Mechanism of a redox coupling of seleninic acid with thiol

Article abstract of DOI:10.1021/jo300156x

Equimolar quantities of 2-ethoxyethaneseleninic acid and p-thiocresol react rapidly in dichloromethane solution to give the selenosulfide along with disulfide, diselenide, and two products oxidized at sulfur, the thiosulfonate and the selenosulfonate. The latter two are new for this sort of coupling; their formation may be the result of an early thioseleninate to selenosulfinate isomerization. A radical chain mechanism is proposed to account for all five products, as well as their relative amounts.

Full text of DOI:10.1021/jo300156x

Article
pubs.acs.org/joc
Mechanism of a Redox Coupling of Seleninic Acid with Thiol
Mohannad Abdo and Spencer Knapp*
Department of Chemistry & Chemical Biology, Rutgers, The State University of New Jersey, 610 Taylor Road, Piscataway, New
Jersey 08854, United States
S
* Supporting Information
ABSTRACT: Equimolar quantities of 2-ethoxyethanesele-
ninic acid and p-thiocresol react rapidly in dichloromethane
solution to give the selenosulfide along with disulfide,
diselenide, and two products oxidized at sulfur, the
thiosulfonate and the selenosulfonate. The latter two are
new for this sort of coupling; their formation may be the result of an early thioseleninate to selenosulfinate isomerization. A
radical chain mechanism is proposed to account for all five products, as well as their relative amounts.
and disulfide 6. As for certain other transformations of SeO
INTRODUCTION
Coupling reactions that occur rapidly and efficiently under mild
■
species,7−9 the fate of the oxygen atoms remained mysterious3
until Kice and Purkiss, who described the extraction of crude
conditions, sometimes called “click reactions,” have applications
not available to other transformations.1,2 For example, the
product with aqueous sodium carbonate, were able to isolate an
oxidized product, the seleninic acid 7 (=1).6 Yields based on
redox coupling of seleninic acid and thiol to give the
selenosulfide succeeds with a variety of coupling partners and
moles of 3 are shown in Scheme 1 and together account for
in a wide range of solvents.3 Applications to the efficient
94% of the Se, 97% of the S, and 52% of the O atoms, assuming
that the latter arose by hydrolysis upon extraction of
coupling of biomolecular components or biomimetics resem-
benzeneseleninic anhydride, PhSe(O)OSe(O)Ph. A rad-
bling amino acids, peptides, nucleosides, phosphatidic acids,
and carbohydrates can be envisioned.4 Indeed, a tyrosine-
derived seleninic acid, as a substrate mimic, couples with a
cysteine thiol residue in the active site of protein tyrosine
phosphatase to effectively deactivate the enzyme by covalent
modification.5
ical chain mechanism featuring a mixed selenenic−seleninic
anhydride intermediate was advanced to account for the
formation of products 4−7.6
In this paper, we analyze the results of a seemingly analogous
seleninic acid/thiol coupling reaction that takes an entirely
different course and suggest a mechanism that accounts for the
differences.
Two authoritative papers by Kice and co-workers established
the coupling reaction of benzeneseleninic acid 1 with t-
butanethiol 2 to proceed by initial formation of the
thioseleninate intermediate 3 (Scheme 1).3,6 When two
RESULTS AND DISCUSSION
■
We find that 2-ethoxyethaneseleninic acid10 (8, Scheme 2)
couples with 3 equiv of thiocresol (9, added last) in
dichloromethane solution to give the selenosulfide 10 along
with bis-(2-ethoxyethyl) diselenide (11)10,11 and di-p-tolyl
disulfide (12). No particular efforts were made to exclude
oxygen, water, or light other than capping the reaction vessel.
The reaction solution immediately becomes foggy, presumably
with the water of reaction, and is over in less than a minute,
judging from TLC analysis. No gas evolution is observed, nor
Scheme 1. Coupling Reaction of t-Butanethiol and
Benzeneseleninic Acid Described by Kice et al.
