Organic Letters
Letter
cinnamic acid derivatives,12d or alkenyl zinc reagents11f using a
decarboxylative approach with a redox-active ester.
a,b
Scheme 3. Decarboxylative Alkenylation
Choosing redox-active N-(acyloxy)phthalimide 1q as the
radical precursor, we were pleased to find that 44% yield of
alkenylated product was obtained using Na2−eosin Y as the
photoredox catalyst and N,N-diisopropylethylamine (DIPEA)
as the base in dioxane under the photoirradiation of 24 w blue
optimizing the reaction conditions, the yield was further
improved to 81% in the presence of eosin Y (4.3 mol %),
DIPEA (10 equiv), and (n-Bu)4NPF6 (2.2 equiv) in a mixed
solvent of Et2O and methyl tert-butyl ether. Control experi-
ments demonstrated that all reaction parameters are crucial for
this transformation (Scheme 2). The reaction did not proceed
Scheme 2. Evaluation of Reaction Parameters for
a,b
Decarboxylative Alkenylation.
a
Reaction conditions: 1q (0.3 mmol), vinyl sulfonium salt 2 (0.6
mmol), photocatalyst (4.3 mol %), DIPEA (10 equiv), Et2O/TBME
b
(1.4 mL/0.3 mL), (n-Bu)4NPF6 (2.2 equiv), blue LED, N2, 9 h. The
yield was determined by crude 1H NMR using CH2Br2 as the internal
a
Reaction conditions: 1 (0.3 mmol), vinyl sulfonium salt 2a (0.6
mmol), eosin Y (4.3 mol %), DIPEA (10 equiv), Et2O/TBME (1.4
mL/0.3 mL), (n-Bu)4NPF6 (2.2 equiv), blue LED, rt, N2, 9 h.
c
standard. Isolated yield.
b
c
Isolated yield. The reaction was conducted on 2.0 mmol.
in the absence of blue light, a photocatalyst of eosin Y, or
DIPEA (entries 2−4). Although a similar yield was obtained
without the addition of (n-Bu)4NPF6, this additive helped to
improve the reproducibility of this reaction, probably due to its
capability of enhancing the solubility of the sulfonium salts in
ether (entry 5). The use of ether or TBME alone resulted in
inferior results (entries 6 and 7). To our surprise, this radical
alkenylation could partially tolerate the moisture and O2,
although a lower yield was obtained under air atmosphere
(entry 8). Other organic and metal photocatalysts, solvents,
and bases all led to insufficient reaction outcomes (entries 9−
12, 15−20). Notably, the high loading of DIPEA are also
crucial for the high yield of this reaction. We have also
investigated the efficiency of sulfonium salt variants (2a−d), in
which tetrahydrothiophene-derived sulfonium salt 2a gave the
best outcome.
synthetic useful yields. The tolerance of alkyl chloride (3h),
alkyl bromide (3i), and unmasked alkyne (3j) was noteworthy
because those functionalities were normally sensitive under
photoredox conditions. In addition, this procedure is also
compatible with complex natural products and pharmaceuticals
(3x−ac), such as myristic acid, oleic acid, fenbufen,
dehydrocholic acid, isoxepac, and indomethacin, which further
demonstrated it is highly valuable for the late-stage
functionalization of bioactive molecules. The scalability of
this process was demonstrated using 1q as model substrate,
giving the desired alkenylated product 3q in 60% yield.
Next, the scope of the vinyl sulfonium salts was examined
using N-(acyloxy)phthalimide 1a as the alkyl radical source
(Scheme 4). Substituents at the ortho-, meta-, and para-
positions of the arene are all tolerant (5a−c and 5d−f), and
meta-substituted styrene-derived sulfonium salts normally
provided desired 1,2-trans-alkene in higher yields. Fluorinated
vinyl sulfonium salts are also suitable alkenyl sources for this
reaction (5g−h), while tert-butylated vinyl sulfonium salt gave
With the optimal conditions in hand, the generality of the
decarboxylative alkenylation reaction was evaluated. As
summarized in Scheme 3, primary (3a−j), secondary (3k−
t), and tertiary (3u−w) carboxylic acids are all compatible with
this procedure to afford the desired alkenylation products in
B
Org. Lett. XXXX, XXX, XXX−XXX