Intermolecular coupling of alkenes to heterocycles via C-H bond activation

Article abstract of DOI:10.1021/jo048666p

The intermolecular coupling of unactivated alkenes to a range of heterocycles using a Rh(I) catalyst was investigated. A variety of functional groups were incorporated into the alkene, including esters, nitriles, acetals, and phthalimide. Furthermore, the heterocycle tolerated substitution with both electron-rich and electron-deficient groups. The intermolecular coupling became possible after it was discovered that weak acids dramatically increase the rate of both the inter- and intramolecular reactions. An extensive optimization of additives was performed, and HCl·PCy₃ (Cy = cyclohexyl) and HCl·P-t-Bu₂Et were in general found to be the best additives for the reaction.

Full text of DOI:10.1021/jo048666p

In ter m olecu la r Cou p lin g of Alk en es to Heter ocycles via C-H Bon d
Activa tion
Kian L. Tan, Steve Park, J onathan A. Ellman,* and Robert G. Bergman*
Department of Chemistry, University of California and Division of Chemical Sciences,
Lawrence Berkeley National Laboratory, Berkeley, California 94720
jellman@uclink.berkeley.edu; bergman@cchem.berkeley.edu
Received August 2, 2004
The intermolecular coupling of unactivated alkenes to a range of heterocycles using a Rh(I) catalyst
was investigated. A variety of functional groups were incorporated into the alkene, including esters,
nitriles, acetals, and phthalimide. Furthermore, the heterocycle tolerated substitution with both
electron-rich and electron-deficient groups. The intermolecular coupling became possible after it
was discovered that weak acids dramatically increase the rate of both the inter- and intramolecular
reactions. An extensive optimization of additives was performed, and HCl‚PCy3 (Cy ) cyclohexyl)
and HCl‚P-t-Bu2Et were in general found to be the best additives for the reaction.
In tr od u ction
catalysts to perform hydroarylations with electron-rich
aromatics.6 Fujiwara has demonstrated that electron rich
aromatics will undergo nondirected coupling to electron
deficient alkenes and alkynes at room temperature.7
Sames extended the scope of this methodology to an
intramolecular cyclization using a Ru(III) catalyst, al-
lowing access to chromanes, tetralins, and terpenoids.8
Several groups have focused on directed C-H activa-
tion to increase both the selectivity and reactivity of these
processes. In 1993, Murai reported that a ketone could
be used to direct C-H activation at the ortho position of
an aromatic ring, and subsequent coupling to an alkene
yielded the ortho alkylated product.9 Since this initial
report, several groups have expanded the scope of this
reaction to include a variety of directing groups such as
imines,10 nitriles,11 ketones,12 esters,13 pyridines,14 and
other heterocycles. The Bergman and Ellman labs have
developed an intramolecular cyclization using an imine
Metal-catalyzed C-H bond activation1 reactions have
become important methods for the formation of carbon-
carbon bonds.2 The addition of aryl C-H bonds to
alkenes, known as hydroarylation, has been a rapidly
developing field. Two major approaches have been de-
veloped to accomplish hydroarylation:3 nondirected C-H
activation and directed C-H activation by a pendant
functional group or embedded heteroatom. Nondirected
C-H activation processes have the benefit of not requir-
ing a functional handle on the starting material; however,
the regioselectivity of the functionalization is a major
challenge in this area. Periana has developed Ir catalysts
that couple benzene derivatives to alkenes.4 While Peri-
ana’s Ir-based catalysts have shown good turnover num-
bers and alkene generality, the process has suffered from
modest regioselectivity on both the arene and alkene
component. Recently, Gunnoe reported a Ru catalyst for
this process, which functions at a lower temperature.5
Several groups have used electrophilic C-H activation
(6) (a) Ferreira, E. M.; Stoltz, B. M. J . Am. Chem. Soc. 2003, 125,
9578-9579. (b) Beccalli, E. M.; Broggini, G. Tetrahedron Lett. 2003,
44, 1919-1921.
(1) For reviews of C-H activation see the following references: (a)
Shilov, A. E.; Shul’pin, G. B. Chem. Rev. 1997, 97, 2879-2932. (b)
Ryabov, A. D. Chem. Rev. 1990, 90, 403-424.