are other nonvolatile products detected, according to H, 13C,
1
and 77Se NMR spectroscopic examination of the concentrated
crude reaction mixture. Reversing the order of addition, or
changing the rate of addition, or conducting the reaction at 0
°C did not change the results appreciably. Careful chromatog-
raphy enabled the isolation of 10−12 in the yields shown
(yields shown are based on the respective chalcogen-containing
precursor), which together account for 98% of the Se (i.e., 81%
+ 17%) and 96% of the S (i.e., one-third of 81%, plus 69%)
additional equivalents of 2 are present, 3 is reduced efficiently
to the selenosulfide product 4. However, with equimolar
amounts of 1 and 2, the reaction stops at 3, which can then
with care be isolated and characterized, and its decomposition
studied separately. In concentrated acetone solution, 3 is
rapidly converted to a mixture of selenosulfide 4, diselenide 5,
Received: February 1, 2012
Published: February 29, 2012
© 2012 American Chemical Society
3433
dx.doi.org/10.1021/jo300156x | J. Org. Chem. 2012, 77, 3433−3438

The Journal of Organic Chemistry
Article
(12% + 39% + 44%) and 99% of the S (12% + 32% + 44% +
11%) atoms; 13 and 14 together account for 99% [respectively,
88% (two O’s per 44% S) plus 11% (one O per 11% S)] of the
expected oxygen atoms from 8, exclusive of the water of
reaction of coupling. No other products are apparent in the
crude concentrated reaction mixture, according to TLC and 1H,
13C, and 77Se NMR analysis. As far as we are aware, S-oxidized
products have not been previously isolated from a seleninic
acid/thiol coupling reaction.
Scheme 2. Redox Coupling Reaction of 2-
Ethoxyethaneseleninic Acid and Thiocresol
The reaction of Scheme 3 was scaled up to 1.5 mmol (0.75
mmol/mL) in order to quantitate the water produced.
Extraction of the dichloromethane solution with D2O and
then analysis of the 1H NMR spectrum with integration relative
to N-acetylglycine as an internal standard allowed assessment15
of the approximate amount of water produced: 1.1 equiv based
on 1 equiv of 8. Details of this analysis as well as the
accompanying calibration of the method are given in the
Experimental Section. A trace (about 1%) of a 2-ethoxyetha-
neseleninic product (δ 4.01, t, J = 6.5 Hz), distinct from 8 (δ
3.90, t, J = 6.5 Hz) and possibly the anhydride, was also
detected in the aqueous layer. Examination of the 1H, 13C, and
77Se NMR spectra of the concentrated crude dichloromethane
soluble reaction mixture showed the exclusive formation of the
previous five products, and integration of the proton signals
(the well-resolved tolyl methyl singlets at 2.2−2.5 ppm and the
distinct methylene quartet of 11 at 3.53 ppm) allowed
assignment of the relative yields (based on 100%), which are
virtually identical to the isolated yields of Scheme 3 (10−14,
respectively: 14, 40, 32, 44, and 10%). Because of this
agreement, the isolated yields may be taken as quite accurately
reflective of the relative amounts of products formed.
atoms. Selenosulfide 10, although stable to isolation and
characterization, does slowly disproportionate in solution to
give a mixture of 10, 11, and 12. Analogous chalcogen
scrambling behavior has been seen with other selenosul-
fides.3,12−14
In comparison to the reaction of Scheme 2, the coupling
reaction of equimolar equivalents of 8 and 9 under otherwise
identical conditions (0.111 mmol/mL in dichloromethane
solution) gave the previously observed products 10−12, as well
as two new S-oxidized products, the selenosulfonate 13 and the
thiosulfonate 14 (Scheme 3). The products were isolated by
Several control reactions (Scheme 4) were undertaken to
support the product assignments and the transformations
Scheme 3. Stoichiometric Coupling Reaction of 2-
Ethoxyethaneseleninic Acid and Thiocresol, and Isolation of
Two New S-Oxidized Products
Scheme 4. Control Reactions Confirming the Identity of
Products and the Susceptibility of 10 and 12 toward S-
Oxidation
implied by Scheme 3. Independent oxidation of selenosulfide
10 to selenosulfonate 13 was carried out by treating 10 with
dimethyldioxirane (DMDO, eq 1). Analogous oxidation of
disulfide 12 gave the thiosulfonate 14 (eq 2). The products
were spectroscopically and chromatographically identical to
those produced earlier from the coupling of 8 and 9 (Scheme
3). Both 10 and 12 were shown to be unreactive toward 2-
ethoxyethaneseleninic acid 8 (eq 3). The structure of
column chromatography on silica (eluent 9:1 hexanes/ethyl
acetate; order of elution: 12, 10, 11, 14, and finally 13) and
1
were characterized by mass spectrometry; H, 13C, and 77Se
NMR spectroscopy; and by comparison to authentic samples.
Taken together, the isolated yields (each based on the
respective chalcogen precursor) account for 95% of the Se
3434
dx.doi.org/10.1021/jo300156x | J. Org. Chem. 2012, 77, 3433−3438

The Journal of Organic Chemistry
Article
selenosulfonate 13 was confirmed by independent synthesis:
redox coupling16,17 of 8 with p-toluenesulfonylhydrazide (15)
gave 13 quantitatively (eq 4).
