Enantioselective hydrogenation of ethyl pyruvate over diop modified Pt nanoclusters. Determination of geometry of the ligand adsorption mode via DRIFTS

Article abstract of DOI:10.1039/b615470k

Immobilization of chiral ligands on the surface of metal nanoparticles is one concept for heterogenation of enantioselective catalysts. Diop modified Pt nanoclusters were found to be able to induce enantioselectivity and enhancement of the reaction rate in hydrogenation of ethyl pyruvate. The model of geometry of the diop adsorption has been proposed based on DRIFTS and molecular modelling investigations. The ability of diop to induce enantioselectivity and enhanced reaction rate was explained through the adsorption geometry and comparison with known models of enantioselectivity established for the cinchonidine on Pt system. the Owner Societies 2007.

Full text of DOI:10.1039/b615470k

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/pccp | Physical Chemistry Chemical Physics
Enantioselective hydrogenation of ethyl pyruvate over diop modified Pt
nanoclusters. Determination of geometry of the ligand adsorption mode
via DRIFTSwz
Alexander Kraynov and Ryan Richards*
Received 25th October 2006, Accepted 13th December 2006
First published as an Advance Article on the web 10th January 2007
DOI: 10.1039/b615470k
Immobilization of chiral ligands on the surface of metal nanoparticles is one concept for
heterogenation of enantioselective catalysts. Diop modified Pt nanoclusters were found to be able
to induce enantioselectivity and enhancement of the reaction rate in hydrogenation of ethyl
pyruvate. The model of geometry of the diop adsorption has been proposed based on DRIFTS
and molecular modelling investigations. The ability of diop to induce enantioselectivity and
enhanced reaction rate was explained through the adsorption geometry and comparison with
known models of enantioselectivity established for the cinchonidine on Pt system.
Introduction
In this work, we chose the chiral ligand diop which is well
7–29
2
known in enantioselective homogeneous catalysis
and the
Heterogeneous enantioselective catalysis has attracted atten-
tion in science and in industry because of advantages in
catalyst separation, recycling etc. However, practical applica-
tions are limited to asymmetrical hydrogenation of a-keto
non-chiral diphos (due to the similarities in their structures,
Fig. 1) as modifiers for Pt nanoclusters and studied their
geometrical orientations on the surface of Pt nanoparticles
via DRIFTS and molecular modelling. The DRIFT spectra
were chosen for analysis because it is the more applicable
technique for observing adsorbed species on finely divided
1
esters over cinchonidine modified Pt catalysts. Enantioselec-
2
tive homogeneous catalysis is further developed and has been
3
implemented in industry. However, due to its homogeneous
17,30
metal powder.
4
nature, catalyst recycling and reuse are problematic.
We found that the diop modified Pt nanoclusters system can
induce enantioselectivity and rate enhancement in the hetero-
geneous ethyl pyruvate hydrogenation reaction.
There are several advantages to using nanoscale metal
particles as they have a large surface area and generally high
5
,6
7
8
activity and selectivity. Studies by Zuo et al. have revealed
that polymer (poly-N-vinyl-2-pyrrolidone) stabilized nano-
clusters exhibited novel catalytic properties for the hydrogena-
tion of a-ketoesters (e.g. structure insensitivity which differed
from their heterogeneous counterparts). Nanoscale metals
have demonstrated excellent catalytic properties e.g. in hydro-
The ability of diop modified Pt nanoclusters to induce
enantioselectivity was qualitatively explained through com-
parison with the cinchona on Pt system.
