Rhodium-catalysed racemisation of N-acyl α-amino acids

Article abstract of DOI:10.1016/S0040-4039(00)00539-6

The first transition metal-catalysed racemisation of N-acyl α-amino acids, which is of importance for kinetic resolution processes, is described. Enantiomerically pure N-acyl α-amino acids were efficiently racemised under mild conditions using various rhodium complexes as catalysts, e.g. [Rh(cod)Cl]₂, in the presence of phosphines. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4039(00)00539-6

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 41 (2000) 3821–3824
Rhodium-catalysed racemisation of N-acyl α-amino acids
Martin J. Hateley, Daniel A. Schichl, Hans-Jörn Kreuzfeld and Matthias Beller ∗
Institut für Organische Katalyseforschung an der Universität Rostock e.V., Buchbinderstr. 5-6, D-18055 Rostock, Germany
Received 11 February 2000; accepted 30 March 2000
Abstract
The first transition metal-catalysed racemisation of N-acyl α-amino acids, which is of importance for kinetic
resolution processes, is described. Enantiomerically pure N-acyl α-amino acids were efficiently racemised under
mild conditions using various rhodium complexes as catalysts, e.g. [Rh(cod)Cl]2, in the presence of phosphines.
© 2000 Elsevier Science Ltd. All rights reserved.
Keywords: racemisation; rhodium; tricyclohexylphosphine; N-acyl α-amino acids.
Racemisation, that is the interconversion between enantiomers, is an important but overlooked reac-
tion. Racemisation of enantiomerically pure compounds is often an unwanted side reaction. Nevertheless,
the ability to racemise enantiomerically pure compounds under mild conditions is important as a part of
the kinetic resolution process.1 In industry a number of important enantiomerically pure compounds are
still produced by kinetic resolution procedures,2 despite major advances in stereoselective synthesis. In a
classic kinetic resolution only one enantiomer is converted into product and thus the maximum obtainable
yield is 50%.3 Racemisation of the remaining enantiomer followed by further kinetic resolutions leads,
after several repetitions, to a maximum obtainable yield approaching 100%.
Recently, we became interested in the racemisation of N-acyl α-amino acids since we developed a
convenient one-step procedure (amidocarbonylation) for the synthesis of this class of compounds.4 It has
long been known that N-acyl α-amino acids can be racemised using acetic anhydride at high temperatures
via azlactone formation;5 however, we wanted to find a milder, catalytic alternative to this procedure. To
the best of our knowledge we report here the first transition metal-catalysed racemisation of N-acyl α-
amino acids.
In order to find an active catalyst we tested a variety of transition metal complexes which are known
to be active in transfer hydrogenation6 (Rh, Ru, Ir complexes), since we postulated that a possible
racemisation pathway might proceed through dehydrogenation of the α-H and the NH of the amino acid
derivative and subsequent hydrogenation of the resulting imine. The catalyst activities were assessed by
dissolving the catalyst, enantiomerically pure (S)-N-acetyl phenylalanine, our test substrate, in the chosen
solvent and heating the mixture in a pressure tube for a given period of time. The enantiomeric excess
∗
Corresponding author.
0040-4039/00/$ - see front matter © 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(00)00539-6
tetl 16818

3822
of the substrate was determined by GC analysis using a chiral column after derivatisation of the N-acyl
α-amino acid using (trimethylsilyl)diazomethane in methanol.7
Interestingly, it was found that several rhodium complexes in acetonitrile catalysed the racemisation
of (S)-N-acetyl phenylalanine (Table 1). The most efficient catalyst was found to be Wilkinsons’ catalyst
(entries 1–3), although both the cationic Rh(cod)2BF4 complex and the [Rh(cod)Cl]2 complex exhibited
some activity. Entries 1 to 3 show that significant racemisation already proceeds at 60°C. In addition,
the catalyst seems to be very robust and does not appear to lose substantial activity after 24 h. With the
exception of Rh(cod)acac all the complexes gave consistent results when the experiments were repeated.
Entry 7 (Table 1) shows that when no catalyst is present only very little racemisation takes place at 60°C
in the course of 48 h.
Table 1
Rhodium complex-catalysed racemisation of (S)-N-acetyl phenylalanine8
Since Wilkinsons’ catalyst, a phosphino–rhodium complex, had proven to be the most active catalyst,
the effect of the addition of a variety of phosphines was explored, in an attempt to form a more active
phosphino–rhodium species in situ. Although a variety of phosphines were tested, (including tris(o-
tolyl)phosphine, tris(p-trifluorophenyl)phosphine, tris-pyrolidinephosphine, tris(t-butyl)phosphine and
the chelating phosphines DPPE and DPPB), it was found that tricyclohexylphosphine gave the best
results (Table 2). The difference in the rate of racemisation between the systems using PPh3 and PCy3 is
clearly visible in entries 6 to 9 (Table 2). With (S)-N-acetyl phenylalanine methyl ester as the substrate
no racemisation was observed, indicating that the presence of a free carboxylic group is important. The
addition of 1 equivalent of α-cinnamic acid or the presence of hydrogen led to a much reduced rate of
racemisation, possibly due to the formation of inactive rhodium species.
In order to see what effect the acyl moiety played on the racemisation we tested [Rh(cod)Cl]2 in
the presence of PPh3 and PCy3 for the racemisation of N-benzoyl, N-trifluoroacetyl and N-phenacetyl
phenylalanine (Table 3). While the trifluoroacetyl derivative showed no racemisation and the N-benzoyl
derivative did not racemise as well as the N-acetyl compound, the N-phenacetyl compound racemised
at a comparable rate to the N-acetyl compound. In all cases the racemisation in the presence of PCy3
was faster than in the presence of PPh3. Next, the standard system was tested on a series of other N-acyl
α-amino acids, with the enantiomeric excess of the substrate being measured at 24 and 48 h after the start

