Three consecutive steps over the chirally modified Pt surface: Asymmetric catalytic cascade reaction of 2-nitrophenylpyruvates

Article abstract of DOI:10.1039/c4cy00883a

The influence of the reaction conditions on the asymmetric heterogeneous cascade reaction of 2-nitrophenylpyruvates over Pt catalysts modified with cinchonidine leading to (R)-3-hydroxy-3,4-dihydroquinolin-2(1H)-one derivatives has been studied. Results of studies on the amount of acetic acid or catalyst, nature of the Pt support, kinetic examinations, effect of H₂ pressure, and modifier and substrate concentrations showed that all three steps of this catalytic cascade take place on the Pt surface, with the nitro group reduction immediately following the enantioselective hydrogenation of the keto group, whereas the final intramolecular amidation was preceded by desorption after complete reduction of the substrate and re-adsorption of the corresponding intermediate.

Full text of DOI:10.1039/c4cy00883a

Catalysis
Science &
Technology
View Article Online
View Journal
PAPER
Three consecutive steps over the chirally modified
Pt surface: asymmetric catalytic cascade reaction
of 2-nitrophenylpyruvates
Cite this: DOI: 10.1039/c4cy00883a
a
György Szőllősi, Lenke Kovácsb and Zsolt Makrac
*
The influence of the reaction conditions on the asymmetric heterogeneous cascade reaction of
2-nitrophenylpyruvates over Pt catalysts modified with cinchonidine leading to IJR)-3-hydroxy-3,4-
dihydroquinolin-2IJ1H)-one derivatives has been studied. Results of studies on the amount of acetic
acid or catalyst, nature of the Pt support, kinetic examinations, effect of H2 pressure, and modifier
and substrate concentrations showed that all three steps of this catalytic cascade take place on the
Pt surface, with the nitro group reduction immediately following the enantioselective hydrogenation of
the keto group, whereas the final intramolecular amidation was preceded by desorption after complete
reduction of the substrate and re-adsorption of the corresponding intermediate.
Received 7th July 2014,
Accepted 8th September 2014
DOI: 10.1039/c4cy00883a
www.rsc.org/catalysis
Recently, we have developed a novel asymmetric heteroge-
neous catalytic cascade reaction for the enantioselective
1. Introduction
Optically pure partially saturated quinoline derivatives are
intermediates in the preparation of natural products and
pharmaceuticals.1,2 For the asymmetric synthesis of these
valuable chiral building blocks several catalytic methods were
designed.3 Besides enantioselective catalytic hydrogenations
of quinoline derivatives using chiral metal complexes,4
recently, asymmetric organocatalytic or metal catalysed cas-
cade reactions have been applied for the construction of the
hydroquinoline moiety from readily available multifunctional
compounds.5–7 Due to recent requirements of using sustain-
able, atom-economical and environmentally benign processes
in the preparation of fine chemicals, special attention is paid
to the application of heterogeneous asymmetric catalysts as
easily separable and recyclable materials.8 Modifications
of supported metal catalysts with optically pure compounds
are among the simplest procedures for obtaining chiral
heterogeneous catalysts.9 However, chirally modified catalysts
were efficient mostly in enantioselective hydrogenations of
certain prochiral compounds and were found to have limited
applicability.10–13 Moreover, asymmetric catalytic cascade
reactions over heterogeneous chiral catalysts were scarcely
reported,14,15 and only few reactions are known in which the
stereoselective step occurs on the solid catalyst surface.16,17
preparation of 3-hydroxy-3,4-dihydroquinolin-2IJ1H)-one deriv-
atives from 2-nitrophenylpyruvates over Pt catalyst modified
with cinchona alkaloids.18 The formation of hydroquinolone
derivatives over PtO2 catalyst was previously reported,19 and
the presence of the cinchona alkaloid, besides inducing
enantiodifferentiation, also increased the yield of the product.
The latter was suggested to be due to acceleration of the
enantioselective hydrogenation by the modifier, a phenome-
non called ligand acceleration,20,21 and also due to decelera-
tion of the nitro group reduction. Close to exclusive formation
of quinolones was observed from derivatives substituted next
to the nitro group due to the decreased reduction rate of the
nitro group by substituents (Scheme 1).
We found that tuning of the reaction conditions has a
major influence on the product composition, as the enantio-
selective hydrogenation of the keto group and the reduction
of the nitro group are competing reactions, both catalysed on
the Pt surface. It was also suggested that the final cyclization
may also take place on the metal surface.18 In the present
study, our aims were to examine the effect of several reaction
parameters on the product distribution, to obtain informa-
tion on the sequence leading to the desired product, and ulti-
mately to develop other asymmetric heterogeneous catalytic
cascade processes.
a MTA-SZTE Stereochemistry Research Group, University of Szeged, Department of
Organic Chemistry, H-6720, Szeged, Dóm tér 8, Hungary.
