Mild and practical method for the α-arylation of nitriles with heteroaryl halides

Article abstract of DOI:10.1021/jo051737f

A mild and transition-metal-free method for theα-arylation of a liphatic nitriles with activated heteroaryl halides was developed using NaHMDS o r KHMDS as base at ambient temperature. The key to the success of this method is generation of the nitrile anion in the presence of the heteroaryl halide. The method is applicable to both primary and secondary carbonitriles and a wide range of heteroaryl halides. Selective monoarylation was observed with primary carbonitriles. The operational simplicity and the mild reaction conditions add to the value of this methodas a practical alternative to the preparation of α-heteroaryl carbonitriles.

Full text of DOI:10.1021/jo051737f

TABLE 1. Arylation of Nitrile 1 with
Mild and Practical Method for the
r-Arylation of Nitriles with Heteroaryl
Halides
2
,5-Dibromopyridine 2ba
Artis Klapars,* Jacob H. Waldman,* Kevin R. Campos,
Mark S. Jensen, Mark McLaughlin, John Y. L. Chung,
Raymond J. Cvetovich, and Cheng-yi Chen
Department of Process Research, Merck Research
Laboratories, P.O. Box 2000, Rahway, New Jersey 07065
base,
equiv
conv of conv of yield of
b
entry catalyst
T, °C
1, %
2b, %
3, %
c
artis_klapars@merck.com; jacob_waldman@merck.com
Received August 17, 2005
1
2
3
4
5
6
7
Verkade 1.2
60
n.d.
90
86
54
86
93
89
n.d.
80
74
31
82
94
86
44
53
57
5
65
42
61
c
Verkade 1.2
25
none
none
none
none
none
1.2
1.2 (sep)
1.0
1.5
1.2
25
d
25
25
25
e
-40 to 25
a
General procedure: A 1.0 M solution of NaHMDS was added
to a solution of 1 (1.0 equiv) and 2b (1.0 equiv) in toluene at 25
C, and the reaction was aged for 15-20 h. HPLC assay yield.
b
°
c
4
mol % of Pd(OAc)2 and 8 mol % of 2,8,9-triisobutyl-2,5,8,9-
d
tetraaza-1-phosphabicyclo[3.3.3]undecane. NaHMDS was added
to a solution of nitrile 1 in toluene at 25 °C, and the mixture was
aged at 25 °C for 1 h and then added to bromide 2b. e NaHMDS
was added to a solution of 1 and 2b in toluene at -40 °C, and the
reaction mixture was allowed to warm to 25 °C over 15 h.
A mild and transition-metal-free method for the R-arylation
of aliphatic nitriles with activated heteroaryl halides was
developed using NaHMDS or KHMDS as base at ambient
temperature. The key to the success of this method is
generation of the nitrile anion in the presence of the
heteroaryl halide. The method is applicable to both primary
and secondary carbonitriles and a wide range of heteroaryl
halides. Selective monoarylation was observed with primary
carbonitriles. The operational simplicity and the mild reac-
tion conditions add to the value of this method as a practical
alternative to the preparation of R-heteroaryl carbonitriles.
only one example of a heteroaryl fluoride was included,
and applicability of this method to heteroaryl halides
other than fluorides was not addressed.
While developing a route to a drug candidate, we
encountered a particularly challenging R-arylation reac-
tion of a functionalized cyclopropanecarbonitrile 1 (avail-
able as a 1:1 mixture of diastereomers) with 2,5-
dibromopyridine (2b) (eq 1). Several Pd-catalyzed coupling
The R-arylation of aliphatic nitriles is a valuable
synthetic transformation that has remained largely
undeveloped. The Pd-catalyzed R-arylation of nitriles has
1
recently emerged as a major advancement; however, its
utility is still limited by the relatively harsh coupling
conditions (i.e., strong base, high reaction temperature)
which may not be compatible with complex substrates
containing base-sensitive groups. Direct nucleophilic
conditions were applied to this transformation, yet only
the procedure reported by Verkade1a provided any of the
desired product in a modest 44% yield (Table 1, entry 1).