(2) For reviews on hydroarylation see the following references: (a)
Miura, M.; Nomura, M. In Cross-Coupling Reactions; Springer-Verlag
Berlin: Berlin, 2002; Vol. 219, pp 211-241. (b) Guari, Y.; Sabo-Etienne,
S.; Chaudret, B. Eur. J . Inorg. Chem. 1999, 1047-1055. (c) Dyker, G.
Angew. Chem., Int. Ed. 1999, 38, 1699-1712.
(3) (a) Kakiuchi, F.; Murai, S. Topics in Organometallic Chemistry;
1999; Vol. 3. (b) Ritleng, V.; Sirlin, C.; Pfeffer, M. Chem. Rev. 2002,
102, 1731-1769. (c) J ia, C. G.; Kitamura, T.; Fujiwara, Y. Acc. Chem.
Res. 2001, 34, 633-639.
(7) (a) J ia, C. G.; Piao, D. G.; Oyamada, J . Z.; Lu, W. J .; Kitamura,
T.; Fujiwara, Y. Science 2000, 287, 1992-1995. (b) Fujiwara, Y.; J ia,
C. G. Pure Appl. Chem. 2001, 73, 319-324.
(8) Youn, S. W.; Pastine, S. J .; Sames, D. Org. Lett. 2004, 6, 581-
584.
(9) (a) Murai, S.; Kakiuchi, F.; Sekine, S.; Tanaka, Y.; Kamatani,
A.; Sonoda, M.; Chatani, N. Nature 1993, 366, 529-531. (b) Murai,
S.; Kakiuchi, F.; Sekine, S.; Tanaka, Y.; Kamatani, A.; Sonoda, M.;
Chatani, N. Pure Appl. Chem. 1994, 66, 1527-1534. (c) Kakiuchi, F.;
Sekine, S.; Tanaka, Y.; Kamatani, A.; Sonoda, M.; Chatani, N.; Murai,
S. Bull. Chem. Soc. J pn. 1995, 68, 62-83.
(10) J un, C. H.; Hong, J . B.; Kim, Y. H.; Chung, K. Y. Angew. Chem.,
Int. Ed. 2000, 39, 3440.
(4) (a) Matsumoto, T.; Taube, D. J .; Periana, R. A.; Taube, H.;
Yoshida, H. J . Am. Chem. Soc. 2000, 122, 7414-7415. (b) Periana, R.
A.; Liu, X. Y.; Bhalla, G. Chem. Commun. 2002, 3000-3001. (c)
Matsumoto, T.; Periana, R. A.; Taube, D. J .; Yoshida, H. J . Catal. 2002,
206, 272-280. (d) Matsumoto, T.; Periana, R. A.; Taube, D. J .; Yoshida,
H. J . Mol. Catal. A: Chem. 2002, 180, 1-18. (e) Oxgaard, J .; Muller,
R. P.; Goddard, W. A.; Periana, R. A. J . Am. Chem. Soc. 2004, 126,
352-363.
(11) Kakiuchi, F.; Sonoda, M.; Tsujimoto, T.; Chatani, N.; Murai,
S. Chem. Lett. 1999, 1083-1084.
(12) Lenges, C. P.; Brookhart, M. J . Am. Chem. Soc. 1999, 121,
6616-6623.
(13) Trost, B. M.; Imi, K.; Davies, I. W. J . Am. Chem. Soc. 1995,
117, 5371-5372.
(14) (a) Lim, Y. G.; Kim, Y. H.; Kang, J . B. Chem. Commun. 1994,
2267-2268. (b) Lim, Y. G.; Kang, J . B.; Kim, Y. H. Chem. Commun.
1996, 585-586. (c) Lim, Y. G.; Kang, J . B.; Kim, Y. H. J . Chem. Soc.,
Perkin Trans. 1 1998, 699-707.
(5) (a) Lail, M.; Arrowood, B. N.; Gunnoe, T. B. J . Am. Chem. Soc.
2003, 125, 7506-7507. (b) Oxgaard, J .; Goddard, W. A. J . Am. Chem.
Soc. 2004, 126, 442-443.
10.1021/jo048666p CCC: $27.50 © 2004 American Chemical Society
Published on Web 09/23/2004
J . Org. Chem. 2004, 69, 7329-7335
7329

Tan et al.