Scheme 6. Proposed Radical Chain Reaction for Redox
Coupling of 8 and 9
a
The equimolar condensation of 8 and 9 was repeated as in
Scheme 3, but with the respective prior addition of 1 equiv of
dodecyl sulfide (eq 5, Scheme 5) and trans-stilbene (eq 6,
Scheme 5. Attempted Trapping of Reactive Intermediates
with Dodecyl Sulfide and trans-Stilbene
Scheme 5). Neither additive reacts with 8 under these
conditions. For both reactions, analysis of the crude reaction
1
mixtures by H NMR spectroscopy indicated the same five
products (10−14) in approximately the same ratio. Further-
more, no dodecyl sulfoxide was formed (eq 5), and no CC
trapped product was detected (eq 6). Thus, it may be
concluded that any oxidizing, electrophilic, or radical
intermediates reacted faster with species present anyway in
the course of the reaction rather than with the traps, even
though the latter are present in relative excess. Sulfides have
been oxidized to the sulfoxide by seleninic acids,18 and a
selenoxide,10 but only in the presence of strong acid. Seleninic
acid 8 is a weak acid, pKa ∼ 5,19 as is ArSO2H. Traces of
ArSO3H might be formed here, although its salt with 8 would
likely precipitate from dichloromethane solution.20 Electron-
rich alkenes can add selenenic acids RSeOH if the latter survive
contact with other reducing species in the reaction mixture.9
A mechanism for the formation of coupled products 10−14
is proposed in Scheme 6 (isolated products are shown in the
boxes). The initial condensation of seleninic acid 8 and
thiocresol 9 to give the thioseleninate 16 and water (eq 7) is
suggested to proceed rapidly and completely, consistent with
the observations of Kice and co-workers.3,6 A new isomer-
ization, the conversion of thioseleninate 16 into selenosulfinate
17 (eq 8), can account for the appearance of S-oxidized
products, as will be seen. This isomerization is analogous to the
apparent scrambling of aryl groups in the thermal rearrange-
ment of S-aryl arenethiosulfinates,21 although this one would
occur at a lower temperature. Caged radical recombination
processes21 may be involved, and the mixed selenenic−sulfenic
anhydride (EtOCH2CH2Se-O-SAr), although not expected to
be stable itself,6 is a likely intermediate. Initiation of the radical
chain occurs with selenenyl abstraction from 17 (eq 9). The
resulting arenethioxyl radical 18 can combine with thioselenen-
ate 16 to give the seleninic−sulfenic anhydride 19 (eq 10), or
equivalently, the α-selenoxide-sulfoxide (EtOCH2CH2Se(
O)-S(O)Ar), either of which should rearrange rapidly to
the selenosulfonate 13 (eq 11). This rearrangement finds
precedent in the analogous behavior of α-disulfoxides, which
rearrange to thiosulfonates so fast at −20 °C as to be
undetectable.22,23 The same intermediates (16, 17, and the α-
selenoxide-sulfoxide) could also be involved in the rapid
DMDO oxidation of selenosulfide 10 to selenosulfonate 13
a
Products are shown in the boxes.
(Scheme 4, eq 1), assuming that kinetic oxidation of 10 occurs
on Se. The arenethiyl radical 20 produced in eq 10 can react
with selenosulfinate 17 (eq 12) to give the selenosulfide 10 and
arenethioxyl radical 18, which continues the chain. The mixed
chalcogen coupled products 10 and 13 are accounted for by eqs
9−12, while a separate, parallel radical chain sequence is
necessary to explain the formation of the same-chalcogenide
products 11, 12, and 14.
Selenosulfide 10 can be further transformed, in part, by
reaction with arenethiyl radical 20 to produce the product
disulfide 12 and a selenyl radical 21 (eq 13). Further reaction of
21 with more of 10 leads to the product diselenide 11 and
regenerates 20 to continue the chain (eq 14). Chalcogen
scrambling of selenosulfides in the presence of radicals is well-
precedented.3,24 The fifth product 14 can be formed by
reaction (eq 15) of arenethioxyl radical 18 with selenosulfinate
17, the proposed source of SO groups in both sulfonate
products 13 and 14. The resulting sulfenic−sulfinic anhydride
3435
dx.doi.org/10.1021/jo300156x | J. Org. Chem. 2012, 77, 3433−3438

The Journal of Organic Chemistry
Article
22 (or equivalently, the α-disulfoxide) would be expected22,23
to rapidly rearrange to thiosulfonate 14 (eq 16). The remaining
selenosulfinate 17 is converted (eq 17) to additional diselenide
11 and enough arenethioxyl radical 18 to continue the chain by
reentry at eq 15.
logical choice given that the 44% isolated yield of product 13
(the selenosulfonate) comes only from 16 (eqs 19 and 20).