A single diastereomer of 3 was obtained, which was
presumably due to severe steric hindrance from the
concave side of the metalated nitrile intermediate. In-
terestingly, the yield of the coupling reaction could be
improved to 53% if the reaction was performed at room
temperature instead of 60 °C (entry 2). This result was
somewhat unexpected because the reported examples of
Pd-catalyzed arylation of nitriles have typically required
elevated temperatures. In our case, the heteroaryl halide
N
aromatic substitution (S Ar) provides an alternative
approach but also suffers from the use of strong bases
n
2
3
such as BuLi or NaNH
temperatures and moderate yields. An improved proce-
dure for the S Ar arylation of secondary nitriles with
2
as well as high reaction
4
N
various aryl fluorides, even including electron-rich aryl
5
fluorides, was recently disclosed by Caron. However,
(
1) (a) You, J.; Verkade, J. G. Angew. Chem., Int. Ed. 2003, 42, 5051.
(
b) You, J.; Verkade, J. G. J. Org. Chem. 2003, 68, 8003. (c) Culkin, D.
A.; Hartwig, J. F. Acc. Chem. Res. 2003, 36, 234. (d) Culkin, D. A.;
Hartwig, J. F. J. Am. Chem. Soc. 2002, 124, 9330.
(
2) Skerlj, R. T.; Bogucki, D.; Bridger, G. J. Synlett 2000, 1488.
2
b was highly activated, and we suspected that the nitrile
(
3) (a) Sperber, N.; Papa, D.; Schwenk, E.; Sherlock, M.; Fricano,
anion could undergo an uncatalyzed substitution at the
R. J. Am. Chem. Soc. 1951, 73, 5752. (b) Carver, D. R.; Greenwood, T.
D.; Hubbard, J. S.; Komin, A. P.; Sadcheva, Y. P.; Wolfe, J. F. J. Org.
Chem. 1983, 48, 1180.
6
pyridine ring. Indeed, a comparable yield of 57% and
>
99% de was obtained if the Pd catalyst was omitted
(
4) (a) Loupy, A.; Philippon, N.; Pigeon, P.; Galons, H. Heterocycles
1
991, 32, 1947. (b) Sleevi, M. C.; Cale, A. D. Jr.; Gero, T. W.; Jaques,
L. W.; Welstead, W. J.; Johnson, A. F.; Kilpatrick, B. F.; Demian, I.;
Nolan, J. C.; Jenkins, H. J. Med. Chem. 1991, 34, 1314.
(5) Caron, S.; Vazquez, E.; Wojcik, J. M. J. Am. Chem. Soc. 2000,
122, 712.
10.1021/jo051737f CCC: $30.25 © 2005 American Chemical Society
10186
J. Org. Chem. 2005, 70, 10186-10189
Published on Web 10/21/2005

TABLE 2. Stability Tests in the Presence of NaHMDS at
ing was carried out at -78 °C (eq 2). The deprotonated
a
2
5 °C
entry
b
compd
time
15 h
15 h
15 h
15 h
5 min
1 h
15 h
assay, %
1
2
3
4
5
6
7
1
2b
3
40
17
18
96
93
45
0.5
2-bromopyridine
3-bromopyridine
3-bromopyridine
3-bromopyridine
nitrile was stable only at low temperatures because a
significantly lower assay (35% total of both diastereo-
mers) was obtained at 0 °C along with a complex mixture
a
General procedure: A 1.0 M solution of NaHMDS in THF was
added to a solution of the compound in toluene at 25 °C and aged
b
at 25 °C for the time indicated in the table. HPLC assay of the
9
of unidentified decomposition products. In an attempt
compound remaining in the reaction mixture.
to minimize the decomposition of the deprotonated nitrile,
one R-arylation experiment was performed wherein the
NaHMDS base was added to a solution of both starting
materials 1 and 2b at -40 °C, and the reaction mixture
was allowed to slowly reach room temperature (Table 1,
entry 7). However, very little improvement in the product
yield was observed, which could be due to an insufficient
rate of R-arylation at -40 °C.
(
entry 3). This promising lead invited further explorations
of the uncatalyzed nucleophilic aromatic substitution
reaction.
In the preceding examples (Table 1, entries 1-3),
sodium hexamethyldisilazide base (NaHMDS) was added
to a mixture of both starting materials. When the
deprotonation of nitrile 1 was carried out separately and
was then followed by addition of bromide 2b, a dramati-
cally reduced yield of product 3 was obtained (Table 1,
entry 4). Even when the base was added to a mixture of
both starting materials 1 and 2b, only a partial conver-
sion was observed using 1.2 and 1.0 equiv of base (Table
It became clear that in order to overcome the base-
promoted decomposition of the starting materials and the
product, the relative rate of the R-arylation had to be
10
11
increased. Several solvents and bases were screened
with the bromide 2b under the premise that an S Ar
N
reaction would exhibit a strong dependence on the solvent
polarity and the counterion, yet no improvements over
the original conditions (THF/toluene solvent, NaHMDS
base) could be achieved. We then resorted to a different
approach whereby the reactivity of the pyridine electro-
phile was tuned through the choice of the leaving group.
