Asymmetric Catalysis with CO2: The Direct α-Allylation of Ketones

Article abstract of DOI:10.1002/anie.201601545

Quaternary stereocenters are found in numerous bioactive molecules. The Tsuji-Trost reaction has proven to be a powerful C-C bond forming process, and, at least in principle, should be well suited to access quaternary stereocenters via the α-allylation of ketones. However, while indirect approaches are known, the direct, catalytic asymmetric α-allylation of branched ketones has been elusive until today. By combining "enol catalysis" with the use of CO₂ as a formal catalyst for asymmetric catalysis, we have now developed a solution to this problem: we report a direct, highly enantioselective and highly atom-economic Tsuji-Trost allylation of branched ketones with allylic alcohol. Our reaction delivers products bearing quaternary stereocenters with high enantioselectivity and water as the sole by-product. We expect our methodology to be of utility in asymmetric catalysis and inspire the design of other highly atom-economic transformations.

Full text of DOI:10.1002/anie.201601545

Angewandte
Communications
Chemie
International Edition: DOI: 10.1002/anie.201601545
German Edition: DOI: 10.1002/ange.201601545
Synthetic Methods
Asymmetric Catalysis with CO2: The Direct a-Allylation of Ketones
Gabriele Pupo, Roberta Properzi, and Benjamin List*
Abstract: Quaternary stereocenters are found in numerous
bioactive molecules. The Tsuji–Trost reaction has proven to be
amine/chiral Brønsted acid/Pd0 catalytic system,[12] in which
the enantioselectivity was controlled solely by the enantio-
pure counteranion.[13] Furthermore, the Carreira group devel-
oped an enantio- and diastereodivergent a-allylation of
branched aldehydes, catalyzed by a chiral amine in combina-
tion with a chiral iridium p-complex.[14]
The use of unsymmetrical cyclic ketones presents a chal-
lenging regioselectivity problem, as the more hindered enol
needs to selectively react. We recently proposed the concept
of enol catalysis,[15] in which a nucleophilic enol is formed
from the interaction between a branched ketone and a chiral
phosphoric acid catalyst,[16] mimicking an enzymatic enoliza-
tion.[17] This activation mode appeared ideally suited for the
targeted direct a-allylation reaction as it preferentially
proceeds via the more substituted enol. Furthermore, towards
maximizing atom economy, we aimed at using simple allylic
alcohols as electrophiles, an unprecedented feature in combi-
nation with ketones [Eq. (1)].
À
a powerful C C bond forming process, and, at least in
principle, should be well suited to access quaternary stereo-
centers via the a-allylation of ketones. However, while indirect
approaches are known, the direct, catalytic asymmetric a-
allylation of branched ketones has been elusive until today. By
combining “enol catalysis” with the use of CO2 as a formal
catalyst for asymmetric catalysis, we have now developed
a solution to this problem: we report a direct, highly
enantioselective and highly atom-economic Tsuji–Trost allyla-
tion of branched ketones with allylic alcohol. Our reaction
delivers products bearing quaternary stereocenters with high
enantioselectivity and water as the sole by-product. We expect
our methodology to be of utility in asymmetric catalysis and
inspire the design of other highly atom-economic transforma-
tions.
Q
uaternary stereocenters are common motifs in natural
products and they can be found in more than ten percent of
the top 200 prescription drugs.[1] In the last decade, progress
has been made towards the stereoselective construction of
quaternary stereocenters[2] and among other methods, the
asymmetric Tsuji–Trost[3] a-allylation has proven to be of use
in this context.[4–6] However, when unsymmetrical a-branched
ketones (e.g. 2-phenyl or 2-methylcyclohexanone) are
employed as substrates, an additional problem arises from
the difficult control of isomeric enolates (tetra- vs. tri-
substituted). Since the pioneering work by the Trost
group,[7] alternative indirect approaches to this challenge
have been developed. Phase transfer catalysis conditions
(PTC) with activated substrates have also been disclosed.[8]
Most of the reported approaches employ Pd0 and chiral
ligands and have until now involved specific classes of
substrates (e.g. b-ketoesters, 1,3-diketones, tetralones),[7,9]
preformed enolates (e.g. enol ethers or metal enolates),[10]
and/or decarboxylative allylic alkylation conditions.[5,11] These
indirect approaches are less atom-economic, require addi-
tional, sometimes low yielding synthetic steps for the
substrate preparation, and often suffer from a lack of
generality. Hence, despite major advances in the field,
a general and direct catalytic enantioselective a-allylation of
unsymmetrical ketones has never been achieved.