SCHEME 1
In addition to an examination of substrate scope, we
provide an account of the optimization of the additive and
phosphine, as well as postulate a rationale of the additive
effect.
Resu lts a n d Discu ssion
In ter m olecu la r Cou p lin g of Neoh exen e to Het-
er ocycles. In earlier work, we had reported the intramo-
lecular coupling of alkenes to benzimidazole, using
[RhCl(coe)2]2 (coe ) cis-cyclooctene) and PCy3 (Cy )
cyclohexyl) as the catalyst. Application of this catalyst
system to the intermolecular coupling of benzimidazole
to neohexene resulted in no reaction. However, the
addition of certain Lewis or Brønsted acids resulted in a
dramatic increase in the rate of the cyclization chemistry
and made the intermolecular coupling accessible.21 Though
several Lewis and Brønsted acids accelerated the reac-
tion, weak Brønsted acids afforded the highest yields for
the largest range of heterocycle substrates. Using luti-
dinium chloride as a cocatalyst, the heterocycle generality
was expanded to include benzthiazole, benzoxazole, 4,5-
dimethylthiazole, and purines (Table 1). 1-Methylbenz-
imidazole also functioned as a coupling partner, although
higher catalyst loading was required to obtain high
conversion (Table 1, entry 6). The success of the reaction
with these substrates is consistent with the formation
of an N-heterocyclic carbene complex as a critical inter-
mediate on the catalytic cycle. No reaction was observed
between neohexene and either pyrimidine or indole,
which is expected since these compounds are not known
to form N-heterocyclic carbenes. However, isoxazole,
indazole, and 1-phenylpyrazole also did not couple with
neohexene, even though they could form carbene com-
plexes. Recently, Herrmann and co-workers showed that
in the formation of Rh(I) carbene complexes with azolium
salts the reaction rate decreased in the order benzimi-
dazole > triazole > imidazole > pyrazole, which cor-
relates with the decreasing acidity of the salt.22 We
believe this may also explain the lack of reactivity of
1-phenylpyrazole, indazole, and isoxazole in our system.
Interestingly, purine was found to undergo multiple
alkylations, unlike the other heterocycles tested. Moni-
directing group,10 which has allowed access to a variety
of functionalized carbo- and heterocyclic products.15
We became interested in further developing our hy-
droarylation methodology using heterocycles as the cou-
pling partner. The selective functionalization of hetero-
cycles is of particular importance due to the ubiquity of
these structures in natural products, as well as in
pharmaceutical agents. Prior to our initial communica-
tion, the only account of a catalytic hydroarylation of a
heterocycle at the ortho position was by Taylor and
J ordan, who reported the ortho alkylation of pyridine
with ethylene and propene using a Zr catalyst.16 We
initially developed an intramolecular cyclization of
alkenes to benzimidazole and imidazole scaffolds.17 This
methodology allowed for the cyclization of numerous
substituted alkenes, yielding a variety of heterocyclic
scaffolds. In the course of studying the mechanism of this
reaction, an N-heterocyclic carbene Rh(I) complex was
identified and characterized (Scheme 1).18 The interme-
diacy of this NHC complex suggested that a broad range
of heterocycles could function under this mechanism,
thereby increasing the scope of the C-C bond-forming
reaction dramatically.19 Due to the limitations of the
intramolecular reaction, which requires a tethered alkene
proximal to the site of C-H activation, we were inter-
ested in exploring conditions for a general intermolecular
coupling reaction. After an initial screening of additives,
it was discovered that addition of weak Lewis and
Brφnsted acids dramatically accelerated the rate of the
reaction, thus making the intermolecular coupling of
alkenes to heterocycles possible (eq 1).20
1
toring the reaction by H NMR spectroscopy established
that the reactions proceed sequentially, with initial
alkylation occurring at the 8 position, followed by reaction
at the 6 position (Table 1, entry 5).23
Su r vey of th e F u n ction a l Gr ou p Com p a tibility on
th e Alk en e Cou p lin g P a r tn er . With these encouraging
results, a survey of alkene generality was undertaken
using benzimidazole as the model heterocycle. One of the
major limitations in the C-H coupling to alkenes thus
far has been the lack of success with isomerizable
alkenes. We were delighted to find that with the new
conditions for intermolecular coupling, 1-hexene reacts
efficiently with benzimidazole giving the linearly coupled
product 9 in 80% isolated yield (Table 2, entry 2).
