Hydrogenation of 1-phenyl-1,2-propanedione over Pt catalysts modified by cinchona alkaloid O-ethers and the kinetic resolution of the 1-hydroxyketones generated

Article abstract of DOI:10.1016/j.jcat.2008.01.012

Nine cinchona alkaloid O-ethers together with cinchonidine and cinchonine were studied as chiral modifiers in the enantioselective hydrogenation of 1-phenyl-1,2-propanedione over Pt/Al₂O₃. The influence of the O-substituent on the reaction rate, selectivity and product distribution was investigated. Apparent rate constants for all hydrogenation steps were calculated using a first-order kinetic approach resulting in a good agreement between the experimentally recorded and predicted concentrations. The experimentally observed structure-selectivity effects indicate that the mechanisms of enantiodifferentiation over the catalyst modified by parent cinchona alkaloids and their ether derivatives differ from each other. Moreover, the modifier structure-selectivity dependence and the solvent effect were different for enantio- and diastereoselection in the 1-phenyl-1,2-propanedione and 1-hydroxyketone hydrogenations. Correlation between the modifier substituent bulkiness and diastereoselectivity of the 1-hydroxyketone hydrogenation was observed. Data on the inversion of enantioselectivity of 1-phenyl-1,2-propanedione hydrogenation, diastereoselectivity and the sense of kinetic resolution of the 1-hydroxyketones were presented. Due to the complexity of the reaction network, several competing mechanistic pathways may be present in a single reaction system.

Full text of DOI:10.1016/j.jcat.2008.01.012

Journal of Catalysis 254 (2008) 339–348
www.elsevier.com/locate/jcat
Hydrogenation of 1-phenyl-1,2-propanedione over Pt catalysts modified
by cinchona alkaloid O-ethers and the kinetic resolution
of the 1-hydroxyketones generated
Igor Busygin a, Johan Wärnå b, Esa Toukoniitty b, Dmitry Yu. Murzin b,∗, Reko Leino a,∗
a
Laboratory of Organic Chemistry, Åbo Akademi University, 20500 Åbo, Finland
b
Laboratory of Industrial Chemistry, Åbo Akademi University, 20500 Åbo, Finland
Received 18 December 2007; revised 10 January 2008; accepted 12 January 2008
Available online 28 January 2008
Abstract
Nine cinchona alkaloid O-ethers together with cinchonidine and cinchonine were studied as chiral modifiers in the enantioselective hydrogena-
tion of 1-phenyl-1,2-propanedione over Pt/Al O . The influence of the O-substituent on the reaction rate, selectivity and product distribution
2
3
was investigated. Apparent rate constants for all hydrogenation steps were calculated using a first-order kinetic approach resulting in a good
agreement between the experimentally recorded and predicted concentrations. The experimentally observed structure–selectivity effects indicate
that the mechanisms of enantiodifferentiation over the catalyst modified by parent cinchona alkaloids and their ether derivatives differ from each
other. Moreover, the modifier structure–selectivity dependence and the solvent effect were different for enantio- and diastereoselection in the
1-phenyl-1,2-propanedione and 1-hydroxyketone hydrogenations. Correlation between the modifier substituent bulkiness and diastereoselectivity
of the 1-hydroxyketone hydrogenation was observed. Data on the inversion of enantioselectivity of 1-phenyl-1,2-propanedione hydrogenation,
diastereoselectivity and the sense of kinetic resolution of the 1-hydroxyketones were presented. Due to the complexity of the reaction network,
several competing mechanistic pathways may be present in a single reaction system.
© 2008 Elsevier Inc. All rights reserved.
Keywords: Cinchona alkaloids; Heterogeneous catalysis; Hydrogenation kinetics; 1-Phenyl-1,2-propanedione; Inversion of enantioselectivity; Inversion of
diastereoselectivity
1. Introduction
report on the enantioselective hydrogenation of methyl pyru-
vate over a heterogeneous catalyst by Orito and co-workers in
1978 [3], applications of the Pt/cinchona alkaloid system have
been successfully extended to a number of substrates. Several
recent reviews are available on this subject [4–10].
Three functional parts of the cinchona alkaloids can be dis-
tinguished, namely, the bicyclic quinuclidine part, the aromatic
quinoline ring, and the stereogenic region which couples the
two rigid moieties together by two carbon–carbon bonds (Ta-
ble 1). It is well documented that cinchonidine (CD) and cin-
chonine (CN) induce an excess of opposite product chirality in
hydrogenations. Recently, O-ether derivatives of cinchona al-
kaloids have attracted considerable attention as modifiers for
enantioselective hydrogenations of activated ketones [11–17].
This interest originates from the capacity of changing the sense
of product chirality by derivatization of the hydroxyl group in
Chiral induction over heterogeneous metal catalysts is one of
the most rapidly progressing methods for generating asymmet-
ric centers in organic molecules [1,2]. Enantioface discriminat-
ing hydrogenation of a prochiral C=X functionality produces
one enantiomer of the molecule in excess over the other. Cin-
chona alkaloids are efficient chiral surface modifiers for enan-
tioselective hydrogenation of various activated ketones over
heterogeneous catalysts. A micromolar quantity of the chiral
modifier in the reaction milieu allows producing one product
enantiomer in high enantiomeric excess (ee). Since the first
*
Corresponding authors. Faxes: +358 2 2154479, +358 2 2154866.
E-mail addresses: dmitry.murzin@abo.fi (D.Yu. Murzin),
reko.leino@abo.fi (R. Leino).
0021-9517/$ – see front matter © 2008 Elsevier Inc. All rights reserved.
doi:10.1016/j.jcat.2008.01.012

340
I. Busygin et al. / Journal of Catalysis 254 (2008) 339–348
Table 1
Structures of the modifiers
the kinetics of butane-2,3-dione [27,28] and cyclohexane-1,2-
dione hydrogenation [29], at least six rate constants should
be accounted for. The hydrogenation of unsymmetrical dike-
tones, such as 1-phenyl-1,2-propanedione (PPD), is even more
complex and consist of nine steps (Scheme 1). In our labora-
tories, kinetics of the enantioselective hydrogenation of PPD
(Scheme 1), have been studied for some years [14,15,30–32].
Some characteristic features have been revealed: (1) The pres-
ence of two inequivalent keto groups in the molecule raises the
issue of regioselectivity; Hydrogenation of the carbonyl adja-
cent to the phenyl ring is always preferred over hydrogenation
of the aliphatic one [33]; (2) Under the currently optimal condi-
tions, the main product, (R)-1-hydroxy-1-phenyl-2-propanone
(depicted as (1R) in Scheme 1), can be obtained in 65% ee
using cinchonidine as the chiral catalyst modifier; (3) When
O-methyl or O-silyl ethers of cinchona alkaloids are used as
chiral modifiers, a loss or inversion of enantioselectivity re-
sults [14,15,34]. This observation is indicative of a significant
difference in the enantioselection mechanisms operating in the
diketone and the classical keto ester hydrogenations, as in the
latter case, the O-methyl derivatives of CD and CN induce, in
toluene [13], enantioselectivities similar to those obtained with
the parent alkaloids, and in acetic acid even higher selectivities
than the parent modifiers [22].
