J. M.; Holm, R. H. J. Am. Chem. Soc. 1985, 107, 917-925.
(14) Caradonna, J. P.; Reddy, P. R.; Holm, R. H. J. Am. Chem. Soc. 1988,
110, 2139-2144.
(15) Holm, R. H.; Berg, J. M. Acc. Chem. Res. 1986, 19, 363-370.
(16) Holm, R. H. Chem. ReV. 1987, 87, 1401-1449.
(17) Schultz, B. E.; Holm, R. H. Inorg. Chem. 1993, 32, 4244-4248.
(18) Sung, K.-M.; Holm, R. H. Inorg. Chem. 2000, 39, 1275-1281.
(19) Sung, K.-M.; Holm, R. H. J. Am. Chem. Soc. 2001, 123, 1931-1943.
(20) Laughlin, L. J.; Young, C. G. Inorg. Chem. 1996, 35, 1050-1058.
(21) Smith, P. D.; Millar, A. J.; Young, C. G.; Ghosh, A.; Basu, P. J. Am.
Chem. Soc. 2000, 122, 9298-9299.
(31) Kropp, K. G.; Andersson, J. T.; Fedorak, P. M. EnViron. Sci. Technol.
1997, 31, 1547-1554.
(32) Collins, F. M.; Lucy, A. R.; Sharp, C. J. Mol. Catal. A: Chem. 1997,
117, 397-403.
(33) Yatsu, K.; Miki, H.; Ukegaw, K.; Yamamoto, N. (Ministry of
Economy, Trade and Industry; National Industrial Research Institute,
Japan). Jpn. Kokai Tokkyo Koho 2001; Patent No. JP2001/29376, p
12.
(22) Smith, P. D.; Slizys, D. A.; George, G. N.; Young, C. G. J. Am. Chem.
Soc. 2000, 122, 2946-2947.
(23) Arias, J.; Newlands, C. R.; Abu-Omar, M. M. Inorg. Chem. 2001,
40, 2185-2192.
(34) Otsuki, S.; Nonaka, T.; Qian, W.; Ishihara, A.; Kabe, T. Sekiyu
Gakkaishi 1999, 42, 315-320.
(35) Herrmann, W. A.; Kratzer, R. M.; Fischer, R. W. Angew. Chem., Intl.
Ed. Engl. 1997, 36, 2652-2654.
Inorganic Chemistry, Vol. 41, No. 5, 2002 1273

Wang et al.
resulted in the respective mononuclear compounds MeReO-
and 1 (6.3 mg, 0.05%) in 0.2 mL of toluene. The solution was
refluxed for 2 h, during which time a white solid precipitated. It
was filtered and washed with cold dichloromethane to eliminate
the trace of catalyst. The dioxide was isolated in 96% yield by this
procedure and was characterized by its known 1H NMR spectrum.42
For DMBTO, DMDBT (0.42 g, 2.0 mmol) and TBHP (2.5 mmol)
were dissolved in 5 mL of CHCl3 at room temperature. Catalyst 1
(39 mg, 3.0 mol %) was dissolved in 2 mL of CHCl3 and added to
the solution. The reaction was stirred for 1.5 h. The solvent was
then evaporated, and DMDBTO was isolated by flash chromatog-
(mtp)L: 1, 5, and 6. The first two of these are already known,5-7
1
and the latter was characterized by its H and 31P NMR spectra.
Addition of hexane precipitated the product, which was filtered
and washed with hexane to produce MeReO(mtp)L in >90% yield.
Compound 7 was obtained as a brown solid by adding 2.1 equiv
of 1,2-bis(diphenylphosphinobenzene), dppb, to a toluene solution
of 2 that had first been treated with 6 equiv of 4-(tert-butyl)pyridine.
The product was obtained after allowing the solution to stand
overnight.36 Compounds 3 and 4 were prepared as reported earlier.9
Compound 10 was synthesized from 2 and 2.1 equiv (+)-
neomenthyldiphenylphosphine (Aldrich). The NMR spectrum of
10 in CDCl3, recorded on a Bruker DRX 400 MHz spectrometer,
is characterized by these resonances: 1H δ (ppm) 8.00 (m, 2H),
7.53-7.66 (m, 6H), 7.19-7.32 (m, 6H), 4.98 (d, 1H, J ) 12 Hz),
4.23 (m, 1H), 1.60-2.20 (m, 9H), 2.90 (d, 1H, J ) 12 Hz), 2.45
(d, 3H, Re-CH3, J ) 8 Hz), 1.23 (d, 3H, J ) 8 Hz), 0.62 (d, 3H,
J ) 8 Hz), 0.18 (d, 3H, J ) 8 Hz); 31P δ 28.86.
