Heterogeneous Enantioselective Hydrogenation of Activated Ketones Catalyzed by Modified Pt-Catalysts: A Systematic Structure-Selectivity Study

Article abstract of DOI:10.1002/adsc.200303104

A systematic structure-selectivity study was carried out for the enantioselective hydrogenation of activated ketones with chirally modified Pt/Al₂O₃ catalysts. For this, 18 modifiers containing an extended aromatic system able to form a strong adsorption complex with the Pt surface, and a suitable chiral group with an amino function capable to interact with the keto group of the substrate (HCd, Qd, HCn, Qn, and semi-synthetic derivatives, as well as synthetic analogues) were prepared and tested on 8 different activated ketones in AcOH and toluene under standard conditions. It was found that relatively small structural changes of the substrate and/or modifier structures strongly affected the enantioselectivity, and that no "best" modifier exists for all substrates. The highest ees for all substrates were obtained with quinuclidine-derived modifiers in combination with naphthalene or quinoline rings, either in AcOH (substrates 1-5 and 8, all carrying an sp³ carbon next to the keto group) or toluene (6 and 7, with an sp² carbon next to the ketone). The presence and nature of the substituent R′ at the quinuclidine significantly affected the ee (positive and negative effects). Certain combinations of an aromatic system and an amino function were preferred: For the quinuclidine moiety, quinoline and to a somewhat lesser extent naphthalene were a better match, while for the pyrrolidinylmethyl group anthracene was better suited. Methylation of the OH group often had a positive effect for hydrogenations in AcOH but not in toluene. With the exception of 8, higher ees were obtained for the Cd/ Qn series [leading to (R)-products] than for the Cn/ Qd series [leading to (S)-products]. In several cases, opposite structure-selectivity trends were detected when comparing reactions in toluene and AcOH, indicating a significant influence of the solvent.

Full text of DOI:10.1002/adsc.200303104

FULL PAPERS
Heterogeneous Enantioselective Hydrogenation of Activated
Ketones Catalyzed by Modified Pt-Catalysts: A Systematic
Structure-Selectivity Study
Christian Exner,a Andreas Pfaltz,*,a Martin Studer,b Hans-Ulrich Blaser*b
a
University of Basel, Department of Chemistry, St. Johanns-Ring 19, 4056 Basel, Switzerland
Fax: ( ‡ 41)-61-267-1103, e-mail: Andreas.Pfaltz@unibas.ch
b
Solvias AG, P. O. Box, 4002 Basel, Switzerland
Fax: ( ‡ 41)-61-686-6100, e-mail: Hans-Ulrich.blaser@solvias.com
Received: May 28, 2003; Accepted: August 18, 2003
Supporting information for this article is available on the WWW under http://asc.wiley-vch.de.
Abstract: A systematic structure-selectivity study was carbon next to the ketone). The presence and nature
carried out for the enantioselective hydrogenation of of the substituent R× at the quinuclidine significantly
activated ketones with chirally modified Pt/Al2O3 affected the ee (positive and negative effects). Certain
catalysts. For this, 18 modifiers containing an extend- combinations of an aromatic system and an amino
ed aromatic system able to form a strong adsorption function were preferred: For the quinuclidine moiety,
complex with the Pt surface, and a suitable chiral quinoline and to a somewhat lesser extent naphtha-
group with an amino function capable to interact with lene were a better match, while for the pyrrolidinyl-
the keto group of the substrate (HCd, Qd, HCn, Qn, methyl group anthracene was better suited. Methyl-
and semi-synthetic derivatives, as well as synthetic ation of the OH group often had a positive effect for
analogues) were prepared and tested on 8 different hydrogenations in AcOH but not in toluene. With the
activated ketones in AcOH and toluene under stand- exception of 8, higher ees were obtained for the Cd/
ard conditions. It was found that relatively small Qn series [leading to (R)-products] than for the Cn/
structural changes of the substrate and/or modifier Qd series [leading to (S)-products]. In several cases,
structures strongly affected the enantioselectivity, and opposite structure-selectivity trends were detected
that no ™best∫ modifier exists for all substrates. The when comparing reactions in toluene and AcOH,
highest ees for all substrates were obtained with indicating a significant influence of the solvent.
