A versatile ruthenium precursor for biphasic catalysis and its application in ionic liquid biphasic transfer hydrogenation: Conventional vs task-specific catalysts

Article abstract of DOI:10.1021/ja048886k

The synthesis of a novel imidazolium-tagged ruthenium complex, which represents a versatile precursor for aqueous and ionic liquid biphasic catalysis, is reported. Its utility is demonstrated in the highly enantioselective ionic liquid biphasic transfer h

Full text of DOI:10.1021/ja048886k

Published on Web 06/11/2004
A Versatile Ruthenium Precursor for Biphasic Catalysis and Its Application in
Ionic Liquid Biphasic Transfer Hydrogenation: Conventional vs Task-Specific
Catalysts
Tilmann J. Geldbach and Paul J. Dyson*
Institute des Sciences et Inge´nierie Chimiques, Ecole Polytechnique Fe´de´rale de Lausanne, EPFL-BCH,
CH-1015 Lausanne, Switzerland
Received February 27, 2004; E-mail: paul.dyson@epfl.ch
Scheme 1a
Ionic liquids are receiving growing attention as a means to
immobilize catalysts, facilitating product isolation and offering an
opportunity to reuse the catalyst, with their potential already
demonstrated in numerous catalytic processes.1 Modification of
known catalysts may, however, be necessary to obtain a catalytic
system which combines high activity with good catalyst retention.
-
Phosphine ligands containing ionic groups such as -SO3 or
-NR3+ probably represent the most common approach to alter the
solubility of a complex to match ionic liquids.
Since the introduction of highly active and selective ruthenium
catalysts based on chiral diamine or amino alcohol ligands, e.g. 1
and 2, transfer hydrogenation has attracted considerable interest.2
Immobilization of the chiral catalyst onto a solid support often
affords good recyclability, but the obtained ee is usually lower than
in the homogeneous or biphasic case.3 Examples for aqueous
biphasic transfer hydrogenation are rare,4 and to date only two
reports on transfer hydrogenation in ionic liquids exist. One
describes microwave-assisted reduction with palladium on charcoal
and formate salts,5 and the other reports on the rhodium-catalyzed
hydrogenation of acetophenone in phosphonium tosylates with
2-propanol as proton source.6 Catalyst recycling is possible in both
cases, but yields decrease considerably with each cycle in the first
example, whereas the high reaction temperatures, viz. 120-150
a Reagents and conditions: (i) 1,2-dimethyl imidazole, toluene, 110 °C;
(ii) NaBF4, CH2Cl2; (iii) RuCl3, MeOH, 80 °C; (iv) (1S,2R)-2-amino-1,2-
diphenylethanol, DMF, rt; (v) (1R,2R)-N-tosyl-1,2-diphenylethylenediamine,
DMF, rt.
°C, in the second case are somewhat problematic and conversion
is usually below 50%. The scarcity of transfer hydrogenation in
ionic liquids might to some extent be due to the fact that only few
will form a separate phase with the common proton source
2-propanol. In addition, the presence of strong basesoften essential
for the reactionsfurther limits the solvent choice.
dazole with 3 and subsequent anion exchange with NaBF4 yields
the functionalized imidazolium salt 4, which itself can be classified
as an ionic liquid (mp ) 85 °C). The latter step is necessary to
increase the solubility for the remaining transformations. Reduction
of RuCl3 with three equivalents of 4 (excess diene is easily
reclaimed after the reaction) in methanol under reflux conditions
leads to the dinuclear complex 5 in good yield.
This methodology offers a significant advantage in that 5 can
be regarded as a versatile starting material for the synthesis of a
wide range of catalysts and conventional, easily accessible ligands
can be used without necessitating further modification. Depending
on the nature of the anion, 5 is insoluble in most common organic
solvents but is highly soluble in water and ionic liquids. As
numerous ruthenium-based catalysts are derived from such dimers,7
it represents an ideal precursor for both aqueous and ionic liquid
biphasic catalysis. To the best of our knowledge, this is the first
time that charged moieties on a complexed arene have been used
to alter the solubility properties of a complex, although ruthenium
dimers with functionalized arenes have been recently synthesized
by other groups.8
Addition of (1S,2R)-2-amino-1,2-diphenylethanol or (1R,2R)-N-
tosyl-1,2-diphenylethylenediamine to 5 in DMF affords the cationic
complexes 6 and 7, respectively, in near quantitative yield. Both
complexes have been tested in the biphasic, enantioselective transfer
hydrogenation of acetophenone and the results compared with those
of the pre-catalysts 1 and 2.
In most examples of ruthenium-catalyzed transfer hydrogenation,
the active catalyst is a neutral 16-electron complex, which is usually
formed in situ from precursors such as 1 and 2 by reaction with
base.2g Hence, the presence of charged ligands may be required to
adequately immobilize the catalyst in an ionic liquid phase. Herein,
we report the synthesis of new ruthenium catalysts with attached
imidazolium tags and compare their performance to well-established
complexes using both 2-propanol and formic acid as the proton
source in the presence of ionic liquids.
The route to the new complexes, bearing η6-arenes with pending
imidazolium tags is depicted in Scheme 1. Starting from phenyl-
ethanol, Birch reduction and subsequent chlorination affords
chloroethyl cyclohexadiene 3. Quaternization of 1,2-dimethylimi-
9
8114
J. AM. CHEM. SOC. 2004, 126, 8114-8115
10.1021/ja048886k CCC: $27.50 © 2004 American Chemical Society

C O M M U N I C A T I O N S
Table 1. Asymmetric Transfer-hydrogenation of Acetophenone in
Table 2. Performance and Recycling of Catalysts 2 and 7 in the
Reduction of Acetophenone with Formic Acid/Triethylamine
Azeotrope as Proton Source; ee g 99 % in all casesa
2-propanol/KOH Catalyzed by Ruthenium Complexes 1, 2, 6 and
7a
c
c
entry
cat.
run
% convb
%ee
% leachingd
2b, %
2b, %
7b, %
7 , %
1
2
3
4
5
6
7
8
9
10
1
1
2
1
2
1
2
1
2
3
4
97
1
95
5
95
15
80
66
57
21
58
61
cycle 1
cycle 2
cycle 3
cycle 4
cycle 5
>99
>99
>99
99
>99
>99
80
>99
68
19
1
>99
>99
80
2
6
7
98
27
98
63
5
45
52
96
a Reaction at 40 °C for 24 h in a 1.5 M solution of acetophenone in
formic acid/triethylamine azeotrope and S/C ) 200; catalyst immobilized
in 0.5 mL of [C4C1C1Im]PF6. b Product extracted with hexane, ionic liquid
washed with H2O, and dried in vacuo. c Product extracted with hexane and
ionic liquid dried in vacuo.
8
a Reaction at 35 °C, 24 h in a 0.1 M solution of acetophenone in
2-propanol/KOH (2 mol %) and S/C ) 200; catalyst immobilized in 1.5
mL of [C4C1C1Im]PF6. b Determined by GC. c Determined by GC using a
Chromapack CP-Cyclodex B column. d Determined by ICP-OES.
enantioselectivity, although this procedure is not applicable to
complex 7, which is too soluble in water and consequently extracted
from the ionic liquid.
In the course of the recycling experiments, the color of the
solution changes from orange to deep red which probably indicates
some decomposition of the catalyst. This is, however, not reflected
in the enantioselectivity, which remains above 99% throughout all
catalytic cycles.
In conclusion, a versatile precursor for aqueous and ionic liquid
biphasic catalysis has been synthesized and tested in biphasic ionic
liquid transfer hydrogenation. With 2-propanol/KOH, the beneficial
effect of additional charged groups in terms of catalyst retention
and recycling was demonstrated. With formic acid, conventional
catalysts afford better results due to a necessary extraction step
with water. Studies to further optimize the reaction conditions and
to apply 5 to other catalytic reactions such as, for example, biphasic
olefin metathesis are currently underway.
In 2-propanol with 5 mol % KOH and in the absence of an ionic
liquid phase, 6 and 7 afford results comparable to those from their
p-cymene analogues, albeit 6 and 7 are slightly less active. However,
the obtained ee with 6 is significantly lower than for its neutral
counterpart 1 (25 vs 58%), which is in accordance with previous
findings that for these amino alcohol complexes, enantioselectivity
is highly dependent on the arene substitution pattern.2f
After screening numerous ionic liquids, 1-butyl-2,3-dimethylimi-
dazolium hexafluorophosphate, [C4C1C1Im]PF6, was chosen for the
immobilization of the catalysts as it forms a phase separate from
2-propanol and appears to be stable toward base. At the beginning
of the reaction, 6 and 7 are found exclusively in the ionic liquid,
whereas the neutral complexes 1 and 2 are present in both phases.
However, during the course of the reaction, leaching does occur,
probably due to base-induced catalyst degradation. Accordingly,
when the amount of base is reduced from five to two equivalents,
catalyst loss is markedly reduced (i.e. from 22 to 5% for 6), and
enantioselectivities increase slightly, although at the cost of reduced
activity. Relative to the neutral complexes 1 and 2, catalyst loss is
up to 10 times lower in the cationic analogues 6 and 7, demonstrat-
ing the positive effect of the imidazolium tag. Yet, even under these
milder, less basic conditions, 1 and 6 appear to become deactivated
quickly, and reuse of the ionic liquid phase is not viable, see Table
1. In contrast, 7 is stable for at least 72 h and recycling of the