1
raphy using 1:1 hexane/ethyl acetate as eluent (52%). The H and
13C spectra were determined; no literature report could be found.
NMR: 1H δ 7.60 (d, 2H, J ) 8 Hz), 7.46 (t, 2H, J ) 8 Hz), 7.23
(d, 2H, J ) 8 Hz), 2.74 (s, 6H); 13C δ 18.7, 119.5, 124.8, 131.0,
132.7, 137.6, 139.2, 142.7.
Results
Sulfides to Sulfoxides. Useful amounts of the pure
sulfoxides could be obtained by defined laboratory proce-
dures. These reactions, when catalyzed by 1 in chloroform
(except for DMDBT where methylene chloride and toluene
were also used), gave excellent conversions of RSR′
selectively to the sulfoxide RS(O)R′ in essentially quantita-
tive yields when carried out on a 2 mmol scale. Several
protocols were explored. A syringe pump was used to
introduce 1 slowly as the reaction was intermittently
monitored by TLC until it had reached completion. Com-
parable results and reaction times were realized in many cases
with the same quantity of 1 added at the outset. Typically,
we used 1-3 mol % (6-40 mg) of 1 and found reaction
times of 40 min-2.5 h at 25 °C. A mild increase in
temperature to 50 °C was helpful. To prevent sulfone
formation, an excess of TBHP was strictly avoided in
experiments at 50 °C.
This procedure was tested for 14 sulfides: (a) dialkyl (R,
R′ ) Me, R ) Me, R′ ) tert-butyl; R ) R′ ) tert-butyl;
cyclo-C4H8); R ) Me, R′ ) Bn; (b) alkyl, aryl (R ) Me,
C6H4-4-X, X ) Me, NO2, Br, AcO, and CN); (c) diaryl (R
) R′ ) Ph); (d) R ) Ph, R′ ) vinyl); (e) dibenzothiophene;
and (f) thianthrene. Table 1 presents the results. In every
case, the sulfoxide was obtained almost quantitatively.
This method proved tolerant of the tested functional
groups. Vinyl phenyl sulfoxide was formed without an
epoxidation side product.43,44 In the workup, the TBA
produced evaporated with the solvent. The 1H NMR spectrum
showed sulfoxide but not sulfone. That was true even for
diphenyl sulfide, where overoxidation can occur.24,25 In no
case did the unreacted sulfide exceed 5%, and generally, it
was absent. This method avoids an aqueous workup, which
is often required when peracids are used,24 and is thus
particularly useful for water-soluble sulfoxides.
General Procedure for Sulfoxides. The rhenium catalyst was
usually 1, but others were tested as well. Just 2.05 mmol of TBHP
was added to 2 mL of chloroform containing 2.0 mmol of sulfide.
A solution of 1 in 1-2 mL of chloroform was added to provide
the desired catalyst concentration. In some cases, the solution of 1
was introduced slowly by syringe pump, and in others, it was added
all at once. Likewise, experiments were carried out at 25 or 50 °C.
The progress of the reaction was monitored in a tiny sample that
was withdrawn periodically and checked by TLC or by1H NMR
spectroscopy. When the reaction was found to be complete, the
1
solvent and TBA were evaporated. The H and 13C NMR spectra
of the products were determined in comparison with literature
values,37,38 which sufficed for their identification because they are
all known materials. High-purity sulfoxides were obtained by flash
chromatography with ethyl acetate as the eluent.
