Oxidation of thioanisole by hydrogen peroxide: Activation by nitrites

Article abstract of DOI:10.1002/poc.646

The oxidation of thioanisole (PhSMe) by H₂O₂ is activated by acetonitrile (MeCN) and propionitrile (EtCN) and involves the formation of a transient peroxyimidate, 1, by reaction of HO₂ ^- and RCN, and 1 can be rapidly trapped by PhSMe. The rate of oxidation of PhSMe is then independent of the concentrations of PhSMe and of H₂O₂, but varies linearly with [HO₂ ^-] and [RCN]. In very dilute PhSMe it and H₂O₂ compete in reacting with 1, and the rate then depends on [PhSMe]. The initial reaction gives PhSOMe, and subsequent formation of PhSO₂Me is slow. The rates of oxidation are slightly higher than that expected from the MeCN-activated decomposition of H₂O₂, which involves a second molecule of H₂O₂ in conversion of the peroxyimidate into amide and oxygen. Copyright

Full text of DOI:10.1002/poc.646

JOURNAL OF PHYSICAL ORGANIC CHEMISTRY
J. Phys. Org. Chem. 2003; 16: 603–607
Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/poc.646
Oxidation of thioanisole by hydrogen peroxide:
activation by nitriles
Nicholas D. Gillitt, Josiel Domingos and Clifford A. Bunton*
Department of Chemistry and Biochemistry, University of California, Santa Barbara, California 93106-9510, USA
Received 29 October 2002; revised 4 March 2003; accepted 5 March 2003
ABSTRACT: The oxidation of thioanisole (PhSMe) by H O is activated by acetonitrile (MeCN) and propionitrile
2
2
ꢀ
EtCN) and involves the formation of a transient peroxyimidate, 1, by reaction of HO and RCN, and 1 can be rapidly
2
(
trapped by PhSMe. The rate of oxidation of PhSMe is then independent of the concentrations of PhSMe and of H O ,
2
2
ꢀ
but varies linearly with ½HO ꢁ and [RCN]. In very dilute PhSMe it and H O compete in reacting with 1, and the rate
2 2
2
then depends on [PhSMe]. The initial reaction gives PhSOMe, and subsequent formation of PhSO Me is slow. The
2
rates of oxidation are slightly higher than that expected from the MeCN-activated decomposition of H O , which
2
2
involves a second molecule of H O in conversion of the peroxyimidate into amide and oxygen. Copyright # 2003
2
2
John Wiley & Sons, Ltd.
KEYWORDS: sulfide oxidation; thioanisole; nitriles; peroxyimidates; zero-order kinetics
INTRODUCTION
were interpreted in terms of competition between the
0a
1
sulfide and H O for the short-lived intermediate, 1,
2
2
Decontamination of chemical weapons under mild condi-
tions is complicated, because although nucleophiles react
rapidly with phosphonofluoridate nerve agents, decompo-
sition of the chlorosulfide blister agent HD (Mustard),
involves oxidation, or the use of strong bases in aprotic
which we show as a peroxyimidate, although we do not
know the extent of its protonation (Scheme 1).
1
ꢀ
media. The hydroperoxide ion, HO , formed by depro-
2
2
tonation of H O , is a very reactive nucleophile, but H O
2
2
2
2
Scheme 1
is a weak oxidant towards sulfides and has to be activated,
for example, by conversion into a peroxy acid. Hydro-
gencarbonate ion and H O generate a peroxocarbonate
Competitive trapping in a mixture of sulfides was favored
by electron-donating substituents in MeOH–K CO ,
although there was no such effect on trapping by the
2
2
2
3
ꢀ
which is an effective oxidant at a pH where HO is a
3
2
nucleophile. Molybdate ion and H O generate a variety
2
2
less reactive sulfoxides and other reactions might have
10a
4
of peroxomolybdates, and the tetraperoxomolybdate dia-
intervened.
For example, organic solvents promote
ꢀ
nion rapidly oxidizes sulfides at pHꢂ 10 where HO
ꢀ
formation of peroxocarbonates from HCO and H O ,
3
2 2
3
2
5
,6
reacts nucleophilically. Decomposition of peroxomo-
lybdates, especially the triperoxomolybdate dianion, also
generates singlet oxygen, which reacts rapidly with many
and the reaction of 1 with H O could generate singlet
2
2
7,11
oxygen, which would react rapidly with sulfides.
