Enantioselective construction of highly substituted pyrrolidines by palladium-catalyzed asymmetric [3+2] cycloaddition of trimethylenemethane with ketimines

Article abstract of DOI:10.1021/ja102102d

The transition-metal catalyzed trimethylenemethane [3+2] cycloaddition provides a direct route to functionalized heterocycles. Herein, we describe a protocol for the reaction between 1-cyano-2-((trimethylsilyl)methyl)allyl acetate and a series of ketimines to generate highly substituted pyrrolidines. This methodology showcases a catalytic, asymmetric addition of a carbon nucleophile to ketimines, examples of which are still rare. The corresponding pyrrolidines were obtained in excellent yields and selectivities making use of our novel phosphoramidite ligands L2 - L3 .

Full text of DOI:10.1021/ja102102d

Published on Web 05/26/2010
Enantioselective Construction of Highly Substituted Pyrrolidines by
Palladium-Catalyzed Asymmetric [3+2] Cycloaddition of Trimethylenemethane
with Ketimines
Barry M. Trost* and Steven M. Silverman
Department of Chemistry, Stanford UniVersity, Stanford, California 94305-5080
Received March 11, 2010; E-mail: bmtrost@stanford.edu
Table 1. Selected Optimization Studiesa
The transition-metal catalyzed [3+2] trimethylenemethane (TMM)
cycloaddition reaction is a versatile method for the chemo-, regio-,
and diastereoselective construction of highly substituted five-
membered rings.1 Utilization of Pd-TMM complexes derived from
3-acetoxy-2-trimethylsilylmethyl-1-propene has resulted in the
efficient catalytic syntheses of carbocycles2 and heterocycles.3
entry
R
Ligand T, °C
% yield
3/4
3 % ee 4 % ee
Although this methodology was disclosed over 30 years ago by
our laboratory,4 a general asymmetric variant of the cycloaddition
remained elusive until 2006.5 Using a new class of chiral phos-
phoramidite ligands, we were able to affect the efficient syntheses
of cyclopentanes,5,6 pyrrolidines,7 and bicyclo[4.3.1]decadienes8
through reaction of olefins, aldimines, and tropones respectively.9
While catalytic, enantioselective addition of carbon nucleophiles
to aldimines is well precedented, the corresponding additions to
ketimines are rare.10 The intrinsic lower reactivity of ketimines due
to both electronic and steric factors undoubtedly plays a significant
role. In addition to the opportunity for novel selectivity afforded
by our phosphoramidite ligands, we also observed a significant
increase in reactivity of various substrate classes. Many cycload-
ditions requiring elevated temperatures and high catalyst loading
in the racemic reaction were found to proceed under milder
conditions, requiring as little as 1 mol % metal at low temperatures.
This enhanced reactivity inspired us to examine the asymmetric
cycloaddition to ketimines, despite the relatively rare and specialized
nature of reactive ketimines in the achiral TMM reaction.3b Such
a reaction (eq 1, R1, R2 * H) would provide a novel route to highly
substituted pyrrolidines containing a tetrasubstituted center.
1
2
3
4
5
6
7
8
