Second-generation organocatalysts for the highly enantioselective dynamic kinetic resolution of azlactones

Article abstract of DOI:10.1039/b418666d

Bifunctional organocatalysts of the thiourea-tert-amine type, carrying two "matched" elements of chirality, effect the alcoholytic dynamic kinetic resolution of a variety of azlactones with up to 95% ee. The Royal Society of Chemistry 2005.

Full text of DOI:10.1039/b418666d

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Second-generation organocatalysts for the highly enantioselective
dynamic kinetic resolution of azlactones
Albrecht Berkessel,* Santanu Mukherjee, Felix Cleemann, Thomas N. Mu¨ller and Johann Lex
Received (in Cambridge, UK) 14th December 2004, Accepted 26th January 2005
First published as an Advance Article on the web 15th February 2005
DOI: 10.1039/b418666d
Bifunctional organocatalysts of the thiourea-tert-amine type,
carrying two ‘‘matched’’ elements of chirality, effect the
alcoholytic dynamic kinetic resolution of a variety of azlactones
with up to 95% ee.
Enantiomerically pure natural and non-natural a-amino acids find
numerous applications in the synthesis of chiral ligands, catalysts,
pharmaceuticals, peptides and many other valuable target
molecules. Therefore, the development of practical methods for
the enantioselective synthesis of a-amino acids is a challenging
task, in particular for non-natural amino acids.1 Besides
asymmetric synthesis, the dynamic kinetic resolution (DKR) of
suitable racemic precursors is a very practical avenue leading to
Scheme 2 Urea/thiourea-based bifunctional organocatalysts.
the desired enantiomerically pure product in (theoretically)
quantitative yield.2
(besides the chiral diamine moiety). As it turned out, excellent
In the area of a-amino acids, the alcoholytic dynamic kinetic
enantioselectivities were achieved even at room temperature.
resolution of azlactones 1 is an attractive way to generate the
Our previous work was based on urea/thiourea catalysts
enantiomerically enriched and N,C-doubly protected a-amino acid
carrying an electron-poor aromatic moiety on the urea nitrogen
atom (as in 3a).4 In the first stage of the current study, we
derivatives 2 (Scheme 1). Azlactones racemize readily, but only a
few examples of chemically catalyzed DKRs of azlactones appear
to exist.3 The planar-chiral DMAP-derivatives by Fu et al.
aliphatic (3b, c) or chiral benzylic substituents (4a, b) on the urea/
represented the most efficient catalysts, affording a maximum ee
of 78%, albeit at the expense of very long reaction times.3c
examined a series of bifunctional organocatalysts bearing simple
thiourea nitrogen atom. The DKR of the phenylalanine-derived
azlactone 1a served as the test reaction (Table 1). To our surprise,
Recently, we applied (inter alia) the (thio)urea-based bifunc-
we found that the methyl substituted catalyst 3b showed
tional organocatalyst 3a for this purpose and achieved significantly
comparable reactivity and higher enantioselectivity for the ring
improved enantioselectivities, e.g. 91% ee at 220 uC in the reaction
of the tert-leucine-derived azlactone (1, R 5 t-Bu, Scheme 1) with
allyl alcohol.4 NMR-spectroscopic studies indicated that the
Table 1 Optimization of catalyst structure
catalyst activates the azlactone by hydrogen bonding of the
quasi-Lewis-acidic (thio)urea moiety to the carbonyl oxygen
atom.4 The alcohol nucleophile is activated and steered by the
Brønsted-basic tertiary amine. Here we report the use of the
catalysts 3b–8 (Scheme 2) in the dynamic kinetic resolution of
azlactones. With the exception of 3b and 3c, these second
generation catalysts harbour an additional element of chirality
Entry
Catalyst
Conversion (%)
eea (%)
1
2
3
4
5
6
7
8
9
3a
3b
3c
4a
4b
5a
5b
6a
6b
7
90
81
96
87
70
27
22
76
59
24
84
51
67
75
74
76
73b
47b
73
78
59
76
10
11
a
8
Scheme 1 Alcoholytic DKR of racemic azlactones.
b
Conversion and ee determined by chiral HPLC. (S)-Enantiomer
formed in excess.