General Procedure for Synthesizing Sulfones. TBHP (2.0
mmol, but usually 3.0 mmol, an excess) was added to 1.0 mmol of
sulfide in 3 mL of chloroform. Catalyst 1 contained in 1 mL of
chloroform was added by syringe pump over 2 h at 50 °C, as the
reaction was intermittently monitored by TLC. Upon completion,
the sulfones were obtained by evaporation and again identified by
comparison with the reported NMR spectra.39,40
Oxidative Cycloaddition of 2,5-Dimethylthiophene with N-
Phenylmaleimide. TBHP (2.00 mmol) was added to 3 mL of a
chloroform solution of 2,5-dimethylthiophene (112 mg, 1.00 mmol)
at 50 °C) and N-phenylmaleimide (320 mg, 1.85 mmol). The latter
was added to trap the otherwise labile thiophene monoxide. Catalyst
1 (6.3 mg, 1%) dissolved in 1 mL of chloroform was added by
syringe pump over 2 h. The reaction was further stirred 1 h and
then evaporated to yield the product, N-phenyl-1,4-dimethyl-7-
thiabicyclo[2.2.1]hept-5-ene-2,3-dicarboxamide 7-oxide, which was
separated by flash chromatography using 2:1 hexane/ethyl acetate
as eluent. The product (78%) was identified from 1H and 13C NMR
data and from its mass spectrum in comparison with literature
values.41
Oxidation of DMDBT to DMDBTO2 and DMDBTO. DMDBT
(0.105 g, 0.50 mmol) was dissolved in 2 mL of toluene which was
brought to reflux. To this was added TBHP (1.75 mmol, 3 equiv)
Convenience is also a factor. The attractive features of 1
are its convenient synthesis, long shelf life (>3 months), and
stability toward the humid air of an Iowa summer. Not only
is the catalyst a forgiving one, but the catalytic reactions
themselves can also be carried out on the benchtop. At room
(36) Espenson, J. H.; Shan, X.; Lahti, D. W.; Rockey, T. M.; Saha, B.;
Ellern, A. Inorg. Chem. 2001, 40, 6717-6724.
(37) Ali, M. H.; Bohnert, G. J. Synthesis 1998, 1238-1240.
(38) Ali, M. H.; Stevens, W. C. Synthesis 1997, 764-768.
(39) Rozen, S.; Bareket, Y. J. Org. Chem. 1997, 62, 1457-1462.
(40) Ali, M. H.; Bohnert, G. J. Synth. Commun. 1998, 28, 2983-2998.
(41) Li, Y.; Thiemann, T.; Sawada, T.; Mataka, S.; Tashiro, M. J. Org.
Chem. 1997, 62, 7926-7936.
(42) Sato, K.; Hyodo, M.; Aoki, M.; Zheng, X. Q.; Noyori, R. Tetrahedron
2001, 57, 2469-2476.
(43) Guertin, K. R.; Kende, A. S. Tetrahedron Lett. 1993, 34, 5369-5372.
(44) Su, W. Tetrahedron Lett. 1994, 35, 4955-4958.
1274 Inorganic Chemistry, Vol. 41, No. 5, 2002

Oxorhenium(V) Dithiolate Catalysts
Table 1. Oxidation of Sulfides to Sulfoxides by TBHP, Catalyzed by 1a
a At 25 °C in CHCl3; TBHP (2.05 mmol) was added to 2.0 mmol of sulfide in 2 mL chloroform; catalyst 1 in 1-2 mL CHCl3 was added by syringe pump
(or all at once, if so noted). b Yields are NMR yields of the crude product after solvent and TBA evaporation. NMR yields were determined by integration.
The samples were very pure by NMR, so this is a satisfactory procedure. In a few cases, the method was checked by GC-MS. The isolated yield is given
in parentheses. c The catalyst was added all at the beginning. d 2.00 mmol TBHP. e Other materials are 5,10-thianthrene dioxide (2%); 2% unconverted
thianthrene was also present.
temperature, they proceed in an hour or two, depending on
several factors: temperature and the concentrations of 1 and
TBHP.
A mild increase in temperature to 50 °C permitted the use
of a smaller amount of 1 and allowed a shortened reaction
time. The amount of TBHP then needed to be even more
tightly controlled at eRSR′, lest some sulfone also be
formed. At the extreme, a much less reactive sulfide,
4-nitrophenyl methyl sulfide, was used with TBHP, 1.0 mmol
each. With 0.05 mol % (0.3 mg) of 1, 95% sulfoxide was
Inorganic Chemistry, Vol. 41, No. 5, 2002 1275

Wang et al.
Table 2. Oxidation of Sulfides to Sulfones by 3 equiv of t-BuOOH
Catalyzed by 1
Table 3. Oxidation of Methyl Tolyl Sulfoxide (MTSO) to Sulfone by
TBHP under Different Conditionsa
MTSO/mmol
TBHP/mmol
1 (mol %)
MTSO2, %
1.0
1.0
1.0
2.0
2.0
1.5
1.0
0
1.0
100
12
82
a At 50 °C in 3 mL of chloroform.
Scheme 1. Oxidation of Thianthrene
at the expense of a longer reaction time, ∼4 h, although the
yield of some sulfones was reduced to 85%.
