Regioselective heck vinylation of electron-rich olefins with vinyl halides: Is the neutral pathway in operation?

Article abstract of DOI:10.1021/jo802781m

Highly regioselective vinylation of electron-rich olefins by bromo- as well as chlorostyrenes is effected by palladium catalysis with either mono- or bidentate phosphines in a molecular solvent, with no need for halide scavengers, ionic liquids, or ionic additives. The use of the hemilabile 1,3-bis(diphenylphosphino)propane monoxide (dpppO) as a ligand led to faster reactions of more challenging 2-substituted vinyl ethers and reduced Pd loadings. In contrast to the related arylation reaction, evidence suggests that the vinylation may proceed via the neutral Heck mechanism.

Full text of DOI:10.1021/jo802781m

Regioselective Heck Vinylation of Electron-Rich Olefins with Vinyl
Halides: Is the Neutral Pathway in Operation?
†
†
‡
,†
Matthew McConville, Ourida Saidi, John Blacker, and Jianliang Xiao*
LiVerpool Centre for Materials and Catalysis, Department of Chemistry, UniVersity of LiVerpool,
LiVerpool L69 7ZD, U.K., and NPIL Pharma UK, Ltd., Leeds Road, Huddersfield HD1 9GA, U.K.
j.xiao@liV.ac.uk
ReceiVed December 23, 2008
Highly regioselective vinylation of electron-rich olefins by bromo- as well as chlorostyrenes is effected
by palladium catalysis with either mono- or bidentate phosphines in a molecular solvent, with no need
for halide scavengers, ionic liquids, or ionic additives. The use of the hemilabile 1,3-bis(diphenylphos-
phino)propane monoxide (dpppO) as a ligand led to faster reactions of more challenging 2-substituted
vinyl ethers and reduced Pd loadings. In contrast to the related arylation reaction, evidence suggests that
the vinylation may proceed via the neutral Heck mechanism.
Introduction
p-RC
palladacycle and Fu’s Pd-P(t-Bu)
R/ꢀ selectivities of 10/13 and 4/1, respectively. It is widely
6
H
4
Br (R ) OMe, NMe ) catalyzed by Herrmann-Beller’s
2
6
7
3
system, which gave poor
The Heck reaction has become an indispensable tool for the
formation of C-C bonds, finding its place in many different
areas of synthetic organic chemistry. For the reaction concern-
1
(4) For examples of regioisomer formation, see: (a) Berthiol, F.; Doucet,
ing electron-deficient olefins (R′ ) EWG, Scheme 1), the
products usually result from aryl/vinylation at the terminal ꢀ
H.; Santelli, M. Appl. Organomet. Chem. 2006, 20, 855. (b) Kondolff, I.; Doucet,
H.; Santelli, M. Organometallics 2006, 25, 5219. (c) Berthiol, F.; Doucet, H.;
Santelli, M. Eur. J. Org. Chem. 2005, 1367. (d) Park, S. B.; Alper, H. Org. Lett.
2
position. However, under similar conditions with electron-rich
2
003, 5, 3209. (e) Calo, V.; Nacci, A.; Monopoli, A.; Spinelli, M Eur. J. Org.
olefins (R′ ) EDG, Scheme 1), regioselectivity can be poor,
and in fact, a mixture of products is often obtained.
This is exemplified by the arylation of butyl vinyl ether with
Chem. 2003, 1382. (f) Hierso, J. C.; Fihri, A.; Amardeil, R.; Meunier, P.; Doucet,
H.; Santelli, M.; Donnadieu, B. Organometallics 2003, 22, 4490. (g) Maeda,
K.; Farrington, G. E.; John, B. D.; Brown, J. M. AdV. Synth. Catal. 2002, 344,
104. (h) Schnyder, A.; Aemmer, T. A.; Indolese, F.; Pittelkow, U.; Studer, M.
AdV. Synth. Catal. 2002, 344, 495. (i) Feuerstein, M.; Doucet, H.; Santelli, M.
Tetrahedron Lett. 2002, 43, 2191. (j) Littke, A.; Fu, G. J. Am. Chem. Soc. 2001,
123, 6989.
2
e,g-i,3-5
†
University of Liverpool.
NPIL Pharma UK, Ltd.
‡
(5) (a) Cabri, W.; Candiani, I.; Bedeschi, A.; Santi, R. J. Org. Chem. 1993,
58, 7421. (b) Cabri, W.; Candiani, I.; Bedeschi, A.; Penco, S.; Santi, R. J. Org.
Chem. 1992, 57, 1481. (c) Cabri, W.; Candiani, I.; Bedeschi, A.; Santi, R. J.
Org. Chem. 1992, 57, 3558. (d) Cabri, W.; Candiani, I.; DeBernardinis, S.;
Francalanci, F.; Penco, S.; Santi, R. J. Org. Chem. 1991, 56, 5796. (e) Cabri,
W.; Candiani, I.; Bedeschi, B. J. Org. Chem. 1990, 55, 3654.