Experimental
9
10
genation, oxidation, carbon–carbon bond formation reac-
12,13
and asymmetrical
Materials
1
tions, enantioselective hydrogenations
1
1
4
Ethylene bis(diphenylphosphine) (diphos) 96% (Aldrich),
allylic alkylation. Soluble nanoclusters as ‘‘quasi-homoge-
neous’’ (or ‘‘soluble-heterogeneous’’) catalysts represent sup-
port-free analogues to the classical heterogeneous systems and
allow study of the catalytic behavior in the absence of support
(
4S, 5S)-4, 5-bis(diphenylphosphinomethyl)-2, 2-dimethyl-1, 3-
dioxolane (or (S,S)-Diop) and (4R,5R)-4,5-bis(diphenylpho-
sphinomethyl)-2,2-dimethyl-1,3-dioxolane (or (R,R)-Diop),
98% (Aldrich) were used as received. Pt (DBA) (DBA—bis-
1
5–17
effects.
2
3
dibenzylidene acetone) was synthesized according to the pub-
3
It is known for the system of cinchona family ligands on Pt
that the geometrical orientation of the ligand is a critical factor
for enantioselectivity in heterogeneous reactions over Pt and
1,32
lished literature protocol.
Conventional 5% Pt/Al
2 3
O
(
Aldrich) was activated under N
2
/H flow for 2 h at 400 1C
2
1
8,19
before use.
Pd catalysts.
Spectroscopic methods (ATR, RAIRS, etc.)
are widely used for studying molecules on the metal
2
surfaces.
Sample preparation
0–26
In order to prepare diphos on Pt nanoclusters, 560 mg
Pt (DBA) (1 mmol Pt) in 40 ml THF was added to 400 mg
2
3
School of Engineering and Science, International University Bremen,
Campus Ring 8, 28759 Bremen, Germany. E-mail: r.richards@
iu-bremen.de; Fax: +49 421 2003229; Tel: +49 421 2003236
w The HTML version of this article has been enhanced with colour
images.
z Electronic supplementary information (ESI) available: TEM image
of diop modified Pt nanoclusters (Fig. S1). See DOI: 10.1039/
b615470k
diphos (1 mmol) in 30 ml THF. The mixture was purged with
Ar for 10 min in a metal reactor and a hydrogen pressure of
4.5 bar was applied for 2.5 h under constant stirring. After-
wards, the hydrogen pressure was released and the mixture
was transferred into 400 ml cyclohexene. After 30 h the
supernatant was decanted and the precipitated black
8
84 | Phys. Chem. Chem. Phys., 2007, 9, 884–890
This journal is ꢀc the Owner Societies 2007

View Article Online
to 70 ml of THF, then 20 mg of catalyst was added, in the case
of colloidal catalysts the mixture was sonicated for 10 min,
transferred into the reactor, flushed with Ar for 10 min and
finally after the pressure of hydrogen had been stabilized for
1
0–15 s, the initial time (t = 0) was set and the reaction
kinetics were monitored. GC measurements were performed
on a Varian 3900, with a Lipodex-E chiral column. In order to
prevent contamination of the GC column, Pt colloids were
removed from the mixture before GC injection by alumina
powder. Fine alumina powder was heated at 120 1C for one
hour, then cooled down and added to the colloidal solution.
After 10 min of stirring, the almost colorless supernatant was
decanted and after repeating this procedure a second time,
injected into the GC. THF was used as a solvent, because diop
and diphos modified Pt nanoclusters are well redispersable in
THF, and at the same time THF was used in preparation of
nanoclusters, thus the further use of it prevents contact with
possible external effects from other solvents.
Fig. 1 Diphos (left) and diop (right) molecules.
substance was transferred into 400 ml THF:cyclohexene mix-
ture (1 : 7), and stirred for 10 min. After 4 h the precipitate was
collected and this procedure was repeated until no trace of the
yellow DBA remained in the supernatant. Finally, the black
powder was washed with 400 ml cyclohexene and dried in air
at 65 1C.
(
R, R) and (S, S) enantiomers of diop were used as modifiers
The metal content was found from thermogravimetric ana-
for Pt nanoclusters. Typically, 250 mg (0.5 mmol) e.g. (S, S)
lysis (TGA) over the 30–1200 1C temperature interval under
diop in 35 ml THF was transferred into the metal reactor with
ꢁ1
nitrogen 100 ml min with an SDTQ 600 instrument from TA
Instruments.