We therefore examined the kinetics of oxidation of
7
organic compounds, including sulfides.
ꢀ
thioanisole by nitrile-activated H O2 in water with
2
The reaction of HO with nitriles generates short-lived
peroxyimidates which convert alkenes into epoxides and
2
[H O ] ꢃ [NaOH] to avoid possible catalysis by carbo-
2
2
3
nate or hydrogencarbonate ion. Under these conditions
8–10
oxidize sulfides to sulfoxides and sulfones. Mixtures
of H O and nitriles are preparatively useful as alter-
natives to peroxy acids in oxidations of alkenes and
sulfides.
Oxidations of arylmethyl sulfides by acetonitrile–H O
have been examined in methanol–K CO and the results
ꢀ
[
HO ] is approximately equal to [NaOH]. The postulated
2
2
2
mechanism of activation indicates that reactions should
be zeroth order in [PhSMe] provided that it is at a
sufficient concentration to compete effectively with
H O for 1. Reaction products were identified by their
9
,10
2
2
2
2
1
2
3
H NMR spectra.
*
Correspondence to: C. A. Bunton, Department of Chemistry and
Biochemistry, University of California, Santa Barbara, CA 93106-
510, USA. E-mail: bunton@chem.ucsb.edu
EXPERIMENTAL
9
Contract/grant sponsors: US Army Office of Research; US–Brazil
Cooperative Program; Contract/grant number: INT-9722458.
Contract/grant sponsor: CAPES.
Materials. Hydrogen peroxide, 30% or 50%, stabilized or
unstabilized, was used as in earlier work, and the results
5
Copyright # 2003 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2003; 16: 603–607

6
04
N. D. GILLITT, J. DOMINGOS AND C. A. BUNTON
were the same with the different samples. Thioanisole
5
was material used earlier, acetonitrile (MeCN) and
NMR spectroscopy. Products of reaction in 5 vol.%
MeCN were identified by their H NMR spectra in a
1
1
Varian Inova (Unity) instrument (400 MHz for H), with
propionitrile (EtCN) (Aldrich) were of spectrophoto-
metric grade and benzonitrile was obtained from AC-
ROS. Solutions were prepared in redistilled, deionized,
1
suppression of the signal of H O. Chemical shifts were
5a
2
referred to external sodium 3-(trimethylsilyl) propionate
(TSP) in D O, and are: PhSOMe, 7.70 (3H), 7.79 (2H)
ꢄ
water and reactions were followed at 25.0 C.
2
and CH , 2.94 ppm; PhSO Me, o-H, 8.02 (d), m-H, 7.75
3
2
Kinetics. Reactions with MeCN or EtCN were followed
spectrophotometrically on an HP 8451A spectrometer by
measuring the decrease in absorbance as PhSMe is
(t), p-H, 7.85 (t) and CH , 3.32 ppm; PhSMe, 7.42 (4H,
3
multiplet), 7.28 (1H, t) and CH , 2.55 ppm. Signals of the
3
amides were: acetamide, CH , 2.04 (s) ppm, and for
3
12
1
oxidized. The wavelength was varied between 270
and 300 nm, depending upon [PhSMe] and [H O ]. The
propionamide: H, 1.12 ppm (t), J ¼ 8 Hz and 2.88 ppm
1
3
(q); C, 12.19 (CH CH ꢀ) and 31.26 (CH CH ꢀ) ppm.
2
2
3
2
3
2
absorbance change during reaction increases with
decreasing wavelength, but the absorbance of H O at
The product of the reaction in the presence of PhCN was
identified as PhSOMe, by its CH signal.
2
2
3
low wavelengths is such that we followed the reaction
with 1 M H O at 290 and 300 nm.
