Fast oxidation of organic sulfides by hydrogen peroxide by in situ generated peroxynitrous acid

Full text of DOI:10.1002/(SICI)1521-3773(19980817)37:15<2088::AID-ANIE2088>3.0.CO;2-1

COMMUNICATIONS
of ONOOH) and (3) (oxidation of the substrate by
ONOOH).
Fast Oxidation of Organic Sulfides by
Hydrogen Peroxide by In Situ Generated
Peroxynitrous Acid**
For solubility reasons, organic substrates require organic
media. To our knowledge, the generation of ONOOH
according to Equation (1) in organic media has not yet been
investigated. The time dependence of the absorbance at
260 nm for the reaction of H2O2 with HNO2 in acetonitrile [6]
at 293 K is shown in Figure 1.[7] The fast formation of
Â
Â
Stephane Vayssie and Horst Elias*
Dedicated to Professor Bernt Krebs
on the occasion of his 60th birthday
The oxidation of organic sulfides RSR' by hydrogen
peroxide to form sulfoxides and, in a consecutive slower step,
sulfones is a well established procedure.[1] The rate of such
oxidations can be enhanced by the use of organic peroxides or
peroxy acids instead of H2O2 and, catalytically, by the
activation of H2O2 through coordination to metal centers.[1]
The reaction of H2O2 with nitrous acid at low pH leads to
peroxynitrous acid ONOOH,[2] an isomer of nitric acid which
is a much more powerful oxidant than H2O2.[3] The use of
ONOOH for the oxidation of organic substrates is limited by
the fact, however, that this peroxy acid is an unstable species
that isomerizes at ambient temperature with a half-life of
about 1 s to form nitric acid. Since ONOOH is also generated
by the reaction of superoxide with NO, its role in biological
systems is currently of great interest.[4]
The peroxynitrite anion ONOO is considerably more
stable than the acid ONOOH. As a result, most of the studies
with the oxidant peroxynitrous acid have been carried out in
such a way that, after the fast generation of ONOOH from
H2O2 and HNO2 at low pH,[2] the decay of ONOOH was
blocked by the addition of alkali. The cooled, stable alkaline
peroxynitrite solutions thus prepared were taken as stock
material.[5] The approach presented here is basically different
in that the short-lived oxidant ONOOH is generated in situ as
a very reactive intermediate according to Equation (1). This
Figure 1. Time dependence of the absorbance at 260 nm for the reaction of
H2O2 with nitrous acid in acetonitrile at 293 K ([H2O2] ˆ 5mm, [NaNO2] ˆ
0.5mm, [HClO4] ˆ 0.1m, I ˆ 0.6m (NaClO4), [H2O] ˆ 5m). The rate con-
1
1
stants kf ˆ 1.2 s
and kd ˆ 0.61 s
for the formation and decay of
peroxynitrous acid were obtained from the sum of two exponentials[7]
(solid line).
ONOOH with an experimental pseudo-first-order rate con-
1
stant kf' ˆ 1.2 Æ 0.2 s is followed by its first-order decay to
1
HNO3 with kd ˆ 0.61 Æ 0.05 s . Under comparable conditions,
the rate constant kd for the reaction in water is 1.07 Æ
kf
1 [9]
HNO2 ‡ HOOH ! ONOOH ‡ H2O
(1)
(2)
(3)
0.05 s . This means that acetonitrile (with an admixture of
5m H2O) is suitable for the generation of ONOOH according
to Equation (1), and that the rate of decay of ONOOH in
acetonitrile is close to that in water.
kd
ONOOH ! HNO3
kA
ˆ
A ‡ ONOOH ! A O ‡ HNO2
We have tested the practical applicability of in situ
generated peroxynitrous acid as an oxygen-transfer agent in
acetonitrile for several classes of organic substrates under
standardized conditions.[10] These conditions were such that
there was a 2.5-fold excess of organic substrate relative to
H2O2, and the reaction was monitored cerimetrically by the
loss of H2O2. Houk et al.[11] reported recently the results of
calculations suggesting that oxygen transfer from ONOOH to
ethylene is energetically feasible. This led us to study the
epoxidation of several olefins according to reaction sequence
(1) ± (3) at [HNO2]:[H2O2] ˆ 1:4.4. The result was negative in
the sense that the consumption of H2O2 for substrate
oxidation was zero for cyclohexene and allyl alcohol, and
very minor for 2-methoxypropene, ethylvinyl ether, and 2,3-
dimethyl-2-butene. Under the given conditions, the reaction
of these olefins with ONOOH is apparently too slow to
compete with the isomerization of ONOOH to HNO3
[Eq. (2)]. It was not surprising to find that N-oxidation of
amines such as pyridine and substituted pyridines was not to
be observed. In the given acidic medium, protonation reduces
the concentration of free amine to a minimum.