Products are shown in solid boxes, and chain carriers [ArS•
(20), ArSO• (18), and EtOCH2CH2Se• (21)] are shown in
dashed boxes. Note that the yields shown are based on the
respective S and Se starting materials 8 and 9, as in Scheme 3. A
44% portion of 17 gives rise to a 44% initial yield of
selenosulfide 10, as shown (eq 21), but the latter is partitioned
into 16% (which leads to disulfide 12, eq 22), an additional
16% (which leads to diselenide 11, eq 23), and the remaining
12%, which is isolated. The remaining selenosulfinate 17 (12%)
goes in equal parts, at least formally, to the formation of
thiosulfonate 14 (eqs 24 and 25) and also additional diselenide
11 (eq 26). The amounts of carrier species ArS• (20), ArSO•
(18), and EtOCH2CH2Se• (21) and other intermediates
balance on both sides of the arrows. The product yields
predicted by Scheme 7 (given the 16/17 partitioning) for 10−
14 are 12, 44, 32, 44, and 12%, respectively, whereas the
corresponding isolated yields are 12, 39, 32, 44, and 11%, a very
close match. Not all the predicted diselenide 11 was isolated
(39% vs 44% predicted), possibly because some was lost to the
aqueous layer as the seleninic anhydride (also a Kice product;
see Scheme 1). This accounting lends credence to the proposed
mechanism of Scheme 6, in that the chalcogen atoms and
oxidation states are properly conserved.
Is the proposed mechanism of Scheme 6 consistent with the
trapping experiments of Scheme 5? No strong oxidizing species
(stronger than seleninic acid 8, that is) is proposed as an
intermediate, so the noninvolvement of dodecyl sulfide is
understandable. Direct transfer of [O] from thioseleninate 16
as an oxidant might be possible, but ought to require strong
acid catalysis (to protonate SeO).10 Whether ArSeO• can
serve as an oxidant is not known at this time, but it would have
to be selective for 10 and 12 in the presence of thioether. We
propose that the respective precursors of selenosulfonate 13
and thiosulfonate 14 are probably not the corresponding
dichalcogenides, 10 and 12, each of which is stable to 8, if not
DMDO. The selenenic acid EtOCH2CH2SeOH, which might
react with trans-stilbene as an electrophile,9 is not invoked as an
intermediate. Seleninic acids are also known to combine with
sulfinates to produce selenosulfonates (compare 13) as well as
sulfonic acid,20 but the latter is not found here among the
products.
Selenosulfonates, such as 13, can also react with alkenes by
photoactivated formation and addition of ArSO2•,20 but no
such adducts are seen, and ArSO2• is not a required
intermediate in Scheme 6. Indeed, trans-stilbene and dodecyl
sulfide are found to coexist with 13 in solution under the
conditions of the reaction. The radical chain carriers ArS• (20),
ArSO• (18), and EtOCH2CH2Se• (21) might be expected to
add to an alkene, but the addition could be reversible, and other
radicalophiles, including 16 and 17, are present and can react
rapidly by rupture of their weak chalcogen−chalcogen bonds.
Of course, it cannot be assumed that a radical trap more
reactive than trans-stilbene might not have intercepted, say,
arenethioxyl radical 18. Under certain conditions, ArS• (but
perhaps not RSe•25) can trap dissolved molecular oxygen,26 but
no product attributable to such a reaction was found.
An analogous radical chain mechanism (Scheme 8) can be
drawn for the formation of the Kice coupling products (shown
Scheme 8. Proposed Radical Mechanism for the Kice
a
Decomposition of S-tert-Butyl Benzenethioseleninate
The mechanism proposed in Scheme 6 also formally
accounts for the relative yields of the five products 10−14.
Scheme 7 delineates the consequences of the mechanism of
Scheme 6 if one assumes a 44/56 partitioning of intermediates
thioseleninate 16 and selenosulfinate 17 (eq 18). This is a
Scheme 7. Formal Partitioning of Scheme 6 Intermediates,
Radical Carrier Species (in Dashed Boxes), and Coupled
Products (in Solid Boxes)
a
Products are shown in solid boxes, and radical carrier species in
dashed boxes.
in Scheme 1) that allows a comparison between the two
reactions. In this case, no rearrangement of the thioseleninate
(here, 3) to a selenosulfinate occurs. Instead, 3 initiates the
radical chain by forming benzeneselenoxyl radical 23 (eq 27),
which can combine with more 3 to produce a mixed selenenic−
seleninic anhydride 24 as well as t-butanethiyl radical 25 (eq
28). The latter abstracts the t-butylthio group from 3 to give di-
t-butyl disulfide 6 and regenerate the radical carrier species 23
3436
dx.doi.org/10.1021/jo300156x | J. Org. Chem. 2012, 77, 3433−3438

The Journal of Organic Chemistry
Article
(eq 29). Combination of 23 with 24 leads to benzeneseleninic
anhydride 26, and benzeneselenyl radical 27 (eq 30), and 27, in
turn, produces the other two products (selenosulfide 4 and
diphenyl diselenide 5, eqs 31 and 32) by radical exchange.
