Deracemization of quaternary stereocenters by Pd-catalyzed enantioconvergent decarboxylative allylation of racemic β-ketoesters

Article abstract of DOI:10.1002/anie.200502018

(Chemical Equation Presented) Stereochemical alchemy! Racemic allyl β-ketoesters allow the regiocontrolled formation of enolates, where the same catalyst is intimately involved in both the stereoablative (C-C bond-breaking) and stereoselective (C-C bond-f

Full text of DOI:10.1002/anie.200502018

Communications
Asymmetric Catalysis
stereoablative process.[5,6] Herein, we describe the first
catalytic system for the deracemization of quaternary
carbon stereocenters. Our enantioconvergent method con-
verts racemic a-substituted 2-carboxyallylcyclohexanones
into highly enantioenriched cycloalkanones that bear quater-
nary stereocenters through catalytic asymmetric decarboxy-
lative allylation.
We recently reported the first catalytic asymmetric
allylation methods for the synthesis of 2-alkyl-2-allylcyclo-
alkanones (Scheme 1).[7] These reactions, based on racemic
DOI: 10.1002/anie.200502018
Deracemization of Quaternary Stereocenters by
Pd-Catalyzed Enantioconvergent
Decarboxylative Allylation of Racemic
b-Ketoesters**
Justin T. Mohr, Douglas C. Behenna,
Andrew M. Harned, and Brian M. Stoltz*
Enantioconvergent catalysis is a powerful
synthetic method that converts a racemic
stereogenic substrate into an enantiomerically
enriched product with a theoretical yield of
100% in a single operation.[1] Conceptually, a
catalytic system for such a reaction must
achieve a stereomutation[2] or stereoabla-
tion[3,4] of the substrate (or an intermediate),
Scheme 1. Enantioselective Tsuji allylations. dba=dibenzylideneacetone; TBAT=tetra-
butylammonium difluorotriphenylsilicate; TMS=trimethylsilyl.
followed by an enantioselective conversion
into product (Figure 1, compare pathways I
transformations reported in the early 1980s by Tsuji,[8] use
enol carbonates and silyl enol ethers along with various allyl
carbonates and a Pd0 catalyst supported by a chiral phosphi-
nooxazoline ligand (e.g., 1).[9–11]
To demonstrate the utility of this methodology, we have
started to employ these allylations as the key enantioselective
reaction in multistep syntheses. In one such project, we
required the substituted cyclohexenone
6 (Scheme 2).
Unfortunately, the preparation of the allylation precursor 4
(R = CO2allyl or SiMe3) was hampered by nonselective
enolization of 3 to form inseparable mixtures of 4 and 5,
which resulted in significant amounts of 7 after allylation.
Prompted by the need for better position control in these
synthetic sequences, we sought mechanistically guided alter-
natives to the reactions depicted in Scheme 1. To this end, we
synthesized dideuterated carbonate 8 and trideuterated
carbonate 9. When 8 was subjected to our standard conditions
for allylation, the deuterium label was almost evenly scram-
bled between the termini of the allyl fragment in the product
(Scheme 3a).[12] In a separate crossover experiment, the
reaction of equimolar amounts of carbonates 8 and 9 was
performed under our standard allylation conditions. As
expected, NMR spectroscopic analysis of the product
showed deuterium scrambling between the allyl termini, but
interestingly, mass spectrometric analysis of the product
showed an almost perfect statistical distribution of enolate
and allyl fragment pairs with four possible masses (Sche-
me 3b).[12] Thus, all six possible products were formed in the
reaction including those derived from crossover reactions.
Although the specific details associated with the enantiodis-
crimination remain unclear, we hypothesized that a discrete
achiral ketone enolate (namely, 10) must exist in solution for
some period to explain our observations.[13]
Figure 1. Stereomutative versus stereoablative enantioconvergent cat-
alysis.
and II). A number of catalytic processes of this type have
been developed, including chemical, enzymatic, and chemo-
enzymatic strategies.[1,2,4] Racemic compounds that contain
quaternary carbon centers are typically unsuitable substrates
for enantioconvergent catalysis because of the difficulty
ꢀ
associated with C C bond cleavage in the stereomutative or
[*] J. T. Mohr, D. C. Behenna, Dr. A. M. Harned, Prof. B. M. Stoltz
The Arnold and Mabel Beckman Laboratories of Chemical Synthesis
Division of Chemistry and Chemical Engineering
California Institute of Technology
1200 E. California Boulevard
MC 164-30, Pasadena, CA 91125 (USA)
Fax: (+1)626-564-9297
E-mail: stoltz@caltech.edu
[**] The authors wish to thank the Fannie and John Hertz Foundation
(predoctoral fellowship to D.C.B.), the NIH (postdoctoral fellowship
to A.M.H.), Research Corporation, the Camille and Henry Dreyfus
Foundation, Merck, Pfizer, and Lilly for financial support, and Prof.
D. A. Dougherty for helpful discussions.
We took this mechanistic insight into account, and so
reasoned that the putative prochiral enolate needed for the
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
6924
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2005, 44, 6924 –6927