quinuclidine-derived modifiers in combination with
naphthalene or quinoline rings, either in AcOH Keywords: enantioselective hydrogenation of ke-
(substrates 1 5 and 8, all carrying an sp3 carbon tones; heterogeneous modified catalyst; modifier; Pt-
next to the keto group) or toluene (6 and 7, with an sp2 Cinchona; structure-ee relationship
Introduction
hydrogenation of the model substrate ethyl pyruvate
and it has been demonstrated that an effective modifier
Empirical structure-selectivity correlations are an im- must have two structural elements:[2] an (extended)
portant tool for understanding and developing enantio- aromatic system which is able to form a strong adsorp-
selective catalysts. This is even more so for heteroge- tion complex with the Pt surface, and a suitable chiral
neous systems because of a serious lack of understand- group with an amino function able to interact with the
ing of their mode of action. A significant shortcoming of keto group of the substrate. For different types of
most published studies in this area is the usually narrow activated ketones, the most selective modifiers up to
range of substrates and reaction conditions employed. now have been shown to be the naturally occurring
Due to an often high substrate specificity, this can cinchonidine Cd and its pseudo-enantiomer cinchonine
severely limit the value of such investigations.
Cn (with different relative configurations at C8 and C9)
The hydrogenation of activated ketones with a leading to the corresponding (R)- and (S)-alcohols,
chirally modified platinum catalyst is the most thor- respectively (Figure 1). Generally, the (S)-enantiomer is
oughly investigatedasymmetric heterogeneouscatalytic formed with a lower ee compared to the (R)-enantiom-
process.[1] The effect of the modifier structure on the er. The difference can be small to medium as shown for 1
enantioselectivity has been studied extensively for the and 2, but is very significant in other cases.
Adv. Synth. Catal. 2003, 345, 1253 1260
DOI: 10.1002/adsc.200303104
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1253

Christian Exner et al.
FULL PAPERS
To the best of our knowledge, no systematic study sions were 90 100% for the keto esters and the keto
comparing different modifiers for various activated acid. Because of this, a discussion of the effect of the
ketones has been carried out. In most cases, only ethyl or modifier structure on the reaction rates and acceleration
methyl pyruvate was tested, and if other substrates were factors is not possible. Lower conversions were usually
used, usually only one modifier was investigated. obtained with keto acetals, but even here they were
Furthermore, much more data are available for modi- always > 20%. Selected enantioselectivities are sum-
fiers of the cinchonidine series than for the pseudo- marized in Table 1. Since for practical reasons the
enantiomeric cinchonine modifiers. Here we report a standard reaction conditions chosen for our investiga-
systematic study of 18 modifiers with 8 substrates in 2 tion differ from the optimized ones described in refs.[3 9]
solvents carried out under standard conditions. Evalua- (Figure 1), our ees usually did not reach the best
tion of the results has led to empirical structure- literature values. We are aware that changes in the
selectivity relationships for various ketone/modifier reaction conditions (p, T and especially modifier con-
combinations which show some general trends not centration) can affect different substrate-modifier com-
documented so far.
binations differently. However, the fact that we ob-
served the same trends as described in refs.[3 9] gives us
confidence that the main effects are really due to an
inherent structure-selectivity correlation and are not
much distorted by unsuitable reaction conditions.
For reasons of clarity, only the series for the solvent
Results and Discussion
Our study addressed three questions: i) What are the
effects of varying the aromatic moiety (presence of N, giving higher enantioselectivity is depicted in Table 1 (a
R'', size) and the amino function (R' at C-3, bicycle vs. complete list of ees is given in the supporting informa-
monocycle)? ii) Is it possible tofind a ™best modifier∫ for tion) with the exception of 5. Here, the results for both
a series of activated ketones for producing either the solvents are given because the highest ees are obtained
(R)- or the (S)-alcohol in high ees? iii) What is the effect in toluene but the averages are higher in AcOH. In the
of the substrate structure, especially the different func- upper part, results with modifiers based on the pseudo-
tional groups adjacent to the keto group?
enantiomeric Cd and Cn, and in the lower part with
For this purpose we investigated 18 different modi- synthetic modifiers [only (S)-series at C-OH] are shown.
