Cinchona methyl ethers as modifiers in the enantioselective hydrogenation of (E)-2,3-diphenylpropenoic acids over Pd catalyst

Article abstract of DOI:10.1016/j.jcat.2010.09.020

The enantioselective hydrogenation of (E)-2,3-diphenylpropenoic acids over Pd/Al₂O₃ modified by (R)C⁸(S)C⁹ cinchona ethers resulted in the inversion of the sense of the enantioselectivity. To find the explanation of the phenomenon, the interaction of acids bearing different substituents with cinchona alkaloids was studied in solution by NMR spectroscopy and experiments using mixtures of modifiers were carried out. The prominent non-linear behaviour obtained revealed the altered adsorption of the cinchona methyl ethers when compared with the parent alkaloids. The investigations indicated that the interaction of the ether derivatives with the unsaturated acids is more flexible and the presence of the methyl group reshapes the chiral surface sites. The combination of these effects complemented by the bulkiness of the diaryl substituted acrylic acids may lead to the inversion of the docking preference of the substrates in the altered chiral pocket of the adsorbed modifier and consequently results in decrease in the enantioselectivity or even in the inversion of its sense. Novel evidence on the ligand-accelerated mechanism in the enantioselective hydrogenation of (E)-2,3-diphenylpropenoic acids over cinchona alkaloid-modified Pd in the presence of benzylamine was also presented.

Full text of DOI:10.1016/j.jcat.2010.09.020

Journal of Catalysis 276 (2010) 259–267
Contents lists available at ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
Cinchona methyl ethers as modifiers in the enantioselective hydrogenation
of (E)-2,3-diphenylpropenoic acids over Pd catalyst
a,
b
György Szollosi , Beáta Hermán , Ferenc Fülöp a,b, Mihály Bartók a,c
⇑
} }
a Stereochemistry Research Group of the Hungarian Academy of Sciences, University of Szeged, Dóm tér 8, Szeged H-6720, Hungary
b Institute of Pharmaceutical Chemistry, University of Szeged, Eötvös utca 6, Szeged H-6720, Hungary
c Department of Organic Chemistry, University of Szeged, Dóm tér 8, Szeged H-6720, Hungary
a r t i c l e i n f o
a b s t r a c t
Article history:
The enantioselective hydrogenation of (E)-2,3-diphenylpropenoic acids over Pd/Al2O3 modified by
(R)C8A(S)C9 cinchona ethers resulted in the inversion of the sense of the enantioselectivity. To find the
explanation of the phenomenon, the interaction of acids bearing different substituents with cinchona
alkaloids was studied in solution by NMR spectroscopy and experiments using mixtures of modifiers
were carried out. The prominent non-linear behaviour obtained revealed the altered adsorption of the
cinchona methyl ethers when compared with the parent alkaloids. The investigations indicated that
the interaction of the ether derivatives with the unsaturated acids is more flexible and the presence of
the methyl group reshapes the chiral surface sites. The combination of these effects complemented by
the bulkiness of the diaryl substituted acrylic acids may lead to the inversion of the docking preference
of the substrates in the altered chiral pocket of the adsorbed modifier and consequently results in
decrease in the enantioselectivity or even in the inversion of its sense. Novel evidence on the ligand-
accelerated mechanism in the enantioselective hydrogenation of (E)-2,3-diphenylpropenoic acids over
cinchona alkaloid-modified Pd in the presence of benzylamine was also presented.
Received 14 July 2010
Revised 30 August 2010
Accepted 18 September 2010
Keywords:
Hydrogenation
Cinchona alkaloids
(E)-2,3-Diphenylpropenoic acids
Palladium
Enantioselective
Inversion
Non-linear effect
Cinchona methyl ethers
Ó 2010 Elsevier Inc. All rights reserved.
1. Introduction
excess (ee) can be obtained in the hydrogenation of a,b-dialkyl
substituted acids [8,19], while in the reaction of b-aryl substituted
The asymmetric catalytic hydrogenation of the prochiral
unsaturated acids is a widely used method for the preparation of
optically pure chiral carboxylic acids used as intermediates in the
fine chemical industry [1,2]. Extended efforts were devoted to
develop heterogeneous catalytic systems for replacing the soluble
metal complex catalysts used for these purposes [3]. The adsorption
of chiral compounds so-called modifiers on the surface of metals
provided efficient heterogeneous asymmetric catalytic systems
such as the tartaric acid-modified Raney-Ni or the supported Pt
and Pd catalysts modified by cinchona alkaloids [4–6]. Pd catalysts
modified by cinchonidine (CD) were found efficient in the enantio-
selective hydrogenation of prochiral unsaturated carboxylic acids.
The early report on the enantioselective hydrogenation of
acids good optical purities could be reached, especially with aryl
group in the
a position, too [13,18,20]. Parallel with these investi-
gations, studies aimed the elucidation of the nature of the modi-
fier–unsaturated acid interactions to reveal the structures of the
surface intermediate complex responsible for enantiodifferentia-
tion were also conducted. However, in situ methodologies neces-
sary to reveal the real interactions established on the surface in
this complex three-phase system are not yet available. Thus, valu-
able information was gained from the effects of the alterations in
the structure of the modifier [21,22] or the substrate [11,13,20],
the non-linear behaviour obtained by using modifier mixtures
[21,23] combined with the results of spectroscopic and computa-
tional methods [22,24,25].
a
,b-unsaturated carboxylic acids over CD-modified Pd [7] was fol-
These investigations resulted in several similarities and differ-
ences in the interpretation of the enantiodifferentiation occurring
in the hydrogenations of acids bearing aliphatic or aromatic
substituents. The protonation of the quinuclidine moiety of the
modifier by the acids was demonstrated with both acid types
by the loss of the enantioselectivity in the hydrogenation of
the methyl esters of the acids or over N-methyl cinchonidinium
salt-modified catalyst [21,22,24] and even by the effect of the
acidity of the substrate tuned by substituting the b-phenyl group
lowed by efforts to increase the optical purity of the products
either by optimization of the catalytic system or by extension of
the scope on structurally different acids [8–20]. The studies of
several research groups converged to similar conclusions concern-
ing the effect of the substrate structure, i.e. moderate enantiomeric
⇑
Corresponding author. Fax: +36 62 544200.
}
}
E-mail address: szollosi@chem.u-szeged.hu (G. Szollosi).
0021-9517/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved.
doi:10.1016/j.jcat.2010.09.020

} }
G. Szollosi et al. / Journal of Catalysis 276 (2010) 259–267
260
[13,17,18,20]. The involvement of the C9AOH group of the
cinchona alkaloid in the interaction with the aliphatic acids
was demonstrated by using cinchonidine methyl ether (CDM)
and FT-IR, ATR-IR, NMR spectroscopy combined with theoretical
calculations [22,24,25]. These results also indicated that the most
stable complexes are formed by the participation of two or three
aliphatic acid molecules interacting with CD [25]. In contrast, the
hydrogenation of (E)-2,3-diphenylpropenoic acid (1) was assumed
to occur through a 1:1 CD–acid complex, and the interaction of
C9AOH group with the carboxylic group was assumed based on
the result obtained by using dihydro-CDM [21]. While the differ-
ences in the stoichiometry of the complexes of the two acid types
may be rationalized by the solvent effect, the interpretation did
not explain the formation in the presence of dihydro-CDM of
the opposite enantiomer in small excess when compared with
CD.