3823
Table 2
Rhodium complex-catalysed racemisation of (S)-N-acetyl phenylalanine in the presence of phosphine
ligands
Table 3
Rhodium-catalysed racemisation of different N-acyl amino acids

3824
of the reaction (see Table 3). All the N-acetyl amino acids tested racemised, expect for N-acetyl aspartic
acid. In all cases a control reaction was conducted with just the N-acetyl amino acid in acetonitrile and
in each case no significant racemisation was observed (ee >95%).
In conclusion, we have demonstrated for the first time that N-acyl amino acids can be efficiently
racemised using rhodium complexes under comparably mild condition. These results are not only of
significance to kinetic resolution processes, but also of major importance to asymmetric hydrogenations
of N-acylamido acrylic and cinnamic acids. It is possible that at low hydrogen pressure and at high
conversion, small amounts of racemisation take place, thus decreasing the enantiomeric excess of the
reaction.
Acknowledgements
We would like to thank the state of Mecklenburg-Vorpommern for providing financial support for
M.J.H. We also thank Dr. T. Riermeier, Dr. H. Trauthwein and Dr. B. Bosch (all Aventis R&T) for
helpful discussions.
References
1. Ebbers, E. J.; Ariaans, G. J. A.; Houbiers, J. P. M.; Bruggink, A.; Zwanenburg, B. Tetrahedron 1997, 53, 9417.
2. Crosby, J. In Chirality in Industry; Collins, A. N.; Sheldrake, G. N., Eds.; Wiley, 1992, p. 20 (see also Ref. 50).
3. Bornscheuer, U. T.; Kazlauskas R. J. Hydrolases in Organic Synthesis — Regio- and Stereoselective Biotransformations;
Wiley-VCH: Weinheim, 1999, p. 122.
4. (a) Beller, M.; Eckert, M.; Vollmüller, F.; Bogdanovic, S.; Geissler, H. Angew. Chem., Int. Ed. Engl. 1997, 36, 1494. (b)
Beller, M.; Eckert, M.; Geissler, H.; Napierski, B.; Rebenstock, H.-P.; Holla, E. W. Chem. Eur. J. 1998, 4, 935.
5. Ebbers, E. J.; Ariaans, G. J. A.; Houbiers, J. P. M.; Bruggink, A.; Zwanenburg, B. Tetrahedron 1997, 53, 9443, and patent
Ref. 194–200.
6. Gladiali, S.; Mestroni G. In Transition Metals for Organic Synthesis; Beller, M.; Bolm C., Eds.; Wiley-VCH: Weinheim,
1998; Vol. 2, p. 97.
7. Ishida, T.; Inoue, M.; Harusawa, S.; Hamada, Y.; Shiori, T. Chem. Pharm. Bull. 1981, 29, 1475, 1495.
8. General procedure for racemisation experiments: A solution of PCy3 was made up in absolute THF under argon and the
appropriate amount was pipetted into an ACE pressure tube and the THF removed using a stream of argon. The residue was
then dissolved in dry acetonitrile (10 mL) and the N-acyl α-amino acid (0.48 mmol) was added followed by the transition
metal complex as either a solid or a solution in MeCN. The mixture was then heated for between 24 and 48 h at 40–60°C.
A portion of the solution was then taken and derivatised using (trimethylsilyl)diazomethane (2 M in hexane) and methanol
and the enantiomeric excess determined by GC or HPLC analysis.