fiers (HCd, Qd, HCn, Qn, semi-synthetic derivatives, as Besides these, a variety of other modifiers was tested
well as synthetic analogues, many of them new, see like the corresponding threo-QQ, QN and QA, MeO-
Figures 1 and 2) for the hydrogenation of eight ketones isoCn, 2-quinolinyl and 1-anthracenyl derivatives. In all
(Figure 1). Six aliphatic ketones 1 5 and 8 carrying an cases, the ees were much lower than those described
sp3 and two aromatic ketones 6 and 7 carrying an sp2 here and these results will not be discussed further.
ˆ
carbon next to the C O function were chosen. As
First conclusions can be drawn by cursory inspection
activating groups, ester, acid and acetal functions were of Table 1: i) The highest enantioselectivities for differ-
investigated. Acetic acid and toluene were used as ent substrates were obtained with different modifiers,
solvents. Under our standard conditions, the conver- there is definitively no ™best modifier∫ for all substrates.
Table 1. Selected ee values for various combinations of substrates, modifiers and solvents [upper part left column % ee for
(S)-, right column % ee for (R)-product; lower part % ee for (S)-product; bold: highest ee for a particular substrate].
Substrate/solvent
Modifier
1
2
3
4
5
5
6
7
8
Average
AcOH
AcOH
AcOH
AcOH
AcOH
toluene
toluene
toluene
AcOH
HCn/HCd
84/88
88/88
85/91
86/93
84/88
90/94
82
88
79
70
73
78/81
81/80
81/88
66/85
76/78
82/90
78
86
63
57
58
78/91
82/93
93/97
66/83
81/72
92/98
69
84
80
8
54
64/70
67/84
59/69
49/62
60/58
73/69
61
74
56
30
51
54/69
62/65
62/72
34/60
39/49
61/76
58
67
54
29
42
58/78
71/80
18/54
23/23
40/56
4/23
66
35
32
31
42
78/89
80/91
31/55
52/87
66/84
8/45
90
70
60
52
74
47/80
55/83
0/0
9/34
9/32
0/0
72
19
3
43
60/69
64/34
75/44
46/36
51/33
77/64
59
70
31
34
29
68/81
74/79
55/62
50/63
58/63
53/60
72
66
51
41
52
IsoCn/isoCd
MeOCn/MeOHCd
Qd/Qn
IsoQd/isoQn
MeOQd/MeOQn
(1S,2R)-(‡)-QQ
(1S,2R)-(‡)-QN
(1S,2R)-(‡)-QA
(S)-(‡)-PQ
(S)-(‡)-PN
32
n.a.
26/36
(S)-(‡)-PA
76
82/84
71
73/76
70
71/75
61
59/62
51
51/58
69
41/49
84
62/73
29
52/44
66
58/63
Average
Reaction conditions: 1.0 mg modifier, 10 mg 5% Pt/Al2O3 (JM 94), 1.0 mL solvent, 0.1 mL substrate, 50 bar H2, 258C, 1 h.
Reactions were carried out in glass vials (4 parallel experiments in a 50-mL autoclave).
1254
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2003, 345, 1253 1260

Heterogeneous Enantioselective Hydrogenation of Activated Ketones
FULL PAPERS
Figure 1. Structure and numbering of substrates, structure of cinchona modifiers, and highest ees reported in the literature.
a)
Only results for the corresponding ethyl ester reported.
Figure 2. Structural elements (unless indicated otherwise, R and R'' ˆ H, R' ˆ vinyl) and abbreviation of modifiers (Cd
cinchonidine, Cn cinchonine, Qn quinine, Qd quinidine).
ii) It was confirmed that the Cn/Qd derivatives as well as modifiers in which only one element was changed and
the corresponding synthetic analogues with (S)-config- plotted the ee differences (ee obtained with the upper
uration at C-OH always give an excess of the (S)- modifier minus ee obtained with the lower modifier) for
enantiomer, whereas Cd/Qn derivatives always give all substrates in both toluene and acetic acid as depicted
predominantly the (R)-alcohol. iii) For ester and acetal in Figures 3 7. We will discuss one analysis in more
substrates, Cinchona-derived modifiers were generally detail and summarize the others in analogy.