During this procedure, sulfides were found to have
disappeared within 15-30 min. To no surprise, the reaction
starting with a sulfide is a sequential one, with a sulfoxide
intermediate. To show whether 1 is needed for the second
oxidation, methyl tolyl sulfoxide (MTSO) was used. We
obtained the sample commercially, lest some adventitious
rhenium be present in a sample that we had prepared. With
1 mol % of 1, a reaction between TBHP (2.0 mmol) and
MTSO (1.0 mmol) at 50 °C in chloroform gave a 100% yield
of MTSO2 in 2 h. Without 1, only 12% sulfone was formed.
Recognizing that the 2:1 molar ratio represents an excess,
the reaction was repeated stoichiometrically, 1.5:1 TBHP/
MTSO, yielding 82% MTSO2. The data in Table 3 suggest
that 1 does indeed catalyze the sulfoxide-to-sulfone step but
that this step, unlike sulfide-to-sulfoxide, occurs slowly on
its own.
Oxidation of Thianthrene. Several products can be
formed in principle, but only two were found, thianthrene
5-oxide (SSO) with 1:1 TBHP and RSR′, and thianthrene
5,10-dioxide (SOSO) with a 3:1 ratio. No trace was found
of 5,5-dioxide (SSO2), 5,5,10-trioxide (SOSO2), or 5,5,10,-
10-tetraoxide (SO2SO2). No other oxidation products were
detected. The results (Table 2) are summarized in Scheme
1.
a The yields were determined by taking the 1H NMR of the crude product
after evaporation. b The balance is thianthrene monoxide. c Compound 1
added at the start; complete reaction in 2 h. d Compound 1 added at the
start; complete reaction in 1 h.
obtained in 1 h. A control reaction lacking 1 showed 2%
sulfoxide in that time.
In the absence of sulfide, 1 is decomposed by TBHP. The
modes of catalyst decomposition are complex; we represent
them by an approximate chemical equation that shows
formation of RSSR (the cyclic organic disulfide, mtp) and
MTO (which does not catalyze sulfide oxidation by TBHP),
in eq 4. This is a simplification, however, because some Ph3-
PS and other byproducts are formed as well.
MeReO(mtp)PPh3 + 3TBHP f
Oxidation of Thiophenes. Heavily substituted thiophenes,
such as dibenzothiophene, were successfully oxidized to
dibenzothiophene monoxide (Table 1, entry 13) and diben-
zothiophene dioxide (Table 2, entry 5), as governed by the
TBHP:DBT ratio. On the other hand, with a 1:1 or a 3:1
reagent ratio, 2,5-dimethylthiophene both at room temper-
ature and at 50 °C gave <20% 2,5-dimethylthiophene 1,1-
dioxide; unreacted thiophene and unidentified products
remained, and the use of additional 1 did not improve
matters.
MeReO3 (MTO) + RSSR + 3TBA + Ph3PO (4)
Sulfides to Sulfones, via Sulfoxides. This transformation
requires working at 50 °C. The syntheses were done with
3:1 TBHP/RSR′, because with 2 equiv of TBHP, only partial
(75%-85%) conversions to sulfones were obtained. Sulfones
were obtained in excellent yields in 2 h with only 0.3-0.5
mol % of 1. Six sulfides were tested to demonstrate tolerance
to functional group variations, including the vinylic group
in phenyl vinyl sulfide, which remained unchanged. These
results are summarized in Table 2. With syringe pump
addition of 1, all of the oxidations were complete within 2
h. The amount of 1 could be further reduced to 0.05 mol %
Thiophene oxides lacking attached ring systems are known
to cyclize rapidly in solution.45-47 One can capture the
thiophene monoxide by addition of a dienophile trap.41,48,49
1276 Inorganic Chemistry, Vol. 41, No. 5, 2002

Oxorhenium(V) Dithiolate Catalysts
Table 4. Oxidation of 4,6-DMDBT with Different Reaction Conditions and Catalystsa
time
entry
DMDBT (mmol)
TBHP (mmol)
catalyst (mol %)
temp (°C)
(h)
solvent
“SO” (%)
“SO2” (%)
“SO” + “SO2” (%)
1
2
3
4
5
6
7
8
9
10
11
12
13
1.0
2.0
2.0
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
2.0
2.5
2.5
1.0
1 (1.0)
1 (3.0)
2 (3.0)
3 (5.0)
5 (5.0)
5 (3.8)
5 (5.0)
1 (3.8)
1 (3.8)
1 (1.0)
1 (0.5)
1 (0.05)
(0)
50
25
25
50
50
50
50
50
50
50
70
111
111
2
C
C
C
C
C
C
M
C
T
T
T
T
T
71
58
61 (52)
20
3
4
15
31
91
61
65
82
95
100
100
100
100
100
100
100
7
1.5
1.5
2.5
2.5
2
1.5
2
2
67
64
0
0
0
3
9
7
0
1.0
1.75
1.75
1.75
1.75
1.50
1.75
1.75
1.75
100 (95)
100 (92)
100 (95)
97
91
93
100 (96)
0
2.5
0.7
2
2
7
a Solvents: C ) chloroform; T, toluene; M, methylene chloride. “SO” ) DMDBTO; “SO2” ) DMDBTSO2.
