Imprinting and locking chiral memory for stereoselective catalysis

Article abstract of DOI:10.1039/b611709k

A salen ligand based Co(iii) complex ±1 with imprintable chiral memory was locked-in and used for stereoselective catalysis. The Royal Society of Chemistry.

Full text of DOI:10.1039/b611709k

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Imprinting and locking chiral memory for stereoselective catalysis{
Jik Chin,*a Yong S. Chong,b Rhiana Bobb,a Lisa Studnickia and Jong-In Hong*b
Received (in Cambridge, UK) 14th August 2006, Accepted 9th October 2006
First published as an Advance Article on the web 30th October 2006
DOI: 10.1039/b611709k
complexes of salen ligands.17 Although the meso-diamine based
A salen ligand based Co(III) complex ¡1 with imprintable
ligand itself is achiral, the cobalt complex derived from this ligand
is chiral and is produced as a racemic mixture ¡1. As shown in
Fig. 1, only one of the two phenols can coordinate at a time to the
metal. It is this coordination of the phenolic oxygen that makes the
cobalt complex chiral. 1H NMR of the metal complex shows four
distinct tert-butyl group signals consistent with the proposed
chiral memory was locked-in and used for stereoselective
catalysis.
In principle, achiral or racemic molecules with imprintable chiral
memory could be converted to stereoselective catalysts. However,
it would be difficult to design a polymer1 or a supramolecular2,3
structure with imprintable chiral memory that is a stereoselective
catalyst. A metal complex with imprintable chiral memory that can
be locked-in would stand a much better chance for such an
application.4 In the field of stereoselective catalysis, salen ligand
(Schiff base formed from a 2 : 1 mixture of salicylaldehyde and
ethylene diamine derivatives) based metal complexes have received
considerable attention over the last couple of decades.5 We became
interested in adapting the concept of imprintable chiral memory
onto a system with obvious advantages for stereoselective
recognition6 and catalysis.7 Jacobsen,8 Katsuki9 and others have
shown that salen based metal complexes are highly effective as
stereoselective catalysts for a wide variety of important chemical
reactions including (a) epoxidation of alkenes with manganese–
salen complexes;9,10 (b) hydrolysis of epoxides with cobalt–salen
complexes;8 (c) Strecker synthesis of amino acids with aluminium–
salen complexes;11 (d) Diels–Alder reaction with chromium–salen
complexes;12 (e) cyanohydrin formation with titanium–
salen complexes;13 and (f) Michael addition with aluminium–salen
complexes.14,15 We developed a salen based Co(III) complex ¡1
with imprintable chiral memory (Fig. 1) that can be locked-in and
used for stereoselective catalysis.
1
structure (ESI{). H NMR reveals that addition of two equiv of
(2R,3S)-2-phenyl-3-methylaziridine18 to the racemic cobalt com-
plex ¡1 results in complete coordination of the amine within
minutes at ambient temperature (Fig. 2). One of the imine
hydrogen signals of the initial cobalt complex appears as a clean
singlet (Fig. 2a) while the other imine signal is buried (ESI{). Upon
coordination of the aziridine, the singlet signal changes into two
new singlet signals (Fig. 2b). Initially, a 1 : 1 ratio of the
diastereomeric complex is formed, however after equilibration this
ratio changes and stabilizes at about 2 : 1 (Fig. 2c). The rate
constant at 40 uC for the equilibration reaction as measured by
NMR methods is 2.4 6 1024 s21 (t1/2 y 2 d at 25 uC). Thus, the
chiral aziridine shifts the equilibrium in Fig. 1 by coordinating6
and stabilizing one complex more than the other. Possible origin of
the stereoselectivity is described in the ESI.{
The diimine ligand in ¡1 was synthesized from meso-1,2-bis(2-
hydroxyphenyl)-ethylenediamine16 and 3,5-di-tert-butylsalicyl-
aldehyde.17 The neutral ligand was reacted with cobalt acetate
and air oxidized following standard procedures for making Co(III)
Fig. 1 Co(III)–salen ligand complex ¡1.
aDepartment of Chemistry, University of Toronto, 80 St. George Street,
Toronto, M5S 3H6, Canada. E-mail: jchin@chem.utoronto.ca;
Fax: (+1)-416-946-7335
Fig. 2 1H NMR spectra in DMSO-d6 of (a) one of the two imine signals
of the initial racemic Co(III) complex ¡1, (b) the initial diastereomeric
complex formed from the addition of (2R,3S) 2-phenyl-3-methylaziridine,
(c) the diastereomeric complex after equilibration by heating and (d) the
complex after displacement of the coordinated aziridine with benzylamine.
bDepartment of Chemistry, Seoul National University, Seoul, 151-747,
Korea. E-mail: jihong@snu.ac.kr; Fax: (+82)-2-889-1568
{ Electronic supplementary information (ESI) available: Experimental
procedures, characterization data, and 1H NMR data. See DOI: 10.1039/
b611709k
120 | Chem. Commun., 2007, 120–122
This journal is ß The Royal Society of Chemistry 2007