A valuable method for obtaining in situ information on the sur-
face reaction was found to be the application of modifier mixtures
or sequentially added modifiers. The non-linear behaviour ob-
tained in the hydrogenation of activated ketones over Pt [26–31]
or of 2-pyrone derivatives over Pd [32,33] indicated differences
in the adsorption strength and geometry of the modifiers during
the reactions. The striking non-linear behaviour observed in the
enantioselective hydrogenation of 1 using mixtures of CD with
dihydro-CDM was interpreted as a confirmation of the H-bonding
between the alkaloid C9AOH and the acid [21]. Very recently, mix-
tures of CD and CN were also applied in the reaction of the same
acid; the lack of deviation from the linearity demonstrated that
the lower ee obtained with CN when compared with CD is not
due to the different adsorption properties of the two alkaloids [23].
In our present study, we sought to obtain further details on the
interaction of (E)-2,3-diphenylpropenoic acid derivatives with the
modifier by investigating the hydrogenations of six (E)-2,3-diphe-
nylpropenoic acids (unsubstituted, OCH3 and/or F substituted) over
Pd catalyst modified by the four parent cinchona alkaloids, i.e. CD,
cinchonine (CN), quinine (QN) and quinidine (QD), their C9AOCH3
ethers (CDM, CNM, QNM, QDM) and b-isocinchonine (b-ICN). The
interpretation of the results was aided by the NMR spectroscopic
investigation of the interaction of cinchona alkaloids with the acids
in solution and by experiments using mixtures of the chiral
modifiers.
(1000 rpm). The H2 uptake was followed by a gas burette, and the
consumption between 15% and 25% of the total uptake was used
for calculating the initial rates (Ri, mmol hꢀ1 gꢀ1). In a typical
run, 0.025 g catalyst and 3 cm3 DMF containing 2.5 vol.% dist.
H2O were introduced into the reactor, the apparatus was flushed
with H2 and the catalyst was pretreated for 0.5 h by stirring the
slurry. After pretreatment, 0.025 mmol modifier ([modi-
fier] = 5 mM), 0.5 mmol acid ([acid] = 0.1 M), 0.5 mmol BA (when
used; [BA] = 0.1 M) and another 2 cm3 solvent were added, the sys-
tem was flushed with H2 and the reaction started by stirring the
mixture. After the specified time (6–8 h), 5 cm3 CH3OH was added,
and the catalyst was filtered and washed with another 5 cm3
CH3OH. The resulting compounds were identified by GC–MS anal-
ysis (Agilent Techn. 6890N GC – 5973 MSD, 60 m HP-1MS capillary
column) and by 1H and 13C NMR spectroscopy (Bruker Avance DRX
400 spectrometer in d6-DMSO, see Supporting information).
Portions of these solutions were used for transforming the acids
in methyl esters by reaction with CH3OH using conc. H2SO4 and/or
with CH2N2 ethereal solution. Conversions (X%) and enantioselec-
tivities expressed as ee (%) were calculated from the gas chromato-
graphic analysis of these samples (YL6100 GC equipped with FID
and Cyclosil-B, 30 m ꢁ 0.25 mm, J & W Sci. Inc., chiral capillary col-
umn) by the formulae: X (%) = 100 ꢁ ([S] + [R])/[acid]i; ee (%) = 100
ꢁ |[S] ꢀ [R]|/([S] + [R]); where [acid]i is the initial concentration of
the unsaturated acid; [S] and [R] are the concentrations of the
product enantiomers. The experiments were repeated giving re-
sults reproducible within 1%. The absolute configurations of the
excess enantiomers obtained in presence of CD were assigned in
previous studies to be S [7,13,17,18] based on the rotation sign of
the saturated products. Optical rotation measurements (Polamat
A polarimeter, l 0.5 dm, c 1, methanol) showed that the use of CD
results in the excess formation of the dextrorotatory (S) enantio-
mers, while with CN the levorotatory (R) enantiomers resulted in
excess. The configuration of the excess enantiomers formed by
using the other cinchona derivatives were assigned by chiral GC
measurements using as reference the products obtained with CD
and CN.
The hydrogenations in the presence of modifier mixtures and
the product analysis were carried out similarly as with a single
modifier except the corresponding mixture of modifier was added
to the pretreated catalyst. The theoretical ee values (eecalc) corre-
sponding to the linear behaviour for a modifier mixture were cal-
culated with the formulae: eecalc (%) = (x1 ꢁ Ri1 ꢁ ee1 + x2 ꢁ
Ri2 ꢁ ee2)/ (x1 ꢁ Ri1 + x2 ꢁ Ri2); where x1 and x2 are the molar frac-
tions of modifiers 1 and 2; Ri1 and Ri2 (mmol hꢀ1 gꢀ1) are the initial
rates and ee1 and ee2 (%) are the enantiomeric excesses obtained
with the modifiers 1 and 2, respectively. The theoretical initial
rates (Ricalc) were calculated with the formulae: Rci alc (mmol hꢀ1
gꢀ1) = x1 ꢁ Ri1 + x2 ꢁ Ri2.
2. Experimental
2.1. Materials
The catalyst used in this study was commercial 5% Pd/Al2O3
(Engelhard, 40692), which was pretreated in H2 flow at 523 K as
described before [17,18]. The parent cinchona alkaloids were com-
mercial products (CD, CN, QN, QD all Fluka, P98%). The cinchona
methyl ethers and b-ICN were prepared by previously described
methods [34,35]. Benzylamine (BA, Fluka, P99.5%), N,N-dimethyl-
formamide (DMF, Scharlau, Multisolvent grade) and H2 gas (Linde
AG, 99.999%) were used as received. (E)-2,3-Diphenylpropenoic
acid (1, Aldrich, P97%) was purified by crystallization in
acetone–water. The preparation by the Perkin condensation,
purification and characterization of the OCH3 and/or F substituted
(E)-2,3-diphenylpropenoic acid derivatives has been described pre-
viously [17,18].
2.3. 1H and 13C NMR investigations
The interaction of the substrates with the modifiers and BA in
liquid phase was studied by 1H and 13C NMR spectroscopy. Spectra
were recorded on a Bruker Avance DRX 500 NMR instrument oper-
ated at 500 MHz (1H) or 125 MHz (13C) using the solvent signals as
reference. The 0.025 mmol cinchona alkaloid was dissolved in
0.5 mL d7-DMF + 2.5 vol.% D2O solvent mixture, and their 1H and
13C NMR spectra were recorded. To these samples, 0.1 mmol acid
was added and the spectra were recorded followed by addition
of 0.1 mmol BA and recording again the spectra. Changes in the
2.2. Hydrogenation procedure and product analysis
chemical shift (Dd, ppm) of the H or C atom signals were calculated
by subtracting the chemical shift (d, ppm) of the H or C atom sig-
nals in the mixtures from the d of the corresponding atom signals
The hydrogenations were carried out in batch reactors under
atmospheric H2 pressure and room temperature (unless otherwise
noted) in a glass hydrogenation apparatus using magnetic agitation
in the spectra of the pure compounds using the formulae:
D
d(Hx or
Cx) = d(Hx or Cx)Cinchona alkaloid ꢀ d(Hx or Cx)Cinchona alkaloid + acid
.