Allylic alcohols have proven to be suitable for the a-
allylation of aldehydes,[12b] but failed in initial attempts with
ketones. To solve this problem, we envisioned an activation
mode for allylic alcohols, in which CO2 would be utilized as
a catalytic reagent, potentially generating a more reactive p-
We have recently developed a direct a-allylation of
branched aldehydes with allylic alcohols using an achiral
[*] G. Pupo, Dr. R. Properzi, Prof. Dr. B. List
Max-Planck-Institut für Kohlenforschung
Kaiser-Wilhelm-Platz 1, 45470 Mülheim an der Ruhr (Germany)
E-mail: list@kofo.mpg.de
Scheme 1. Design of a direct asymmetric a-allylation of branched
ketones with allylic alcohols. ACDC=asymmetric counteranion-
directed catalysis.
Supporting information for this article can be found under:
http://dx.doi.org/10.1002/anie.201601545.
Angew. Chem. Int. Ed. 2016, 55, 6099 –6102
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
6099

Angewandte
Communications
Chemie
allyl precursor species, such as a carbonic acid ester.[18]
Accordingly, we designed the following triple catalytic cycle
(Scheme 1). The chiral phosphoric acid (HA*) initially
interacts with the ketone, preferentially enabling reactivity
from the thermodynamically more stable enol. In turn, the
allylic alcohol is activated by CO2 to form the corresponding
carbonic acid ester, potentially via acid or Pd catalysis. In
a parallel Tsuji–Trost catalytic cycle, the p-allyl-Pd electro-
phile is generated in an oxidative addition into the carbonic
acid ester. Upon release of water and regeneration of CO2,
a [Pd-allyl]+A*À complex would be formed in this step. The
successive nucleophilic attack of the enol onto the p-allyl
cation, reductive elimination, and release of the chiral acid
would afford the desired product bearing a quaternary
stereocenter, and simultaneously regenerate the catalysts.
To investigate if the overall ketone a-allylation was
feasible, we first proceeded to develop the methodology
with allyl methyl carbonate as electrophile, by reacting 2-
phenylcyclohexanone in the presence of 10 mol% of com-
mercially available (S)-TRIP phosphoric acid, an excess of
allylating agent, and 2.5 mol% of different Pd0 sources
(Supporting Information (SI), Table S1). In the absence of
a phosphine ligand, only traces of the desired allylated
product were obtained, probably due to the instability of the
Pd complex and the rapid formation of Pd black. However,
promising enantioselectivities could be observed (up to 89:11
er). We hypothesized that the Pd ligand should influence both
the yield and enantioselectivity as it would modify the sterics
and stability of the electrophile. Indeed, when different
phosphines were tested, a strong effect, both on reactivity and
enantioselectivity, was found (SI, Table S2). Bulky and
electron-rich biphenyl-based phosphine ligands almost exclu-
sively afforded the desired regioisomer and, when highly
hindered tBuXPhos[19] was employed in combination with the
chiral acid, good enantioselectivities could also be achieved
(88:12 er). To our delight, upon fine tuning of the catalyst and
optimization of the reaction conditions (SI, Tables S3–S6),
product 2a could be isolated in 81% yield and an er of 97:3
using (S)-H8-TRIP as catalyst at 358C.
