Enantioselective rhodium-catalyzed C-C bond activation of cyclobutanones

Article abstract of DOI:10.2533/chimia.2015.187

The activation of carbon-carbon bonds has attracted much attention in the past decade. Despite important progress, the development of asymmetric reactions lags behind. For the first time, asymmetric rhodium(i)-catalyzed direct oxidative additions into ena

Full text of DOI:10.2533/chimia.2015.187

Laureates: Junior Prizes, sCs FaLL Meeting 2014
CHIMIA 2015, 69, No. 4 187
doi:10.2533/chimia.2015.187
Chimia 69 (2015) 187–190 © Schweizerische Chemische Gesellschaft
Enantioselective Rhodium-catalyzed C–C
Bond Activation of Cyclobutanones
Laetitia Souillart§ and Nicolai Cramer*
§SCS-Metrohm Award for best oral presentation
Abstract: The activation of carbon–carbon bonds has attracted much attention in the past decade. Despite
important progress, the development of asymmetric reactions lags behind. For the first time, asymmetric
rhodium(i)-catalyzed direct oxidative additions into enantiotopic C–C bonds of cyclobutanones could be realized.
Subsequentcarboacylationoftetheredolefinsandcarbonylgroupsofthegeneratedrhoda(iii)cyclopentanonegive
an efficient access to complex polycyclic scaffolds in high yields. Despite operating at high reaction temperatures,
the processes are characterized by outstanding enantioselectivities of generally greater than 99.5:0.5 er.
Keywords: Asymmetric catalysis · C–C bond activation · Cyclobutanone · Rhodium
(Scheme 1). In 2012, Murakami disclosed ing step. Processes in which the key oxida-
an enantioselective nickel-catalyzed syn- tive addition into the C–C bond is the en-
Introduction
thesis of benzobicyclo[2.2.2]octanones 2 antiodetermining step were elusive.[7] C–C
from 3-styryl-substituted cyclobutanones bond activations require forcing reaction
1.[5] In the presence of [Ni(cod)2] and a temperatures, rendering the development
chiral phosphoramidite ligand, an enan- of enantioselective processes challenging.
tioselective oxidative cyclization delivers In the following, we demonstrate asym-
nickel(ii) cyclobutanolate I. Subsequent metric oxidative additions of rhodium(i)
diastereoselective β-carbon elimination complexes into enantiotopic C–C bonds of
and reductive elimination close the cata- cyclobutanones, leading to efficient meth-
lytic cycle yielding benzobicyclo[2.2.2] ods for the preparation of chiral bicyclic
octanone 2. Dong reported rhodium(i)- scaffolds.
The catalytic activation of carbon–car-
bon single bonds is a prime challenge in
organometallic chemistry, since the lack of
prefunctionalization steps opens the way
to new, economically and ecologically at-
tractive reaction pathways.[1] Strained ring
substrates occupy a privileged role in C–C
bond activations as the release of their ring
strain facilitates the desired metal inser-
tion. Important progress has been made in
the field during the last decade.[2] However,
the development of asymmetric variants
lags behind. Transition metal-catalyzed
C–C bond cleavages fall into two major
mechanistic categories: oxidative addition
or β-carbon elimination. Whereas exam-
ples for enantioselective β-carbon elimina-
tion processes have recently become more
frequent,[3,4] asymmetric reactions of direct
insertions into C–C bonds are scarce. For
reactions involving oxidative additions
of transition metals as the C–C cleavage
mechanism, strained ketones such as cy-
clobutanones have proven versatile. So
far, only two asymmetric transformations
of cyclobutanones have been reported
catalyzed C–C bond activations of ben-
zocyclobutanones 3.[6] This process is
initiated by an achiral oxidative insertion Results and Discussion
into the aryl–acyl C–C bond of benzocy-
clobutanone substrate 3 generating rho- Development of an Enantio-
dacyclopentenone II. An enantioselective selective Olefin Carboacylation
migratory insertion of the alkene moiety
Murakami and Ito reported rhodium(i)-
and reductive elimination affords fused te- catalyzed intramolecular olefin carbo-
tralones 4. In both of these examples, the acylations.[8] The process converted cy-
C–C cleavage is not the enantiodetermin- clobutanone 1 (R = H) into symmetrical
Murakami (2012): Enantiodetermining Oxidative Cyclization
R1
R1
R
10 mol% [Ni(cod)2]
O
R2
12 mol% L*
R2
R2
O
O
[NiII]
*
1
I
2
77-97%, 80-93% ee
Dong (2013): Enantiodetermining Migratory Insertion
R2
n
n
*
R2
O
n
O
O
5 mol% [{Rh(cod)Cl}2]
R2
O
[RhIII]
12 mol% L*
O
R1
O
*Correspondence: Prof. N. Cramer
Ecole Polytechnique Fédérale de Lausanne
EPFL SB ISIC LCSA
R1
R1
3
II
4
44-97%, 92-99% ee
BCH 4305
CH-1015 Lausanne
Tel.: +41 21 693 98 39
E-mail: nicolai.cramer@epfl.ch
Scheme 1. Asymmetric transformations of cyclobutanones involving a non-enantiodetermining
C–C bond cleavage.