Attempts at coupling cyclohexene with benzimidazole
In this paper, a detailed study of the intermolecular
coupling reaction of heterocycles to alkenes is reported.
(15) (a) Thalji, R. K.; Ahrendt, K. A.; Bergman, R. G.; Ellman, J . A.
J . Am. Chem. Soc. 2001, 123, 9692-9693. (b) Ahrendt, K. A.; Bergman,
R. G.; Ellman, J . A. Org. Lett. 2003, 5, 1301-1303.
(16) (a) J ordan, R. F.; Taylor, D. F. J . Am. Chem. Soc. 1989, 111,
778-779. (b) Dagorne, S.; Rodewald, S.; J ordan, R. F. Organometallics
1997, 16, 5541-5555. (c) J ordan, R. F.; Bradley, P. K.; Lapointe, R.
E.; Taylor, D. F. New J . Chem. 1990, 14, 505-511.
(17) Tan, K. L.; Bergman, R. G.; Ellman, J . A. J . Am. Chem. Soc.
2001, 123, 2685-2686.
(18) Tan, K. L.; Bergman, R. G.; Ellman, J . A. J . Am. Chem. Soc.
2002, 124, 3202-3203.
(19) Lewis, J . C.; Wiedemann, S. H.; Bergman, R. G.; Ellman, J . A.
Org. Lett. 2004, 6, 35-38.
(21) Addition of Bu4NCl did not result in acceleration of the reaction
showing that the additive effect is not a result of ionic strength of the
medium.
(22) Kocher, C.; Herrmann, W. A. J . Organomet. Chem. 1997, 532,
261-265.
(23) The regioselectivity of the alkylation of 6 was determined by
X-ray crystallography.
(20) (a) Tan, K. L.; Bergman, R. G.; Ellman, J . A. J . Am. Chem.
Soc. 2002, 124, 13964-13965. (b) Tan, K. L.; Vasudevan, A.; Bergman,
R. G.; Ellman, J . A.; Souers, A. J . Org. Lett. 2003, 5, 2131-2134. (c)
Wiedemann, S. H.; R. G. Bergman; Ellman, J . A. Org. Lett. 2004, 6,
1685-1687.
7330 J . Org. Chem., Vol. 69, No. 21, 2004

Coupling of Alkenes to Heterocycles via C-H Bond Activation
TABLE 1. In ter m olecu la r Cou p lin g of Neoh exen e to
Heter ocycles
TABLE 2. Su r vey of F u n ction a l Gr ou p Toler a n ce on
th e Alk en e Cou p lin g P a r tn er
a
2.5 mol % [RhCl(coe)2]2, 7.5 mol % PCy3, 5 mol % lutidinium
Cl-, and 5 equiv of neohexene, at 150 °C in THF. 5 mol %
b
[RhCl(coe)2]2, 15 mol % PCy3, 5 mol % lutidinium Cl-, and 5 equiv
of neohexene at 150 °C in THF. c 5 mol % [RhCl(coe)2]2, 10 mol %
HClPCy3, and 5 equiv of neohexene, at 150 °C in THF.
a
Reactions run with 2.5 mol % [RhCl(coe)2]2, 7.5 mol % PCy3,
5 mol % lutidinium Cl-, and 5 equiv of alkene at 150 °C, in THF.
b
Reactions run with 5 mol % [RhCl(coe)2]2, 15 mol % PCy3, 5 mol
% lutidinium Cl-, and 5 equiv of alkene at 150 °C, in THF.
c Reactions run with 5 mol % [RhCl(coe)2]2, 10 mol % HClPCy3,
and 5 equiv of alkene at 150 °C, in THF.
were unsuccessful, suggesting that only terminal alkenes
are effective substrates.
With this initial success, we explored the electronic
requirements of the alkene and the functional group
compatibility of the substrates under the reaction condi-
tions. A silyl-protected alcohol was stable under the
C-H/alkene coupling conditions (Table 2, entry 3). Ester
11 was synthesized by coupling an allylic ester to
benzimidazole, although an increase in the catalyst
loading was necessary to obtain a high yield (Table 2,
entry 4). A poor yield was observed when styrene was
used as the coupling partner (Table 2, entry 5). Polym-
erization of styrene was observed under the conditions,
which may have accelerated the catalyst decomposition.
