Single-Flask Multicomponent Palladium-Catalyzed α,γ-Coupling of Ketone Enolates: Facile Preparation of Complex Carbon Scaffolds

Article abstract of DOI:10.1002/anie.201505895

A three-component palladium-catalyzed reaction sequence has been developed in which γ-substituted α,β-unsaturated products are obtained in a single flask by an α-alkenylation with either a subsequent γ-alkenylation or γ-arylation of a ketone enolate. Coup

Full text of DOI:10.1002/anie.201505895

.
Angewandte
Communications
International Edition: DOI: 10.1002/anie.201505895
German Edition: DOI: 10.1002/ange.201505895
Cross-Coupling
Single-Flask Multicomponent Palladium-Catalyzed a,g-Coupling of
Ketone Enolates: Facile Preparation of Complex Carbon Scaffolds
Michael Grigalunas, Per-Ola Norrby, Olaf Wiest, and Paul Helquist*
Abstract: A three-component palladium-catalyzed reac-
tion sequence has been developed in which g-substituted
a,b-unsaturated products are obtained in a single flask
by an a-alkenylation with either a subsequent g-
alkenylation or g-arylation of a ketone enolate. Cou-
pling of a variety of electronically and structurally
different components was achieved in the presence of
a Pd/Q-Phos catalyst (2 mol%), usually at 228C with
yields of up to 85%. Most importantly, access to these
products is obtained in one simple operation in place of
employing multiple reactions.
S
equential multicomponent reactions have long been
recognized as an efficient approach to the synthesis of
complex molecules, especially when they can be con-
ducted as single-flask (one-pot) operations. The first
reaction of a multicomponent sequence serves to
generate a substrate that enters into the next step or
steps of the sequence. Among the implementations of
this concept are the sequential addition of reactants to
a single flask or tandem reactions in which all
components may be present simultaneously.[1] Overall
transformations are accomplished more efficiently than
would be the case through use of individual reactions.
In recent years, this concept has been developed
Figure 1. a) Palladium-catalyzed a-alkenylation of enolates. b, c) Palladium-
catalyzed g-arylation and g-alkenylation of dienolates. d) Palladium-catalyzed
sequential a,g-coupling. dba=dibenzylideneacetone, TES=triethylsilyl,
THF=tetrahydrofuran, Q-Phos=1-di-tert-butylphosphino-1’,2’,3’,4’,5’-penta-
phenylferrocene.
extensively through use of metal-promoted reaction sequen-
ces,[2] with carbon–carbon bond-forming reactions playing
especially important roles. Potentially amenable to multi-
component sequences are transition-metal-catalyzed a-alke-
nylation of enolates[3] (Figure 1a) and the related g-arylation
and g-alkenylation of dienolates generated from either a,b- or
b,g-unsaturated carbonyl compounds (Figure 1b and c).[4]
Individually, these reactions provide access to b,g-unsaturated
and g-substituted, a,b-unsaturated carbonyl compounds,
respectively. Facile access to both of these classes of scaffolds
is desirable because of their appearance in biologically
important compounds and their versatility in further synthetic
transformations.[3a,5] Greater synthetic efficiency could be
achieved if the a- and g-coupling reactions were employed in
single-flask sequences (Figure 1d).
Previous studies of palladium-catalyzed coupling reac-
tions of dienolates have demonstrated ligand-dependent
selectivity for achieving g-coupling as opposed to either a-
or b-coupling. Work by Miura and co-workers led to
a procedure for coupling a,b-unsaturated aldehydes and
ketones with aryl bromides at 1208C to afford g-arylated
tertiary products.[6] Later, Hyde and Buchwald employed
both a,b- and b,g-unsaturated ketones to form g-arylated and
g-alkenylated quaternary products at 1008C (Figure 1b).[7]
More recently, Huang and Hartwig reported a procedure
using milder reaction conditions for the g-arylation and g-
alkenylation of silyl ketene acetals at 228C (Figure 1c).[8]
While good yields have been achieved utilizing these proce-
dures, there is ample room for improvement because of
drawbacks which include high temperatures or the need to
prepare, in separate steps, the required unsaturated carbonyl
or their ketene acetal derivatives. Also, the scope of g-
alkenylation has been much more limited than for g-
arylation.[5d,7–9]
[*] M. Grigalunas, Prof. Dr. P.-O. Norrby, Prof. Dr. O. Wiest,
Prof. Dr. P. Helquist
Department of Chemistry and Biochemistry
University of Notre Dame, Notre Dame, IN 46556 (USA)
E-mail: phelquis@nd.edu
Prof. Dr. P.-O. Norrby
Pharmaceutical Development, Global Medicines Development
AstraZeneca, 43183 Mçlndal (Sweden)
Prof. Dr. O. Wiest
Lab of Computational Chemistry and Drug Design
School of Chemical Biology and Biotechnology, Peking University
Shenzhen Graduate School, Shenzhen 518055 (China)
To overcome the need for the separate synthesis of the
dienolate precursor, we envisioned a palladium-catalyzed a-
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201505895.