2
2
RESULTS AND DISCUSSION
Kinetics
Plots of absorbance against time were linear for up to
over 90% reaction with dilute H O and higher [PhSMe],
but the extent of linearity decreased as [H O ] was
2
2
2
2
increased and [PhSMe] decreased. However, the number
of data points in the linear region (30–70) was adequate
for calculation of the slopes. The slopes were converted
ꢀ
2
13
Both H O and HO absorb at up to 320 nm, but except
2
2
with high [H O ] we followed the oxidations at 270 and
2
80 nm. Very little sulfone was formed, and it was
2
2
into reaction rates, i.e. zeroth-order rate constants, k
0
1
ꢀ
1
ꢀ1
identified by its H NMR spectrum, which was taken
0–15 min after initiation of the reaction and was never
more than 5% of PhSOMe.
With H O in large excess over NaOH, [HO ] ꢂ
l mol s , by extrapolating absorbance back to the time
of mixing and equating the decrease of absorbance in the
course of reaction with the initial concentration of
PhSMe. The reaction was started by adding PhSMe in a
small volume of MeCN or EtCN to the stirred reaction
mixture with a spring-loaded Hamilton syringe.
1
ꢀ
2
2
2
13
[
NaOH], but we made the (small) correction by taking
ꢀ
3
the base dissociation constant K ¼ 3 ꢆ 10 , and using
b
the equation
Formation of bubbles due to decomposition of perox-
ides is a problem when reactions are followed spectro-
5
photometrically. We did not have this problem in the
½NaOHꢁ
ꢀ
½
HO ꢁ ¼
ð1Þ
2
1
þ K =½H O ꢁ
present work provided that PhSMe is trapping all the
peroxyimidate, but there is bubbling once PhSMe has
reacted, and absorbance at complete reaction of PhSMe
was estimated after tapping the cuvette and using a gentle
vacuum to remove bubbles. However, we could not
follow the reaction with [H O ] > 1 M.
b
2
2
We assume that K
by dilute MeCN or EtCN and H
will not be significantly affected
acts as its own
b
O
2 2
2c
buffer.
All the reactions give good linear plots of absorbance
against time for much of the reaction, but there is
uncertainty in the absorbance after complete reaction of
PhSMe, in part because of bubbling, but also because of
2
2
In all experiments H O and nitrile were in large
2
2
excess over PhSMe and their concentrations did not
change significantly during reaction. As a control, absor-
bance was monitored in the absence of PhSMe and the
decrease was small relative to that with PhSMe. The
reference solution contained acetamide and NaOH,
although their absorbances should be negligible at 270–
the continuing decomposition of H
relatively unimportant (Experimental). For reaction in
5 vol.% MeCN, 0.1 M H and 0.001 M NaOH, oxidation
O , which should be
2 2
O
2 2
of PhSMe is complete within ca 100 s, and in that time
with no PhSMe the absorbance decreases by 0.001 and
0.0006 at 270 and 280 nm, respectively. Decreases in
absorbance due to oxidation of PhSMe shown in Fig. 1
are very much larger than those due to decomposition of
2
80 nm.
The decomposition of H O by reaction with MeCN
was monitored titrimetrically following the general pro-
2
2
8
cedure of Wiberg. Aliquots were removed and the
reaction was quenched by 3 M H SO , KI was added
2
H O , and some of the data points have been omitted for
2 2
4
and I was titrated against Na S O , after addition of
2
clarity. These figures also illustrate how the extents of
linearity in plots of absorbance against time depend on
reactant concentrations.
2 2 3
0
.001 M Na MoO . In one experiment samples were
2 4
analyzed immediately after the acid quench, but gener-
ally analysis was made after removal of all the samples.
Reactions were followed for up to 40% in dilute H O
Values of k for reaction in MeCN–H O are in given
0 2
Table 1 and those for reaction mediated by EtCN in
Table 2, with [EtCN] equivalent to [MeCN] in 5 vol.%
MeCN.
2
2
and ꢅ10% in more concentrated H O , to limit the
2
2
depletion of MeCN.
Copyright # 2003 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2003; 16: 603–607

OXIDATION OF THIOANISOLE BY HYDROGEN PEROXIDE
Table 2. Reaction with propionitrile
605
a
4
0 [PhSMe] (M)
1
2.81
9.07 (10.3)
4.50
10.50 (9.88)
16.5
11.5
6
0 k (l mol
ꢀ1 ꢀ1
s )
1
0
a
ꢄ
At 25 C, 6.83 vol.% EtCN, 0.1 M H
measured at 280 nm (270 nm in parentheses).
