Kinetics and Mechanism of the Methyltrioxorhenium-Catalyzed Sulfoxidation of Thioketones and Sulfines

Article abstract of DOI:10.1021/jo990621e

Kinetic studies have been carried out on the oxidation of thiobenzophenones and thiocamphor by hydrogen peroxide and on the second step in the sequence, oxidation of the resulting sulfines (Ar₂C= S=O). The first reaction follows the pattern now common to methyltrioxorhenium/H₂O₂ reactions, in that the rate constant for the reaction between the peroxorhenium intermediate and the thioketone follows a Hammett correlation such that electron-releasing substrates react more rapidly. The reaction constant is p = -1.12. However, the plot of log(k_X/k_H) for the sulfines against 2σ, determined over the range -1.6 < 2σ < +1.5, is markedly U-shaped. This suggests a mechanism for the sulfines in which the direction of the electron flow in the transition state changes with the electron demand of the substituents on the sulfines. The product of oxidation of the sulfine is a transient sultine (from epoxidation of the C=S bond). It cannot be detected, however, because it so rapidly yields sulfur monoxide. The SO was oxidized to SO₂ under these conditions rather than undergoing disproportionation to SO₂ and S. Also, SO was trapped with a 1,3-diene as a thiophene-1-oxide.

Full text of DOI:10.1021/jo990621e

6
374
J . Org. Chem. 1999, 64, 6374-6379
Kin etics a n d Mech a n ism of th e Meth yltr ioxor h en iu m -Ca ta lyzed
Su lfoxid a tion of Th iok eton es a n d Su lfin es
Ruili Huang and J ames H. Espenson*
Ames Laboratory and the Department of Chemistry, Iowa State University of Science and Technology,
Ames, Iowa 50011
Received April 13, 1999
Kinetic studies have been carried out on the oxidation of thiobenzophenones and thiocamphor by
hydrogen peroxide and on the second step in the sequence, oxidation of the resulting sulfines (Ar
2
Cd
SdO). The first reaction follows the pattern now common to methyltrioxorhenium/H reactions,
2
O
2
in that the rate constant for the reaction between the peroxorhenium intermediate and the
thioketone follows a Hammett correlation such that electron-releasing substrates react more rapidly.
The reaction constant is F ) -1.12. However, the plot of log(k /k ) for the sulfines against 2σ,
X H
determined over the range -1.6 < 2σ < +1.5, is markedly U-shaped. This suggests a mechanism
for the sulfines in which the direction of the electron flow in the transition state changes with the
electron demand of the substituents on the sulfines. The product of oxidation of the sulfine is a
transient sultine (from epoxidation of the CdS bond). It cannot be detected, however, because it so
rapidly yields sulfur monoxide. The SO was oxidized to SO
undergoing disproportionation to SO and S. Also, SO was trapped with a 1,3-diene as a thiophene-
-oxide.
2
under these conditions rather than
2
1
Sulfines (thioketone S-oxides) are attractive heterocu-
mulenes that participate in the thioepoxidation of alk-
is activated for nucleophilic or electrophilic attack. We
have studied the kinetics and mechanism of both steps
in eq 1, with a particular reference to the substituent
effects on the rate constants of ring-substituted thioben-
zophenones.
1
2
3-6
enes, nucleophilic addition, and cycloaddition reactions.
They are usually prepared by oxidizing the thioketones
7
-13
10,14
15
with peracids,
ozone,
or singlet oxygen. The
sulfines can be further oxidized, leading ultimately to the
ketone. The sequence of conversions, disregarding other
intermediates at this point, is outlined in eq 1.
Exp er im en ta l Section
Ma ter ia ls. The starting thiobenzophenones were prepared
according to the literature method,1 except for commercially
available thiocamphor, which was used as received. The
thiobenzophenones are deep blue materials. The absorption
5,16
(1)
maxima for the compounds (XC H ) CdS, given as λmax/nm (ꢀ/L
6 4 2
We have found that hydrogen peroxide can carry out
both stages of this transformation when methyltrioxo-
rhenium (CH ReO , abbreviated as MTO) is used as the
3 3
catalyst. We have undertaken a study of these reactions
to describe their mechanisms. A point of particular
mechanistic interest in this system is whether peroxide
-
1
-1
4
mol cm ), are X ) p-NMe
2
458 (2.7 × 10 ), p-OMe 355 (2.46
4
×
10 ) and 570 (407), p-Me 587 (262), H 590 (170), p-Cl 600
(
215), m-NO
10 L mol cm ).
In str u m en ta tion . The reaction progress for thioketones
2
-1
598 (203). Thiocamphor has λmax 250 nm (ꢀ 8.7
3
-1
×
was monitored at these wavelengths spectrophotometrically
with a Shimadzu UV-3101PC instrument. The second stage,
oxidation of sulfines, did not give suitable absorbance changes.