R
CD
MeOCD
PhOCD
TMSOCD
TBDMSOCD
ADMSOCD
DPMSOCD
–H
–CH
–C H
–Si(CH )
–Si(CH ) t-Bu
–Si(CH ) A
–Si(C H ) CH
3
CN
MeOCN
PhOCN
TMSOCN
–
–
–
3
6
5
3 3
3 2
a
3 2
6
5 2
a
A = –CH C(H)=CH .
2
2
the alkaloid while retaining the same the absolute configura-
tions of the alkaloid asymmetric centers. Isocinchonas, which
are cyclic O-ether derivatives of cinchona alkaloids, have been
studied actively as well [18–20]. Studies on the structure–
selectivity relationships provide valuable information for un-
derstanding the experimentally feasible reaction mechanisms
and potentially broaden the application area of heterogeneous
catalysts in enantioselective hydrogenation [21–23].
Study of the reaction kinetics is one of the approaches for
elucidating the reaction mechanisms and interpreting the role
of the chiral modifier in enantiodifferentiation. Several reports
on studies of kinetics of ethyl pyruvate hydrogenation [24–26]
are available. Kinetic analysis of the complex reaction sys-
tems consisting of parallel and consecutive steps is represented
in the literature only by a few examples. When considering
Baiker and co-workers [35] reported the phenomenon of
enantioselectivity inversion in the hydrogenation of 4,4,4-tri-
fluoroacetoacetate which is related to the hydrate formation
from the ketone in THF. This opened a new approach to re-
search on the nature of modifier–substrate–metal surface in-
teractions in the Orito reaction [4–10]. Since then several in-
vestigation have been published reporting inversion of enan-
tioselectivity due to changes in the modifier structure, change
Scheme 1. 1-Phenyl-1,2-propanedione hydrogenation reaction scheme.

I. Busygin et al. / Journal of Catalysis 254 (2008) 339–348
341
of a solvent [36–40] and even changes of the modifier con-
2.3. Kinetic model
centration have resulted in inversion of chirality of the main
product [41–43]. Studying the kinetics of complex asymmet-
ric reactions raises the issues of enantioselectivity and regios-
electivity for the initial steps; kinetic resolution and diastere-
oselectivity for the consecutive steps. In the present study, we
observed the inversion of enantioselectivity of PPD hydrogena-
tion, diastereoselectivity and the sense of kinetic resolution of
the 1-hydroxyketones by O-derivatization of cinchona alkaloid
modifiers. Nine derivatives of cinchonidine and cinchonine (Ta-
ble 1), possessing substituents with different nature and bulk-
iness, were examined as chiral modifiers in the enantioselec-
tive hydrogenation. Reaction kinetics of the PPD hydrogenation
was analyzed by first-order kinetic approach, considering for
the first time the complete reaction network.
The reaction kinetics can be described as follows. The dis-
appearance of the reactants and the generation of the products
can be calculated by the solution of the molar balances for the
components in a batch reactor:
V
d[PPD]
rPPD
r(1R)
r(1S)
r(2R)
r(2S)
=
=
=
=
=
= −(k1 + k2 + k3 + k4)[PPD],
= k1[PPD] − (k5 + k6)[1R],
= k1[PPD] − (k7 + k8)[1S],
= k1[PPD] − (k9 + k10)[2R],
= k1[PPD] − (k11 + k12)[2S],
mcat dt
V
d[1R]
mcat dt
V
d[1S]
mcat dt
V
d[2R]
mcat dt
V
d[2S]
mcat dt
2. Experimental
V
d[1R, 2S]
r(1R,2S)
r(1R,2R)
r(1S,2R)
r(1S,2S)
=
=
=
=
= k5[1R] + k11[2S],
mcat
V
dt
2.1. Materials
d[1R, 2R]
= k6[1R] + k10[2R],
= k7[1S] + k9[2R],
= k8[1S] + k12[2S].
mcat
V
dt
For use in the catalytic hydrogenation reactions, 1-phenyl-
1,2-propanedione (Aldrich, 99%) was vacuum distilled, where-
as toluene (J.T. Baker, >99.5%), acetic acid (J.T. Baker,
99.9%), cinchonidine (Fluka, 98%) and cinchonine (Fluka,
98%) were used as received.
d[1S, 2R]
mcat
V
dt
d[1S, 2S]
mcat
dt
The system of differential equations was solved numerically
with the backward difference method using the Odessa solver
in the ModEst program package [45]. The differential equation
solver operated under a parameter estimation routine, which
2.2. Catalytic hydrogenation
Hydrogenations were carried out using a pressurized batch
reactor (Parr, 300 cm3) equipped with constant flow–constant
pressure equipment. According to the standard procedure,
the 5 wt% Pt/Al2O3 catalyst (Strem Chemicals, 78–1660)
(100 mg) was pre-reduced in situ by flushing with hydrogen
(50 cm3 min−1) for 2 h at 400 ◦C. After cooling to room tem-
perature in hydrogen flow, a solution of the chiral modifier
(3.4 × 10−5 mol in 50 mL of toluene or acetic acid), prelim-
inary saturated with hydrogen (50 cm3 min−1) for 10 min in a
separate injection chamber, was injected into the reactor. The
reaction was commenced by starting the agitation immediately
after the injection of the solution of 1-phenyl-1,2-propanedione
(0.74 g, 5 mmol) pre-treated in the same manner. The reac-
tions were carried out under kinetic regime in the absence
of external and internal mass transfer limitations [31]. The
constant hydrogen (AGA, 99.999%) pressure of 10 bar, tem-
perature of 15 or 17 ◦C (for the hydrogenations in toluene
and acetic acid, respectively) and stirring rate of 2000 rpm
was applied. The initial concentrations of chiral modifi−e3rs and
minimized an objective function, the residual sum of squares
ꢀ
Q =
w(cexp − cmodel)2,
where cexp and cmodel denote the experimental and predicted
concentrations, and w is a weighting factor. A hybrid Simplex–
Levenberg–Marquardt algorithm was used in the minimization
of the objective function Q. The numerical algorithm is in-
cluded in the program package ModEst [45]. The quality of the
model fit and the model parameters were tested by calculating
the standard deviations of the parameters and the degree of ex-
planation of the model. The values of estimated rate constants
(ki) are lumped constants, which might include the reactant
and modifier adsorption coefficients as well as concentrations
of modifier and hydrogen. The first-order kinetic approach es-
sentially assumes nonsignificant coverage of adsorbed reactant,
which can be justified by utilization of low substrate concen-
tration, as well as experimentally observed kinetic regularities
(Fig. 1), following closely first-order behavior.
1-phenyl-1,2-propanedione were 3.4 × 10−4 moldm
and
2.4. Definition of selectivities
0.05 mol dm−3, respectively. The reaction was followed by a
Varian 3300 Gas Chromatograph (GC) equipped with a chiral
column (β-Dex 225). Details of the catalyst characterization
and analytical procedure and modifier synthesis can be found
elsewhere [30–32,44]. The reproducibility of catalytic results
was examined by carrying out of repeated experiment with
O-methyl derivatives in toluene and acetic acid.