E-mail: szollosi@chem.u-szeged.hu; Fax: +36 62 544200
b University of Szeged, Institute of Pharmaceutical Chemistry, H-6720, Szeged,
Eötvös utca 6, Hungary
2. Experimental
2.1 Materials
c University of Szeged, Department of Organic Chemistry, H-6720, Szeged,
Dóm tér 8, Hungary
Commercial 5% Pt/Al2O3 catalyst (Engelhard, 4759, Pt dispersion
0.2–0.3 by TEM, support: γ-Al2O3 with acidity close to
This journal is © The Royal Society of Chemistry 2014
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
Scheme 2 Products formed in the asymmetric catalytic cascade
reaction of 1 over Pt/Al2O3 modified with CD.
Scheme 1 Asymmetric catalytic cascade reaction over Pt catalyst
modified by cinchona alkaloids (ee: enantiomeric excess).
the catalyst was filtered and washed with solvent, and the
organic phase was analyzed.
Identification of the products was performed by GC-MSD
(Agilent Techn. 6890N GC with 5973 inert MSD) using an
HP-1MS 60 m column. Products identified in reactions of 1
are shown in Scheme 2. Quantitative analysis was performed
using GC-FID (Agilent Techn. 6890N GC with FID) equipped
with a chiral capillary column (Cyclodex-B, 30 m) and tetra-
decane as the internal standard. These results were used to
calculate the conversion (X), product selectivities IJSIJPi)) and
the enantioselectivities, expressed as enantiomeric excess (ee),
according to the following formulae:
pure flame-made Al2O3)22–24 was used following reduction in
a fixed-bed flow reactor at 673 K in 30 mL min−1 H2 flow, as
described earlier.21 Other catalysts used were either commer-
cial products (5% Pt/C, 80982; 10% Pt/C, 80980; PtO2 with
≥60 m2 g−1, 206032 all supplied by Sigma-Aldrich) or were
prepared using known methods and commercial supports
(montmorillonite K10 having 250 m2 g−1 surface area and
pH 3–4,25 Nafion® SAC-13 fluorosulfonic acid with Nafion®
polymer on amorphous silica, 10–20%, >200 m2 g−1 surface
area, >10 nm pore diameter,26 both from Sigma-Aldrich;
Cab-O-Sil® M-5 SiO2 from Cabot Corp. with 256 m2 g−1 surface
area and 151 μmol g−1 total acidity27). The characteristics
of some of these catalysts were previously published, i.e.
0.14 metal dispersion of 3% Pt/SiO2, and the Pt particles
deposited on montmorillonite K10 (5% Pt/K10) have a
3.8 nm mean diameter.28,29 Cinchonidine (CD, Alfa-Aesar,
99%) was used as received. Solvents of analytical grade and
H2 gas (99.999%, Linde AG) were used as received. Ethyl IJ3-
methyl-2-nitrophenyl)pyruvate (1), ethyl IJ5-methyl-2-nitro
phenyl)pyruvate (6) and ethyl IJ6-methyl-2-nitrophenyl)pyruvate
(7) were prepared by condensation of 2,6-dimethyl-
nitrobenzene, 2,4-dimethylnitrobenzene and 2,3-dimethyl-
nitrobenzene (Sigma-Aldrich), respectively, with diethyl
oxalate (Sigma-Aldrich) using potassium tert-butoxide as
earlier reported.18 Ethyl 2-nitrophenylpyruvate (8) was
prepared from 2-nitrophenylpyruvic acid (Sigma-Aldrich)
as described in our previous report.18
X (%) = 100 × (1 − ct(1)/c0(1));
S(Pi) (%) = 100 × c(Pi)/(c0(1) − ct(1));
ee (%) = 100 × |c(R-2) − c(S-2)|/c(R-2) + c(S-2);
where c0IJ1) and ctIJ1) are the initial concentration and the
concentration at time t of 1, respectively; cIJPi) is the concen-
tration of product i; and cIJR-2) and cIJS-2) are the concentra-
tions of the R and S enantiomers of product 2, respectively.