11822
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2015, 54, 11822 –11825

Angewandte
Chemie
alkenylation and subsequent g-arylation or g-alkenylation
sequence that would generate a g-substituted a,b-unsaturated
scaffold more directly in a single operation (Figure 1d). This
sequence would employ a ketone, an alkenyl halide to
generate a bridging alkene unit, and either an aryl halide or
an alkenyl halide as a capping group. Herein, we report a mild
palladium-catalyzed procedure to effect this type of sequen-
tial a,g-coupling in one flask with excellent regioselectivity
and control of the sequence of incorporation of components
for the rapid construction of complex carbon scaffolds, thus
providing facile access to molecular diversity.
of the second electrophile (capping group) needed because of
the presence of unreacted starting ketone enolate. We
therefore focused on developing reaction conditions that
utilize a 1:1 ratio of 1a and 2a to afford high conversion into
the initial coupling product 3aa with minimal side product
formation (Table 2). The reaction parameters examined
Table 2: Optimization of the reaction conditions for the formation of
3aa.
Initial success was achieved upon the isolation of an a,g-
alkenylated product (3aaa) when acetophenone (1a) and
a small excess of 2a were subjected to our recently published
a-alkenylation conditions (Table 1, entry 1).[3b] To optimize
Entry[a]
X[b]
Y[b]
Z[b]
Solvent (m)
T [8C]
Yield [%][c]
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
2
1
1
1
2
2
2
2
2
2
4
2
2
2
220
220
220
220
220
220
220
280
340
400
toluene (0.25)
THF (0.25)
THF (0.50)
THF (0.13)
THF (0.06)
THF (0.25)
THF (0.25)
THF (0.25)
THF (0.25)
THF (0.25)
22
22
22
22
22
0
16
79
66
83
89
29
72
89
95
96
Table 1: Optimization of the reaction conditions for the formation of
3aaa.[a]
0
22
22
22
10
[a] Reactions were conducted on a 0.25 mmol scale. [b] Reported in
mol%. [c] Determined by NMR spectroscopy using an internal standard.
Entry[b]
2a[c]
Base[c]
T [8C]
Yield [%][d]
t [h]
1
2
3
130
250
250
LiHMDS (110)
LiHMDS (250)
LiOtBu (250)
0
22
22
35
0
99
2
12
12
included the choice of solvent, molarity, temperature, and
base loading. The use of a large excess of LiOtBu was the
most beneficial reaction component examined, thus affording
96% yield of 3aa (entry 10). Optimized reaction conditions
employed [Pd2(dba)3] (1 mol%), Q-Phos (2 mol%), and
LiOtBu (400 mol%) in THF (0.25m) at 228C, and they
were employed for the subsequent study of the desired
sequential a,g-couplings.
[a] Here and in subsequent figures, the multiple letter designations in the
compound numbering refer to the ketone, bridging, and capping groups
in that order. [b] Reactions were conducted on a 0.5 mmol scale.
[c] Reported in mol%. [d] Yield of isolated products. HMDS=hexame-
thyldisilazide.
For the sequential couplings, an alkenyl bromide
(100 mol%) was added as a bridging alkene to a solution of
the starting ketone enolate, base, and catalyst in THFat 228C.
After 45 minutes, either an alkenyl bromide or an aryl halide
(110 mol%) was added as a capping group and reacted for
16 hours at either 228C or 458C. The scope of ketones was
explored by employing 2a as a bridging alkene and either 2b
or 2c as a capping group (Figure 2). For most examples,
employing either 2b or 2c as a capping group led to similar
yields of the isolated products unless otherwise stated.