Angewandte
Chemie
information, coupled with
the product yield of over
50% and high enantiomeric
excess, clearly demonstrates
the
nature
enantioconvergent
of the system
(Figure 1, pathway II).
With the enantioconver-
gent decarboxylative allyla-
tion of (ꢁ )-12 established,
we investigated the scope of
this unusual asymmetric
transformation. In general,
the system is highly tolerant
to an array of functionalities
and substitution patterns
(Tables 1 and 2). A variety
of a-substituted 2-carboxy-
allylcyclohexanones can be
converted into products with
high enantiopurity in excel-
lent yields under the condi-
tions for decarboxylative
allylation (Table 1). The
presence of enolizable func-
tional groups (entries 4 and
5) and b-heteroatom substi-
tution (entry 9) is of partic-
ular note. Furthermore, the
2-fluoro derivative (entry 10)
allows for the high-yielding
synthesis of stereodefined
tertiary fluorides.
Scheme 2. Nonselective enol formation. R=anion.
Scheme 3. a) Deuterium-labeling experiment with enol carbonate 8. b) Crossover experiment with deuterium-
labeled enol carbonates 8 and 9.
synthesis of 6 (namely, 4; R = anion) could be derived through
decarboxylation of a racemic b-ketoester precursor, such as
(ꢁ )-12 (Scheme 4).[14] Moreover, this route would solve the
enolization problem outlined above because the position-
In addition to changes at the a position, variation of the
cyclic and allylic portions of the quaternary b-ketoester is
tolerated (Table 2). Examples in which the cyclohexane ring
has been substituted (entries 1–4), unsaturated (entry 5),
appended
(entry 6),
or
enlarged (entry 7) proceed
to products in high yield and
enantioselectivity.
substrates with
Finally,
internal
allylic substitution (entries 8
and 9) and a racemic hetero-
cycle (entry 10) also afford
products in excellent yield
and enantioselectivity.
With a general procedure
for enantioconvergent decar-
boxylative allylation in hand
Scheme 4. b-Ketoester synthesis and enantioconvergent decarboxylative allylation. LDA=lithium diisopropyl-
amide.
selective a-alkylation of ketoester 11 would be facile. We
prepared b-ketoester (ꢁ )-12 by acylation and alkylation of
enone 2 to test the reaction (Scheme 4).[15] To our delight,
treatment of (ꢁ )-12 with the complex derived from [Pd2-
(dba)3] and (S)-tBu-phox ligand 1 in Et2O exclusively
produced allyl ketone (ꢀ)-6 in 73% yield and with 86% ee.
Importantly, b-ketoester 12 remained racemic throughout the
course of the reaction, thus indicating that significant kinetic
resolution of the starting material had not occurred. This
and given our interest in tandem reactions and catalysis,[16] we
designed a substrate that could undergo a double allylation
cascade to afford a product that possesses two all-carbon
quaternary stereocenters. Thus, we prepared (ꢁ )-13, a
substrate that contains a reactive enol carbonate and a
latent b-ketoester moiety (Scheme 5).[12] Treatment of (ꢁ )-13
under our catalytic asymmetric conditions provided the
double alkylation product in 76% yield as a 4:1 mixture of
C2/meso diastereomers.[17] Gratifyingly, the major diaster-
Angew. Chem. Int. Ed. 2005, 44, 6924 –6927ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 6925