2
2 3
74 mg Pt (DBA) in 35 ml THF. The system was purged with
Ar for 10 min and finally a hydrogen pressure of 4 bar was set
for 6 h. The mixture was then transferred into 400 ml
cyclohexene. The precipitate collected after 20 h was washed
with 400 ml THF : cyclohexene mixture (1 : 4) and then with
2 3
Hydrogen chemisorption over the conventional Pt/Al O
was performed on an Autosorb 1-C instrument supplied by
Quantachrome, Germany. Typically the sample was preheated
ꢁ2
at 400 1C for 1 h under vacuum (10 mbar) and for 3 h under
a flow of hydrogen then chemisorption started at 40 1C in the
B0–600 mbar hydrogen pressure interval.
2
00 ml cyclohexene. Finally, the sample was dried at 40 1C
under vacuum.
Characterisation
Transmission electron microscopy (TEM) micrographs were
obtained by using a TECNAI F20 instrument. Specimens were
prepared by placing a drop of the colloid onto a copper grid
with a perforated carbon film and then allowing the solvent to
evaporate.
Results and discussion
General characterization
XRD analysis of diop and diphos modified Pt nanoclusters
clearly demonstrate crystalline nature of the samples (Fig. 2).
In fact, both spectra have a strong peak at 391 corresponding
to the (111) crystal face. The shoulder at 461 (diop on Pt, curve
Powder XRD spectra were recorded on a Siemens D5000
instrument with Cu Ka radiation.
DRIFT spectra of samples were measured with a Thermo
Nicolet 4700 FT-IR spectrometer with a liquid nitrogen cooled
detector using the following parameters: 300 scans, 600–
3
4
1
) corresponds to the (200) crystal face, whereas in the case
of diphos on the Pt sample the corresponding peak (or
ꢁ
1
ꢁ1
3
500 cm scan range, 4 cm resolution.
Molecular modelling including geometry optimisation and
IR spectra calculation has been performed with Gaussian 03
3
3
software for diop and diphos molecules by using the DFT
method (B3LYP) and (6-31G) basis set. Visualisation of the
geometry and IR vibrations was performed using Gaussview
software. In order to facilitate assignment of the peaks of the
experimental IR spectra for the free molecule, its IR spectra
were compared (in terms of intensities of the signals and their
vibrational frequencies) with calculated IR spectra using MS
Excel (intensities of both spectra were normalized, frequencies
of the calculated spectra were multiplied by a correlation
coefficient). The correlation coefficient dotheory/doexp of 1.02
was found to provide the best fit.
Hydrogenation of ethyl pyruvate was carried out in a metal
reactor (Parr) under a pressure of 10 bars of hydrogen at room
temperature (22 ꢂ 1 1C). The reactor was cleaned every time
before use with acetone and then with THF. The stirring speed
was 1000 rpm. The ethyl pyruvate (4 ml, 36 mmol) was added
Fig. 2 XRD spectra of diop (1) and diphos (2) modified Pt nano-
clusters.
This journal is ꢀc the Owner Societies 2007
Phys. Chem. Chem. Phys., 2007, 9, 884–890 | 885

View Article Online
Table 1 Average crystal size of diphos and diop modified Pt na-
noclusters estimated according to the Scherrer model from XRD
spectra and obtained from the TEM image
A very similar situation was found with the diop molecule,
in fact, the energy gap DE2,1 = 7.5 meV indicates that both
conformations (Fig. 4) coexist at room temperature.
Sample
hsizei, nm (Scherrer model)
hsizei, nm (TEM)
Diop on Pt
Diphos on Pt
1.7
1.3
2
—
Determination of geometrical orientation of diphos and diop on
Pt nanoclusters
In order to determine the orientation of the ligands on the
metal surface, DRIFT spectra of the free and adsorbed ligand
were compared and analyzed with the help of the metal surface
shoulder) is not seen, probably due to the noise in the
spectrum and broadening of the neighboring peak at 391.