Gratifyingly, replacing the 2-bromo substituent in 2b
with a 2-fluoro substituent (2d) resulted in a dramatic
1
, entries 3 and 5). When 1.5 equiv of the NaHMDS base
was used in an effort to drive the reaction to completion,
a significantly lowered yield of 3 was actually observed
despite some of the starting material still remaining
(
Table 1, entry 6). These results strongly indicated that
a base-promoted decomposition was competing with the
desired R-arylation reaction.
1
2
The decomposition pathways were investigated in more
detail. When starting materials 1, 2b, and product 3 were
separately treated with 1.5 equiv of NaHMDS under the
conditions mimicking the R-arylation reaction, a signifi-
cant decomposition was observed in all three cases (Table
increase in the product yield (Table 3, entry 9). Inter-
estingly, the effects of the metal cation and the leaving
group were interrelated because NaHMDS provided the
best results with the bromide 2b (Table 3, entry 3) while
KHMDS gave the highest yield with fluoride 2d (Table
3, entry 10). The latter example was demonstrated on a
2 g scale in a 96% assay yield, and the product was
isolated in a 92% yield (99 wt % purity) after crystalliza-
tion from 2:1 water-ethanol.
2
, entries 1-3). The bromide ion was detected by HPLC
as the only identifiable decomposition product of com-
pounds 2b and 3. This observation was consistent with
a decomposition mechanism involving an unstable pyri-
dyne intermediate.7 A brief exploration of two model
systems, 2-bromopyridine and 3-bromopyridine, showed
that 2-bromopyridine was perfectly stable in the presence
of NaHMDS (Table 2, entry 4) while 3-bromopyridine
decomposed rapidly (Table 2, entries 5-7). This allowed
us to pinpoint the 3-bromo substituent as the most likely
source of instability in the starting material 2b and
product 3.
Although the optimized conditions for the R-arylation
of the sensitive nitrile 1 with fluoropyridine 2d resembled
5
the procedure developed by Caron, which mainly tar-
geted the R-arylation of nitriles with unactivated aryl
fluorides,13 we found that in our case the reaction
proceeded under significantly milder conditions. The
reaction could be performed at room temperature instead
of heating at 70 °C; furthermore, we were able to use 1.0
equiv of the nitrile and 1.0 equiv of base instead of 4.0
and 1.5 equiv, respectively, without compromising the
Decomposition of nitrile 1 was addressed next. Starting
with a pure convex diastereomer of 1, clean epimerization
and the expected 50% deuterium incorporation were
8
observed when the deprotonation and deuterium quench-
(9) The concave diastereomer of 1 provided comparable results in
the deprotonation-deuteration experiments.
(
10) The following solvents provided lowered yields of 3 compared
t
(6) Displacement of halo substituents at R- and γ-positions of
to PhMe/THF: DMSO, DMA, PhCN, BuCN, THF.
pyridines via an addition-elimination mechanism is well precedented
with other nucleophiles. See: Joule, J. A.; Mills, K.; Smith, G. F.
Heterocyclic Chemistry; Chapman and Hall: London, 1995; p 82.
(11) A low yield of 3 was obtained with the following bases: DBU/
t
t
t
PhMe at 80 °C, NaO Bu/ BuOH at 50 °C, NaO Bu/PhMe at 80 °C, NaH/
t
n
THF at 60 °C, KO Bu/PhMe at rt, BuLi at -78 °C to rt, LiTMP at
-78 °C to rt.
(7) (a) Zoltewicz, J. A.; Smith, C. L. Tetrahedron 1969, 25, 4331. (b)
Mallet, M.; Queguiner, G. Tetrahedron 1982, 38, 3035. (c) Connon, S.
J.; Hegarty, A. F. J. Chem. Soc., Perkin Trans. 1 2000, 1245.
(12) The same product 3 was obtained from either 2b and 2d, which
is consistent with the 2-halo and not the 5-bromo substituent being
displaced on the pyridine ring.
(
8) We expect only 50% deuterium incorporation because hexa-
methyldisilazane, which was formed in the NaHMDS deprotonation
step, contains an exchangeable proton.
(13) Only one example of a heteroaryl fluoride was reported by
Caron.