Gratifyingly, our reaction design indeed enabled the first
general, highly enantioselective and regioselective direct a-
allylation of branched ketones (Scheme 2). Low pKa ketones
(e.g. a-aryl substituted ones), represent a challenging sub-
strate class that has previously been shown to be unsuitable
under decarboxylative asymmetric allylic alkylation condi-
tions, and only rare examples with preformed metal enolates
have been reported.[10e] By employing our method, both
electron-rich and electron-poor a-aryl substituted ketones
reacted smoothly and afforded the desired products 2a–2g in
excellent yields and unprecedented enantioselectivities (up to
97:3 er). Higher pKa substrates such as a-alkyl substituted
ketones also afforded the desired products 2h–2i in excellent
enantioselectivities (up to 98:2 er), and in this case moderate
to good yields were obtained, along with small amounts of the
corresponding regioisomer. Other classes of substrates such
as b-ketoester, indanone, and tetralone derivatives all proved
to be compatible with our general protocol and afforded
products 2j–2l in good yields and enantioselectivities (up to
94.5:5.5 er).
Scheme 2. Direct catalytic asymmetric a-allylation of branched ketones
with allyl methyl carbonate. [a] Isolated yields of reactions run on 0.1
or 0.2 mmol scale with 3 equiv of carbonate. Absolute configurations
were determined by comparison of optical rotations with literature
data. [b] Reaction run at 708C (10 mol%, 5 h). [c] 20 mol% acid. All
enantiomeric ratios were determined by HPLC analysis using a chiral
stationary phase.
Having established a general asymmetric protocol, we
turned our attention towards our final goal: the use of allylic
alcohols as ideal atom-economic alternative to carbonates. As
expected, under our optimized conditions, either under argon
or air, the conversion dropped to less than 15%, yet the
enantioselectivity was only slightly diminished. However,
when the reaction was run under an atmosphere of CO2,
reactivity, good conversion, and enantioselectivity were
restored, and product 2a could be isolated in 93% yield and
with an er of 95:5 (see SI).
The generality of this CO2-based protocol and the
suitability of our design were confirmed by the successful
reaction of other representative a-substituted ketones
(Scheme 3). These results suggest the in situ generation of
a p-allyl precursor which resembles allyl carbonates. The
exact mechanism by which CO2 activates allylic alcohol is
unclear at this point, but we envision the establishment of an
equilibrium between CO2/allylic alcohol and the correspond-
ing allyl carbonic acid ester.[18] This would then be followed by
oxidative addition of Pd, regeneration of CO2 and release of
a molecule of H2O. Interestingly, only catalytic amounts of
6100 www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 6099 –6102

Angewandte
Communications
Chemie
product and drug synthesis, and the concepts described herein
may inspire other highly selective and atom-economic
reactions.
Experimental Section
Catalytic asymmetric synthesis of (S)-2a with CO2-activated allylic
alcohol: in a Schlenk equipped with a condenser/cold finger, 2-
phenylcyclohexanone (0.1 mmol, 1 equiv), MS 3 Š (100 mg), (S)-H8-
TRIP (10 mol%), tBuXPhos (11 mol%), Pd2(dba)3 (2.5 mol%) were
dissolved in 1 mL of cyclohexane (0.1m). After three freeze/vacuum/
thaw/CO2 backfill cycles, the reaction mixture was stirred at 458C for
96 h under a CO2 atmosphere (1 atm, balloon). The crude mixture
was then cooled to room temperature and directly purified by flash
column chromatography. For complete experimental details and
compound characterization, see Supporting Information.
Scheme 3. Direct asymmetric a-allylation of branched ketones with
CO2 and allylic alcohol. [a] Isolated yields of reactions run on 0.1 mmol
scale with 3 equiv of allylic alcohol. [b] NMR yield determined by using
triphenylmethane as internal standard.