(
1) Adam, W.; Deeg, O.; Weinkotz, S. J . Org. Chem. 1997, 62, 7084.
1
(2) Alayrac, C.; Cerreta, F.; Chapron, I.; Corbin, F.; Metzner, P.
Those reactions were followed by H NMR with a Bruker DRX-
Tetrahedron Lett. 1996, 37, 4507.
3) Mloston, G.; Linden, A.; Heimgartner, H. Helv. Chim. Acta 1996,
9, 31.
4) Braverman, S.; Grinstein, D.; Gottleib, H. J . Chem. Soc., Perkin
Trans. 1 1998.
5) Manoharan, M.; Venuvanalingam, P. J . Phys. Org. Chem. 1997,
0, 768.
6) Saito, T.; Takekawa, K.; Nishimura, J .-I.; Kawamara, M. J .
Chem. Soc., Perkin Trans. 1 1997, 2957.
7) Battaglia, A.; Dondoni, A.; Giorgianni, P.; Maccagnani, G.;
Mazzanti, G. J . Chem. Soc. (B) 1971, 1547.
8) Zwanenburg, B.; Thijs, L.; Strating, J . Recl. Trav. Chim. Pays-
Bas 1967, 86, 577.
9) Hasserodt, J .; Pritzkow, H.; Sundermeyr, W. Chem. Ber. 1993,
27, 1701.
1
4
00 spectrometer. The H NMR chemical shifts were measured
(
relative to the residual resonance of the solvent, CD
.92 ppm. The solvent in both cases was water-acetonitrile
or, in 1H experiments, D
O) 1:4 by volume, and it contained
2
HCN, δ
7
1
(
(
2
1
7
(
0.10 M trifluoromethanesulfonic acid to stabilize the catalyst.
Throughout, the temperature was maintained at 25 °C.
1
(
Equ ilibr iu m Mea su r em en ts. The equilibrium constants
for the two stepwise MTO-H interactions defined in
Scheme 1 were evaluated in this medium by H NMR methods.
The intensities of the CH groups of the three species MTO,
A, and B were integrated relative to a standard. The deter-
2
O
2
(
1
(
3
(
-1
-1
mined constants are K
2
1
) 91 L mol and K
2
) 347 L mol at
1
5 °C, with the activity of water being ignored. These values
are much as one might interpolate from other media. As in
other circumstances, the values reflect the phenomenon of
(
(
10) Zwanenburg, B.; J anssen, W. A. J . Synthesis 1973, 617.
11) Carlsen, L.; Snyder, J . P.; Holm, A.; Pederson, E. Tetrahedron
1
8
1
981, 37, 1257.
12) Adiwidjaja, G.; Sawluk, A.; Volz, W.; Voss, J . Phosphorus,
Sulfur, Silicon 1973, 74, 451.
(
cooperativity, in that K
the rhenium-oxo multiple bonds are strengthened relative to
1
> K . We interpret that to mean that
2
(13) Cerreta, F. Bull. Soc. Chim. Fr. 1995, 67, 132.
(
14) Tabuchi, T.; Nojima, M.; Kusabayashi, S. J . Chem. Soc., Perkin
Trans. 1 1991, 12, 3043.
15) Ramnath, N.; Ramesh, V.; Ramamurthy, V. J . Org. Chem. 1983,
8, 214.
(16) Scheeren, J .; Ooms, P.; Nivard, R. Synthesis 1973, 149.
(17) Abu-Omar, M.; Hansen, P. J .; Espenson, J . H. J . Am. Chem.
Soc. 1996, 118, 4966-4974.
(
4
1
0.1021/jo990621e CCC: $18.00 © 1999 American Chemical Society
Published on Web 07/27/1999

Sulfoxidation of Thioketones and Sulfines
J . Org. Chem., Vol. 64, No. 17, 1999 6375
Sch em e 1
previously unknown compounds (p-FC
CF CSO. The synthesis and isolation of those
compounds has been reported separately.
