Hydrogen-bonding asymmetric metal catalysis with α-amino acids: A simple and tunable approach to high enantioinduction

Article abstract of DOI:10.1021/ja904185b

(Chemical Equation Presented) While asymmetric transition-metal catalysis has become a powerful method for constructing chiral products, a challenge in this field is the identification of the correct ligand for high selectivity. We report here a simple ap

Full text of DOI:10.1021/ja904185b

Published on Web 07/24/2009
Hydrogen-Bonding Asymmetric Metal Catalysis with r-Amino Acids: A Simple
and Tunable Approach to High Enantioinduction
Yingdong Lu, Tim C. Johnstone, and Bruce A. Arndtsen*
Department of Chemistry, McGill UniVersity, 801 Sherbrooke Street West, Montreal, QC H3A 2K6, Canada
Received May 22, 2009; E-mail: bruce.arndtsen@mcgill.ca
Scheme 1. Proposed Mechanism for Imine/Alkyne Coupling
Asymmetric transition-metal catalysis has become a major area of
research in synthetic chemistry.1 One challenge in this field, however,
is how to identify the correct chiral ligand for high enantioselectivity.
The subtle nature of asymmetric catalysis can make it difficult to predict
the optimal ligand to use in a particular reaction.2 This presents a
significant issue, since the ligands found to lead to high enantioselec-
tivity are often themselves complex and typically require a multistep
synthesis prior to screening and tuning. Both others and ourselves have
been actively pursuing alternatives to chiral ligands in asymmetric
catalysis (e.g., chiral counteranions,3 Bronsted acids,4,5 chiral environ-
ments6), though these can again require synthesizing chiral scaffolds
for high enantioselectivity. An alternative would be to employ naturally
occurring sources of chirality in metal catalysis without structural
modification, such as R-amino acid derivatives. The latter are inex-
pensive and commercially available in a variety of forms. The difficulty
is how to associate such units to catalysts in a manner that can lead to
high enantioinduction without, as is often the case, ultimately moving
beyond the natural chiral pool to tune their selectivity.7,8
In considering this issue, one possibility would be to change how
amino acids are used in metal catalysis. There exist a range of metal-
catalyzed reactions with substrates that can interact with Bronsted acids.
As such, we postulated that coupling metal catalysis with the ability
of amino acids to hydrogen bond could provide an easy route for
inducing both enantioselectivity (via the amino acid) and the tunability
often required to obtain high selectivity (by changes to the metal
catalyst) (Figure 1). Bronsted acid catalysts have recently attracted
significant attention,5 including in metal catalysis,4 though these latter
typically employ chiral phosphoric acid scaffolds. In contrast, we
describe herein how simple amino acid hydrogen bonding can provide
a method for introducing chirality into metal catalysis. In addition,
the modularity of this system makes tuning it for high enantioselectivity
a straightforward process. This includes the potential for designing
substrate-specific chiral catalysts.
amino acid Fmoc-valine was examined. As hoped, a similar accelera-
tion was observed, and 2a was formed in 49% ee.
The above demonstrates that an amino acid can catalyze the coupling
of imines and alkynes, albeit with only moderate selectivity. However,
it is straightforward to exploit the advantage of this system: its
tunability. Changing the commercial amino acid is trivial and can
significantly influence the enantioselectivity. For example, histidine
and asparagine led to <25% ee, while several other derivatives
increased the selectivity relative to valine. A single day of screening
allowed N-Boc-proline to be identified as the optimal amino acid for
the reaction (61% ee).
In addition to the amino acid, the copper catalyst can also be tuned,
as a route for modulating the selectivity beyond that provided by
available amino acid derivatives. A variety of ligands were found to
influence the enantioselectivity (Table 1), likely by changing the steric
bulk of the copper catalyst for reaction with 1. Overall, the use of
P(o-tolyl)3 with N-Boc-proline leads to 2a in high yield and 96% ee
(entry 11) (Figure 2).
Figure 1. Hydrogen-bonding amino acids in metal catalysis.
The reaction we examined is the copper-catalyzed coupling of
alkyne and imine. This has been found to form propargylamines in
good enantioselectivity with synthetic ligands (e.g., Py-BOX deriva-
tives).9 Our previous research has shown that imines can be activated
toward metal-catalyzed coupling reactions by in situ generation of
N-acyliminium salts.10 As such, it seemed viable that this reaction
might proceed more rapidly upon hydrogen bonding to the imine to
create the electrophile 1 (Scheme 1).11 This effect is illustrated in eq
1. While CuPF6 catalyst alone leads to 2a in low yield, the addition of
a mild Bronsted acid (10% PhCO2H) results in near-quantitative 2a
formation. In view of the catalytic influence of carboxylic acids, the
Figure 2. N-Boc-protected amino acid influence on enantioselectivity.
Superscripts a, b, c, and d indicate O-benzyl, O-t-Bu, N-tosyl, and N-xan,
respectively.
Preliminary studies were consistent with R-amino acid/imine
association during catalysis. 1H NMR analysis of N-Boc-proline with
(p-tolyl)HCdN(Bn) showed a significant downfield shift in the -CO2H
resonance (from 11.58 to 14.66 ppm), as anticipated for hydrogen
bonding. Titration studies provided a Kassoc of 14 M-1. Consistent with
this weak, equilibrium association to form 1 during catalysis, kinetic
9
11284 J. AM. CHEM. SOC. 2009, 131, 11284–11285
10.1021/ja904185b CCC: $40.75  2009 American Chemical Society