The acetal of acrolein reacted with benzimidazole to give
the functionalized heterocycle 13 (Table 2, entry 6). The
pinacol protecting group was used because other acetals,
e.g., acrolein dimethyl acetal and 2-vinyl-1,3-dioxolane,
led either to decomposition or poor yields. The steric bulk
of the pinacol may hinder isomerization and provides a
more robust protecting group, thereby eliminating po-
tential decomposition pathways. While reaction with
alkenes incorporating a carbamate protecting group
resulted in decomposition or poor yields, use of a vinyl
phthalimide as a coupling partner resulted in good yields
of the desired amine-protected product 14 (Table 2, entry
7).24 This reaction is also notable because it demonstrates
that electron-rich alkenes are effective coupling partners.
Investigation of electron deficient R,â-unsaturated esters
with benzimidazole as a coupling partner resulted in
hydroamination rather than the desired C-H coupling
products. We then tested the Michael acceptors with 4,5-
dimethylthiazole, which is unable to undergo the hy-
droamination reaction. Acrylonitrile and tert-butyl acry-
late both underwent coupling to 4,5-dimethylthiazole,
demonstrating that electron-deficient alkenes are also
suitable substrates for the reaction (Table 2, entries 8
and 9). Acrylonitrile provided mostly the desired linear
isomer, but a small amount of branched isomer was also
observed (L/B ) 3.8:1).
Su r vey of th e F u n ction a l Gr ou p Com p a tibility on
th e Heter ocycle Cou p lin g P a r tn er . The functional
group tolerance on the heterocycle component was also
(24) A carbamate protecting group was used successfully in the
coupling with an oxazoline derivative; see ref 20c.
J . Org. Chem, Vol. 69, No. 21, 2004 7331

Tan et al.
TABLE 3. F u n ction a l Gr ou p Com p a tibility on th e
Heter ocycle Cou p lin g P a r tn er
SCHEME 2
ize back to the terminal alkene and then couple, the
concentration of desired 1-hexene is dramatically low-
ered, thereby causing a large decrease in the overall rate
of reaction. At 105 °C, the reaction will go to completion
after 2.5 days, demonstrating that the catalyst is robust.
These results suggested that the reaction could likely be
improved if the rate of coupling was increased relative
to the rate of isomerization.
Because it was unknown if the use of an additive
affected the rate of isomerization and coupling indepen-
dently, a screen of additives and solvents was undertaken
using benzthiazole and 1-hexene as the model com-
pounds at a reaction temperature of 75 °C. A set of Lewis
acids and lutidinium bromide was screened in THF,
CH3CN, toluene, 1,2-dichloroethane, and o-dichloroben-
zene (o-DCB) (Figure 1). Generally, the additive effect
was most dramatic in THF, with TiCl4‚THF providing
the highest yield; however, none of the additives were
found to be more effective than HClPCy3. Other solvents
appeared to suppress the coupling reaction. With these
initial leads, several other metal complexes were screened,
and a variety of Ti derivatives were evaluated (Table 4).
Unfortunately, none of the new metals screened improved
the yield of the reaction at this lower temperature.
Furthermore, changing the ligands on the Ti center was
detrimental to the reaction. Use of the Lewis acidic
Ti(O-i-Pr)4 gave almost no conversion, demonstrating the
importance of the chloride ligand.
P r op osed Mech a n ism for In ter m olecu la r Cou -
p lin g w ith HClP Cy3 a s Ad d itive. The inability to
identify a Lewis acid that accelerated the reaction more
effectively than HClPCy3 raised the possibility that the
Lewis acids screened were themselves simply generating
HX in situ rather than acting as Lewis acids. The
reactions were performed with dry solvents and reagents;
however, the additives may be metallating at the acidic
N-H bond of the NHC-Rh intermediate thereby gener-
ating HX. A possible mechanism by which a Brønsted
acid could accelerate the coupling reaction is via the
formation of a Rh(III) hydride (Figure 2). Several groups
have reported the isolation of Rh(III) and Ir(III) hydrides
from the addition of HCl to Rh(I) and Ir(I) phosphine
complexes.26 We speculate that N-heterocyclic carbene/
alkene complex 23 initially forms, which then oxidatively
a
Reactions run with 2.5 mol % [RhCl(coe)2]2, 5 mol % HClPCy3,
b
and 5 equiv of alkene at 150 °C, in THF. Reactions run with 5
mol % [RhCl(coe)2]2, 10 mol % HClPCy3, and 5 equiv of alkene at
150 °C, in THF.
examined. For this series of experiments HClPCy325 was
employed as both the additive and phosphine. This
modification simplifies the reaction setup, and it also
renders the phosphine air stable for long-term storage.
