.
Angewandte
Communications
DOI: 10.1002/anie.201200370
Synthetic Methods
Pseudoephenamine: A Practical Chiral Auxiliary for Asymmetric
Synthesis**
Marvin R. Morales, Kevin T. Mellem, and Andrew G. Myers*
Pseudoephedrine is widely employed as a chiral auxiliary in
diastereoselective alkylation reactions and provides ready
access to enantiomerically enriched carboxylic acids, alde-
hydes, ketones, and alcohols.[1] Because pseudoephedrine can
be transformed into methamphetamine and other illegal drug
substances, many countries restrict or ban its sale and
distribution, thus complicating its use in industrial and
academic settings.[2] Herein we report that (1S,2S)- and
(1R,2R)-2-methylamino-1,2-diphenylethanol (synonymously,
(1S,2S)- and (1R,2R)-pseudoephenamine,[3] respectively)
have a broad range of utilities in asymmetric synthesis that
meet or exceed those that previously characterized the
pseudoephedrine system alone, with several advantages.
Scheme 1. Synthesis of (À)-(1S,2S)-pseudoephenamine by a modified
Specifically, 1) these auxiliaries are free from regulatory
restrictions and are not known to be transformable into illicit
substances, 2) asymmetric alkylation reactions that employ
pseudoephenamine as a directing group proceed with equal or
greater diastereoselectivities in relation to the corresponding
reactions that employ pseudoephedrine, with notable
improvements in the selectivities of alkylation reactions that
form quaternary stereocenters, and 3) amides derived from
pseudoephenamine exhibit a greater propensity to be crys-
talline substances compared with the corresponding pseu-
doephedrine derivatives and provide sharp, well-defined
signals in NMR spectra.
Tishler protocol followed by N-methylation.
both enantiomers of threo-1,2-diphenyl-2-aminoethanol from
the appropriate erythro-diastereomer (both erythro diaste-
reomers are commercially available in enantiomerically pure
form and are widely used as chiral auxiliaries themselves, for
example, in the Williams amino acid synthesis).[6–8] Subse-
quent N-methylation of threo-1,2-diphenyl-2-aminoethanol
was achieved in 97% yield by N-formylation with acetic
formic anhydride followed by reduction with lithium alumi-
num hydride.[9] The product was recrystallized from hot
ethanol to produce large, orthorhombic, colorless crystals
(m.p. 109–1108C).[10] We have routinely prepared 20–40 g
batches of (1R,2R)- or (1S,2S)-pseudoephenamine by the
described four-step sequence, which proceeds in 87% yield
and requires no column chromatography.[11,12] X-ray crystallo-
graphic analysis revealed that pseudoephenamine adopts
a conformation identical to pseudoephedrine in the solid
state, with gauche orientations between both the amino-
methyl and hydroxy substituents as well as the two phenyl
substituents (Figure 1).
Both enantiomeric forms of pseudoephenamine are easily
prepared with well-established methods (Scheme 1). In 1951,
Tishler and co-workers at Merck reported a process for the
transformation
of
erythro-1,2-diphenyl-2-aminoethanol
(1R,2S or 1S,2R) into the corresponding threo diastereomer
(1S,2S or 1R,2R, respectively) by N-formylation with form-
amide, invertive cyclization to form the corresponding oxazo-
line using thionyl chloride, and hydrolytic ring-opening under
acidic conditions.[4,5] By employing a small but important
modification (that is, the use of formamide containing
approximately 0.2 equivalents ammonium formate for N-
formylation rather than pure formamide, the use of which
leads to a reduced yield and yellowing of the product), we
have applied the Tishler protocol for large-scale synthesis of
Amide derivatives of pseudoephenamine were prepared
from the corresponding carboxylic acid chlorides or anhy-
drides by routine methods and, in most cases, were crystalline
solids (see the Supporting Information). Pseudoephenamine
[*] M. R. Morales, K. T. Mellem, Prof. A. G. Myers
Department of Chemistry and Chemical Biology, Harvard University
Cambridge, MA 02138 (USA)
E-mail: myers@chemistry.harvard.edu
[**] We gratefully acknowledge the NSF (CHE-0749566) and the NIH
(CA-047148) for financial support of this research. We also wish to
thank Dr. Shao-Liang Zheng for X-ray crystallographic analyses and
Dr. David Kummer for helpful discussions.
Figure 1. X-ray crystal structures of (À)-(1S,2S)-pseudoephenamine
(left) and (+)-(1S,2S)-pseudoephedrine[13] (right). Thermal ellipsoids
are at 50 % probability.
Supporting information for this article is available on the WWW
4568
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2012, 51, 4568 –4571