Communication
for the carbonylative coupling of bromobenzene and iso-
butyronitrile, because it had been successfully employed in the
direct a-arylation of nitriles by Hartwig and co-workers.[11a] Un-
fortunately, attempts under different pressures of carbon mon-
oxide (5 and 30 bar) did not lead to the formation of the de-
sired b-ketonitrile, and only direct arylation of isobutyronitrile
was observed. Instead, starting the reaction from iodobenzene
at a CO-pressure of 30 bar enabled the formation of 2,2-di-
methyl-3-oxo-3-phenylpropanenitrile (2) for the first time in
21% yield with a remarkable 21:1 selectivity towards the car-
bonylated product (Table 1, entry 1). Next, further optimiza-
tions were undertaken by using this model system. Changing
the solvent did not lead to any positive effect and only nonpo-
lar aromatic solvents such as toluene and benzene turned out
to be suitable (Table 1, entries 1–6). Testing different bases
showed that only strong ones provide sufficient deprotonation
of the nitrile to enable the coupling process. Thus, organic and
inorganic bases of moderate strength did not lead to any de-
sired product formation, and lithiumdiisopropylamide (LDA)
and sodium hexamethyl disilazide (NaHMDS) gave the desired
product in 14 and 21% yield, respectively (Table 1, entries 1
and 7–12). Surprisingly, similarly basic KHMDS resulted in low
conversion and no product formation.
Next, we turned our attention to determine the influence of
different ligands on our model system. The use of standard
monodentate phosphines, such as triphenylphosphine and
di(1-adamantyl)-n-butylphosphine, as well as bidentate phos-
phine ligands, such as 1,2-bis(diphenylphosphino)ethane
(DPPE), 1,2-bis(diphenylphosphino)butane (DPPB), and 1,1’-
bis(di-phenylphosphino)ferrocene (DPPF), did not promote the
model reaction (Table 1, entries 13–15 and 17–18). However, 2-
bis(diphenylphosphino)propane gave 2 in 10% yield (Table 1,
entry 16). Applying 4,5-bis(diphenylphosphino)-9,9-dimethyl-
xanthene (xantphos) resulted in an enhanced conversion and
yield, albeit a lack of selectivity induced the formation of both
coupling products 1 and 2 in 35 and 27% yield, respectively
(Table 1, entry 19). In further studies by using xantphos,
a higher catalyst loading of 4 mol% was found to be advanta-
geous for the selective formation of the b-ketonitrile in 33%
yield (Table 1, entry 20). Multiple side reactions were account-
ing for the differences between conversion and desired prod-
uct yield; thus, dehalogenation, formylation to benzaldehyde,
as well as formation of benzonitrile and a-(trimethylsilyloxy)-
phenylacetonitrile were observed as side-reactions. Applying
milder reaction conditions (808C, 5 bar of CO) suppressed
these side reactions in favor of the CꢀC-bond formation to-
wards products 1 and 2, which were formed in 26 and 52%
yield, respectively (Table 1, entry 21). Finally, a high selectivity
towards the formation of the b-ketonitrile (73% yield) could be
achieved by using a slight excess of the ligand (5 mol%) with
respect to the palladium precursor (4 mol%; Table 1, entry 21).
This suggests that the loss of ligand owing to the reduction of
PdII to the catalytically active Pd0 species causes a minimal defi-
cit of ligand, which eventually leads to the formation of less-
selective catalyst species.
Table 1. Optimization of the model system.[a]
Entry
Ligand
Solvent
Base
Conv. [%][b]
Yield [%][b]
1
2
Once the optimized reaction conditions had been estab-
lished, we turned our focus on the scope and limitations of
the reaction. The model reaction with iodobenzene led to the
isolation of 2,2-dimethyl-3-oxo-3-phenylpropanenitrile in 71%
yield (Table 2, entry 1). Notably, 4-tert-butyliodobenzene was
converted into the corresponding product in a gratifying yield
of 83% (Table 2, entry 2). Similarly, ortho-, meta- and para-
mono- as well as dimethyl-substitutions were well tolerated,
resulting in good to excellent yields of 75–87% (Table 2, en-
tries 3–6). The bicyclic 1-iodonaphthalene provided a slightly
lower yield of 67% (Table 2, entry 7). Also the presence of
alkoxy substituents on the aryl iodide substrate was tolerated
in the coupling process and resulted in yields ranging from 69
to 78% (Table 2, entries 8 and 9). 4-Chloroiodobenzene proved
to be an excellent substrate, generating the corresponding b-
ketonitrile in 87% yield, and the products of 3- and 4-fluoro-
iodobenzene were obtained in good yields of 71 and 73%, re-
spectively (Table 2, entries 10–12). Electron-deficient trifluoro-
methyl substitution was also tolerated in the ortho-, meta- and
para-positions (63 to 79% yield; Table 2, entries 13–15). Addi-
tionally, we demonstrated that heterocyclic (thiophenyl)aryl io-
dides afforded the desired b-ketonitriles in good to excellent
isolated yields (93 and 71%; Table 2, entries 16 and 17). At-
tempts to apply other aryl halides under similar reaction condi-
tions resulted in consistently lower selectivities towards the
1
2
3
4
5
6
7
8
9
10
11
12
13[c]
14[c]
15
16
17
18
19
20[d]
21[e]
22[f]
BINAP
BINAP
BINAP
BINAP
BINAP
BINAP
BINAP
BINAP
BINAP
BINAP
BINAP
BINAP
PPh3
nBuPAd2
DPPE
DPPP
DPPB
DPPF
xantphos
xantphos
xantphos
xantphos
toluene
dioxane
DMF
DMSO
benzene
THF
NaHMDS
NaHMDS
NaHMDS
NaHMDS
NaHMDS
NaHMDS
KHMDS
LDA
KOtBu
Cs2CO3
NaOAc
NEt3
NaHMDS
NaHMDS
NaHMDS
NaHMDS
NaHMDS
NaHMDS
NaHMDS
NaHMDS
NaHMDS
NaHMDS
25
30
59
72
59
73
47
93
60
84
5
1
1
0
3
0
1
0
2
0
0
0
0
0
0
0
0
0
21
0
4
0
15
0
0
14
0
0
0
0
0
0
0
10
0
0
27
33
52
73
toluene
toluene
toluene
toluene
toluene
toluene
toluene
toluene
toluene
toluene
toluene
toluene
toluene
toluene
toluene
toluene
2
30
58
55
68
42
62
90
90
>99
>99
0
35
7
26
2
[a] General reaction conditions: iodobenzene (1 mmol), isobutyronitrile
(1.2 mmol), Pd(OAc)2 (2 mol%), ligand (2 mol%), base (1.3 mmol), solvent
(2 mL), CO (30 bar), 1008C, 16 h. [b] Conversions and yields were deter-
mined on the basis of calibrated GC data by using hexadecane as an in-
ternal standard. [c] 8 mol% of ligand. [d] Pd(OAc)2 (4 mol%), xantphos
(4 mol%). [e] Pd(OAc)2 (4 mol%), xantphos (4 mol%), CO (5, bar), 808C.
[f] Pd(OAc)2 (4 mol%), xantphos (5 mol%), CO (5, bar), 808C.
&
&
Chem. Eur. J. 2014, 20, 1 – 6
2
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!