Under these conditions, a variety of substituted benz-
imidazoles were successfully coupled to neohexene. Both
electron-rich and electron-deficient substituents were
tolerated on the ring, and excellent yields were obtained
for both types of substitution (Table 3, entries 1 and 5).
A free carboxylic acid on the heterocycle resulted in a
lower yield of the desired product compared to the other
functional groups screened (Table 3, entry 2). The lower
yield may be attributed to the poor solubility of the
carboxylic acid in THF because replacing the carboxylic
acid functionality with a more soluble ester derivative
resulted in an excellent yield of the alkylated product
(Table 3, entry 3). An aryl chloride was also coupled
efficiently, leaving the halide substituent intact for
subsequent elaboration via cross-coupling chemistry.
Exp a n d ed Solven t a n d Ad d itive Scr een . While the
use of HClPCy3 did solve many technical problems, it did
not improve the reactivity of the catalyst significantly,
which led to an investigation of the causes of the modest
reactivity. Monitoring the reaction of benzthiazole with
1-hexene at 75 and 105 °C by 1H NMR spectroscopy
showed that the major impediment to efficient coupling
was olefin isomerization (Scheme 2). Within 2 h at 75
and 105 °C, 1-hexene had equilibrated to a mixture
of alkene isomers. Under these conditions, only 2-4%
(0.1-0.2 equiv) of the initial amount of 1-hexene re-
mained. Though the internal alkene isomers can isomer-
(25) In our synthesis of HCl‚PCy3 we found that a variable amount
(1-2 equiv) of HCl was incorporated. However, this did not appear to
affect the efficiency of the reaction.
(26) (a) Grushin, V. V.; Alper, H. Organometallics 1991, 10, 1620-
1622. (b) Gusev, D. G.; Bakhmutov, V. I.; Grushin, V. V.; Volpin, M.
E. Inorg. Chim. Acta 1990, 175, 19-21. (c) J ames, B. R.; Preece, M.;
Robinson, S. D. Inorg. Chim. Acta 1979, 34, L219-L221. (d) Intille, G.
M. Inorg. Chem. 1972, 11, 1, 695.
7332 J . Org. Chem., Vol. 69, No. 21, 2004

Coupling of Alkenes to Heterocycles via C-H Bond Activation
F IGURE 1. Screen of additives and solvents.
TABLE 4. Exp a n d ed Scr een of Meta l Com p lex
Ad d itives
additive
benzthiazole (%)
22 (%)
total (%)
Lut. Br-
AlCl3
(O)MoCl4
MoCl5
(O)VCl3
V2O5
BCl3 SMe2
TiBr4
Cp2TiCl2
CpTiCl3
Cp*TiCl3
Ti(O-i-Pr)4
75
76
78
73
75
92
73
72
85
72
71
89
20
15
14
14
18
0
17
18
5
95
92
92
87
93
92
90
91
90
93
93
90
20
22
1
a
Reaction carried out with 5 mol % [RhCl(coe)2]2, 10 mol %
PCy3, and 10 mol % additive in THF at 75 °C; yields were
determined by GC versus an internal standard.
adds HCl forming Rh(III) hydride 24 (Figure 2). The
hydride then inserts into the alkene yielding Rh alkyl
complex 25. Alkyl migration onto the C-2 position of the
heterocycle (complex 26, Figure 2) followed by â-hydride
elimination would yield alkylated product 27. Alterna-
tively, metal alkyl 25 could reductively eliminate an
imidazolium salt, which through proton transfer would
reform Rh(III) hydride 28. Cavell and co-workers have
shown that stoichiometric reductive elimination of imi-
dazolium salts occur from Pd(II) complexes,27 and that
Ni(0) complexes can undergo direct oxidative addition to
imidazolium C-H bonds.28
Scr een of P h osp h on iu m Sa lts in th e Cou p lin g of
Ben zth ia zole a n d 1-Hexen e a t 75 °C. Because we
believe that only Brφnsted acids accelerate the reaction,
a systematic survey of phosphonium salts was under-
taken that examined the effects of increasing the size and
branching of the phosphine. This screen was limited to
bulky trialkylphosphines, which an initial screen had
F IGURE 2. Proposed mechanism of reaction with HClPCy3
and [RhCl(coe)]2.
shown to be the most active ligands. The coupling of
benzthiazole and 1-hexene at 75 °C for 16 h was used as
the model reaction, utilizing GC to determine the yield.
