A tunable class of chiral Cp ligands for enantioselective rhodium(III)-catalyzed C-H allylations of benzamides

Article abstract of DOI:10.1021/ja311956k

The lack of robust and tunable chiral versions of cyclopentadienyl (Cp) ligands hampers progress in the development of catalytic asymmetric versions of a myriad of reactions catalyzed by this ubiquitous ligand. Herein, we describe of a class of chiral Cp ligands with tunable steric parameters. Coordinated to transition metals, the ligand creates a well-defined chiral pocket, able to imprint its chirality onto the metal. The corresponding Rh complexes are shown to be excellent catalysts for enantioselective allylation of N-methoxybenzamides via directed C-H functionalizations at very mild conditions. The obtained enantioselectivities are excellent and demonstrate the viability of chiral Cp complexes as selective transition metal catalysts.

Full text of DOI:10.1021/ja311956k

Communication
pubs.acs.org/JACS
A Tunable Class of Chiral Cp Ligands for Enantioselective
Rhodium(III)-Catalyzed C−H Allylations of Benzamides
Baihua Ye and Nicolai Cramer*
́ ́
Laboratory of Asymmetric Catalysis and Synthesis, Institute of Chemical Sciences and Engineering, Ecole Polytechnique Federale de
Lausanne, Lausanne CH-1015, Switzerland
S
* Supporting Information
Scheme 1. Conceptual Design of the Chiral Cp Ligands
ABSTRACT: The lack of robust and tunable chiral
versions of cyclopentadienyl (Cp) ligands hampers
progress in the development of catalytic asymmetric
versions of a myriad of reactions catalyzed by this
ubiquitous ligand. Herein, we describe of a class of chiral
Cp ligands with tunable steric parameters. Coordinated to
transition metals, the ligand creates a well-defined chiral
pocket, able to imprint its chirality onto the metal. The
corresponding Rh complexes are shown to be excellent
catalysts for enantioselective allylation of N-methoxyben-
zamides via directed C−H functionalizations at very mild
conditions. The obtained enantioselectivities are excellent
and demonstrate the viability of chiral Cp complexes as
selective transition metal catalysts.
yclopentadienyl ligands are among the most versatile and
Coften used ligands to access robust and catalytically
competent transition metal complexes.1 Whereas most often
the parent cyclopentadienyl itself (Cp) or pentamethylcyclo-
pentadienyl (Cp*) is used, chiral versions enabling catalytic
asymmetric reactions remain underdeveloped. Cp ligands
combined with other ligands on the metal carrying chiral
information,2 ansa-metallocenes,3 or multidentate Cp ligands
with further coordinative groups4 are established motifs in
asymmetric catalysis. In stark contrast, few chiral Cp precursors
and complexes were reported for catalytic reactions with Cp
being the only permanent ligand on the metal and which
require all remaining coordination sides of the metal for
substrate and reactants binding or turnover.5,6 We recently
introduced a new class of Cp ligands and demonstrated their
potential in Rh(III)-catalyzed enantioselective C−H function-
alizations leading to chiral dihydroquinolones.7,8 Despite the
success of this ligand family, an expanded range of ligand
scaffolds would be of high synthetic value. Along these lines, we
seek to enlarge the available toolbox with ligands that have a
complementary profile. Besides electronic and steric tuning of
the Cp fragment itself, the reactivity and selectivity of chiral 1,2-
disubstituted Cp ligands can be adjusted by manipulation of its
chiral pocket. In a simplified view, this pocket consists of a back
and a side wall (Scheme 1). Whereas the back wall forces an
approach of the incoming reactant from the unsubstituted side,
the side wall is responsible for aligning the three-coordinated
intermediate to minimize steric interactions (B preferred over
A). Once the ligand chirality is transferred to chirality-at-metal
(C),9 an asymmetric reaction can take place. Our previously
introduced cyclohexene-derived ligand family, with 1 as the
most efficient member, has an adjustable back wall. We now
report a class of tunable chiral Cp ligands from a common
synthetic intermediate that allows us to modulate the A:B
alignment by adapting the size of the side wall.
The envisaged ligand class 2 is based on the very successful
C2-symmetric atrop-chiral biaryl moiety, which was used by
Halterman for the parent chiral Cp ligand architecture lacking
its critical 3,3′ substituents.5c−e The synthesis started with 3,
used by Maruoka for the preparation of chiral phase-transfer
catalysts.10 Radical bromination followed by double alkylation
of cyclopentadiene gave a mixture of ligand 5a and the spiro
product 4a (Scheme 2). Cleavage of the methoxy groups
provided phenol 4b, which served as a common platform to
introduce adjustable bulk in both ortho positions.