superior for the (R)-alcohols, whereas the highest ees for
In order to analyze the effect of R' at the quinuclidine
(S)-products were often observed with the synthetic system we compared HCn with (‡)-QQ, isoCn with (‡)-
modifiers. iv) The preferred solvent for the alkyl ketones QQ, and HCn with isoCn for the (S)-series (Figure 3a),
1 5 and 8 was found to be AcOH, whereas aryl ketones and HCd with (À)-QQ, isoCd with (À)-QQ, and HCd
6 and 7 gave higher ees in toluene. v) In general, HCd with isoCd for the (R)-series (Figure 3b). The left parts
and isoCd were better modifiers for the R-series, while of Figure 3a and Figure 3b allow a comparison of R' ˆ
isoCn and QQ were better for the S-series. vi) With the Et and H, respectively. For the (S)-alcohols in toluene,
exception of 8, higher ees were obtained for the Cd/Qn the derivative with R' ˆ Et gave significantly lower ees
series [(R)-products] than for the Cn/Qd series. vi) The than the modifier with R' ˆ H, whereas no uniform
highest ees were obtained with the aliphatic ketones 1 3. trend was observed in AcOH; for the (R)-products the
Et substituent was superior in both solvents. The two
middle parts allow a comparison of R' ˆ vinylidene and
Specific Modifier Structure-Selectivity Effects
H, respectively. R' ˆ vinylidene instead of H gave mixed
resultsfor the (S)-products in toluene but, except for one
In order to analyze the effects of the various structural substrate had a positive effect in AcOH. The two right
elements in more detail, we compared different pairs of parts finally allow a comparison of R' ˆ ethyl and
Adv. Synth. Catal. 2003, 345, 1253 1260
asc.wiley-vch.de
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1255

Christian Exner et al.
FULL PAPERS
Figure 3. Effect of the nature of R© on enantioselectivity. The first five bars (white) always represent 1 5 (aliphatic substrates),
the next two bars (gray) represent 6 and 7 (aromatic substrates), the black bar represents 8 (free acid). Y axis:Dee (%) of the
two modifiers depicted. Figure 3a (left): (S)-series. Figure 3b (right): (R)-series.
Figure 4. Effect of varying R'' on enantioselectivity.
vinylidene, respectively. R' ˆ Et instead of vinylidene picture emerged for HCd and isoCd (Figure 4, right). In
had (with the notable exception of 8 in toluene) a a few cases a weak inverse effect was found, especially
negative effect for the (S)- and mixed effects for the (R)- for 1 and 2 in AcOH. Independent experiments showed,
products. From these comparisons we conclude that the however, that this was due to an artifact. Because of the
modifiers with R' ˆ Et or vinyl are superior for produc- high rates obtained with, e.g., the HCd/ethyl pyruvate/
ing (R)-alcohols, while for (S)-alcohols the ees usually AcOH combination, this reaction was diffusion limited
increase in the order vinylidene > H > Et. This means in our experimental set-up, whereas the slower combi-
that the presence of an R© substituent has a positive nation Qn/ethyl pyruvate/AcOH was not. Since it is well
effect on the ee in the Cd/Qn series, while it has a known that lowering the hydrogen concentration (e.g.,
negative effect in the Cn/Qd series. The same results by diffusion limitation) decreases the enantioselectiv-
were seen with the analogous Qn and Qd derivatives (no ity,[10,11] this explains the lower ee observed for HCd than
details shown, averages see Table 1).
for Qn. Control experiments with better mass transfer
The effect of R©© at the quinoline moiety (Figure 4) can indeed gave higher ees for HCd.
be summarized as follows: the unsubstituted Cn and
The effect of O-methylation (R ˆ Me) can be seen in
isoCn derivatives (Figure 4, left) generally gave a much Figure 5. In toluene, the non-methylated derivatives
better performance than their counterparts carrying a usually gave a significantly higher ee for the (R)- as well
methoxy group on the quinoline system. A similar as for the (S)-products. In AcOH the methoxy deriva-
1256
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2003, 345, 1253 1260

Heterogeneous Enantioselective Hydrogenation of Activated Ketones
FULL PAPERS
Figure 5. Effect of varying R on enantioselectivity.
tives were generally better for the aliphatic ketones 1
5, whereas for the other substrates both negative and
positive effects were observed, strongly depending on
the nature of R'' at the quinoline system.