Unless otherwise given, the reactions led to the complete
transformation of 2-nitrophenylpyruvates. In reactions using
CD as the modifier, the R enantiomer is formed in excess, as
shown in our previous report.18 The reproducibility of the
results was within 2% as demonstrated by reactions repeated
at least three times. The main product, 3-hydroxy-8-methyl-
3,4-dihydroquinolin-2IJ1H)-one, was isolated by flash chroma-
tography using a petroleum ether/ethyl acetate 4/1 eluent and
was characterized by 1H and 13C NMR spectroscopy.18
2.2 Asymmetric catalytic cascade reaction: typical procedure
Reactions were performed in a 45 cm3 glass tube introduced
in a stainless steel autoclave. Given amounts of catalyst,
solvent, modifier and substrate were introduced into the tube
and placed in the autoclave which was flushed with H2 and
filled to the desired H2 pressure, and the slurry was stirred
magnetically at 1000 rpm at room temperature (25 °C).
Following the given reaction time the H2 was released,
3. Results and discussion
In the present study we investigated the transformation
of 3-methyl-substituted ethyl 2-nitrophenylpyruvate (1),
which provided higher hydroquinolone yield than the com-
pound lacking a methyl substituent on the aromatic moiety
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2014

View Article Online
Catalysis Science & Technology
Paper
(see Scheme 1).18 The product mixture resulted from the reac-
tion of 1 over Pt/Al2O3 modified by CD contained the four
compounds shown in Scheme 2, i.e. the corresponding indole
(4) and oxindole (5) derivatives formed by complete or partial
reduction of the nitro group and condensation with the keto
group, as occurs in the Reissert indole synthesis. These com-
pounds are formed due to easy reduction of aromatic nitro
compounds to hydroxylamines and anilines over metal cata-
lysts such as Pd, Pt or Ni.30,31 However, the main products in
our reactions were the corresponding hydrogenated amino-
alcohol derivative (3) and the hydroquinolone (2). Our initial
studies showed that the choice of solvent has a major influ-
ence on the product distribution; the highest 2 selectivities
were obtained in toluene with addition of small amounts of
acetic acid (AcOH).18 Similarly, these two solvents (toluene
and AcOH) and mixtures thereof are the best performing
and most often used in the enantioselective hydrogenation
of α-keto esters over Pt catalysts modified with cinchona
alkaloids.9,21,23,32 Thus, we decided to investigate the amount
of AcOH necessary to obtain the best results in this asymmetric
cascade reaction.
accelerating effect on the reduction of the nitro group, as
indicated by the selectivity for 4. On the basis of these
results, 2 vol% AcOH was sufficient to obtain the best 2
selectivity.
Thus, in our next study designed to investigate the role of
the catalyst in the individual steps of the cascade reaction we
have used the above solvent mixture. Initially, we focused on
examining the effect of the catalyst amount while keeping
the CD/Pt ratio constant at 2.89 mmol of CD per millimole of
surface Pt (0.27 Pt dispersion22) (Fig. 2).
The main products over small amounts of catalyst
(10 or 30 mg) were also 3 and 2 with both reducible groups
(CO and NO2) completely hydrogenated, even when the
conversion of 1 was not complete. The indole 4 selectivity
remained low over the whole range of the catalyst amount.
The ee values of the two chiral compounds 2 and 3 were within
the limit of the determination error (not shown). A significant
decrease in the ee was detected by decreasing the catalyst
amount below 50 mg Pt/Al2O3, i.e. when 1 was not completely
transformed. This decrease was probably due to decreased
CD concentration, which may diminish the ee even when the
CD/surface Pt ratio is kept constant due to dilution effect;
thus, a higher ratio of surface sites is available for the race-
mic hydrogenation of the strongly adsorbing 1.33,34 Although
the effect of diffusion on the two reduction steps is difficult
to examine due to fast and possible competitive occurrence
of these steps, the constant ee over larger amounts of catalyst
(50–100 mg) is indicative of enantioselective hydrogenation
without hydrogen diffusion limitation; otherwise, a decrease
in ee is to be expected.35 We also mention that once formed,
the chiral centre is not further affected in the successive
transformations; thus, the obtained ee values cannot provide
further information on the kinetics of the following steps.
However, an increase in the amount of 2 at the expense of 3
over a larger catalyst amount showed that the intramolecular
amidation took place on the catalyst surface, although also
accelerated to some extent by using a small amount of AcOH
(see Fig. 1).
3.1 Influence of the AcOH and catalyst amounts
Results obtained using different amounts of AcOH are plotted
in Fig. 1. Under the experimental conditions, complete conver-
sion of 1 was always obtained. It was observed on one hand
that the relatively high amount of 3 obtained in toluene
decreased by addition of small amounts of AcOH concomi-
tant with the increase in 2 selectivity. On the other hand, the
indole (4) selectivity increased by increasing the amount
of AcOH, which ultimately reached 38% in pure AcOH
(not shown in Fig. 1). A small increase in ee was obtained
by increasing the amount of AcOH up to 5 vol%; however,
the ee values did not change significantly when the AcOH
content of the solvent was further increased.