It is well known that the origin of enantioselectivity is di-
rectly related to the altered formation rates of the enantiomers
in the presence of the chiral modifier [31]. Therefore, concen-
trations of the (R)- and (S)-enantiomers in the classic enan-
tiomeric excess definition (ee1 in Eq. (1)) were replaced by ap-
parent rate constants of the single hydrogenation step (Eq. (2)),

342
I. Busygin et al. / Journal of Catalysis 254 (2008) 339–348
culated (Eq. (6)) erythro/threo ratios are defined as
[(1R, 2S)] + [(1S, 2R)]
[(1R, 2R)] + [(1S, 2S)]
erythro/threo =
erythro/threo =
,
(5)
(6)
k5 + k7 + k9 + k11
.
k6 + k8 + k10 + k12
The regioselectivity (rs) is defined as shown in
[(1R)] + [(1S)]
rs =
.
(7)
[(2R)] + [(2S)]
3. Results and discussion
In the present study, the kinetics of PPD hydrogenation in
toluene and acetic acid was explicitly addressed using several
new modifiers. A typical experiment was followed for 8–14 h
which allows studying the hydrogenation of hydroxyketones,
the primary hydrogenation products of PPD. All of the modi-
fiers presented in Table 1 were hydrogenated in toluene. Insta-
bility of the O-silyl ether derivatives under acidic conditions
prevented their use as chiral modifiers in acetic acid. The sta-
bility of O-phenyl derivatives of cinchona alkaloids under the
conditions employed for catalysis was previously questioned by
Baiker and co-workers [46]. Hence, in the present study, stabili-
ties of the O-phenyl ethers were investigated in both acetic acid
and toluene under the actual hydrogenation conditions. The re-
sults are discussed in the following section.
Fig. 1. Comparison of the model prediction (solid lines) with experimental
data (symbols) of PPD hydrogenation over cinchonidine modified Pt/Al O
2
3
in toluene: (a) conversion versus time (2) PPD; (Q) (1R); (a) (1S);
(r) (1R,2S)-diol and (s) (1S,2R)-diol. 2-Hydroxyketones and threo-diols are
not displayed. (b) Enantioselectivity versus time data.
obtained from the kinetic model.
[(1R)] − [(1S)]
ee1 =
Er1 =
× 100%,
(1)
(2)
[(1R)] + [(1S)]
k1 − k2
× 100%,
k1 + k2
where k1 and k2 are the rate constants identified in Scheme 1.
Considering the mechanistic aspect, enantioselectivity ex-
pressed as Er results in more accurate interpretation of enan-
tioselectivity in the first hydrogenation step. The Er value is
independent of the conversion whereas the ee based on exper-
imental concentrations is altered by kinetic resolution taking
place in the second hydrogenation step.
The kinetic resolution of the (1R)/(1S) mixture was quanti-
fied by the rate constants (Eq. (3)) for estimation of the modifier
structure effect:
3.1. Modifier stability and influence of the impurities
The stability of a chiral modifier under the employed reac-
tion conditions is particularly important for mechanistic studies.
Column purification of silyl ethers promotes partial cleavage of
the Si–O bond [13]. Therefore, micro amounts of cinchonidine
and cinchonine may be present in the O-silyl ether modifier
samples. Thus, in the present study, O-silyl derivatives of cin-
chona alkaloids were purified by adsorption over the Pt/Al2O3
catalyst [13] and the results were compared with the data ob-
tained from experiments with column-purified modifiers. The
kinetics of the hydrogenation reaction, as demonstrated for
O-(allyldimethylsilyl)cinchonidine (ADMSOCD), was not af-
fected by the purification procedure, indicating that amounts of
free cinchonidine sufficient to influence the results of the catal-
ysis experiments were not present in the modifier samples.
This observation can easily be rationalized by considering
the effect of modifier concentration in the hydrogenation of
PPD. Enantioselective hydrogenation of PPD over cinchona-
modified oxide-supported platinum catalysts requires higher al-
kaloid concentrations than commonly applied for enantioselec-
tive hydrogenations of α-keto esters [31]. As little as 0.7 mol%
of cinchonidine in admixture with PhOCD can control the enan-
tioselection in ketopantolactone hydrogenation [47]. For PPD,
a cinchonidine concentration of 1.4 × 10−5 moldm−3 (corre-
sponding to approx. 4 mol% with the commonly utilized mod-
ifier amounts and constant catalyst loading) is required for
obtaining a minimal noticeable effect of chiral modification
(ee = 8% in ethyl acetate) [31]. Thus, taking into considera-
tion the experimental conditions, it can be concluded that the
k5 + k6
k(1R)/k(1S) =
.
(3)
k7 + k8
At high conversion of 1-hydroxyketones, a first-order kinetic
approach is not able to describe all the processes on the surface.
As will be discussed later, the selectivity of the second hydro-
genation step depends on the conversion of the hydroxyketones.
Consequently, and in order to avoid overestimation of the accu-
racy of the values reported, k(1R)/k(1S) will be represented as
a proper fraction.
Diastereoselectivity of the (R)-1-hydroxy-1-phenyl-2-pro-
panone hydrogenation (dr(1R)) is defined in
k5 − k6
dr(1R) =
× 100%.
(4)
k5 + k6
Diastereoselectivity of the (S)-enantiomer (dr(1S)) hydrogena-
tion is defined in an analogous fashion.
The ratio of the erythro- ((1R,2S)- and (1S,2R)-diol) and
threo-diols ((1R,2R)- and (1S,2S)-diol) was quantified by the
rate constants. The experimentally observed (Eq. (5)) and cal-

I. Busygin et al. / Journal of Catalysis 254 (2008) 339–348
343
PPD hydrogenation is less sensitive to minor impurities in the
modifier than the hydrogenation of keto esters.
Impurities may also be formed during the experiment by
cleavage or hydrogenolysis of the O-ether substituent. In
PhOCD, the phenoxy moiety is cleaved under strongly acidic
conditions [46]. Consequently, in the present study, stability
of the O-ether bond under currently employed reaction con-
ditions was investigated. Solutions of one silyl ether (ADM-
SOCD) in toluene and two O-phenyl derivatives (PhOCD
and PhOCN), both in toluene and acetic acid, were exposed
to hydrogen pressures of 10 bar over the platinum cata-
lyst for 12–16 h. After work-up, the samples were analyzed
by NMR. The compounds observed were identified as O-
(propyldimethylsilyl)-10,11-dihydrocinchonidine (PDMSOD-
HCD), O-phenyl-10,11-dihydrocinchonidine and O-phenyl-
10,11-dihydrocinchonine, respectively. The samples were also
examined for the presence of decomposition product traces.
