Organic Letters
Letter
coupling using sodium aryl sulfinates as relatively safe com-
modity sulfur surrogates (Scheme 1).
Scheme 3. Control Experiments
4-Methoxyphenyl boronic acid 1a and sodium benzenesulfinate
2a were chosen as model substrates to optimize the reaction
conditions (Tables 1, 2, S1, and S2). The initial reactions were
performed under our previous conditions for the coupling
between sulfinates and iodoarenes,9 and the new conditions
were based on the literature for the coupling between thiols
and boronic acids11 plus DABCO as the base, which was
shown to be critical for the deoxygenation of sulfinates (Table 1,
entry 1). 5% of the desired thioether 3a was obtained under
the new conditions. Encouraged by the result, we performed
the further screening of amine bases, which showed no obvious
improvement in the reaction yield while the best result was
obtained in the presence of 2,2,6,6-tetramethylpiperidine
(TMP) (Table 1, entries 2 and 3). By turning to inorganic
reducing agents, we observed that the yield of the coupling
product 3a increased significantly to 38% when Na2SO3
was used (Table 1, entry 4). Subsequently, different sulfite
salts were screened (Table 1, entries 4−7), and K2SO3 gave the
highest yield of 61% (Table 1, entry 7), probably because of its
relatively higher solubility in the solvent compared to those of
other sulfites. Using K2SO3 as the base, various types of copper
catalysts were tested (Table 1, entry 8−10). Cu(CO2CF3)2
provided a slightly higher yield of 3a (Table 1, entry 10), and
the yield increased to 69% when 2.5 equiv of K2SO3 was used
(Table 1, entry 12).
Further exploration on the choice of ligands demonstrated
that 1,10-phenanthroline (1,10-Phen) was an appropriate coor-
dination agent in this reaction system, while other bidentate
N,N- and N,O-ligands were less effective (Table 2, entries
2−8). Finally, the reaction performance was enhanced with the
addition of alcohol (Table 2, entries 9−11). Notably, the yield
of 3a increased significantly to 82% when 100 μL of EtOH
was added (Table 2, entry 10), probably due to the improved
solubilities of both K2SO3 and sodium benzenesulfinate in the
reaction mixture.
their instability. Finally, a series of substituted sodium benzene-
sulfinates were tested. 4-Methyl, 4-chloro, and 4-fluoro benzene-
sulfinates could be used as sulfenylating agents to afford the
desired thioethers in 69−78% yields (4o−4q, respectively).
Other aryl- and heteroarylsulfinates also gave reasonable yields.
We then assessed the tolerance for different aryl boron reagents
as coupling partners. Both aryl boronic acid pinacol esters and
potassium aryl trifluoroborates can be converted to the corre-
sponding thioethers with ≥50% yields (Scheme 2b and c).
A series of control experiments were performed to elucidate
the reaction mechanism (Scheme 3). Oxygen gas or air was
shown to be important for this reaction (Schemes 2 and 3a).
When sodium benzenesulfinate 2a was treated under the standard
conditions, disulfide 5 was isolated in only a 22% yield (Scheme 3b)
and sodium benzenesulfonate 6 was detected in the HRMS analysis
(Figure S1). With the addition of TEMPO, no desired product
can be obtained with or without the presence of boronic acid
1a (Scheme 3c and d), suggesting a radical mechanism for the
deoxygenation process of sulfinate.
By adding TEMPO to the standard reaction while stirring
after 4 h, the thiyl radicals were trapped by TEMPO (Scheme 3e
and Figure S2). By treating disulfide 5 with 1a, 3a was formed
with only a 53% yield. The reaction yield decreased to 28%
with the addition of TEMPO (Scheme 3f), indicating multiple
coupling mechanisms. The thiyl radicals were also trapped
when TEMPO was added to the reaction mixture between
disulfide 5 and 1a after 4 h (Scheme 3g and Figure S3). These
results suggest a radical mechanism in parallel to the coupling
reaction with the aryl boronic acid.
With the establishment of the optimized conditions, the
scope of the reaction was then explored with an array of
substituted aryl boronic acids (Scheme 2a). In the presence of
para-substituted aryl boronic acids, the reaction was compatible
with a series of electron-donating groups and electron-
withdrawing groups, with the yields from 52% to 86% (3a−3n).
Substituents at the meta-position displayed a similar substituent−
reactivity relationship with an improved isolated yield, especially
for 2-cyano and 2-chloro groups (3o−3u). For ortho-substituted
substrates (3v−3z), similar isolated yields were obtained from the
reaction. While boroxine showed a similar reactivity compared to
those of boronic acid (4a) and 2,4,6-trimethyl-substituted aryl
boronic acid gave the corresponding product in a good yield
(4b), the effect of the substitution position toward the reactivity
was more significant when dimethoxy-substituted aryl boronic
acids underwent sulfenylation. 3,5-Dimethoxyphenyl boronic
acid underwent the reaction with an excellent isolated yield
of 92% (4d), whereas the yields diminished for the 2,3- and
2,4-dimethoxy-substituted substrate(4e and 4f, respectively),
and no desired product was isolated for the 2,6-dimethoxy-
substituted substrate, reflecting the steric impact on the reaction.
Attempts for the one-pot disulfenylation of aryl diboronic
acid were also successful with phenyl diboronic acid despite
the product yield decreasing to ∼40% (4j and 4k). Additionally,
heteroaryl boronic acids can also be used as the coupling partner
(4l−4n) except 2-heteroaryl boronic acids, probably due to
Based on the results, the following reaction mechanism is
proposed (Scheme 4). Through oxidation by copper(II), sulfite
6166
Org. Lett. 2021, 23, 6164−6168