Accordingly, the presence of the acid promotes cyclization,
shown by the obtained amount of 3, and also has an
Fig. 1 Effect of the AcOH amount on the product selectivities and
enantioselectivity for 2. Reaction conditions: 100 mg of Pt/Al2O3, 5 mL
Fig. 2 Effect of the Pt/Al2O3 catalyst amount on the reaction mixture
composition and enantioselectivity for 2. Reaction conditions: 5 mL of
toluene with 2 vol% AcOH, 2 × 10−4 mmol of CD per gram of catalyst,
of solvent, cIJCD) 4 mM, c0IJ1) 80 mM, pH 4 MPa, 25 °C, 3 h, full
2
conversions of 1.
c0IJ1) 80 mM, pH 4 MPa, 25 °C, 3 h.
2
This journal is © The Royal Society of Chemistry 2014
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
Table 1 Influence of catalyst support on the reaction of 1a
3.2 Kinetics of the cascade reaction
Catalystb (mg)
S (3) (%)
S (2) (%)
ee (2) (%)
Results of kinetic investigations of the cascade reaction over
50 mg catalyst are illustrated in Fig. 3. Similar trends were
also obtained over 100 mg of catalyst (not shown). Complete
conversion of 1 was obtained (except after 0.5 h) and the
main products in all of these experiments were 3 and 2. The
former was in excess following shorter reactions and was
transformed to hydroquinolone 2 as the reaction time was
extended. Obviously, under such conditions, the rates of the
first two competitive steps could not be examined, and only
information on the final step was gathered. The product com-
positions clearly showed that the amino-alcohol 3 desorbs
from the surface following the reduction steps. The forma-
tion of 2 by cyclization of 3 was also indicated by the identi-
cal ee values of 2 and 3 (not shown in Fig. 3). Desorption was
probably assisted by the presence of 1 in the slurry, the latter
having higher adsorption strength than 3, as shown by the
adsorption strengths of nitro compounds in comparison with
amines and the adsorption mode of ketones and alcohols.36,37
Thus, 3 is a stable intermediate through which 2 is formed
and which is transformed following the complete consump-
tion of 1.
5% Pt/Al2O3 (50)
5% Pt/Al2O3 (100)
5% Pt/C (100)
10% Pt/C (50)
3% Pt/SiO2 (100)
PtO2 (50)
5% Pt/K 10 (100)
5% Pt/SAC-13 (50)
5% Pt/Al2O3 (50) + SAC-13 (50)
20
6
77
90
71
77
68
88
3
71
76
36
36
36
24
10
23
59
<1
<1
29
<1
65
95
57
3
42
a
Reaction conditions: a given amount of catalyst, 5 mL of toluene
with 2 vol% AcOH, 2 × 10−4 mmol of CD per gram of catalyst, c0IJ1)
b
80 mM, pH 4 MPa, 25 °C, 3 h, complete conversions of 1. For the
2
origin and properties of the catalysts see subsection 2.1.
whereas the worst was with Pt supported over solid acids.
Although the enantioselectivity is strongly influenced by the
particle size and the morphology of Pt,9,10 the decrease in ee
observed after addition of Nafion® SAC-13 to the slurry using
Pt/Al2O3 indicated that the acidic material probably bonded
a significant part of the modifier on its surface, impeding
the adsorption on the metal. Thus, a similar ee value was
obtained in the presence of a lower amount of the modifier.18
Surprisingly, over catalysts using acidic supports, the selectiv-
ity for the quinolone 2 was very low, whereas good selectiv-
ities were obtained over carbon- or silica-supported catalysts
or even over in situ reduced PtO2. Accordingly, the support
was not necessary to obtain the cyclic compound 2. The high
3 selectivities observed after addition of acidic material to the
Pt/Al2O3 containing slurry indicate an unfavorable effect of
such materials on the cyclization. Thus, one may exclude the
cyclization of 3 to 2 over the acidic support and consequently
over Al2O3, and it could be assumed that this final step also
proceeds on the metal surface. The intramolecular amidation
of esters as a subsequent step of reduction over a metal cata-
lyst is a known lactam preparation procedure even in neutral
solvents.19,30,39 The significant effect of the support charac-
teristics may be attributed to the electronic charge transfer
that occurred between metal and support, as was suggested
to be the reason for the selectivities obtained in the competi-
tive hydrogenation of different functional groups.40
3.3 Effect of the catalyst support
According to the above results the final step of the cascade
reaction, i.e. the amidation, occurs on the catalyst surface
following the consumption of nitrophenylpyruvate 1. This
step may be catalyzed by acids. Thus, we examined the possi-
ble involvement of the catalyst support in the final reaction
step, which may also have an acidic character, similar to the
Pd catalysts supported over γ-Al2O3.38 In order to investigate
the role of the support, the reaction was performed using
Pt deposited on various supports, including a Lewis and
Brønsted acidic clay (montmorillonite K10) or Brønsted
acidic Nafion® SAC-13. Results obtained are summarized in
Table 1.
Large variations in enantioselectivities were obtained over
these catalysts; the best values were attained over Pt/Al2O3,
3.4 Influence of the H2 pressure
Confirmation of the suggestion that all three steps of the
cascade reaction proceed over the Pt particles covered by
chemisorbed hydrogen was obtained from studies on the
effect of H2 pressure on the product composition shown in
Fig. 4.
The best results were obtained under 1 MPa H2 pressure.
The increase in pressure over this value resulted in an
increase in the selectivity for the unreacted 3 and a decrease in
quinolone 2 selectivity. Similarly, under pressures over 1 MPa
the ee also slightly decreased by increasing the pressure,
as observed in enantioselective hydrogenations of activated
Fig. 3 Effect of the reaction time on the transformation of 1 over
Pt/Al2O3. Reaction conditions: 50 mg of catalyst, 5 mL of toluene
with 2 vol% AcOH, cIJCD) 2 mM, c0IJ1) 80 mM, pH 4 MPa, 25 °C.
2
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2014

View Article Online
Catalysis Science & Technology
Paper
Fig. 4 Effect of H2 pressure on the reaction of 1. Reaction conditions:
50 mg (closed symbols) or 100 mg (open symbols) of Pt/Al2O3, 5 mL of
toluene with 2 vol% AcOH, 2 × 10−4 mmol of CD per gram of catalyst,
c0IJ1) 80 mM, 25 °C, 3 h (over 50 mg of catalyst) or 2 h (over 100 mg of
catalyst) reactions, complete conversions of 1.
Fig.
5
Effect of cinchonidine concentration on the product
selectivities and enantioselectivity for 2. Reaction conditions: 50 mg of
Pt/Al2O3, 5 mL of toluene with 2 vol% AcOH, c0IJ1) 80 mM, pH 1 MPa,
25 °C, 3 h, complete conversions of 1.
2
ketones.10,41 The most plausible explanation for the decrease
in ee is the consumption of the modifier by hydrogenation
of the anchoring quinoline moiety, resulting in formation of
partially hydrogenated cinchonidine derivatives with lower
enantiodifferentiating ability.41 Considering that the selectivity
for 2 followed the same trend as that of ee as a function
of H2 pressure, one may speculate that the cyclization is also
unfavorably affected by the hydrogenation of CD. Possible
interaction of 3 with partially hydrogenated CD on the Pt sur-
face may decrease the cyclization rate and may lead to higher
concentration of the amino-alcohol product in the mixture
following identical reaction time. Although this assumption
should be verified in future experiments, if confirmed, these
results support the assumption of the third step taking place
on the Pt surface.
Observations also worth discussing are the selectivities
obtained under low, i.e. 0.1 MPa, H2 pressure. The high
amount of 3 remaining in the reaction mixture should be
due to low rates of the first two steps; thus, following the
given reaction time the third cyclization step of the cascade
was not complete. This possible explanation could also ratio-
nalize the low 2 selectivities obtained over home-made Pt/K
10 and Pt/SAC-13, which probably were less active in the
reduction steps than the commercial catalysts. Accordingly,
the third step over these materials was not complete after
identical reaction times as was used in experiments over
commercial catalysts.
previously published experiments. An increase in the modi-
fier coverage by increasing the CD concentration up to 2 mM
caused a shift of the reaction pathway from the partial forma-
tion of 4 to the exclusive formation of 2, whereas a further
increase in the cIJCD) resulted in a gradual decrease in 2
selectivity and the formation of increasing amounts of 3. The
ee value also increased up to 2 mM cIJCD) followed by con-
stant values upon further increasing the cIJCD). It should be
mentioned that in these experiments the ee of 3 (not shown
in the figure) was always identical or very close to the ee of 2.
On the basis of these observations one may reach the conclu-
sion that the presence of CD hindered the fast reduction of
the nitro group and the formation of 4. However, at a certain
CD concentration, i.e. coverage, the adsorbed modifier also
hindered the cyclization step. Although one may speculate
that this may occur due to alterations in the adsorption mode
of the cinchona alkaloids as a function of coverage,42–44 the
most plausible explanation is that the surface of the metal
should be only partially covered by the modifier to be able to
adsorb the intermediate product 3 in order to obtain 2. This
confirmed that the cyclization occurs on the metal surface.
The effect of the concentration of nitrophenylpyruvate 1
on the product selectivities and the ee of 2 is presented in
Fig. 6. No significant variations in ee were observed; thus, ee
was influenced by the catalyst/CD ratio (at low CD concentra-
tions, see Fig. 5) and was practically independent of the 1/CD
ratio (in the c0IJ1) range studied, see Fig. 6). An almost exclu-
sive formation of 2 was detected up to 80 mM c0IJ1); however,
at this concentration cyclization was not complete and the
intermediate 3 was detected in the product mixture. As the
reduction steps were complete in these experiments, even
when the highest c0IJ1) was used, selectivities were governed
by the cyclization rate of 3. Thus, under these conditions
the first two steps of the cascade, i.e. the enantioselective
hydrogenation and the reduction of the nitro group, occurred
without desorption of partially reduced intermediates even
at a ratio of 8 mmol of substrate per gram of catalyst. The
decrease in 2 selectivity at a certain c0IJ1) (80 mM) showed
3.5 Effect of the modifier and substrate concentration
The present study was also extended to investigate the effect
of CD and substrate concentration and performed under
1 MPa H2 pressure (based on the above). Although the
influence of the modifier concentration has already been
discussed,18 data obtained over lower catalyst amounts
presented in Fig. 5 better showed the obtained tendencies
due to amplification of the CD concentration effect by
increasing the modifier/catalyst ratio, as compared to our
This journal is © The Royal Society of Chemistry 2014
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
the functional groups, especially that of the –NO2 group
bonded to the phenyl ring. The influence of the position of
the methyl substituent has been evidenced by comparison of
results obtained under identical conditions in the reaction
of the three methyl derivatives (1, 6, 7) and the compound
lacking a methyl substituent on the phenyl ring (8). Results
obtained are summarized in Table 2. Variations in the amount
of indole derivative (In) formed during the reactions may be
considered as an indication of the effect of the substituent
on the reduction rate of the –NO2 group, considering that the
methyl substituent has less effect on the rate of the enantio-
selective hydrogenation of the –CO group. Accordingly,
the data showed that the 3-methyl substituent decelerated
the reduction of the –NO2 group, whereas the substituent
in position 5 has no influence when compared with the
unsubstituted compound 8. On the contrary, the methyl sub-
stituent in position 6 increased the selectivity for In. This
may be explained by tilting of the adsorbed substrate on
the metal due to steric repulsions between the methyl group
and the surface, thus allowing easier interaction of the
–NO2 group with the Pt and the chemisorbed hydrogen.
The selectivities for the hydroquinolone derivatives (Hq)
increased in the presence of the methyl substituent in
positions 5 or 3; in the case of the latter, an almost exclusive
formation of the target product was obtained. Only in the
reaction of the compound substituted in position 6 the
amount of Hq decreased. It must be noted that very similar
ee values were obtained with the exception of the reaction
of 6, which may be ascribed to possible hindering of the
CD–substrate interaction on the surface by the methyl group
in this position. One can also observe that under these condi-
tions only the reaction of 7 resulted in a significant amount
Fig.
6 Effect of substrate initial concentration on the product
selectivities and enantioselectivity for 2. Reaction conditions: 100 mg of
Pt/Al2O3, 5 mL of toluene with 2 vol% AcOH, cIJCD) 4 mM, pH 1 MPa,
25 °C, full conversions in 2 h.
2
that the cyclization rate was not influenced significantly by
c0IJ1) and longer reaction times are necessary for the complete
transformation of larger amounts of 3 to 2.
3.6 Influence of the position of the methyl substituent
The 3-methyl-substituted ethyl 2-nitrophenylpyruvate (1) was
chosen as the substrate for these examinations on the basis
of results obtained over unmodified Pt19 and our previous
study.18 The 3-methyl substituent was suggested to hinder to
some extent the reduction of the nitro group, which altered
the ratio of the –CO hydrogenation/–NO2 reduction rates.
However, the position of the substituent may alter the
adsorption of the substrate and thus affect the reactivity of
Table 2 Influence of the methyl substituent position on the asymmetric cascade reaction of 2-nitrophenylpyruvatesa
Substrate
S (Inb) (%)
S (Redb) (%)
S (Hqb) (%)
ee (Hq) (%)
20
1
55
80
(8)
2
20
45
<1
1
97
65
35
80
52
79
(1)
(6)
8
(7)
a
Reaction conditions: 100 mg of Pt/Al2O3, 5 mL of toluene with 2 vol% AcOH, cIJCD) 4 mM, c0IJ1) 80 mM, pH 1 MPa, 25 °C, 2 h,
2
full conversions. b Abbreviations: In, the corresponding indole; Red, the reduced intermediate; and Hq, the hydroquinolinone.
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2014

View Article Online
Catalysis Science & Technology
Paper
Fig. 7 Reaction pathway leading to the formation of chiral 3-hydroxy-3,4-dihydroquinolin-2IJ1H)-one derivatives.
of reduced uncyclized product (Red). Thus, the methyl group
only in this position hindered the re-adsorption of the amino-
alcohol intermediate.
shown that the use of a support was not necessary; thus, this
final step may occur on the metal surface. The latter con-
clusion was supported by the influence of H2 pressure and
modifier and substrate concentrations on the selectivities of
the products. The accumulation of the intermediate 3 in the
reaction mixture showed that the cyclization takes place on
the Pt surface following desorption and re-adsorption of this
intermediate.
On the basis of these conclusions, the steps of this asym-
metric heterogeneous catalytic cascade reaction may be illus-
trated, as shown in Fig. 7. Eventually, it was concluded that
all three steps of this unique cascade reaction, which leads to
optically enriched N-heterocyclic compounds, take place on
the Pt surface.
Conclusions
Results disclosed in this report on the effect of the reaction
conditions on the recently published asymmetric cascade
transformation of 2-nitrophenypyruvates over Pt catalyst
modified with cinchonidine18 allowed us to draw novel con-
clusions compared to our previous report, which could be
edifying for future design of asymmetric cascade reactions,
particularly those using chiral surfaces as heterogeneous
catalyst.
Although the first two steps of the reaction studied are
competitive, both may occur only on the metal surface and
their rates are comparable; modification of the Pt particles
by cinchona alkaloids besides imparting asymmetry to the
surface also increased the rate difference between these
two steps by decelerating the –NO2 group reduction and accel-
erating the enantioselective hydrogenation of the activated
keto group. This difference is even more accentuated when a
substituent adjacent to the –NO2 group further decelerates
the reduction of this group. Thus, in the reaction of 1 these
competitive steps became subsequent reactions over a chirally
modified Pt catalyst, which proceeded without desorption
of the partially reduced intermediates between the steps,
as indicated by the effect of the substrate concentration.
The kinetic study shown here also indicated that the amino
alcohol 3 is a stable intermediate through which 2 is formed
and which is transformed mostly following the complete
consumption of 1 from the reaction mixture. Results of exam-
ining the influence of the catalyst amount indicated that
the third and final step of the reaction, i.e. intramolecular
amidation, also proceeded on the catalyst. However, it was
Acknowledgements
Financial support from the Hungarian Scientific Research Fund
(OTKA grant K 109278) is highly appreciated. This research was
supported in the framework of TÁMOP 4.2.4. A/2-11-1-2012-0001
“National Excellence Program – Elaborating and operating
an inland student and researcher personal support system”
key project. The project was subsidized by the European Union
and co-financed by the European Social Fund.
Notes and references
1 Heterocycles in Natural Product Synthesis, ed. K. C. Majumdar
and S. K. Chattopadhyay, Wiley-VCH, Weinheim, 2011.
2 Chiral Drugs: Chemistry and Biological Action, ed. G.-Q. Lin,
Q.-D. You and J.-F. Cheng, John Wiley & Sons, Hoboken,
New Jersey, 2011.
3 V. Sridharan, P. A. Suryavanshi and J. C. Menéndez, Chem.
Rev., 2011, 111, 7157–7259.
This journal is © The Royal Society of Chemistry 2014
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
4 Y.-G. Zhou, Acc. Chem. Res., 2007, 40, 1357–1366.
5 H. Pellissier, Adv. Synth. Catal., 2012, 354, 237–294.
6 H. Clavier and H. Pellissier, Adv. Synth. Catal., 2012,
354, 3347–3403.
7 L.-Q. Lu, J.-R. Chen and W.-J. Xiao, Acc. Chem. Res., 2012,
45, 1278–1293.
25 T. Cseri, S. Békássy, F. Figueras, E. Cseke, L.-C. de Menorval
and R. Dutartre, Appl. Catal., A, 1995, 132, 141–155.
26 M. A. Harmer and Q. Sun, Appl. Catal., A, 2001, 221, 45–62.
27 J. A. Cecilia, I. Jiménez-Morales, A. Infantes-Molina,
E. Rodríguez-Castellón and A. Jiménez-López, J. Mol. Catal. A:
Chem., 2013, 368–369, 78–87.
8 Handbook of Asymmetric Heterogeneous Catalysis, ed. K. Ding
and Y. Uozumi, Wiley-VCH, Weinheim, 2008.
9 T. Mallat, E. Orglmeister and A. Baiker, Chem. Rev., 2007,
107, 4863–4890.
28 Gy. Szőllősi, Á. Mastalir, Á. Molnár and M. Bartók, React.
Kinet. Catal. Lett., 1996, 57, 29–36.