*berkessel@uni-koeln.de
1898 | Chem. Commun., 2005, 1898–1900
This journal is ß The Royal Society of Chemistry 2005

View Article Online
opening of the Phe-derived azlactone 1a with allyl alcohol (Table 1,
entry 2). Replacement of the methyl group by a cyclohexyl group
(3c) increased both reactivity and selectivity (entry 3).
Incorporation of a 1-phenethyl group furnished the two diaster-
eomeric catalysts 4a, 4b, none of them showing enhanced
selectivity (Table 1, entries 4, 5). The same holds for the
diastereomeric catalysts 5a, b which, in addition, showed fairly
low activity (entries 6, 7). The latter observation demonstrates the
importance of the rigid diamine scaffold. In summary, it appears
that the sense of stereoinduction is determined predominantly by
the absolute configuration of the chiral diamine. It is interesting to
note that the C2-symmetric thiourea 8 showed selectivity compar-
able to 3c (Table 1, entries 3, 11).
Table 2 Substrate screening using the thiourea catalyst 6b
Entry
Substrate (R)
Time/h
Conversiona (%)
eea,b (%)
1
2
3
4
5
a
1a (PhCH2)
1b (Me)
1c (i-Pr )
1d (t-Bu)
1e (i-Bu)
48
24
48
48
48
77 (98)
94
59 (89)
28
77
78 (77)
80
92 (90)
95
91
The values in parentheses were obtained for catalysis using 1.7-fold
We expected further enhancement in selectivity by increasing the
steric demand at the additional chiral centre. For this purpose, we
synthesized the tert-leucine amide-derived catalysts 6a, 6b and 7
(Scheme 2). The tert-leucine amide motif was employed previously
by Jacobsen et al. in the design of chiral organocatalysts.5 To our
delight we found that the catalyst 6b indeed furnished 78% ee,
which is the highest enantioselectivity ever achieved in the
chemically catalyzed DKR of the Phe-azlactone 1a. The X-ray
crystal structure of the related catalyst 6a (Fig. 1){ aids in
rationalizing the improved selectivity: We assume that the chiral
information present at the ‘‘left’’ end of the thiourea moiety
enhances the stereodifferentiation effected by the diaminocyclo-
hexane moiety. Clearly, changing from N,N-diethyl to N-methyl-
N-benzyl in 6b makes the latter effect even more pronounced, and
thus explains the improved selectivity of 6b over 6a. In line with the
above argument, the ‘‘mismatched’’ diastereomer 7 (derived from
D-tert-leucine) of 6b provides significantly lower enantioselectivity
and activity (Table 1, entry 10).
b
concentrations of the substrates and catalyst. Enantiomeric excess
was determined by chiral HPLC analysis.
esters 2c–e were obtained (Table 2, entries 3–5). In the case of the
tert-leucine derived azlactone 1d, the product ester 2d was obtained
with 95% ee which marks a significant improvement compared to
all other catalysts studied.3,4 Lowering the reaction temperature
from ambient to 220 uC resulted in even higher enantioselectivity
(ca. 99% ee), but also in unacceptably slow conversion (ca. 5%
after 48 h). At room temperature, the reactions go to completion
without drop in enantioselectivity in all cases, indicating the
dynamic character of these kinetic resolutions. In addititon to the
reaction times listed in Table 2, 96 h were required to achieve
. 90% conversion in the case of the Leu-azlactone 1e. For the
t-Leu-azlactone 1d, 63% conversion was achieved after 7 d.
The initial rates of substrate conversion in the DKR of the Phe-
azlactone (1a) with allyl alcohol, using the three catalysts 3a, 3c
and 6b are compared in Fig. 2. The order of reactivity is 3c . 3a .
