Palladium-catalyzed cross-coupling reactions of benzyl indium reagents with aryl iodides

Article abstract of DOI:10.1021/jo802280m

Herein described is an operationally simple procedure for generating benzyl indium species from readily available benzyl bromides and indium metal followed by in situ palladium-catalyzed coupling with aryl halides. The procedure provides diarylmethanes in

Full text of DOI:10.1021/jo802280m

a bimetallic catalyst consisting of [Ir(COD)Cl]2 and SnCl42 that
promotes the reaction of aldehydes with electron-rich benzenes.
The resulting benzyl alcohol continues to react giving substituted
diarylmethanes. The reaction worked best with electron-donating
groups in the arene and electron-withdrawing groups in the
aldehyde. We targeted compounds that would contain electron-
withdrawing groups in both aromatic rings. Furthermore, we
could not rely on classical directing-group effects to control
regiochemistry due to the varying substitution patterns in our
substrates.
Palladium-Catalyzed Cross-Coupling Reactions of
Benzyl Indium Reagents with Aryl Iodides
Louis S. Chupak,*,† Joanna P. Wolkowski,‡ and
Yves A. Chantigny§
Groton Laboratories, Pfizer Inc., Eastern Point Road,
8220/2255, Groton, Connecticut 06340
louis.chupak@bms.com
Transition-metal catalysis has become a powerful tool for the
formation of diarylmethanes from aryl zinc or aryl boronic acids.
A recent report for coupling aryl zincs with benzyl halides
showed excellent tolerance of functional groups on the aryl rings
and on the benzylic position.3 The only drawback to this method
was the requirement for in situ zinc activation. The most
promising methods for diarylmethane synthesis are the Suzuki-
Miyaura-type couplings of benzylic halides,4a benzylic carbon-
ates,4 or benzylic phosphates.4 These methods show excellent
functional-group tolerance, proceed under mild conditions, and
provide predictable connectivity regardless of the aryl substitu-
tion pattern. The only limitation is the availability of the desired
aryl boronic acids. This limitation is rapidly fading as more
aryl boronic acids become commercially available. 5 We sought
a simple procedure that would take advantage of the many
benzyl halides and aryl halides that are commercially available.
We chose to prepare a benzyl metal species from a benzyl
halide and couple it with an aryl halide. Benzyl Grignard,6a
benzyl zinc,6b or benzyl indium6c reagents can be coupled with
aryl halides to provide diarylmethanes. Benzyl Grignards couple
in high yields with aryl and heteroaryl iodides under copper
catalysis. Unfortunately, sensitive functional groups are not
compatible with these conditions. Organozinc and organoindium
reagents are well-known for having excellent functional group
compatibility, but these reagents are often prepared from more
reactive benzyl metal species.7 Knochel described the use of
zinc dust and lithium chloride to generate reactive benzyl zinc
chlorides directly from the benzyl chlorides.6b These organozinc
reagents were used in a variety of palladium-catalyzed coupling
reactions. Minehan described a lithium chloride-promoted
insertion of indium into aryl halides that avoids using more
ReceiVed October 16, 2008
Herein described is an operationally simple procedure for
generating benzyl indium species from readily available
benzyl bromides and indium metal followed by in situ
palladium-catalyzed coupling with aryl halides. The proce-
dure provides diarylmethanes in modest to excellent yield
and tolerates a variety of functional groups in both coupling
partners.
As part of a drug discovery program, we required diaryl-
methanes as synthetic intermediates. We sought a simple
procedure with broad functional-group compatibility. Our goal
was to have a method that was compatible with parallel-
synthesis techniques, tolerant of most functional groups, and
avoided using highly reactive organometallic intermediates. In
particular, our synthetic targets had to contain an ester or nitro
group on one of the aryl rings.
To the best of our knowledge, there are three useful methods
for making diarylmethanes: (1) reduction of diarylketones or
diarylcarbinols; (2) Friedel-Crafts alkylation; and (3) transition-
metal-catalyzed couplings of either a benzyl halide with an aryl
metal species or an aryl halide with a benzyl metal species.