Communications
Table 1: Catalytic enantioconvergent decarboxylative allylation of a-substituted 2-carboxyallylcyclohexanones.
Entry
R
Solvent
T [8C]
t [h]
Yield [%][a]
ee [%][b]
1
2
3
CH3
CH3
prenyl
CH2CH2CN
CH2CH2CO2Et
CH2C6H5
CH2(4-CH3OC6H4)
CH2(4-CF3C6H4)
CH2OTBDPS
F
THF
Et2O
Et2O
Et2O
Et2O
THF
THF
THF
THF
Et2O
25
25
30
25
25
25
25
25
25
30
7.5
4.75
6
6.5
6
0.5
10
0.5
5
3.5
85
89
97
97
96
99
80
99
86
80
88
88
91
88
90
85
86
82
81
91
4
5[c]
6
7
8
9[c]
10
[a] Yield of isolated product from reaction of 1.0 mmol substrate at 0.033m in solvent, unless otherwise noted. [b] Determined by chiral GC or HPLC.[12]
[c] Conditions: 4 mol% [Pd2(dba)3], 10 mol% (S)-tBu-phox (1), 0.021m. phox=phosphinooxazoline; TBDPS=tert-butyldiphenylsilyl.
Table 2: Enantioconvergent decarboxylative allylation of b-ketoesters.
Entry
Substrate
Product
T [8C]
t [h]
Yield [%][a]
ee [%][b]
1
25
25
1.5
24
94
94
85
86
2[c]
3
4
30
25
9
5
89
90
90
85
5[d,e]
6[d]
7
30
25
25
35
35
4
77
97
83
87
87
90
92
87
92
91
10
9.5
6.5
2.5
8[d]
9[d,e]
10
25
2.5
91
92
[a] Yield of isolated product from reaction of 1.0 mmol substrate, 2.5 mol% [Pd2(dba)3], and 6.25 mol% (S)-tBu-phox (1) at 0.033m in THF, unless
otherwise noted. [b] Determined by chiral GC or HPLC.[12] [c] Conditions: 25 mmol substrate, 1.5 mol% [Pd2(dba)3], and 3.75 mol% (S)-tBu-phox (1).
[d] Performed in Et2O. [e] Conditions: 4 mol% [Pd2(dba)3] and 10 mol% (S)-tBu-phox (1) at 0.021m. Bn=benzyl.
6926 www.angewandte.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2005, 44, 6924 –6927