}
}
G. Szollosi et al. / Journal of Catalysis 276 (2010) 259–267
261
previous studies using CD as modifier [17,18], and for the sake of
comparison, all the reactions were carried out under identical
conditions, although these conditions may not be the optimal for
all the examined cinchona alkaloid–acid pairs, similarly with the
hydrogenation of activated ketones [34].
Using CN instead of CD resulted in the formation of the R satu-
rated acids in excess as expected based on the opposite configura-
tion of the C8 and C9 chiral centres of CN when compared to CD and
having in mind the accepted assumption that these chiral atoms
form the stereogenic centre responsible for enantiodifferentiation
[5,16,21]. The ees obtained on CN-modified catalyst were always
lower when compared with CD-modified Pd, similarly with the re-
cent report on the hydrogenation of 1 [23] or other unsaturated
acids [16], though in the hydrogenation of 2-pyrone derivatives
[32] or some N containing unsaturated acids [36,37] opposite ten-
dencies were also observed. The presence of BA in the reaction
3. Results
3.1. Effect of the modifier structure
The hydrogenation of 1 and its substituted derivatives over CD-
modified Pd catalyst results in the excess formation of the corre-
sponding (S)-2,3-diphenylpropionic acids (see Scheme 1). The ees
obtained in the hydrogenation of the carboxylic acids selected for
this study (Fig. 1) in the absence and presence of BA are presented
in Table 1. The reaction conditions were selected based on our
X
X
+H2
Y
Y
Pd/Al2O3 + CD
COOH
COOH
S
mixture increased the ee and the Ri over CN-modified Pd, too.
0
The cinchona alkaloids bearing OCH3 substituent in the C6 position
Scheme 1. Enantioselective hydrogenation of (E)-2,3-diphenylpropenoic acids over
were found much less efficient in inducing enantioselection,
though in the hydrogenation of 5 even with QN up to 65% ee was
obtained. In the hydrogenations using the four parent cinchona
alkaloids, the acid structure had similar effect on the ee, i.e. the
ee decreased in the order: 5 > 3 > 4 > 2 > 1 > 6.
Pd catalyst modified by CD.
X
Based on the suggested CD-acid intermediate [21], one would
expect the formation of close-to-racemic products over Pd modi-
fied by the cinchona methyl ethers. However, the report on the
hydrogenation of 1 using dihydro-CDM already prognosticated a
peculiar behaviour in the presence of such modifiers. When CDM
was used in the absence of BA, the results obtained were more or
less in line with the expected low ees. However, in the presence
of BA even with this modifier, moderate ees were obtained (3, 4
and 5 up to 42%, 38% and 54% ee, respectively). Completely different
behaviour was observed in the hydrogenations over CNM-modified
catalyst. Moderate ees were obtained with each substrate, more-
over, in the hydrogenation of 6, the ee exceeded even the value
obtained in the presence of CD. Moreover, inversion of the sense
of the ee was obtained in the hydrogenation of each acid when
compared with the parent CN. The highest ees over CNM-modified
catalyst were obtained in the hydrogenation of acids 2 and 3, (52%
and 55%) in contrast with the parent cinchonas. The ee and Ri
decreased in the presence of BA (except 1), though this additive
3
Y
N
COOH
O
H
8
9
H
R
(1) X = H, Y = H
S
(2) X = 2-CH3O, Y = H
(3) X = H, Y = 4-CH3O
(4) X = H, Y = 4-F
N
(5) X = 2-CH3O, Y = 4-F
β-ICN
(6) X = 2-CH3O, Y = 2-F
5
3
4
6
11
10
2
N
S
N
O
R
R
O
7
8
8
9
9
H
H
R
H
H
R
S
5'
4'
R'
R'
6'
7'
6'
3'
2'
8'
N
N
usually has opposite effect [9,18].
0
The C6 AOCH3 substituted methyl ethers (QNM and QDM) gave
(CN) R = H, R' = H
(CD) R = H, R' = H
(CNM) R = CH3, R' = H
(QD) R = H, R' = CH3O
(QDM) R = CH3, R' = CH3O
(CDM) R = CH3, R' = H
(QN) R = H, R' = CH3O
(QNM) R = CH3, R' = CH3O
lower ees than CDM and CNM, as expected based on the results ob-
tained with the four parent cinchona alkaloids. Although the
hydrogenation of 1 and 2 resulted in racemic products over
QNM-modified catalyst, the other acids led to small excess of the
R enantiomers in the absence of BA, opposite to QN. Over catalyst
modified by QDM, the inversion of the sense of the ee observed by
Fig. 1. Structure of the examined substituted 2,3-diphenylpropenoic acids and
cinchona alkaloid derivatives.
Table 1
Effect of alterations in the modifier structure on the ee and Ri in the hydrogenation of (E)-2,3-diphenylpropenoic acids 1–6.a,b
Modifier
1
2
3
4
5
6
CD
CN
QN
QD
CDM
CNM
QNM
QDM
b-ICN
70S/73S (8/12)b
34R/37R (13/16)
11S/6S (11/20)
3S/2R (12/20)
3S/22S (10/12)
47S/34S (9/10)
0/1S (22/33)
12S/2S (17/32)
–
76S/85S (6/8)
29R/45R (4/7)
15S/5S (4/8)
3S/1R (5/9)
7R/10S (8/7)
52S/41S (5/3)
0/0 (6/11)
83S/89S (5/11)
47R/56R (6/7)
40S/17S (6/6)
11R/14R (5/6)
16S/42S (7/7)
55S/37S (6/5)
16R/2S (7/8)
27S/1S (6/7)
–
73S/85S (10/15)
38R/53R (8/9)
27S/16S (7/8)
6R/8R (7/9)
6S/38S (8/10)
42S/22S (7/6)
13R/2S (8/10)
20S/1R (7/12)
–
86S/93S (3/7)
55R/74R (4/6)
65S/36S (3/5)
21R/30R (4/7)
25S/54S (6/5)
44S/34S (7/7)
9R/2S (5/4)
20S/29S (5/7)
9R/11R (4/5)
11S/4S (4/10)
2R/4R (4/4)
4R/3S (4/4)
29S/15S (5/5)
–
12S/2S (5/8)
22S/16S (4/5)
30S/0 (4/4)
3S/1R (5/4)
–
–
a
Reaction conditions: 25 mg 5% Pd/Al2O3, 5 cm3 DMF + 2.5 vol.% H2O, 0.025 mmol modifier ([modifier] = 5 mM), 0.5 mmol substrate (for the structure of the acids see
Fig. 1), 0.1 MPa H2, 295 K, ee determined at X = 95–100%.
b
ee (%) and absolute configuration of the product obtained in the absence/presence of 0.5 mmol BA, in brackets the Ri (mmol hꢀ1 gꢀ1) in the absence/presence of BA
determined from the H2 uptake curves.