As shown in Table 1, the highest ees for each substrate
were obtained with quinuclidine-derived modifiers in
combination with naphthalene or quinoline rings. Fig-
ure 6 shows the performance of different amino func-
tion/aryl group combinations. Here, modifiers with an
N-pyrrolidinylmethyl group are compared with their
quinuclidine analogues. The quinuclidine/quinoline
combination gave higher ees in both solvents with only
one exception. With naphthalene-derived modifiers,
quinuclidine was superior in toluene while in AcOH the
inverse effect was observed. Surprisingly, the quinucli-
dine ring in combination with anthracene was generally
inferior in both solvents.
Figure 6. Effect of varying the amino function with different
aromatic rings on enantioselectivity [(S)-series].
Finally, the effect of replacing quinoline by anthracene
or naphthalene is depicted in Figure 7. With quinucli-
Figure 7. Effect of the aromatic moiety on enantioselectivity. Left part modifiers with quinuclidine, right part with N-
pyrrolidinemethyl [(S)-series].
Adv. Synth. Catal. 2003, 345, 1253 1260
asc.wiley-vch.de
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1257

Christian Exner et al.
FULL PAPERS
dine as chiral amino function, quinoline (toluene) or
naphthalene (AcOH) were usually the best choice,
while the ees with anthracene were generally lower.
With the N-pyrrolidinylmethyl moiety, however, either
anthracene or, somewhat less pronounced, naphthalene
showed a better over-all performance than quinoline.
From these data, the following conclusions can be
drawn
Figure 8. Activated complex between cinchona type modi-
fiers and trans a-keto ester adsorbed on the Pt surface.[1,2,3c]
(a) Cd in the open(3) conformation, (b) Cd in the open(4)
conformation, (c) QA, open(3) conformation.
*
It was confirmed that the presence of an extended
aromatic system with a chiral group carrying an
amino function are necessary but not sufficient
prerequisites for high ees.
*
For every substrate the highest ees were obtained
depicted for Cd in Figure 8b), which might lead to lower
ees, is not possible in this case.[2a] In addition, the
adsorption of the modifier on the Pt surface might also
be stronger. It was therefore disappointing that anthra-
cene gave much lower ees when combined with quinu-
clidine than either quinoline or naphthalene. The reason
for this is most likely that the very rigid quinuclidine
moiety cannot adopt an optimal conformation when
attached to the anthracene group, maybe because of
unfavorable interactions with the hydrogen atom in the
1' position (see Figure 8c). With the much smaller and
more flexible pyrrolidinylmethyl group such interac-
tions are less likely, and the beneficial effect of the large
aromatic anthracene dominates.
with quinuclidine-derived modifiers in combination
with naphthalene or quinoline rings.
The substituent R× at the quinuclidine system has a
more important influence than previously thought
and can significantly affect the ee compared to the
unsubstituted derivatives (positive and negative
effects!).
*
*
Both bicyclic and tricyclic aromatic systems rings can
lead to high ees. For the sterically more demanding
and more rigid quinuclidine, quinoline and to some
lesser extent naphthalene were a better match, while
for anthracene the smaller pyrrolidinylmethyl group
was superior. The combination of the previously
identified ™best structural elements∫ anthracene and
quinuclidine does not give the highest ees.
*
HCd and HCn derivatives usually gave higher ees
than the corresponding Qn and Qd which carry a
Specific Modifier Structure-Selectivity Effects
methoxy substituent at the quinoline system.