2
O
2
and 0.001 M NaOH. Rates were
a
Table 3. Products without and with NaOH
Time (min) PhSMe PhSOMe PhSO Me MeCONH2
2
b
5
2
1
87
60
13
40
65
27
b
0
0
15
35
35
73
100
188 (135)
333 (173)
1060 (200)
Figure 1. Variation of absorbance at 270 nm with time for
ꢄ
reaction of thioanisole at 25.0 C, 0.1 M H O , 0.001 M NaOH
2
2
ꢀ4
in H O–MeCN (95:5, v/v) for 14.1 ꢆ 10
(*) and
2
a
ꢄ
2 2
Mole % of reactants and products, at 25 C, 5 vol.% MeCN, 0.1 M H O ,
ꢀ
4
2
.80 ꢆ 10 (*) M PhSMe
0
.001 M PhSMe and 0.001 M NaOH.
Without NaOH.
b
The data in Table 1 show that k increases with
0
ꢀ
1
increasing [MeCN] and [HO ] and, except for a few
ꢀ
oxide, and was monitored by H NMR spectroscopy
(Table 3). The extent of reaction is in the range expected
2
outlying points, values of k /[MeCN][HO ] are not very
0
2
sensitive to [H O ] or [PhSMe]. There are uncertainties
5a
from data for reaction in H O–t-BuOH and the rate
2
2
2
in k due to experimental limitations in following the
0
constant extrapolated from data for the acid-catalyzed
oxidation of 4-nitrothioanisole.
The reaction was started in a neutral solution where
14
reaction spectrometrically, and difficulties in estimating
absorbances after complete oxidation of PhSMe (Experi-
mental). We assume that MeCN and EtCN are not
exerting significant kinetic solvent effects, and it is
oxidation of PhSMe is slow and no MeCOꢇNH is
2
formed. After 5 min, 0.001 M NaOH was added to one
portion. All the PhSMe had been oxidized within 5 min
after addition of NaOH, the time required for acquisition
of the NMR spectrum, and with time PhSOMe was
difficult to decide whether effects of an increase in
ꢀ
[PhSMe] on values of k and k /[MeCN][HO ] are real
or due to experimental error, because there are differ-
0
0
2
ences in k from experiments at different wavelengths.
0
converted into PhSO Me (Table 3), although this reaction
2
1
5
However, with a few exceptions, values of k /
0
does not necessarily involve the peroxyimidate. Acet-
amide is formed after addition of NaOH to an extent
greater than that expected in terms of oxidations of
ꢀ
[
7
MeCN][HO ] are within 10% of the mean value of
s (Table 1).
2
.8 ꢆ 10ꢀ l mol
4
ꢀ1 ꢀ1
8
2
sulfide and sulfoxide due to reaction of 1 with H O
2
Slow reactions with and without NaOH
(Table 3). The amounts of acetamide in parentheses are
calculated on the assumption that the overall formation of
PhSOMe generates 1 equiv. and of PhSO Me 2 equiv. of
There is a slow oxidation of PhSMe by H O in H O–
2
2
2
2
MeCN in the absence of NaOH, which gives only sulf-
MeCONH . The excess MeCONH increases sharply
2
2
a
Table 1. Reactions with MeCN
4
6
10 k
(l mol
4
10 k /[MeCN][HO
(l mol
ꢀ
MeCN (vol.%)
[NaOH]
M)
[H O ]
2
(M)
10 [PhSMe]
(M)
]
2
0
1
0
2
ꢀ
ꢀ1
ꢀ1 ꢀ1
s )
(
s
)
2
.5
.0
0.001
0.001
0.1
3.10
.90
3.14 (3.24)
3.40 (3.25)
3.84 (3.44)
6.69 (7.11)
8.49 (7.87)
7.03 (7.42)
8.50 (8.67)
8.00 (7.61)
11.3 (12.1)
13.4 (13.6)
17.2 (15.5)
7.54 (7.29)
7.18 (6.29)
15.6 (14.3)
6.9
4
7.2
7.8
7.4
9.5
7.8
9.2
8.4
6.3
7.3
9.0
7.8
7.1
7.8
1
5.5
5
0.01
4.50
4.1
1
0
.1
2.80
4
.50
14.1
2.80
0
0
.002
.001
0.1
0.5
1.0
0.1
4
.50
1
4.1
b
15.5
15.5
15.5
b
1
0.0
a
ꢀ1 ꢀ1
s
Followed at 300 nm (290 nm in parentheses).