In short, the NMR detection limit for eventual impurities is
considerably lower than the amounts of modifier needed for
obtaining enantioselection in the PPD hydrogenation. Products
of C(9)–O hydrogenolysis or O–R hydrolysis were not detected
Fig. 2. Comparison of the model prediction (solid lines) with experimental
data (symbols) of PPD hydrogenation over TBDMSOCD modified Pt/Al O
2
3
in toluene. Designations are identical to Fig. 1.
the calculated and experimental concentrations. In the present
investigation, for the first time, the complete hydrogenation
scheme is evaluated. The other features of the earlier model, ad-
sorption of the modifier, active and spectator species and mul-
ticentered adsorption, are, however, not considered here since
experiments were conducted only at one concentration level
both for the substrate and for the modifier.
1
in the H and 13C NMR spectra of the hydrogenated modi-
fiers, demonstrating that the C(9)–O and O–R bonds are stable
under the conditions employed in the catalytic hydrogenation
experiments and, accordingly, the impurity traces should have
no effect on the reaction kinetics.
3.2. Catalytic hydrogenation
3.2.2. Influence of the modifier structure on the first
hydrogenation step
Hydrogenation of PPD proceeds in two steps (Scheme 1).
First, two regioisomers, 1-hydroxy-1-phenyl-2-propanone and
2-hydroxy-1-phenyl-1-propanone are formed, which both ex-
ist as their (R)- and (S)-enantiomers. Further hydrogenation
of the hydroxyketones produces the corresponding diastere-
omeric (1R,2R)-, (1S,2S)-, (1R,2S)- and (1S,2R)-1-phenyl-
1,2-propanediols as two pairs of enantiomers. Kinetic resolu-
tion of the 1-hydroxy-1-phenyl-2-propanones is commonly ob-
served during the second hydrogenation step.
The hydrogenation results are presented in Table 2. The rate
of the PPD hydrogenation is lower in the presence of a chi-
ral modifier being in line with our previous reports [14,15,30,
31]. Generally, but with some exceptions, the rate of hydro-
genation in toluene is gradually decreasing with increase in the
C(9)-O-substituent bulkiness, while in acetic acid it is almost
independent on the modifier structure.
Regioselectivity (rs) of the hydrogenation in toluene was al-
ways lower when O-ethers were used as chiral modifiers (rs =
2.0–3.1) instead of the parent cinchona alkaloids (rs = 3.0 and
4.8). With these modifiers, the rs was also lower than in the
hydrogenation in the absence of modifier (racemic hydrogena-
tion rs = 4.1), indicating that the cinchona alkaloid O-ether
modified chiral sites slightly disfavor the hydrogenation of the
C1=O1 carbonyl group in comparison with the unmodified sur-
faces. The nature of the O-ether substituent has only marginal
effect on the regioselectivity in toluene within the ROCD and
ROCN series.
The modifier hydroxyl group is essential for achieving high
enantioselectivity in the hydrogenation of the PPD C1=O1
keto group in toluene. This observation is in line with our
earlier investigations [14,15]. In all examples presented here,
the cinchona alkaloid O-ethers induced either an inversion or
complete loss of enantioselectivity in comparison with their
unsubstituted counterparts (Fig. 3). Similar observations have
been described in the literature for hydrogenation of vari-
ous aromatic α-hydroxy ketones [12], substituted acetophe-
nones [49] and 1,1,1-trifluoro-2,4-diketones [50]. Obviously,
3.2.1. Kinetic modeling
The apparent rate constants of all hydrogenation steps were
estimated using the first-order kinetic approach (Tables S1
and S2, Supplementary material). Quantitative treatment of the
hydrogenation kinetics is based on the reaction scheme depicted
in Scheme 1. Despite the apparent simplicity, the first-order ki-
netic model, which neglects the differences in adsorption and,
strictly speaking, is valid in the cases of low surface coverage
only, still allows the estimation of the rate constants in PPD
hydrogenation with high accuracy. Examples of the fit are pro-
vided in Figs. 1 and 2 showing very good agreement between
the experimentally recorded and predicted concentrations. Fur-
thermore, the selectivities calculated from experimentally mea-
sured concentrations and from estimated rate constants are in
good agreement with each other (Tables 2 and 3) giving further
support for the reliability of the kinetic modeling.
In the model developed in our earlier study [31], kinetic
resolution was not considered resulting in deviation between

344
I. Busygin et al. / Journal of Catalysis 254 (2008) 339–348
Table 2
The efficiency of ether derivatives of cinchonidine and cinchonine used as chiral modifiers of Pt/Al O in the heterogeneous hydrogenation of ethyl pyruvate (EtPy)
2
3
and 1-phenyl-1,2-propanedione (PPD)
EtPy
Modifier
PPD
c
a
b
a
ee, %
rs
Initial rate
k
ee , %
Er , %
1
PPD
1
d
d
d
d
d
74 [46]; 85 [23]; (89) [23]
72 [13] (91) [22]
CD
MeOCD
PhOCD
TMSOCD
TBDMSOCD
ADMSOCD
ADMSOCD
DPMSOCD
CN
4.8 (3.2)
2.2 (2.5)
3.1 (3.1)
2.4
2.6
2.5
2.3
2.2
3.0 (3.0)
2.1 (2.5)
2.3 (2.8)
2.0
11.3 (17.3)
7.6 (16.2)
10.9 (13.7)
12.2
11.1
7.8
7.8
8.6
7.3 (12.9)
13.1 (13.3)
12.6 (13.1)
10.8
31.8 (52.2)
19.8 (48.6)
30.1 (39.1)
31.2
31.6
20.6
21.1
22.4
15.0 (38.5)
37.9 (36.3)
33.4 (42.4)
30.9
63 (13)
60 (11)
f
e
f
−10 (1)
−10 (2)
−27 (−5)
0 (−)
14 [13]; −7 [46]; −21 [48]
−28 (−5)
0 (−)
g
−28 [13] ; 26 [46]
–
–
–
−27 (−)
−1 (−)
2 (−)
−28 (−)
0 (−)
g
0 (−)
g,h
−20 [13]
−3 (−)
−19 (−3)
32 (−2)
33 (3)
−3 (−)
−18 (−3)
32 (−2)
30 (3)
−70 (−81) [23]
e
−32 [22]; (−85) [22]
MeOCN
PhOCN
TMSOCN
–
–
30 (−)
28 (−)
i
–
4.1
21.2
0
a
At 50% conversion of PPD.
−1
−1
g
cat
b
−4
×10 molmin
.
−1
c
d
e
f
−1
k
PPD
= (k + k + k + k ) [h
g
].
1
2
3
4
cat
The first number attributed to the hydrogenation in toluene, the number in parentheses to the hydrogenation in acetic acid.
Dihydro derivative.
Negative value corresponds to the excess of (S)-enantiomer.
Purified by adsorption over the metal catalyst.
g
h
i
Ph SiOCD.
3
Racemic hydrogenation; taken from [15].