In this case, N-phenylmalelimide was used. Its adduct was
isolated in 78% yield, as described by Scheme 2:
Scheme 2. Oxidation and Trapping of 2,5-Dimethylthiophene
Monoxide
Oxidation of DMDBT. This compound proved more
difficult to oxidize, but good results were obtained in three
Figure 1. Formation of MTSO, showing fitted vs experimental data during
the reaction of 33.9 mM MST and 83 mM TBHP in the presence of 0.31
mM MeReO(mtp)PPh3, 1. The experimental progress curve was modeled
by a kinetics simulations routine that gave the optimum fit to data from six
such experiments. The fitted curve is shown as a smooth curve. Other
experiments and their fits are given in Figure S-1.
solvents with several oxorhenium(V) catalysts. The data are
summarized in Table 4. A modest excess of TBHP was
needed to ensure complete conversion to the dioxide. Several
oxorhenium catalysts were used; of these, 1 and 5 were used
the most, and both gave high yields in a reasonable time,
e2.5 h. The reaction runs well in chloroform, methylene
chloride, and toluene; the last is particularly noteworthy in
that it represents a nonpolar hydrocarbon environment.
Further reductions in the catalyst level could be realized by
working at a higher temperature. At the extreme, the reaction
proceeded well in refluxing toluene with only 0.05% of the
catalyst. The separation of the dioxide proved trivial because
it is nearly insoluble in all of these solvents. It precipitates
from toluene even at 111 °C.
Product Buildup Curves. The formation of methyl tolyl
sulfoxide (MTSO) was monitored as a function of time under
several sets of conditions in which methyl tolyl sulfide (MTS)
was used as a representative substrate. The curves all showed
similar features, most notably a distinct induction period
whose length is dependent on several factors. One illustration
of the product buildup curve is displayed in Figure 1, and
others are illustrated in Figure S-1. In each case, the rate
increased, and the induction period decreased with increasing
[TBHP] and [1], whereas it slowed dramatically and the
induction time grew longer as [MTS] was increased, other
concentrations being constant.
Independent of this, PPh3 on its own is oxidized by TBHP
in the presence of catalyst 1 faster than MTS is.50 When 3%
of PPh3 relative to MTS was added, added [PPh3] ) 1.0 mM,
the oxidation of MTS slowed; this phenomenon can be
represented by t1/2 values of ∼3900 and 1900 s, the first for
the reaction with phosphine added and the second free of
phosphine save for the 0.3 mM PPh3 introduced as coordi-
nated to 1. The implications of these variations will be taken
up in the Discussion, where an analysis of the reaction
mechanism will be presented.
Comparing Different Rhenium Catalysts. A comparison
among several of the catalysts shown in Chart 1 for the
conversion of MTS to MTSO is presented in Figure 2. Table
5 shows the time required with different catalysts for 50%
conversion of MTS with 1 mol % catalyst. All of these
catalysts carry the reaction to completion with rates in the
order 2 > 6 > 3 > 1. Compound 4, which lacks the
stabilization provided by a chelating dithiolate, acts as a
(45) Nagasawa, H.; Sugihara, Y.; Ishii, A.; Nakayama, J. Bull. Chem. Soc.
Jpn. 1999, 72, 1919-1926.
(46) Nakayama, J.; Nagasawa, H.; Sugihara, Y.; Ishii, A. J. Am. Chem.
Soc. 1997, 119, 9077-9078.
(47) Nakayama, J. Bull. Chem. Soc. Jpn. 2000, 73, 1-17.
(48) Li, Y.; Matsuda, M.; Thiemann, T.; Sawada, T.; Mataka, S.; Tashiro,
M. Synlett 1996, 461-464.
(49) Naperstkow, A. M.; Macaulay, J. B.; Newlands, M. J.; Fallis, A. G.
Tetrahedron Lett. 1989, 30, 5077-5080.
(50) Saha, B.; Espenson, J. H. To be submitted for publication.
Inorganic Chemistry, Vol. 41, No. 5, 2002 1277

Wang et al.
1
in this paragraph were monitored by H and 31P NMR
spectroscopies.