6b. When a relative rate (krel) of 1 is assigned to the most
enantioselective catalyst 6b, the bis(trifluoromethyl)phenyl-
substituted thiourea 3a shows a krel of 1.43. The cyclohexyl-
substituted thiourea 3c is more than twice as fast as 6b (krel 5 2.14).
Please note that our organocatalytic6 DKR can also be applied
to the clean stereoinversion of natural and non-natural a-amino
acids. For example, the azlactone 1d prepared from
In order to demonstrate the general applicability of the catalyst
6b, we screened several azlactones 1a–e derived both from natural
and non-natural a-amino acids. The results are summarized in
Table 2. In all cases, the enantiomeric excesses were higher than
75%, using 5 mol% of the catalyst 6b at room temperature. For the
valine-, tert-leucine- and leucine-derived azlactones 1c–e, excellent
enantiomeric purities of the resulting N-benzoylamino acid allyl
Fig. 2 Initial rates of substrate conversion in the DKR of the Phe-
azlactone (1a) with allyl alcohol (rt, 5 mol% catalyst).
Fig. 1 X-Ray crystal structure of the thiourea catalyst 6a (ORTEP).
This journal is ß The Royal Society of Chemistry 2005
Chem. Commun., 2005, 1898–1900 | 1899

View Article Online
added. After addition of a solution containing 167 mmol of the azlactone
enantiomerically pure L-tert-leucine gave the N-benzoyl-D-tert-
leucine allyl ester 2d after alcoholytic DKR with catalyst 6b.
In summary, we have described a method for the alcoholytic
DKR of azlactones effected by the organocatalyst 6b, providing
direct access to a wide range of protected natural and non-natural
a-amino acids in high enantiomeric excess. The catalyst is readily
accessible from commercially available starting materials.
Experiments to reveal the precise mechanism of this catalytic
process are currently underway in our laboratory and will be the
subject of further reports.{
(1.00 eq) in 1.0 ml abs. toluene, the homogeneous reaction mixture was
stirred at ambient temperature. For analysis, 100 ml samples were
withdrawn, diluted with 900 ml dichloromethane, and conversion and
enantiomeric excess were determined immediately by HPLC (Daicel
Chiralpak AD or Merck (S,S)-Whelk O1, n-hexane/2-propanol).
Quantification was based on UV detection at l 5 230 nm and 210 nm,
respectively. Conversion was determined by comparison with the peak
areas of stock solutions of the azlactones and the corresponding N-benzoyl
amino acid esters in dichloromethane.
1 (a) M. Breuer, K. Ditrich, T. Habicher, B. Hauer, M. Keßeler, R. Stu¨rmer
and T. Zelinski, Angew. Chem., 2004, 116, 806; M. Breuer, K. Ditrich,
T. Habicher, B. Hauer, M. Keßeler, R. Stu¨rmer and T. Zelinski, Angew.
Chem. Int. Ed., 2004, 43, 788; (b) H. Gro¨ger and K. Drauz, in Asymmetric
Catalysis on Industrial Scale, eds. H. U. Blaser and E. Schmidt, Wiley-
VCH, Weinheim, 2004, p. 131.
2 (a) H. B. Kagan and J. C. Fiaud, Top. Stereochem., 1988, 18, 249; (b)
K. Faber, Chem. Eur. J., 2001, 7, 5004; (c) H. Pellissier, Tetrahedron,
2003, 59, 8291.
3 (a) K. Gottwald and D. Seebach, Tetrahedron, 1999, 55, 723; (b) L. Xie,
W. Hua, A. S. C. Chan and Y.-C. Leung, Tetrahedron: Asymmetry, 1999,
10, 4715; (c) J. Liang, J. C. Ruble and G. C. Fu, J. Org. Chem., 1998, 63,
3154.