(
1) For applications of the Heck reaction, see: (a) Nicolaou, K. C.; Bulger,
P. G.; Sarlah, D. Angew. Chem., Int. Ed. 2005, 44, 4442. (b) Farina, V. AdV.
Synth. Catal. 2004, 346, 1553. (c) Blaser, H.-U.; Indolese, A.; Naud, F.;
Nettekoven, U.; Schnyder, A. AdV. Synth. Catal. 2004, 346, 1583. (d) Zapf, A.;
Beller, M. Top. Catal. 2002, 19, 101. (e) de Vries, J. G. Can. J. Chem. 2001,
¨
7
9, 1086. (f) Link, J. T.; Overman, L. E. In Metal-Catalyzed Cross-Coupling
(6) Herrmann, W. A.; Brossmer, C.; Reisinger, C. P.; Riermeier, T. H.; Ofele,
Reactions; Diederich, F., Stang, P. J., Eds.; Wiley-VCH: New York, 1998;
Chapter 6.
K.; Beller, M. Chem. sEur. J. 1997, 3, 1357.
(7) Littke, A. F.; Fu, G. C. J. Am. Chem. Soc. 2001, 123, 6989.
(
2) For reviews, see: (a) Knowles, J. P.; Whiting, A. Org. Biomol. Chem.
(8) (a) Arefalk, A.; Larhed, M.; Hallberg, A. J. Org. Chem. 2005, 70, 938.
(b) Vallin, K. S. A.; Zhang, Q.; Larhed, M.; Curran, D. P.; Hallberg, A. J. Org.
Chem. 2003, 68, 6639. (c) Nilsson, P.; Larhed, M.; Hallberg, A. J. Am. Chem.
Soc. 2003, 125, 3430. (d) Bengston, A.; Larhed, M.; Hallberg, A. J. Org. Chem.
2002, 67, 5854. (e) Olofsson, K.; Sahlin, H.; Larhed, M.; Hallberg, A. J. Org.
Chem. 2001, 66, 544. (f) Nilsson, P.; Larhed, M.; Hallberg, A. J. Am. Chem.
Soc. 2001, 123, 8217. (g) Vallin, K. S. A.; Larhed, M.; Johansson, K.; Hallberg,
A. J. Org. Chem. 2000, 65, 4537. (h) Olofsson, K.; Larhed, M.; Hallberg, A. J.
Org. Chem. 2000, 65, 7235. (i) Olofsson, K.; Larhed, M.; Hallberg, A. J. Org.
Chem. 1998, 63, 5076. (j) Larhed, M.; Hallberg, A. J. Org. Chem. 1997, 62,
7858. (k) Larhed, M.; Hallberg, A. J. Org. Chem. 1996, 61, 9582.
2
007, 5, 31. (b) Oestreich, M. Eur. J. Org. Chem. 2005, 783. (c) Bedford, R. B.;
Cazin, C. S. J.; Holder, D. Coord. Chem. ReV. 2004, 248, 2283. (d) Shibasaki,
M.; Vogl, E. M.; Ohshima, T. AdV. Synth. Catal. 2004, 346, 1533. (e) Larhed,
M.; Hallberg, A. In Handbook of Organopalladium Chemistry for Organic
Synthesis; Negish, E.-I., Ed.; Wiley-Interscience: New York, 2002; Vol. 1, p
1
133. (f) Whitcombe, N. J.; Hii, K. K.; Gibson, S. E. Tetrahedron 2001, 57,
449. (g) Beletskaya, I. P.; Cheprakov, A. V. Chem. ReV. 2000, 100, 3009. (h)
7
Hegedus, L. S. In Transition Metals in the Synthesis of Complex Organic
Molecules, 2nd ed.; University Science Books: Sausalito, CA, 1999; Chapter
4
.6. (i) Crisp, G. T. Chem. Soc. ReV. 1998, 27, 427.
3) (a) Cabri, W.; Candiani, I. Acc. Chem. Res. 1995, 28, 2. (b) Daves, G. D.,
Jr.; Hallberg, A. Chem. ReV. 1989, 89, 1433.
(
(9) (a) Ozawa, F.; Kubo, A.; Hayashi, T. J. Am. Chem. Soc. 1991, 113, 1417.
(b) Sato, Y.; Sodeoka, M.; Shibasaki, M. Chem. Lett. 1990, 1953.
2
692 J. Org. Chem. 2009, 74, 2692–2698
10.1021/jo802781m CCC: $40.75  2009 American Chemical Society
Published on Web 03/06/2009

RegioselectiVe Heck Vinylation of Electron-Rich Olefins
believed that this problem arises from competing reaction
pathways, which result in different regioisomers (Scheme
SCHEME 1. Heck Reaction and the Key Ionic and Neutral
Species, Where R ) Aryl/Vinyl, R′ ) EDG/EWG, and X )
Halide/Triflate
3
-5,7,8
1
).
Several methods have been successfully employed to over-
come this problem, bringing the selectivity up to a synthetically
useful level. This success is usually attributed to promoting the
1
0-12
“
ionic” pathway A, which leads to internal R substitution.
