Massive Parallel Catalyst Screening: Toward Asymmetric MCRs

Article abstract of DOI:10.1021/ol035010u

(Equation presented) Hundreds of Lewis acid/ligand combinations have been screened for stereochemical induction in the Passerini multicomponent reaction. The combination of titan tetraisopropylate and (4S,5S)-4,5- bis(diphenylhydroxymethyl)-2,2-dimethyldioxolane was found to give enantiomeric excesses between 32% and 42% in several examples. The absolute stereo induction of one example was determined chemically and by means of X-ray crystallography. This comprises the first asymmetric Passerini reaction and the first example of a stereochemical induction in an isocyanide based multicomponent reaction by a chiral Lewis acid.

Full text of DOI:10.1021/ol035010u

ORGANIC
LETTERS
2003
Vol. 5, No. 22
4021-4024
Massive Parallel Catalyst Screening:
Toward Asymmetric MCRs
Ulrike Kusebauch,† Barbara Beck,† Kim Messer,† Eberhardt Herdtweck,‡ and
,†
Alexander Do1mling*
Morphochem AG, Gmunderstr.37-37a, 81379 Mu¨nchen, Germany,
and Anorganisch Chemisches Institut der Technischen, UniVersita¨t Mu¨nchen,
Lichtenbergstr.4, 85747 Garching, Germany
alexander.doemling@morphochem.de
Received June 5, 2003 (Revised Manuscript Received September 2, 2003)
ABSTRACT
Hundreds of Lewis acid/ligand combinations have been screened for stereochemical induction in the Passerini multicomponent reaction. The
combination of titan tetraisopropylate and (4S,5S)-4,5-bis(diphenylhydroxymethyl)-2,2-dimethyldioxolane was found to give enantiomeric excesses
between 32% and 42% in several examples. The absolute stereo induction of one example was determined chemically and by means of X-ray
crystallography. This comprises the first asymmetric Passerini reaction and the first example of a stereochemical induction in an isocyanide
based multicomponent reaction by a chiral Lewis acid.
Isocyanide-based MCRs belong to the most interesting
diversity-generating reactions in organic chemistry.1 Mostly
in a one-pot manner, very many different scaffolds are
accessible in a large number of examples from simple start-
ing materials.2 The most prominent of these isocyanide
based MCRs are the Passerini reaction3 (P-3CR) of iso-
cyanides discovered in 1921, oxo components and car-
boxylic acids, and the Ugi reaction4 (U-4CR) from 1959
of isocyanides, oxo components, primary amines, and
carboxylic acids, respectively. During isocyanide-based
MCRs, a new stereocenter is usually generated at the posi-
tion of the former carbonyl carbon.5 Unfortunately, no
conclusive and general way to induce stereochemical in-
formation in the products of MCRs has been developed
in the past. Ugi6 early on recognized that the highest
induction of the new stereocenter formed (Scheme 1) can
be obtained from chiral amines in the case of U-4CR. In
the first instance, chiral phenylethylamines have been utilized
that could be cleaved hydrogenolytically or under acidic
conditions.7 Then chiral ferrocenylakylamines have been
initially used as cleavable chiral ammonia equivalents.8
† Morphochem AG.
‡ Universita¨t Mu¨nchen.
(1) (a) Ugi, I. Proc. Estonian Acad. Sci. Chem. 1995, 44, 237. (b) Hulme,
C.; Gore, V. Curr. Med. Chem. 2003, 10, 51.
(2) Recent examples include: (a) Brahmachary, E.; Ling, F. H.; Svec,
F.; Frechet, J. M. J. J. Comb. Chem. 2003, 5, 441. (b) Marcaccini, S.; Miguel,
D.; Torroba, T.; Garcia-Valverde, M. J. Org. Chem. 2003, 68, 3315. (c)
Kim, Y. B.; Park, S. J.; Keum, G.; Janh, M. S.; Kang, S. B.; Lee, D. H.;
Kim, Y. Bull. Korean Chem. Soc. 2002, 23, 1277. (d) Mironov, M. A.;
Mokrushin, V. S.; Maltsev, S. S. Synlett 2003, 943. (e) Beck, B.; Larbig,
G.; Mejat, B.; Magnin-Lachaux, M.; Picard, A.; Herdtweck, E.; Do¨mling,
A. Org. Lett. 2003 5, 1047.
(4) Ugi, I.; Meyr, R.; Fetzer, U.; Steinbru¨ckner, C. Angew. Chem. 1959,
71, 386.
(5) During Ugi-4CR and Passerini-3CR symmetrical carbonyl compounds
(e.g., formaldehyde, acetone, or cyclohexanone) do not result in a new
stereocenter.
(6) Ugi, I. Rec. Chem. Progr. 1969, 30, 289.
(7) Ugi, I.; Offeremann, K.; Herlinger, H.; Marquarding, D. Liebigs Ann.
Chem. 1967, 709, 1.
(3) Passerini, M. Gazz. Chim. Ital. 1921, 51, 126.
10.1021/ol035010u CCC: $25.00 © 2003 American Chemical Society
Published on Web 10/03/2003