3
5
37
Based on the Scherrer equation, the average size of the
metal crystals was estimated considering the width at half
height of the strongest peak at 391. The data are summarized
in Table 1 together with the average size of particles obtained
from the TEM image (available in the electronic supplemen-
tary informationz) of diop modified Pt nanoclusters.
selection rule and the orientation of the corresponding
dipole moment obtained from the molecular modelling.
The orientation of the phenyl rings in diphos and diop
molecules is not rigidly fixed due to the free rotation around
the Ph–P and P–C bonds, thus we do not have to take into
account the orientations of the dipole moments induced by
any vibration of any phenyl ring in diphos and diop in
determination of the adsorption mode from analysis of
DRIFT spectra of the corresponding samples.
The data obtained from the XRD pattern of diop modified
Pt nanoclusters are in good agreement with the data obtained
from the TEM image of the corresponding sample. The
average crystal size together with evaluation of the conversion
of ethyl pyruvate in time were used for the estimation of the
reaction rate normalized to the number of metal surface atoms
It is important to note that in the spectrum of free diop and
ꢁ
1
diphos some peaks (e.g. from in plane 1480–1430 cm
,
ꢁ
1
ꢁ1
1160–1000 cm and out of plane 846–675 cm range) are
superpositions of vibrations induced by single phenyl rings
and due to their close proximity they can not be well resolved
and thus have to be excluded from the analysis. Thus, only
vibrations of e.g. diphos which do not affect the vibrations of
the phenyl rings were taken into account. Among these (1276
3
6
(
estimated from the full-shell model).
It is important to note, that due to the specific washing
procedure the samples have no free (non-adsorbed) ligand or
by-products (e.g. hydrogenated DBA) as was confirmed with
3
XRD, DRIFTS, UV-Vis and P-NMR techniques. The metal
1
ꢁ
1
ꢁ1
ꢁ1
ꢁ1
content of the samples was found to be B43% (TGA).
cm , 1099 cm and 907 cm ) the vibration at 1099 cm
inducing a dipole moment orientated (for all three conforma-
tions) along the line connecting the two phosphorus atoms
Conformation analysis of diphos and diop
(
Fig. 3) is the strongest. Since this vibration is present in the
As was found from molecular modelling, the diphos molecule
can exist in several geometrical conformations (Fig. 3) at room
temperature. The presence of different conformations is con-
ditioned by free rotation around the Ph–P and P–C bonds.
At least three conformations of diphos were found with the
energy gaps DE2,1 = 0.014 eV, DE2,3 = 0.003 eV that are less
than the energy corresponding to the thermal motion of
spectrum of diphos adsorbed on Pt nanoclusters as a strong
ꢁ
1
peak (at 1106 cm ) (Fig. 5) we propose that the axis con-
necting the two phosphorus atoms is not parallel to the metal
surface, i.e. it is either parallel to the surface normal vector, or
has a significant angle of tilt, as is schematically shown in
Fig. 6.
The same approach was applied to the analysis of the
DRIFT spectra of adsorbed diop on Pt nanoclusters (Fig.
7), thus only those vibrations which do not affect vibrations of
phenyl rings were considered. Among them, the strongest
vibrations (marked in Fig. 7) were chosen and analyzed (Table
molecules (k
B
T = 0.023 eV, T = 300 K), thus these three
(
and probably more) conformations are likely to coexist,
especially in solution.
2
). As can be seen from the orientation of the corresponding
dipole moment, the main axis (line connecting both P atoms)
of the diop molecule can not exclusively have orientations
perpendicular or parallel to the normal of surface. Otherwise,
all of these vibrations (absorption peaks) will not be obser-
vable simultaneously in one spectrum. This means that the
molecule has more than one adsorption mode, or what we
think is most probable, the axis of the molecule has a
significant tilt with respect to the three coordinate axes (one
normal and two tangent vectors of the metal plane). The
adsorption mode is shown schematically in Fig. 8.