13
Taking the lead from Hallberg’s pioneering work, triflates,
tosylates, and mesylates have been utilized as labile counterions
in Heck reactions of electron-rich olefins. Unfortunately,
problems such as poor availability, high cost, and thermal lability
5
,8,14
limit their synthetic utility.
bidentate ligands such as 1,3-bis(diphenylphosphino)propane
dppp) could effect the aryl/vinylation of electron-rich olefins
It was shown by Cabri that
a
TABLE 1. Vinylation of BVE in Ionic Liquid vs DMSO
(
5
with almost exclusive R substitution. Ligands of this type are
now used as the standard for internal arylation reactions. In
addition to this, halide scavengers such as silver or thallium
salts drastically improve the regioselectivity for reactions of aryl/
vinyl bromides. However, the cost of silver and toxicity of
thallium prevent their use on larger scales, particularly as more
than stoichiometric quantities are often needed.
8
g,j
Recent work in this group has revealed that the use of
imidazolium-based ionic liquids in conjunction with Pd-dppp
catalysis offers excellent levels of regioselectivity with aryl
b
c
entry
solvent
additive
none
time (h) conv (%) selectivity
1
2
3
4
[bmim][BF
[bmim][BF
DMSO
4
]
]
20
20
0.5
0.5
40
33
61
66
>99:1
>99:1
>99:1
>99:1
1
5
4
[HNEt
none
3
][BF
4
]
]
bromides. It is assumed that the ionic environment provided
by the solvent makes generation of the cationic species more
facile (Scheme 1), and this is echoed by the recent kinetic studies
DMSO
[HNEt
3
][BF
4
12
a
of Amatore, Jutand, and co-workers. In addition, it was found
that hydrogen-bond-donating salts such as [HNEt ][BF ] could
Reaction conditions: p-methoxy-ꢀ-bromostyrene (1 mmol), BVE (3
(3 mol %), dppp (6 mol %), and NEt (3 mmol) in 2
mL of solvent at 115 °C; conversion measured by H NMR. 1.5 mmol
added where appropriate. Molar ratio of 1/2 >99:1 assigned when 2
was not detected by H NMR and GC.
mmol), Pd(OAc)
2
3
3
4
1
b
increase both the rate and selectivity in ionic as well as
molecular solvents, presumably by facilitating dissociation of
c
1
15d
bromide from Pd(II). Further development led to the discov-
ery that alcohol solvents such as ethylene glycol and i-PrOH,
with their large hydrogen-bond-donating ability, allow expedient
While generally treated as analogous to arylation, the viny-
lation of electron-rich olefins is relatively unexplored. Early
work by Hallberg et al. showed that although high selectivity
for branched products could be achieved in the reaction of vinyl
16,17
and highly selective arylation of electron-rich olefins.
Larhed
recently showed that water can also be an excellent solvent for
the arylation reaction in terms of rate and selectivity, although
this method seems to be limited to 2-hydroxy vinyl ether as the
3 4
iodides and triflates with 3 mol % of Pd(PPh ) in DMSO, the
1
8
13a
olefin.
selectivity was highly substrate-dependent. Thus, complete
reversal of the selectivity was observed when switching from
an electron-rich triflate to an electron-deficient bromide. The
realization by Cabri that bisphosphine ligands increase the
R-selectivity of arylation reactions was taken on board by those
studying the vinylations. These ligands have been routinely used
since then to obtain the required regioisomer.
using Pd-dppp as catalyst, Skrydstrup et al. addressed some
of the problems associated with the use of vinyl triflates by
using the less expensive mesylates and tosylates as vinylating
(10) For theoretical studies, see: (a) Surawatanawong, P.; Fan, Y.; Hall, M. B.
J. Organomet. Chem. 2008, 693, 1552. (b) Deeth, R. J.; Smith, A.; Brown, J. M.
J. Am. Chem. Soc. 2004, 126, 7144. (c) von Schenck, H.; Akermark, B.; Svensson,
M. J. Am. Chem. Soc. 2003, 125, 3503. (d) Hii, K. K.; Claridge, T. D. W.;
Brown, J. M.; Smith, A.; Deeth, R. J. HelV. Chim. Acta 2001, 84, 3043. (e)
Deeth, R. J.; Smith, A.; Hii, K. K.; Brown, J. M. Tetrahedron Lett. 1998, 39,
8a,g,14a
For instance,
3
229. (f) Albert, K.; Gisdakis, P.; Rosch, N. Organometallics 1998, 17, 1608.
11) A cationic pathway is expected to show preference for electron-rich
olefins: Hahn, C. Chem.sEur. J. 2004, 10, 5888.
12) (a) Amatore, C.; Godin, B.; Jutand, A.; Lamaitre, F. Organometallics
007, 26, 1757. (b) Amatore, C.; Godin, B.; Jutand, A.; Lemaitre, F. Chem.sEur.
J. 2007, 13, 2002.
13) (a) Andersson, C. M.; Hallberg, A. J. Org. Chem. 1989, 54, 1502. (b)
Andersson, C. M.; Hallberg, A. J. Org. Chem. 1989, 53, 2112.