Methylation of the OH group often had a positive
effect for hydrogenations in AcOH but not in
toluene.
a) Substrates carrying an sp3 carbon next to the
activated ketone (aliphatic ketones)
*
For the monoketones 1 3 the highest enantioselectiv-
While this study was not designed to answer mechanistic ities were always observed in AcOH with ees well above
questions, we would nevertheless like to briefly address 80%: Relatively high ees could be achieved with most
the structure of the activated complex formed between modifiers. The ees were somewhat lower for the
the functionalized ketone and the modifier. There is diketones 4 and 5 and they seemed to be more sensitive
sufficient evidence that the productive activated com- to the combination of modifier and solvent. For 5, the
plex is created by the adsorbed modifier forming an H average ee was higher in AcOH but the highest ees were
bond between the amino moiety and the carbonyl obtained in toluene. In all cases, the best modifiers were
oxygen of the adsorbed ketone.[1,2,3c] For the combina- either derived from the natural alkaloids or closely
tion of ethyl pyruvate and cinchonidine (and also some related synthetic analogues like QN. While the differ-
simpler synthetic modifiers) evidence from calcula- ences between the two pseudo-enantiomers were usu-
tions[3c] and synthetic variations[12 14] have led to the ally not very large, O-methylation generally had a small
proposal that for an optimal complex, the modifier has beneficial effect in AcOH.
to be able to adopt the open(3) conformation as
depicted in Figure 8a. Our results are basically compat-
ible with this proposal, even though it is not possible to b) Substrates carrying an sp2 carbon next to the
interpret the sometimes very strong influence of R, R©, activated ketone (aromatic ketones)
R'', the substrate structure and the solvent with this
simple model.
The trends observedfor 6 and 7were different compared
Earlier it was found that synthetic modifiers with the to 1 5. Here, the highest ees were observed in toluene,
larger anthracene gave higher ees in combination with a usually with the iso-derivatives or the closely related
pyrrolidinylmethyl[2a] or piperidine-derived moiety.[14] It QQ. In contrast to 1 5, the ees showed enormous
was suggested that the open(4) conformation (as fluctuations when applying different modifiers. While
1258
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2003, 345, 1253 1260

Heterogeneous Enantioselective Hydrogenation of Activated Ketones
FULL PAPERS
the HCn/HCd and the isoCN/isoCd series gave high ees, Experimental Section
derivatives carrying a methoxy group at the quinoline
ring or at the stereogenic center led to much lower ees
(sometimes zero!). The differences between the pseudo-
Synthesis of the Modifiers
HCn and HCd were synthesized according to ref.[2b] and PQ,
PN and PA according to ref.[2a] All synthetic derivatives
carrying a quinuclidine group were synthesized according to
Scheme 1 (upper part), all methylated derivatives according to
Scheme 1 (lower part, right side) and all isomerized derivatives
according to Scheme 1 (lower part, left side). For a detailed
description of the synthesis and the characterization of the
modifiers, see supporting information.
enantiomers were also larger than for the aliphatic
substrates.
c) Free ketoacid 8
In general, 8 showed a similar behavior as 1 5, with
relatively small ee variations for different modifiers.
Interestingly, the (S)-series gave higher ees here in
contrast to all other substrates. However, it has to be
noted that AcOH and toluene might not be the best
solvent for 8. Previous studies suggested an isopropanol/
water mixture to be the optimum for keto acids.[9]
Typical Hydrogenation Experiment
1.00 mg modifier, 10 mg catalyst [5% Pt/Al2O3 (JMC 94),
pretreated for 2 h at 4008C under hydrogen], 1.00 mL solvent
Scheme 1. Synthesis of the modifiers.
and 100 mL substrate (2 6), 100 mg substrate (7, 8) or 250 mL
substrate (1) were placed into a 3-mL glass vial equipped with a
magnetic stirring bar. Four such vials (three tests and one
reference experiment) were placed into a 50-mL autoclave
thermostatted at 258C. The autoclave was closed, purged with
Conclusion
Our systematic structure-selectivity study revealed the
very high substrate specificity of this catalytic system
and should be taken as a warning to propose a general argon (3 times) and hydrogen (3 times) and then pressurized to
50 bar hydrogen pressure. The reactions were started by
turning the magnetic stirrer on (1100 rpm). After 1 h, the
pressure was released, and the autoclave was purged with
argon (3 times). The vials were discharged, and the reaction
mixtures were filtered through disk filters and analyzed. For
analytical methods, see supporting material.
reaction mechanism based on just a few isolated
observations. Obviously, relatively small structural
changes in substrate and/or modifier can strongly affect
the enantioselectivity but to a varying degree and often
in opposite manner for different substrates or when
comparing reactions in toluene and AcOH. This indi-
cates that the interactions between the substrate and the
modifier depend on many factors. Nevertheless, we did
find some common trends and our data serve as a good
basis to select the optimal modifier for a particular
transformation. Moreover, some of the new modifier-
substrate combinations gave significantly higher ees
than previously reported, especially for the (S)-prod-
ucts. Our results should also be a useful basis for further
optimization of the modifier structure as well as the
development of modifiers for other substrates.