} }
G. Szollosi et al. / Journal of Catalysis 276 (2010) 259–267
262
using CNM was also detected, though the ees were lower (up to
30%) and further decreased in the presence of BA. Replacing the
methyl ethers with the rigid, cyclic ether derivative b-ICN in the
reaction of 2 and 5, the ee decreased when compared with CNM,
however, inversion of the sense of the enantioselectivity was also
observed.
The interpretation of these results should take into account
both the alterations in the interaction of the cinchona alkaloids
with the acids and the metal surface. Methods for investigating
simultaneously these interactions under the conditions of the
hydrogenations are not yet available. However, useful hints may
be obtained by examining the alkaloid–acid complex in solution
and treating the results carefully due to the essential effect of the
metal surface and adsorbed hydrogen on the structure of the
intermediate.
selected modifiers with the acids 1, 2, 3 and 5 was investigated
in d7-DMF + 2.5 vol.% D2O solvent. The signals in 1H and 13C NMR
spectra were assigned using published data [16,38–40]. Variations
in the chemical shift (Dd) of the H and C signals were used to indi-
cate the relative strength of the cinchona–acid interactions. Alter-
ations may also result due to the conformational mobility of the
0
alkaloids by rotation around the C4 AC9 and C9AC8 axis [38,39,41],
which occur when the cinchona is protonated on the quinuclidine
N [38,42].
The
D
d obtained in the 1H NMR spectra of CD, in the presence of
1, 2, 3 and 5 and CDM and CNM in the presence of 5 showed the
formation of the acid salts as expected (see Fig. 2a and Supporting
information: Table SI-1). However, in the spectra of CD, the H9 sig-
nal was shifted the most, which is attributed to the involvement of
the C9AOH of CD in H-bonding, as was previously assumed [21].
Similarly, in the 13C NMR spectra besides the signals of the quinu-
0
clidine C atoms, the C9 and C4 signals were also shifted (see Fig. 2b
3.2. NMR spectroscopic investigation of the modifier–acid–BA
interaction
and Table SI-2), supporting the involvement of the C9AOH in the
interaction. In the spectra of the methyl ethers, the smaller
Dd of
the H9 and C9 signals (Fig. 2) when compared with CD showed that
the acid does not or interacts much weaker with the C9AOCH3
group than with the C9AOH.
In the present study, the cinchona–acid interaction in liquid
phase was examined by NMR spectroscopy. The interaction of
The
Dd values in the spectra of CD were dependent on the struc-
Sample
ture of the acids (Figs. 2 and 3). As the acidity of 1 is higher than of
3 (pKa(1) 7.00; pKa(3) 7.23 [43]), the stronger shift of the H9 signal
in presence of the former is attributed to the looser contact be-
tween the N+AH and the carboxylate group, resulting in a stronger
H-bonding to the C9AOH. The shift of this signal was found to be
much smaller in the presence of the acids ortho-OCH3 substituted
CD+1
CD+3
CD+2
CD+5 CDM+5 CNM+5
(a)
0
-0.1
-0.2
-0.3
-0.4
-0.5
-0.6
-0.7
-0.8
H3’
on the
a-phenyl ring (2 and 5). Accordingly, this substituent in-
creases the N+AHꢂ ꢂ ꢂOAC@O interaction strength when compared
with 3 and decreases the strength of the C9AOAHꢂ ꢂ ꢂO@CAO bond.
H8
Addition of BA to the samples decreased the
Dd values, however,
remained still shifted when compared with the spectra of the mod-
ifiers (Figs. 3 and 4 and Fig. SI-2).
H2t
H6en
H9
H11c/H11t
H9 (6.4)
H10
Δδ (ppm)
CD
Δδ (ppm)
H9 (~1.8)
(b)
5
C7
CD+1
H9
4
3
CD+3
C9
H9
CD+2
C4’
C2
H9
2
CD+5
H9 (3.7)
CD+1+BA
C8
1
H9 (5.0)
CD+3+BA
0
H9 (5.0)
CD+2+BA
C6
H9 (5.0)
-1
CD+5+BA
CD+1
CD+3
CD+2
CD+5
CDM+5 CNM+5
Sample
6.3 6.2 6.1
6
5.9 5.8 5.7 5.6 5.5 5.4 5.3 5.2 5.1
5
4.9
D
d (ppm), in the 1H NMR (a) and 13C NMR (b)
δ (ppm)
Fig. 2. Chemical shift changes,
spectra of cinchona alkaloids in the presence of (E)-2,3-diphenylpropenoic acids
(for atom numbering see Fig. SI-1; the lines between the data points are just to
guide the eye).
Fig. 3. 1H NMR traces of CD, CD + acid and CD + acid + BA solutions (see Section 2.3;
3J(H8AH9) values (Hz) are given in brackets).

}
}
G. Szollosi et al. / Journal of Catalysis 276 (2010) 259–267
263
H9 (6.4)
H11c/H11t
Table 2
H10
Enantioselectivities obtained in the hydrogenation of (E)-2,3-diphenylpropenoic acid
derivatives 1, 2 and 4 in presence of 1/1 modifier mixtures.
CD
H9
Modifier mixture
Additive
(E)-2,3-diphenylpropenoic acid derivatives
CD+5
1
2
4
eea (eecalc
)
ee (eecalc
)
ee (eecalc
)
H9 (5.0)
CD+5+BA
CD + CN
CD + CN
–
BA
–
BA
–
BA
–
BA
–
BA
19S (5S)
22S (32S)
25S (26S)
68S (40S)
67S (39S)
30R (7R)
38R (18R)
20R (14R)
30R (16R)
15R (14S)
31R (18R)
19S (25S)
36S (33S)
73S (41S)
82S (49S)
35R (7R)
33R (21R)
24R (15R)
39R (6R)
28R (0)
H9 (6.9)
H11c/H11t
22S (11S)
67S (29S)
68S (27S)
31R (14R)
37R (13R)
23R (18R)
32R (11R)
20R (1R)
45R (9R)
H10
CD + QDb
CD + QD
CN + QN
CN + QN
CN + CDM
CN + CDM
CN + CNM
CN + CNM
CDM
H9 (3.7)
H9 (5.5)
CDM+5
CDM+5+BA
47R (25R)
H9
H11c/H11t
Reaction conditions: 25 mg 5% Pd/Al2O3, 5 mL DMF + 2.5 vol.% H2O, 0.025 mmol 1/1
mixture of modifier ([modifier] = 5 mM), 0.5 mmol substrate (see Fig. 1), 0.1 MPa
H2, 295 K, ee determined at X = 90–100%.
H10
CNM
a
ee – enantiomeric excess obtained experimentally and the configuration of the
excess enantiomer, eecalc – theoretical ee corresponding to the linear behaviour (see
Section 2).
H9 (4.1)
H9 (5.9)
CNM+5
b
The ‘‘dominating” modifier in determining the enantioselectivity with bold
characters.