188 CHIMIA 2015, 69, No. 4
Laureates: Junior Prizes, sCs FaLL Meeting 2014
Scheme 2.
benzobicycloheptanone 5 via rhoda(iii)
cyclopentanone III. An efficient enantio-
selective access to the valuable benzobicy-
cloheptanone scaffold 5 would be of great
interest (Scheme 2). We thus selected cy-
clobutanone 1a (R ≠ H) as model substrate
for the development of the enantioselective
methodology (Scheme 3). Executing the
reaction in the presence of [{Rh(cod)Cl}2]
and BINAP (L1) in dioxane at 130 °C
gave benzobicycloheptanone 5a with an
excellent enantiomeric ratio of 98.5:1.5.
However, a moderate 50% yield was ob-
served due to a limited conversion. The re-
activity could be increased with ligands of
the Segphos family. The bulkiest member
DTBM-Segphos (L4) resulted in the most
active catalyst. Under these conditions,
product 5a was obtained in 94% yield with
an exceptionally high enantiomeric ratio
of 99.7:0.3. Related DTBM-MeOBiphep
(L5) was less reactive leading to moderate
conversion. The counterion of the rhodium
complex was found to be of critical impor-
tance. The use of other rhodium sources
such as [{Rh(cod)OH} ] or the cationic
[{Rh(cod)}BF ] was d2etrimental to the
yield and the e4nantioselectivity.
R
R
R
O
[RhIII
Envisioned enantio-
topic C–C bond acti-
vation followed by an
olefin carboacylation.
[RhI], L*
O
*
]
Enantioselective
C-C Activation
O
1
III
5
Scheme 3.
Me
Me
2.5 mol% [{Rh(cod)Cl}2]
Optimization of the
enantioselective C–C
bond activation of
cyclobutanone 1a.
O
6 mol% L*
dioxane, 130 °C, 12 h
O
1a
5a
O
O
O
PAr2
PAr2
MeO
MeO
P(DTBM)2
P(DTBM)2
PPh2
PPh2
O
L1
Ar = Ph (L2): 44%, 99.0:1.0 er
Ar = 3,5-Xylyl (L3): 72%, 99.5:0.5 er
Ar = DTBM (L4): 94%, 99.7:0.3 er
L5
50%, 98.5:1.5 er
53%, 99.5:0.5 er
DTBM = 3,5-tBu-4-MeO-C6H2
O
R1
R1
R4
2.5 mol% [{Rh(cod)Cl}2]
O
O
6 mol% L4
P(DTBM)2
R3
R2
P(DTBM)2
O
dioxane, 130 °C, 12-24 h
R2
R3
R5
O
R4
L4
O
R5
1
5
With the optimized reaction condi-
tions, the generality of the process was
explored (Scheme 4). The influence of dif-
ferent substituents (R1) at the 3-position
of the cyclobutanone including several
aliphatic and aromatic groups as well as
esters, nitriles and protected ethers were
minimal and delivered the desired benzo-
bicycloheptanones 5 with a variety of func-
tionalized bridgeheads in high yields and
enantioselectivities. Electronic modifica-
tions of the aryl moiety (R2) has no influ-
ence on the reaction outcome. Importantly,
1,1-disubstituted alkenes (R4, R5 = H) pro-
vide benzobicycloheptanones 5j and 5k
bearing quaternary stereogenic centers at
both bridgehead positions in similar yields
and enantioselectivities. Moreover, 1,2-di-
substituted alkenes react well and fully
maintain their stereochemical information.
Cyclobutanones bearing a trans-olefin (R3
and R5 = H) deliver ketones 5l and 5m in
excellent diastereomeric ratios (> 20:1).
On the other hand, substrate bearing a cis-
alkene (R3, R4 = H) did not yield the de-
sired benzobicycloheptanone and slowly
degrades under the reaction conditions. A
tri-substituted olefin is well tolerated and
provides a rapid and efficient access to tet-
racyclic ketone 5n.