14) (a) Hansen, A. L.; Skrydstrup, T. Org. Lett. 2005, 7, 5585. (b) Hansen,
(
(
2
1
4a
agents. Excellent regioselectivity and good to excellent yields
were obtained for the coupling of enamides and electron-
deficient mesylates and tosylates. However, with vinyl bromides,
halide scavengers such as TlOAc appear still necessary to
(
(
A. L.; Skrydstrup, T. J. Org. Chem. 2005, 70, 5997. (c) Tu, T.; Hou, X.-L.; Dai,
L.-X. Org. Lett. 2003, 5, 3651. (d) Barcia, J. C.; Cruces, J.; Est e´ vez, J. C.;
Est e´ vez, R. J.; Castedo, L. Tetrahedron Lett. 2002, 43, 5141. (e) Maik, L.;
Stroemberg, S.; Svensson, M.; Åkermark, B. Organometallics 1999, 18, 970.
8g
achieve high regioselectivity. Thus, there is a need to improve
the protocols most commonly employed for the vinylation of
electron-rich olefins.
(
15) (a) Mo, J.; Xu, L. J.; Ruan, J. W.; Liu, S. F.; Xiao, J. L. Chem. Commun.
2
006, 3591. (b) Liu, S. F.; Berry, N.; Thomson, N.; Pettman, A.; Hyder, Z.;
Mo, J.; Xiao, J. L. J. Org. Chem. 2006, 71, 7467. (c) Hyder, Z.; Mo, J.; Xiao,
J. L. AdV. Synth. Catal. 2006, 348, 1699. (d) Mo, J.; Xiao, J. L. Angew. Chem.,
Int. Ed. 2006, 45, 4152. (e) Mo, J.; Xu, L. J.; Xiao, J. L. J. Am. Chem. Soc.
Results and Discussion
2
005, 127, 751. (f) Xu, L. J.; Chen, W. P.; Ross, J.; Xiao, J. L. Org. Lett. 2001,
, 295. (g) Liu, S. F.; Xiao, J. L. J. Mol. Catal. A: Chem. 2007, 270, 1.
As a starting point for our study on the vinylation of electron-
rich olefins, we applied the arylation conditions developed within
this group to the reaction of p-methoxy-ꢀ-bromostyrene with
n-butyl vinyl ether (BVE). Table 1 shows the results obtained
with 1-butyl-3-methylimidazolium tetrafluoroborate ([bmim]-
3
(
(
16) Hyder, Z.; Ruan, J.; Xiao, J. L. Chem.sEur. J. 2008, 14, 5555.
17) Vallin, K. S. A.; Larhed, M.; Hallberg, A. J. Org. Chem. 2001, 66,
4
340. This paper also shows that the regioselective Heck reaction is feasible
15
using a DMF/MeOH mixture.
18) Arvela, R. K.; Pasquini, S.; Larhed, M. J. Org. Chem. 2007, 72, 6390.
(
J. Org. Chem. Vol. 74, No. 7, 2009 2693

McConville et al.
a
TABLE 2. Regioselective Heck Vinylation of Electron-Rich Olefins with Pd-dppp in DMSO
a
Reaction conditions: vinyl halide (1 mmol), vinyl ether (3 mmol), Pd(OAc)
2
(3 mol %), dppp (6 mol %), NEt
3
(3 mmol) in 2 mL of DMSO at 115
1
b
°
C. Following the coupling reaction, HCl was added at rt. Isolated yields given. No ꢀ-vinylation product was detected by H NMR. As in footnote a
but without hydrolysis. Conditions as in footnote a, but Pd(OAc)
c
d
1
2
(5 mol %) and dppp (10 mol %) were used. Conversion measured by H NMR.
1
5
[
BF
4
]) as solvent. In contrast to the analogous arylation, the
p-acetyl-ꢀ-bromostyrene (entry 7), the product of which would
be difficult to achieve via a conventional aldol methodology.
In addition to BVE, 2-hydroxyethyl vinyl ether allowed access
to cyclic ketals under identical conditions (entries 8-10). The
formation of a protected ketone in this way could be potentially
useful when chemoselectivity would otherwise be troublesome.
Unfortunately, ketals bearing methoxy groups on the aromatic
ring were somewhat unstable in solution, and traces of the
Pd-dppp-catalyzed Vinylation was sluggish in the ionic liquid,
and addition of hydrogen-bond-donating salts did not have an
accelerating effect on the rate (entries 1 and 2).
However, the reaction proceeded very well in DMSO,
affording over 60% conversion in just 0.5 h (entry 3). And
significantly, the reaction exhibited remarkable selectivity for
the products resulting from R substitution, as no linear product
could be detected by H NMR or GC. Again, the addition of
hydrogen-bond-donating salts had only a negligible effect on
the rate (entry 4). A limited screening of conditions revealed
1
1
corresponding ketone were detectable by H NMR after several
hours. For convenience, the 2-methoxy compound was therefore
hydrolyzed and isolated as the ketone (entry 11).
3
that DMSO as solvent and NEt as base gave the best results.
Remarkably, vinyl chlorides also reacted, affording exclu-
sively branched products, albeit over longer reaction times.
Again, these products were hydrolyzed and isolated as the
ketones in good yields (entries 12-15). Very few examples of
These results show that the R-vinylation of electron-rich
olefins catalyzed by Pd-dppp can work in a common, molecular
solvent with no need for any additives. By applying the protocol
used for Table 1, it was found that the reaction of ꢀ-bromosty-
rene was finished in 4 h. The only side product of the reaction
was trace amounts of the dimer resulting from homocoupling
of the starting bromide. As can be seen in Table 2 (entry 1),
hydrolysis of 1 and subsequent purification led to the desired
ketone 3 in an excellent 96% yield.