CNM+5+BA
6.3 6.2 6.1
6
5.9 5.8 5.7 5.6 5.5 5.4 5.3 5.2 5.1
5
4.9
Table 3
δ (ppm)
Enantioselectivities obtained in the hydrogenation of 5 using 1/1 modifier mixtures.
Fig. 4. 1H NMR traces of cinchona derivatives in the absence and presence of 5 and
BA (see Section 2.3; 3J(H8AH9) values (Hz) are given in brackets).
Modifier mixture
Without additive
Using BA additive
ee (eecalc
eea (eecalc
)
)
CD + CNb
CD + QD
CD + QN
CD + CNM
CN + QN
CN + QD
CN + CDM
CN + CNM
QD + QN
QD + CDM
QD + CNM
31S (2S)
33S (10S)
94S (27S)
93S (68S)
89S (63S)
75R (26R)
74R (50R)
56R (15R)
71R (18R)
4S (4R)
89S (27S)
87S (75S)
72S (56S)
47R (3R)
53R (40R)
22R (10R)
43R (6S)
19S (21S)
10S (7S)
The values of 3J(H8AH9) coupling constant may be used for the
approximation of the equilibrium population of the cinchona con-
formers [38,39]. In our solvent mixture, the 6.4 Hz value (Fig. 3)
corresponds to a relatively low population (ꢃ40%) of open(3) con-
former [38], and closed conformers, which are irrelevant from the
point of view of the hydrogenation over the metal surface, are in
excess. Addition of 1 to the CD solution decreased significantly
the 3J(H8AH9) to 1.8 Hz, which is close to that calculated for the
open(3) conformer [38]. The shape of the H9 signal in the presence
of the substituted acids used in our study indicated the split of this
signal to superimposed doublets with low 3J(H8AH9). Based on the
3J(H8AH9) values (Fig. 4) in our solvent mixture, CDM was found to
be in ꢃ33% (6.9 Hz) as the open(3) conformer, which increased in
the presence of 5 to 75% (3.7 Hz), For CNM, a similar value of
ꢃ70% (4.1 Hz) was found. The H9 signals appeared as doublets in
contrast to the superimposed double doublets found in the spectra
of CD.
42S (6S)
35S (1S)
46S (22S)
Reaction conditions: 25 mg 5% Pd/Al2O3, 5 mL DMF + 2.5 vol.% H2O, 0.025 mmol 1/1
mixture of modifier ([modifier] = 5 mM), 0.5 mmol substrate (see Fig. 1), 0.1 MPa
H2, 295 K, ee determined at X = 90–100%.
a
ee – enantiomeric excess obtained experimentally and the configuration of the
excess enantiomer, eecalc – theoretical ee corresponding to the linear behaviour (see
Section 2).
b
The ‘‘dominating” modifier in determining the enantioselectivity with bold
characters.
By addition of BA to the samples containing CD, the 3J(H8AH9)
increased indicating that the additive may be attached to the
CD–acid complex and alters the population of the different con-
formers. The increase in the coupling constant was also detected
in the samples containing CDM and CNM, thus the interaction of
BA with the cinchona–acid complex occurs even if the C9AOH is
protected by the methyl group.
the lower ee obtained with CN when compared with CD. Similarly,
a mixture of QN and QD in the hydrogenation of 5 resulted only in
small deviations from the linear behaviour. Prominent non-linear
0
behaviour was obtained using mixtures of the C6 -substituted QN
or QD with CN or CD. The ee obtained with these mixtures were
close to those attained by using the latter modifiers alone. Interest-
ingly, in some reactions, the 1/1 mixtures led to slightly higher ee
than the sole CD or CN, i.e. CD + QD, CN + QN (see Table 3). The use
of the methyl ethers (CDM or CNM) in mixture with CD or CN also
resulted in marked deviation from the linear behaviour, the latter
modifiers dominating the chiral induction. Using QD + methyl
ether (CDM or CNM) mixtures, the enantiodifferentiation was
dominated by the methyl ethers and the ees deviated from the cal-
culated values. Moreover, in the hydrogenations using QD + CNM,
the ees were higher than with CNM alone.
As in some reactions higher ees were obtained than with the
sole ‘‘dominating” modifier, we investigated the effect of the CD
amount to ascertain that the results are not caused by variations
in the modifier concentration. The effect of the CD concentration
([CD]) on the ee and Ri in the hydrogenation of 5 is presented in
3.3. Hydrogenations using modifier mixtures
The relative adsorption strength of the four parent cinchona
alkaloids and their methyl ethers during the hydrogenation of 1,
2, 4 and 5 was investigated by using 1/1 modifier mixtures both
in the absence and presence of BA (see Tables 2 and 3). The exper-
imentally obtained ee values were compared with those calculated
presuming a linear behaviour (for Ri see Supporting information
Tables SI-3 and SI-4).
The deviations from the linear behaviour obtained by the use of
the mixture of CD and CN were much smaller as expected based on

}
}
264
G. Szollosi et al. / Journal of Catalysis 276 (2010) 259–267
100
ee(BA)273
ee
xQD
15
12
9
(a)
1
0.8
0.6
0.4
0.2
0
ee % 100
ee(BA)
90
80
80
60
40
20
0
Ri
70 ee (%)
S
Ri(BA)
6
60
50
40
eecalc
Ri
3
Ri(BA)273
4
0
0
2
6
8
10
[CD] (mM)
Fig. 5. Effect of [CD] on the ee (closed symbols) and Ri (mmol hꢀ1 gꢀ1
, open
-20
-40
R
eecalc(BA)
symbols) in the hydrogenation of 5 in the absence (N, 4) and in the presence of BA
(j, h at 295 K; ꢀ, } at 273 K). For reaction conditions see Table 1 and Section 2.
0
1
0.2
0.8
0.4
0.6
0.4
0.8
0.2
1
0
xCD
xQD
Fig. 5. A continuous increase in the ee was obtained by increasing
the [CD] up to 2.5 mM, followed by constant or very slow increase
by further increase in the [CD]. Accordingly, with cinchona mix-
tures, the increase in the ee over the value obtained with CD should
be due to the presence of the second alkaloid. Surprising was the
low Ri attained during the racemic hydrogenation in the presence
of BA. Although the increase in the [CD] led to decrease in the Ri,
in the presence of BA, the racemic hydrogenation proceeded at
much lower rate than at low [CD].
The non-linear behaviour in the hydrogenation of 5 was studied
in detail using the modifiers pairs: CD + QD, CN + QN and CN + CNM,
respectively. The ee and Ri values vs the modifier composition were
plotted in Figs. 6–8. Using CD + QD, prominent alterations from
the linear behaviour was observed (Fig. 6). CD dominated the
enantioselection, similarly with the hydrogenation of 4-methoxy-
6-methyl-2-pyrone [32]. The Ri values deviated substantially from
the values corresponding to the linear behaviour in the absence of
BA at low CD/QD ratio, approaching those obtained in the presence
of BA (see Fig. 6b). Similar behaviour showed the CN + QN pair,
with CN dominating the enantioselection (Fig. 7). The found Ri
values were close to the values reached with CN alone, except in
the absence of BA at low CN amounts. Pronounced non-linear
behaviour was also obtained with the CN + CNM mixtures, CN
determining the enantioselection (Fig. 8). The Ri reached the value
0.6
(b)
9
8
7
6
5
4
3
2
Ricalc(BA)
Ri
Ricalc
0
0.2
0.4
0.6
0.8
1
xCD
Fig. 6. Effect of CD + QD mixture composition on the ee (%) (a) and Ri
(mmol hꢀ1 gꢀ1) (b) obtained in the hydrogenation of 5 in the absence (N) and in
the presence (j) of BA. For reaction conditions see Table 3, [modifier] = 5 mM; eecalc
and Rci alc – calculated ee and Ri curves corresponding to the linear behaviour (see
Section 2).
corresponding to pure CN even at low CN amounts (up to xCN
0.25) and remained constant to this value on further increase in
the xCN
=
.