Electron-neutral and electron-rich methyl ketones afforded
high yields of the desired products (3aab, 3aac, 3bab, 3bac),
while an electron-deficient ketone was not as reactive and
resulted in moderate yields at 458C with 2c as a capping group
(3cac). Attempts to synthesize 3cab resulted in substantially
lower yields than 3cac under similar reaction conditions. An
ortho-substituted methyl aryl ketone afforded high yields at
458C (3dab, 3dac). Aliphatic ketones were not as effective as
aryl ketones for the subsequent g-coupling and resulted in
a low yield of the desired a,g-coupled product 3eab. An
acyclic a-substituted ketone resulted in moderate yields at
458C (3 fab, 3 fac). When a-tetralone was subjected to our
reaction conditions at 458C, the desired products were
obtained in moderate yields without any detection of
the formation of 3aaa, the amount of LiHMDS and alkenyl
halide were both increased to 250 mol%. However, neither
an a-alkenylated nor an a,g-alkenylated product was detected
(entry 2). A screening of bases identified LiOtBu as a suitable
base, thus resulting in a quantitative yield of 3aaa as a single
regioisomer (entry 3). The crude product was remarkably
pure as shown by the 1H NMR spectrum of the material
obtained after a simple work-up procedure without further
purification. While 3aa could be obtained very efficiently
under these optimized reaction conditions, a sequential
coupling employing two different electrophiles (bridging
alkene and capping group) would allow a greater variation
in scope.
To employ two different electrophiles as bridging alkene
and capping groups, the use of the reactants must proceed
with a good level of control in their order of incorporation.
The formation of the initial a-alkenylated intermediate must
result in a high yield before the incorporation of a second
electrophilic component and must avoid further reaction of
this intermediate with the first electrophile to give the
previously observed a,g-alkenylated product 3aaa. A facile
way to overcome the formation of 3aaa is to employ excess
ketone, but this tactic is wasteful because of the large excess
Angew. Chem. Int. Ed. 2015, 54, 11822 –11825
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 11823

.
Angewandte
Communications
Figure 3. Reaction scope of bridging alkene. Yields are those of the
isolated products. Reactions were conducted on a 0.25 mmol scale
relative to 1. TBS=tert-butyldimethylsilyl.
Figure 2. Reaction scope of ketones. Yields are those of the isolated
products. Reactions were conducted on a 0.25 mmol scale relative to
1. [a] Conducted on a 5 mmol scale. [b] Reaction was run at 458C in
THF (0.125m).
aromatized naphthol products (3gab 3gac). Employing
indanone at 458C resulted in high yields with both styryl
and phenyl capping groups (3hab, 3hac). Our protocol was
applicable to scale-up, thus producing 1.38 grams (82% yield)
of 3bac.
Figure 4. Reaction scope of capping group. Yields are those of the
isolated products. Reactions were conducted on a 0.25 mmol scale
relative to 1. [a] Reaction was run at 458C in THF (0.125m).
Next, the scope of the bridging alkenyl halide was
examined (Figure 3). A high yield of a g-quaternary product
(3adb) was obtained when R2 is an extended alkyl chain.
Products with tertiary g-carbon atoms could be generated in
high yields (3 fea, 3 fef) by employing propiophenone, cis-1-
bromopropene, and an alkenyl capping group with cis sub-
stitution. An altered outcome was observed when an a-
unsubstituted ketone was employed and cis-1-bromopropene
was utilized as a bridging alkene. With bromobenzene as the
capping group, a nonconjugated ketone product 4aec was
produced in contrast with the normally obtained conjugated
ketones.[10] a-Bromostyrene and 2-bromo-3-methyl-2-butene
did not result in significant product formation when employed
as bridging alkenes.
Finally, the scope with respect to the capping group to
form a quaternary g-center was examined with 1a as the
ketone and 2a as the bridging alkene (Figure 4). An activated
ester alkenyl bromide capping group resulted in moderate
yields (3aag) and was the only example in which a competing
a,a-coupling product was observed, but still favored the usual
a,g-coupling product by a factor of 2:1. When the ester group
was replaced by a silyl-protected alcohol, complete selectivity
for the a,g-coupled product was observed in good yield
without desilylation (3aaf). Employing a capping group
containing an alkenyl bromide and an aryl bromide moiety
afforded a moderate yield of 3aah with 3:1 chemoselectivity
favoring the g-alkenylation over the arylation product. A
more sterically hindered ortho-substituted aryl bromide
afforded a good yield at 458C (3aai). A range of para-
substituted electron-deficient and electron-rich aryl bromides
all coupled in high yields (3aaj–aam). Employing chloroben-
zene as a capping group resulted in an insignificant yield at
228C, and led to the use of 4-chloro-1-bromobenzene as
a capping group to afford a high yield of 3aan. A thiophene
capping group was tolerated and gave a high yield of 3aao.
Employing 3-bromopyridine, 2-bromoaniline, 5-bromoindole,
and a-bromostyrene as capping groups resulted in low yields
of the desired products.
A
final preliminary observation suggests that this
approach to a,g-coupling may not be limited to the use of
11824 www.angewandte.org
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2015, 54, 11822 –11825

Angewandte
Chemie
Rev. 1996, 96, 115; d) L. F. Tietze, Domino Reactions: Concept
palladium as a catalyst. When the b,g-unsaturated ketone 3 fa,
obtained by our previously reported nickel- or palladium-
catalyzed ketone alkenylation,[3b,c] was subjected to further
coupling with (E)-b-bromostyrene with a nickel catalyst, 3 fab
was obtained in 83% yield as an overall a,g-coupling product
(Figure 5). The same product was obtained above in 57%
yield from the palladium-catalyzed sequential procedure
(Figure 2). We are unaware of previously reported nickel-
catalyzed g-alkenylations or g-arylations.
for Efficient Organic Synthesis Wiley-VCH, Weinheim, 2014;
e) Y. Wang, H. Lu, P. -F. Xu, Acc. Chem. Res. 2015, 48, 1832.
[2] a) P. -F. Xu, W. Wang, Catalytic Cascade Reactions, Wiley,
Somerset, 2014; b) N. T. Patil, Y. Yamamoto, Top. Organomet.
Chem. 2006, 19, 91; c) P. Von Zezschwitz, A. De Meijere, Top.
Organomet. Chem. 2006, 19, 49; d) D. Alberico, M. E. Scott, M.
Lautens, Chem. Rev. 2007, 107, 174; e) T. Vlaar, E. Ruijter, R. V.
Orru, Adv. Synth. Catal. 2011, 353, 809; f) H. Ohno, Asian J. Org.
Chem. 2013, 2, 18; g) C. J. Ball, M. C. Willis, Eur. J. Org. Chem.
2013, 425; h) A. De Meijere, P. Von Zezschwitz, S. Bräse, Acc.
Chem. Res. 2005, 38, 413; i) H. Zhou, C. Moberg, J. Am. Chem.
Soc. 2012, 134, 15992; j) J. Buendia, B. Darses, P. Dauban,
Angew. Chem. Int. Ed. 2015, 54, 5697; Angew. Chem. 2015, 127,
5789; k) Y. Horino, A. Aimono, H. Abe, Org. Lett. 2015, 17,
2824.
[3] a) T. Ankner, C. C. Cosner, P. Helquist, Chem. Eur. J. 2013, 19,
1858; b) M. Grigalunas, T. Ankner, P.-O. Norrby, O. Wiest, P.
Helquist, Org. Lett. 2014, 16, 3970; c) M. Grigalunas, T. Ankner,
P. -O. Norrby, O. Wiest, P. Helquist, J. Am. Chem. Soc. 2015, 137,
7019; d) Z. Huang, L. H. Lim, Z. Chen, Y. Li, F. Zhou, H. Su, J. S.
Zhou, Angew. Chem. Int. Ed. 2013, 52, 4906; Angew. Chem.
2013, 125, 5006; Z. Huang, Z. Chen, L. H. Lim, G. C. P. Quang,
H. Hirao, J. Zhou, Angew. Chem. Int. Ed. 2013, 52, 5807; Angew.
Chem. 2013, 125, 5919.
[4] For a review of palladium-catalyzed g-couplings, see: a) I.
Franzoni, C. Mazet, Org. Biomol. Chem. 2014, 12, 233; For
examples of individual reports of Pd-catalyzed g-couplings, see:
b) Y. Terao, T. Satoh, M. Miura, M. Nomura, Tetrahedron 2001,
56, 1315; c) Y. Terao, Y. Kametani, H. Wakui, T. Satoh, M.
Miura, M. Nomura, Tetrahedron 2001, 57, 5967; d) G. N. Varseev,
M. E. Maier, Org. Lett. 2005, 7, 3881; e) A. M. Hyde, S. L.
Buchwald, Org. Lett. 2009, 11, 2663; f) T. Imahori, T. Tokuda, T.