Further chalcogen scrambling of 4, isolated as a minor product,
can be expected to occur under the radical conditions, leading
to the formation of additional amounts of disulfide 6 and
diselenide 5. Hydrolysis of benzeneseleninic anhydride 26 upon
aqueous base workup accounts for the isolation of benzenese-
leninic acid 7. Kice proposed a mechanism6 similar to Scheme 8
for radical decomposition of 3.
EXPERIMENTAL SECTION
■
S-p-Tolyl 2-Ethoxyethaneselenenylthioate (10), Bis-(2-
ethoxyethyl)diselenide (11),10,11 and Di-p-tolyl Disulfide
(12).28,29 A solution of 2-ethoxyethaneseleninic acid 8 (11.1 mg,
0.060 mmol) in dichloromethane (0.5 mL) was treated, over about 5 s,
with a solution of p-thiocresol 9 (22.4 mg, 0.180 mmol) in
dichloromethane (0.5 mL). After 1 min of stirring at room
temperature, TLC analysis indicated the consumption of starting
materials. The reaction mixture was concentrated and chromato-
graphed on silica with 9:1 hexanes/ethyl acetate as the eluant to give
13.4 mg (81%) of selenosulfide 10 as a colorless oil, 1.5 mg (17%) of
diselenide 11 as a yellow oil, and 15.3 mg (69%) of disulfide 12 as a
colorless oil.
The Kice reaction6 produces no S-oxidized products, possibly
because (a) steric hindrance at tert-butyl-S suppresses the
thioseleninate-to-selenosulfinate rearrangement (eq 8, Scheme
6), and/or (b) the chalcogen−chalcogen bond of 3 (eq 27,
Scheme 8) is stronger than that of 16 (eq 7, Scheme 6).
Because the early coupled products differ, the radical chain
carriers in the two reactions are different: ArSO• in the case of
Scheme 6, and PhSeO• in the reaction of Scheme 8. In an
earlier investigation of the same reaction, Kice and Lee3
combined benzeneseleninic acid 1 and t-butanethiol 2 in
equimolar amounts and observed that the thioseleninate
intermediate 3 decomposed rapidly in concentrated acetone
solution to give three products only: 4, 5, and 6. No oxygen-
containing products were found, and preliminary speculation3
that the other product is molecular oxygen (O2) was withdrawn
in the later investigation,6 where the formation of benzenese-
leninic acid 7 as the oxygen-containing product from separate
decomposition of 3 was reported. In our earlier studies of
alkaneseleninic acids,4 we have also found that the seleninic
acid oxygens sometimes “disappear” during coupling reactions
with thiols, even though all the chalcogen atoms are accounted
for in the products. At present, we have no general explanation
for this phenomenon, but would note that possible conversion
of the carrier species RSeO• to RSe• can theoretically account
for the absence of oxygen-containing coupled products. A
somewhat analogous process has been proposed for oxygen
insertion from dimethyldioxirane into R−H.27 The missing
oxygen atom could well be incorporated into solvent or other
nonchalcogen species.
1
Selenosulfide 10: Rf 0.47 (9:1 hexanes/ethyl acetate); H NMR
(500 MHz, CDCl3) δ 7.44 (d, 2H, J = 8.4 Hz), 7.08 (d, 2H, J = 8.4
Hz), 3.71 (t, 2H, J = 6.8 Hz), 3.46 (q, 2H, J = 7.2 Hz), 3.09 (t, 2H, J =
6.8 Hz), 2.32 (s, 3H), 1.17 (t, 3H, J = 7.2 Hz); 13C NMR (100 MHz,
CDCl3) δ 137.6, 133.8, 130.5, 129.9, 69.4, 66.5, 31.8, 21.2, 15.4; 77Se
NMR (76 MHz, CDCl3) δ 437.4 [vs PhSeSePh at 460.0 ppm as an
external standard]; HRMS (ESI) Calcd for C11H16ONaSSe (MNa+)
298.9985, found 298.9970.