C O M M U N I C A T I O N S
Table 1. Ligands in CuPF6/N-Boc-Proline-Catalyzed Coupling (eq 1)
with even seemingly small changes leading to loss in enantioselectivity
and the potential need to redesign the chiral ligand. An example is
with 1-hexyne (2j), which led to product in 80% ee, consistent with
reports of low selectivity with alkyl alkynes.9 However, because the
availability of amino acids and phosphines, it was straightforward to
identify the correct ligand for this system (i.e., PCy3). This modularity
can also allow the expansion of this reaction to new classes of imines,
such as C-alkylimines, which in this case provided 2k in 91% ee with
P(1-napthyl)3 as the ligand. As far as we are aware, this represents the
most general and easily tunable catalytic system for the synthesis of
optically active propargylamines.
entry
L (10 mol %)
yield (%)
% eea
1
2
3
4
5
6
7
8
9
-
94
95
78
89
88
92
96
52
48
96
89
61
71
80
81
73
81
85
65
80
93
96b
PPh3
PBu3
P(OPh)3
P(O(2,4,6-(t-Bu)3C6H2))3
P(cyclohexyl)3
P(1-napthyl)3
P(t-Bu)2(2-N-phenyllpyrrole)
P(t-Bu)2(o-biphenyl)
P(o-tolyl)3 (20%)
P(o-tolyl)3
10
11
In conclusion, these results suggest what is to our knowledge a new
approach for incorporating chirality into metal catalysis, namely, the
use of hydrogen-bonding amino acid derivatives with metal catalysts.
In contrast to complex chiral ligands, these catalysts are accessible
and commercially available, and their modularity can be used to create
a large number of different catalysts by using different members of
the available pools of amino acids and phosphines, with screening often
limited only by the rate of HPLC analysis. This provides a facile
approach for tuning chiral catalysts for high enantioselectivity, even
for specific substrates. Experiments directed toward determining the
applicability of this approach to other catalytic reactions are currently
underway.
a Determined using a Chiral Pak AD-H column. b Conditions: 2.5%
CuPF6/5% L, 0 °C, 3 days.
experiments showed the rate to have a first-order dependence on the
N-Boc-proline concentration (from 3 to 50 mol %) and to be
independent of the CuPF6 concentration (10 mol %).12 Notably, no
change in % ee was observed over this range. Alternatively, the
addition of base (20% NEt3) completely inhibited catalysis, presumably
by blocking this Bronsted acid activation.
From a synthetic perspective, this combined amino acid/copper
catalyst provides a simple system for synthesizing a range of
propargylamines with high enantioselectivity. This includes coupling
with various N- and C-aryl imines (2b-h) as well as vinyl, alkyl, and
functionalized alkynes, each forming 2 in high yield and up to 99%
ee (Table 2). The accelerating influence of the amino acid also allowed
the use of substrates previously considered incompatible with coupling.
For example, despite considerable efforts,9 electron-rich N-alkylimines
have been reported as unreactive toward alkynylation, likely because
of their reduced electrophilicity. In contrast, N-Boc proline activates
these toward coupling, forming N-protected 2n in 93% ee.
Acknowledgment. The authors thank the NSERC (Canada)
Discovery and AGENO programs for support of this research.
Supporting Information Available: Synthesis and characterization
of 2 and kinetic/titration data. This material is available free of charge via
the Internet at http://pubs.acs.org.