Yb. As chiral ligands we used 12 different commercially
available compounds with a known potential to induce other
stereoselective reactions (Scheme 2).
Scheme 1
Scheme 2. The Passerini Model Reaction
As a model and screening reaction we chose benzyl
isocyanide 13, isobutyric aldehyde 14, and benzoic acid 15
as starting materials giving product 16 (Scheme 2). The
reaction is high yielding (86%), does not show significant
side products, and can be followed by HPLC using an UV
detector. Baseline separation of the enantiomers with reten-
tion times of less than 30 min could be accomplished after
some experimentation with Chiralcel-ODH (4.6 × 250 mm)
as chiral stationary phase. This column performed better than
a evaluation set of five short Pirkle-type Chirex columns (50
mm), although Chirex 3020 gave nearly separated peaks of
the enantiomers. In contrast to the Ugi reaction, the Passerini
reaction is preferably carried out in apolar aprotic solvents.16
Therefore, as solvent we investigated ether, THF, DCM,
toluene, and dioxane. In accordance with the experiments,
the screening concentration of the educts was held at 0.5 M.
Three ligand/Lewis acid ratios were used: 1/1, 1/2, and 2/1.
The catalyst-to-educt ratio was either 1/1 or 0.5/1. The order
of educt addition was aldehyde, isocyanide, and carboxylic
acid. The reaction mixture was stirred overnight at room
temperature and then quenched with aqueous NaHCO3 and
THF. The organic layer was separated and purified by a
silica-based solid-phase extraction column. The enantiomeric
excess was determined by HPLC, if necessary, after a further
chromatographic separation.
Later, Kunz et al.9 and others10 recognized that high induc-
tion could be obtained from glycosylamines. Thus, stereo-
chemically pure R-amino acids have been synthesized.
Unfortunately, none of these methods is generally applic-
able to the plethora of isocyanide-based MCRs. More-
over, the cleavage conditions are quite harsh and thus not
compatible with a wide range of functional groups. Recently,
Ugi et al. described a greatly improved stereoinduction
using a thioglycosylamine with very mild cleavage con-
dition.11 For the P-3CR, no enantioselective procedure at
all has been described in the past.12 Moreover, chiral Lewis
acids have never been used successfully in isocyanide-
based MCRs to induce stereochemistry in the resulting
products.13
Herein, we would like to present for the first time a chiral
Lewis acid based approach toward the solution of the
problem of stereoinduction in isocyanide-based MCRs.
To discover a way to induce stereoselectivity in the
Passerini reaction, we chose to screen various Lewis acid/
chiral ligand combinations.14 We selected oxophilic Lewis
acids15 that have been reported to interact with the compo-
nents of the P-3CR, aldehydes, carboxylic acids, and
importantly isocyanides: Ti, Zr, Mg, Al, B, Cu, Zn, Sc, and
(8) Marquarding, D.; Hoffmann, P.; Heitzer, H.; Ugi, I. J. Am. Chem.
Soc. 1970, 92, 1969.
Using 16 Lewis acids, 12 chiral ligands, and 5 solvents,
an abundance of reactions was performed and screened
(Figure 1). Many of the reactions gave a lot of side products
as compared to the reaction without any additives. Often,
the desired product was not found anymore. In some cases,
the P-2CR product was formed instead, whereby water acts
formally as acid component yielding the corresponding
R-hydroxyamide. Several combinations especially including
the Lewis acids G Yb(OTf)3, H Zn(OTf)2, K MgBr2, M
MgBr2-OEt2, and N BF3-OEt2 showed encouraging ee’s.
Unfortunately, several byproducts were formed. Therefore,
(9) (a) Kunz, H.; Pfrengle, W. J. Am. Chem. Soc. 1988, 110, 651. (b)
Kunz, H.; Ru¨ck, K. Angew. Chem. 1993, 105, 355.
(10) (a) Goebel, M.; Ugi, I. Synthesis 1991, 1095. (b) Linderman, R. J.;
Binet, S.; Petrich, S. R. J. Org. Chem. 1999, 64, 336.
(11) Thiosugar amines, as chiral inductors, display the advantage of being
easily cleavable (Hg2+, 20 °C): Ross, G.; Ugi, I.; Herdtweck, E. Tetrahedron
2002, 58, 6127.
(12) A Camphor derived isocyanide has been described to induce
diastereoselectivity in the P-3CR: Bock, H.; Ugi, I. J. Prakt. Chem. 1997,
339, 385.
(13) For the unsuccessful use of chiral Lewis acids in the P-2CR, see:
Seebach, D.; Adam, G.; Gees, T.; Schiess, M.; Weigand, W. Chem. Ber.
1988, 121, 507.
(14) For a recent review on high-throughput methods for the development
of new catalytic asymmetric reactions, see: Traverse, J. F.; Snapper, M. L.
DDT 2002, 7, 1002.
(15) Yamamoto, H., Ed. Lewis Acids in Organic Synthesis,; Wiley:
Weinheim, 2000; Vols. 1 and 2.
(16) Do¨mling, A.; Ugi, I. Angew. Chem., Int. Ed. Engl. 2000, 39,
3168.
4022
Org. Lett., Vol. 5, No. 22, 2003