The independently established model of adsorption of diop
on Pt nanoclusters is very similar to the model of diphos
adsorption. This is not surprising because both molecules have
the same (Ph) P–X–P(Ph) structure.
2
2
Fig. 3 Conformations of diphos molecules, hydrogen atoms are not
Based on the proposed adsorption geometry, we think that
the lone pair of the phosphorus atom does not bind to the flat
shown.
8
86 | Phys. Chem. Chem. Phys., 2007, 9, 884–890
This journal is ꢀc the Owner Societies 2007

View Article Online
Fig. 4 Conformations of diop molecule. For classification of the dipole moment orientations please see the text. Hydrogens are not shown.
metal surface because of steric hindrance induced by both of
the neighbouring phenyl rings. The possible distance between
the metal surface and the phosphorus atom can be roughly
analysis and due to the larger number of atoms (making
adsorption sites) on the plane of the crystals we consider
adsorption to occur on a flat surface only.
(
without considering changes in the geometry of the adsorbed
molecule) estimated as the length of the phenyl ring (B1.4–
1
Ethyl pyruvate hydrogenation over diop and diphos modified Pt
nanoclusters
˚
.5 A, hydrogens were ignored). Based on obtained geome-
trical models of diop and diphos it is clear that due to the
almost perpendicular (with respect to each other) orientation
of phenyl rings bound to the same phosphorus atom (and the
steric hindrance of the rest of the molecule), the ring is unable
to be adsorbed in the flat mode (via p-electron system) on the
flat metal surface, at least without significant alteration of the
geometry of the adsorbed molecule. In this case, two phenyl
rings are involved in the binding with the metal in the tilted
Results of ethyl pyruvate hydrogenation over diop and diphos
modified Pt and nanoclusters are summarized in Table 3.
As can be seen from Table 3, surprisingly, diop modified Pt
nanoclusters were found to be able to induce low (4.4%) but
reproducible enantioselectivity in the ethyl pyruvate hydro-
genation reaction. Moreover, this type of catalyst demon-
strates enhancement of the specific reaction rate and thus
initial rate values by a factor of E5–7 with respect to Pt
nanoclusters with similar average sizes modified by non-chiral
(
upright) orientation (the flat mode is likely not possible, see
above). We think this adsorption mode (Fig. 6 and 8) can be
2 3
diphos. In contrast with unmodified conventional Pt/Al O ,
considered as an analog of the a-pyridyl bonded species
formed by the cleavage of the C–H bond(s) on Pt.
3
8
no conversion was found in ethyl pyruvate hydrogenation with
the same catalyst activated (400 1C, H /N , 2 h) in the
2
2
We do not exclude the adsorption of diop and diphos on the
corners and edges of the Pt nanoclusters, that likely occurs via
phosphorus bound to the Pt atom which is located at the
intersection of terraces of the nanocluster, one or two phenyl
ring(s) can be adsorbed via the p-electron system on different
terraces of the nanocluster. However, in order to facilitate
presence of free diop (1 mM), probably due to the differences
in surface chemistry (surface contaminations and external
impurities) between diop modified nanoclusters and conven-
2 3
tional Pt/Al O . Because of the same reason, we do not
compare the activity of colloidal and conventional Pt/Al
systems.
2 3
O
Fig. 6 Proposed orientation of adsorbed diphos on Pt nanoclusters
ꢁ
1
demonstrates a vibration mode at 1106 cm with a dipole moment
(big arrow) orientated with significant projection on the normal of the
surface. The most intensive displacement of atoms is shown by small
arrows.
Fig. 5 DRIFT spectra of free (1) and adsorbed on Pt nanoparticles
(2) diphos. Intensity of spectrum of free diphos is decreased for better
comparison.