Acknowledgements
The authors thank S. Burkhardt for technical support. Financial
support by the Federal Commission for Technology and
Innovation (KTI Project No. 5189.2 KTS) is gratefully acknowl-
edged.
References
[1] For a recent overview, see M. Studer, H. U. Blaser, C.
Exner, Adv. Synth. Catal. 2003, 345, 45 and references
cited therein.
Adv. Synth. Catal. 2003, 345, 1253 1260
asc.wiley-vch.de
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1259

Christian Exner et al.
FULL PAPERS
[2] a) A. Pfaltz, T. Heinz, Topics Catal. 1997, 4, 229; b) H. U.
Blaser, H. P. Jalett, W. Lottenbach, M. Studer, J. Am.
Chem. Soc. 2000, 122, 12675 and references cited therein;
c) A. Baiker J. Mol. Catal. A: Chemical. 2000, 163, 205.
[9] H.-U. Blaser, H.-P. Jalett, Stud. Surf. Sci. Catal. 1993, 78,
139.
[10] H.-U. Blaser, H.-P. Jalett, M. Garland, M. Studer, H.
Thies, A. Wirth-Tijani, J. Catal. 1998, 173, 282.
[11] U. K. Singh, R. N. Landau, Y. Sun, C. LeBlond, D. G.
Blackmond, S. K. Tanielyan, R. L. Augustine, J. Catal.
1995, 154, 91.
¬
¬
¬
[3] a) M. Bartok, B. Tˆrˆk, K. Balazsik, T. Bartok, Catal.
Lett. 2001, 73, 127; b) W.-R. Huck, T. Mallat; A. Baiker,
Adv. Synth. Catal. 2003, 345, 255.
¬
¬
¬
[12] a) M. Bartok, B. Tˆrˆk, K. Balazsik, T. Bartok, Catal.
¬
¬
[4] B. Tˆrˆk, K. Balazsik, M. Tˆrˆk, G. Szˆllˆsi, M. Bartok,
¬
Lett. 2001, 73, 127; b) M. Bartok, K. Felfˆldi, B. Tˆrˆk, T.
Ultrasonics Sonochemistry 2000, 7, 151.
¬
¬
Bartok, Chem. Commun. 1998, 2605; c) M. Bartok, K.
¬
¬
[5] B. Tˆrˆk, K. Felfˆldi, K. Balazsik, M. Bartok, Chem.
¬
Felfˆldi, G. Szˆllˆsi, T. Bartok, Catal. Lett. 1999, 61, 1;
Commun. 1999, 1725.
[6] M. Studer, S. Burkhardt, A. F. Indolese, H.-U. Blaser,
Chem. Commun. 2000 1327.
¬
¬
d) G. Szˆllˆsi, K. Felfˆldi, T. Bartok, M. Bartok, React.
¬
Kin. Catal. Lett. 2000, 71, 99; e) M. Bartok, M. Sutyinsz-
ki, K. Felfˆldi, G. Szˆllˆsi, Chem. Commun. 2002, 1130.
[13] T. B¸rgi, Z. Zhou, N. K¸nzle, T. Mallat, A. Baiker, J.
Catal. 1999, 183, 405.
¬
[7] M. Sutyinszki, K. Szˆri, K. Felfˆldi, M. Bartok, Catal.
Lett. 2002, 81, 281.
[8] M. Studer, S. Burkhardt, H.-U. Blaser, Chem. Commun.
1999, 1727.
[14] A. Solladie-Cavallo, C. Marsol, F. Garin, Tetrahedron
Letters 2002, 43, 4733.
1260
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2003, 345, 1253 1260