For instance, several alkyl and silyl groups were introduced
(5b−5d). Alternatively, conversion to the corresponding triflate
allowed for cross-couplings to access aryl-substituted dienes
such as 5e. Thermal rearrangement yielded the cyclopentadiene
ligand progenitors 6a−6f (Scheme 3). Subsequent complex-
ation with {[Rh(C2H4)Cl]2} led to the chiral Cp**Rh(I)
complexes 2a−2f.
A representative X-ray crystal structure analysis of 2a shows
the structural characteristics of this complex family (Figure
1).11 The lower naphthyl portion mainly acts as the back wall,
preventing any coordination approach from this face. The top-
view ORTEP representation illustrates the importance of the
Received: December 7, 2012
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ja311956k | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
ortho substitution of the naphthyl group (the OMe groups for
2a). Variability in this position modulates the chiral pocket and
hence is expected to strongly influence the ratio of the two
three-coordinated intermediates A and B.
Scheme 2. Synthesis of the Spiro Precursors 5
Given our longstanding interest in enantioselective C−H
functionalizations12 and allenes,13 we selected as a challenging
model reaction a directed aromatic C−H allylation14 based on
the Cp*Rh(III)-catalyzed process recently reported by Ma and
co-workers.15−17 With our developed chiral rhodium com-
plexes, this process allows for the first time for a directed
enantioselective aromatic C(sp2)−H allylation. Based on our
previous findings, in situ oxidation of the Rh(I) olefin
complexes 2 with dibenzoyl peroxide provides direct access
to the catalytically competent Rh(III) carboxylate species,
avoiding the use of any basic carboxylate additive.18 The
reaction of methyl hydroxamate 7a and trisubstituted allene 8a
was first explored with our cyclohexene-derived catalyst 1. The
low selectivity for the monoallylated product 9aa underscores
the needs for additional ligand scaffolds (Table 1, entry 1).
Scheme 3. Synthesis of the Cp**Rh(I) Complexes 2
a
Table 1. Optimization of the Rh-Catalyzed Allylation
b
c
d
entry [Rh]
solvent
T (°C) 9aa:10aa
yield (%) er of 9aa
1
1
MeOH
MeOH
MeOH
MeOH
MeOH
CH2Cl2
MeOH
CH2Cl2
acetone
toluene
CH2Cl2
CH2Cl2
CH2Cl2
0
2:1
3.5:1
2.5:1
3:1
62
71
60
80
63
70
40
82
42
67
90
76
80
61:39
2
2f
0
67.5:32.5
77.5:22.5
84.5:15.5
90:10
3
2a
2b
2c
2d
2e
2c
2c
2c
2c
2c
2c
0
4
0
0
5
3.2:1
5.4:1
2:1
6
0
94:6
7
0
72:28
8
0
5.8:1
2.3:1
6:1
94.5:5.5
90:10
9
0
10
11
12
13
0
94.5:5.5
96.5:3.5
92:8
e
e
−20
0
30:1
6.9:1
10:1
, f
−20
96.5:3.5
a
Conditions: 0.05 mmol 7a, 0.05 mmol 8a, 5 mol% [Rh], 5 mol%
b
c
1
(BzO)2, 0.2 M in solvent, 18 h. Determined by H NMR. Isolated
d
yields of 9a. Determined by HPLC with a chiral stationary phase.
e
0.05 mmol 7a, 0.075 mmol 8a, 2 mol% [Rh], and 2 mol% (BzO)2.
Complex 2f, with unsubstituted ortho positions of the Cp
ligand,5c catalyzed the reaction well, albeit in moderate
enantiomeric ratio of 67.5:32.5 and with significant amounts
of the double allylation product 10aa (entry 2). The selectivity
was significantly better with 2a, having two methoxy
substituents (entry 3). Larger ortho groups improved the
selectivity further (entries 4−6), the best obtained with 2c,
having large OTIPS moieties (entry 5). Phenyl substitution of
the ligand (2e) was not successful for this reaction (entry 7). A
solvent switch to dichloromethane and reduction of the ratio
7a:8a to 1:1.5 mitigated competing double allylation and
further improved the selectivity (entry 11). With 2 mol%
catalyst loading, a slightly lower selectivity was observed (entry
12). Conducting the reaction at −20 °C improved the
enantiomeric ratio further, providing 9a with 96.5:3.5 er in
80% isolated yield at a 7a:8a ratio of 1:1 (entry 13).
Figure 1. (I) X-ray crystal structure of Rh(I) complex 2a. (II) Top-
view of 2a showing the important modulation of the chiral space
around O2 and C2.