Reduction protocols do not address the carbon-carbon bond
formation, often require the presence of electron-rich aromatic
rings, and lack the broad functional-group compatibility we
required. While hydride donors are the typical reductants,
recently Provot and co-workers1 described a high-yield dispro-
portionation reaction of diarylmethylisopropyl ethers under acid
catalysis. The authors demonstrated the selective reduction of
a diarylcarbinol in the presence of a diaryl ketone. This method
did not meet our criteria because the diarylcarbinols were formed
from aryl Grignards or aryllithium intermediates. Recently, Roy
reported a milder version of the Friedel-Crafts alkylation using
benzyl alcohol to produce diarylmethanes. The authors described
(2) For bimetallic catalysis see: Podder, S.; Choudhury, J.; Roy, U. K.; Roy,
S. J. Org. Chem. 2007, 72, 3100. Choudhury, J.; Podder, S.; Roy, S. J. Am.
Chem. 2005, 127, 6162. For Friedel-Crafts to prepare biarylmethanes see:
Kobayashi, T.; Rahman, S. M. Synth. Commun. 2003, 33, 3997. Olah, G. A.;
Kobayashi, S.; Tashiro, M. J. Am. Chem. Soc. 1972, 94, 7488.
(3) Amator, M.; Gosmini, C. Chem. Commun. 2008, 5019.
(4) (a) Burns, M. J.; Fairlamb, I. J. S.; Kapdi, A. R.; Sehnal, P.; Taylor,
R. J. K. Org. Lett. 2007, 9, 5397. Chahen, L.; Doucet, H.; Santelli, M. Synlett
2003, 1668. Nobre, S. M.; Monteiro, A. L. Tetrahedron Lett. 2004, 45, 8225.
Langle, S.; Abarbri, M.; Duchene, A. Tetrahedron Lett. 2003, 44, 9255. (b)
Kuwano, R.; Kondo, Y.; Shirahama, T. Org. Lett. 2005, 7, 2973. (c) McLaughlin,
M. Org. Lett. 2005, 7, 4875.
(5) A current SciFinder search shows 1797 commercially available phenyl-
boronic acids.
(6) (a) Ku, Y.; Patel, R. R.; Sawick, D. P. Tetrahedron Lett. 1996, 37, 1949.
(b) Metzger, A.; Schade, M. A.; Knochel, P. Org. Lett. 2008, 10, 1107. (c)
Mosquera, A.; Riveiros, R.; Perez Sestelo, J.; Sarandeses, L. A. Org. Lett. 2008,
10, 3745. (d) Perez, I.; Perez Sestelo, J.; Sarandeses, L. A. J. Am. Chem. Soc.
2001, 123, 4155.
† Current address: Early Discovery Chemistry, Bristol-Myers Squibb Co., 5
Research Parkway, Wallingford, CT 06492.
‡ Current address: Sierra Nevada Brewing Company.
§ Current address: MethylGene Inc.
(7) (a) Lee, P. H.; Sung, S.; Lee, K. Org. Lett. 2001, 3, 3201–3204. (b) Lee,
P. H.; Lee, S. W.; Seomoon, D. Org. Lett. 2003, 5, 4963. (c) Riveiros, R.; Saya,
L.; Sestelo, J. P.; Sarandeses, L. A. Eur. J. Org. Chem. 2008, 1959.
(1) L’Hermite, N.; Giraud, A.; Provot, O.; Peyrat, J.-F.; Alami, M.; Brion,
J.-D. Tetrahedron 2006, 62, 11994, and references cited therein.