4. Discussions
alkaloid–diphenylpropenoic acid interaction is not yet fully under-
stood, we attempted to explain this inversion based on investigat-
ing the alterations in the cinchona–acid interactions in solution
and the competitive adsorption of the cinchona derivatives on
the Pd combined with the effect of substitution of the acid on
the phenyl rings.
Although the effect of the substituents of both the acid phenyl
rings and the cinchona C9AO on the alkaloid–acid interaction
was studied by NMR spectroscopy in solution, we presume that
the found effects on the modifier–acid bonding may also occur un-
der the reaction conditions, complemented by the effect of the
hydrogen saturated metal surface. The results demonstrated the
binding of the acids to both the quinuclidine N and the C9AOH of
the alkaloid, as presumed previously [21]. Decrease in the acidity
(by the b-phenyl-para-OCH3) increased the strength of the
N+AHꢂ ꢂ ꢂOAC@O bond and weakened the C9AOHꢂ ꢂ ꢂO@CAO bond,
The present study on the hydrogenation of (E)-2,3-diphenyl-
propenoic acids over Pd/Al2O3 modified by the four parent cin-
chona alkaloids and their methyl ethers showed inversion of the
sense of the ee by using the (R)C8A(S)C9 cinchona ethers. Inver-
sions of the sense of the ee were reported in asymmetric catalytic
reactions inclusive in hydrogenations over modified metal cata-
lysts [44]. Most of these occurred in the hydrogenation of activated
ketones over Pt [28–30,35,44–51]. Inversion of the ee in the hydro-
genation of prochiral olefins over Pd was seldom reported
[21,36,52,53]. Our previous study demonstrated that the adsorp-
tion of the unsaturated acids in presence of CD is directed by
the b-substituent [11]. According to the present study in the
hydrogenation of diaryl substituted acids, the adsorption may
also be inverted by methylation of the modifier, without changing
the configuration of the stereogenic centre. As the cinchona

}
}
G. Szollosi et al. / Journal of Catalysis 276 (2010) 259–267
265
xQN
xCNM
(a)
1
0.8
0.6
0.4
0.2
0
1
0.8
0.6
0.4
0.2
0
(a)
ee % 80
60
40
ee %
eecalc(BA)
60
eecalc(BA)
eecalc
S
S
40
20
20
eecalc
0
0
-20
-40
-60
-80
-20
-40
-60
-80
R
R
0
1
0.2
0.8
0.4
0.6
0.8
0.2
1
0
0
1
0.2
0.8
0.4
0.6
0.4
0.8
0.2
1
0
xCN
xQN
xCN
xCNM
(b)
(b)
0.6
0.6
0.4
7
6
5
8
7
6
Ricalc(BA)
Ricalc(BA)
Ri
Ri
4
3
2
5
4
3
Ricalc
Ricalc
0
0.2
0.4
0.6
0.8
1
0
0.2
0.4
0.6
0.8
1
xCN
xCN
Fig. 7. Effect of CN + QN mixture composition on the ee (%) (a) and Ri
(mmol hꢀ1 gꢀ1) (b) obtained in the hydrogenation of 5 in the absence (N) and in
the presence (j) of BA. For reaction conditions see Table 3, [modifier] = 5 mM; eecalc
and Rci alc – calculated ee and Ri curves corresponding to the linear behaviour (see
Section 2).
Fig. 8. Effect of CN + CNM mixture composition on the ee (%) (a) and Ri
(mmol hꢀ1 gꢀ1) (b) obtained in the hydrogenation of 5 in the absence (N) and in
the presence (j) of BA. For reaction conditions see Table 2, [modifier] = 5 mM; eecalc
and Rci alc – calculated ee and Ri curves corresponding to the linear behaviour (see
Section 2).
indicated by the
D
d of the H9 signal. The
a
-phenyl-ortho-OCH3 sub-
explain the acid structure-dependent split of the H9 signal leading
to superimposed doublets. In contrast, the H9 doublet was not split
in the spectra of the methyl ethers in the presence of 5, suggesting
that these conformers are not formed when the C9AOH is
protected.
stituent (2, 5) increases more the strength of the N+AHꢂ ꢂ ꢂOAC@O
bond and further decreases the C9AOHꢂ ꢂ ꢂO@CAO interaction
strength. Although it is not yet clarified by what means, the ee
increasing effect of the
a-phenyl-ortho-OCH3 group is due to this
change in the strengths of the two H-bonds.
In the presence of BA, the less shifted NMR signals may be due
to the interaction of the CD–acid complex with the additive as
suggested in Fig. 10 (S3–S5). However, the present experiments
cannot differentiate between the structures; attempts to identify
the real intermediate(s) are in progress. Similarly, using the cinchona
methyl ethers in the presence of BA structures containing the
additive may form (Fig. 10, Fig. S6), the illustrated flexible contact
could explain the decrease in the ee in the presence of BA. The
increase in the 3J(H8AH9) in the presence of BA showed either
a decrease in the population of the open(3) conformer or a trans-
formation of the suggested ‘‘bridged” conformers by interaction
with BA.
Accordingly, the CD–acid interaction may be sketched similarly
as previously supposed [21] (Fig. 9, Fig. S1), complemented with
the mentioned equilibration in the strengths of the two H-bonds
vital for obtaining high ee. In the presence of the cinchona methyl
ethers, we assume the formation of structures S2 and S20 (Fig. 9)
which allow higher mobility to the acids. The decrease in the
3J(H8AH9) by the acids showed transformation of the modifiers
either to open(3) conformer or to other stable conformers, such
as the so-called ‘‘bridged” conformers resulted in the presence of
0
HCl or HF, having the C4 AC9AC8AN torsional angle rotated with
almost 360° (when compared with the open(3)) [42]. This may

}
}
266
G. Szollosi et al. / Journal of Catalysis 276 (2010) 259–267
yδ-
yδ-
O
O
xδ-
xδ-
xδ-
O
O
O
yδ-
+
H
O
+
R
+
R
R
H
H
N
N
N
H
O
H
O
H
CH3
O
H
CH3
H
H
H
9
9
9
(S1)
(S2)
(S2')
Fig. 9. Proposed structures of the modifier–acid complexes in liquid phase with the parent cinchona alkaloids (S1) and with their methyl ethers (S2, S20).
xδ-
H
O
H
Oyδ-
xδ-
vδ+O
N
H
zδ+
N
H
R
Oyδ-
H
H
zδ+
N
R
Nvδ+
O
H
H9
H
O
(S4)
(S3)
H
H9
H
H
yδ-
O
yδ-
O
xδ-
xδ-
O
O
H
H
N vδ+
zδ+
N
Nvδ+
H
zδ+
N
H
H
R
R
H
CH3
H
O
O
H9
H9
H
(S6)
(S5)
Fig. 10. Proposed structures of the modifier-acid complexes in liquid phase in the presence of BA with the parent cinchona alkaloids (S3–S5) and with their methyl
ethers (S6).