This journal is ꢀc the Owner Societies 2007
Phys. Chem. Chem. Phys., 2007, 9, 884–890 | 887

View Article Online
Fig. 8 Proposed adsorption model of diop on Pt nanoclusters,
Fig. 7 DRIFT spectra of free (1) and adsorbed on Pt nanoclusters (2)
hydrogens are not shown.
diop.
demonstrate comparable reaction rates. These facts clearly
indicate the importance of the chiral nature of the modifier on
the resulting enantioselectivity. The rate enhancement effect is
ꢁ1
Table 2 Assignment of selected frequencies (cm ) in DRIFT spectra
of free and adsorbed on Pt nanoclusters diop
a
Free diop
Diop on Pt
Diop calculated
2 3
well known for the cinchonidine over conventional Pt/Al O
1
8,40
12,13,41,42
b
1376 (st)
c
systems
and colloidal Pt.
The quantum chemical
1
1
1
385 (st)
371 (st)
251 (st)
1461 z1,5c
b
1376 (st)
1236 (st)
origin of the reactivity was shown by Vargas and co-work-
43
ers, the ability to form hydrogen bonds could be an explana-
1447 ꢃ1,2
c
bc
1282 z1,6
bc
1
275 z1,5
tion for the enhancement of the reaction rate because of
stabilization of the keto carbonyl p orbitals of the substrate.
Below we provide an explanation for the enantioselectivity
of diop modified Pt nanoclusters through comparison with the
c
1
1
8
a
216 (st)
087 (st)
91 (st)
1217 (sh)
1101 (st)
885 (st)
1210 21 c, 4
1016 z1,5
c
874 21,4
Type of the vibration summarized in this table and corresponding
orientation of the dipole moment are the same for all conformations of
1
8,40,43
model developed for cinchonidine on Pt.
b
c
diop. Unambiguous assignment is difficult. Orientation of the di-
pole moment was characterized as ‘‘2’’ horizontal (in plane of the
figure), ‘‘?’’ vertical and ‘‘ꢃ’’ horizontal (perpendicular to the figure’s
plane) with numerical subscript according to the scheme in Fig. 4, left.
According to the proposed adsorption geometry of diop, the
lone pair of phosphorus is not involved in binding to the metal
surface (at least for adsorption on a flat surface), thus it may
be possible that the phosphorus atom (Fig. 9, right) can
‘
‘catch’’ the ethyl pyruvate molecule adsorbed on Pt via
1
8
It is important to note that the reaction conditions (e.g.
hydrogen pressure, solvent used) were not optimized and
enantiomeric excess could be increased by use of higher
H-bonding, as unprotonated cinchonidine does in non-ionic
solvents (toluene, THF, etc.). In both cases, the a-carbonyl
group of the ethyl pyruvate molecule can form a half-hydro-
genated state with the modifier via H-bonding, that can result
in stabilization of the keto carbonyl p orbitals of the ethyl
pyruvate and thus increase the reactivity. Being ‘‘caught’’ via
the carbonyl group, ethyl pyruvate is in an asymmetric envir-
onment produced by the OH and H groups of a modifier. For
cinchona alkaloids on Pt it is known that reversing the
positions of the OH and H groups leads to an inverse of
(
410 bar) hydrogen pressure or another solvent. The same
conventional Pt/Al O (Aldrich) demonstrated excellent activ-
3
2
4
3
ity and enantioselectivity for the same hydrogen pressure,
ethyl pyruvate and catalyst loadings, in AcOH, but with
3
9
cinchonidine as modifier.