Initial zero-order rate constants, l mol
, followed at 280 nm (270 nm in parentheses unless specified otherwise).
b
Copyright # 2003 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2003; 16: 603–607

6
06
N. D. GILLITT, J. DOMINGOS AND C. A. BUNTON
towards the end of the reaction, which is the MeCN-
8
oxidation faster or decomposition of H O slower than
2 2
promoted decomposition of H O .
2
expected from the reactions shown in Scheme 1.
There is a direct reaction of H O with PhSMe, which
2
2
2
from other data we expect to be of minor importance
except with the highest [PhSMe], and if we exclude these
values (Table 1) the ratio increases to 1.8:1, which is
almost within the scatter of the data.
Kinetic form
The linearity of the plots of absorbance against time
Fig. 1), provided that [PhSMe] is sufficient for quanti-
(
Another possibility is that we are underestimating the
ꢀ
tative trapping of intermediate 1, is as expected for the
reactions in Scheme 1, and is inconsistent with significant
contributions by other reactions.
rate of the reaction of HO with MeCN in the absence of
2
PhSMe, and two factors could be involved. First, the
formation of peroxyimidate, 1, may be reversible, but its
trapping by PhSMe is generally faster than reversion to
ꢀ
HO and MeCN, and decomposition by attack of H O is
2
2 2
Decomposition of peroxyimidate, 1, by reaction with
H O should generate O , which should rapidly oxidize
1
2
2
2
7,11
PhSMe,
but there is also unproductive conversion
not fast enough to prevent some return in the absence of
PhSMe. Second, the decomposition of peroxy acids
involves attack of a peroxy anion on the peroxy acid,
1
of O into O , and partitioning should depend on
3
2
2
[
PhSMe], which would preclude the observation of
1
15
zeroth-order kinetics and the rate of formation of O2
would depend on [H O ]. We conclude that this reaction
e.g. reaction of H O with anionic peroxyimidate, but
2
2
2
2
in the absence of PhSMe, [peroxyimidate] may be such
that decomposition involves reaction of peroxyimidic
acid and peroxyimidate. In that event, the rate of decom-
position of H O would be the rate of formation of
path could, at most, make only a minor contribution to the
oxidation of PhSMe. The direct oxidation of PhSMe by
H O , or formation of a peroxyimidic species from non-
2
2
2 2
ionic H O and MeCN, appear to be much slower than the
2
peroxyimidate, 1, rather than twice that rate as in the
classical reaction in the absence of PhSMe.
We cannot distinguish between these explanations and
in view of the uncertainties in the rate data (Table 1) the
discrepancies between the two sets of data may not be
real.
2
8
reactions at higher pH (Tables 1 and 3). However,
reaction of H O with PhSMe (Table 3) should be first
2
2
order rather than zeroth-order in [PhSMe], and might
make a minor contribution at higher [PhSMe], although
the rate would also depend significantly on [H O ] and
2
2
we see no such effect (Table 1).
Reactivity of peroxyimidates
Decomposition of H O2
2
9
,10
Peroxyimidates react rapidly with alkenes and sulfides
In agreement with existing evidence, the rate of decom-
ꢀ1
and the observation of zeroth-order kinetics in the oxida-
tion of PhSMe shows that peroxyimidates, or the imidic
acids, are considerably more reactive than typical peroxy
acids. Sulfide oxidations by peroxycarboxylic acids and
by peroxymonosulfate ion are first order in sulfide and
ꢀ
1
position of H O , v l mol s , depends on [MeCN] and
2
2
ꢀ
HO ], but not on [H O ], and similar values are obtained
2
2 2
[
titrimetrically and spectrophotometrically (Table 4).
Provided that PhSMe traps peroxyimidate, 1, quantita-
tively, the rate of oxidation, k (Table 1), should be half
0
1,12,16,17
can be followed by conventional methods.