6 4 2
H ) CSO and (m-
3
6 4 2
C H )
2
9
Kin etics of Th iok eton e Oxid a tion . Control experi-
ments for reactions between thioketones and hydrogen
3
peroxide, without MTO, proceeded more than 10 times
more slowly than those measured with the catalyst
present. Thus the uncatalyzed reaction can be neglected
entirely. It was also shown that oxidation by atmospheric
oxygen or photoactivation in laboratory light were not
important. No effort was made to exclude either.
those in the predecessor compound as the number of RedO
bonds is reduced at each stage.
Kin etic Da ta . Both initial rate and full time course
analyses were employed to determine the kinetics. To deter-
mine initial rates, the absorbance-time data from spectro-
photometric kinetics or the integrated intensities-time data
The typical ranges of concentrations used for kinetics
experiments on thioketones in these studies were 0.01-
0.1 mM thioketone, 0.1-1 mM hydrogen peroxide, and
1
from H NMR kinetics were converted to concentration-time
values. The initial rate for each experiment was obtained by
1
-100 µM MTO. The analysis of the kinetic data requires
fitting each curve to a polynomial function: [Y]
t
) [Y]
was taken as the
. Under some circumstances the reaction
0
- m
1
t
an examination of the general scheme by which MTO
activates hydrogen peroxide. The general mechanism has
2
3
-
m
2
t - m
3 1
t ... The value of parameter m
initial reaction rate, v
i
3
0-32
now been established.
The reactions, shown in
was first-order with respect to the thioketone concentration.
In such cases the absorbance-time data were fitted to the
equation
Scheme 1,30 feature (a) conversion of MTO to two suc-
cessive peroxorhenium complexes, (b) attack of substrate
Y at an oxygen of the peroxo group, and (c) transfer of
that oxygen to Y. This general picture requires a few
refinements in special cases, as recently reviewed, but
it otherwise suffices quite well.
Abs ) Abs + (Abs - Abs ) exp(-k t)
(2)
t
∞
0
∞
ψ
3
3
Tr a p p in g of SO w ith Dien es. 4,4′-Difluorothiobenzophe-
none S-oxide (13 mM) and MTO (12 mM) were mixed in
acetonitrile containing 0.1 M HOTf to stabilize the MTO.17
Hydrogen peroxide (250 mM) was then added to the solution,
The kinetic data for the thioketones could be analyzed
by accounting for only the left-hand loop in Scheme 1.
That is so because these experiments were carried at low
followed immediately by 2,3-dimethyl-1,3-butadiene (100 mM).
The reaction was monitored by 1H NMR. After 1.5 h, 2,5-
[H O ], where any contribution from B can be ignored.
2
2
dihydro-3,4-dimethylthiophene-1-oxide (5.0 mM) was formed.
This compound is the known product of the reaction between
Under these circumstances the rate of the reaction takes
the form
1
9
sulfur monoxide and the diene. The detected spectrum
1
matched that of the known compound: H (CD
2
3
CN) δ 3.85 (d,
k k [H O ][Y][Re]
k k [H O ][Y][Re]
1 3 2 2 T
1
3
2
2
T
H), 3.61 (d, 2H), 1.77 (s, 6H). The remainder of the diene (46
v )
=
(3)
mM) had been oxidized to diol products. MTO-catalyzed
oxidations of dienes with hydrogen peroxide have been well-
k-1 + k [Y] + k [H O ]
k-1 + k [Y]
3
3
1
2
2
2
0
characterized.
in which [Re]
Under the conditions used for the thioketones, the k
] denominator term was entirely negligible. The
T
stands for the total rhenium concentration.
The same procedure using 4,4′-dimethoxythiobenzophenone
and thiobenzophenone S-oxides with the same trapping re-
agent, 2,3-dimethyl-1,3-butadiene, was used. In these cases
the same SO trapped product was obtained.
1
-
2 2
[H O
expression for the reaction rate in eq 3 states that the
variation of the initial rate with the concentration of the
thioketone will take the form of a rectangular hyperbola,
with the rate attaining a thioketone-independent plateau
Resu lts
Id en tifica tion a n d Isola tion of th e Su lfin es. The
initially formed sulfines were identified by their UV-
at high concentrations. The same is true for values of k
as a function of [R CS]. These experiments were carried
out with (p-MeOC CdS and with thiocamphor. Data
for both compounds are displayed in Figure 1. From both
ψ
2
1
vis, H NMR, and mass spectra in comparison with those
6
4 2
H )
7
,8,11,14,15,21-28
of the known compounds.
