Mixtures of ionic liquids and water as a medium for efficient enantioselective hydrogenation and catalyst recycling

Article abstract of DOI:10.1002/adsc.200404113

In a screening of ligands, ionic liquids and reaction conditions in the Rh-catalyzed hydrogenation of enamides, a novel multi-phase reaction system consisting of an ionic liquid (IL) and water (wet ILs) was found to give the most promising results. In many cases such IL/water combinations were superior compared to conventional organic solvents and biphasic ILs/organic co-solvents media with respect to catalytic performance as well as to catalyst separation and recycling. So far, the best results were obtained with Rh-ferrocenyl- diphosphine catalysts (>99% ee). Generally, somewhat lower ees were observed at higher pressure. However, this effect was less pronounced with wet ILs than with conventional solvents. It is shown that IL/water combination allow repeated catalyst recycling without significant loss of activity and that industrially relevant turnover numbers of > 10,000 can be obtained.

Full text of DOI:10.1002/adsc.200404113

FULL PAPERS
Mixtures of Ionic Liquids and Water as a Medium for Efficient
Enantioselective Hydrogenation and Catalyst Recycling
B. Pugin,a M. Studer,a E. Kuesters,b G. Sedelmeier,b X. Fenga,*
a
Solvias AG, Postfach, 4002 Basel, Switzerland
Fax: (þ41)-61-686-6311, e-mail: xiangdong.feng@solvias.com
b
Novartis Pharma AG, Postfach, 4002 Basel Switzerland
Fax: (þ41)-61-324-5274, e-mail: ernst. kuesters@pharma. novartis. com
Received: April 19, 2004; Revised: September 10, 2004; Accepted: September 13, 2004
Abstract: In a screening of ligands, ionic liquids and Generally, somewhat lower ees were observed at
reaction conditions in the Rh-catalyzed hydrogena- higher pressure. However, this effect was less pro-
tion of enamides, a novel multi-phase reaction system nounced with wet ILs than with conventional sol-
consisting of an ionic liquid (IL) and water (wet ILs) vents. It is shown that IL/water combination allow re-
was found to give the most promising results. In many peated catalyst recycling without significant loss of ac-
cases such IL/water combinations were superior com- tivity and that industrially relevant turnover numbers
pared to conventional organic solvents and biphasic of >10,000 can be obtained.
ILs/organic co-solvents media with respect to catalytic
performance as well as to catalyst separation and re-
cycling. So far, the best results were obtained with Keywords: asymmetric hydrogenation; catalyst recy-
Rh-ferrocenyl-diphosphine catalysts (>99% ee). cling; homogeneous catalysis; ionic liquids; rhodium
Introduction
were selected because they have different polarities
and miscibilities with various inorganic and organic sol-
In the last few years, ionic liquids have been attracting vents and because they are supposed to be non-coordi-
much attention as “green and designer” media for chem- nating.[6] Pyridinium ILs and phosphonium ILs were
ical reactions.[1] Their application in catalysis has been considered but not chosen, because the first are solids
extensively explored and found to be promising,[2,3] at room temperature and the latter gave only very
also in enantioselective hydrogenations.[4,5]
poor conversions in preliminary hydrogenation experi-
The goal of this work was to investigate the scope of ments.
ILs in enantioselective hydrogenation with various li-
gands/catalysts with respect to catalytic performance
and to product/catalyst separation and to identify suita- Results and Discussion
ble reaction systems and conditions. To study the effects
of ILs we performed a screening with various rhodium- Before starting our screening, we investigated the influ-
diphosphine catalysts, using the two standard enamides ence of the quality of the commercial ILs. Significant dif-
test substrates methyl a-acetamidoacrylate (2a; MAA) ferences in catalyst performance could be observed
and methyl a-acetamidocinnamate (2b; MAC) when different batches of ILs were used. We found no
(Scheme 1), seven classes of diphosphine ligands correlation between catalyst performance and the col-
(Fig. 1) and four imidazolium ILs (Fig. 2). These ILs our of the ILs, which can vary from yellow to deep or-
ange depending on the batch. However, formation of a
precipitate upon addition of aqueous AgNO3 indicated
that many batches of ILs were contaminated with signif-
icant amounts of ClÀ or BrÀ anions, which have an im-
portant effect on catalyst performance. As a conse-
quence, all commercial ILs were purified as described
in the literature[7] before use. With such purified ILs, re-
producibility was no longer a problem.
In Table 1 we summarize the results of hydrogenations
that were carried out with the ferrocenyl-diphosphine li-
Scheme 1. Enantioselective hydrogenation of enamides.
Adv. Synth. Catal. 2004, 346, 1481–1486
DOI: 10.1002/adsc.200404113
ꢀ 2004WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1481

B. Pugin et al.
FULL PAPERS
Figure 1. Structures of the ligands used for enantioselective hydrogenation in this study.
biphasic reaction system was studied. We found that cat-
alyst separation and recycling by simple decantation is in
principle feasible, but unsatisfactory, since in most cases
significant catalyst leaching occurred, as indicated by
the yellow colour of the co-solvent phase (Table 1, en-
1
tries 7 and 8). In addition, H NMR measurements re-
vealed significant leaching of ILs into the i-PrOH phase.
The promising results in the MeOH-H2O mixtures
mentioned above prompted us to study combinations
of ILs with water. Mixtures of ILs and water in any ratio
Figure 2. Ionic liquids used in this study.
gands 1a and 1b.[8] The classical solvents for cationic have been called “wet ILs” in the literature[12] and can be
Rh-catalysts are alcohols (most often methanol). Refer- single or multi-phasic systems. Such “wet ILs”have been
ence reactions without ILs are given for the same ligands shown to give good results in several reactions such as
(Table 1, entries 1–4, 12, 13, 20, 26). Experiments with fluorination with metal fluorides,[13] catalytic hydroge-
ligand 1a illustrate the influence of the polarity of the nation of alkynes,[14] and catalytic epoxidation of a,b-un-
solvent on catalytic performance: In alcohols good activ- saturated carbonyl compounds.[15] In addition, several
ities and enantioselectivties are obtained but they drop studies indicate that the presence of water can strongly
significantly with decreasing solvent polarity (Table 1, affect the physical properties of ILs.[7,16]
entries 1–4). It is also interesting to note that addition
of water to MeOH has no negative effect (Table 1, en- tion has only been mentioned for the Ru-catalyzed hy-
try 1). drogenation of tiglic acid, where wet [bmim]PF6 was re-
The use of “wet ILs” in enantioselective hydrogena-
ILs have been shown to have similar polarities as low ported to give better ees (up to 92%) than pure
alcohols.[9,10] Indeed, the reactions in ILs gave similar ees [bmim]PF6 (88%) or i-PrOH/[bmim]PF6 (up to
as in alcohol, but significantly lower rates (Table 1, en- 40%).[12] This beneficial effect was attributed to a lower
tries 5, 6, 14, 21). It is known, that ILs have much lower hydrogen concentration in [bmim]PF6 and wet
hydrogen solubilities and higher viscosities than alco- [bmim]PF6 than in the organic solvent. Unfortunately,
hols.[11] The lower rates may therefore be caused by a this report gives only very little information about cata-
lower hydrogen concentration due to limited mass trans- lyst activity.
port.
We tested this concept with two different biphasic sys-
Addition of i-PrOH as co-solvents to [bmim]BF4 and tems. We thereby avoided the use of PF6-ILs which are
[bmim]PF6 yields a known biphasic reaction system. In known to easily decompose in water.[17] We chose the
this medium the rates were generally higher than in more stable [omim]BF4 and [bmim]Tf2N which both
pure ILs but lower than in alcohols (Table 1, entries 7, have a poor solubility in water, in order to prevent a pos-
8, 15, 16 and 22). Furthermore, the ees were significantly sible leaching of the IL into the water-phase and to sim-
lower in several cases (Table 1, entries 7 and 8). The pos- plify the following work-up. We also worked with anoth-
sibility to immobilize the catalysts in the IL phase of this er biphasic system, where one phase was a mixture of the
1482
ꢀ 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2004, 346, 1481–1486

Mixtures of Ionic Liquids and Water
FULL PAPERS
Table 1. Enantioselective hydrogenation of enamides catalyzed by Rh-Taniaphos in various Ils/water combinations and in con-
ventional solvents.