29 Gy. Szőllősi, B. Török, L. Baranyi and M. Bartók, J. Catal.,
1998, 179, 619–623.
10 D. Yu. Murzin, P. Mäki-Arvela, E. Toukoniitty and T. Salmi,
Catal. Rev.: Sci. Eng., 2005, 47, 175–256.
11 M. Bartók, Curr. Org. Chem., 2006, 10, 1533–1567.
12 H.-U. Blaser and M. Studer, Acc. Chem. Res., 2007, 40,
1348–1356.
13 Gy. Szőllősi, B. Hermán, K. Felföldi, F. Fülöp and M. Bartók,
Adv. Synth. Catal., 2008, 350, 2804–2814.
14 J. Muzart, F. Hénin and S. J. Aboulhoda, Tetrahedron: Asymmetry,
1997, 8, 381–389.
15 L. Deiana, S. Afewerki, C. Palo-Nieto, O. Verho, E. V. Johnston
and A. Córdova, Sci. Rep., 2012, 2, 851.
16 K. Felföldi, K. Szőri and M. Bartók, Appl. Catal., A, 2003,
251, 457–460.
17 Gy. Szőllősi and M. Bartók, ARKIVOC, 2012, (v), 16–27.
18 Gy. Szőllősi, Zs. Makra, L. Kovács, F. Fülöp and M. Bartók,
Adv. Synth. Catal., 2013, 355, 1623–1629.
30 S. Nishimura, Handbook of Heterogeneous Catalytic Hydrogenation
for Organic Synthesis, John Wiley & Sons, Hoboken, New Jersey,
2011, ch. 9, pp. 315–389.
31 H. U. Blaser, U. Siegrist, H. Steiner and M. Studer, Fine
Chemicals through Heterogeneous Catalysis, ed. R. A. Sheldon
and H. van Bekkum, Wiley-VCH, Weinheim, 2001, pp. 389–406.
32 H. U. Blaser, H. P. Jalett and J. Wiehl, J. Mol. Catal., 1991,
68, 215–222.
33 K. Balázsik, K. Szőri, Gy. Szőllősi and M. Bartók, Chem. Commun.,
2011, 47, 1551–1552.
34 K. Balázsik, K. Szőri, Gy. Szőllősi and M. Bartók, Catal.
Commun., 2011, 12, 1410–1414.
35 U. K. Singh, R. N. Landau, Y. Sun, C. LeBlond, D. G. Blackmond,
S. K. Tanielyan and R. L. Augustine, J. Catal., 1995, 154, 91–97.
36 Q. Chen, S. Haq, B. G. Frederick and N. V. Richardson,
Surf. Sci., 1996, 368, 310–317.
19 H. Suzuki, H. Gyoutoku, H. Yokoo, M. Shinba, Y. Sato,
H. Yamada and Y. Murakami, Synlett, 2000, 1196–1198.
20 M. Garland and H.-U. Blaser, J. Am. Chem. Soc., 1990,
112, 7048–7050.
21 Gy. Szőllősi, Sz. Cserényi, F. Fülöp and M. Bartók, J. Catal.,
2008, 260, 245–253.
37 A. Vargas, S. Reimann, S. Diezi, T. Mallat and A. Baiker,
J. Mol. Catal. A: Chem., 2008, 282, 1–8.
38 M. Casagrande, S. Franceschini, M. Lenarda, O. Piccolo and
A. Vaccari, J. Mol. Catal. A: Chem., 2006, 246, 263–267.
39 E. R. Blout and D. C. Silverman, J. Am. Chem. Soc., 1944,
66, 1442–1443.
22 M. von Arx, T. Mallat and A. Baiker, J. Catal., 2000,
193, 161–164.
40 S. Santiago-Pedro, V. Tamayo-Galván and T. Viveros-García,
Catal. Today, 2013, 213, 101–108.
23 C. Exner, A. Pfaltz, S. Studer and H.-U. Blaser, Adv. Synth. Catal.,
2003, 345, 1253–1260.
24 B. Schimmoeller, F. Hoxha, T. Mallat, F. Krumeich,
S. E. Pratsinis and A. Baiker, Appl. Catal., A, 2010,
374, 48–57.
41 Gy. Szőllősi, P. Forgó and M. Bartók, Chirality, 2003, 15, S82–S89.
42 D. Ferri and T. Bürgi, J. Am. Chem. Soc., 2001, 123, 12074–12084.
43 D. Ferri, T. Bürgi and A. Baiker, J. Catal., 2002, 210, 160–170.
44 M. Bartók, M. Sutyinszki, K. Balázsik and G. Szőllősi, Catal.
Lett., 2005, 100, 161–167.
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2014