View Article Online
The slow shift in equilibrium indicates that the rate of exchange
in position of the two phenolic groups in Fig. 1 is slow. Such slow
ligand exchange rates are common with octahedral Co(III)
complexes and they are said to be substitutionally inert.19
Although weakly coordinating epoxides are known to dissociate
from octahedral (salen)Co(III) complexes at unusually rapid
rates,11 the strongly coordinating intramolecular phenoxy groups
in ¡1 exchange slowly due to the slow dissociation rate. The
coordinated aziridine can be rapidly displaced with an achiral
amine such as benzylamine at ambient temperature (Fig. 2d). The
chiral aziridine bound complexes are diastereomeric while the
benzylamine bound complexes are enantiomeric. Thus the two
imine C–H signals of the aziridine bound cobalt complexes
(Fig. 2c) are replaced by a single peak (Fig. 2d) upon exchange of
the aziridine with benzylamine.
(ESI{). The chiral memory can be erased by heating the
benzylamine coordinated complex. The CD bands of the
imprinted complex disappear when the sample is heated at 70 uC
for 30 min (t1/2 y 2 d at 25 uC). A more stable chiral memory
would be desirable if we were to use the imprinted complex as a
stereoselective catalyst.
In order to obtain a more lasting chiral memory, we decided to
lock in the chiral memory. For the purpose of making the locked
complex, phenylalanine was used as the imprinting agent rather
than the chiral aziridine because the amino acid is easier to remove
after imprinting. We locked in the chiral memory by acylating the
free phenolic group of the imprinted complex with acetic
anhydride. The amino acid was then removed by aqueous workup
and the locked complex was purified by chromatography. The
enantiopurity of the locked complex (36% ee) was determined by
1H NMR methods using R-phenethylamine. We showed that
although R-phenethylamine binds to ¡1 with no observable
stereselectivity, the two diastereomeric complexes show distinct 1H
NMR signals (ESI{). Addition of excess R-phenethylamine to the
locked complex results in generation of two diastereomeric
The chiral aziridine cobalt complex (after equilibration of
phenol coordination (Fig. 2c)) shows a distinct circular dichroism
(CD) band around the absorption band for the metal complex
(Fig. 3). Neither the complex by itself nor the aziridine by itself
gave any CD bands (Fig. 3e and 3f). Thus the metal complex is a
reporter or sensor of chirality that can generate or amplify the CD
signal from a chiral guest.
1
complexes (Fig. 4) with distinct H NMR signals for one of the
imine C–H. As expected, the enantiopurity of the locked complex
is the same when D- or L-phenylalanine is used as an imprinting
agent. Opposite enantiomers of the cobalt complex are enriched
depending on the chirality of the imprinting agent.
Interestingly, the CD spectrum does not change appreciably
when the coordinated chiral aziridine is replaced with the achiral
amine (Fig. 2c and 3a to 2d and 3b). The CD spectrum in Fig. 3b
was taken after confirming displacement of the aziridine and
coordination of benzylamine by NMR methods (Fig. 2d). This
indicates that chiral memory can be imprinted on the metal
complex with the chiral aziridine. The memory effect is made
possible by the slower rate of exchange of the coordinated phenol
compared to the rate of exchange of the aziridine with
benzylamine. We have also been able to demonstrate the chiral
memory effect purely from 1H NMR without the use of CD
The locked complex (0.5 mol %) was used to catalyze the ring
opening of styrene oxide (1 eq) with N-methylaniline (0.55 eq). The
solvent free reaction was complete in 12 h at ambient temperature
to give the amino alcohol and the unreacted epoxide (Fig. 5a). The
remaining styrene oxide was isolated by fractional distillation and
found to be enantioenriched.
Detection of the enantiomeric excess (ee) of styrene oxide was
1
determined by H NMR and by optical rotation measurements.
Fig. 3 Circular dichroism (CD) spectra in DMSO at 20 uC of (a) ¡1 (25 mM) + (2R,3S) 2-phenyl-3-methylaziridine (50 mM), (b) Same as (a) except after
adding benzylamine (250 mM) to displace the aziridine, (c) ¡1 (25 mM) + (2S,3R) 2-phenyl-3-methylaziridine (50 mM), (d) same as (c) except after adding
benzylamine (250 mM) to displace the aziridine, (e) ¡1 (25 mM), (f) (2S,3R) 2-phenyl-3-methylaziridine (50 mM).
This journal is ß The Royal Society of Chemistry 2007
Chem. Commun., 2007, 120–122 | 121