J. Org. Chem, Vol. 70, No. 24, 2005 10187

TABLE 3. Effect of the Leaving Group and the Cation
on the Arylation of Nitrile 1a
entries 9 and 10), presumably because 1 equiv of the base
was consumed to deprotonate the product. Acetonitrile
suffered extensive dimerization when 2-chloropyridine
was used as the arylating agent (entry 12). Fortunately,
the dimerization was suppressed when the more reactive
2
1
-fluoropyridine was used as the arylating agent (entry
1).
The reaction could be extended to other activated
heteroaryl halides (Table 4, entries 13-18). Although
-chloropyrimidine (entry 15) turned out to be a chal-
b
entry
X
M
conv of 1, % conv of 2, % yield of 3, %
2
1
2
3
4
5
6
7
8
9
I
Na
86
81
78
18
76
75
7
75
65
36
92
96
48
10
63
35
8
62
40
29
83
96
lenging substrate, the arylation was highly successful
with other heterocycles including an isoquinoline, py-
ridazine, pyrazine, benzoxazole, and benzothiazole. These
reactions could be readily carried out with heteroaryl
chlorides, which are generally more available and less
expensive than the corresponding heteroaryl fluorides.
In summary, we have developed a mild, practical, and
transition-metal-free method for the R-arylation of ali-
phatic nitriles with activated heteroaryl halides using
NaHMDS or KHMDS as base at ambient temperature.
The key to the success of this method is generation of
the nitrile anion in the presence of a heteroaryl halide
whereby any competing decomposition pathways are
minimized. The method is applicable to both primary and
secondary carbonitriles and a wide range of heteroaryl
halides. Selective monoarylation is observed with pri-
mary carbonitriles. In most cases, either heteroaryl
bromides, chlorides, or fluorides could be used as the
arylating agents although the more reactive aryl fluorides
provided better results with particularly sensitive sub-
strates. We believe that the operational simplicity and
the mild reaction conditions add to the value of this
method as a practical alternative to the preparation of
R-heteroaryl carbonitriles.
Br Li
Br Na
82
Br
Cl
Cl
Cl
F
F
F
K
71
Li
Na
K
71
73
58
Li
Na
K
97
c
96
d
10
>99
a
General procedure: A solution of MHMDS was added to a
solution of 1 (1.0 equiv) and 2b (1.0 equiv) in toluene at 25 °C,
b
and the reaction mixture was aged for 15-20 h. HPLC assay
yield. c Demonstrated on 1 g of 1 in 89% assay and 86% isolated
d
yield (99 wt % purity, heptane as crystallization solvent). Demon-
strated on 2 g of 1 in 96% assay and 92% isolated yield (99 wt %
purity, 2:1 water-ethanol as crystallization solvent).
product yield. Consequently, we became interested if
these mild conditions were applicable to R-arylation of
nitriles with other heteroaryl halides.
The R-arylation of cyclopropanecarbonitrile (4) with
bromide 2b turned out to be even more challenging than
the arylation of 1 (Table 4, entry 2). Nevertheless, the
use of the fluoride 2d provided a 75% isolated yield (Table
4
, entry 1). These results were in contrast to the arylation
of isobutyronitrile, which proceeded in a high yield even
with the base-sensitive bromide 2b (Table 4, entry 3).14
Exploring the substrate scope further, we observed that
a 2-chloro group reacted exclusively in the presence of a
Experimental Section
(
1R*,5S*,6S*)-6-(5-Bromopyridin-2-yl)-6-cyano-3-aza-
bicyclo[3.1.0]hexane-3-carboxylic Acid tert-Butyl Ester (3).
An oven dried 100 mL round-bottom flask with magnetic stirring
and an internal temperature probe was evacuated and backfilled
three times with nitrogen. The flask was charged with nitrile 1
(a 1:1 mixture of diastereomers, 2.00 g, 9.60 mmol, 1.0 equiv)
and 10 mL of dry toluene and cooled to 0 °C. 2-Fluoro-5-
bromopyridine (1.69 g, 9.60 mmol, 1.0 equiv) was added followed
by KHMDS (20.4 mL of 0.47 M solution in toluene, 9.60 mmol,
3
-fluoro group on the pyridine ring (Table 4, entry 4). The
reaction also tolerated relatively hindered secondary
nitriles (Table 4, entries 5 and 6). We then turned our
attention to primary nitriles, which according to Caron
5
are unreactive toward unactivated aryl fluorides. In our
case, primary nitriles turned out to be excellent sub-
strates for arylation with heteroaryl halides (entries
1
.0 equiv) over 20 min, maintaining the internal temperature
7
-12). Moreover, only the monoarylation products were
observed even if 2 equiv of the arylating agent was used
entry 8). This selectivity is interesting considering that
below 3 °C. The reaction was stirred at 0 °C for 1 h, after which
time the cold bath was removed and the viscous brown mixture
was stirred at room temperature overnight. The solution dark-
ened as it warmed to room temperature. The reaction was
checked for completeness by TLC analysis, and the reaction
mixture was diluted with 80 mL of MTBE and washed with
water (3 × 40 mL). The organic layer was separated and
concentrated to ∼30 mL and the solvent switched to ethanol (3
× 30 mL). Water (40 mL) was added dropwise with stirring for
90 min at room temperature to precipitate the product. A single
diastereomer of the product (3.24 g, 93%) was then isolated as
a tan powder: 98.9 HPLC area %; 99.4 wt % by HPLC; mp
(
the products of the monoarylation reactions are sterically
similar to 2-phenylbutyronitrile in entry 6, which is a
competent arylation substrate. Apparently, the electron-
withdrawing pyridyl group is detrimental to the nucleo-
philicity of the nitrile anion. In fact, the reactions with
primary nitriles proceeded best with 2 equiv of base (cf.