These spectra and
those of the starting thiones and final product ketones
are given in the Supporting Information. The only
exceptions to that are two of the sulfines, which have
evidently not been previously reported. These are the
-1 -1
experiments, the value k
1
) 15.5 ( 0.6 L mol
s
was
-1
1
obtained. Given K
1
) 91 L mol from H NMR experi-
ments cited previously, we obtain the value k-1 ) 0.17
-1
s
. Such determinations were needed to substantiate the
1
model and establish values of k and k-1 in this medium,
(
18) Espenson, J . H.; Abu-Omar, M. M. Adv. Chem. Ser. 1997, 253,
9-134.
19) Tardif, S. L.; Rys, A.; Abrams, C. B.; Abu-Yousef, I. A.; Leste-
Laserre, P. B.; Schults, E. K. V.; Harpp, D. N. Tetrahedron 1997, 53,
2225.
but the kinetic data at the plateau provided no new data
and no information about the reaction of the thioketone
itself.
In most of the determinations, therefore, the thioketone
concentration was kept deliberately low, such that eq 3
9
(
1
(
(
20) Tan, H.; Espenson, J . H. Inorg. Chem. 1998, 37, 467-472.
21) Maccagnani, G.; Innocenti, A.; Zani, P.; Battaglia, A. J . Chem.
Soc., Perkin Trans. 2 1987, 1113.
22) Dahn, H.; Pechy, P.; Toan, V. V.; Bonini, B. F.; Lunazzi, L. J .
Chem. Soc., Perkin Trans. 2 1993, 10, 1881.
3
could be simplified even further by dropping the k term
in the denominator. The initial rate could then be
accurately represented by
(
(
23) Zwanenburg, B. Tetrahedron 1971, 27, 1731.
24) Tangerman, A.; Zwanenburg, B. J . Chem. Soc., Perkin Trans.
(
2
2
1
1973, 458.
25) Tangerman, A.; Zwanenburg, B. J . Chem. Soc., Perkin Trans.
1974, 1141.
26) Veenstra, G. E.; Zwanenburg, B. Recl. Trav. Chim. Pays-Bas
976, 95, 37.
27) Huisgen, R.; Mloston, G.; Polborn, K.; Palacios-Gambra, F.
Liebigs Ann. Org. Bioorg. Chem. 1997, 1, 187.
28) Kuipers, J . A. M.; Lammerink, B. H. M.; Still, K. W. J .;
Zwanenburg, B. Synthesis 1981, 4, 295.
v ) K k [R CdS] [H O ] [Re]
T
(4)
(
i
1
3
2
0
2
2 0
(
The test of this kinetic equation is presented in Figure
(
(29) Huang, R.; Espenson, J . H., submitted for publication.
(30) Abu-Omar, M. M.; Appleman, E. H.; Espenson, J . H. Inorg.
Chem. 1996, 35, 7751-7757.
(

6
376 J . Org. Chem., Vol. 64, No. 17, 1999
Huang and Espenson
sulfene:
(5)
These reactions occur rather more slowly. To conduct the
kinetics measurements, the thioketones were first treated
with 1.0 equiv of hydrogen peroxide, in the presence of
1
MTO. The conversion to the sulfine was shown by H
NMR in each case to be quantitative, within the precision
of the measurement. Additional hydrogen peroxide was
2 2 2
then added to provide [H O ]/[Ar CSO] g 20. The disap-
1
pearance of the sulfine was monitored by H NMR,
because the absorbance changes were not suitable. The
overall concentration ranges in these experiments were
F igu r e 1. Kinetic data for the oxidation of thioketones to
sulfines by hydrogen peroxide with MTO as a catalyst. The
initial rates at constant catalyst and hydrogen peroxide
concentrations describe a rectangular hyperbola against the
concentration of the thioketone. Data were acquired in aqueous
2
-20 mM sulfine, 0.2-0.5 M H
2 2
O , and 0.1-10 mM
MTO.
The conditions under which a thioketone is oxidized
give reactions with a typical lifetime of ∼10 min; under
similar conditions the resulting sulfine requires 670 h.
The addition of the electronegative oxygen atom to the
sulfur greatly reduces the rate of a subsequent oxidation.