Attempted Chiral Induction. We also used 10 as a
catalyst. It bears a chiral phosphine, (+)-neomenthyldiphe-
nylphosphine, a ligand that has been used successfully in
asymmetric catalysis.51,52 The oxidation of MTS to MTSO
by 1 equiv of TBHP occurred efficiently and completely,
>95%. Chiral induction was not observed in several trials,
however, as no enantiomeric excess was obtained as deter-
mined by a chiral chromatography.53
Discussion
The success of catalysis speaks for itself. Sulfoxides and
sulfones can be obtained nearly quantitatively without cross-
contamination. Even the least reactive substrates (DMDBT
and 4-nitrophenyl methyl sulfide) can be oxidized.
The principal issue at this point is to devise a reaction
scheme consistent with the data and with the known
chemistry and reactivity of 1. We have relied upon the new
data as well as the information gleaned from our study of
rhenium-catalyzed OAT from pyridine N-oxides to PAr3.54
The latter set of reactions occurs considerably faster than
sulfide oxidations do.
Figure 2. Different oxorhenium(V) compounds in this family serve as
catalysts for the oxidation of MTS by TBHP. Data are shown for (a) [2] )
0.157 mM; (b) [6] ) 0.31 mM; (c) [3] ) 0.31 mM; (d) [1] ) 0.31 mM; (e)
[7] ) 0.31 mM.
Table 5. Comparisons among Different Oxorhenium(V) Catalysts for
the Oxidation of MTS and 4-Nitrophenyl Methyl Sulfidea
catalyst
t1/2 (h)b,c
time (h)d
2.5
yieldd
95
2
6
3
1
4
5
7
0.42
0.58
0.97
1.05
2.5
1
4
95
95
75e
97
2.5
Kinetic data were collected for a single sulfide, MTS, that
is representative of the class. The buildup of MTSO was
NR
1
a At 25 °C in CHCl3, 2.0 mmol substrate by TBHP (2.05 mmol). b Listed
in order of decreasing reactivity for 50% conversion of MTS; the rhenium
content, relative to sulfide, is 1 at. %. c Substrate is MTS; t1/2 is for
comparison only, because each progress curve shows an induction period
of different length. d Time and yield for complete oxidation to 4-O2NC6-
H4S(O)Me. e Catalyst 4 has decomposed after 4 h.
followed by H NMR spectroscopy. All of the experiments
were characterized by an induction period of varying length,
even though 100% conversion to product was recorded with
a stoichiometric excess of either MTS or TBHP. Reactions
were not followed to the point where either sulfone formation
or decomposition of 1 became important.
The ligand displacement (not dissociation) reactions of eq
1 (the 5 ligands give a set of 10 equilibrium reactions, 4 of
which are independent) occur more rapidly than sulfide
oxidation. Catalyst 1 was the only rhenium species detected
by NMR spectroscopy during oxidation catalysis. Phosphine
coordinates to Re(V) much more strongly than do any of
the other Lewis bases present. As a result, one important
kinetic barrier is the step in which TBHP displaces PPh3.
Our analysis starts with this premise: ligand displacement
steps other than phosphine displacement remain at equilib-
rium. This is one of the assumptions we have made to reduce
the kinetic steps to a tractable number.
catalyst, but it gave only 75% conversion, owing to its
eventual decomposition. Table 5 also presents the time
needed for the complete conversion of 4-nitrophenyl methyl
sulfide to its sulfoxide.
Detection and Characterization of the ReVII Intermedi-
ate. A direct reaction between TBHP and 1 was monitored
by 1H NMR spectroscopy at room temperature and at 240-
260 K. The room-temperature products were those shown
in eq 4 without any intermediate being detected.
Dimer 2 is also a catalyst for sulfoxide formation. Addition
of 8.0 mM TBHP to 2.0 mM 2 at 260 K gave a red
compound 8. It was formed nearly quantitatively in about
1
30 min. Its H NMR spectrum, given in Figure S-2, is
Insofar as the chemistry critical to catalysis is concerned,
another kinetic barrier arises at the step in which a diox-
orhenium(VII) intermediate is formed by elimination of TBA
from an intermediate containing coordinated hydroperoxide:
characterized by these resonances: δ (ppm): 2.71 (s, 3H,
Re-CH3), 4.55 (d, 1H, J ) 8 Hz), 5.36 (d, 1H, J ) 8 Hz),
and 7.1-7.7 (m, 4H). Its structure is assigned as shown in
1
Chart 1 because the H NMR spectrum showed that TBA
had been formed concurrently, signaling that 8 is a Re(VII)
compound, for proper redox balance.