Acknowledgements
We are grateful for the generous support from the Max-
Planck-Society, the Fonds der Chemischen Industrie (fellow-
ship to G.P. ) as well as the GC department of the MPI für
Kohlenforschung for analytics and Hendrik van Thienen for
the synthesis of (S)-TRIP.
carbon dioxide are formally required for this type of
activation.
To highlight the synthetic potential of our method for the
synthesis of natural products, we have developed a short
enantioselective formal total synthesis of the Amaryllidaceae
alkaloid, (+)-crinane [Eq. (2)] from commercially available
ketone 1g. By employing our novel CO2-promoted a-
allylation reaction, product 2g was obtained in 78% isolated
yield and 95.5:4.5 er. The successive oxidation afforded
compound 3, which can be readily converted into (+)-crinane
in 2 steps.[8b] Our four step approach represents the shortest
synthesis of (+)-crinane.
Keywords: allylic alcohol · atom economy · enol catalysis ·
quaternary stereocenters · a-allylation
How to cite: Angew. Chem. Int. Ed. 2016, 55, 6099–6102
Angew. Chem. 2016, 128, 6204–6207
[1] a) J. Christoffers, A. Baro, Quaternary Stereocenters. Challenges
and Solutions for Organic Synthesis, Wiley-VCH, Weinheim,
2006; b) K. W. Quasdorf, L. E. Overman, Nature 2014, 516, 181 –
191.
[2] For reviews, see a) E. J. Corey, A. Guzman-Perez, Angew. Chem.
Int. Ed. 1998, 37, 388 – 401; Angew. Chem. 1998, 110, 402 – 415;
b) B. M. Trost, C. Jiang, Synthesis 2006, 369 – 396.
[3] For selected reviews on allylic alkylations and their applications
in total synthesis, see: a) B. M. Trost, D. L. Van Vranken, Chem.
Rev. 1996, 96, 395 – 422; b) B. M. Trost, M. L. Crawley, Chem.
Rev. 2003, 103, 2921 – 2943.
[4] M. Braun, T. Meier, Angew. Chem. Int. Ed. 2006, 45, 6952 – 6955;
Angew. Chem. 2006, 118, 7106 – 7109.
[5] Y. Liu, S.-J. Han, W.-B. Liu, B. M. Stoltz, Acc. Chem. Res. 2015,
48, 740 – 751.
[6] M. Braun, T. Meier, F. Laicher, P. Meletis, M. Fidan, Adv. Synth.
Catal. 2008, 350, 303 – 314.
[7] a) B. M. Trost, R. Radinov, E. M. Grenzer, J. Am. Chem. Soc.
1997, 119, 7879 – 7880; b) B. M. Trost, G. M. Schroeder, J. Am.
Chem. Soc. 1999, 121, 6759 – 6760.
[8] a) E. J. Park, M. H. Kim, D. Y. Kim, J. Org. Chem. 2004, 69,
6897 – 6899; b) T. Kano, Y. Hayashi, K. Maruoka, J. Am. Chem.
Soc. 2013, 135, 7134 – 7137.
Key features for the design of new methodologies for
chemical synthesis are generality, selectivity, efficiency, and
atom-economy. Guided by these principles, we designed and
successfully realized the first solution to the long-standing
problem of the direct a-allylation of branched ketones. The
successful combination of palladium catalysis, enol catalysis,
asymmetric counteranion directed catalysis, and an unprece-
dented CO2-catalyzed substrate activation, led to a method to
access challenging quaternary stereocenters, a key motif in
numerous bioactive molecules. We believe that our opera-
tionally simple protocol will be widely applicable in natural
[9] T. Hayashi, K. Kanehira, T. Hagihara, M. Kumada, J. Org. Chem.
1988, 53, 113 – 120.
[10] For selected examples, see: a) B. M. Trost, G. M. Schroeder,
Chem. Eur. J. 2005, 11, 174 – 184; b) B. M. Trost, J. Xu, T.