B
dx.doi.org/10.1021/ja311956k | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
When the allylation reaction was carried out with 2.2 equiv of
allene 8a at 0 °C, the double allylated product 10a became
dominant and was isolated in excellent enantioselectivity of
99.5:0.5 er (eq 1).
reduced selectivity observed (entry 9), whereas a m-CF3
substituent behaves normally (entry 4). The allene component
8 can be varied over a wide range of trisubstituted allenes
(entries 10−17). For instance, protected or free hydroxyl
groups and esters at various distances from the allene can be
used. Simple unfunctionalized allenes like 8g provide similar
selectivity (entry 15), leading to the conclusion that the
selectivity is by and large determined by the steric difference
between the H and R2 substituents of allene 8. The conditions
are very mild, and even delicate products, such as 9af with a
rather acidic stereogenic center, are accessed in good selectivity
(entry 14).
The absolute configuration of the allylated products was
unambiguously established by X-ray crystallographic analysis of
9ic to be (R) (Figure 2).11 The selectivity leading to this isomer
can be well realized with the stereochemical model. The allene
coordinates with the lesser substituted double bond to the
cyclometalated intermediate C, and the R2 group points away
from the Cp moiety.19 With respect to the ligand backbone, the
hydroxamate section of the substrate is preferentially oriented
With the optimized conditions of Table 1, we next evaluated
the scope of the enantioselective allylation. The reaction is
largely independent of the substitution patterns of the
benzamide. No double allylation was observed with ortho- or
meta-substituted substrates 7 (Table 2, entries 1−5). Electron-
withdrawing and -donating substituents generally have little
influence on the reactivity and selectivity of the allylation
(entries 1−9). Only for 7j, with a p-CF3 group, is a somewhat
a
Table 2. Scope of the Rh-Catalyzed Allylation
a
b
c
Conditions: 0.12 mmol 7x, 0.10 mmol 8y, 5 mol% 2c, 5 mol% (BzO)2, 0.2 M in CH2Cl2, −20 °C, 18 h. Isolated yields. Determined by HPLC
d
with a chiral stationary phase. 0.10 mmol 7, 0.12 mmol 8y.
C
dx.doi.org/10.1021/ja311956k | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
REFERENCES
■
(1) Crabtree, R. H. The Organometallic Chemistry of the Transition
Metals, 3rd ed.; Wiley: New York, 2001.
(2) For an overview, see: Ikariya, T. Bull. Chem. Soc. Jpn. 2011, 84, 1.
(3) For an overview, see: Hoveyda, A. H.; Morken, J. P. Angew.
Chem., Int. Ed. 1996, 35, 1262.
(4) For an overview, see: Siemeling, U. Chem. Rev. 2000, 100, 1495.
(5) For reports of chiral Cp−metal complexes, see: (a) Halterman, R.
L. Chem. Rev. 1992, 92, 965. (b) Halterman, R. L.; Vollhardt, K. P. C.
Organometallics 1988, 7, 883. (c) Colletti, S. L.; Halterman, R. L.
Tetrahedron Lett. 1989, 30, 3513. (d) Colletti, S. L.; Halterman, R. L.
Organometallics 1992, 11, 980. (e) Colletti, S. L.; Halterman, R. L.
Organometallics 1991, 10, 3438. (f) Ramsden, J. A.; Milner, D. J.;
Adams, H.; Bailey, N. A.; Smith, A. J.; White, C. Organometallics 1995,
14, 2575. (g) Paquette, L. A.; Bzowej, E. J.; Kreuzholz, R.
Organometallics 1996, 15, 4857. (h) Ramsden, J. A.; Milner, D. J.;
Adams, H.; Bailey, N. A.; Hempstead, P. D.; White, C. J. Organomet.
Chem. 1998, 551, 355. (i) Schumann, H.; Stenzel, O.; Girgsdies, F.;
Haltermann, R. L. Organometallics 2001, 20, 2215. (j) Schumann, H.;
Stenzel, O.; Dechert, S.; Girgsdies, F.; Haltermann, R. L. Organo-
metallics 2001, 20, 5360. (k) Schumann, H.; Stenzel, O.; Dechert, S.;
Girgsdies, F.; Blum, J.; Gelman, D.; Halterman, R. L. Eur. J. Inorg.
Chem. 2002, 211. (l) Gutnov, A.; Heller, B.; Drexler, H.-J.;
Spannenberg, A.; Oehme, G. Organometallics 2003, 22, 1550.
(m) Gutnov, A.; Drexler, H.-J.; Spannenberg, A.; Oehme, G.; Heller,
B. Organometallic 2004, 23, 1002. (n) Rios, R.; Paredes, S.; Pericas, M.
A.; Moyano, A. J. Organomet. Chem. 2005, 690, 358. (o) McGlacken,
G. P.; O’Brien, C. T.; Whitwood, A. C.; Fairlamb, I. J. S.
Organometallics 2007, 26, 3722.
Figure 2. X-ray crystal structure of 9ic and stereochemical model of
the chiral-at-metal complex C accounting for the observed selectivity
(R = OTIPS).