R1
Me
Et
Me
MeO
Me
F
O
O
O
O
R1= Me (5a) 97%, 99.7:0.3 er
R1= Et (5b) 91%, 99.6:0.4 er
R1= CN (5c) 95%, 99.6:0.4 er
78%, 99.8:0.2 er
5g
5h
81%, 99.6:0.4 er
5i
96%, 98.2:0.2 er
92%, 99.7:0.3 er
R1= CO2Me (5d)
R1= CH2OBn (5e)
R1= CH2OTIPS (5f)
96%, 99.6:0.4 er
89%, 99.6:0.4 er
Me
Me
Me
Ph
Ph
Ph
Me
OBn
Me
n-Bu
H
H
O
O
O
O
O
H
5j
5k
89%, 99.4:0.6 er
5l
5m
5n
73%, 98.3:1.7 er
95%, 99.6:0.4 er
90%, 97.7:2.3 er
> 20:1 dr
85%, 99.8:0.2 er
> 20:1 dr
Scheme 4. Scope of the enantioselective C–C bond activation of cyclobutanones 1.
Scheme 5. Selective
Dong (2008): Rhodium(I)-Catalyzed Carbonyl Hydroacylation
C–C bond activations
in the presence of
aldehydes for asym-
metric carbonyl car-
boacylations.
O
O
5 mol% [{Rh(cod)}BF4]
O
12 mol% L*
H
Ph
O
H
O
O
Ph
6
7
Oxidative
Addition
Reductive
Elimination
[RhI]
O
O
O
[RhIII
O
]
[RhIII
] H
O
Hydrometallation
Ph
O
V
H
IV
Ph
Development of an Enantio-
selective Carbonyl Carboacylation
The utility of the asymmetric C–C
bond activation process could be extended
to carbonyl carboacylations, thus expand-
ing the accessible scaffold range. In this
case, the reaction provides an efficient
access to lactones – a ubiquitous and im-
Our idea: Rhodium(I)-Catalyzed Carbonyl Carboacylation
untouched
H
O
H
O
O
H
[RhIII
]
Enantioselective
C-C Activation
R
O
R
O
O
R
8
VI
9

Laureates: Junior Prizes, sCs FaLL Meeting 2014
CHIMIA 2015, 69, No. 4 189
portant structural motif – from uncommon
synthetic precursors. A related rhodium(i)-
catalyzed intramolecular asymmetric car-
bonyl hydroacylation from 6 providing
Tishchenko-type lactone products 7 was
reported by Dong (Scheme 5).[9] The major
limitation is that carbonyl hydroacylations
are strictly limited to the transfer of a hy-
dride to the accepting carbonyl group. In
contrast, our envisioned carbonyl carboac-
ylation would allow for the formation of
C–C bonds during the lactonization event.
From a mechanistic point, these carbonyl
carboacylations require the opposite re-
action order compared to the carbonyl
hydroacylations. In the latter case, the re-
action is initiated by oxidative addition of
rhodium(i) into the aldehyde C–H bond
leading to the acyl rhodium(iii)hydride
intermediate IV. Ketone hydrometallation
then delivers acyl rhodium(iii) species V.
Finally, a C–O bond forming reductive
elimination closes the catalytic cycle. In
our case, the enantioselective C–C bond
activation of the cyclobutanone 8 giving
rhodium(iii) intermediate VI must pro-
ceed first, leaving the generally more re-
active aldehyde untouched. We envisioned
that the superior reactivity of the strained
cyclobutanone would enable such reactiv-
ity reversal.
The evaluation of this hypothesis was
conducted on model substrate 8a. Again,
the chloride counteranion was critical for
the reactivity. Different chiral ligands were
examined using [{Rh(cod)Cl} ] as rho-
dium source (Scheme 6). Sim2ilar trends
as for the olefin carboacylation were ob-
served. BINAP (L1) as chiral ligand gave
lactone 9a in a promising enantiomeric
ratio of 93.2:6.8, however in a very poor
yield of 8%. Ligands of the Segphos fam-
ily resulted in higher reactivity as well as
enantioselectivities. DTBM-Segphos (L4)
proved to be the most efficient and afford-
ed lactone 9a in 94% yield and excellent
enantiomeric ratio of 99.4:0.6. The related
DTBM-MeOBiphep (L5) was less reac-
tive. Despite outstanding enantioselectiv-
ity of 99.8:0.2 er, Difluorphos (L6) gives
a poorly reactive catalyst.