Table 3
The efficiency of ether derivatives of cinchonidine and cinchonine in the second step of PPD hydrogenation
a
b
c
Yield of diols
after 8 h
k(1R)/k(1S)
dr(1R), %
dr(1S), %
erythro/threo
d
Calculated
Experimental
e
e
e
e
e
CD
MeOCD
PhOCD
TMSOCD
TBDMSOCD
ADMSOCD
DPMSOCD
CN
51 (38)
48 (30)
51 (19)
43
45
45
1/3 (1/3)
58 (33)
38 (100)
4.1 (3.6)
5.2 (3.2)
5.8 (4.4)
4.9
4.2
3.2
3.7 (4.9)
10/9 (2/5)
3/2 (3/5)
7/6
33 (21)
2 (−14)
49.5 (100)
22.5 (35)
23
4.0 [5.5 (38)]
2.8 [2.9 (21)]
2.8
2
12
2
3
3/2
5/6
1/1
7.4
26.2
27.1
38 (58)
4.6 (60)
−17 (7)
−21
2.4
2.3
2.7
37
4.1
32 (32)
55 (30)
54 (21)
39
1/1 (2/1)
3/4 (17/6)
5/7 (1/1)
2/3
25 (96.5)
29 (100)
26 (3)
34
4.2 (1.5)
3.6 (2.3)
1.7 (2.9)
2.7
3.5 (36) (3.6)
2.9 [6.3 (36)]
2.6 [1.9 (27)]
2.2
MeOCN
PhOCN
TMSOCN
a
k(1R)/k(1S) = (k + k )/(k + k ).
7
5
6
8
b
c
d
e
dr(1R) = [(k − k )/(k + k )] × 100, %.
5
6
5
6
dr(1S) = [(k − k )/(k + k )] × 100, %.
7
7
8
8
When diols yield is 50%, unless otherwise specified as a superscript value.
The first number attributed to the hydrogenation in toluene, the number in parentheses to the hydrogenation in acetic acid.
on the Pt surface, the cinchona alkaloid O-ethers adopt an ad-
sorption mode or conformation that favors the formation of
the product (S)-enantiomer in the case of the substituted cin-
chonidine derivatives and the product (R)-enantiomer in the
case of the substituted cinchonine derivatives. This is an im-
portant observation concerning the nature of the chiral site
on the Pt surface. Clearly, the observed inversion of enan-
tioselectivity induced by the cinchona alkaloid ether modifiers
in toluene indicates that the enantiodifferentiating steps over
the cinchona alkaloid/Pt and cinchona alkaloid ether/Pt sys-
tems do not proceed via the same type of substrate–modifier
complexes. Besides of the hydroxyl group blocking and sub-
sequent inability to form hydrogen bonding between the re-
actant carbonyl group and the modifier C(9)–OH group, the
bulky substituent could induce differences in the modifier ad-
sorption mode [46]. If the latter is the case, the enantiose-
lectivity should correlate the substituent bulkiness. The size
of the C(9)-O-substituent increases in the series MeOCD–
TMSOCD–ADMSOCD–TBDMSOCD–DPMSOCD. As it was
studied for this series (Fig. 3 and Table 2), the substituent bulk-
iness does not correlate with enantioselectivities observed in
toluene which supports the interaction between the modifier hy-
droxyl group and the reactant carbonyl group on the platinum
surface [51].

I. Busygin et al. / Journal of Catalysis 254 (2008) 339–348
345
The enantiomeric excess, however, appears to show a correla-
tion.
3.2.3. Kinetic resolution of the
1-hydroxy-1-phenyl-2-propanones
Due to the consecutive reaction network, the relative concen-
trations of the 1-hydroxy-1-phenyl-2-propanone enantiomers
change significantly during the course of the reaction. Follow-
ing the reaction over a long period of time enables to study
the effect of the C(9) substituent on the kinetics of the second
hydrogenation step. Kinetic resolution of the primary hydro-
genation products is influenced by the presence of the modi-
fier. When cinchonidine is utilized as a chiral modifier, both
in toluene and acetic acid, the enantiomeric excess increases
with higher conversion of PPD being indicative of higher rate
of (S)-enantiomer hydrogenation (Fig. 1b) [15]. The opposite
is observed in the cinchonine case where the (R)-enantiomer is
reacting faster than its (S)-enantiomer, but only in acetic acid.
In toluene both enantiomers are reacting almost with the same
rate (Fig. 3).
Fig. 3. Enantioselectivities of PPD hydrogenation in toluene over Pt/Al O
catalysts modified with: (") CD; (!) CN; (2) MeOCD; (1) MeOCN; (Q)
PhOCD; (P) PhOCN; (F) TMSOCD; (E) TMSOCN.
2
3
In toluene the substitution of the modifier hydroxyl group
did not affect the rate of hydroxyketone hydrogenation, while
in acetic acid the substituent bulkiness appears to be crucial.
The yield of the diols after 8 h of reaction is gradually decreas-
ing with increase in the substituent group size (OH–OMe–OPh,
Table 3). The relationship holds for both cinchonidine and cin-
chonine derivatives.
Kinetic resolution of the 1-hydroxyketones results from the
lower hydrogenation rate of enantiomer produced in excess
during the first hydrogenation step. Thus, (1S)- reacts faster
than its (1R)-enantiomer when the catalyst surface is modified
by cinchonidine, and an inversion in the sense of the enan-
tiomer preference is observed in toluene when the hydroxyl
group is substituted. The most effective resolution was observed
when cinchonidine was used as a chiral modifier and the (1S)-
enantiomer reacts three times faster than its (1R)-enantiomer.
Almost no preference for one of the 1-hydroxyketones was
detected with MeOCD, TMSOCD, ADMSOCD and CN mod-
ifiers in toluene, while PhOCD and TBDMSOCD modifiers
induced (1R)-enantiomer higher reaction rate (Table 3).
In acetic acid, the (S)-enantiomer also reacts faster than
the (R)-enantiomer in the presence of cinchonidine and its
derivatives and slower in the presence of cinchonine deriva-
tives (Fig. 4). The effect of substituent bulkiness is evident
in the case of cinchonidine, as the increase in the size of the
C(9)-O-substituent results in less effective resolution of the
1-hydroxyketones. Cinchonine and its derivatives do not show
clear correlation. The lack of correlation may originate from the
structure of cinchonine, where the vinyl group at C(3) is spa-
tially closer to the hydroxyl group than in cinchonidine. The
present observations explain the high selectivity towards the
formation of (1R,2S)- and (1S,2R)-diols in acetic acid at the
early stage of the reaction when MeOCN [15] and O-methyl-
10,11-dihydrocinchonidine were used as chiral modifiers, re-
spectively [14,15].
Fig. 4. Resolution of 1-hydroxyketones in acetic acid over Pt/Al O catalysts
modified with: (") CD; (!) CN; (2) MeOCD; (1) MeOCN; (Q) PhOCD; (P)
PhOCN.
2
3
In the present study, the complete reaction profile was also
followed in acetic acid with O-phenyl and -methyl ethers of CD
and CN utilized as modifiers. Clearly, for both CD and CN and
their corresponding ether derivatives, a correlation between the
substituent bulkiness and ee can be observed (Table 2, Fig. 4).
The correlation is more pronounced in the cinchonidine case
while being considerably less pronounced in the cinchonine se-
ries. In acetic acid, both O-phenyl ethers induce an inversion
of enantioselectivity. The solvent employed plays a significant
role in the enantiodifferentiation of PPD over chirally modi-
fied platinum catalysts [52]. For hydrogenation in acetic acid,
the observed enantiomeric excesses are low, and the results ob-
tained should be critically assessed when drawing mechanistic
conclusions on the hydrogenation.