View Article Online
Fig. 4 1H NMR in DMSO-d6 of one of the imine C–H signals of the locked complex after adding excess R-phenethylamine. Complex imprinted with (a)
D-phenylalanine and (b) L-phenylalanine.
Fig. 5 (a) Schematic representation of the stereoselective ring opening reaction of styrene oxide catalyzed by locked 1 after being imprinted by L-phenyl
alanine. (b) 1H NMR in DMSO-d6 of the C–H signals of the enantioenriched styrene oxide and 0.15 equiv. of NMR shift reagent, europium tris[3-
(heptafluoropropylhydroxy-methylene)-(+)-camphorate].
Chem. Soc., 1997, 119, 5267–5268; (d) Y. Kubo, T. Ohno, J. Yamanaka,
S. Tokita, T. Iida and Y. Ishimaru, J. Am. Chem. Soc., 2001, 123,
12700–12701.
5 T. P. Yoon and E. N. Jacobsen, Science, 2003, 299, 1691–1693.
6 R. Bobb, G. Alhakimi, L. Studnicki, A. Lough and J. Chin, J. Am.
Chem. Soc., 2002, 124, 4544–4545.
7 M. Tokunaga, J. F. Larrow, F. Kakiuchi and E. N. Jacobsen, Science,
1997, 277, 936–938.
8 E. N. Jacobsen, Acc. Chem. Res., 2000, 33, 421–431.
9 T. Katsuki, in Catalytic Asymmetric Synthesis, ed. I. Ojima, Wiley-
VCH, New York, 2nd edn, 2000, ch. 6B.
10 E. N. Jacobsen, in Catalytic Asymmetric Synthesis, ed. I. Ojima, VCH,
New York, 1st edn, 1993, ch. 4.2.
11 M. S. Sigman and E. N. Jacobsen, J. Am. Chem. Soc., 1998, 120,
5315–5316.
12 S. E. Schaus, J. Bra¨nalt and E. N. Jacobsen, J. Org. Chem., 1998, 63,
403–405.
Addition of a small amount of (0.15 eq) europium tris[3-
(heptafluoropropylhydroxymethylene)-(+)-camphorate] resulted
in separation of the three C–H signals of one enantiomer of
styrene oxide from those of the other (Fig. 5b).20 The unreacted
styrene oxide is enriched in the S form (15% ee) if the catalyst is
imprinted with L-phenylalanine. Furthermore, optical rotation
studies performed on the same enantioenriched sample of styrene
oxide gave a reading of +5u. This equates to an ee of 15%, based
on the optical rotation value of pure styrene oxide (¡34u).17 As
expected, D-phenylalanine imprinting agent gave the same value,
equal in magnitude but opposite in sign. While the stereoselectivity
is only modest, this study demonstrates for the first time how
achiral catalysts can be easily imprinted and locked.{
13 Y. N. Belokon, J. Am. Chem. Soc., 1999, 121, 3968–3973.
14 G. M. Sammis and E. N. Jacobsen, J. Am. Chem. Soc., 2003, 125,
4442–4443.
Notes and references
{ Support of this work by NSERC of Canada and a grant from the
MOCIE (Grant No. 10024945) are gratefully acknowledged. Y.S.C thanks
the Korean Ministry of Education for the BK 21 postdoctoral fellowship.
15 M. S. Taylor and E. N. Jacobsen, J. Am. Chem. Soc., 2003, 125,
4442–4443.
16 F. Vo¨gtle and E. Goldschmitt, Chem. Ber., 1976, 109, 1–40.
17 S. E. Schaus, B. D. Brandes, J. F. Larrow, M. Tokunaga, K. B. Hansen,
A. E. Gould, M. E. Furrow and E. N. Jacobsen, J. Am. Chem. Soc.,
2002, 124, 1307–1315.
18 A. Galindo, L. Orea, D. Gnecco, R. G. Enriquez, R. A. Toscano and
W. F. Reynolds, Tetrahedron: Asymmetry, 1997, 8, 2877–2879.
19 C. H. Langford and H. B. Gray, Ligand Substitution Processes, W. A.
Benjamin, New York, 1965.
20 H. J. C. Yeh, S. K. Balani, H. Yagi, R. M. E. Greene, N. D. Sharma,
D. R. Boyd and D. M. Jerina, J. Org. Chem., 1986, 51,
5439–5443.
1 E. Yashima, K. Maeda and Y. Okamoto, Nature, 1999, 399, 449–451.
2 L. J. Prins, F. De Jong, P. Timmerman and D. N. Reinhoudt, Nature,
2000, 408, 181–184.
3 T. Ishi-I, M. Crego-Calama, P. Timmerman, D. N. Reinhoudt and
S. Shinkai, J. Am. Chem. Soc., 2002, 124, 14631–14641.
4 (a) Y. S. Chong, M. D. Smith and K. D. Shimizu, J. Am. Chem. Soc.,
2002, 123, 7463–7464; (b) A. Sugasaki, M. Ikeda, M. Takeuchi,
A. Robertson and S. Shinkai, J. Chem. Soc., Perkin Trans. 1, 1999,
3259–3264; (c) Y. Furusho, T. Kimura, Y. Mizuno and T. Aida, J. Am.
122 | Chem. Commun., 2007, 120–122
This journal is ß The Royal Society of Chemistry 2007