Me
2.5 mol% [{Rh(cod)Cl}2]
O
6 mol% L*
O
dioxane, 110 °C, 12 h
Me
O
O
8a
9a
O
O
F
F
O
O
O
O
PAr2
PAr2
MeO
MeO
P(DTBM)2
P(DTBM)2
PPh2
PPh2
PPh2
PPh2
F
F
O
O
L1
Ar = Ph (L2): 30%, 99.2:0.8 er
L5
65%, 99.5:0.5 er
L6
8%, 93.2:6.8 er Ar = 3,5-Xylyl (L3): 45%, 99.5:0.5 er
Ar = DTBM (L4): 94%, 99.4:0.6 er
15%, 99.8:0.2 er
Scheme 6. Optimization of the asymmetric carbonyl carboacylation.
O
R1
R3
O
2.5 mol% [{Rh(cod)Cl}2]
O
O
O
6 mol% L4
dioxane, 110 °C, 12-24 h
P(DTBM)2
P(DTBM)2
R3
O
R2
R2
R1
O
L4
O
8
9
MeO
Me
O
O
O
O
R1
Me
9g
Et
9h
Bu
9i
F
O
O
O
O
O
O
1
R1= Me (9a) 89%, 99.4:0.6 er
R1= Ph (9b) 80%, 99.4:0.6 er
R = Bu (9c) 76%, 99.3:0.7 er
78%, 99.7:0.3 er
83%, 99.6:0.4 er
83%, 99.5:0.5 er
69%, 99.7:0.3 er
R1= CO2Me (9d)
CO2Et
O
Me
R1= CH2OBn (9e)
R1= CH2OTIPS (9f)
R2
85%, 98.9:1.1 er
64%, 97.2:2.8 er
O
O
Ph
Me
9l
Bu
O
R2= Cl (9j) 86%, 99.5:0.5 er
9m
R2= CF3 (9k) 85%, 99.3:0.7 er
89%, 99.4:0.6 er
81%, 99.8:0.2 er
Scheme 7. Scope for the asymmetric synthesis of bicyclic lactones 9.
Scheme 8. Proposed
mechanistic picture.
R
R
X
O
X
O
5 (X = CR2)
9 (X= O)
Reductive
Elimination
1 (X = CR2)
8 (X= O)
[RhI]
Coordination
R
R
O
O
X
[RhIII]
[RhI]
O
X
R
IX
VII
The generality of the process for the
carbonyl carboacylation was subsequently
investigated (Scheme 7). The influence of
different substituents at the 3-position of
cyclobutanones 8 (R1) including aliphatic
and aromatic groups, methyl ester or pro-
Migratory
Insertion
Enantioselective
Oxidative Addition
[RhIII]
X
VIII
Mechanistic Picture
tected alcohols was limited and the de- active towards migratory insertion. Very
sired polycyclic lactones 9 were obtained activated ketones, such as α-ketoester 8l
in good yields and excellent enantiose- gave benzo[c]oxepinone 9l bearing two
lectivities. Modification of the electronic different quaternary stereogenic centers at
properties of the aryl moiety with electron- the bridgehead positions with no erosion
withdrawing or -donating groups (R2) did of the high enantiomeric ratio. By increas-
not influence the reaction outcome. The ing both the catalyst loading and the reac-
reactivity of ketones as accepting group tion time, a simple methyl ketone reacted
was also investigated (R3 ≠ H). Due to their as well and provided lactone 9m in good
lower electrophilicity, ketones are less re- yields.
The proposed mechanism for both
presented carboacylations is depicted
in Scheme 8. An initial coordination of
rhodium(i) to the carbonyl group as well
as to the unsaturated acceptor (X = CR2 or
X = O) of the cyclobutanone would lead
to complex VII. This double coordination
induces a relatively rigid transition state,
enabling a good enantiodiscrimination in

190 CHIMIA 2015, 69, No. 4
Laureates: Junior Prizes, sCs FaLL Meeting 2014
Chem. Soc. 2003, 125, 8862; b) M. Waibel, N.
the C–C cleavage step. Oxidative addition with outstanding enantioselectivities and
of rhodium into the acyl-carbon bond of are giving an efficient access to complex
cyclobutanone would deliver rhoda(iii) polycyclic scaffolds in high yields. On-
cyclopentanone species VIII. Subsequent going research is focused on the devel-
migratory insertion of either the appended opment of further asymmetric C–C bond
olefin or the carbonyl group would form activations.