Schmidt, J. Am. Chem. Soc. 2009, 131, 18343 – 18357; c) F. Nahra,
Y. MacØ, A. Boreux, F. Billard, O. Riant, Chem. Eur. J. 2014, 20,
10970 – 10981; d) T. Graening, J. F. Hartwig, J. Am. Chem. Soc.
2005, 127, 17192 – 17193; e) B. M. Trost, G. M. Schroeder, J.
Angew. Chem. Int. Ed. 2016, 55, 6099 –6102
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
6101

Angewandte
Communications
Chemie
Kristensen, Angew. Chem. Int. Ed. 2002, 41, 3492 – 3495; Angew.
Chem. 2002, 114, 3642 – 3645.
[15] I. Felker, G. Pupo, P. Kraft, B. List, Angew. Chem. Int. Ed. 2015,
54, 1960 – 1964; Angew. Chem. 2015, 127, 1983 – 1987.
[16] a) G. Pousse, F. Le Cavelier, L. Humphreys, J. Rouden, J.
Blanchet, Org. Lett. 2010, 12, 3582 – 3585; b) G. A. Shevchenko,
G. Pupo, B. List, Synlett 2015, 26, 1413 – 1416; c) X. Yang, D.
Toste, J. Am. Chem. Soc. 2015, 137, 3205 – 3208; For recent
reviews on chiral phosphoric acids for asymmetric catalysis, see:
d) T. Akiyama, Chem. Rev. 2007, 107, 5744 – 5758; e) M. Terada,
Synthesis 2010, 12, 1929 – 1982; Also see: f) M. R. Monaco, R.
Properzi, B. List, Synlett 2016, 27, 591 – 594; g) T. Akiyama,
Synlett 2016, 27, 542 – 545.
[17] R. Pollack, Bioorg. Chem. 2004, 32, 341 – 353.
[18] M. Sakamoto, I. Shimizu, A. Yamamoto, Bull. Chem. Soc. Jpn.
1996, 69, 1065 – 1078.
[19] C. H. Burgos, T. E. Barder, X. Huang, S. L. Buchwald, Angew.
Chem. Int. Ed. 2006, 45, 4321 – 4326; Angew. Chem. 2006, 118,
4427 – 4432.
[11] a) D. C. Behenna, B. M. Stoltz, J. Am. Chem. Soc. 2004, 126,
15044 – 15045; b) D. C. Behenna, J. T. Mohr, N. H. Sherden, S. C.
Marinescu, A. M. Harned, K. Tani, S. Masaki, S. Ma, Z. Novµk,
M. R. Krout, R. M. McFadden, J. L. Roizen, J. A. Enquist, Jr.,
D. E. White, S. R. Levine, K. Petrova, A. Iwashita, S. C. Virgil,
B. M. Stoltz, Chem. Eur. J. 2011, 17, 14199 – 14223; c) C. M.
Reeves, D. C. Behenna, B. M. Stoltz, Org. Lett. 2014, 16, 2314 –
2317.
[12] a) S. Mukherjee, B. List, J. Am. Chem. Soc. 2007, 129, 11336 –
11337; b) G. Jiang, B. List, Angew. Chem. Int. Ed. 2011, 50, 9471 –
9474; Angew. Chem. 2011, 123, 9643 – 9646.
[13] For recent reviews on chiral anions for asymmetric catalysis, see:
a) M. Mahlau, B. List, Angew. Chem. Int. Ed. 2013, 52, 518 – 533;
Angew. Chem. 2013, 125, 540 – 556; b) G. L. Hamilton, E. J.
Kang, M. Mba, F. D. Toste, Science 2007, 317, 496 – 499; c) R. J.
Phipps, G. L. Hamilton, F. D. Toste, Nat. Chem. 2012, 4, 603 –
614.
[14] S. Krautwald, D. Sarlah, M. A. Schafroth, E. M. Carreira,
Science 2013, 340, 1065 – 1068.
Received: February 12, 2016
Published online: April 13, 2016
6102 www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 6099 –6102