1
Diselenide 11: Rf 0.32 (9:1 hexanes/ethyl acetate); H NMR (400
MHz, CDCl3) δ 3.71 (t, 2H, J = 6.8 Hz), 3.53 (q, 2H, J = 7.2 Hz), 3.11
(t, 2H, J = 6.8 Hz), 1.21 (t, 3H, J = 7.2 Hz); 13C NMR (100 MHz,
CDCl3) δ 70.7, 66.5, 29.6, 15.4; 77Se NMR (76 MHz, CDCl3) δ 295.4
[vs PhSeSePh at 460.0 ppm as an external standard]; ESI-MS m/z 329
MNa+, molecular ion cluster: two Se.
1
Disulfide 12: Rf 0.64 (9:1 hexanes/ethyl acetate); H NMR (500
MHz, CDCl3) δ 7.40 (d, 4H, J = 8.0 Hz), 7.12 (d, 4H, J = 7.5 Hz),
2.33 (s, 6H); 13C NMR (125 MHz, CDCl3) δ 137.7, 134.2, 130.0,
128.8, 21.3; ESI-MS m/z 269 MNa+.
Equimolar Redox Coupling of 2-Ethoxyethaneseleninic acid
(8) and p-Thiocresol (9). A solution of seleninic acid 8 (20.6 mg,
0.111 mmol) in dichloromethane (1 mL) was treated with p-thiocresol
9 (13.8 mg, 0.111 mmol) in one aliquot. After 1 min of stirring at
room temperature, TLC analysis indicated the consumption of starting
materials. The reaction mixture was concentrated and chromato-
graphed on silica with 9:1 hexanes/ethyl acetate as the eluant to give in
order of elution: 4.3 mg (32%) of disulfide 12 as a colorless oil, 3.7 mg
(12%) of selenosulfide 10 as a colorless oil, 6.6 mg (39%) of diselenide
11 as a yellow oil, 1.7 mg (11%) of thiosulfonate 14 as a white solid,
and 14.9 mg (44%) of selenosulfonate 13 as a colorless oil.
Se-(2-Ethoxyethyl)-p-toluenesulfonylselenylate (13): Rf 0.13 (9:1
hexanes/ethyl acetate); 1H NMR (500 MHz, CDCl3) δ 7.77 (d, 2H, J
= 8.5 Hz), 7.33 (d, 2H, J = 8.5 Hz), 3.76 (t, 2H, J = 6.5 Hz), 3.46 (q,
2H, J = 7.0 Hz), 3.38 (t, 2H, J = 6.5 Hz), 2.45 (s, 3H), 1.15 (t, 3H, J =
7.0 Hz); 13C NMR (125 MHz, CDCl3) δ 144.8, 144.7, 130.0, 126.7,
69.1, 66.8, 33.3, 21.9, 15.2; 77Se NMR (76 MHz, CDCl3) δ 857.2 [vs
PhSeSePh at 460.0 ppm as an external standard]; HRMS (ESI) Calcd
for C11H16O3NaSSe (MNa+) 330.9883, found 330.9875.
The radical chain mechanism of Scheme 6 might not operate
effectively in dilute solution, nor in the active site of an
enzyme,5 since bimolecular encounters and radical initiation
events would be infrequent. In these cases, an alternative
reducing agent might be serving to transform the putative
thioseleninate intermediate to products. However, attempts to
improve the efficiency of such equimolar seleninic acid/thiol
redox couplings by intentionally adding an in situ reducing
agent have thus far been unsuccessful.
S-p-Tolyl-p-toluenesulfonylthioate (14).30 Rf 0.27 (9:1 hexanes/
1
ethyl acetate); H NMR (500 MHz, CDCl3) δ 7.47 (d, 2H, J = 8.0
Hz), 7.25 (d, 2H, J = 8.5 Hz), 7.22 (d, 2H, J = 8.0 Hz), 7.15 (d, 2H, J
= 8.0 Hz), 2.43 (s, 3H), 2.38 (s, 3H); 13C NMR (125 MHz, CDCl3) δ
144.8, 142.3, 140.7, 136.7, 130.4, 129.6, 127.8, 124.9, 21.9, 21.7; ESI-
MS m/z 301 MNa+.
Oxidation of Selenosulfide 10 to Selenosulfonate 13.
Dimethyldioxirane (total ∼196 μL of a 0.30 M titrated solution of
dimethyldioxirane (DMDO) in chloroform31 (∼2.0 equiv) was added
to a stirred solution of selenosulfide 10 (8.1 mg, 0.029 mmol) in moist
dichloromethane (1 mL). After 2 min of stirring at room temperature,
the reaction mixture was concentrated and chromatographed on silica
with 4:1 hexanes/ethyl acetate as the eluant to give 6.8 mg (76%) of
selenosulfonate 13 as a white solid.