(6) For Co(I)-catalyzed cyclotrimerizations, good enantioselectivities
were achieved: (a) Gutnov, A.; Heller, B.; Fischer, C.; Drexler, H.-J.;
Spannenberg, A.; Sundermann, B.; Sundermann, C. Angew. Chem., Int.
Ed. 2004, 43, 3795. (b) Heller, B.; Gutnov, A.; Fischer, C.; Drexler, H.-
J.; Spannenberg, A.; Redkin, D.; Sundermann, C.; Sundermann, B.
Chem. Eur. J. 2007, 13, 1117. (c) Hapke, M.; Kral, K.; Fischer, C.;
Spannenberg, A.; Gutnov, A.; Redkin, D.; Heller, B. J. Org. Chem.
2010, 75, 3993.
in an antiparallel fashion, pointing away from the bulky OTIPS
group.
In summary, we have reported a class of chiral Cp ligand
precursors and their corresponding Rh(I) complexes, based on
a sterically adjustable biaryl backbone. They are excellent
scaffolds for enantioselective Rh(III)-catalyzed C−H allylations.
This process proceeds with excellent selectivity and is
characterized by its mildness and good functional group
compatibility. Further work focuses on expanding the chiral
Cp ligand platform to other transition metals and trans-
formations.
(7) Ye, B.; Cramer, N. Science 2012, 338, 504.
(8) For a supramolecular approach, embedding a Cp*Rh(III)
complex fragment in an enzyme, that promotes the same reaction,
see: Hyster, T. K.; Knorr, L.; Ward, T. R.; Rovis, T. Science 2012, 338,
̈
500.
(9) (a) Brunner, H. Angew. Chem., Int. Ed. 1999, 38, 1194. (b) Bauer,
E. B. Chem. Soc. Rev. 2012, 41, 3153.
(10) Ooi, T.; Kameda, M.; Maruoka, M. J. Am. Chem. Soc. 2003, 125,
5139.
(11) Crystallographic data for 2a and 9ic are available as Supporting
Information or as CCDC 914717 and 914718. These data can be
obtained free of charge from The Cambridge Crystallographic Data
Centre via www.ccdc.cam.ac.uk/data_request/cif.
(12) (a) Albicker, M. R.; Cramer, N. Angew. Chem., Int. Ed. 2009, 48,
9139. (b) Seiser, T.; Roth, O. A.; Cramer, N. Angew. Chem., Int. Ed.
2009, 48, 6320. (c) Tran, D. N.; Cramer, N. Angew. Chem., Int. Ed.
2011, 50, 11098. (d) Saget, T.; Lemouzy, S.; Cramer, N. Angew. Chem.,
Int. Ed. 2012, 51, 2238. (e) Saget, T.; Cramer, N. Angew. Chem., Int.
Ed. 2012, 51, 12842.
(13) (a) Seiser, T.; Cramer, N. Angew. Chem., Int. Ed. 2008, 47, 9294.
(b) Saget, T.; Cramer, N. Angew. Chem., Int. Ed. 2010, 49, 8962.
(c) Tran, D. N.; Cramer, N. Angew. Chem., Int. Ed. 2010, 49, 8181.
(14) For the first example of a directed aromatic C−H prenylation
with an Ir(I) catalyst, see: Zhang, Y. J.; Skucas, E.; Krische, M. J. Org.
Lett. 2009, 11, 4248.
(15) Zeng, R.; Fu, C.; Ma, S. J. Am. Chem. Soc. 2012, 134, 9597.
(16) For a related reaction with allenes, see: Wang, H.; Glorius, F.
Angew. Chem., Int. Ed. 2012, 51, 7318.
ASSOCIATED CONTENT
* Supporting Information
Synthetic procedures, characterization data for all new
compounds, HPLC traces of the allylation products, and
crystallographic data (CIF) for 2a and 9ic. This material is
available free of charge via the Internet at http://pubs.acs.org.
■
S
AUTHOR INFORMATION
Corresponding Author
nicolai.cramer@epfl.ch
Notes
■
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work is supported by the Swiss National Science
Foundation (no. 137666) and the European Research Council
under the European Community’s Seventh Framework
Program (FP7 2007−2013)/ERC Grant agreement no.
257891. We thank Dr. R. Scopelliti for X-ray crystallographic
analysis of 2a and 9ic.
(17) For an overview on Rh(III)-catalyzed C−H functionalizations,
see: Song, G.; Wang, F.; Li, X. Chem. Soc. Rev. 2012, 41, 3651.
(18) Lapointe, D.; Fagnou, K. Chem. Lett. 2010, 39, 1118.
(19) Gladysz, J. A.; Boone, B. J. Angew. Chem., Int. Ed. 1997, 36, 550.
D
dx.doi.org/10.1021/ja311956k | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX