The importance of complex stability for asymmetric copper-catalyzed cyclopropanations in [emim][OTf] ionic liquid: The bis(oxazoline)- azabis(oxazoline) case

Article abstract of DOI:10.1016/j.tetlet.2004.07.030

Azabis(oxazoline)-Cu complexes are more stable than their analogues based on bis(oxazoline) ligands. This increased stability leads to improved recoverability (up to eight times) when these systems are used in an ionic liquid medium. The solution of the c

Full text of DOI:10.1016/j.tetlet.2004.07.030

Tetrahedron
Letters
Tetrahedron Letters 45 (2004) 6765–6768
The importance of complex stability for asymmetric
copper-catalyzed cyclopropanations in [emim][OTf] ionic liquid: the
bis(oxazoline)–azabis(oxazoline) case
a
Jose M. Fraile, Jose I. Garcıa, Clara I. Herrerıas, Jose A. Mayoral,
a
a
a,
*
´
´
´
´
´
Oliver Reiserb and Michel Vaultierc
a
´ ´
Departamento de Quımica Organica, Instituto de Ciencia de Materiales de Aragon, Facultad de Ciencias,
Universidad de Zaragoza-CSIC, E-50009 Zaragoza, Spain
´
bInstitut fu¨r Organische Chemie, Universita¨t Regensburg, Universita¨tstrasse 31, D-93053 Regensburg, Germany
´
U.M.R. 6510, Institut de Chimie, Universite de Rennes I, Campus de Beaulieu, F-35042 Rennes Cedex, France
c
Received 2 June 2004; revised 5 July 2004; accepted 9 July 2004
Available online 29 July 2004
Abstract—Azabis(oxazoline)–Cu complexes are more stable than their analogues based on bis(oxazoline) ligands. This increased
stability leads to improved recoverability (up to eighttimes) when these systems are used in an ionic liquid medium. The solution
of the chiral catalyst can even be reused in different enantioselective cyclopropanation reactions and still lead to high enantiomeric
excess (>90%).
ꢀ 2004 Elsevier Ltd. All rights reserved.
Enantioselective reactions promoted by chiral catalysts
constitute an area of great interest in organic chemistry.1
The choice between homogeneous and heterogeneous
catalysts is a subject of great debate and is influenced
by the potential advantages associated with each type
of catalyst.
between styrene and ethyl diazoacetate has been studied
and the crucial roles of the IL anion and the presence of
impurities have been demonstrated. In spite of the fact
that moderate yields and high enantioselectivities have
been obtained, catalyst recovery is not problem-free
and the recovery process leads to a noticeable reduction
in enantioselectivity after the catalyst has been reused no
more than twice.
The use of ionic liquids (ILs) offers the possibility of
combining the positive aspects of both homogeneous
and heterogeneous catalysts. On the one hand, the reac-
tion in the IL takes place in a homogeneous phase with
high activity and selectivity. On the other hand, the easy
separation of the products after the reaction makes it
possible to recover and reuse the catalyst, as in the case
of heterogeneous catalysis.2 Comparatively few of these
examples concern the use of chiral catalysts to promote
enantioselective reactions and most deal with hydrogen-
ation3 and oxidation4 processes. In recentyears ht e use
in ILs of catalysts bearing chiral bis(oxazoline) ligands
has noticeably increased. These catalysts have been
tested in Diels–Alder5 and, to a greater extent, in cyclo-
propanation6,7 reactions. Only the benchmark reaction
Leaching of the catalyst or ligand from the ionic liquid
phase markedly reduces the utility of these solvents in
catalytic reactions. In view of this, several authors have
designed new cationic ligands in an attempt to increase
the solubility of the catalyst in the ionic liquid.8 How-
ever, this strategy has seldom been applied to chiral
catalysts.
In the case of bis(oxazoline)–copper catalysts, both the
complex and the metal precursor are quite soluble in
the ionic liquid and are not leached from this phase.
The main problem with these systems arises from the
equilibrium between complexed and free ligand. The lat-
ter is extracted during the product extraction process
and the presence of free copper precursor in the ionic
liquid causes a reduction in the enantioselectivity. This
effectbecomes more pronounced afetr each recovery
cycle (Fig. 1). One option to solve this problem involves
Keywords: Asymmetric catalysis; Ionic liquids.
*
Corresponding author. Tel./fax: +34-976-762077; e-mail: mayoral@
unizar.es
0040-4039/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2004.07.030