The single time point yields were assumed to represent
a difference in rates among the reactions, though the
increase could potentially be attributed to either an
increase in the rate of coupling or a decrease in the rate
of isomerization of the alkene. Two independent runs
using HClPCy3 (Table 5, entries 1 and 2) showed only
minor variation in yield (Table 5). A plot of yield versus
the cone angle of the phosphines, which have been used
to evaluate steric effects of phosphorus ligands,29-30
(27) McGuinness, D. S.; Saendig, N.; Yates, B. F.; Cavell, K. J . J .
Am. Chem. Soc. 2001, 123, 4029-4040.
(28) Clement, N. D.; Cavell, K. J . Angew. Chem., Int. Ed. 2004, 43,
3845-3847.
J . Org. Chem, Vol. 69, No. 21, 2004 7333

Tan et al.
total (%)
TABLE 5. Scr een of P h osp h on iu m Sa lts w ith
Ben zth ia zole a n d 1-Hexen ea
TABLE 6. Scr een of P h osp h on iu m Sa lts w ith
Ben zim id a zole a n d 1-Hexen e
entry
PR3
HClPCy3
HClPCy3
HClPCy3
HClPCy2Me
HClPCy2Et
HClPCp3
HClP(i-Pr)3
HClPCy2-t-Bu
HClP(t-Bu)2Et
HClP(t-Bu)2Et
HClP(t-Bu)2(i-Pr)
HClP(t-Bu)3
HClPCy2Biphenyl
benzthiazole (%)
22 (%)
PR3
HClPCy3
HClPCy2Me
HClPCy2Et
HClP(t-Bu)2Me
HClPCp3
HClP(i-Pr)3
HClP(t-Bu)2Et
HClPCy2-t-Bu
HClP(t-Bu)2-i-Pr
HClP(t-Bu)3
benzimidazole (%)
9 (%)
1
2
57
58
54
103
71
74
52
49
23
31
64
82
96
36
33
38
2
24
18
42
43
75
69
23
3
33
78
71
73
58
53
42
73
71
72
81
40
73
78
74
79
69
69
59
76
71
72
81
3b
4
3
6
11
16
17
3
5
6
7
8
9
10b
11
12
13
HClPCy2Biphenyl
a
Reaction performed with 5 mol % [RhCl(coe)2]2 and 10 mol %
2
1
HClPR3 in THF at 105 °C for 16 h. Yields were determined by H
a
Reaction performed with 5 mol % [RhCl(coe)2]2 and 10 mol %
HClPR3 in THF at 75 °C for 16 h. Yields were determined by GC.
NMR spectroscopy
b
Reaction monitored by 1H NMR in THF-d8 and yield determined
benzimidazole to 1-hexene at 105 °C (Table 6). The
optimal phosphonium salt in this case was found to be
HClPCy3, and again there was a dependence on phos-
phine cone angle, such that use of either larger or smaller
phosphines led to a decrease in yield. It is worth noting
that over the course of these reactions material precipi-
tates from the reaction solution. The differential solubili-
ties of the components in the reaction mixtures of
benzimidazole and benzthiazole make it difficult to
compare the optimization results. The effect of increasing
the ratio of phosphonium salt to [RhCl(coe)2]2 was also
studied, and doubling the amount of phosphonium salt
had no significant impact on this reaction.
by 1H NMR.
In summary, the discovery of the profound effects of
additives on this type of C-H activation reaction has led
to an efficient intermolecular coupling of unactivated
alkenes to heterocycles. The reaction functions success-
fully with a variety of heterocycles and alkenes, including
both electron-deficient and electron-rich alkenes. Fur-
thermore, the optimized conditions are mild enough to
allow for the introduction of a large breadth of functional
groups on both the alkene and heterocycle. These im-
provements suggest a significant potential for drug
discovery and process chemistry. We are currently con-
ducting experiments designed to confirm the hypothesis
that the active catalyst in the reaction is a Rh(III)
hydride. Using the information gained from that work,
we hope to identify even more active and robust catalysts.