1
9
Heck reactions of vinyl chlorides exist in the literature. To
our knowledge, these are the first examples with an electron-
rich olefin.
Another reaction that was interesting to us was that of
2
-substituted vinyl ether with a vinyl halide, as this would lead
With the conditions for the vinylation in hand, we next
investigated the scope of the reaction with respect to vinyl
halides and electron-rich olefins. The results are summarized
in Table 2. In general, reactions featuring electron-rich bromides
such as p-methoxy-ꢀ-bromostyrene were complete in slightly
shorter times. Worthy of particular mention is the reaction of
to products other than methyl ketones. However, reactions with
(19) Examples of Heck coupling with vinyl chlorides: (a) Littke, A. F.; Fu,
G. C. J. Am. Chem. Soc. 2001, 123, 6995. (b) Voigt, K.; Schick, U.; Meyer,
F. E.; de Meijere, A. Synlett 1994, 189. (c) Horino, H.; Inoue, N.; Asao, T.
Tetrahedron Lett. 1981, 22, 741.
2
694 J. Org. Chem. Vol. 74, No. 7, 2009

RegioselectiVe Heck Vinylation of Electron-Rich Olefins
a
TABLE 3. Ligand Effects on the Heck Vinylation of BVE
TABLE 4. Regioselective Heck Vinylation of 2-Substituted Vinyl
a
Ethers
entry
ligand
conv (%)
selectivity 3:4
1
2
3
4
5
6
7
8
9
PPh
P(4-OMePh)
P(2-OMePh)
P[3,5-(CF
PCy
dppm
dppe
dppp
dppb
(2-CF
3
80
86
86
94
55
37
36
61
90
86
95
80:20
90:10
90:10
>99:1
>99:1
>99:1
>99:1
>99:1
95:5
3
3
3 2 3
) Ph]
3
1
1
0
1
3
Ph)
2
P(CH
2
)
3
P(2-CF
3
Ph)
2
>99:1
>99:1
dpppO
a
Reaction conditions: p-methoxy-ꢀ-bromostyrene (1 mmol), BVE (3
(3 mol %), ligand (6 mol % of bisphosphine or 9 mol
of monophosphine), NEt (3 mmol) in 2 mL of solvent at 115 °C for
0 min; no 2 was detected.
mmol), Pd(OAc)
%
3
2
3
20
these olefins, which are known to be less reactive, were much
slower than with BVE. Thus, even with the catalyst loading
increased to 5 mol %, 1-propenyl ether took 16 h to complete
(entry 16); with 1-butenyl ethyl ether, complete conversion could
not be reached, even with long reaction times (entry 17).
Because the reactions of substituted vinyl ethers were
sluggish, it was decided to undertake a ligand screening. The
results are shown in Table 3. Much to our surprise, the
regioselectivity of the reaction was unaffected by the choice of
ligand; be it monodentate or bidentate, the linear product was
never detected, and only varying amounts of the homocoupling
products were produced. Remarkably, the monophosphines
significantly out-performed dppp, a ligand which has been
universally used in regiocontrol of the Heck reaction of electron-
rich olefins. Of the ligands tested, the bis(trifluoromethyl)-
a
Reaction conditions: vinyl halide (1 mmol), vinyl ether (3 mmol),
Pd(OAc)2 (5 mol %), dppp (10 mol %), NEt3 (3 mmol) in 2 mL of
DMSO at 115 °C. Isolated yields given. No type 2 or homocoupling
products were detected in the crude H NMR.
substituted PPh
3
(entry 4) and the hemilabile dpppO (entry 11)
1
gave the best results, both affording ∼95% conversion and
2
1
complete R selectivity for the desired product.
Encouraged by this success, we decided to continue the study
of substituted vinyl ethers with dpppO as ligand, due to its lower
cost and ease of availability compared to the monophosphine
homogeneous catalysis greener and more economic. Hence,
when we observed the higher activity of Pd-dpppO in the
reaction of 2-substituted vinyl ethers, we decided to investigate
if a lower catalyst loading in the reactions of other vinyl ethers
would be feasible. Table 5 shows the results obtained with BVE
and 2-hydroxyethyl vinyl ether. To our delight, the reactions at
1 mol % of Pd (entries 1-8) were complete in around 1 h and
those with 0.1 mol % of Pd in 18 h (entries 9-11). For these
reactions, the ligand to metal ratio was increased from 2 to 4;
this appeared to be necessary for reproducibility at the reduced
palladium loading. However, a lower yield was observed when
p-methoxy-ꢀ-bromostyrene was coupled with BVE at the 0.1
mol % Pd loading (not shown). It was noticed previously that
if the reaction mixture was allowed to stir overnight, the yield
dropped from 82% to ca. 35% under the conditions of Table 2
(entry 3). This indicates that the initially formed diene from
BVE and the methoxy-substituted ꢀ-bromostyrene is unstable
under the chosen conditions, explaining the lower yield observed
with BVE at the lower Pd loading where a longer reaction time
was necessary. The ketal is seemingly much more stable under
these conditions, with excellent yields being obtained for all
substrates (Table 5, entries 9-11). These results compare
2
2,23
in entry 4.