6766
J. M. Fraile et al. / Tetrahedron Letters 45 (2004) 6765–6768
%
ments in the reusability of chiral cyclopropanation cata-
lysts when used in an IL.
100
Experimental binding constants are difficult to measure
for these copper complexes and so we estimated the rel-
ative coordinating ability of the two ligands by theoret-
ical calculations. To this end, we calculated the
equilibrium between the bis(oxazoline)–copper complex
and free azabis(oxazoline) (Scheme 1), where the copper
complexes correspond to the catalytic precursor.10 Sol-
vent effects are difficult to take into account in ILs.
However, in an attempt to gain some insight into the
possible effectof solventon ht e above equilibrium, we
considered the endo/exo selectivity results of the reaction
80
After supplementary
addition of chiral ligand
60
40
20
between cyclopentadiene and methyl acrylate carried
14
run
0
2
4
6
8
10
outin [emim][OTf].
This selectivity has previously
been used to define the solvent polarity parameter X,15
which has proven to be mainly related to the solvent di-
polarity in the case of aprotic solvents.16 The X value
measured in an IL corresponds to a polarity close to that
of acetonitrile or nitromethane.15 We therefore used the
dielectric permittivity of acetonitrile to estimate solvent
effects.17 The results obtained show that the equilibrium
is shifted to the right-hand side by 5.7kcalmolÀ1, which
is consistent with a higher coordinating ability of the
azabis(oxazoline) ligand.
Figure 1. Reuse of the solutions of azabis(oxazoline)–CuCl (1-CuCl
(––)) and bis(oxazoline)–CuCl (2-CuCl (Á Á Á)) in [emim][OTf]: (j) %ee
for trans isomers; (r) %yield.
the addition of a further amount of chiral ligand to the
IL solution,6 as shown in the sixth run with the bis(ox-
azoline)–copper catalyst in Figure 1. Another possibility
is to use an excess of ligand from the outset. However,
this is not an ideal solution from the point of view of
ligand economy. The introduction of an ionic group into
the bis(oxazoline) ligand, in order to increase its solubil-
ity in the IL, is not synthetically straightforward and
may have unpredictable effects on the enantioselectivity,
as demonstrated by the significant reduction in the
enantioselectivity observed after modification of a Salen
ligand.9 A more suitable strategy would therefore
involve modification of the chiral ligand to increase
the binding constant of the complex, but without
modifying the structural properties responsible for the
enantioselectivity.10
The copper(I) complexes of N,N-bis[(S)-4-tert-butyl-4,5-
dihydro-1,3-oxazol-2-yl]methylamine18 (1) and the anal-
ogous bis(oxazoline) 2 were tested in the aforementioned
benchmark cyclopropanation reaction between styrene
(3a) and ethyl diazoacetate with [emim][OTf] as the IL
solvent( Scheme 2).19 In a previous study, CuII com-
plexes were used despite the fact that CuI is the active
species.10 This modification is due to the higher stability
of Cu(OTf)2 in comparison with CuOTf and the role of
the counter ion, which precludes the use of CuCl. How-
ever, CuCl can be used in the IL because chloride is re-
placed by the IL counter ion.6 The advantages of
azabis(oxazoline) in comparison with bis(oxazoline) as
the chiral ligand are represented in Figure 1. The enan-
tioselectivity is consistently high at 90–92% ee for the
trans isomer (4) in a series of eightreactions and is also
stable for the cis isomers (5) at82% ee (notshown). The
chemical yield decreases from 62% to 40% in the first
reuse, but then remains constant in the range 35–40%
Azabis(oxazoline) ligands (1)11 (Scheme 1) that bear an
electron-donating group in the central bridge were envi-
sioned as more coordinating ligands than bis(oxazo-
lines) (2), which contain an isopropylidene bridge.
Indeed, these ligands led to better results when cationic
copper complexes were immobilized onto anionic sup-
ports.12,13 In this communication we describe how the
use of azabis(oxazoline) ligands gives rise to improve-
Scheme 1. Equilibrium between free and complexed bis(oxazoline) and
azabis(oxazoline) ligands.
Scheme 2. Cyclopropanation reactions catalyzed by azabis(oxazo-
line)–copper and bis(oxazoline)–copper complexes.

