CHEMCATCHEM
COMMUNICATIONS
additive to tune the acidity of the reaction system. Accordingly,
a range of alkali salts were examined.[17] The results revealed
that addition of NaOAc or KOAc improved the yield of 4b to
higher than 50% (Table 1, entries 5–7). In addition, only a slight-
ly increased yield was observed if the reaction was performed
under a nitrogen atmosphere as opposed to an oxygen atmos-
phere (data not shown). Finally, by optimizing the number of
equivalents of the catalyst and additives (Table 1, entries 8–10),
we obtained product 4b in 55% yield in the presence of PdCl2
(10 mol%), AgOAc (1.5 equiv.), and NaOAc (5.0 equiv.; Table 1,
entry 9). Further extensive efforts toward improving the reac-
tion efficiency were futile, because the homocoupling of 2b
was observed as a major side reaction and it could not be suc-
cessfully suppressed under various conditions.
Scheme 1. Reaction of phosphinate 1 with aryl iodide 2a in acetic acid.
(Scheme 1). In the presence of Pd(OAc)2 and Ag2CO3, arylated
product 3 was obtained in 52% yield based on 1. In addition,
tolylacetic acid (4a) was isolated as a byproduct in approxi-
mately 6% yield based on iodotoluene (2a). A literature survey
showed that in a 2007 report by Daugulis,[14] the same side re-
action was observed in their study on the Pd-catalyzed aryla-
tion of benzoic acids in acetic acid solvent. Motivated by these
observations, we envisioned that the reaction of carboxylic
acids with aryl halides under acidic conditions would offer
a new opportunity to realize the a-arylation of carboxylic acids
[Eq. (7)]. The successful demonstration of this method and
a preliminary mechanistic inspection will be presented herein.
Thus, we examined the scope and limitations of the reaction
by varying the aryl iodides (Table 2). The results show that for
aryl iodides with an unsubstituted phenyl ring or with
a phenyl ring substituted with a weak electron-donating Me
group, the a-arylacetic acids were obtained in moderately high
yields (see 4a–4d). However, the yield was somewhat dimin-
ished if the Me group was replaced by a stronger electron-do-
nating OMe group (see 4 f). In contrast, for aryl iodides bearing
electron-withdrawing groups, the reactions proceeded more
efficiently than those with electron-donating groups. Generally,
moderately high yields of approximately 60–70% were ob-
served for substrates modified by a rich variety of electron-de-
ficient groups such as F (see 4g and 4h), Cl (see 4j and 4k), Br
(see 4l and 4m), CO2Me (see 4n and 4o), CF3 (see 4p and
4q), and NO2 (see 4r). 4’-Iodoacetophenone afforded a some-
what diminished yield (see 4s). Moreover, the reaction exhibit-
ed good chemoselectivity, as the F, Cl, and Br groups remained
intact (see 4g–4m) during the reaction. Finally, the yields were
lower for ortho-substituted aryl iodides 4e and 4i. These re-
sults indicate that the reaction is sensitive to steric hindrance.
Notably, for most of the reactions, products resulting from the
homocoupling of the aryl iodides were observed as major by-
products.
The optimization of the reaction parameters was performed
through the reaction of iodobenzene (2b) in neat acetic acid.
Representative data are summarized in Table 1. Extensive
screening of various Pd catalysts[15] showed that Pd(OAc)2, Pd-
(CF3CO2)2, and PdCl2 in the presence of silver salts such as
silver acetate (AgOAc) or silver trifluoromethanesulfonate
(AgOTf)[16] were promising combinations that could afford ary-
lated product 4b in approximately 40% yield (Table 1, en-
tries 1–3 and 7). We then investigated the effect of a weak
base on this transformation, which was assumed to act as an
Next, to gain some mechanistic information for this new
transformation, a series of control experiments were per-
formed. Upon treatment of aryl iodide 2o with CD3CO2D under
standard conditions (Table 3, entry 1), analysis of product 4o-D
Table 1. Optimization of the reaction conditions.[a]
1
by H NMR spectroscopy revealed that the deuterated and un-
deuterated ratio for the H-a proton was approximately D-a/
H-a=5.45:1, which indicated the generation of 15.5% undeu-
terated H-a. This is understandable because both AgOAc and
NaOAc may also serve as the acetic acid sources to participate
in competitive arylation with CD3CO2D. However, an interesting
observation was that the H-b and H-c protons in the phenyl
ring were partially deuterated. The deuterated ratio for
H-b and H-c was approximately D-b/H-b=4.56:1 and D-c/
H-c=1:5.06, respectively. The data showed that the ortho
H-b proton could be deuterated much more easily than the
meta H-c proton. A similar trend was also observed by replac-
ing CD3CO2D with CH3CO2D (Table 3, entry 2). These experi-
ments imply that proton exchange of H-a, H-b, and H-c with
D+ takes place during the arylation reaction. A similar deutera-
tion reaction was also reported very recently by Yu.[18]
Entry
Pd catalyst
Pd(OAc)2
Silver salt [equiv.] Additive [equiv.] Yield [%][b]
1
2
3
4
5
6
7
8
9
10
AgOAc (2.0)
–
–
42
42
Pd(O2CCF3)2 AgOAc (2.0)
PdCl2
PdCl2
PdCl2
PdCl2
PdCl2
PdCl2
PdCl2
PdCl2
AgOAc (2.0)
–
–
–
43
trace
50
53
50
53
55
45
AgOAc (2.0)
AgOAc (2.0)
AgOTf (2.0)
AgOAc (2.0)
AgOAc (1.5)
AgOAc (1.0)
KOAc (5.0)
NaOAc (5.0)
NaOAc (5.0)
NaOAc (10.0)
NaOAc (5.0)
NaOAc (5.0)
[a] Reaction conditions: 2b (1.0 mmol), Pd catalyst (10 mol%), AcOH
(3.0 mL), 1308C, 24 h. [b] The yield was determined by 1H NMR spectros-
copy owing to contamination by trace diarylacetic acid 5b after isolation.
The structure of 5b was determined by analysis of the mixture of 4b and
5b by HRMS and 1H NMR and 13C NMR spectroscopy.
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2014, 6, 1589 – 1593 1590