Higher peroxide concentrations were used for the
sulfines to avoid a prolonged reaction time. This change
means that the simplified kinetic treatment given above
will not suffice, because B now attains a high concentra-
tion and contributes to the overall catalysis. With all the
reactions in Scheme 1, the rate law takes the form
acetonitrile (1:4) at 25 °C. Values are shown for (4-MeOC
CS (squares) and for thiocamphor (circles).
6 4 2
H ) -
d[Y]
dt
-
)
2
k k k [Re] [Y][H O ]
1
2
4
T
2
2
k k [Re] [H O ][Y] +
1
3
T
2
2
k4[Y] + k
-2
(6)
2
k k [H O ]
1
2
2
2
k-1 + k [Y] + k [H O ] +
3
1
2
2
F igu r e 2. Kinetics of oxidation of thiobenzophenones by
hydrogen peroxide. The y-axis displays the initial rate divided
k4[Y] + k-2
by [Ar
Re] , in accord with eq 4. Data, acquired in aqueous aceto-
nitrile (1:4) at 25 °C, are shown for (p-XC CS, with X )
CH O (a), CH (b), and Cl (c).
2 0 2 2 0
CS] [H O ] . This quantity is directly proportional to
At these high peroxide concentrations, however, a dif-
ferent simplification is valid. This provides a form that
is an accurate representation under the circumstances:
[
T
6
4 2
H )
3
3
Ta ble 1. Kin etic Da ta for th e MTO-Ca ta lyzed Oxid a tion
v ) k [R CSO][Re]
T
(7)
4
2
of Th ioben zop h en on es (XC6H4)2CdS a n d Su lfin es
(
XC6H4)2CdSdO by Hyd r ogen P er oxid e
To check this equation, tests were carried out to affirm
the first-order dependence on sulfine concentration, the
first-order dependence on the total MTO concentration
(see Figure 3), and the absence of a kinetic term in the
hydrogen peroxide concentration, which was varied with-
thioketones k3 sulfines k4
-
1
-1
(L mol-1
-1
X
(L mol
s
)(25 °C)
s
)(22.5 °C)
3
1
p-NMe2
p-OMe
p-Me
H
(9.2 ( 0.4) × 10
(3.3 ( 0.1) × 10
(4.5 ( 0.1) × 10
(9.0 ( 0.1) × 10
(1.93 ( 0.12) × 10
(3.00 ( 0.08) × 10
(5.46 ( 0.13) × 10
(2.02 ( 0.07) × 10
(3.03 ( 0.16) × 10
(2.71 ( 0.09) × 10
(2.90 ( 0.06) × 10
(3.37 ( 0.12) × 10
(1.06 ( 0.01) × 10
3
2
1
-1
-2
-2
out effect in the range 0.2-0.5 M. The values of k
4
-
2
determined for the sulfines by this procedure are sum-
marized in Table 1.
p-F
p-Cl
p-Br
1
-2
(5.2 ( 0.1) × 10
-
-
2
2
The oxidation of sulfines to ketones may proceed
through a sultine or sulfene intermediate, as written in
eq 5, or in some other manner. It had been suggested
previously, in the oxidation of sulfines with monoper-
phthalic acid8 and perbenzoic acid,7 that the second
oxygen was added to the CdS double bond. Theoretical
calculations showed that the sultine intermediate so
obtained was more stable than its isomer, the sulfene. It
was further suggested that the sultine would rapidly
m-CF3
m-NO2
0
3
-1
(5.4 ( 0.3) × 10
(3.0 ( 0.1) × 10
thiocamphor
2
, which displays plots of v
three substrates. The values of k
and for others are summarized in Table 1.
Kin etics of Su lfin e Oxid a tion . The next step of
oxidation converts the sulfine to a sultine and/or a
i
/[Y]
0
[H
2
O
2
]
0
against [Re]
T
for
3
for these compounds
7
,8,34,35
eliminate the ketone, forming sulfur monoxide.
(
(
31) Gable, K. P. Adv. Organomet. Chem. 1997, 41, 127-161.
32) Herrmann, W. A.; K u¨ hn, F. E. Acc. Chem. Res. 1997, 30, 169-
(34) Lyaschchuk, S. I.; Strypnik, Y. G.; Bezvodnyi, V. P. Russ. J .
Org. Chem. 1997, 33, 9958.