J. M. Fraile et al. / Tetrahedron Letters 45 (2004) 6765–6768
6767
up to the eighth reaction. Given that total conversion of
the trans cyclopropanes (80% ee) buta higher level in
the cis compounds (98% ee).
ethyl diazoacetate occurs, the decrease in the yield of cy-
clopropanes is due to an increase in the competitive di-
azoacetate dimerization and subsequent reactions of the
resulting maleate and fumarate. The high molecular
weight by-products are probably not extracted com-
pletely into the hexane phase and they are then able to
poison the catalyst, thus favouring the non-catalyzed
dimerization. The recoverability of the azabis(oxazol-
ine)-based catalyst confirms that the copper is more
strongly bound to this ligand than to the corresponding
bis(oxazoline).
In summary, it has been demonstrated that azabis(ox-
azolines) have clear advantages over bis(oxazolines)
for use as chiral ligands in enantioselective reactions
carried out in an IL. These advantages are due to
the electron-donating properties of the aza bridge, which
increases the stability of the copper complex and
improves the reusability of the chiral catalyst solution.
Acknowledgements
After the eighth reaction, the IL solution containing the
chiral azabis(oxazoline)–copper complex was leftin ht e
open air for 72h in order to assess its stability. As can
be seen from Figure 1, the absorption of moisture due
to the hygroscopic character of [emim][OTf] causes a
noticeable decrease in both yield and enantioselectivity.
The detrimental effect of water has already been de-
scribed for bis(oxazoline)–copper complexes.6 This phe-
nomenon was attributed to the competitive coordination
of water and chiral ligand, a process that leads to non-
chiral catalytic copper centres. This behaviour can be re-
versed by simply drying the [emim][OTf] solution under
vacuum and the change is possibly due to the negligible
vapour pressure of the IL. Evidence for this reversibility
can be seen from the tenth reaction in Figure 1.
This work was made possible by the generous financial
support of the CICYT (project PPQ2002-04012), the
´
MCYT-DAAD (Accion Integrada HA2001-0096) and
the DGA. C.I.H. is indebted to the MCYT for a grant.
References and notes
1. Comprehensive Asymmetric Catalysis; Jacobsen, E. N.,
Pfaltz, A., Yamamoto, H., Eds.; Springer: Berlin–Heidel-
berg, 1999.
2. Wasserscheid, P. In Ionic Liquids in Synthesis; Wassersc-
heid, P., Welton, T., Eds.; Wiley-VCH: Weinheim, 2000;
p 213.
3. (a) Chauvin, Y.; Mussmann, L.; Olivier, H. Angew. Chem.,
Int. Ed. Engl. 1995, 34, 2698; (b) Monteiro, A. L.; Zinn, F.
K.; de Souza, R. F.; Dupont, J. Tetrahedron: Asymmetry
1997, 8, 177; (c) Berger, A.; de Souza, R. F.; Delgado, M.
R.; Dupont, J. Tetrahedron: Asymmetry 2001, 12, 1825; (d)
Guernik, S.; Wolfson, A.; Herskowitz, M.; Greenspoon,
N.; Geresh, S. Chem. Commun. 2001, 2314; (e) Brown, R.
A.; Pollet, P.; McKoon, E.; Eckert, C. A.; Liotta, C. L.;
Jessop, P. G. J. Am. Chem. Soc. 2001, 123, 1254.
4. (a) Song, C. E.; Roh, E. J. Chem. Commun. 2000, 837; (b)
Song, C. E.; Jung, D.; Roh, E. J.; Lee, S.; Chi, D. Y.
Chem. Commun. 2002, 3038; (c) Branco, L. C.; Afonso, C.
A. M. Chem. Commun. 2002, 3036.
It would be of great interest to ascertain whether the
same catalyst solution, which can be considered as a liq-
uid phase-immobilized catalyst, could be used in differ-
ent cyclopropanation reactions. With this aim in mind,
the solution of azabis(oxazoline)-CuCl in [emim][OTf]
used in the cyclopropanation of styrene, was recovered
and reused in successive cyclopropanations with differ-
ent alkene substrates; namely 1-octene (3b), a-methylstyr-
ene (3c) and 1,1-diphenylethylene (3d) (Scheme 2). The
solution was then finally reused in the reaction with styr-
ene in order to assess its performance in the benchmark
reaction initially studied. The results of these experi-
ments are gathered in Table 1. As can be seen, all the
yields were acceptable considering that only 1equiv of
alkene was used. The only exception involved the use
of the aliphatic alkene 1-octene and this is probably
due to its lower reactivity. The high enantioselectivity
obtained in the five runs is remarkable and values over
90% were routinely obtained even in the final reuse with
styrene. Once again, the only exception is 1-octene,
which led to a somewhat lower enantioselectivity in
5. Meracz, I.; Oh, T. Tetrahedron Lett. 2003, 44, 6465.
´
6. (a) Fraile, J. M.; Garcıa, J. I.; Herrerıas, C. I.; Mayoral, J.
A.; Carrie, D.; Vaultier, M. Tetrahedron: Asymmetry 2001,
´
´
´
´
12, 1891; (b) Fraile, J. M.; Garcıa, J. I.; Herrerıas, C. I.;
Mayoral, J. A.; Gmough, S.; Vaultier, M. Green Chem.
2004, 6, 93.
7. Davies, D. L.; Kondola, S. K.; Patel, R. K. Tetrahedron:
Asymmetry 2004, 15, 77.
8. (a) Brasse, C. C.; Englent, U.; Salzer, A.; Waffenschmidt,
H.; Wasserscheid, P. Organometallics 2000, 19, 3818; (b)
Sirieix, J.; Ossberger, M.; Betzemeier, B.; Knochel, P.
Synlett 2000, 1613; (c) Wasserscheid, P.; Waffenschmidt,
H.; Machnitzki, P.; Kottsieper, K.; Stelzer, O. Chem.
Commun. 2001, 451; (d) Kottsieper, K. W.; Stelzer, O.;
Wasserscheid, P. J. Mol. Catal. A 2001, 175, 285; (e)
Brauer, D. J.; Kottsieper, K. W.; Liek, C.; Stelzer, O.;
Waffenschmidt, H.; Wasserscheid, P. J. Organomet. Chem.
2001, 630, 177; (f) Favre, F.; Olivier-Bourbigou, H.;
Commereuc, D.; Saussine, L. Chem. Commun. 2001, 1360.
9. Baleizao, C.; Gigante, B.; Garcia, H.; Corma, A. Tetra-
hedron Lett. 2003, 44, 6813.
Table 1. Recycling of the 1-CuCl solution in [emim][OTf] for the
enantioselective cyclopropanation of different alkenesa
Run
Alkene
Yield (%)
4/5
%ee 4b
%ee 5b
1
2
3
4
5
3a
3b
3c
3d
3a
62
20
44
52
30
73:27
71:29
57:43
––
91
80
91
92
91
82
98
84
––
83
73:27
´
´
10. Fraile, J. M.; Garcıa, J. I.; Martınez-Merino, V.; Mayoral,
J. A.; Salvatella, L. J. Am. Chem. Soc. 2001, 123,
7616.
a Results determined from the hexane extract by gas chromatography
on 5% phenylsilicone and cyclodex-B columns.
b Compounds 4R and 5R were the major enantiomers.
11. Glos, M.; Reiser, O. Org. Lett. 2000, 2, 2045.

