Combinatorial ligand development based on mass spectrometric screening and a double mass-labeling strategy

Article abstract of DOI:10.1021/ja0740317

Mixtures of chiral ligands were prepared from simple precursors and evaluated in the asymmetric Pd-catalyzed allylation by simultaneous screening using ESI-MS and quasienantiomeric substrates. Copyright

Full text of DOI:10.1021/ja0740317

Published on Web 02/20/2008
Combinatorial Ligand Development Based on Mass Spectrometric Screening
and a Double Mass-Labeling Strategy
Christian Markert, Pirmin Ro¨sel, and Andreas Pfaltz*
Department of Chemistry, UniVersity of Basel, St. Johanns Ring 19, CH-4056 Basel, Switzerland
Received June 4, 2007; E-mail: Andreas.Pfaltz@unibas.ch
Scheme 1. Pd-Catalyzed Allylic Substitution4 Starting from
Quasienantiomeric Substrates 1a and 1b
The development and fine-tuning of chiral catalysts which
provide effective asymmetric induction in a specific transformation
is often a labor-intensive and time-consuming process. Thus,
screening methods that accelerate the identification and optimization
of new catalysts are of considerable interest.1
We have recently developed a screening method for chiral
catalysts which induce kinetic resolution of racemic allyl esters 1.2
Starting from a 1:1 mixture of quasienantiomeric substrates 1a and
1b, the selectivity of chiral Pd catalysts can be determined by mass
spectrometric quantification of the corresponding allyl-Pd inter-
mediates 2a and 2b (Scheme 1).3
Electrospray ionization mass spectrometry (ESI-MS) is used as
an analytical tool, as this technique allows selective detection of
charged species in the presence of a large excess of neutral com-
pounds.5 Our method benefits from operational simplicity and a
fast, direct readout while avoiding completely any workup, isolation,
or purification steps. All components are easily handled as solutions,
and the detected ratio of mass-labeled intermediates 2a and 2b
Scheme 2. Preparation of Modular Ligands 5a
directly reflects the catalyst’s intrinsic selectivity (kB/kA),6 unaffected
by catalytically active impurities, partial dissociation of the chiral
ligand-metal complex, or a noncatalytic background reaction
leading to racemic product.
In previous experiments, we were able to show that this screening
method can also be applied to catalyst mixtures. The enantio-
selectivities of five Pd catalysts were determined simultaneously
in a single experiment.2 Here we report on an extension of this
concept and introduce a double mass-labeling strategy in the context
of a catalyst optimization study.
Starting from a set of readily available chiral diamines 3 and a
chiral diol 4, a library of ligands 5 can be prepared in two steps, as
illustrated in Scheme 2 for two diamines 3a and 3b. The
a
Conditions: (a) PCl3, NEt3, THF, -78 °C f rt; filter, evaporate; (b)
chiral diol 4, NEt3, THF, -78 °C f rt; filter, evaporate.
conventional way to identify the most effective ligand would be to
prepare all possible products 5 from a given set of precursors
3 and to test them individually. Although symmetrical ligands
such as 5aa or 5bb derived from two identical precursors 3 can be
readily prepared, the synthesis of unsymmetrical derivatives such
as 5ab is less straightforward and most likely would require
additional protection and deprotection steps in order to avoid
formation of mixtures of symmetrical and unsymmetrical ligands.
In addition, because the diamines and the diol are chiral, the
influence of the relative configuration on the enantioselectivity has
to be examined.
We thought that our mass spectrometric screening method would
enable a much more effective approach to identify the optimal
ligand, without the need of preparing each ligand separately. By
statistical condensation of a mixture of diamines under the
conditions shown in Scheme 2, it should be possible to prepare a
library of ligands of type 5 in a single batch and to screen the
corresponding catalyst mixture, prepared in situ by complexation
with Pd, simultaneously in a single experiment. To demonstrate
this concept, we chose sulfonamides of trans-1,2-cyclohexanedi-
amine (3) and trans-(3,4)-1-benzyl-3,4-pyrrolidinediol (4) as readily
available chiral building blocks.7
The first experiment was designed to identify the optimal
(matched) relative configuration. Accordingly, the (R,R)-diamine
3a (R1 ) PhSO2) and the quasienantiomeric (S,S)-diamine 3b (R2
) 4-Me-PhSO2) were mixed and condensed with phosphorus
trichloride and (S,S)-diol 4 (Scheme 2 and Figure 1). Filtration
through a plug of silica gel and removal of all volatiles under
vacuum yielded the expected mixture of three ligands 6aa, 6ab,
and 6bb. As our screening method is quite tolerant to impurities,
we subsequently found that simple filtration without the use of silica
gel provided ligand mixtures of sufficient purity for complexation
with [(MeCN)2Pd(allyl)]OTf and subsequent screening.2,6 The
spectra of the precatalysts and the screening results are shown in
Figure 1.
The combination of mass-labeled substrates 1a and 1b and mass-
labeled quasidiastereomeric ligands allowed us to determine the
selectivity of each catalyst in the mixture. With a 6:94 ratio of allyl
intermediates, the Pd complex of ligand 6bb was clearly the most
selective catalyst. As the three ligands bear sterically and electroni-
9
3234
J. AM. CHEM. SOC. 2008, 130, 3234-3235
10.1021/ja0740317 CCC: $40.75 © 2008 American Chemical Society