1
4
80.
(
33) Espenson, J . H. J . Chem. Soc., Chem. Commun. 1999, 479-
(35) McIntosh, C. L.; De Mayo, P. J . Chem. Soc., Chem. Commun.
1969, 32.
88.

Sulfoxidation of Thioketones and Sulfines
J . Org. Chem., Vol. 64, No. 17, 1999 6377
Ta ble 2. Com p a r ison s of th e Ra te Con sta n ts k3 for A
a n d k4 for B in Th eir Rea ction s w ith Differ en t
Su bstr a tes
1
k4/k3a
substrate
PhSMe
Ph2S
PPh3
k3 (L mol- s-1)a
ref
3
2.65 × 10
0.36
0.27
3.0
37
37
48
48
49
49
50
51
52
2
1.18 × 10
5
7.3 × 10
3
P(C6F5)3
1.30 × 10
2.7
trans-PhCHdCHOMe
trans-PhCHdCHMe
14.2
0.51
0.90
0.43
2.3
0.49
0.58
-
-2
Cl
5.6 × 10
-
2
Br
3.5 × 10
0
.19
1
0.2
2.1
53
F igu r e 3. Values of the pseudo-first-order rate constants in
aqueous acetonitrile (1:4) at 22.5 °C for the oxidation of
thiobenzophenone oxides are a linear function of [Re]
agreement with eq 7. Values are shown for (p-XC CSO,
for X ) CH O (a), CH (b), and Cl (c).
T
, in
a
From the original references compiled in ref 33.
6
4 2
H )
3
3
4
0% being detected as the product of eq 9 and the balance
as its oxidation product.
That sulfur monoxide could disproportionate, and it was
the previous basis for suggesting the sultine. The oxida-
tion of SO by peroxorhenium complexes seems more
probable in this medium; no elemental sulfur was found
in our experiments.
Discu ssion
The most important issue to be dealt with is the
comparison of the mechanism for thioketones versus
sulfines. A seeming complication is that the experiments
3 4
yielded values of k for the one but k for the other. In
(8)
fact, this need be of no concern. For numerous substrates
both rate constants have been determined. Relatively
little variation has been found. The summary of rate
constants in Table 2 makes this point very clearly. There
is, indeed, one exception, but that refers to the case of
allyl alcohols, where hydrogen bonding with B plays a
special role.36 With this justification, in what follows the
values will be discussed interchangeably.
Close attention was therefore paid to the NMR spectra
taken during the course of this reaction. Signals from the
thioketone, sulfine, and ketone were easily detected. No
signal corresponding to a sultine (or sulfene) was de-
tected. Because no intermediate or any other new species
was observed, it can be concluded that the subsequent
step(s) must be much more rapid than the first.
In earlier studies, the electronic structure of the
transition state has been probed by the selection of
different substrates. For example, phosphines and sul-
Tr a p p in g of Su lfu r Mon oxid e. Sultine formation
was not proved by trapping the sulfur monoxide released;
evidence for it was the formation of elemental sulfur,
3
fides have been studied. With compounds Ar P and
ArSMe, certain adjustments can be made that will alter
the electron density at the accepting atom without
causing an important structural change. In these par-
ticular cases, the introduction of substituents in meta or
para positions provides a useful indicator. A substituent
that is electron-attracting relative to H decreases the rate
constant and vice versa. When correlated by Hammett’s
suggested to come from SO disproportionation.7,8 Elemen-
tal sulfur was not seen in the present study because even
if it was formed under these conditions, it would be
oxidized very quickly to sulfur dioxide. To establish this
mechanism, experiments were done to trap SO from the
oxidation of sulfines. Dienes were known to be able to
react with SO fast enough that SO could be trapped as
thiophene 1-oxides.19 Thus 2,3-dimethyl-1,3-butadiene
was put into the reaction mixture to trap the SO
produced from sulfine oxidation (eq 9). Because dienes
equation, the values are F ) -0.63 (PAr ) and -0.98
3
(ArSMe).