6768
J. M. Fraile et al. / Tetrahedron Letters 45 (2004) 6765–6768
´
´
12. Fraile, J. M.; Garcıa, J. I.; Harmer, M. A.; Herrerıas, C. I.;
Mayoral, J. A.; Reiser, O.; Werner, H. J. Mater. Chem.
2002, 12, 3290.
level. Solvent effects were considered through the PCM
continuum model.
18. Werner, H.; Vicha, R.; Gissibl, A.; Reiser, O. J. Org.
Chem. 2003, 68, 10166.
´
13. Fraile, J. M.; Garcıa, J. I.; Herrerıas, C. I.; Mayoral, J. A.;
Reiser, O.; Socuellamos, A.; Werner, H. Chem. Eur. J.
2004, 10, 2997.
14. Sethi, A.; Welton, T.; Wolff, J. Tetrahedron Lett. 1999, 40,
793.
15. Berson, J. A.; Hamlet, Z.; Mueller, W. A. J. Am. Chem.
Soc. 1962, 84, 297.
´
16. (a) Ruiz-Lopez, M. F.; Assfeld, X.; Garcıa, J. I.; Mayoral,
J. A.; Salvatella, L. J. Am. Chem. Soc. 1993, 115, 8780; (b)
´
´
19. The complexes were prepared in situ by dissolving CuCl
(0.038mmol) and the ligand (0.038mmol) in the IL
(0.5mL). The mixture was stirred at room temperature
until a clear pale green solution was obtained. To this
solution, under an Ar atmosphere, was added styrene
(3.8mmol). In some cases a clear solution was obtained
only after the addition of styrene. Ethyl diazoacetate
(3.8mmol) was slowly added (2h) using a syringe pump.
The reaction mixture was stirred at room temperature for
20h. After this time the products were extracted with
hexane (3 · 4mL) and n-decane (100mg) was added to the
hexane solution as an internal standard for the GC
analysis. The remaining solution of the catalyst in the IL
was reused following the same method.
´
´
Cativiela, C.; Garcıa, J. I.; Mayoral, J. A.; Salvatella, L. J.
Chem. Soc., Perkin Trans. 2, 1994, 847; (c) Cativiela, C.;
´
Garcıa, J. I.; Mayoral, J. A.; Salvatella, L. Chem. Soc.
Rev. 1996, 25, 209.
17. All theoretical calculations were carried out with the
GAUSSIAN 03 program atthe B3LYP/6-31G*//HF/6-31G*