Organic Letters
Letter
a
from the inexpensive and commercially available salicylalde-
hydes via a two-step synthesis and without purification, would
be a very good starting material for this purpose; first, the N-
tosylhydrazone moiety could be converted into Cu-carbene in
the presence of base and Cu catalysis, which further
intramolecularly reacts with thiocarbamate unit via a
Barton−Kellogg9 type reaction, leading to o-amino benzofur-
anthiols as the key intermediate, which has been proven by the
formation of disulfide compounds via the dimerization of thiols
under oxidative conditions.7a The whole process looks just like
a formal 1,5-sulfur atom migration. Therefore, we imagine if an
appropriate electrophile is added before the dimerization
reaction, thioether products might be resulted once the
formation of 2-aminobenzofurane-3-thiol. Adding allyl bro-
mide or propargyl bromide to the system did not get the
corresponding thioether product, diazo, and other electrophiles
could not get the desired product either (see Supporting
Information for details). After a preliminary screening on
various electrophiles, silylaryl triflate demonstrated the best
efficiency as the fact that it could be readily converted into
benzyne species10,11 in situ without the interference of the
prior cyclization reaction. However, there are three selectivity
issues making the proposed pathway highly challenging, which
might greatly affect the efficiency of the transformation
(Scheme 2): (1) N-tosylhydrazone is easily deprotonated in
Table 1. Optimization of the Reaction Conditions
Scheme 2. Our Design and Challenges in Our Strategy
a
Reaction conditions: 1a (0.2 mmol), 2a (1.5 equiv), catalyst (Cu, 20
mol %; Rh, 5 mol %), “F” (2−3 equiv), base (1.2−2 equiv), solvent (1
b
c
mL), 1h, 70 °C, under N2. Isolated yield. No reaction. 12 h, 90 °C.
d
Under air. DBU = 1,8-Diazabicyclo[5.4.0]undec-7-ene; TBAF =
tetrabutylammonium fluoride; TBAT = tetrabutylammonium triphe-
nyldifluorosilicate.
Therefore, the optimized conditions were eventually found as
CuBr2 as the catalyst and TBAT as fluorine source with the
help of DBU in 1,4-dioxane under N2 atmosphere (Table 1,
entry 12).
With the optimized reaction conditions in hand, we next
explored the scope of reaction between thiocarbamate 1 and
silylaryl triflate 2 (Scheme 3). The reaction was successful for
both electron-donating (3ba−3ea) and electron-withdrawing
(3fa−3ja) substituents on the aromatic rings. The absolute
structure of 3aa was unambiguously validated by X-ray
crystallographic structure analysis. Various functionalities,
such as fluorine (3fa, 3ga), bromide (3ia), and a nitro group
(3ja), were all tolerated well, leading to the desired products in
decent yields, respectively. In addition, the polyphenyl system,
such as naphthalene, was also compatible with the standard
system (3ka). Alkynyl is a common synthetic unit which could
be handled for further transformation; to our delight, the
substrate containing alkynyl group in our system could also be
successfully transformed into the target product (3la) in good
yield. Gratifyingly, heteroaromatic substrates such as thiophene
(3ma) was good candidate as well. Meanwhile, the substituents
on nitrogen atom have little effect on the reaction; when
methyl is replaced with ethyl group, the target product was
obtained in 60% yield (3na). Furthermore, when the benzyne
precursor containing two methyl substitutions was used, the
corresponding 4-amino-3-thioaryl benzofuran (3ab) was
obtained in 62% yield.
basic conditions, which might trap the arylative electrophiles to
lead to N-arylation byproducts. (2) Bamford−Stevens12
reaction of p-methylbenzenesulfonyl in basic conditions
releases an anion Ts−, which might further react with the
arylative electrophile to form a Ts-arylation byproduct,
consuming the electrophiles. (3) After the formation of diazo
compounds, a [3 + 2] cycloaddition13 reaction might occur
between it and benzyne species generated in situ from silylaryl
triflate with the diazo compound as 1,3-dipole (Scheme 2).
With these challenges in mind, we began our studies using
N-tosylhydrazone thiocarbamate (1a) and silylaryl triflate 2a as
starting materials in the presence of CuBr2 and DBU as a base
and cesium fluoride as a fluorine source in THF at 70 °C
(Table 1). To our delight, the product of 3aa was obtained
from the above reaction with 37% isolated yield. Further
catalysis screening suggested that CuBr2 was the optimal one
compared with CuCl2 and Rh2(OAc)4 (Table 1, entries 2 and
3). The formation proved to be the best one (Table 1, entry
6). Alternative fluoride sources other than TBAT indicated less
efficiency in promoting yields (Table 1, entries 6−9). Base
exploration suggested that DBU was the superior one (Table 1,
entries 10−12). When the reaction was carried out in air, no
corresponding product was detected (Table 1, entry 13).
Although silylaryl triflate 2 as an arylative electrophile
demonstrated good reactivity in our transformation, the
B
Org. Lett. XXXX, XXX, XXX−XXX
Li, Xue