Table 1. Enantiomeric Excesses and Yields of Selected P-3CR
Figure 1.
a Isolated yield of the Passerini reaction performed on a 1 mmol scale
with A7. b In brackets: isolated yield of the Passerini reaction performed
on a 5 mmol scale without A7. c Enantiomeric excess of the Passerini
reaction performed with A7.
these reactions were not investigated. Only the Lewis acid
A Ti(i-OPr)4 in combination with ligands 7, 9, and 12 showed
significant ee’s and clean reactions that allowed the isolation
of the product at the same time. These reactions were
repeated and examined more closely.
To get a first idea about scope and limitation of this
catalyst A7, a couple of reactions varying the components
(Table 1) were performed. Stereoinduction from 32% to
42% has been reached in these reactions. Generally, one
observes that the reaction with additive suffers from poorer
yields. During the hydrolysis of the reaction, considerable
amounts of R-hydroxy methyl amides (P-2CR product) are
formed.
The absolute stereoinduction was determined by perform-
ing the P-3CR between 2S-N-Boc-phenylalanine, isobutyral-
dehyde, and benzylisocyanide yielding the correspond-
ing product 22 in a diastereomeric ratio of 1:1.27. The
structure of the preferentially crystallizing isomer 22
was investigated by X-ray crystallography (Scheme 3) and
the stereochemistry determined to be 2-S,5-R. This com-
pound with known absolute stereochemistry was then
subjected to saponification with aqueous sodium hydroxide
solution, yielding the corresponding stereochemically
pure, R-configured R-hydroxyamide 23. The stereoisomeric
mixture of the P-3CR with catalyst A7 yielding compound
16 with an ee: 36% was also subject to saponification.
The resulting mixture of the two alcohols (R- and S-23)
as compared with the reference isomer R-23 obtained
from the crystal 22. This clearly indicates that the major
stereoisomer is S-configured, proven by coelution of the
isomer.
The reaction can be performed with sub-stoichiometric
amounts of Ti and ligand, but a considerable loss of
enantiomeric excess occurs. Clearly, in the stereochemistry-
inducing step of the reaction the aldehyde has to be
complexed by Ti (Scheme 4). Generally, titanium complexes
are known to exist in complex mixtures of monomeric and
oligomeric structures.17 The configuration displayed in the
rate-determining step causes the latter stereochemistry of the
Passerini product. A possible complex 24 could account for
the observed induction. The addition of the isocyanide onto
the aldehyde carbon giving the imminium ion 25 could
happen via a pre-coordination of the isocyanide on the
titanium. Alternatively, the isocyanide inserts directly without
precoordination with the titanium atom. Addition of the
carboxylic acid yields the R-adduct 26, which upon rear-
rangement forms the Passerini product.
(17) (a) Mahrwald, R.; Ziemer, B. Tetrahedron Lett. 2002, 43, 25, 4459.
(b) Review: Weidman, B.; Seebach, D. Angew. Chem., Int. Ed. Engl. 1983,
22, 31.
Org. Lett., Vol. 5, No. 22, 2003
4023

Scheme 3
Scheme 4
are not yet synthetically useful, chiral Lewis acids provide
a promising and not yet explored tool toward chiral MCRs.
The possibility to perform chiral MCRs would tremendously
broaden the synthetic usefulness of these reactions. Further
work is ongoing to improve the enantioselectivity of the
P-3CR by modifying the ligands and reaction conditions and
to investigate the reaction mechanism.
Acknowledgment. This work is dedicated to Lutz Weber
on the occasion of his 45th birthday.
Supporting Information Available: General procedure
for the Passerini reaction and 1H and 13C NMR, X-ray, and
HPLC data are given. This material is available free of charge
via the Internet at http://pubs.acs.org.
OL035010U
In conclusion, we are reporting for the first time an
enantioselective Passerini MCR using a chiral Lewis acid
catalyst.18 This catalyst system was found by screening of a
multitude of different reactions. Although the observed ee’s
(18) During the submission process, two more examples of stereoselective
Passerini reaction where published: (a) Frey, R.; Galbraith, S. G.; Guelfi,
S.; Lamberth, C.; Zeller, M. Synlett 2003, 1536; (b) Denmark, S.; Fan, Y.
J. Am. Chem. Soc. 2003, 125, 7825.
4024
Org. Lett., Vol. 5, No. 22, 2003