Oxidation of Disulfide 12 to Thiosulfonate 14: DMDO (total
∼366 μL of a 0.30 M titrated solution in chloroform, ∼2.0 equiv) was
added to a stirred solution of disulfide 12 (13.5 mg, 0.0549 mmol) in
moist dichloromethane (1 mL). After 2 min of stirring at room
temperature, the reaction mixture was concentrated and chromato-
CONCLUSION
■
The remarkably fast equimolar redox coupling reaction of 2-
ethoxyethaneseleninic acid and p-thiocresol (Scheme 3) is
shown to take a new course explainable by isomerization of the
initial coupled product, a thioseleninate (16), to a selenosulfi-
nate (17). A radical chain mechanism (Scheme 6) is proposed
to rationalize the formation of the five products 10−14, with
intermediate 17 suggested as the source of the novel S-oxidized
products 13 and 14. The water of reaction is quantified for the
first time.
3437
dx.doi.org/10.1021/jo300156x | J. Org. Chem. 2012, 77, 3433−3438

The Journal of Organic Chemistry
Article
graphed on silica with 9:1 hexanes/ethyl acetate as the eluant to give
14.1 mg (92%) of thiosulfonate 14 as a white solid.
ACKNOWLEDGMENTS
■
We are grateful to the Prusoff Foundation for financial support,
and to Rohm & Haas Company for a graduate fellowship to
M.A.
Independent Synthesis of Selenosulfonate 13 by the Redox
Coupling of 8 with p-Toluenesulfonylhydrazide. Seleninic acid 8
(8.4 mg, 0.045 mmol) in dichloromethane (1 mL) was added
dropwise to a solution of p-toluenesulfonylhydrazide (9.3 mg, 0.050
mmol) in dichloromethane (1 mL). After 30 min of stirring at room
temperature, the reaction mixture was concentrated and chromato-
graphed on silica with 4:1 hexanes/ethyl acetate as the eluant to give
13.9 mg (∼100%) of selenosulfonate 13 as a white solid.
REFERENCES
■
(1) Kolb, H. C.; Finn, M. G.; Sharpless, K. B. Angew. Chem., Int. Ed.
2001, 40, 2004−2021.
(2) Finn, M. G.; Fokin, V. V. Chem. Soc. Rev. 2010, 39, 1231−1232
Water of Reaction Produced by the Equimolar Coupling of 8
and 9. A solution of seleninic acid 8 (277 mg, 1.50 mmol) in
dichloromethane (2 mL) was treated with thiocresol 9 (186 mg, 1.50
mmol) in one aliquot. After 1 min of stirring at room temperature (the
solution became slightly foggy), 2.0 mL of deuterium oxide was added.
and references cited therein.
(3) Kice, J. L.; Lee, T. W. S. J. Am. Chem. Soc. 1978, 100, 5094−5102.
(4) Abdo, M.; Knapp, S. J. Am. Chem. Soc. 2008, 130, 9234−9235.
(5) Abdo, M.; Liu, S.; Zhou, B.; Walls, C. D.; Wu, L.; Knapp, S.;
Zhang, Z.-Y. J. Am. Chem. Soc. 2008, 130, 13196−13197.
(6) Kice, J. L.; Purkiss, D. W. J. Org. Chem. 1987, 52, 3448−3451.
(7) Woodbridge, D. T. J. Chem. Soc. B 1966, 5052.
(8) Sharpless, K. B.; Lauer, R. F. J. Org. Chem. 1974, 39, 429−430.
(9) Reich, H. J.; Wollowitz, S.; Trend, J. E.; Chow, F.; Wendelborn,
D. F. J. Org. Chem. 1978, 43, 1697−1705.
1
The organic layer was concentrated in vacuo; examination of the H,
13C, and 77Se NMR spectra of the crude product showed the exclusive
formation of compounds 10−14 with relative spectroscopic yields of
14, 40, 32, 44, and 10%, respectively (based on the integration of
proton signals of the tolyl methyls and the methylenes of 11). The
aqueous layer was separated, and 0.75 mL was combined with N-
acetylglycine (5.9 mg, 0.050 mmol) as an internal NMR integration
standard. The resulting aqueous solution was transferred to an NMR
tube, and the 1H NMR spectrum was examined. The integration value
of the HOD signal relative to the methylene of N-acetylglycine was
7.86 to 2.21, corresponding to 1.1 equiv of water produced (based on
1.0 equiv of 8), after background correction. The control experiments
as well as the calibration of the method (i.e., quantitation of known
added amounts of water) are summarized in Table 1 in the Supporting
Information.