References
(1) (a) Jacobsen, E. N.; Pfaltz, A.; Yamamoto, H. ComprehensiVe Asymmetric
Catalysis; Springer: Heidelberg, Germany, 1999. (b) Walsh, P. J.; Ko-
zlowski, M. C. Fundamentals in Asymmetric Catalysis; USB: Sausalito,
CA, 2008.
Perhaps the most significant advantage of this approach is its
flexibility. Asymmetric catalysis is notorious for its substrate specificity,
(2) For example: Yoon, T. P.; Jacobsen, E. N. Science 2003, 299, 1691.
(3) (a) Llewellyn, D. B.; Adamson, D.; Arndtsen, B. A. Org. Lett. 2000, 2,
4165. (b) Dorta, R.; Shimon, L.; Milstein, D. J. Organomet. Chem. 2004,
689, 751. (c) Hamilton, G. L.; Kang, E. J.; Mba, M.; Toste, F. D. Science
2007, 317, 496. (d) Mukherjee, S.; List, B. J. Am. Chem. Soc. 2007, 129,
11336. (e) Lacour, J.; Hebbe-Viton, V. Chem. Soc. ReV. 2003, 32, 373. (f)
Llewellyn, D. B.; Arndtsen, B. A. Organometallics 2004, 23, 2838.
(4) (a) Rueping, M.; Antonchick, A. P.; Brinkmann, C. Angew. Chem., Int.
Ed. 2007, 46, 6903. (b) Li, C.; Wang, C.; Villa-Marcos, B.; Xiao, J. J. Am.
Chem. Soc. 2008, 130, 14450.
Table 2. Diversity of Asymmetric Imine Alkynylationa
(5) Review: Doyle, A. G.; Jacobsen, E. N. Chem. ReV. 2007, 107, 5713.
(6) (a) Brunkan, N. M.; Gagne, M. R. J. Am. Chem. Soc. 2000, 122, 6217. (b)
Boersma, A. J.; Feringa, B. L.; Roelfes, G. Angew. Chem., Int. Ed. 2009,
48, 3346. (c) Walsh, P. J.; Balsells, J. J. Am. Chem. Soc. 2000, 122, 1802.
(7) Examples of amino acid ligands: (a) Shi, B.-F.; Maugel, N.; Zhang, Y.-H.;
Yu, J.-Q. Angew. Chem., Int. Ed. 2008, 47, 4882. (b) Ma, D.; Cai, A. Acc.
Chem. Res. 2008, 41, 1450. (c) Paradowska, J.; Stodulski, M.; Mlynarski,
J. Angew. Chem., Int. Ed. 2009, 48, 4288.
(8) (a) Cole, B. M.; Shimizu, K. D.; Krueger, C. A.; Harrity, J. P. A.; Snapper,
M. L.; Hoveyda, A. H. Angew. Chem., Int. Ed. 1996, 35, 1688. (b)
Hargaden, G. C.; Guiry, P. J. Chem. ReV. 2009, 109, 2505.
(9) (a) Wei, C.; Mague, J. T.; Li, C.-J. Proc. Natl. Acad. Sci. U.S.A. 2004,
101, 5749. (b) Wei, C.; Li, C.-J. J. Am. Chem. Soc. 2002, 124, 5638. (c)
Bisai, A.; Singh, V. K. Org. Lett. 2006, 8, 2405. (d) Liu, J.; Liu, B.; Jia,
X.; Li, X.; Chan, A. S. C. Tetrahedron: Asymmetry 2007, 18, 396. (e)
Colombo, F.; Benaglia, M.; Simonetta, O.; Usuelli, F.; Celentano, G. J.
Org. Chem. 2006, 71, 2064. (f) Irmak, M.; Boysen, M. M. K. AdV. Synth.
Catal. 2008, 350, 403.
(10) (a) Black, D. A.; Arndtsen, B. A. Org. Lett. 2004, 6, 1107. (b) Black, D. A.;
Arndtsen, B. A. Org. Lett. 2006, 8, 1991. (c) Fischer, C.; Carreira, E. M.
Org. Lett. 2004, 6, 1497. (d) Gommermann, N.; Koradin, C.; Polborn, K.;
Knochel, P. Angew. Chem., Int. Ed. 2003, 42, 5763.
(11) This effect has recently been postulated with synthetic phosphate coun-
teranions and protonated C-ester-substituted imines.4a
.
(12) While amino acid/copper secondary coordination is possible, the catalysis
is zeroth-order in [CuPF6], also suggesting separate roles for Cu and the
amino acid in the reaction (see the Supporting Information). Notably, other
H-bonding acids can also lead to enantioselectivity, though at lower ee’s
[(S)-mandelic acid, 19% ee; L-2-pyrrolidone-5-carboxylic acid, 0% ee].
(13) See the Supporting Information for configuration assignment (for 2b and
2g, see ref 9c).
a See the Supporting Information for conditions. b O.R. ) optical
rotation.13 c Room temperature.
JA904185B
9
J. AM. CHEM. SOC. VOL. 131, NO. 32, 2009 11285