To summarize the results from the hydrogenation of PPD in
acetic acid, neither the regioselectivity nor the initial rate of re-
action directly correlates with the size of the C(9) substituent.
It is noteworthy, that in the case of the O-methyl ether deriv-
atives the solvent plays a determinative role in favoring the

346
I. Busygin et al. / Journal of Catalysis 254 (2008) 339–348
reaction of one enantiomer over another. In the second hydro-
genation step, the (1R)-enantiomer is consumed slightly faster
than the (1S)-enantiomer when MeOCD is employed as the chi-
ral modifier in toluene, whereas in acetic acid the opposite is
observed. For MeOCN the solvent-dependent inversion of the
sense of kinetic resolution is well pronounced.
an excess of threo-(1S,2S)-diol. Moreover, a correlation be-
tween the substituent bulkiness and diastereoselectivity of the
(1R)-hydroxyketone hydrogenation was observed for cinchoni-
dine and its derivatives. When the bulkiness of the group at
the C(9) position of the modifier was increased in the order
OH → OMe → OR (R = Ph, Me3Si, Me2(CH2=CHCH2)Si,
t-BuMe2Si, Ph2MeSi), the diastereoselective excess gradually
decreased from 58% to almost 0% (Table 3). In the cinchonine
series, however, diastereoselectivity of the (1R) hydrogenation
did not depend on the C(9)-substituent. A striking resemblance
was observed for the (1S)-hydroxyketone hydrogenation, tak-
ing into consideration that for the cinchonine derived modifiers,
diastereoselectivity correlates with the substituent bulkiness
(Table 3), whereas in the case of cinchonidine and its deriva-
tives direct correlation was not observed. It is evident that the
modifier structure–selectivity dependence is different for the
enantio- and diastereoselection in PPD and 1-hydroxyketone
hydrogenations in toluene. Substitution of the modifier hy-
droxyl group results in a dramatic change in enantioselectivity
of the PPD hydrogenation. In contrast, the diastereoselectiv-
ity of 1-hydroxyketone hydrogenation gradually decreases with
increase in the substituent bulkiness. Therefore, it can be con-
cluded that the OH group of the modifier is not directly involved
in the formation of the complex between the 1-hydroxyketone
and the chiral modifier on the catalyst surface, whereas for PPD
opposite observations have been reported [14,15,51].
The model presented predicts highly diastereoselective hy-
drogenation of 1-hydroxyketones in acetic acid (Table 3). This
is in line with the previous observations [15]. The high di-
astereomeric excess in the hydrogenation of 1-hydroxyketones
in acetic acid and the low ee observed in the hydrogenation of
PPD in the same media also support the hypothesis that the
mechanisms for stereoselection are different for the aromatic
and aliphatic ketones.
Experimentally, it was observed that the ratio of the diols
strongly depends on the conversion of 1-hydroxyketones both
in acetic acid and toluene. When the conversion is low, ery-
thro-diols are produced in large excess over the threo-diols.
With an increase in conversion, the ratio is gradually decreas-
ing. Obviously, the first-order kinetic approach cannot describe
all the processes on the surface at higher metal coverage by di-
ols (cf. the values for experimentally observed and calculated
erythro/threo in Table 3). Earlier it has been shown by in situ
ATR-IR monitoring of a Pt/Al2O3 film during the hydrogena-
tion of ethyl pyruvate in supercritical CO2 [54] that desorption
of ethyl lactate is slower than the adsorption of ethyl pyruvate
and the hydrogenation reaction. Clearly, the solvent plays an
important role in the adsorption–desorption processes over het-
erogeneous catalysts which has been observed for ethyl pyru-
vate hydrogenation [55]. Nevertheless, the accumulation of the
diol products on the catalyst surface and the subsequent alter-
ation of the chiral site cannot be excluded entirely.
3.2.4. Diastereoselective hydrogenation of
1-hydroxy-1-phenyl-2-propanone
In the present investigation, kinetics of 1-hydroxy-1-phenyl-
2-propanone hydrogenation and the influence of the modifier
structure effect on it has been considered for the first time
(Table 3). In particular, it is interesting to compare the mech-
anisms of enantio- and diastereoselection for PPD and the 1-
hydroxyketones. More specifically, the same type of modifier
structure–selectivity dependence and solvent effect in the two
cases indicate similar modifier–reactant complexes active in
enantio- and diastereoselection, and vice versa.
Recently, we have also investigated the hydrogenation
of enantiomerically enriched (>90% ee) (R)-1-hydroxy-1-
phenyl-2-propanone (1R in Scheme 1) over cinchonidine mod-
ified platinum catalyst, the results of which will be addressed in
more detail in a separate study [53]. In brief, the experimentally
observed diastereoselectivity in this reaction was approximately
60% in favor of the (1R,2S)-diol when cinchonidine was used
as the chiral modifier in toluene. Moreover, the hydrogenation
of (1R) proceeds diastereoselectively even over the unmodified
platinum catalyst (de = 49%) [53]. In the present study, we ap-
ply the kinetic approach to estimate the diastereoselectivities
in the hydrogenation of the 1-hydroxy-1-phenyl-2-propanones
being present as component of the complex mixture of PPD
hydrogenation products. This task is very challenging since
the hydrogenation of two different isomeric hydroxyketones
results in the formation of identical diols. For example, hydro-
genation of both (1R)-hydroxyketone and (2S)-hydroxyketone
potentially provides the (1R,2S)-diol as one of the two prod-
ucts. Diastereoselectivities in the hydrogenation of the (1R)-
and (1S)-hydroxyketones (dr(1R) and dr(1S), respectively)
were here calculated using the apparent rate constants (Table 3).
Diastereoselectivity of the (1R)-hydroxyketone hydrogenation
(dr(1R) value in Table 3) calculated from the PPD hydrogena-
tion experiments over the CD-modified catalyst in toluene in
the present work is in very good agreement with the diastere-
oselectivity observed experimentally under similar conditions
in the hydrogenation of (R)-1-hydroxy-1-phenyl-2-propanone
alone [53]. This observation supports the reliability of the
diastereoselectivity estimation method utilized in the present
study.
Hydrogenation of the 1- and 2-hydroxyketone mixture al-
ways gives rise to the preferential formation of erythro-diols
(Table 3). In toluene, the O-ether cinchonidine derivatives
are less effective for diastereoselective hydrogenation of (1R)
than the parent alkaloid, while the O-ether cinchonine deriv-
atives are as effective as CN. The hydrogenation of (1S) is
also diastereoselective, with MeOCD being the most efficient
modifier in toluene. Diastereoselectivity inversion was ob-
served for two modifiers, PhOCN and TMSOCN, resulting in
Another possible explanation is that kinetic resolution of the
hydroxyketones results in regio and enantiomers reacting sep-
arately during the time scale studied. When CD was used as a
modifier in the hydrogenation of PPD in toluene (Fig. 1) the di-
ols were observed in 1% yield already after 15 min of reaction.

I. Busygin et al. / Journal of Catalysis 254 (2008) 339–348
347
The amounts of the 1-hydroxyketones and 2-hydroxyketones
were 25.7% and 5.7%, respectively, and increased until almost
all of PPD was consumed. Thus, after 2 h of reaction the yield of
the hydroxyketones was at its maximum (79.6%) and gradually
decreased after this point. After 8 h, both (1S) and (2S) were al-
most completely consumed indicating that the diols obtained at
this point and thereafter are produced from the hydrogenation
of the (1R)- and (2R)-hydroxyketones, while the diols obtained
earlier result from the hydrogenation of all four hydroxyke-
tones.
of cinchonine were better kinetic resolution agents than the
parent alkaloid. In acetic acid, the sense of kinetic resolution
was retained with the highest value obtained for the O-methyl
derivatives as modifiers within the cinchonidine and cincho-
nine series. Correlation between the substituent bulkiness and
diastereoselectivity of the 1-hydroxyketone hydrogenation was
observed for the reactant–modifier pairs: (1R)-hydroxyketone–
cinchonidine O-ethers and (1S)-hydroxyketone–cinchonine
O-ethers. Hydrogenation of (1S)-hydroxyketone was highly di-
astereoselective when cinchonidine and O-methyl cinchonidine
were used as modifiers in acetic acid. Moreover, in the same
solvent hydrogenation of (1R)-hydroxyketone was also highly
diastereoselective with cinchonine and its O-methyl derivative.
It should be emphasized that the hydrogenation of PPD
is very complex and mechanistically different from the well
established keto ester hydrogenation. The experimentally ob-
served structure–selectivity effects indicate that the mechanism
of enantiodifferentiation over the catalyst modified by parent
cinchona alkaloids and their ether derivatives differ from each
other.
3.3. Conformational equilibria versus selectivity
The central role of the Open(3) conformation of cinchoni-
dine is commonly accepted as crucial for obtaining high enan-
tioselectivity in the hydrogenation of carbonyl compounds over
cinchona-modified platinum catalysts [56]. The decrease of ee
for O-substituted cinchona alkaloids in enantioselective hydro-
genation of ethyl pyruvate in apolar solvents was related to the
population of the Open(3) conformation in the liquid phase and
consequently on the metal surface [56]. As a result of modifier
adsorption–desorption processes, the conformational equilibria
in the solution can provide some valuable information on the
distribution of the adsorbed species.
Although the conformational equilibria of the parent cin-
chona alkaloids in the liquid phase have been studied in detail
earlier [57,58], only a few studies have addressed the confor-
mational behavior of the ether derivatives [58–60]. Considering
that the modification of the cinchona alkaloid structure induces
changes in the conformational equilibria [58], it is challenging
to investigate whether the selectivity over the cinchona alkaloid
O-ether modified metal catalysts correlates with the popula-
tion of Open(3) or any other conformation in the liquid phase.
Structures and conformational behavior of selected modifiers
in solution (NMR), in vacuum (DFT) and in the solid state (X-
ray) were also investigated and will be reported in a separate
paper [44]. The conformational equilibria of cinchona alkaloid
O-ethers will be discussed in the light of the results presented
in this paper.
Acknowledgments
Financial support by the Magnus Ehnrooth Foundation (I.B.)
and by the Academy of Finland (project 204724, E.T.) is grate-
fully acknowledged.
Supplementary material
The online version of this article contains additional supple-
mentary material.
Please visit DOI: 10.1016/j.jcat.2008.01.012.
References
[1] See Curr. Org. Chem. 10 (13) (2006), volume dedicated to the topic of the
use of heterogeneous catalysts in organic synthesis.
[2] M. Heitbaum, F. Glorius, I. Escher, Angew. Chem. Int. Ed. 45 (2006) 4732.
[3] Y. Orito, S. Imai, S. Niwa, in: 43rd Catalyst Forum, Japan, 1978, p. 30.
[4] M. Bartók, Curr. Org. Chem. 10 (2006) 1533.
[5] D.Yu. Murzin, P. Mäki-Arvela, E. Toukoniitty, T. Salmi, Catal. Rev. 47
(2005) 175.
[6] G.J. Hutchings, Annu. Rev. Mater. Res. 35 (2005) 143.
[7] T. Bürgi, A. Baiker, Acc. Chem. Res. 37 (2004) 909.
[8] M. Studer, H.-U. Blaser, C. Exner, Adv. Synth. Catal. 345 (2003) 45.
[9] C.J. Baddeley, Top. Catal. 25 (2003) 17.
[10] T. Mallat, E. Orglmeister, A. Baiker, Chem. Rev. 107 (2007) 4863.
[11] S. Diezi, A. Szabo, T. Mallat, A. Baiker, Tetrahedron: Asymmetry 14
(2003) 2573.
[12] O.J. Sonderegger, G.M.W. Ho, T. Bürgi, A. Baiker, J. Mol. Catal. A
Chem. 229 (2005) 19.
[13] S. Cserényi, K. Felföldi, K. Balázsik, G. Szöllo˝si, I. Bucsi, M. Bartók,
J. Mol. Catal. A Chem. 247 (2006) 108.
[14] E. Toukoniitty, I. Busygin, R. Leino, D.Yu. Murzin, J. Catal. 227 (2004)
210.
4. Conclusion
In the hydrogenation of 1-phenyl-1,2-propanedione using
cinchona alkaloid O-ethers as chiral modifiers, apparent rate
constants of all hydrogenation steps were estimated by the
first-order kinetic approach. Very good agreement between
the experimentally recorded concentrations and the ones pre-
dicted by the kinetic model was achieved, which is particularly
remarkable considering the complexity of the reaction. The
modifier structure–selectivity dependence and the solvent effect
were different for enantio- and diastereoselection in the PPD
and 1-hydroxyketone hydrogenations. The sense of enantios-
electivity in the PPD hydrogenation was either inverted when
compared with the corresponding unsubstituted alkaloids or,
alternatively, no chiral induction was observed. A sense of ki-
netic resolution of the 1-hydroxyketones was also inverted with
the O-ethers of cinchonidine in toluene, whereas the O-ethers
[15] I. Busygin, E. Toukoniitty, R. Leino, D.Yu. Murzin, J. Mol. Catal. A
Chem. 236 (2005) 227.
[16] N. Bonalumi, A. Vargas, D. Ferri, A. Baiker, Chem. Eur. J. 13 (2007) 9236.
[17] K. Balázsik, I. Bucsi, S. Cserényi, G. Szöllo˝si, M. Bartók, J. Mol. Catal.
A Chem. 280 (2008) 87.
[18] M. Bartók, M. Sutyinszki, K. Felföldi, G. Szöllösi, Chem. Commun.
(2002) 1130.

348
I. Busygin et al. / Journal of Catalysis 254 (2008) 339–348
[19] J.L. Margitfalvi, E. Tálas, Appl. Catal. A Gen. 301 (2006) 187.
[20] F. Hoxha, T. Mallat, A. Baiker, J. Catal. 248 (2007) 11.