The non-linear behaviour obtained using cinchona alkaloid
allowing the inversion of the docking preference of the acid, also
due to the flexibility of the acid–modifier interaction and the steric
hindrances exerted by the bulky substrates, in contrast with the
pairs indicated large differences in the adsorption of the modi-
0
fier-acid salts as effect of the presence of either C6 AOCH3 or
the C9AOCH3 group. The adsorption strength order derived from
the ‘‘domination” order was CD ꢄ CN > CNM ꢄ CDM > QN ꢄ QD,
resembling the one found in the hydrogenation of 4-methoxy-6-
methyl-2-pyrone [33]. Accordingly, the lower ee obtained with
CN or QD when compared with CD or QN cannot be attributed to
the adsorption strength of the modifiers. Thus, this may be as-
cribed to the less efficient CN(QD)Aacid interaction for inducing
enantiodifferentiation, as recently suggested [23]. As the basic
strength of CN is slightly higher than that of CD [54], the decrease
in the N+AHꢂ ꢂ ꢂOAC@O interaction strength in the CN–acid com-
plex explains the lower ee. The studies with the cinchona alkaloid
pairs over the full composition range showed that the adsorption
aliphatic
a
,b-unsaturated carboxylic acids. This suggestion is sup-
ported by the opposite effect on the ee of the
a- and b-phenyl-
ortho substituents of the acid in the reshaped chiral site, when
compared with the parent cinchonas. Thus, better ee was obtained
in the hydrogenation of 6 in the presence of CNM than CD and low-
er ee resulted in the hydrogenation of 5 than in that of 3 or 1 over
CNM-modified catalyst. As the interaction of the acids was proved
to be less efficient with the (R)C8A(S)C9 cinchonas, the reshaping
effect induced by the methyl substituent had stronger effect on
the enantiodifferentiation by using these modifiers than their
(S)C8A(R)C9 stereoisomers. Moreover, the differences in the hydro-
genation of the anchoring quinoline moiety of the cinchonas
belonging to the two series over metal catalyst [55], which under
the present hydrogenation conditions we will study in the future,
may also lead to the observed differences in the results obtained
with CDM and CNM.
Finally, in the presence of BA, the much higher rate observed at
low [CD] when compared with the racemic hydrogenation is a no-
vel evidence on the ligand-accelerated catalytic hydrogenation of
the acids over Pd, similarly with the enantioselective hydrogena-
tion of activated ketones over cinchona-modified Pt [5,56]. The
acceleration effect of the modifier was only very recently observed
for the first time over Pd in the absence of BA [23].
strength of the cinchona-acid salt is influenced decisively by the
0
C6 AOMe substituent, by promoting the tilted adsorption of the
quinoline moiety [27,33]. This weak adsorption leads to the low
ees by using these modifiers, while in mixture with CD, these
derivatives may take the place of the BA additive in increasing
the Ri and the ee (see Figs. 5 and 6, i.e. at [CD] 1.25 mM ee 75%
and 81%, respectively). The deviations from the linearity obtained
with mixtures of CD or CN and cinchona methyl ethers are caused
by both the weakened interaction strength of the acid with the
methyl ethers and the alterations in the adsorption of the methyl
ether–acid complexes on the surface. It is known that the parent
cinchonas adsorb stronger on the Pd surface than their methyl
ethers, as shown by the higher adsorption energy of CD when com-
pared with CDM [50].
5. Conclusions
Accordingly, the origin of the inversion obtained by using the
(R)C8A(S)C9 cinchona ethers may be traced back on the altered
acid–cinchona methyl ether interaction, which should be coupled
with an additional effect of the C9AOCH3 group on the surface
leading to switch in the preferred arrangement of the acid. We
suggest that this occurs due to reshaping of the chiral metal sites
by the methyl substituent by its steric interference within the site,
Our study on the enantioselective hydrogenation of (E)-2,3-
diphenylpropenoic acids over Pd/Al2O3 modified by the four parent
cinchona alkaloids, their methyl ethers and b-isocinchonine re-
vealed the inversion of the sense of the enantioselectivity in pres-
ence of the (R)C8A(S)C9 cinchona ethers. In search for explanation,
we studied the interaction of the acids with CD, CDM and CNM in
liquid phase by NMR spectroscopy. The results demonstrated the

}
}
G. Szollosi et al. / Journal of Catalysis 276 (2010) 259–267
267
[13] T. Sugimura, J. Watanabe, T. Okuyama, Y. Nitta, Tetrahedron: Asymmetry 16
(2005) 1573.
formation of bidentate complexes in which the carboxylate group
is H-bonded to the C9AOH group accompanied by a conformational
change of the CD. The cinchona methyl ethers form flexible mono-
dentate complexes without involvement of the C9–OCH3 group. It
was assumed that the latter group influenced the shape of the sur-
face chiral sites based on the non-linear effect obtained by using
binary mixtures of the parent cinchona alkaloids with the methyl
ethers. Accordingly, the bulkiness and rigidity of these unsaturated
acids may lead to steric interference even with the methyl group
within the chiral site, while the flexibility of the acid–modifier
interaction, which is less efficient with (R)C8A(S)C9 cinchonas than
with the (S)C8A(R)C9 stereoisomers, allows the inversion of the
docking preference of the acid on the surface.
[14] Y. Nitta, J. Watanabe, T. Okuyama, T. Sugimura, J. Catal. 236 (2005) 164.
[15] T. Sugimura, J. Watanabe, T. Uchida, Y. Nitta, T. Okuyama, Catal. Lett. 112
(2006) 27.
}
}
[16] G. Szollosi, K. Balázsik, M. Bartók, Appl. Catal. A: Gen. 319 (2007) 193.
}
}
[17] G. Szollosi, B. Hermán, K. Felföldi, F. Fülöp, M. Bartók, J. Mol. Catal. A: Chem.
290 (2008) 54.
}
}
[18] G. Szollosi, B. Hermán, K. Felföldi, F. Fülöp, M. Bartók, Adv. Synth. Catal. 350
(2008) 2804.
}
}
[19] G. Szollosi, Z. Makra, M. Bartók, React. Kinet. Catal. Lett. 96 (2009) 319.
[20] T. Sugimura, T. Uchida, J. Watanabe, T. Kubota, Y. Okamoto, T. Misaki, T.
Okuyama, J. Catal. 262 (2009) 57.
[21] Y. Nitta, A. Shibata, Chem. Lett. 27 (1998) 161.
[22] K. Borszeky, T. Bürgi, Z. Zhaohui, T. Mallat, A. Baiker, J. Catal. 187 (1999) 160.