MeReO(mtp)(TBHP) f ΤΒΑ + MeRe(O)2(mtp) (5)
In solution, 8 is stable for >1 h. Efforts to obtain it in
solid form for further characterization and analysis failed,
however, probably because of its reactive nature. These
reactions of 8 were characterized, as presented in Scheme
3. With sulfide or phosphine, the dioxorhenium(VII) species
is reduced to oxorhenium(V). With 4-picoline, however,
reduction is not spontaneous; instead, the simple adduct 9 is
formed, as previously reported.15 The reactions referred to
The oxidation of phosphines by pyridine N-oxides requires,
from an analysis of the kinetic data, that a nucleophile assist
(51) Zim, D.; de Souza, R. F.; Dupont, J.; Monteiro, A. L. Tetrahedron
Lett. 1998, 39, 7071-7074.
(52) Morrison, J. D.; Burnett, R. E.; Aguiar, A. M.; Morrow, C. J.; Phillips,
C. J. Am. Chem. Soc. 1971, 93, 1301-1303.
(53) We are grateful to Prof. D. A. Armstrong, J. Anderson, and T. Xiao
for these determinations.
(54) Wang, Y.; Espenson, J. H. Submitted for publication.
1278 Inorganic Chemistry, Vol. 41, No. 5, 2002

Oxorhenium(V) Dithiolate Catalysts
Scheme 3. Formation and Reactions of Dioxorhenium(VII)
Scheme 4. Proposed Nucleophilic Assistance in the Formation of
Dioxorhenium(VII)
two effects: the aforementioned low rate of displacement
of coordinated PPh3 and the partial loss of phosphine due to
oxidation. We therefore introduce a step well-known from
our earlier studies of PyO chemistry. This involves the
oxidation of phosphine by the dioxorhenium(VII) intermedi-
ate:
MeRe(O)2(mtp) + 2PPh3 f
MeReO(mtp)(PPh3) [1] + Ph3PO (k5) (8)
The initial supply of phosphine is no greater than that of
the starting concentration of 1. Thus, any PPh3 oxidized in
reaction 8 reduces the proportion of the rhenium catalyst
present in the phosphine form 1 (which is most stable of all
with respect to ligand displacement equilibria). Other MeRe-
O(mtp)Lj species are formed as a consequence. They are
undoubtedly more reactive catalysts, because their thermo-
dynamic barrier relative to the generation of the reactive
species MeReO(mtp)TBHP in eq 5 is lowered. As phosphine
is oxidized, therefore, the rate increases with time. This is
the essence of the chemistry underlying the induction period.
For Li ) PPh3 and Lj ) Me2S, pK ) 5.3. From the kinetic
modeling, as given subsequently, we can deduce an estimate
pK ) 3.8 for Li ) PPh3 and Lj ) TBHP. The binding of
TBHP is thus favored over that of MTS, which supports the
selection of that step for the modeling.
Kinetic Modeling. To approach this complex system, it
proved desirable to reduce the number of conceivable steps
because none of the rate constants are known directly. For
example, the kinetics and equilibria of the different ligands
that participate in eq 1 were ignored, save for the case Li )
PPh3 and Lj ) TBHP. Reaction 6 was successfully included
with TBHP as the nucleophile, but a parallel step with PPh3
as the nucleophile proved insignificant, no doubt because
phosphine remains at a very low concentration. Reaction 6
produces the Re(VII) compound MeRe(O)2(mtp), 8, which
can react with MTS (higher concentration, lower rate
constant) in competition with PPh3 (vice versa) to restore
MeReO(mtp)L, L ) TBHP, PPh3. These reactions are the
minimum set needed to reproduce the kinetic data. They are
shown in the form of a catalytic cycle in Scheme 5, where
fast steps (e.g., the second step of eq 7 and others) have
been omitted.
this reaction. Each one of phosphine, pyridine N-oxide,
pyridine, tetrabutylammonium bromide, and tmtu was able
to do so.54
MeReO(mtp)O(OPy) + Nuc f
Py + MeRe(O)2(mtp)Nuc (6)
On the basis of that work, it seems likely that a nucleophile
assists reaction 5 as well. In this instance, the nucleophiles
present are PPh3 (except in one instance, however, it was
present only at the level of the catalyst), MTS, MTSO,
TBHP, and TBA. This suggestion is depicted in Scheme 4.
The rationale is that this is a step in which the d2 electrons
of Re(V) are removed by oxidation.