So it is with thioketones. The experimental rate
constants for Ar CdS were analyzed for the effect of
substituent X by a plot of log(k
2
X
H
3
/k
3
X
) versus 2σ . These
points define a linear relation, Figure 4, and afford the
reaction constant F ) -1.12 ( 0.16 (r ) 0.962). If the
(
9)
3
somewhat deviant value for X ) CH O is omitted, then
F ) -1.06 ( 0.07 (r ) 0.994). The value is comparable to
values obtained in other oxidations of thiobenzophe-
can also be oxidized by MTO/H
2
O
2
and in some cases
react faster than sulfines, an excess of sulfines was used
to increase the rate of oxidation of sulfines. 4,4′-Difluoro-
4,4′-dimethoxythiobenzophenone and thiobenzophenone
sulfines were used, and even though part of the dienes
were lost to oxidation, the SO trapped product 2,5-
dihydro-3,4-dimethylthiophene 1-oxide was still detected
7
nones: F ) -0.88 for perbenzoic acid and F ) -0.75 for
21
N-sulfonyloxaziridines.
The data for the thioketones reinforce the already-
established depiction of the transition state. The negative
reaction constant signifies that positive charge has
accumulated in the transition state. The high-valent
rhenium center, an electropositive Lewis acidic center,
,
1
by H NMR up to 10% yield based on diene. Because so
much diene was oxidized, a better basis is to examine
the fate of sulfur monoxide; 100% of it was trapped, about
(
36) Tetzlaff, H. R.; Espenson, J . H. Inorg. Chem. 1999, 38, 881-
885.

6
378 J . Org. Chem., Vol. 64, No. 17, 1999
Huang and Espenson
been activated electrophilically. On the other hand,
peroxides are inherently nucleophilic. Substrates with
substituents that are sufficiently electron-attracting can
be so attacked.
The suggested transition states are shown in these
diagrams:
Whether the changeover will actually be found in any
reaction series cannot be specified in advance. The matter
clearly depends not only on the peroxorhenium species,
common to any series, but also on the nature of the
reaction partner. The sulfine, unlike the sulfide and
many others, at least invites both alternatives. To our
knowledge, this is the first instance of a mechanistic
feature in the MTO activation of peroxide. The phenom-
enon has been recorded previously, however. Perhaps the
closest analogy to these results is provided by the
F igu r e 4. Hammett correlations of log k against 2σ for the
MTO-catalyzed oxidations of thiobenzophenones (upper) and
thiobenzophenone S-oxides (sulfines, lower).
polarizes the peroxide ion coordinated to it. Thus the
peroxo oxygens are less electron-rich than those in
hydrogen peroxide itself. As a result, nucleophilic centers
attack at the peroxo oxygen directly, a point that has
been affirmed by 18O labeling.37
40
oxidation of ArSCH by a peroxotitanium complex. In
3
Against this background, the kinetic data for the
oxidation of the sulfines form a marked contrast. The
Hammett plot, also displayed in Figure 4, shows distinct
curvature. Indeed, further compounds were prepared at
both extremes of σ to establish that the upward curvature
was experimentally and statistically valid over a wide
variation.
Curvature in a Hammett plot signals a certain com-
plexity in the mechanism. By the laws of adding expo-
nentials, concave curvature implicates two pathways in
parallel, whereas convex curvature is consistent with
MTO-peroxide chemistry, the only previous example of
the nucleophilic activation of peroxide appears to be in
Baeyer-Villiger reactions.41 Examples elsewhere include
the OsO
4
-catalyzed oxidation of trans-cinnamic acids by
4
2,43
chloramine-T and -B
and by pyridinium chlorochro-
4
4
mate; oxidation of benzyl alcohols by quinolinium
dichromate;45 oxidation of styrenes by dioxoruthenium-
4
6
(VI) porphyrins and quarternary ammonium perman-
ganates.47
Thioketones are oxidized so much more rapidly than
sulfines that the product of the first stage can be
produced with but little subsequent oxidation. To do so,
thioketone and peroxide were reacted in 1:1 quantities
with a catalytic amount of MTO. Some of the reactions
were monitored by 1H NMR. The spectra showed the
reaction was efficient, reaching completion in (typically)
5 min.
alternative rate-controlling steps along a single path-
way.3
8,39
The first of these options is indicated by these
results.
J ust as with the thioketones, those sulfines that have
electron-donating substituents increase in rate, with log
k varying linearly with σ. The left-hand portion of the
Hammett plot for the sulfines is as steep as the line for
the thioketones and perhaps steeper. This is reasonable
in that the sulfine, with an oxygen atom reducing electron
density on the sulfur, makes the CdSdO group of the
sulfine more electropositive than the CdS group of the
thioketone. The sulfine rates are lower, but their sensi-
tivity to substituents is somewhat greater.