This work was supported by the Fonds der Chemischen
Industrie. In particular, F.C. thanks the Fonds der Chemischen
Industrie for a doctoral fellowship. We gratefully acknowledge the
Degussa AG, Hanau for generous gifts of amino acids.
Albrecht Berkessel,* Santanu Mukherjee, Felix Cleemann,
Thomas N. Mu¨ller and Johann Lex
Institut fu¨r Organische Chemie der Universita¨t zu Ko¨ln, Greinstrasse 4,
D-50939, Ko¨ln, Germany. E-mail: berkessel@uni-koeln.de;
Fax: +49-221-470-5102; Tel: +49-221-470-3283
4 A. Berkessel, F. Cleemann, S. Mukherjee, T. N. Mu¨ller and J. Lex,
Angew. Chem., 2005, 117, 817; A. Berkessel, F. Cleemann, S. Mukherjee,
T. N. Mu¨ller and J. Lex, Angew. Chem. Int. Ed., 2005, 44, 807. Catalysts
of the thiourea-tert-amine type were applied before by Takemoto et al.
for asymmetric aza-Henry reactions and Michael additions: T. Okino,
S. Nakamura, T. Furukawa and Y. Takemoto, Org. Lett., 2004, 6, 625;
T. Okino, Y. Hoashi and Y. Takemoto, J. Am. Chem. Soc., 2003, 125,
12672.
5 (a) G. D. Joly and E. N. Jacobsen, J. Am. Chem. Soc., 2004, 126, 4102;
(b) M. S. Taylor and E. N. Jacobsen, J. Am. Chem. Soc., 2004, 126,
10558; (c) A. G. Wenzel and E. N. Jacobsen, J. Am. Chem. Soc., 2002,
124, 12964; (d) P. Vachal and E. N. Jacobsen, J. Am. Chem. Soc., 2002,
124, 10012; (e) A. G. Wenzel, M. P. Lalonde and E. N. Jacobsen, Synlett,
2003, 1919.
Notes and references
{ Crystal data for 6a: C19H38N4OS, M 5 370.59, colourless needles, 0.30 6
0.30 6 0.25 mm, monoclinic, a 5 11.702(1), b 5 13.351(1), c 5 14.156(1) A,
˚
3
˚
b 5 100.81(1)u, V 5 2172.4(3) A , T 5 100(2) K, space group P21, Z 5 4,
rcalcd. 5 1.133 g cm23, m 5 0.163 mm21. A total of 10636 reflections were
measured, 8211 unique, final residuals were R1 5 0.0439 and vR2 5 0.0887
for 6520 observed reflections with I . 2s(I), 756 parameters, GOF 5 1.023,
maximum residual electron density 0.308 e A23, absolute structure
˚
parameter 20.05(5). Data were collected on a Nonius KappaCCD
˚
diffractometer (2Hmax 5 54u, MoKa-radiation (l 5 0.71073 A), graphite
monochromator, Q/v scans). The structure was solved by using direct
methods,7 followed by full-matrix least-squares refinement (using all unique
reflections) with anisotropic thermal parameters for C, N, O, S and
isotropic parameters for H.8 CCDC 256348. See http://www.rsc.org/
suppdata/cc/b4/b418666d/ for crystallographic data in .cif or other
electronic format.
6 A. Berkessel and H. Gro¨ger, Asymmetric Organocatalysis, Wiley-VCH,
Weinheim, 2005.
7 G. M. Sheldrick, SHELXS-97, Program for the Solution of Crystal
Structures, University of Go¨ttingen, Germany, 1997.
{ Typical procedure for the DKR of azlactones: To a solution of 8.33 mmol
of the catalyst (0.05 eq) in 667 ml abs. toluene, 1.5 eq of allylic alcohol was
8 G. M. Sheldrick, SHELXL-97, Program for the Refinement of Crystal
Structures, University of Go¨ttingen, Germany, 1997.
1900 | Chem. Commun., 2005, 1898–1900
This journal is ß The Royal Society of Chemistry 2005