reaction is a fast process. After its generation, ONOOH is
used for the fast second-order oxidation of suitable organic
substrates A [Eq. (3)]. The oxidation competes with the decay
of ONOOH [Eq. (2)]. As a consequence, this approach is
applicable only when the rate of reaction (3) is higher than
that of (2), that is when kA[A] is much greater than kd. One
should note that reactions (1) and (3) form a catalytic cycle, in
which the catalyst HNO2 converts H2O2 into the more
powerful oxidant ONOOH. The number of cycles is con-
trolled by the competition between reactions (2) (decay
Â
[*] Prof. Dr. H. Elias, Dipl.-Ing. S. Vayssie
Institut für Anorganische Chemie der Technischen Universität
Petersenstrasse 18, D-64287 Darmstadt (Germany)
Fax: (‡49)6151-166-040
E-mail: elias@hrz1.hrz.tu-darmstadt.de
[**] This work was supported by the Deutsche Forschungsgemeinschaft,
the Verband der Chemischen Industrie, and the Otto-Röhm-Stiftung.
This contribution presents the main results of the Dr.-Ing. Disserta-
tion to be submitted to the Technische Universität Darmstadt, D17, by
S.V.
2088
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
1433-7851/98/3715-2088 $ 17.50+.50/0
Angew. Chem. Int. Ed. 1998, 37, No. 15

COMMUNICATIONS
Organic sulfides RSR' turned out to be very suitable
substrates for the fast sulfoxidation with in situ generated
ONOOH [Eq. (4)]. As shown in Figure 2 for the system
RSR' ‡ ONOOH ! RSOR' ‡ HNO2
(4)
PhSMe/H2O2/NaNO2, the sulfoxidation of PhSMe in the
absence of nitrite is a rather slow process.[12] The addition of
15mm of nitrite triggers an instantaneous, partial oxidation of
Figure 3. Time dependence of the concentration of H2O2 for the reaction
of Me2S with H2O2 in acidified acetonitrile at various concentrations of
Me2S at 208C ([H2O2] ˆ 0.005m, [NaNO2] ˆ 0.0005m, [H2SO4] ˆ 0.08m,
[H2O] ˆ 5m; [Me2S] ˆ 0m (a), 0.005m (b), 0.010m (c), 0.050m (d)).
kMe S[Me2S] ꢀ kd is fulfilled and all of the H2O2 is consumed
2
by substrate oxidation within one minute. These results allow
1
the estimate[17] that kMe S is greater than 102 m 1 s . This means
2
that, compared to the second-order rate constant k for the
noncatalyzed Me2S oxidation by H2O2 (Table 1), the reactivity
Table 1. Kinetic data for the sulfoxidation of organic sulfides by H2O2 in
acidified acetonitrile in the absence and presence of nitrous acid.
Figure 2. Time dependence of the concentration of H2O2 for the reaction
of PhSMe with H2O2 in acidified acetonitrile at variable concentrations of
HNO2 under standard conditions[10] ([H2O2] ˆ constant ˆ 0.2m;
[NaNO2] ˆ 0m (a), 0.015m (b), 0.020m (c), 0.030m (d)).