Entry
Reaction Medium
Conversion [%]
ee [%]
Catalyst Separation
No of Phases
Reactions using ligand 1a, substrate 2a
1
2
MeOH-H2O[a]
MeOH
100
100
92
98
97
95
–
–
–
1
1
1
3
i-PrOH
4Toluene
7
4
28
–
1
5
6
7
8
9
[bmim]BF4
[bmim]PF6
[bmim]BF4/i-PrOH
[bmim]PF6/i-PrOH
[omim]BF4/H2O
32
6
6496
12
100
97
>99
91
–
–
1
1
2
2
2
2
2
þ
93
>99
>99
>99
þ
þ þ
þ þ
þ þ
10
11
Reactions using ligand 1b, substrate 2a
[bmim]Tf2N/H2O
[bmim]BF4-H2O[b]/Toluene
100
12
MeOH
53
97
–
1
13
i-PrOH
8495
17
43
47
70
–
1
14[bmim]BF
98
99
98
>99
>99
>99
–
1
2
2
2
2
2
4
15
16
17
18
19
[bmim]BF4/i-PrOH
[omim]BF4/i-PrOH
[omim]BF4/H2O
þ þ
þ þ
þ þ*
þ þ
þ þ
[bmim]Tf2N/H2O
90
100
[bmim]BF4-H2O[b]/Toluene
Reactions using ligand 1a, substrate 2b
20
21
22
23
i-PrOH
[bmim]BF4
[bmim]BF4/i-PrOH
[omim]BF4/H2O
100
52
100
100
100
100
94–
93
93
95
94
95
1
–
1
2
2
2
3
þ
þ þ
þ þ
þ þ
24[bmim]BF
25
Reactions using ligand 1b, substrate 2b
4-H2O[b]/Toluene
[omim]BF4/H2O[c]/Toluene
26
27
i-PrOH
[omim]BF4/H2O
100
100
99
>99
–
þ þ
2
2
ILs: ca. 2 mL, co-solvents: ca. 2–3 mL, S/C¼200, [S]¼0.25 M in co-solvent, room temperature, P(H2)¼1 bar, and t¼20 min.
“–” no, “þ” good (some leaching), “þ þ” excellent (no leaching).
[a]
v:v¼4 : 1.
[b]
v:v¼6: 1.
[c]
v:v¼3: 1.
* ICP-MS: 0.9 ppm Rh content in co-solvent.
water-soluble IL [bmim]BF4 and water and the other a
hydrophobic organic solvent such as toluene.
It is generally known that the ligand is one of the most
important factors that governs the performance of a cat-
To our surprise, hydrogenation of 2a with ligands 1a alyst. To investigate the scope of this new reaction me-
and 1b in [omim]BF4/H2O and [bmim]Tf2N/H2O combi- dium, a series of further ligands with and without ferro-
nations afforded in most cases improved ees and/or re- cenyl backbones 1c–1k (see Figure 1) were tested in wet
action rates compared to conventional organic solvents ILs. The results are given in Table 2.
(Table 1, entries 9 and 10, 17 and 18, 23, and 27). In ad-
The results indicate that the effect of ILs is strongly
dition, the colourless water-phases and orange coloured dependent on the ligands used. While all ferrocene-
ILs phases indicated no or only very low catalyst leach- based ligands (1c–1 h)[19] give better ees and/or activi-
ing.[18]
ties in wet ILs than in MeOH (Table 2, entries 1–16),
Since for more lipophilic substrates such as 2b water is the ligands with other backbones (1i–1k) give inferior
not a good solvent, ternary combinations of ILs, water results (Table 2, entries 17–25). So far, we have no ex-
and an organic solvent (toluene) were also tested (Ta- planation for this very different behaviour. With respect
ble 1, entries 11, 19, 24and 25). Here again, excellent re-
to catalyst separation and catalyst recycling, the IL-wa-
sults with respect to catalytic performance as well as to ter system is superior to the conventional IL/isopropyl
catalyst separation were obtained.
alcohol system for all ligands. This can be explained by
Adv. Synth. Catal. 2004, 346, 1481–1486
asc.wiley-vch.de
ꢀ 2004WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1483

B. Pugin et al.
No of phases
FULL PAPERS
Table 2. Enantioselective hydrogenation of 2a (MAA) with selected chiral diphosphine ligands.
Entry
Ligand
Reaction Medium
Conversion [%]
ee [%]
Catalyst Separation[a]
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
1c
1c
1c
1d
1d
1d
1d
1e
1e
1f
MeOH
8
23
26
100
22
100
70
99
27
100
100
100
95
48
29
75
97
82
>99
>99
14–
63
71
72
93
94
95
97
99
87
66
68
–
1
2
2
1
2
2
2
[bmim]PF6/i-PrOH
[omim]BF4/H2O
MeOH
[bmim]PF6/i-PrOH
[omim]BF4/H2O
[omim]BF4/H2O
MeOH
[omim]BF4/H2O
MeOH
[omim]BF4/H2O
MeOH
[omim]BF4/H2O
MeOH
[bmim]BF4/i-PrOH
[omim]BF4/H2O
MeOH
[bmim]BF4/i-PrOH
[omim]BF4/H2O
MeOH
[bmim]BF4/i-PrOH
[omim]BF4/H2O
MeOH
þ
þ þ
–
–
þ þ
þ þ
1
b
þ þ
2
2
2
1
2
1
2
2
1
2
2
1
2
2
1
2
2
–
1f
þ þ
–
1g
1g
1h
1h
1h
1i
1i
1i
1j
1j
þ þ
63
96
–
þ
100
100
100
97
100
100
90
100
87
70
þ þ
–
þ
þ þ
–
81
73
68
99
99
99
þ
1j
þ þ
–
1k
1k
1k
[bmim]BF4/i-PrOH
[omim]BF4/H2O
þ
þ þ
ILs: ca. 2 mL, co-solvents: ca. 2–3 mL, S/C ¼200, [S] ¼0.25 M in co-solvent, 258C, P(H2)¼1 bar, and t¼20 min.
[a]
“–” no, “þ” good (some leaching), “þ þ” excellent (colourless co-solvent phase).
[b]
Performance of catalyst that has been re-used 6 times.
* ICP-MS: 0.9 ppm Rh content in co-solvent.
the solubility of the catalysts which is generally much
lower in water than in alcohols if ligands without highly effect, a series of hydrogenations with selected ligands
polar or charged groups are used. were compared at low (1 bar) and at high (10 bar) hydro-
To get further insight into this hydrogen concentration
The hydrogen concentration in the liquid phase is an- gen pressure in wet ILs and in MeOH. The reactions
other factor that affects the reaction rate and, depending were stopped before completion, so that average
on the ligand, in many cases also the ee.[4,12,20] The higher TOFs could be estimated. The results are given in Ta-
ees observed with many ligands in wet ILs compared to ble 3. It is interesting to note that generally the ees
organic solvents raises the question of whether this may tend to be lower at 10 bar than at 1 bar and that the de-
be due to a different hydrogen concentration in wet ILs crease is less important in wet ILs than in organic sol-
or to a solvent effect such as a different solvation of the vents. This effect is most pronounced with ligand 1d (en-
catalyst and reactants.
tries 5–8) where differences in ees of 13% in MeOH and
The hydrogen concentration depends on factors such 7% in wet IL are observed. As expected, the average hy-
as hydrogen solubility in the solvent, applied hydrogen drogenation rates increase with increasing hydrogen
pressure and mass transfer effects. As mentioned above, pressure. The fact that with only one exception (entry 6)
hydrogen has a lower solubility in ILs than in organic all rates and ees are higher in wet ILs than in methanol
solvents. Accordingly, as can be seen in Table 1, all hy- strongly indicates that this cannot be due tolower hydro-
drogenations that were performed in alcohols are faster gen availability but that this effect must rather be caused
than those in pure ILs. Interestingly, the results in Ta- by a different solvation of the reactants in the wet ILs.
ble 1 also show that the reaction rates in wet ILs are sim-
An important motive to use ILs is the possibility to
ilar to those in organic solvents. Since the hydrogen sol- separate and reuse the catalyst. Hydrogenation of 2a
ubility is also lower in water than in organic solvents this in [omim]BF4/H2O with the Rh-1d complex was chosen
is a surprising result, demonstrating that a mixture of a as a model reaction to study catalyst recycling and the
IL and water can have different properties than its distribution of the product in the two phase system. Us-
individual components.
ing our standard reaction conditions (see Table 1) the
1484
ꢀ 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2004, 346, 1481–1486

Mixtures of Ionic Liquids and Water
FULL PAPERS
Table 3. Hydrogenation of MAA in MeOH and H2O/[omim]BF4 at 1 bar and 10 bar hydrogen pressure.
Entry
Ligands
Solvents
Ionic liquids
TOF[a] [1/h]
ee [%]
P(H2) bar
1
2
3
4
5
6
7
8
9
1a
1a
1a
1a
1d
1d
1d
1d
1b
1b
1b
1b
MeOH
H2O
MeOH
H2O
MeOH
H2O
MeOH
H2O
MeOH
H2O
MeOH
H2O
No
[omim]BF4
No
[omim]BF4
No
[omim]BF4
No
[omim]BF4
No
[omim]BF4
No
480
816
1645
2229
1536
636
2365
2846
318
98
99
95
98
97
>99
8410
92
97
>99
96
1
1
10
10
1
1
10
1
1
10
10
10
11
12
420
1371
3085
[omim]BF4
>99
S/C¼200, [S in co-solvent]¼0.25 M, room temperature.
[a]
Average TOF during 1–5 minutes reaction time.
catalyst gave complete conversion and ees>99% in 5 within 3 hours and yielded the product 3a with 95%
consecutive runs. In the 6th run some catalyst deactiva- ee. A colourless water and a yellow IL phase indicated
tion was observed (Table 2, entry 7), nevertheless, a sev- no significant leaching. Interestingly the catalyst in the
enth run with complete conversion and unchanged ee IL phase was less but still active and could be reused
could be achieved when the reaction time was increased for a second run without loss of enantioselectivity.
from 20 minutes to 40 minutes. In all runs, the separated
aqueous phases were perfectly colourless. After the first
run (entry 6), the Rh content in the water phase was
checked by ICP-MS and found to be as low as 0.9 ppm.
Conclusion
This corresponds to a catalyst leaching of only 0.5%.
Based on these observations we attribute the observed
loss of activity to catalyst deactivation, e.g., by contam-
ination with traces of air during work-up, rather than
to catalyst leaching.
Catalyst separation was performed by simple decant-
ation of the colourless water phase. According to NMR
and GC measurements 35–40% of the polar hydrogena-
tion product 3a remained in the IL phase. However, af-
ter the last run, the product could efficiently be extract-
ed from the IL phase with t-butyl methyl ether.
Wet ILs constitute a new class of solvents with their own
new and interesting properties. This reaction medium
has an interesting potential in enantioselective catalysis
since it can give excellent catalytic performance, also at
very high S/C ratios, and allow efficient catalyst separa-
tion and easy recycling. Not surprisingly, our screening
showed that the effect of ILs on catalytic performance
is not predictable and strongly depends on the ligand/
catalyst system used.
For substrates with very low water solubility such as
2b, wet ILs/apolar co-solvents combinations were
used. Depending on the ILs and on the amount of water
added, such systems are either biphasic {[bmim]BF4-
H2O (v:v 6:1)/toluene; Table 1, entry 24} or triphasic
{[omim]BF4/H2O (v:v 3:1)/toluene; entry 25}. In both
systems the hydrogenations were fast and yield the lipo-
philic product 3b with high ees. Here, the product 3b
Experimental Section
General Procedure for the Hydrogenations and for
Catalyst Recycling
The ionic liquids [bmim]BF4, [bmim]PF6 and [omim]BF4 were
purchased from Solvent Innovation and purified as described
in literature[7] before use. The ionic liquid [bmim]Tf2N was ob-
could be isolated efficiently and without visible catalyst tained from Dr. R. Giernoth (the Institute for Organic Chem-
istry, University of Cologne) and directly used without further
purification. The ligands BPPM, Prophos and Me-Duphos
were purchased from STREM Chemicals. All manipulations
were performed under Ar and with degassed solvents.
The hydrogenations at 1 bar were performed in a 25 mL vial
equipped with a septum, a connection to Ar, vacuum and H2
and a magnetic stirring bar. To a solution of the ligand
(0.0034mmol) in 2 mL IL, a solution of [Rh(NBD) 2]BF4
(0.003 mmol) and a solution of the substrate in dichlorome-
thane (total approx. 3 mL) was added. The dichloromethane
contamination by direct decantation of co-solvent(s).
To be of industrial interest, enantioselective Rh cata-
lysts should achieve turnover numbers in the range of
several thousands. To the best of our knowledge, enan-
tioselective hydrogenations in ILs so far have only
been performed with low S/C ratios (typically<100).
To check the limits of the wet IL system, we carried
out a hydrogenation of 2a with ligand 1b with an S/C ra-
tio of 10,000 in [omim]BF4/H2O at 10 bar hydrogen pres-
sure. The hydrogenation gave quantitative conversion was removed under reduced pressure and the co-solvent(s)
Adv. Synth. Catal. 2004, 346, 1481–1486
asc.wiley-vch.de
ꢀ 2004WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1485

B. Pugin et al.
FULL PAPERS
added. Typically 2 mL IL and 2–3 mL co-solvent were used.
Then the reaction mixture was set under H2 (1 bar) and the hy-
drogenation started at room temperature by switching on the
magnetic stirrer.
[9] S. K. Poole, C. F. Poole, Analyst 1995, 120, 289 and refer-
ences cited therein.
[10] S. Park, R. J. Kazlauskas, J. Org. Chem. 2001, 66, 8395–
8401.
The reaction at 10 bar H2 pressure were performed in a
50 mL autoclave equipped with a magnetic stirrer. Substrate
concentration was 0.25 M except in the hydrogenation with
an S/C 10,000 (1 M). Care was taken that the mixing in all cases
was turbulent. Conversion and enantiomeric excess were de-
termined by GC (CE instruments MFC 800, equipped with
Chirasil-Val capillary column).
[11] P. J. Dyson, G. Laurenczy, C. A. Ohlin, J. Vallance, T.
Welton, Chem. Commun. 2003, 2418–2419; IUPAC Sol-
ubility Data Series, (Ed.: C. L. Young), Pergamon Press,
Oxford, UK, 1981, Vols. 5–6.
[12] R. A. Brown, P. Pollet, E. McKoon, Ch. A. Eckert, Ch. L.
Liotta, P. G. Jessop, J. Am. Chem. Soc. 2001, 123, 1254–
1255.
[13] D. W. Kim, Ch. E. Song, D. Y. Chi, J. Am. Chem. Soc.
2002, 124, 10278–10279.
[14] P. J. Dyson, D. J. Ellis, T. Welton, Can. J. Chem. 2001, 79,
705–708.
[15] B. Wang, Y. Kang, Li Yang, J. Suo, J. Mol. Catal. A:
Chemical 2003, 203, 29–36.
Acknowledgements
We would like to thank Novartis Pharma AG for financial sup-
port.
[16] A. Mele, Ch. D. Tran, S. H. De Paoli Lacerda, Angew.
Chem. Int. Ed. 2003, 42, 4364–4366; K. E. Gutowski,
G. A. Broker, H. D. Willauer, J. G. Huddleston, R. P.
Swatloski, J. D. Holbrey, R. D. Rogers, J. Am. Chem.
Soc. 2003, 125, 6632–6633.
[17] P. Wasserscheid, T. Welton, Ionic Liquids in Synthesis,
Wiley-VCH, Weinheim, 2003, pp. 26–27; G. S. Fonseca,
A. P. Umpierre, P. F. P. Fichtner, S. R. Teixeira, J. Du-
pont, Chem. Eur. J. 2003, 9, 3263–3269.