Angewandte
Chemie
Angew. Chem. 2005, 117, 2299 – 2301;
Angew. Chem. Int. Ed. 2005, 44, 2259 –
2261.
[7] D. C. Behenna, B. M. Stoltz, J. Am. Chem.
Soc. 2004, 126, 15044 – 15045.
[8] a) J. Tsuji, I. Minami, Acc. Chem. Res. 1987,
20, 140 – 145; b) J. Tsuji, Proc. Jpn. Acad.
Ser. B 2004, 80, 349 – 358.
[9] a) G. Helmchen, A. Pfaltz, Acc. Chem. Res.
2000, 33, 336 – 345; b) J. M. J. Williams,
Synlett 1996, 705 – 710, and references
therein.
Scheme 5. Enantioselective allylation cascade generating two quaternary carbon stereo-
centers.
[10] Subsequent to our initial report, a similar
eomer (ꢀ)-14 was obtained with 92% ee and bears two
transformation using a bisphosphine ligand was reported, see:
B. M. Trost, J. Xu, J. Am. Chem. Soc. 2005, 127, 2846 – 2847.
[11] An elegant alternative catalytic asymmetric alkylation of cyclic
ketones for the synthesis of quaternary stereocenters has been
recently described, see: A. G. Doyle, E. N. Jacobsen, J. Am.
Chem. Soc. 2005, 127, 62 – 63.
quaternary stereocenters set by a single catalytic asymmetric
transformation.
In conclusion, we have developed the first example of a
catalytic enantioconvergent synthesis of quaternary stereo-
centers from racemates with quaternary stereocenters. The
decarboxylative allylation reaction described herein is an
example of an enantioconvergent process in which the same
[12] See Supporting Information for details.
[13] This hypothesis has been experimentally supported for non-
enantioselective systems, see: a) J. Tsuji, T. Yamada, I. Minami,
M. Yuhara, M. Nisar, I. Shimizu, J. Org. Chem. 1987, 52, 2988 –
2995; b) J.-F. Detalle, A. Riahi, V. Steinmetz, F. Hꢀnin, J. Muzart,
J. Org. Chem. 2004, 69, 6528 – 6532.
[14] For an excellent review on metal-catalyzed decarboxylation, see:
J. A. Tunge, E. C. Burger, Eur. J. Org. Chem. 2005, 9, 1715 – 1726.
[15] a) L. N. Mander, S. P. Sethi, Tetrahedron Lett. 1983, 24, 5425 –
5428; b) D. M. X. Donnelly, J.-P. Finet, B. A. Rattigan, J. Chem.
Soc. Perkin Trans. 1 1993, 1729 – 1735.
ꢀ
catalyst is intimately involved in both the stereoablative (C C
ꢀ
bond-breaking) and stereoselective (C C bond-forming)
steps. The availability of racemic b-ketoesters of the type
employed in this study greatly increases the practical value of
these Pd-catalyzed allylations for the enantioselective syn-
thesis of cyclic ketones with one or more quaternary stereo-
centers. The utility of this methodology for enantioselective
synthesis of complex targets is currently under investigation.
[16] a) J. A. May, B. M. Stoltz, J. Am. Chem. Soc. 2002, 124, 12426 –
12427; b) R. Sarpong, J. T. Su, B. M. Stoltz, J. Am. Chem. Soc.
2003, 125, 13624 – 13625.
Received: June 11, 2005
Published online: October 5, 2005
[17] When PPh3 was used as the ligand for the conversion of (ꢁ )-13
into (ꢁ )-14 and the meso isomer of 14 (not shown),
a
diastereomeric ratio of 70:30 was observed.
Keywords: allylation · asymmetric catalysis · enolates ·
.
palladium
[1] For reviews, see: a) R. Noyori, M. Tokunaga, M. Kitamura, Bull.
Chem. Soc. Jpn. 1995, 68, 36 – 56; b) H. Stecher, K. Faber,
Synthesis 1997, 1 – 16; c) F. F. Huerta, A. B. E. Minidis, J.-E.
Bäckvall, Chem. Soc. Rev. 2001, 30, 321 – 331; d) H. Pellissier,
Tetrahedron 2003, 59, 8291 – 8327.
[2] For representative stereomutative examples, see: a) Ref. [1];
b) A. Pfaltz, M. Lautens in Comprehensive Asymmetric Catal-
ysis, Vol. 2 (Eds: E. N. Jacobsen, A. Pfaltz, H. Yamamoto),
Springer, New York, 1999, pp. 833 – 884; c) B. M. Trost, R. C.
Bunt, R. C. Lemoine, T. L. Calkins, J. Am. Chem. Soc. 2000, 122,
5968 – 5976.
[3] The term stereoablation describes the conversion of a chiral
molecule to an achiral molecule (e.g., Figure 1, pathway II, (ꢁ )-
A!C). Ablation from the Oxford English Dictionary meaning
“the action or process of carrying away or removing; removal”.
[4] For representative stereoablative examples, see: a) T. Hamada,
A. Chieffi, J. Šhman, S. L. Buchwald, J. Am. Chem. Soc. 2002,
124, 1261 – 1268; b) N. Mase, F. Tanaka, C. F. Barbas III, Angew.
Chem. 2004, 116, 2474 – 2477; Angew. Chem. Int. Ed. 2004, 43,
2420—2423; c) C. Fischer, G. C. Fu, J. Am. Chem. Soc. 2005, 127,
4594 – 4595; d) D. Magdziak, G. Lalic, H. M. Lee, K. C. Fortner,
A. D. Aloise, M. D. Shair, J. Am. Chem. Soc. 2005, 127, 7284 –
7285.
[5] To the best of our knowledge, there are no examples of the
catalytic asymmetric synthesis of quaternary carbon stereo-
centers from racemates with quaternary stereocenters.
[6] A diastereoselective example of such a process has been recently
reported, see: K. Xu, G. Lalic, S. M. Sheehan, M. D. Shair,
Angew. Chem. Int. Ed. 2005, 44, 6924 –6927ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 6927