C O M M U N I C A T I O N S
Figure 3. Evolution of the substrate ee with conversion in a preparative
kinetic resolution using ligand 7dd (HPLC).
observed for the less easily accessible unsymmetrical ligands 7ac,
7cd, and 7ad. Having identified 7dd as the best ligand in this series,
we finally synthesized and characterized it as a single compound.
ESI-MS testing of the corresponding Pd catalyst confirmed the high
selectivity induced by this ligand, in agreement with a preparative
kinetic resolution8 and HPLC analysis (Figure 3).
In summary, we have demonstrated that mass spectrometric
screening of quasienantiomeric substrates can considerably simplify
structural optimization of chiral catalysts. Starting from a mixture
of suitable building blocks, libraries of modular ligands can be
readily prepared in a single batch and evaluated by simultaneous
screening. Because the time-consuming synthesis and purification
of individual ligands are avoided in this way, catalyst optimization
can be accelerated significantly. We are currently evaluating this
method for other classes of enantioselective reactions.
Figure 1. Structures of quasidiastereomeric ligands 6aa-6bb and ESI-
MS spectra of the corresponding Pd-allyl precatalysts and reaction
intermediates.
Acknowledgment. Financial support by the Swiss National
Science Foundation is gratefully acknowledged.
Supporting Information Available: Experimental procedures and
characterization data for all reactions and products. This material is
available free of charge via the Internet at http://pubs.acs.org.
References
(1) (a) Archibald, B.; Bru¨mmer, O.; Devenney, M.; Gorer, S.; Jandeleit, B.;
Uno, T.; Weinberg, W. H.; Weskamp, T. In Handbook of Combinatorial
Chemistry; Nicolaou, K. C., Hanko, R., Hartwig, W., Eds.; Wiley-VCH:
Weinheim, 2002; pp 885-990. (b) Hoveyda, A. H. In Handbook of
Combinatorial Chemistry; Nicolaou, K. C., Hanko, R., Hartwig, W., Eds.,
Wiley-VCH: Weinheim, 2002; pp 991-1016.
(2) Markert, C.; Pfaltz, A. Angew. Chem., Int. Ed. 2004, 43, 2498-2500.
(3) Mass-labeled enantiomers have been used before for determining the
enantiomeric purity of chiral products: (a) Horeau, A.; Nouaille, A.
Tetrahedron Lett. 1990, 31, 2707-2710. (b) Reetz, M. T.; Becker, M.
H.; Klein, H. W.; Sto¨ckigt, D. Angew. Chem., Int. Ed. 1999, 38, 1758-
1761. (c) Yao, S.; Siuzdak, G.; Finn, M. G. J. Org. Chem. 2003, 68, 2540-
2546. (d) For a review on quasienatiomers, see: Zhang, Q.; Curran, D.
P. Chem.sEur. J. 2005, 11, 4866-4880 and references therein.
(4) Reviews: (a) Trost, B. M.; Lee, C. In Catalytic Asymmetric Synthesis,
2nd ed.; Ojima, I., Ed.; Wiley-VCH: New York, 2000; pp 593-649. (b)
Pfaltz, A.; Lautens, M. In ComprehensiVe Asymmetric Catalysis; Jacobsen,
E. N., Pfaltz, A., Yamamoto, H., Eds.; Springer: Heidelberg, 1999; pp
833-886. (c) Trost, B. M.; Van Vranken, D. L. Chem. ReV. 1996, 96,
395.
(5) Chen, P. Angew. Chem., Int. Ed. 2003, 42, 2832-2847.
(6) Na[15-c-5]CEt(CO2Et)2 is used as a convenient bench-stable malonate
nucleophile that exhibits good solubility in a variety of solvents. Because
2a and 2b react at the same rate, 2a/2b is equal to kA/kB at low conversion
when 1a/1b ) 1. Thus samples were taken in the initial phase after ca.
1-2 turnovers.
Figure 2. Structures of ligands 7cc-7dd and ESI-MS spectra of the
corresponding Pd-allyl precatalysts and reaction intermediates.
cally very similar phenyl- and p-tolyl-groups at their diamine units,
we assumed that they behave like real diastereomers. Thus, we
concluded that the all-S configuration of ligand 6bb corresponds
to the matched combination of stereocenters. As seen from the MS
signals, all three ligands show the same sense of asymmetric
induction. Apparently, the stereoselectivity is mainly controlled by
the (S,S)-diol bridge. A control experiment with interchanged phenyl
and tolyl groups confirmed this result (Supporting Information).
In a subsequent experiment, the structure of ligand 6bb was
further varied by introduction of different sulfonamide groups. As
before, a mixture of all six possible ligands was prepared from
three different sulfonamides and tested simultaneously (Figure 2).
ESI-MS screening revealed that sterically more demanding
sulfonamide substituents induce higher selectivites, a trend also
(7) Hilgraf, R.; Pfaltz, A. AdV. Synth. Catal. 2005, 347, 61-77.
(8) Using racemic 1,3-diphenylallylbenzoate as substrate, a krel of 45 was
observed in the range 0-47% conversion. At higher conversion, the
observed ee values no longer correlated with the catalyst intrinsic
selectivity indicating slow catalyst decomposition or the onset of an
unselective background reaction.
JA0740317
9
J. AM. CHEM. SOC. VOL. 130, NO. 11, 2008 3235