[b]
Sulfide
104 k[a]
[m 1 s
([HNO2]:[H2O2])lim
Catalytic
cycles
1
]
[mm]:[mm]
PhSMe
(PhCH2)2S
Me2S
31 Æ 3
70 Æ 6
30:200
20:200
10:200
6.7
10
20
PhSMe, as monitored by a sharp drop in the H2O2 concen-
tration. The size of this instantaneous drop increases with the
concentration of nitrite. For [NaNO2] ˆ 30mm (which corre-
sponds to [H2O2]: [NaNO2] ˆ 6.7:1), all of the H2O2 oxidant is
consumed within one minute. This means that, in terms of
catalysis, one molecule of HNO2 passes almost seven cycles
according to reactions (1) and (3), thus ªactivatingº seven
molecules of H2O2 by the formation of ONOOH. Owing to
the greater reactivity of ONOOH relative to that of H2O2, the
sulfoxidation of PhSMe is drastically accelerated.
580 Æ 50
[a] Second-order rate constant k for the sulfoxidation at 293 K in the
absence of the catalyst under standard conditions.[10] [b] Limiting concen-
tration ratio at which (and above which) catalyzed sulfoxidation is
complete within one minute under standard conditions.[10]
of ONOOH is more than three orders of magnitude greater
than that of H2O2.
Product formation in the systems PhSMe/H2O2/NaNO2 and
Me2S/H2O2/NaNO2 was additionally controlled by GC anal-
ysis.[18] Under standard conditions ([substrate] > [H2O2]),[10]
the sulfoxides PhSOCH3 and CH3SOCH3 were the only
products. Further oxidation to the corresponding sulfones is
slower by several orders of magnitude than the sulfoxidation
[Eq. (4)],[3] and is only observed when the substrate concen-
tration is less than that of H2O2 and with long reaction times.
The sulfoxidation of dibenzyl sulfide was also studied in
detail. It follows from the data summarized in Table 1 that, at
[H2O2]:[NaNO2] ˆ 10:1, formation of dibenzyl sulfoxide is
complete within one minute. (PhCH2)2S was chosen as a
model compound to test the synthetic potential of the method
for the fast preparation of sulfoxides at the gram level. An
aqueous solution of NaNO2 was quickly added to an acidified,
well-stirred acetonitrile solution of (PhCH2)2S containing an
excess of H2O2.[19] After a reaction time of only five minutes at
ambient temperature, pure dibenzyl sulfoxide was isolated in
a yield of 91%.
The sulfoxidation of dimethyl sulfide according to Equa-
tions (1) ± (3) is accelerated by nitrite in an analogous manner.
The noncatalyzed oxidation in the absence of nitrite is faster
by a factor of 19 than that of PhSMe. Correspondingly, a
nitrite concentration of only 10mm ([H2O2]:[NaNO2] ˆ 20:1)
is sufficient to make the oxidation of Me2S complete within
one minute. These data mean that the catalyst HNO2 cycles
twenty times. As outlined above, the condition for a maximum
number of cycles in the system Me2S/H2O2 is given by
kMe S[Me2S] ꢀ kd.[14] The validity of this condition was tested
2
by a series of experiments with variable concentrations of
Me2S and constant concentrations of H2O2 and NaNO2. As
shown in Figure 3, in the absence of Me2S there is an initial
loss of 10% of H2O2. This loss corresponds to the conversion
of HNO2 into ONOOH, which decays to HNO3 with the rate
constant kd.[16] At [Me2S] ˆ 5mm the oxidation of Me2S by
ONOOH begins to compete with the decay of ONOOH,
an indication that the size of kMe S[Me2S] is getting closer to
2
In conclusion, the present approach provides an exper-
imentally simple and time-saving method for the fast and
convenient preparation of sulfoxides. The method is based on
that of kd. This effect is more pronounced at [Me2S] ˆ
10mm. Finally, at [Me2S] ˆ 50mm the peroxynitrous acid
formed reacts preferentially with Me2S; the condition
Angew. Chem. Int. Ed. 1998, 37, No. 15
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
1433-7851/98/3715-2089 $ 17.50+.50/0
2089

COMMUNICATIONS
[14] The reaction between Me2S and ONOOH in water was recently
shown to be second-order, first-order in both reactants.[3] The
oxidation of organic sulfides by hydroperoxides is generally found
to be a second-order process.[15]
[15] J. O. Edwards in Peroxide Reaction Mechanisms (Ed.: J. O. Edwards),
Wiley-Interscience, New York, 1962, pp. 67 ± 106.