[23] T. Sugimura, H. Ogawa, Chem. Lett. 39 (2010) 232.
[24] D. Ferri, T. Bürgi, A. Baiker, J. Chem. Soc. Perkin Trans. 2 (2002) 437.
[25] D.M. Meier, A. Urakawa, N. Turrà, H. Rüegger, A. Baiker, J. Phys. Chem. A 112
(2008) 6150.
We stress out that the present study is the first in which signif-
icant inversion of the ee, detail study on the non-linear behaviour
and novel evidence on the ligand-accelerated mechanism in the
enantioselective hydrogenation of (E)-2,3-diphenylpropenoic acids
over cinchona alkaloid-modified Pd are presented.
[26] K.E. Simons, P.A. Meheux, A. Ibbotson, P.B. Wells, Stud. Surf. Sci. Catal. 75
(1993) 2317.
[27] W.-R. Huck, T. Bürgi, T. Mallat, A. Baiker, J. Catal. 216 (2003) 276.
[28] S. Diezi, T. Mallat, A. Szabo, A. Baiker, J. Catal. 228 (2004) 162.
}
}
[29] M. Bartók, M. Sutyinszki, I. Bucsi, K. Felföldi, G. Szollosi, F. Bartha, T. Bartók, J.
Catal. 231 (2005) 33.
Acknowledgments
[30] F. Hoxha, L. Königsmann, A. Vargas, D. Ferri, T. Mallat, A. Baiker, J. Am. Chem.
Soc. 129 (2007) 10582.
}
}
[31] K. Balázsik, I. Bucsi, S. Cserényi, G. Szollosi, M. Bartók, J. Mol. Catal. A: Chem.
280 (2008) 87.
Financial support by the Hungarian National Science Foundation
(OTKA Grant K 72065) is highly appreciated. The work was sup-
ported by the János Bolyai Research Scholarship of the Hungarian
[32] W.-R. Huck, T. Mallat, A. Baiker, Adv. Synth. Catal. 345 (2003) 255.
[33] W.-R. Huck, T. Mallat, A. Baiker, Catal. Lett. 87 (2003) 241.
[34] C. Exner, A. Pfaltz, M. Studer, H.-U. Blaser, Adv. Synth. Catal. 345 (2003) 1253.
} }
Academy of Sciences (G. Szollosi).
}
}
[35] M. Bartók, M. Sutyinszki, K. Felföldi, G. Szollosi, Chem. Commun. (2002) 1130.
}
}
[36] G. Szollosi, E. Szabó, M. Bartók, Adv. Synth. Catal. 349 (2007) 405.
}
}
}
[37] G. Szollosi, K. Szori, M. Bartók, J. Catal. 256 (2008) 349.
[38] T. Bürgi, A. Baiker, J. Am. Chem. Soc. 120 (1998) 12920.
[39] I. Busygin, V. Nieminen, A. Taskinen, J. Sinkkonen, E. Toukoniitty, R. Sillanpää,
D.Yu. Murzin, R. Leino, J. Org. Chem. 73 (2008) 6559.
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.jcat.2010.09.020.
[40] B. Halton, A.I. Maidment, D.L. Officer, J.M. Warnes, Austr. J. Chem. 37 (1984)
2119.
[41] G.D.H. Dijkstra, R.M. Kellog, H. Wynberg, J. Org. Chem. 55 (1990) 6121.
[42] R.A. Olsen, D. Borchardt, L. Mink, A. Agarwal, L.J. Mueller, F. Zaera, J. Am. Chem.
Soc. 128 (2006) 15594.
[43] K. Bowden, D.C. Parkin, Can. J. Chem. 46 (1968) 3909.
[44] M. Bartók, Chem. Rev. 110 (2010) 1663.
References
[1] T. Ohkuma, M. Kitamura, R. Noyori, Asymmetric hydrogenation, in: I. Ojima
(Ed.), Catalytic Asymmetric Synthesis, second ed., Wiley-VCH, New York, 2000,
p. 1. Chapter 1.
[2] H.-U. Blaser, C. Malan, B. Pugin, F. Spindler, H. Steiner, M. Studer, Adv. Synth.
Catal. 345 (2003) 103.
[45] M. von Arx, T. Mallat, A. Baiker, Angew. Chem. Int. Ed. 40 (2001) 2302.
}
}
[46] G. Szollosi, C. Somlai, P.T. Szabó, M. Bartók, J. Mol. Catal. A: Chem. 170 (2001)
165.
[3] K. Ding, Y. Uozumi (Eds.), Handbook of Asymmetric Heterogeneous Catalysis,
Wiley-VCH, Weinheim, 2008.
}
}
[47] G. Szollosi, S. Cserényi, M. Bartók, Catal. Lett. 134 (2010) 264.
[48] O.J. Sonderegger, G.M.W. Ho, T. Bürgi, A. Baiker, J. Catal. 230 (2005) 499.
[49] N. Bonalumi, A. Vargas, D. Ferri, T. Bürgi, T. Mallat, A. Baiker, J. Am. Chem. Soc.
127 (2005) 8467.
[4] M. Studer, H.-U. Blaser, C. Exner, Adv. Synth. Catal. 345 (2003) 45.
[5] T. Mallat, E. Orglmeister, A. Baiker, Chem. Rev. 107 (2007) 4863.
}
}
[6] G. Szollosi, Magy. Kem. Foly. 113 (2007) 146.
[50] N. Bonalumi, A. Vargas, D. Ferri, A. Baiker, Chem. Eur. J. 13 (2007) 9236.
[51] I. Busygin, J. Wärnå, E. Toukoniitty, D.Yu. Murzin, R. Leino, J. Catal. 254 (2008)
339.
[7] J.R.G. Perez, J. Malthête, J. Jacques, CR Acad. Sci. Paris, Ser II 300 (1985) 169.
[8] K. Borszeky, T. Mallat, A. Baiker, Tetrahedron: Asymmetry 8 (1997) 3745.
[9] Y. Nitta, Top. Catal. 13 (2000) 179.
[52] W.R. Huck, T. Bürgi, T. Mallat, A. Baiker, J. Catal. 219 (2003) 41.
[53] N.J. Colston, R.P.K. Wells, P.B. Wells, G.J. Hutchings, Catal. Lett. 103 (2005) 117.
[54] C. Drzewiczak, A. Suszko-Purzycka, J. Skolik, Polish J. Chem. 67 (1993) 45.
[10] M. Maris, W.-R. Huck, T. Mallat, A. Baiker, J. Catal. 219 (2003) 52.
}
}
[11] G. Szollosi, S. Niwa, T. Hanaoka, F. Mizukami, J. Mol. Catal. A: Chem. 23 (2005)
91.
}
}
[55] G. Szollosi, P. Forgó, M. Bartók, Chirality 15 (2003) S82.
[56] M. Garland, H.-U. Blaser, J. Am. Chem. Soc. 112 (1990) 7048.
}
}
[12] G. Szollosi, T. Hanaoka, S. Niwa, F. Mizukami, M. Bartók, J. Catal. 231 (2005)
480.