For
the rate, v, in the absence of PhSMe (Table 4), but the
relative values are 1.6:1, based on a mean value of v/
example, oxidations by the peroxocarbonate ion are first
order in sulfide and can be followed readily at room
ꢀ
ꢀ4
ꢀ1 ꢀ1
3
temperature, and for oxidation of PhSMe by 10 M
ꢀ4
[
MeCN][HO ] ¼ 12.4 ꢆ 10 l mol
s
as compared
ꢀ1 ꢀ1
2
ꢀ
ꢀ4
ꢀ
ꢄ
with k =[MeCN][HO ] ¼ 7.8 ꢆ 10 l mol s . Sev-
HSO , t ꢂ 11 s at 25.0 C and the reaction is first order
0.5
0
2
eral factors may account for this discrepancy, by making
5
in sulfide.
17
a
2
Table 4. Acetonitrile-mediated decomposition of H O
2
3
10 [NaOH]
3
10 [HO ]
ꢀ
6
10 v
(l mol
4
10 v/[MeCN][HO ]
ꢀ
MeCN
(vol.%)
2
2
ꢀ
1
ꢀ1
ꢀ1 ꢀ1
s )
(M)
(M)
s
)
(l mol
b
b
5
5
5
0
0
1.0
2.0
1.0
1.0
2.0
0.97
1.94
0.99
0.97
1.94
12.0, 10.7
21.0c
11.7
22.6
48.2
12.9, 11.5
11.8
12.6
12.1
12.7
1
1
a
b
c
ꢄ
At 25 C and 0.1 M H
2
O
2
unless specified otherwise.
ꢀ6
ꢀ1 ꢀ1
s at 270 and 280 nm.
Calculated from four points and immediate titration, rates are measured spectrometrically are 12.5 ( ꢈ0.1) ꢆ 10 l mol
, initial rate determined spectrometrically is 12.8 ( ꢈ0.02) ꢆ 10ꢀ6 l mol
ꢀ1 ꢀ1
at 270 and 280 nm.
s
In 0.5 M H
2
O
2
Copyright # 2003 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2003; 16: 603–607

OXIDATION OF THIOANISOLE BY HYDROGEN PEROXIDE
607
ꢀ
Competition between substituted thioanisoles is fa-
10a
The rate of reaction of HO and MeCN in the absence
2
vored by electron-donating substituents with ꢀ ¼ 1.0,
of PhSMe, giving O þ MeCONH , is higher than that
2
2
showing that oxidation by the peroxyimidate is activated
and therefore much slower than a diffusion-controlled
reaction.
with PhSMe by a factor of 1.6, although a factor of 2
would be expected, and there may be minor contributions
from other oxidative reactions.
The high reactivity of peroxyimidate, relative to other
9
peroxy acids, also applies to alkene epoxidation. We do
Acknowledgements
not know the source of the high reactivity, but the
hydroperoxy group loses a hydrogen ion in the course
of reaction and may be favored by the imidic nitrogen.
Our kinetic data are consistent with the preparative
Support from the US Army Office of Research, the US–
Brazil Cooperative Science Program (INT-9722458)
and CAPES for a fellowship for J. D. is gratefully
acknowledged.
10b
9
results, and Payne et al. have noted the desirability of
avoiding high pH in these nitrile-mediated reactions with
alkenes. Various groups have successfully used K CO to
2
3
10
control the pH in epoxidations and sulfide oxidations,
although here there may have been a contribution of
reaction with peroxocarbonate, especially in organic
REFERENCES
1. (a) Yang Y-C, Baker JA, Ward JR. Chem. Rev. 1992; 92: 1729–
1743; (b) Yang Y-C. Acc. Chem. Res. 1999; 32: 109–115.
3
solvents. The problem with using high pH may be that
2
. (a) Epstein J, Demek MM, Rosenblatt DH. J. Org. Chem. 1956;
21: 796–797; (b) Larsson L. Acta Chem. Scand. 1958; 12:
723–730; (c) Yang Y-C, Szafraniec LL, Beaudry WT, Bunton
CA, Kumar A. J. Chem. Soc., Perkin Trans. 2 1997; 607–613.
. (a) Richardson DE, Yao H, Frank KM, Bennett DA. J. Am. Chem.
Soc. 2000; 122: 1729–1739; (b) Wagner GW, Yang Y-C.