ionic liquid phase is feasible, although limited, as conversion drops
from 80% in the first cycle to 21% in the fourth cycle.
Acknowledgment. We thank the EPFL and Swiss National
Science Foundation for financial support.
Supporting Information Available: Experimental procedures,
additional catalysis experiments, and characterization data. This material
is available free of charge via the Internet at http://pubs.acs.org.
References
(1) (a) Wasserscheid, P., Welton, T., Eds. Ionic Liquids in Synthesis; Wiley-
VCH: Weinheim, 2002. (b) Dupont, J.; de Souza, R. F.; Suarez, P. A. Z.
Chem. ReV. 2002, 102, 3667. (c) Gordon, C. M. Appl. Catal., A 2001,
222, 101. (d) Seddon, K. R., Ed. Ionic Liquids as Green SolVents; ACS
Symposium Series 856; American Chemical Society: Washington, DC,
2003.
(2) (a) Noyori, R. Angew. Chem., Int. Ed. 2002, 41, 2008. (b) Rautenstrauch,
V.; Hoang-Cong, X.; Churlaud, R.; Abdur-Rashid, K.; Morris, R. H. Chem.
Eur. J. 2003, 9, 4954. (c) Ell, A. H.; Johnson, J. B.; Ba¨ckvall, J. E. Chem.
Commun. 2003, 1652. (d) Yi, C. S.; He, Z. J.; Guzei, I. A. Organometallics
2001, 20, 3641. (e) Henning, M.; Pu¨ntener, K.; Scalone M. Tetrahedron:
Asymmetry 2000, 11, 1849. (f) Takehara, J.; Hashiguchi, S.; Fujii, A.;
Inoue, S.; Ikariya, T.; Noyori, R. Chem. Commun. 1996, 233. (g) Haack,
K.-J.; Hashiguchi, S.; Fujii, A.; Ikariya, T.; Noyori, R. Angew. Chem.,
Int. Ed. Engl. 1997, 36, 285.
As these results, although promising, are not satisfactory in terms
of catalyst reuse, formic acid/triethylamine azeotrope was used as
a proton source. Complexes bearing amino alcohol ligands are
known to be inactive with this reagent;8b thus, the screening was
restricted to 2 and 7, and both compounds afforded essentially
quantitative conversion and excellent enantioselectivity (>99%).
As this reagent forms a homogeneous phase together with [C4C1C1-
Im]PF6, a product extraction step with hexane or Et2O is necessary.
The remaining solution can be recharged with ketone and formic
acid and reused. However, this gradually leads to increasing reaction
volume and thus decreased turnover frequency due to dilution.
Accordingly, the solution containing the catalyst was washed with
water after product extraction and dried in vacuo prior to the next
catalytic cycle. In this manner, an ionic liquid solution containing
2 could be used five times without significant decrease in activity,
see Table 2. Furthermore, a range of different substrates including
cyclic ketones and aldehydes can be reduced using the same ionic
liquid/catalyst solution (see Supporting Information). The catalyst
solution could be stored for days without any effect on activity or
(3) (a) Saluzzo, C.; Lemaire, M. AdV. Synth. Catal. 2002, 344, 915. (b)
Bianchini, C.; Barbaro, P. Top. Catal. 2002, 19, 17.
(4) (a) Benyei, A.; Joo, F. J. Mol. Catal. 1990, 58, 151. (b) Sa´nchez-Delgado,
R. A.; Medina, M.; Lo´pez-Linares, F.; Fuentes, A. J. Mol. Catal. A: Chem.
1997, 116, 167. (c) Dwars, T.; Oehme, G. AdV. Synth. Catal. 2002, 344,
239. (d) Ogo, S.; Abura, T.; Watanabe, Y. Organometallics 2002, 21,
2964.
(5) Berthold, H.; Schotten, T.; Honig, H. Synthesis 2002, 1607.
(6) Comyns, C.; Karodia, N.; Zeler, S.; Andersen, J. A. Catal. Lett. 2000,
67, 113.
(7) Bennet, M. A. In ComprehensiVe Organometallic Chemistry II; Wilkinson,
G., Stone, F. G. A., Abel, E. W., Eds.; Pergamon: Oxford, 1995; Vol. 7,
p 549.
(8) (a) Soleimannejad, J.; Sisson, A.; White, C. Inorg. Chim. Acta 2003, 352,
121. (b) Hannedouche, J.; Clarkson, G. J.; Wills, M. J. Am. Chem. Soc.
2004, 126, 986. (c) Marconi, G.; Baier, H.; Heinemann, F. W.; Pinto, P.;
Pritzkow, H.; Zenneck, U. Inorg. Chim. Acta 2003, 352, 189.
JA048886K
9
J. AM. CHEM. SOC. VOL. 126, NO. 26, 2004 8115