[21] A. Pfaltz, T. Heinz, Top. Catal. 4 (1997) 229.
[42] N.F. Dummer, R. Jenkins, X. Li, S.M. Bawaked, P. McMorn, A. Burrows,
C.J. Kiely, R.P.K. Wells, D.J. Willock, G.J. Hutchings, J. Catal. 243 (2006)
165.
[43] N.J. Colston, R.P.K. Wells, P.B. Wells, G.J. Hutchings, Catal. Lett. 103
(2005) 117.
[22] C. Exner, A. Pfaltz, M. Studer, H.-U. Blaser, Adv. Synth. Catal. 345
(2003) 1253.
[44] I. Busygin, J. Sinkkonen, V. Nieminen, R. Sillanpää, E. Toukoniitty, D.Yu.
Murzin, R. Leino, manuscript in preparation.
[23] H.U. Blaser, H.P. Jalett, W. Lottenbach, M. Studer, J. Am. Chem. Soc. 122
(2000) 12675.
[45] H. Haario, Modest 6.1, Profmath Oy, Helsinki, 2002.
[46] S. Diezi, T. Mallat, A. Szabo, A. Baiker, J. Catal. 228 (2004) 162.
[47] N. Bonalumi, A. Vargas, D. Ferri, T. Bürgi, T. Mallat, A. Baiker, J. Am.
Chem. Soc. 127 (2005) 8467.
[48] S. Diezi, S. Reimann, N. Bonalumi, T. Mallat, A. Baiker, J. Catal. 239
(2006) 255.
[49] (a) R. Hess, A. Vargas, T. Mallat, T. Bürgi, A. Baiker, J. Catal. 222 (2004)
117;
(b) R. Hess, F. Krumeich, T. Mallat, A. Baiker, J. Mol. Catal. A Chem. 212
(2004) 205.
[50] R. Hess, S. Diezi, T. Mallat, A. Baiker, Tetrahedron: Asymmetry 15 (2004)
251.
[51] I. Busygin, O.P. Tkachenko, V. Nieminen, V.Yu. Borovkov, R. Sillanpää,
E. Toukoniitty, L.M. Kustov, D.Yu. Murzin, R. Leino, J. Phys. Chem.
C 111 (2007) 9374.
[52] E. Toukoniitty, P. Mäki-Arvela, J. Kuusisto, V. Nieminen, J. Päivärinta,
M. Hotokka, T. Salmi, D.Yu. Murzin, J. Mol. Catal. A Chem. 192 (2003)
135.
[53] I. Busygin, A. Taskinen, V. Nieminen, E. Toukoniitty, T. Stillger, M. Mül-
ler, D.Yu. Murzin, R. Leino, manuscript in preparation.
[54] M.S. Schneider, J.-D. Grunwaldt, T. Bürgi, A. Baiker, Rev. Sci. In-
strum. 74 (2003) 4121.
[55] A. Gamez, J. Köhler, J. Bradley, Catal. Lett. 55 (1998) 73.
[56] T. Bürgi, A. Baiker, J. Am. Chem. Soc. 120 (1998) 12920.
[57] G.D.H. Dijkstra, R.M. Kellogg, H. Wynberg, J.S. Svendsen, I. Marko,
K.B. Sharpless, J. Am. Chem. Soc. 111 (1989) 8069.
[58] G.D.H. Dijkstra, R.M. Kellogg, H. Wynberg, J. Org. Chem. 55 (1990)
6121.
[24] H.U. Blaser, H.P. Jalett, M. Garland, M. Studer, H. Thies, A. Wirth-Tijani,
J. Catal. 173 (1998) 282.
[25] J.L. Margitfalvi, E. Tálas, E. Tfirst, Top. Catal. 39 (2006) 77.
[26] H.U. Blaser, M. Garland, H.P. Jalett, J. Catal. 144 (1993) 569.
[27] J.A. Slipszenko, S.P. Griffiths, P. Johnston, K.E. Simons, W.A.H. Vermeer,
P.B. Wells, J. Catal. 179 (1998) 267.
[28] M. Studer, V. Okafor, H.-U. Blaser, Chem. Commun. (1998) 1053.
[29] O.J. Sonderegger, T. Bürgi, A. Baiker, J. Catal. 215 (2003) 116.
[30] E. Toukoniitty, P. Mäki-Arvela, M. Kuzma, A. Villela, A.K. Neyestanaki,
T. Salmi, R. Sjöholm, R. Leino, E. Laine, D.Yu. Murzin, J. Catal. 204
(2001) 281.
[31] E. Toukoniitty, B. Ševcˇíková, P. Mäki-Arvela, J. Wärnå, T. Salmi, D.Yu.
Murzin, J. Catal. 213 (2003) 7.
[32] E. Toukoniitty, P. Mäki-Arvela, A.N. Villela, A.K. Neyestanaki, R. Leino,
T. Salmi, R. Sjöholm, E. Laine, J. Väyrynen, T. Ollonqvist, P.J. Kooyman,
Catal. Today 60 (2000) 175.
[33] V. Nieminen, A. Taskinen, M. Hotokka, D.Yu. Murzin, J. Catal. 245
(2007) 228.
[34] A. Lindholm, P. Mäki-Arvela, E. Toukoniitty, T.A. Pakkanen, J.T. Hirvi,
T. Salmi, D.Yu. Murzin, R. Sjöholm, R. Leino, J. Chem. Soc. Perkin
Trans. 1 (2002) 2605.
[35] M. von Arx, T. Mallat, A. Baiker, Angew. Chem. Int. Ed. 40 (2001)
2302.
[36] M. Bartók, M. Sutyinszki, I. Bucsi, K. Felföldi, G. Szöllo˝si, F. Bartha,
T. Bartók, J. Catal. 231 (2005) 33.
[37] (a) M. Bartók, K. Balázsik, I. Bucsi, G. Szöllo˝si, J. Catal. 239 (2006) 74;
(b) K. Szo˝ri, K. Balázsik, K. Felföldi, M. Bartók, J. Catal. 241 (2006) 149.
[38] E. Orglmeister, T. Mallat, A. Baiker, J. Catal. 233 (2005) 333.
[39] F. Hoxha, L. Königsmann, A. Vargas, D. Ferri, T. Mallat, A. Baiker, J. Am.
Chem. Soc. 129 (2007) 10582.
[40] M. Maris, T. Mallat, A. Baiker, J. Mol. Catal. A Chem. 242 (2005) 151.
[41] R.L. Jenkins, N. Dummer, X. Li, S.M. Bawaked, P. McMorn, R.P.K. Wells,
A. Burrows, C.J. Kiely, G.J. Hutchings, Catal. Lett. 110 (2006) 135.
[59] (a) A. Vargas, D. Ferri, N. Bonalumni, T. Mallat, A. Baiker, Angew. Chem.
Int. Ed. 46 (2007) 3905;
(b) N. Bonalumni, A. Vargas, D. Ferri, A. Baiker, Chem. Eur. J. 13 (2007)
9236.
[60] A. Vargas, N. Bonalumni, D. Ferri, A. Baiker, J. Phys. Chem. A 110 (2006)
1118.