Cramer, Angew. Chem. Int. Ed. 2010, 49, 4455;
Rh-catalyzed: c) T. Matsuda, M. Shigeno, M.
Murakami, J. Am. Chem. Soc. 2007, 129, 12086;
d) T. Seiser, N. Cramer, Angew. Chem. Int. Ed.
2008, 47, 9294; e) R. Shintani, K. Takatsu, T.
Hayashi, Org. Lett. 2008, 10, 1191; f) T. Seiser,
O. A. Roth, N. Cramer, Angew. Chem. Int. Ed.
2009, 48, 6320; g) M. Shigeno, T. Yamamoto,
M. Murakami, Chem. Eur. J. 2009, 15, 12929;
h) T. Seiser, N. Cramer, J. Am. Chem. Soc.
2010, 132, 5340; i) T. Seiser, N. Cramer, Angew.
Chem. Int. Ed. 2010, 49, 10163; j) M. Waibel,
N. Cramer, Chem. Commun. 2011, 47, 345; k)
N. Ishida, S. Sawano, M. Murakami, Chem.
Commun. 2012, 48, 1973; l) L. Souillart, N.
Cramer, Chem. Sci. 2014, 5, 837; m) A.Yada, S.
Fujita, M. Murakami, J. Am. Chem. Soc. 2014,
136, 7217.
acyl rhodium species IX. In turn, reductive
elimination would give the polycyclic scaf-
fold 5 (X = CR2)or 9 (X = O).
Acknowledgments
Laetitia Souillart
cordially
thanks
Metrohm Schweiz AG and the Swiss Chemical
Society for the SCS Metrohm Best Oral
Presentations Award. This work was supported
by the European Research Council under the
European Community’s Seventh Framework
Program (FP7 2007– 2013) / ERC Grant agree-
ment no. 257891.
Conclusion
Whereas enantioselective β-carbon
elimination processes are well precedent-
ed, the enantioselective direct oxidative
addition into C–C bonds remained a long
standing challenge. We now demonstrated
the possibility for rhodium(i) complexes
to undergo such enantioselective oxida-
tive additions into enantiotopic C–C bonds
of cyclobutanones. Despite the high reac-
tion temperatures, this reactivity was ex-
ploited for an enantioselective rhodium(i)-
catalyzed C–C bond activation of 3-styryl
cyclobutanones giving an efficient access
to bicycloheptanones.[10] Moreover, we re-
ported an enantioselective rhodium(i)-cat-
alyzed carbonyl carboacylation reaction of
cyclobutanones providing an efficient ac-
cess to the benzo[c]oxepinone skeleton.[11]
Both developed methodologies proceed
[5] L. Liu, N. Ishida, M. Murakami, Angew. Chem.
Int. Ed. 2012, 51, 2485.
[6] T. Xu, H. M. Ko, N. A. Savage, G. Dong, J. Am.
Chem. Soc. 2012, 134, 20005.
[7] E. Parker, N. Cramer, Organometallics, 2014,
33, 780.
[8] M. Murakami, T. Itahashi, Y. Ito, J. Am. Chem.
Soc. 2002, 124, 13976.
[9] Z. Shen, H. A. Khan, V. M. Dong, J. Am. Chem.
Soc. 2008, 130, 2916.
[10] L. Souillart, E. Parker, N. Cramer, Angew.
Chem. Int. Ed. 2014, 53, 3001.
[11] L. Souillart, N. Cramer, Angew. Chem. Int. Ed.
2014, 53, 9640.
Received: January 23, 2015
[1] ‘C–C Bond Activation’, Ed. G. Dong, Springer,
Berlin, Top. Curr. Chem. 2014, 346.
[2] a) C. Aïssa, Synthesis 2011, 21, 3389; b) F.
Chen, T. Wang, N. Jiao, Chem. Rev. 2014, 114,
8613; c) I. Marek, A. Masarwa, P.-O. Delaye,
M. Leibeling, Angew. Chem. Int. Ed. 2015, 414.
[3] For reviews on enantioselective β-carbon eli-
mination: a) T. Seiser, N. Cramer, Org. Biomol.
Chem. 2009, 7, 2835; b) N. Cramer, T. Seiser,
Synlett 2011, 4, 449; c) L. Souillart, E. Parker,
N. Cramer, ‘Asymmetric Transformations via
C–C Bond Cleavage’, Top. Curr. Chem. 2014,
Springer, Berlin, pp 163.
[4] For examples of enantioselective β-carbon
elimination: Pd-catalyzed: a) Y. Matsumura,
Y. Maeda, T. Nishimura, S. Uemura, J. Am.