F IGURE 3. Yield of 22 vs cone angle.30
demonstrated that there is a significant dependence of
the phosphine size on the yield (Figure 3). The data
showed that the optimal phosphine size for the coupling
was HClP(t-Bu)2Et (Table 5, entry 9), and increasing or
decreasing the steric bulk of the ligand becomes detri-
mental to the rate of the reaction.
To verify that these single point yields reflected an
increase in the rate of the coupling, two reactions were
1
monitored by H NMR spectroscopy, in which phospho-
nium salts HClPCy3 and HClP(t-Bu)2Et were used. After
16 h, the yield of product corresponded to that observed
by GC (Table 5, entries 3 and 10). By monitoring the
reaction it was discovered that olefin isomerization occurs
rapidly in both cases and the increase in yield can indeed
be attributed to an increase in the rate of coupling of
1-hexene to benzthiazole.
Scr een of P h osp h on iu m Sa lts in th e Cou p lin g of
Ben zim id a zole a n d 1-Hexen e a t 105 °C. The encour-
aging results for coupling with benzthiazole prompted us
to conduct a phosphonium salt screen for the coupling of
Exp er im en ta l Section
Gen er a l Exp er im en ta l P r oced u r e for th e Hyd r oa r y-
la tion Rea ction . In air, to an oven-dried vessel with a Kontes
stopper was added [RhCl(coe)2]2 (5-10 mol % Rh), tricyclo-
hexylphosphonium chloride (5-10 mol %), and the heterocycle.
The vessel was placed under a nitrogen atmosphere. Under a
positive pressure of nitrogen, THF (diluted to 0.1 M in
heterocycle) and the alkene (5 equiv) were added to the vessel.
The vessel was closed with the Kontes stopper and was sub-
jected to three successive freeze-pump-thaw cycles. The ves-
sel was heated for 15 h at 150 °C. The reaction mixture was
dry loaded onto SiO2 and purified by flash chromatography.
Typ ica l P r oced u r e for Op tim iza tion Scr een s. In a
glovebox, a solution of benzthiazole (0.1 M), [RhCl(coe)2]2 (5
mol % Rh dimer, 10 mol % Rh), PCy3 (10 mol %), hexameth-
ylbenzene (internal standard), and 1-hexene (5 equiv relative
to benzthiazole) in THF was made. To glass vials were
distributed the additives (10 mol %) and the above solution
(0.5 mL). The vessels were closed and were heated for 16 h at
75 °C. The crude reaction mixtures were loaded onto a silica
(29) The cone angles were either calculated using the procedure
described by Tolman or were obtained from the following references:
(a) Tolman, C. A. Chem. Rev. 1977, 77, 313-348. (b) Rahman, M. M.;
Liu, H. Y.; Eriks, K.; Prock, A.; Giering, W. P. Organometallics 1989,
8, 1-7.
(30) The cone angles for PCp3 (Cp ) cyclopentyl) was not found, so
it was not plotted in Figure 3. Also PCy2(Biphenyl) is electronically
different from the other phosphines screened, and therefore it was not
plotted on this figure.
7334 J . Org. Chem., Vol. 69, No. 21, 2004

Coupling of Alkenes to Heterocycles via C-H Bond Activation
plugs, eluted with 40% EtOAc/hexanes, and analyzed by GC
or 1H NMR spectroscopy.
Department of Energy, under Contract No. DE-AC03-
76SF00098 (to R.G.B.).
Su p p or tin g In for m a tion Ava ila ble: Experimental de-
tails, including analytical data for all compounds described in
the article, as well as crystallographic data (CIF) for 6. This
material is available free of charge via the Internet at
http://pubs.acs.org.
Ack n ow led gm en t. This work was supported by a
fellowship (to K.L.T.) from the American Chemical
Society, by NIH Grant No. GM069559 (to J .A.E.), and
by the Director, Office of Energy Research, Office of
Basic Energy Sciences, Chemical Sciences Division, U.S.
J O048666P
J . Org. Chem, Vol. 69, No. 21, 2004 7335