In sharp contrast to dppp, dpppO allowed
complete reactions of p-methoxy-ꢀ-bromostyrene with 1-pro-
penyl and 1-butenyl ethyl ether in 0.75 and 4 h, respectively.
The results from expanding the scope of this reaction are shown
in Table 4. Good to excellent yields were obtained for a range
of vinyl bromides in reaction times much shorter than those
possible with the common dppp.
Lowering the quantity of catalysts, which are often based on
expensive transition metals, is one of the main goals in making
(
20) Battace, A.; Zair, T.; Doucet, H.; Santelli, M. Tetrahedron Lett. 2006,
7, 459.
21) In the reaction where dppp was employed, ca. 10% of the ligand was
oxidized to dppp oxide at 60% conversion, with the rest of the ligand coordinating
4
(
31
to palladium according to P NMR (a larger scale reaction was run to facilitate
data collection). This suggests that the coupling reactions in Table 2 were
catalyzed by Pd-dppp rather than Pd-dpppO species.
(
22) (a) DpppO synthesis: Grushin, V. V. J. Am. Chem. Soc. 1999, 121, 5831.
Also see: (b) Amatore, C.; Jutand, A.; Thuilliez, A. Organometallics 2001, 20,
241.
3
(
23) Review on phosphine oxides: Grushin, V. V. Chem. ReV. 2004, 104,
1
629.
(
24) Review on the synthesis of R,ꢀ-unsaturated ketones: Foster, C. E.;
Mackie, P. R. In ComprehensiVe Organic Functional Group Transformations
II; Katritzky, A. R., Taylor, R. J. K., Eds.; Elsevier Ltd.: Oxford, UK, 2005;
Vol. 3, p 215.
(25) Shibuya, M.; Ito, S.; Takahashi, M.; Iwabuchi, Y. Org. Lett. 2004, 6,
4303.
J. Org. Chem. Vol. 74, No. 7, 2009 2695

McConville et al.
TABLE 5. Regioselective Heck Vinylation with Vinyl Bromides
TABLE 6. Effect of Halide Ions on the Vinylation of BVE with
a
a
Using Pd-dpppO at Reduced Catalyst Loading
p-Methoxy-ꢀ-bromostyrene
b
c
d
e
d
entry additive (%) conv (%) selectivity conv (%) selectivity
1
2
3
4
none
10
50
61
58
44
17
>99:1
>99:1
>99:1
>99:1
61
32
12
<5
>99:1
>99:1
>99:1
99:1
200
a
Reaction conditions: p-methoxy-ꢀ-bromostyrene (1 mmol), BVE (3
(3 mol %), dppp (6 mol %), and NEt (3 mmol) in 2
mmol), Pd(OAc)
mL of DMSO at 115 °C for 30 min. NBu
Molar ratio of 1:2. NEt Cl.
4
2
3
b
c
4 4 4
Br or NEt Cl. NBu Br.
d
e
FIGURE 1. Reaction profiles of cis/trans-methoxy-ꢀ-bromostyrene
reacting with BVE in DMSO: 9, trans; 2, cis. Reaction conditions: 1
mmol of bromide, 3 mmol of BVE, 3 mol % of Pd(OAc)
dppp, and 3 mmol of NEt in DMSO at 115 °C.
2
, 6 mol % of
3
B would become more likely under these conditions. Second,
the addition of halide ions to reactions in which the pathway A
5
c,15e,f
is thought to dominate inhibits the catalysis.
This is
particularly evident from arylation work done in this group,
where addition of as little as 10 mol % of bromide halted the
1
5c
arylation of butyl vinyl ether. However, Table 6 shows that
the vinylations in question are relatively insensitive to additional
halide. Most notably, the reaction still proceeds in the presence
4
of 200 mol % of NBu Br, albeit at a reduced rate. Addition of
chloride ions retarded the reaction to a greater extent, although
large quantities were required to completely inhibit the catalysis.
Third, the use of halide scavengers or ionic media enhances
the selectivity and/or the rate of reactions where A is presumed
a
Vinyl bromide (1 mmol), olefin (3 mmol), Pd(OAc)
2
(1 mol %),
dpppO (4 mol %), and NEt
As footnote a but with Pd(OAc)
and hydrolysis.
3
(3 mmol) in 2 mL of DMSO at 115 °C.
b
2
(0.1 mol %) and dpppO (0.4 mol %)
3a,5e,12,15c,e
to be the dominant pathway.
We found that, contrary
to arylation reactions, the vinylation was very slow in ionic
liquids, and a negligible effect on the rate was observed in either
favorably with those shown in Table 2 and leave us with a highly
selective and high-yielding protocol for the synthesis of (E)-
R,ꢀ-unsaturated ketones as an alternative to other methods
[
4 3 4
bmim][BF ] or DMSO on addition of [HNEt ][BF ]. Finally,
in a direct competition reaction between an electron-rich and
an electron-deficient olefin, such as the one shown in Scheme
2, an electron-rich olefin would be more reactive when the
2
4-28
reported in the literature.