Taguchi, H. Takahata, Org. Lett. 2012, 14, 1172; g) I. Franzoni, L.
GuØnØe, C. Mazet, Tetrahedron 2014, 70, 4181.
[5] a) M. Huss, H. Wieczorek, J. Exp. Biol. 2009, 212, 341; b) T. K.
Macklin, G. C. Micalizio, Nat. Chem. 2010, 2, 638; c) Q. Liu, Y.
Zhang, P. Xu, Y. Jia, J. Org. Chem. 2013, 78, 10885; d) J. Yu, T.
Wang, X. Liu, J. Deschamps, J. Flippen-Anderson, X. Liao, J. M.
Cook, J. Org. Chem. 2003, 68, 7565; e) R. V. Edwankar, C. R.
Edwankar, J. M. Cook, J. Org. Chem. 2014, 79, 10030; f) J. M.
Smith, J. Moreno, B. W. Boal, N. K. Garg, J. Am. Chem. Soc.
2014, 136, 4504.
Figure 5. Nickel-catalyzed g-alkenylation. cod=1,5-cyclooctadiene.
In conclusion, we have described the development of
a novel palladium-catalyzed reaction sequence that operates
by a-alkenylation of a ketone with an alkenyl bromide
(bridging group) and subsequent g-coupling of an alkenyl
bromide or aryl bromide (capping group) for rapid construc-
tion of a,b-unsaturated ketones with complex carbon scaf-
folds. The large number of examples in Figures 2–4 illustrates
the broad scope of this a,g-coupling reaction. A variety of
electronically and structurally different ketones, bridging
alkenyl bromides, and capping alkenyl bromides or aryl
bromides provide moderate to high yields with excellent
regioselectivity for g-coupling under relatively mild reaction
conditions (22–458C). The procedure avoids the use of large
excesses of any of the three components, thus facilitating the
isolation of the final products. In some cases, the crude
product is sufficiently pure and further purification is not
necessary. Most importantly, the overall products are
obtained in a single-flask sequence in place of employing
separate reactions for facile access to molecular diversity.
Points for further study of this reaction sequence include
identifying additional factors which affect the reaction out-
come, choice of catalysts, mechanistic details, enantioselective
versions, and applications in synthesis.
[6] Y. Terao, T. Satoh, M. Miura, M. Nomura, Tetrahedron Lett.
1998, 39, 6203.
[7] A. M. Hyde, S. L. Buchwald, Angew. Chem. Int. Ed. 2008, 47,
177; Angew. Chem. 2008, 120, 183.
[8] D. S. Huang, J. F. Hartwig, Angew. Chem. Int. Ed. 2010, 49, 5757;
Angew. Chem. 2010, 122, 5893.
[9] S. Duez, S. Bernhardt, J. Heppekausen, F. F. Fleming, P. Knochel,
Org. Lett. 2011, 13, 1690.
[10] For palladium-catalyzed dienolate coupling reactions leading to
unconjugated products, see: a) M. Yu, Y. Xie, J. Li, Y. Zhang,
Adv. Synth. Catal. 2011, 353, 2933; b) Y. Yamamoto, S. Hatsuya,
J.-I. Yamada, J. Chem. Soc. Chem. Commun. 1988, 86; c) Y.
Yamamoto, S. Hatsuya, J. Yamada, J. Org. Chem. 1990, 55, 3118;
d) K. H. Kim, S. Lee, S. H. Kim, C. H. Lim, J. N. Kim,
Tetrahedron Lett. 2012, 53, 5088.
Acknowledgements
This work was supported by the National Science Foundation
(CHE-1058075 and CHE-1261033). We thank Prof. Dr. K.
Henderson (Notre Dame) for sharing equipment.
Keywords: cross-coupling · ketones ·
multicomponent reactions · palladium · synthetic methods
How to cite: Angew. Chem. Int. Ed. 2015, 54, 11822–11825
Angew. Chem. 2015, 127, 11988–11991
[1] a) T.-L. Ho, Tandem Organic Reactions, Wiley, New York, 1992;
b) R. Bunce, Tetrahedron 1995, 51, 13103; c) L. F. Tietze, Chem.
Received: June 26, 2015
Published online: August 14, 2015
Angew. Chem. Int. Ed. 2015, 54, 11822 –11825
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 11825