1388 J. Org. Chem. 2009, 74, 1388–1390
10.1021/jo802280m CCC: $40.75 2009 American Chemical Society
Published on Web 12/19/2008

TABLE 1. Pd-Catalyzed Cross-Coupling Reactions of Aryl Halides with Benzyl Indiums derived from Benzyl Halides
reactive organometallic precursors. The resulting aryl indium
species react under palladium catalysis with a variety of
electrophiles.8
bromides. Palladium-catalyzed coupling with aryl halides
produced diarylmethanes that contained the desired ester or nitro
groups.
Our best results followed the procedure of Lee for the cross-
coupling of allyl halides with aryl halides.7a In our adaptation
of this procedure, unmodified, commercially available indium
powder was added to a DMF solution of the desired benzyl
halide. Disappearance of the benzyl halide was monitored by
partitioning a reaction aliquot between diethyl ether and water
followed by thin-layer chromatography of the diethyl ether
portion. This analysis showed that the reaction was rapid and
usually complete in less than 1 h. No precautions to exclude
air or moisture were taken during the formation of the benzyl
To the best of our knowledge, diarylmethanes have not been
prepared by the insertion of unactivated indium into a benzyl
halide followed by palladium-catalyzed coupling. Here we report
our results using unactivated indium metal in the absence of
additives to generate benzyl indium species. Our simple
procedure generates benzyl indium reagents containing ester,
carboxylic acid, and cyano groups from readily available benzyl
(8) Papoian, V.; Minehan, T. J. Org. Chem. 2008, 73, 7376.
J. Org. Chem. Vol. 74, No. 3, 2009 1389

indium. After the benzyl halide was consumed, the reaction
mixture was added to the remaining reactants under nitrogen
and heated at 100 °C until the aryl halide was consumed, usually
overnight. We did not attempt to modify or optimize the reaction
when low yields were obtained. Also, we did not attempt to
optimize the catalyst because of our initial success with
tetrakis(triphenylphosphine)palladium(0). It should be noted that
in the absence of any palladium catalyst no product was
observed.
In summary, a simple procedure for generating benzyl indium
species from readily available benzyl bromides and indium metal
followed by in situ palladium-catalyzed coupling with aryl
halides was developed. The method compares favorably with
existing methods in that it uses readily available starting
materials, is compatible with a variety of functional groups, and
is simple to perform. Many benzyl and aryl halides are
commercially available or easily prepared. Access to these
starting materials is an advantage for our method when the
corresponding aryl boronic acids are unavailable, unstable, or
expensive. We have shown that carboxylic acid, ester, cyano,
and amino groups are compatible with the reaction conditions.
The yields were highest when both coupling partners contained
electron-withdrawing groups. This result compares well with
the Friedel-Crafts alkylation where at least one partner must
contain an electron-donating group. While the examples pre-
sented here are limited by our choice of medicinal chemistry
targets, our method provided diarylmethanes containing a variety
of functional groups in useful yields. The reaction is simple to
perform because we prepared the benzyl indium intermediates
without the use of more reactive benzyl metal intermediates. A
detailed, general procedure is given in the Experimental Section.
The examples in Table 1 show that a wide variety of
functional groups are compatible with this procedure. For
example, electron-withdrawing groups are tolerated in both
partners as evidenced by the reaction of 2-iodo-4-nitrocymene
and 3,5-bis(trifluoromethyl)benzyl bromide in 93% yield (entry
20). Electron-donating groups in the aryl halide tended to give
modest yields (entries 9 and 11). The reaction was not sensitive
to steric bulk in the aryl halide coupling partner. An isopropyl
group in the aryl halide ortho position gave a range of yields
from 33% to 93% (entries 3, 6, 7, and 20). The electron-rich o-
and m-methoxy-substituted benzyl bromides failed to couple
as did p-methoxybenzyl chloride (not shown). In contrast,
p-trifluoromethoxybenzyl bromide, in which the electron-
donating ability of the oxygen has been attenuated, reacted to
provide the desired product in 81% yield (entry 21). The
presence of a carboxylic acid resulted in slightly reduced yields
compared to the parent ester (compare entries 10 and 13). The
cross-coupling also worked when basic groups were present,
but the yields were significantly reduced (entries 9 and 12),
although in entry 12 the low yield may be due to the use of an
aryl bromide instead of an aryl iodide as the electrophile.