4-MeOC6H4 (2a) L1
50
50
50
0
0
0
Bn (2b)
OCH3 (2c)
P(O)Ph2 (2d)
Ts (2e)
Ts
L1
L1
L1
L1
L1
L2
L3
50 Complex
50 24b
0:1
1:1.1
2.2:1
-
86
95
53
63
89
-
4
4
4
67
79
91
Ts
Ts
>20:1 >99
a All reactions were performed at 0.2
M in toluene with 5%
CpPd(η3-C3H5), 10% ligand, and 1.5 equiv of 1 and stirred for 4 h.
Yields are isolated values; ee’s were determined by chiral HPLC.
b Mixture also contained 53% yield of the endocyclic double bond with
a 41% ee.
Figure 1. Phosphoramidite ligands examined.
clear that ligand optimization would be required. Azetidine ligand
L2 (Figure 1) provided a sizable increase in enantioselectivity,
although the diastereoselectivity remained low. Our previously
introduced bis-2-naphthyl phosphoramidite L3 solved both the
enantio- and diastereoselectivity issues, providing the desired
product in excellent yield with a >20:1 dr and a >99% ee. Notably,
while N-Boc imines derived from aromatic aldehydes performed
admirably in previous studies, they were ineffective in this case
due to their propensity to isomerize to the enamine tautomer.
We next wanted to examine the reaction scope with respect to
the imine (Table 2). Gratifyingly, the reaction proved very general
under the optimized conditions, performing well in every case
examined. Yields and selectivities proved insensitive to the steric
bulk of the aromatic substituent (entry 2), aliphatic substituent (entry
3), substitution pattern of the aromatic ring (entries 4-6), or
electronic nature of the substituent. Spirocycles could be formed
in good to excellent diastereoselectivity and excellent enantiose-
lectivity (entries 7-8). Heterocycles were tolerated (entry 9), albeit
with moderately decreased enantioselectivity, and equivalently
substituted aliphatic ketimines performed well (entry 10). Conver-
We began our studies with the examination of the Pd-catalyzed
[3+2] cycloaddition of our cyano-TMM donor 1 with a series of
ketimines 2a-e (Table 1). Using CpPd(η3-C3H5) and Feringa ligand
L1 (Figure 1), we were discouraged by the fact that PMP-imine
2a proved completely unreactive under the conditions utilized, as
the corresponding aldimine class had performed well using our
parent donor (eq 1, R4 ) H).7 Equally frustrating, no reactivity
was observed with benzyl imine 2b or oxime ether 2c. Diphe-
nylphosphinoyl imine 2d provided a complex mixture. Analysis
of the crude mixture by NMR indicated the presence of the desired
product 3d and its tautomer containing an endocyclic olefin
conjugated with the nitrile; however purification proved problematic.
These results indicated that a strongly electron-withdrawing group
was necessary for reactivity. Gratifyingly, using N-tosyl imine 2e
we observed formation of the desired cycloadduct, albeit as a
mixture of 4e and its endocyclic olefin tautomer. Reduction of the
temperature prevented isomerization and increased the enantiose-
lectivity, but the diastereoselectivity remained poor and it became
9
8238 J. AM. CHEM. SOC. 2010, 132, 8238–8240
10.1021/ja102102d  2010 American Chemical Society