The dioxorhenium(VII) product of eq 5-6, like other
oxorhenium(VII) compounds, is believed to be in rapid
equilibrium with Lewis bases. Rapid equilibration between
ligands and rhenium(VII) has been shown to be the case not
only here but with MeReO3 in its coordination with pyridine
and pyridine N-oxide.55 Equilibrium between the rhenium
product shown and MeRe(O)2(mtp) occurs on a submilli-
second time scale. In reaction 6, the nucleophile can be any
of PyO, Py, or [Bu4N]Br.
In the step that follows, MTS abstracts an oxygen atom
from the dioxorhenium(VII) intermediate. We also show this
reaction as two steps occurring in sequence, on the basis of
results from molybdenum chemistry.21,22
MTS, k4
MeRe(O)2(mtp) MTS8 {MeReO(mtp)(MTSO)}
8
MeReO(mtp)MTS + MTSO (7)
This is not the complete story, however, because it would
lead to the smooth conversion of MTS to MTSO, without
the distinct induction period. The latter appears to arise from
The success of a model is measured by the extent to which
the calculated and experimental curves agree. Its plausibility
is judged by the chemical scheme itself and the fitted values
(55) Wang, W.-D.; Espenson, J. H. J. Am. Chem. Soc. 1998, 120, 11335-
11341.
Inorganic Chemistry, Vol. 41, No. 5, 2002 1279

Wang et al.
Table 6. Chemical Equationsa and the Fitted Valuesb of Their Rate
Scheme 5. Catalytic Cycle Used for Kinetic Modeling
Constants at 25° in CDCl3
chemical equation
k/L mol-1 s-1
MeReO(mtp)PPh3 +TBHP f
MeReO(mtp)TBHP + PPh3
k1 ) 9.8 ( 1.8 × 10-4
k-1 ) 6.7 ( 1.4
k2 ) 2.2 ( 0.3
MeReO(mtp)TBHP +TBHP f
MeRe(O)2(mtp) + TBA + TBHP
MeRe(O)2(mtp) + MTS (+ TBHP) f
MeReO(mtp)TBHP + MTSO
k4 ) 40 ( 16
MeRe(O)2(mtp) + PPh3 (+ TBHP) f
MeReO(mtp)TBHP + Ph3PO
k5 ) 2.5 ( 1.0 × 102
a A species shown in parentheses is not a part of the rate law, but it
completes the stoichiometry. b The program ZiTa was used for data fitting.56
The fitted rate constants, although of low precision (see
Table 6), deserve comment. As anticipated, k1 ()1 × 10-3
L mol-1 s-1) is small. It can be compared to the value for
another weak nucleophile, with k ) 7 × 10-3 L mol-1 s-1
estimated for pyridine. Values for reactions in which PPh3
displaces a more weakly bound ligand include k-1 ) 7 L
mol-1 s-1 for TBHP and k ∼ 80 L mol-1 s-1 for Py (after a
significant extrapolated from data in ref 10). As anticipated,
k1 is quite low, 10-3 L mol-1 s-1. Also, k5 > k4, although
the margin is not high: 250 versus 40 L mol-1 s-1. This
reflects the selectivity of phosphine over sulfides, consistent
with their nucleophilic strengths.
of the rate constants. In this scheme, there are five constants
to be determined from the six experiments, each with 16-
41 concentration-time values. The fitting to this model by
the program ZiTa56 was satisfactory but not fully precise;
see Figures 1 and S-1. The fitted rate constants are sum-
marized in Table 6. More critical, however, is the result that
the semiquantitative features were properly modeled. Thus,
the occurrence of an induction period, the variation of its
length with concentration, the rapid increase to a maximum
rate, and the result that all of the limiting reagent is converted
to product are indications that the model is satisfactory, in
essence at least.
Acknowledgment. This research was supported by the
U. S. Department of Energy, Office of Basic Energy
Sciences, Division of Chemical Sciences under Contract
W-7405-Eng-82.
Supporting Information Available: Additional figures (S-1-
3), including plots of kinetic data and NMR spectra. This material
is available free of charge via the Internet at http://pubs.acs.org.
(56) Peintler, G. ZITA 5.0., A comprehensiVe program package for fitting
parameters of chemical reaction mechanisms, Version 3.1; Institute
of Chemistry, JATE: Szeged, Hungary, 1999. The first use of this
program was described: Peintler, G.; Nagypal, I.; Epstein, I. R. J.
Phys. Chem. 1990, 94, 2954.
IC011091J
1280 Inorganic Chemistry, Vol. 41, No. 5, 2002