Organic Letters
Letter
of transition metal catalysts is still inevitable. Therefore, an
inexpensive, selective, and efficient method for installation of
polyfluorophenyl groups is strongly desired.
the used solvent remarkably affected the reaction outcome.
Only a trace of 3a was detected at 100 °C in 1,4-dioxane (entry
6). However, product 3a was obtained in 53% yield when the
reaction was conducted in MeCN at 80 °C (entry 7). When
the amount of 2d was decreased, the yield of 3a was slightly
reduced from 85% to 78% (entry 8). The contamination of air
was a negative factor for the reaction probably because of
moisture (entry 9). Most importantly, the reaction with
benzoyl chloride instead of 1a was inefficient, affording 3a in
43% yield (entry 10), which highlights the specific nature of
1a. Finally, the optimization of reaction time revealed that the
reaction could be complete within 12 h and 3a was isolated in
82% yield (entry 11).
As organic electrophiles for cross-couplings, acyl fluorides
can easily be prepared from the corresponding carboxylic
acids.15 In recent years, acyl fluorides have attracted a great
deal of attention due to their powerful capabilities in various
decarbonylative cross-coupling reactions.16,17 Furthermore,
acyl fluorides have served as acyl sources under transition
metal catalysis to generate various aldehydes17a,18 and
ketones19 in a carbonyl-retentive manner. During the course
of our continuing studies of the transition metal-catalyzed
transformations of acyl fluorides,20 we have encountered
unique reactivities of the C(acyl)−F bond. Inspired by
Larrosa’s work of transition metal-free decarboxylative
halogenation of aromatic carboxylic acids,6 we reasoned that
the specific properties of the C(acyl)−F bond of acyl fluorides
could achieve the transition metal-free decarboxylative cross-
coupling reactions. Herein, we describe the transition metal-
free decarboxylative cross-coupling of acyl fluorides and
potassium perfluorobenzoates to generate unsymmetrical
ketones without any catalysts and additives (Scheme 1c).
We commenced our studies by examining the reactions of
benzoyl fluoride (1a) with pentafluorobenzoic acid (2a) or its
benzoates 2b−2d, and the results are summarized in Table 1.
With the optimized reaction conditions in hand, we explored
the scope of acyl fluorides using potassium pentafluoroben-
zoate (2d) as the coupling partner (Scheme 2). A series of acyl
a
Scheme 2. Substrate Scope of Acyl Fluorides 1
a
Table 1. Optimization of the Reaction Conditions
b
entry
M
base
3a (%)
1
2
3
4
5
H (2a)
H
Li (2b)
Na (2c)
K3PO4
K2CO3
0
2
0
11
85
<1
53
78
40
−
−
−
−
−
−
−
−
−
K (2d)
c
6
K
K
K
K
K
K
d
7
e
8
f
9
g
10
11
43
h
85 (82)
a
Reactions were carried out with 1a (0.2 mmol, 1.0 equiv) and 2 (0.3
a
Reaction conditions: acyl fluorides 1 (0.2 mmol, 1.0 equiv),
mmol, 1.5 equiv) in 1,4-dioxane (1 mL) at 140 °C for 24 h under Ar.
b
potassium pentafluorobenzoate 2d (0.3 mmol, 1.5 equiv), 1,4-dioxane
(1 mL), 140 °C, 12 h. Isolated yields. Reaction performed in a 6
mmol scale for 24 h.
GC yields using n-tetradecane as an internal standard; an isolated
b
c
d
yield is shown in parentheses. At 100 °C. MeCN instead of 1,4-
e
f
g
dioxane at 80 °C. With 1.2 equiv of 2d. In air. Benzoyl chloride
h
instead of 1a. For 12 h.
fluorides 1 bearing electron-donating and -withdrawing groups
could be converted into the corresponding ketones 3a−3l in
high yields. In particular, dimethylamino (3h), cyano (3j),
ketone (3k), and ester (3l) substituents were well tolerated.
To our delight, p-nitrobenzoyl fluoride, which is less
mentioned in previous work,16−20 could be transformed to
the desired product 3m in 76% yield. In addition, halogenated
acyl fluorides, especially o-iodobenzoyl fluoride, could also be
applied to this reaction, affording ketones 3n−3q in 85−96%
yields. Not only monosubstituted substrates but also a
trisubstituted benzoyl fluoride was also coupled efficiently to
produce 3r in 95% yield. Furthermore, naphthyl-, quinolyl-,
First, potassium phosphate that showed a superior activity for
decarboxylative halogenation6 was tested, but no desired
product 3a was detected (entry 1). Similarly, the addition of
potassium carbonate as the base gave only a trace of 3a,
although all of the benzoyl fluoride was consumed (entry 2).
We assumed that the acidic hydrogen in 2a may retard the
reaction. Thus, we conducted the reaction with pentafluor-
obenzoates 2b−2d (entries 3−5, respectively). To our delight,
85% of the desired product 3a was obtained when potassium
pentafluorobenzoate (2d) was employed (entry 5). Additional
experimental results showed that the reaction temperature and
B
Org. Lett. XXXX, XXX, XXX−XXX