(14) The lowered acidity of cyclopropanecarbonitrile, caused by the
1
46.7-148.5 °C; IR (film) 1699.78, 1403.86, 1372.92, 1171.97,
nonplanar nature of the carbanion, can be invoked to explain the
reactivity difference between cyclopropanecarbonitrile and isobuty-
ronitrile. See: (a) Peerboom, R. A. L.; de Koning, L. J.; Nibbering, N.
M. M. J. Am. Soc. Mass Spectrom. 1994, 5, 159. (b) Juchnovski, I. N.;
Tsenov, J. A.; Binev, I. G. Spectrochim. Acta A 1996, 52, 1145. However,
we were able to show that the anion of the functionalized cyclopro-
panecarbonitrile 1 could be generated even at -78 °C using NaHMDS.
Interference from an unidentified decomposition pathway thus appears
to be a more likely explanation for the poor reactivity of cyclopropan-
ecarbonitriles in these R-arylation reactions.
-
1 1
1122.46 cm ; H NMR (400 MHz, CDCl
1H, pyridine H-6), 7.82 (dd, J ) 2.4, 8.4 Hz, 1H, pyridine H-4),
.67 (d, J ) 8.4 Hz, 1H, pyridine H-3), 3.96 (d, 1H, J ) 12.2
Hz), 3.87 (d, 1H, J ) 12.1 Hz), 3.73 (m, 2H), 2.66 (m, 2H), 1.49
3
) δ 8.50 (d, J ) 2.4 Hz,
7
13
(
s, 9H); C NMR (100 MHz, CDCl
22.5, 119.3, 116.2, 80.4, 47.3, 47.0, 35.7, 35.1, 28.7, 26.5. Anal.
Calcd for C16 Br: C, 52.75; H, 4.98; N, 11.54. Found: C,
52.72; H, 4.94; N, 11.47.
3
) δ 153.9, 152.4, 150.8, 139.5,
1
18 2 3
H O N
10188 J. Org. Chem., Vol. 70, No. 24, 2005

TABLE 4. Coupling of Various Nitriles with Heteroaryl Halides
General Procedure for Coupling of Nitriles with Het-
eroaryl Halides (Table 4). An oven-dried Schlenk tube with
magnetic stir bar was evacuated and backfilled with nitrogen
three times and charged with nitrile (1.00 mmol), aryl halide
was filtered through a syringe filter (0.45 µm), and the filtrate
was purified by silica gel column chromatography to afford the
coupled products.
(
1.00 mmol), and dry toluene. The reaction mixture was cooled
Acknowledgment. We thank Tom Novak for the
high-resolution mass spectral data and Robert Reamer
and Peter Dormer for help with the NMR analysis.
to 0 °C, and hexamethyldisilazide base (1.00 mmol for tertiary
nitriles and 2.00 mmol for primary and secondary nitriles) was
added dropwise over 5 min. For heteroaryl fluorides, a 0.47 M
solution of KHMDS in toluene was used. For heteroaryl chlorides
or bromides, a 1.0 M solution of NaHMDS in THF was used.
The total reaction volume was 3 mL. The solution was then
stirred at 0 °C for 1 h and then at rt overnight. Reaction
completion was checked by TLC analysis, the reaction mixture
Supporting Information Available: Experimental pro-
cedures and characterization data. This material is available
free of charge via the Internet at http://pubs.acs.org.
JO051737F
J. Org. Chem, Vol. 70, No. 24, 2005 10189