Surprisingly, para-substituted benzyl bromides consistently
coupled in lower yields than their meta or ortho isomers. This
reduced yield was particularly apparent for p-cyanobenzyl
bromide, which cross-coupled in only 23% yield (entry 17).
We identified several side products for this reaction that
accounted for a portion of the starting materials. The homo-
coupled aryl iodide (2,2′-dimethyl-4,4′-dinitrobiphenyl, 16%
yield), reduced benzyl bromide (4-methylbenzonitrile, 4% yield),
reduced aryl iodide (1-methyl-3-nitrobenzene, 8% yield), and
recovered 1-iodo-2-methyl-4-nitrobenzene (2% yield) were
isolated by chromatography on silica gel. The o- and m-
cyanobenzyl bromides gave the desired products in 59% and
57% yield, respectively. Sulfur-containing heterocycles also
cross-coupled, but in low yields (entries 22 and 23). A final
limitation was the failure of R-methylbenzyl bromide to react
under the standard conditions (entry 24) to give R-methyldiaryl-
methane. In this instance, the benzyl bromide was consumed
in the reaction with indium metal, but failed to couple with the
iodobenzene. This failure was a surprising disappointment in
light of the fact that sterics seemed to have little impact on the
reaction when the bulky group was in the electrophile. Such
R-methyldiarylmethanes have been prepared from the reaction
of R-methylbenzyl chloride with an arylzinc by using CoBr2 as
the catalyst.3
Experimental Section
Representative Procedure for the Synthesis of (2-Methyl-5-
nitrobenzyl)benzene (Table 1, Entry 1). To 5 mmol of benzyl
bromide (855 mg, 5 mmol) in 5 mL of DMF was added indium
(861 mg, 7.5 mmol). The reaction mixture was stirred at room
temperature until the disappearance of starting material as assessed
by TLC. To a separate flask were sequentially added 2-iodo-1-
methyl-4-nitrobenzene (526 mg, 2 mmol), LiCl (170 mg, 4 mmol),
Pd(PPh3)4 (116 mg, 0.1 mmol), and the benzyl indium solution.
Additional dimethylformamide (5 mL) was used to complete the
transfer of the benzyl indium to the reaction. The resulting reaction
mixture was placed under a nitrogen atmosphere and stirred at 100
°C until no starting aryl iodide was observed by TLC. After dilution
with Et2O (100 mL), the reaction mixture was washed with 3 N
HCl (5 × 20 mL), water (3 × 20 mL), and brine (1 × 20 mL).
The organic portion was dried over MgSO4 and filtered, and the
solvents were removed by rotary evaporation. Chromatography of
the residue on SiO2 with 1:3 ethyl acetate:hexanes provided the
1
title compound (265 mg, 58%). H NMR (400 MHz, CDCl3) δ
2.33 (s, 3H), 4.04 (s, 2H), 7.10 (d, 7.5 Hz, 2H), 7.20-7.25 (m,
1H), 7.27-7.31 (m, 3H), 7.99 (m, 2H). 13C NMR (500 MHz,
CDCl3) δ 20.2, 36.9, 121.7, 124.1, 126.7, 127.2, 129.2, 131.1, 131.2,
136.7, 136.8, 140.6, 145.0, 146.9. LC/MS 3.0 min. FAB HRMS
calcd for C14H13NO2 (M+) 228.1024, found 228.1036.
Acknowledgment. We thank Jon Bordner and Ivan Samard-
jiev for X-ray crystallography and Pfizer for supporting this
work.
Supporting Information Available: Characterization data
for all compounds and X-ray crystallography for entry 14 in
Table 1. This material is available free of charge via the Internet
at http://pubs.acs.org.
JO802280M
1390 J. Org. Chem. Vol. 74, No. 3, 2009