nonexpensive, commercial chemicals (H2O2, NaNO2, H2SO4,
and CH3CN) and can be classified as a sulfoxidation reaction
with ªactivatedº H2O2. The activation consists of the fast
reaction of H2O2 with HNO2 in acidified CH3CN to form the
unstable, yet highly reactive, oxidant peroxynitrous acid
ONOOH as an intermediate species in situ. The experimental
conditions can be easily set in such a way that the reaction of
the electrophile ONOOH with the nucleophile RSR' is fast
enough to compete successfully with the decay of ONOOH to
nitric acid. Under these optimized conditions, HNO2 acts as a
catalyst, quickly carrying oxygen atoms from H2O2 to the
sulfide in the form of ONOOH.
[16] It was confirmed by experiment that ONOOH, when generated
according to Equation (1), does not decompose H2O2 according to the
reaction ONOOH ‡ H2O2 !HNO2 ‡ H2O.
[17] At the lowest Me2S concentration, [Me2S] ˆ 5mm, almost half of
the H2O2 is consumed for the fast Me2S oxidation by ONOOH
(see Figure 3). This means that the Me2S concentration at
which kMe S [Me2S] ˆ kd is smaller than 5mm and, hence, kMe S
>
2
2
0.61/0.005 ˆ 122m 1 s
.
1
[18] GC analysis was carried out with an Auto System Gas Chromatograph
with FI detector (Perkin-Elmer) and
a 15-m capillary column
(Alltech; Heliflex AT-1000) at 1508C (Me2S) and 1708C (PhSMe).
The samples were neutralized with NaOH before injection.
[19] Concentrated H2SO4 (1 mL, 98%) and technical grade, stabilized
35% H2O2 (4 mL, 47 mmol) were added to a well-stirred solution of
(PhCH2)2S (5.3 g, 25 mmol) in CH3CN (250 mL) at ambient temper-
ature. To start the fast catalyzed reaction, a solution of NaNO2
(320 mg, 4.7 mmol) in water (2 mL) was added in one portion
(caution: the reaction is exothermic! In the case of larger batches
the reaction mixture should be efficiently thermostated). After a
reaction time of five minutes the solution was set to pH 7 with a
solution of NaOH (1m) diluted with water (70 mL) and extracted with
chloroform (3 Â 130 mL). The chloroform phase was dried with
MgSO4 and taken to dryness in vacuo. The residue was recrystallized
from EtOH. Yield: 5.1 g (91%). M.p. 1338C (134 ± 1358C[20]). 1H
NMR and IR spectroscopy as well as C,H,N analysis proved the
identity of the product as dibenzyl sulfoxide.
Received: February 23, 1998 [Z11508IE]
German version: Angew. Chem. 1998, 110, 2246 ± 2249
Keywords: homogeneous catalysis
´ O ± O activation ´
oxidations ´ peroxides ´ synthetic methods
[1] S. Uemura in Comprehensive Organic Synthesis, Vol. 7 (Eds: B. M.
Trost, I. Fleming, S. V. Ley), Pergamon, Oxford, 1991, pp. 762 ± 769.
[2] W. G. Keith, R. E. Powell, J. Chem. Soc. A 1969, 90.
[3] P. Amels, H. Elias, K.-J. Wannowius, J. Chem. Soc. Faraday Trans.
1997, 93, 2537 ± 2544.
.
[4] In vivo, the reaction of O2 with NO leads to the anion ONOO , which
is in equilibrium with the acid ONOOH. The reaction of superoxide
with NO is faster than that with superoxide dismutase (SOD). For the
biological relevance of ONOOH see, for example, W. A. Pryor, G. L.
Squadrito, Am. J. Physiol. 1995, 268, L699 ± L722.
[20] Handbook of Chemistry and Physics, 51st ed., The Chemical Rubber
Co. 1970 ± 1971.
[5] See, for example, a) R. Radi, J. S. Beckman, K. S. Bush, B. A.
Freeman, J. Biol. Chem. 1991, 266, 4244 ± 4250; b) W. A. Pryor, X. Jin,
G. L. Squadrito, Proc. Natl. Acad. Sci. USA 1994, 91, 11173 ± 11177;
c) A. Al-Ajlouni, E. S. Gould, Inorg. Chem. 1996, 35, 7892 ± 7896; d) S.