In Proceedings of the 1998 ERDEC Scientific Conference on
Chemical and Biological Defense Research, ERDEC, 1999;
alkali promotes decomposition of H O rather than af-
2
2
fecting formation and reaction of the peroxyimidate.
3
Reaction with benzonitrile
2
4. Dickman MH, Pope MT. Chem. Rev. 1994; 94: 569–584.
85–291.
The oxidation of 0.001 M PhSMe and 0.01 M H O was
2
2
examined in an aqueous mixture with 0.04 M PhCN. With
no NaOH the mixture remained turbid for up to 4 h at
room temperature, but with 0.001 M NaOH it clarified
5. (a) Bunton CA, Gillitt ND. J. Phys. Org. Chem. 2002; 15: 29–35;
(
b) Chiarini M, Gillitt ND, Bunton CA. Langmuir 2002; 18:
836–3842.
3
6
. Wagner GW, Procell LR, Yang Y-C, Bunton CA. Langmuir 2001;
17: 4809–4811.
1
within 3 min and we saw H NMR signals of the CH3
group of PhSOMe and PhSO Me within the time (10 min)
7. Aubry JM, Bouttemy S. J. Am. Chem. Soc. 1997; 119: 5286–5294.
8. Wiberg KB. J. Am. Chem. Soc. 1955; 77: 2519–2522.
9. Payne GB, Deming PH, Williams PH. J. Org. Chem. 1961; 26:
659–663.
2
required for signal acquisition. The reaction occurs read-
ily, despite the heterogeneous conditions, and Payne
9
et al. had noted that epoxidation of cyclohexene in
10. (a) Bethell D, Graham AE, Heer JP, Markopoulou O, Page PCB,
Park BK. J. Chem. Soc., Perkin Trans. 2 1993: 2161–2162; (b)
Page PCB, Graham AE, Bethell D, Park BK. Synth. Commun.
1993; 23: 1507–1514; (c) Laus G. J. Chem. Soc., Perkin Trans. 2
2001: 864–868.
MeOH is faster with PhCN than with MeCN. We could
not use absorption spectroscopy to follow this reaction
1
and H NMR signals of the products in the aromatic
region are obscured by those of PhCN.
1
1. (a) Liang J-J, Gu G-L, Kacher ML, Foote CS. J. Am. Chem. Soc.
983; 105: 4717–4721; (b) Inoue K, Matsuiira T, Saito I. Tetra-
hedron 1985; 41: 2177–2181; (c) Clennan El. Sulfur Rep. 1996;
9: 171–214.
2. Bunton CA, Foroudian HJ, Kumar A. J. Chem. Soc., Perkin Trans.
1995; 33–39.
3. Everett AJ, Minkoff GJ. Trans. Faraday Soc. 1953; 49: 410–414.
1
1
1
1
CONCLUSIONS
2
ꢀ
14. Bunton CA, Foroudian HJ, Gillitt ND, Kumar A. Can. J. Chem.
999; 77: 895–902.
Reactions of HO with nitriles generates peroxyimidates,
2
1
which are rapidly trapped by PhSMe forming sulfoxide
and subsequently sulfone. In aqueous solutions of MeCN
or EtCN formation of sulfoxide is significantly faster than
1
5. (a) Curci R, Modena G. Tetrahedron 1966; 22: 1227–1233; (b)
Curci R, Giovine A, Modena G. Tetrahedron 1966; 22: 1235–
1239; (c) Curci R, DiPrete P, Edwards JO, Modena G. J. Org.
Chem. 1970; 35: 740–745.
the competing reaction of H O with peroxyimidate.
2
2
16. (a) Zhu W, Ford WT. J. Org. Chem. 1991; 56: 7022–7026; (b)
Webb KS. Tetrahedron Lett. 1994; 35: 3457–3460.
Except with very dilute PhSMe, the rate of oxidation is
that of formation of peroxyimidate, but as [PhSMe]
decreases trapping is no longer quantitative and the rate
of oxidation is no longer independent of [PhSMe].
1
7. Bacaloglu R, Blasko A, Bunton CA, Foroudian HJ. J. Phys. Org.
Chem. 1992; 5: 171.
1
8. Gonzaga F, Perez E, Rico-Lattes L, Lattes A. New J. Chem. 2001;
25: 151–155.
Copyright # 2003 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2003; 16: 603–607