Although we were pleased with the results thus far, they raised
an important question: Is the ionic pathway A or the neutral
pathway B responsible for the observed products? It is generally
accepted that the R arylation of electron-rich olefins proceeds
via a cationic mechanism (A, Scheme 1) and the related
Vinylation is judged under the same criteria, four of which exist
in the literature. First, bidentate ligands are required to obtain
5
a,15e
pathway A is believed to prevail.
We found this was not
the case when a reaction between a vinyl bromide and equimolar
quantities of BVE and methyl acrylate was undertaken; regard-
less of the ligand chosen, ꢀ-vinylation of methyl acrylate was
the dominant reaction (Scheme 2).
All of the results presented thus far were obtained with (E)-
ꢀ-bromostyrenes. Further insight into the mechanism was gained
by observations made with (Z)-ꢀ-bromostyrenes. The com-
mercially available ꢀ-bromostyrene contains approximately 13%
cis isomer, but interestingly, our products contained >99% trans
olefins in all of the previously mentioned reactions. This
prompted us to investigate the reactions of the cis isomer. Using
the same conditions as those shown in Table 1 (entry 3), (Z)-
p-methoxy-ꢀ-bromostyrene was reacted with BVE. Even with
a starting material with an Z/E ratio of >19:1, the E isomer was
5
branched products. However, as revealed in the ligand screen-
ing, not only were bidentate ligands unnecessary for regiocon-
trol, the monodentate or hemilabile phosphines actually gave
significantly faster rates. This is surprising as the neutral pathway
(
(
71.
26) Kim, J. H.; Lim, H. J.; Cheon, S. H. Tetrahedron 2003, 59, 7501.
27) Yu, M.; Li, G.; Wang, S.; Zhanga, L. AdV. Synth. Catal. 2007, 349,
8
(
28) Yadav, J. S.; Subba Reddy, B. V.; Gupta, M. K.; Dash, U.; Pandey, S.
K Synlett 2007, 809.
2
696 J. Org. Chem. Vol. 74, No. 7, 2009

RegioselectiVe Heck Vinylation of Electron-Rich Olefins
2
9
SCHEME 2. Competition Reaction between an
Electron-Rich and -Deficient Olefin
vinyl oxidative-addition product. The usual process of inser-
tion, ꢀ-hydride elimination, and reductive elimination of HX
then follows, regenerating the active Pd(0) catalyst and complet-
ing the catalytic cycle.
DFT calculations by Deeth and Brown have shown that for
a vinyl group migrating to a vinyl ether, R-substitution is
1
0b
preferred even for a neutral Pd species. This is in agreement
with our experimental findings. We note, however, the work of
Amatore and Jutand, who showed that cationic [Pd(dppp)Ar-
+
(
[
solvent)] is more reactive toward olefins than the analogous
12b
Pd(dppp)(Ar)X]. They have also recently demonstrated that,
SCHEME 3. Proposed Neutral Pathway for the Heck
Vinylation of Electron-Rich Olefins
for arylation in DMF, the insertion step always proceeds in an
+
12a
ionic fashion, i.e., from [ArPd(dppp)(DMF)] . Whether or
not this is the case for the vinylation in DMSO remains to be
investigated.
Conclusion
In summary, a highly efficient protocol for the Pd-catalyzed
regioselective Heck vinylation of electron-rich olefins has been
developed. The use of monodentate ligands and dpppO is the
first example of these types of ligands being employed for
regiocontrol in the Heck reaction. In particular, using dpppO
as ligand enables reaction of more challenging olefins in short
times, with lower Pd loadings being feasible. Together with the
results concerning ionic liquid, halide inhibition, and the
competition reaction, our findings indicate that the Heck R
Vinylation and R arylation may proceed via different pathways,
with the former more likely to adopt the neutral pathway B
obtained exclusively according to NMR. This selectivity for the
(
Scheme 1).
trans product may be attributed to a Pd-catalyzed isomerization
2
9
during/after oxidative addition (vide infra). The starting
material does not appear to be isomerized before entering the
catalytic cycle, as no change in the cis/trans ratio of the free
vinyl halides was observed when they were subjected to the
coupling reaction conditions in the absence of olefin.
Experimental Section
General Procedure for the Heck Vinylation of Terminal Vinyl
Ethers with a Pd-dppp Catalyst. An oven-dried, two-necked,
round-bottom flask containing a stir bar was charged with Pd(OAc)
2
The kinetic profiles of the two starting isomers reacting with
BVE are shown in Figure 1, revealing an insignificant difference
between them in rates. Amatore and Jutand showed that cis-
vinyl bromides are significantly slower at oxidative addition than
(0.03 mmol, 7 mg), dppp (0.06 mmol, 25 mg), ꢀ-bromostyrene (1
mmol, 183 mg, 0.12 mL), and 2 mL of DMSO. Following degassing
three times, BVE (3 mmol, 301 mg, 0.4 mL) and NEt
3
(3 mmol,
3
03 mg, 0.4 mL) were injected sequentially. The flask was placed
30
in a parallel reactor at 115 °C and stirred for an appropriate time.