Goldstein, G. Czapski, Inorg. Chem. 1995, 34, 4041 ± 4048.
[6] The use of commercial H2O2 as a reagent necessarily introduces some
water into the system. The solvent acetonitrile with a constant
admixture of water (5m) was therefore taken as the standard reaction
medium.
[7] The absorbance/time data shown in Figure 1 were obtained with a
multi-wavelength stopped-flow spectrophotometer described earli-
er.[8] The absorptivity of ONOOH in the range 250 ± 350 nm is greater
than that of nitrous and nitric acid. The initial increase in absorbance
A at 260 nm corresponds therefore to the formation of ONOOH, the
consecutive decrease to its decay. The rate constants kf and kd were
obtained from Equation (5) using the A/t data (a1, a2 ˆ amplitudes).
Synthesis, Structure, and Reactivity of a
Palladium Hydrazonato Complex: A New Type
of Reductive Elimination Reaction To Form
C N Bonds and Catalytic Arylation of
Benzophenone Hydrazone**
John F. Hartwig*
Reductive elimination reactions which form C N bonds in
amines[1±4] are important primary reactions in practical
catalytic cycles for the synthesis of arylamines from aryl
halides.[1, 5±8] Reductive elimination reactions that result in N-
arylhydrazones would be an important new method for C N
bond formation by means of reductive elimination. The
hydrazone products could be used in Fischer indole syntheses
or, after conversion into the N-arylhydrazine, in condensation
reactions to produce N-arylpyrazoles and N-arylpyrazolones.
A ˆ a1 exp( kf t) ‡ a2 exp( kd t) ‡ A1
[8] C. Drexler, H. Elias, B. Fecher, K. J. Wannowius, Fresenius J. Anal.
Chem. 1991, 340, 605 ± 615.
(5)
[9] D. J. Benton, P. J. Moore, J. Chem. Soc. A 1970, 3179 ± 3182.
[10] Standard procedure and conditions: A solution of H2O2, H2SO4, and
H2O in CH3CN (1 mL) was quickly mixed with a solution of the
organic substrate A and NaNO2 in CH3CN (1 mL) to obtain the
reaction mixture with [H2O2] ˆ 0.2m, [HNO2] ˆ 0 ± 45mm, [A] ˆ 0.5m,
[H2SO4] ˆ 0.08m, and [H2O] ˆ 5m (the pKa of HNO2 is 3.3; to liberate
‡
HNO2 from NaNO2, the condition lg[H ] < 3 has to be fulfilled).
The mixture was stirred at ambient temperature. At adequate time
intervals, samples (0.2 mL) were taken and diluted with water for the
cerimetric determination of H2O2.
[*] Prof. J. F. Hartwig
Department of Chemistry, Yale University
P.O. Box 208107, New Haven, CT 06520-8107 (USA)
Fax: (‡1)203-432-6144
[11] K. N. Houk, K. R. Condroski, W. A. Pryor, J. Am. Chem. Soc. 1996,
118, 13002 ± 13006.
1
[12] The value of k ˆ (31 Æ 3)  10 4 m 1 s (see Table 1) was obtained by
E-mail: john.hartwig@yale.edu
fitting of the [H2O2]/t data shown in Figure 2a to Equation 6. This
[**] I thank the Department of Energy and the NIH for supporting
portions of this work. I also appreciate support through a National
Science Foundation Young Investigator Award, a Camille Dreyfus
Teacher-Scholar award, and the Eli Lilly Grantee Program. J.F.H. is a
fellow of the Alfred P. Sloan Foundation.
[H2O2] ˆ [H2O2]o exp( k[PhSMe]t)
(6)
value is very close to k ˆ 37.3  10 4 m 1 s as reported for the
1
sulfoxidation of PhSMe with H2O2 in EtOH (6% H2O).[13]
[13] G. Modena, L. Maiola, Gazz. Chim. Ital. 1957, 87, 1306 ± 1316.
2090
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
1433-7851/98/3715-2090 $ 17.50+.50/0
Angew. Chem. Int. Ed. 1998, 37, No. 15