Some interferences were checked by adding each of the
substances cyclohexene, PhNMe
2
, Me
2
SO, and Me
2
CO.
This material was added prior to MTO/H
of the sulfine was determined by H NMR. With these
reagents no loss of sulfine was noted.
2
O
2
. The yield
1
(40) Bonchio, M.; Calloni, S.; Di Furia, F.; Licini, G.; Modena, G.;
Moro, S.; Nugent, W. A. J . Am. Chem. Soc. 1997, 119, 6935.
The Hammett plot reverses in slope for substituents
with large σ values. This has not been seen before for
(
41) Herrmann, W. A.; Fischer, R. W.; Correia, J . D. G. J . Mol. Catal.
1994, 94, 213.
42) Sabapathymohan, R. T.; Gopalakrishnan, M.; Sekar, M. Oxid.
Commun. 1995, 18, 65.
43) Sabapathymohan, R. T.; Gopalakrishnan, M.; Sekar, M. Tet-
rahedron 1994, 50, 10945.
44) Sabapathymahan, R. T.; Gopalakrishnan, M.; Sekar, M. Tet-
rahedron 1994, 50, 10933.
45) Aruna, K.; Manikyamba, P.; Sundaram, E. V. Indian J . Chem.
1994, 33A, 854.
46) Ho, C.; Leung, W. H.; Che, C. M. J . Chem. Soc., Dalton Trans.
991, 2933.
47) Lee, D. G.; Brown, K. C.; Karaman, H. Can. J . Chem. 1986,
64, 1054.
(
the oxidations of aromatic thiones or sulfines with other
oxidizing reagents.2
1,33,38
Substituents that are electron-
(
attracting cause a lesser rate diminution and then a rate
increase. This effect can be explained in terms of a model
in which the direction of electron flow in the transition
state has been reversed. This can be explained by the
change of the mode of peroxide activation from electro-
philic at negative σ values to nucleophilic at positive σ
values. The “normal” casesfound for thioketones and for
most other reagents and for those sulfines with electron-
donating substituentssfeatures the nucleophilic reagent
attacking the peroxo oxygen. One says that peroxide has
(
(
(
1
(
(
48) Abu-Omar, M. M.; Espenson, J . H. J . Am. Chem. Soc. 1995,
1
17, 272.
(49) Al-Ajlouni, A.; Espenson, J . H. J . Am. Chem. Soc. 1995, 117,
9243-9250.
(
(
50) Hansen, P. J .; Espenson, J . H. Inorg. Chem. 1995, 34, 5839.
51) Espenson, J . H.; Pestovsky, O.; Huston, P.; Staudt, S. J . Am.
(
(
(
37) Vassell, K. A.; Espenson, J . H. Inorg. Chem. 1994, 33, 5491.
38) Leffler, J . E. J . Org. Chem. 1951, 16, 1785.
39) Exner, O. In Advances in Linear Free Energy Relationships;
Chem. Soc. 1994, 116, 2869.
(52) Abu-Omar, M. M.; Espenson, J . H. Organometallics 1996, 15,
3543.
Chapman, N. B., Shorter, J ., Eds.; Plenum: New York, 1972; pp 12-
7.
(53) Brown, K. N.; Espenson, J . H. Inorg. Chem. 1996, 35, 7211-
7216.
1

Sulfoxidation of Thioketones and Sulfines
The addition of NaHCO stopped the reaction entirely.
J . Org. Chem., Vol. 64, No. 17, 1999 6379
3
use of the facilities of the Ames Laboratory. We are
grateful to Prof. William S. J enks for helpful discussions
and thermochemical values.
This is a general phenomenon for MTO-peroxide chem-
istry, which comes about by the base-catalyzed decom-
position of A and B. (Actually, this occurs by the reaction
-
17
of HOO with MTO). This mode of decomposition is so
Su p p or tin g In for m a tion Ava ila ble: 1H NMR chemical
shifts for substituted thiobenzophenones, (XC H ) CdS, their
6 4 2
corresponding sulfines and ketones. This material is available
free of charge via the Internet at http://pubs.acs.org.
efficient that nonacidified solutions of MTO and H
not long survive intact.
2 2
O will
Ack n ow led gm en t. This research was supported by
a grant from the National Science Foundation (CHE-
9
625349). Some experiments were conducted with the
J O990621E