After the flask was cooled to room temperature, a small sample
was taken for NMR analysis. Ten milliliters of 10% HCl was then
added (except when 2-hydroxyethyl vinyl ether was used) and the
solution extracted with 3 × 15 mL of DCM. The combined organic
their trans counterparts. Hence, the similarity in rates appears
to rule out oxidative addition as rate determining in the Heck
vinylation. As with arylation reactions, the vinylation under
question might be rate-limited by the subsequent migratory
1
2b,15e
insertion.
2 4
layers were washed to neutrality with H O, dried over MgSO ,
The combination of all the pieces of evidence presented above
leads us to believe that the regioselective internal vinylation of
electron-rich olefins is more likely to proceed via the neutral
pathway B. Hence, a modified mechanism for the reaction is
proposed and shown in Scheme 3. Oxidative addition of the
filtered, and concentrated in vacuo. The R,ꢀ-unsaturated methyl
ketone product (or cyclic ketal) was isolated from the crude mixture
by chromatography on silica gel using ethyl acetate and hexane
(
1/99 to 10/90) as eluant. (E)-4-Phenylbut-3-en-2-one as an orange
1
crystalline solid (141 mg, 96%): mp 38-41 °C; H NMR (400
MHz, CDCl ) δ 7.53-7.56 (m, 3H), 7.50 (d, J ) 16.0 Hz, 1H),
.39-7.41 (m, 2H), 6.72 (d, J ) 16.0 Hz, 1H), 2.39 (s, 3H);
NMR (100 MHz, CDCl ) δ 198.8, 143.9, 134.8, 131.0, 129.4, 128.7,
2
3
vinyl halide to a Pd(0) leads to an η vinyl species. It is proposed
1
3
7
C
that this proceeds via dissociation of a neutral ligand and η2
3
coordination of the vinyl ligand. The incoming olefin then
+
+
1
5
27.6, 28.0; CI-MS m/z 147 [(M + H) , 100], 164 [(M + NH
4
) ,
2
1
displaces the η vinyl group, rendering it η coordinating.
Regardless of the geometry of the starting vinyl halide, the
change in the coordination mode is expected to lead to a more
stable trans-palladated olefin and, thus, the trans selectivity
observed above. Alternatively, the isomerization could take place
11O (M + H)+ 147.0810, found
1]; HRMS calcd for C10
H
-
1
147.0813; IR ν (cm ) 1667, 1598, 1494. For details of other
products, see Table 2 and the Supporting Information.
General Procedure for the Heck Vinylation of 2-Substituted
Vinyl Ethers Using a Pd-dpppO Catalyst. An oven-dried, two-
necked, round-bottom flask containing a stir bar was charged with
1
through a carbene species arising from the initial, η -coordinated
Pd(OAc)
-bromostyrene (1 mmol, 183 mg, 0.1 mL), and 2 mL of DMSO.
Following degassing three times, NEt (3 mmol, 303 mg, 0.4 mL)
2
(0.03 mmol, 7 mg), dpppO (0.1 mmol, 52 mg),
ꢀ
(29) For an example of vinyl halide isomerization relevant to our mechanistic
3
proposal, see: Amatore, C.; Bensalem, S.; Ghalem, S.; Jutand, A. J. Organomet.
Chem. 2004, 689, 4642.
was injected, and the flask was placed in a parallel reactor at 115
°C. After 3-4 min, 1-propenyl ethyl ether (3 mmol, 258 mg, 0.3
(
30) Jutand, A.; Negri, S. Organometallics 2003, 22, 4229.
J. Org. Chem. Vol. 74, No. 7, 2009 2697

McConville et al.
) , 69]; HRMS calcd for C11
13O (M + H)+
+
mL) was injected, and the mixture was stirred for an appropriate
time. After being cooled to room temperature, a small sample was
taken for NMR analysis. Ten milliliters of 10% HCl was then added
and the solution extracted with 3 × 15 mL of DCM. The combined
178 [(M + NH
4
H
-
1
161.0966, found 161.0968; IR ν (cm ) 1698, 1600, 1492.
Acknowledgment. We are grateful to NPIL Pharma UK and
the EPSRC for financial support in the form of an industrial
case award (M.M.).
2
organic layers were washed to neutrality with H O, dried over
MgSO , filtered, and concentrated in vacuo. The R,ꢀ-unsaturated
4
methyl ketone product was isolated from the crude mixture by
chromatography on silica gel using ethyl acetate and hexane (1/99
Supporting Information Available: Typical experimental
to 5/95) as eluant. (E)-1-Phenylpent-1-en-3-one as a light orange
1
1
13
oil (147 mg, 91%): H NMR (400 MHz, CDCl
1
3
) δ 7.56 (d, J )
procedures, characterization data, and H and C NMR spectra.
This material is available free of charge via the Internet at
http://pubs.acs.org.
6.1 Hz, 1H), 7.50-7.54 (m, 2H), 7.37-7.39 (m, 3H), 6.74 (d, J
)
16.1 Hz, 1H), 2.69 (q, J ) 7.3 Hz, 2H), 1.17 (t, J ) 7.3 Hz,
) δ 201.3, 142.6, 135.1, 130.7,
3
1
H); 13C NMR (100 MHz, CDCl
3
+
29.3, 128.6, 126.5, 34.4, 8.6; CI-MS m/z 161 [(M + H) , 100],
JO802781M
2
698 J. Org. Chem. Vol. 74, No. 7, 2009