C O M M U N I C A T I O N S
Table 3. Scope of the Aliphatic Imine Cycloadditiona
Table 2. Initial Scope of the Imine Cycloadditiona
a All reactions were performed at 0.2
M in toluene with 5%
CpPd(η3-C3H5), 10% ligand, and 1.5 equiv of 1 and stirred for 2-4 h.
Yields are combined, isolated values; ee’s were determined by chiral
HPLC. b Using Ligand L3. c Using Ligand L1.
the bottom conformer, which is highly disfavored due to severe
steric interactions between the tosyl group and bulky π-allyl.
Nonequivalently substituted aliphatic ketimines presented a
selectivity problem when subjected to the standard conditions (Table
3, entry 1). Using L3 conversion remained high with the cyclohexyl
imine, but the diastereoselectivity dropped below acceptable levels.
Ligand L1 (entry 2) restored the diastereoselectivity, albeit at the
cost of conversion and enantioselectivity. Ligand L2 provided a
nice compromise, providing the desired product in good yield, 10:1
diastereoselectivity, and restored enantioselectivity. A more remote
branch point could be tolerated while maintaining selectivity as
shown by the isobutyl and cyclohexyl imines (entries 4-5).
In summary, we have demonstrated a palladium-catalyzed
asymmetric addition of a carbon nucleophile to ketimines in the
context of a [3+2] TMM cycloaddition, providing highly substituted
pyrrolidines. This transformation showcases novel reactivity and
selectivity that has not been previously observed. We have been
able to affect high enantio- and diastereomeric excesses by matching
substrate classes with the appropriate ligand. Investigations into
the full imine scope and further transformations are ongoing and
will be reported in due course.
a All reactions were performed at 0.2
M in toluene with 5%
CpPd(η3-C3H5), 10% ligand, and 1.5 equiv of 1 and stirred for 2-4 h.
Yields are combined, isolated values; ee’s were determined by chiral
HPLC. b Reaction performed at -15 °C. c Minor diastereomer.
Acknowledgment. We thank the NSF for their generous support
of our programs. S.M.S. thanks Eli Lilly and Roche for graduate
fellowships. We thank Dr. Allen Oliver from the University of Notre
Dame for the X-ray crystal structures and Johnson-Matthey for
generous gifts of palladium salts.
Supporting Information Available: Experimental details and
spectral data for all unknown compounds. This material is available
free of charge via the Internet at http://pubs.acs.org.
Figure 2. Proposed model for reaction diastereoselectivity.
References
sion remained high at low temperatures, in some cases as low
as -15 °C (entry 7). An X-ray crystal structure analysis of 9
(entry 6) unambiguously allowed the determination of absolute
stereochemistry.
(1) (a) Trost, B. M. Angew. Chem., Int. Ed. Engl. 1986, 25, 1. (b) Trost, B. M.
Pure Appl. Chem. 1988, 60, 1615. (c) Lautens, M.; Klute, W.; Tam, W.
Chem. ReV. 1996, 96, 49. (d) Yamago, S.; Nakamura, E. Org. React. 2002,
61, 1.
(2) Chan, D. M. T. Recent Advances in Palladium-Catalyzed Cycloadditions
Involving Trimethylenemethane and its Analogs. In Cycloaddition Reactions
in Organic Synthesis; Kobayashi, S., Jorgensen, K. A., Eds.; Wiley-VCH:
Weinheim, Germany, 2002; pp 57-83.
Our proposed model for the diastereoselectivity of the reaction
is shown in Figure 2 and is based on the imine geometry. We
suggest that the initial bond forming event occurs as depicted from
the top conformer, leading to the observed, thermodynamically less
favored diastereomer.11 The other diastereomer would arise from
(3) (a) Trost, B. M.; King, S. A.; Schmidt, T. J. Am. Chem. Soc. 1989, 111,
5902. (b) Trost, B. M.; Marrs, C. M. J. Am. Chem. Soc. 1993, 115, 6636.
(4) Trost, B. M.; Chan, D. M. T. J. Am. Chem. Soc. 1979, 101, 6429.
(5) Trost, B. M.; Stambuli, J. P.; Silverman, S. M.; Schwo¨rer, U. J. Am. Chem.
Soc. 2006, 128, 13328 Hayashi reported a single account of an asymmetric
9
J. AM. CHEM. SOC. VOL. 132, NO. 24, 2010 8239

C O M M U N I C A T I O N S
TMM reaction: Yamamoto, A.; Ito, Y.; Hayashi, T. Tetrahedron Lett. 1989,
(9) Marquand, P. L.; Tam, W. Angew. Chem., Int. Ed. 2008, 47, 2926.
30, 375.
(10) (a) Vilaivan, T.; Bhanthumnavin, W.; Sritana-Anant, Y. Curr. Org. Chem.
(6) Trost, B. M.; Cramer, N.; Silverman, S. M. J. Am. Chem. Soc. 2007, 129,
2005, 9, 873. (b) Shibasaki, M.; Motomu, K. Chem. ReV. 2008, 108, 2853.
12396.
(11) Semiempirical AM1 calculations performed using Spartan indicate an energy
(7) Trost, B. M.; Silverman, S. M.; Stambuli, J. P. J. Am. Chem. Soc. 2007,
129, 12398.
difference of 1.09 kcal/mol with epi-7 being the more stable isomer.
(8) Trost, B. M.; McDougall, P. J.; Hartmann, O.; Wathen, P. T. J. Am. Chem.
Soc. 2008, 130, 14960.
JA102102D
9
8240 J. AM. CHEM. SOC. VOL. 132, NO. 24, 2010