Equimolar Condensation of 8 and 9 in the Presence of
Dodecyl Sulfide. A solution of seleninic acid 8 (5.0 mg, 0.027 mmol)
and dodecyl sulfide (10.0 mg, 0.027 mmol) in dichloromethane (0.5
mL) was treated in one aliquot with a solution of p-thiocresol 9 (3.4
mg, 0.027 mmol) in dichloromethane (0.5 mL). After 1 min of stirring
at room temperature, TLC analysis indicated the consumption of
starting material. The reaction mixture was concentrated, and the
crude 1H NMR spectrum (Supporting Information) showed the
exclusive formation of compounds 10−14 in addition to unchanged
didodecyl sulfide.
Equimolar Condensation of 8 and 9 in the Presence of
trans-Stilbene. A solution of seleninic acid 8 (11.7 mg, 0.063 mmol)
and trans-stilbene (11.4 mg, 0.063 mmol) in dichloromethane (0.5
mL) was stirred at room temperature for 30 min. No reaction between
seleninic acid 8 and trans-stilbene was observed. The reaction solution
was then treated in one aliquot with a solution of p-thiocresol 9 (7.9
mg, 0.063 mmol) in dichloromethane (0.5 mL). After 1 min of stirring
at room temperature, TLC analysis indicated the consumption of
starting material. The reaction mixture was concentrated, and the
crude 1H NMR spectrum (Supporting Information) showed the
exclusive formation of compounds 10−14 in addition to unchanged
trans-stilbene.
(10) Abdo, M.; Zhang, Y.; Schramm, V. L.; Knapp, S. Org. Lett. 2010,
12, 2982−2985.
(11) Ward, H.; O’Donnell, L. J. Am. Chem. Soc. 1945, 67, 883.
(12) Iwaoka, M.; Komatsu, H.; Tomoda, S. Chem. Lett. 1998, 27,
969−970.
(13) Yoshida, M.; Cho, T.; Kobayashi, M. Chem. Lett. 1984, 13,
1109−1112.
(14) Fisher, H.; Dereu, N. Bull. Soc. Chim. Belg. 1987, 96, 757−768.
(15) For a leading recent reference to methods of water detection,
see: Sun, H.; Wang, B.; DiMagno, S. G. Org. Lett. 2008, 10, 4413−
4416.
(16) Back, T. G.; Collins, S. Tetrahedron Lett. 1980, 21, 2213−2214.
(17) Back, T. G.; Collins, S.; Krishna, M. Can. J. Chem. 1987, 65, 38−
42.
(18) Faehl, L. G.; Kice, J. L. J. Org. Chem. 1979, 44, 2357−2361.
(19) McCullough, J. D.; Gould, E. S. J. Am. Chem. Soc. 1949, 71,
674−676.
(20) Gancarz, R. A.; Kice, J. L. J. Org. Chem. 1981, 46, 4899−4906.
(21) Koch, P.; Ciuffarin, E.; Fava, A. J. Am. Chem. Soc. 1970, 92,
5971−5977.
(22) Chau, M.; Kice, J. L. J. Am. Chem. Soc. 1976, 98, 7711−7716.
(23) Oae, S.; Takata, T.; Kim, Y. H. Bull. Chem. Soc. Jpn. 1982, 55,
2484−2494.
(24) Yoshida, M.; Cho, T.; Kobayashi, M. Chem. Lett. 1984, 1109−
1112.
(25) Ito, O. J. Am. Chem. Soc. 1983, 105, 850−853.
(26) Schafer, K.; Bonifacic, M.; Bahnemann, D.; Asmus, K.-D. J. Phys.
Chem. 1978, 82, 2777−22780.
(27) Curci, R.; Dinoi, A.; Fusco, C.; Lillo, M. A. Tetrahedron Lett.
1996, 37, 249−252.
(28) Maercker, M. Justus Liebigs Ann. Chem. 1865, 136, 88.
(29) Yee, K.; Tan, D.; Kee, J. W.; Fan, W. Y. Organometallics 2010,
29, 4459−4463.
(30) Mizuno, H.; Matsuda, M.; Iino, M. J. Org. Chem. 1981, 46, 520−
ASSOCIATED CONTENT
525.
■
(31) Gilbert, M.; Ferrer, M.; Sanchez-Baeza, F.; Messeguer, A.
Tetrahedron 1997, 53, 8643−8650.
S
* Supporting Information
1H, 13C, and 77Se NMR spectra of reaction products 10−14,
and details of the controls and calibration of the water
detection method. This material is available free of charge via
the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: spencer.knapp@rutgers.edu.
Notes
The authors declare no competing financial interest.
3438
dx.doi.org/10.1021/jo300156x | J. Org. Chem. 2012, 77, 3433−3438