[18] Determination of catalyst leaching: The only strongly
coloured ingredient in the reaction mixtures is the or-
ange-red catalyst. Based on our experience, the colour-
ing of the aqueous phases can reliably be used for a qual-
itative assessment of the degree of catalyst leaching. In
Tables 1–3, the sign “–” stands for no or very poor cat-
alyst separation (co-solvent phase and IL phase have
similar colours), the sign “þ” for good catalyst separa-
tion (slightly collared co-solvent phase) and the sign
“þ þ” for excellent separation (colourless co-solvent
phase). In one reaction (Table 2, entry 6), the Rh content
in a colourless aqueous phase was checked by ICP-MS
and found to be as low as 0.9 ppm. This corresponds to
a catalyst leaching of only 0.5%.
[19] Josiphos: A. Togni, C. Breutel, A. Schnyder, F. Spindler,
H. Landert, A. Tijani, J. Am. Chem. Soc. 1994, 116,
4062–4066; Walphos: T. Sturm, W. Weissensteiner, F.
Spindler, Adv. Synth. Catal. 2003, 345, 160–164; Mandy-
phos: J. J. A. Perea, M. Lotz, P. Knochel, Tetrahedron:
Asymmetry, 1999, 10, 375–384.
References and Notes
[1] P. Wasserscheid, T. Welton, Ionic Liquids in Synthesis,
Wiley-VCH, Weinheim, 2003, pp. 1–6; J. Dupont, R. F.
de Souza, Chem. Rev. 2002, 102, 3667–3691; Ch. E.
Song, Chem. Commun. 2004, 1033–1043; Ch. Baude-
quin, J. Baudoux, J. Levillain, D. Cahard, A. C. Gau-
mont, J. C. Plaquevent, Tetrahedron: Asymmetry 2003,
14, 3081–3093 and references cited therein.
[2] P. Wasserscheid, T. Welton, Ionic Liquids in Synthesis,
Wiley-VCH, Weinheim, 2003, pp. 174–283; P. Wassersc-
heid, W. Keim, Angew. Chem. Int. Ed. 2000, 39, 3773–
3789; Angew. Chem. 2000, 112, 3926–3945.
[3] H. Olivier-Bourbigou, L. Magna, J. Mol. Catal. A: Chem-
ical 2002, 182–183, 419–437; D. Zhao, M. Eu, Y. Kou, E.
Min, Catal. Today 2002, 74, 157–189; R. Sheldon, Chem.
Commun. 2001, 2399–2407 and references cited therein.
[4] A. L. Monteiro, F. K. Zinn, R. F. de Souza, J. Dupont,
Tetrahedron: Asymmetry 1997, 8, 177–179.
[5] S. Guernik, A. Wolfson, M. Herskowitz, N. Greenspoon,
S. Geresh, Chem. Commun. 2001, 2314–2315; S. Lee, Y.
Zhang, J. Piao, H. Yoon, Ch. Song, J. Choi, J. Hong,
Chem. Commun. 2003, 2624–2625; R. Giernoth, M. S.
Krumm, Adv. Synth. Catal. 2004, 346, 989–992; M. Ber-
thod, J. M. Joerger, G. Mignani, M. Vaultier, M. Lemaire,
Tetrahedron: Asymmetry 2004, 15, 2219–2221.
[6] P. Wasserscheid, T. Welton, Ionic Liquids in Synthesis,
Wiley-VCH, Weinheim, 2003, p. 4 2.
[7] L. Cammarata, S. G. Kazarian, P. A. Salter, T. Welton,
Phys. Chem. Chem. Phys. 2001, 3, 5192–5200.
[20] Y. Sun, R. N. Landau, J. Wang, Carl LeBlond, D. G.
Blackmond, J. Am. Chem. Soc. 1996, 118, 1348–1353.
[8] T. Ireland, K. Tappe, G. Grossheimann, P. Knochel,
Chem. Eur. J. 2002, 8, 843–852.
1486
ꢀ 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2004, 346, 1481–1486