Interestingly, (S,S)diop leads to an excess of S-lactate and
R,R)diop to R-lactate and the corresponding catalysts
(
Table 3 Results of ethyl pyruvate hydrogenation over diop and diphos modified Pt catalysts
b
Catalyst
loading mg
Reaction rate
mmol min
Initial
rate min
Conversion (%)
after time min
ee
% (dominated enantiomer)
ꢁ
1
ꢁ1
a
ꢁ1
Sample name
g
c
11 (90)
Diphos on Ptcoll
20
20
20
20
20
2
13
10
E0
0.6
1.2
10
7
E0
16
0
4.4 (S)
4.4 (R)
—
0
c
80 (300)
(
(
S,S)Diop on Ptcoll
c
50 (90)
R,R)Diop on Ptcoll
d
e
0 (240)
Pt/Al
Pt/Al
a
2
O
O
3
f
3
c
2.3 (90)
2
b
). Initial amount of
Values were estimated based on average crystal size and metal content of the samples and H
2
chemisorption data (Pt/Al
O
2 3
c
racemic ethyl lactate (as an impurity in commercial ethyl pyruvate) was taken into account. According to the curve profile (conversion vs. time)
d
the reaction can go further. Reaction was performed with presence of free 34 mg (S,S)diop (1 mM). According to the curve profile (conversion
e
f
vs. time) the reaction cannot go further. Racemic hydrogenation without a modifier.
8
88 | Phys. Chem. Chem. Phys., 2007, 9, 884–890
This journal is ꢀc the Owner Societies 2007

View Article Online
32–34
Fig. 9 Models of the half-hydrogenated state in the case of the known
[ethyl pyruvate-cinchonidine]ad complex (left lower) on Pt and
proposed [ethyl pyruvate-diop]ad complex (right lower) on Pt. Schematic positions of OH and H groups of modifiers with respect to the a-carbonyl
CQO bond of ethyl pyruvate for cinchonidine (left upper) and diop (right upper) modifier.
4
0
enantioselectivity. The same effect was found for diop
modified Pt, as (S,S)diop leads to an excess of S-lactate and
an excess of R-lactate is produced with (R,R)diop.
an alternative to known cinchonidine modified Pt systems (ee
4 90%) at least in ethyl pyruvate hydrogenation. However, it
is important from a fundamental point of view that this system
is capable of inducing enantioselectivity in principle and
reaction rate enhancement could be useful in developing a
molecular understanding of the associated phenomena and the
design of new enantioselective catalysts in the future.
It is interesting to note that, due to the geometry of the
modifiers, the OH and H groups are located significantly
closer to the ‘‘caught’’ CQO group of ethyl pyruvate in the
case of the cinchonidine molecule as compared to diop, thus
their asymmetrical influence on adsorbed substrate may be
stronger resulting in higher enantioselectivity. It is also possi-
ble that the different nature (e.g. energy) of the H–N and H–P
hydrogen bonds also influences the observed ee and reaction
rate enhancing factor.
Acknowledgements
Authors thank Mr Zhi Li (International University Bremen)
for TEM measurements. A. Kraynov and R. Richards thank
IUB for scholarship and research funds, respectively, and EU
COST D24 for support.
Conclusions
The orientations of the chiral diop and non-chiral diphos
molecules on 1.3–2 nm Pt nanoclusters were studied. It was
shown that diop and diphos demonstrate similar adsorption
modes, being attached to the flat Pt surface via the phenyl
rings and having the lone pair of phosphorus atoms unbound
with the surface.
References
1
¨
H.-U. Blaser, M. Eissen, P. Fauquex, K. Hungerbuhler, E.
Schmidt, G. Sedelmeir and M. Studer, in Asymmetric Catalysis
on Industrial Scale: Challenges Approaches and Solutions, ed. H. U.
Blaser and E. Schmidt, Wiley-VCH, Weinheim, 2004, pp. 91–103.
I. Ojima, Catalytic Asymmetric Synthesis, Wiley-VCH, New York,
2
3
2
000.
In contrast to Pt nanoclusters modified with non-chiral
diphos, diop modified Pt nanoclusters were found to be able
to induce enantioselectivity (4.4% ee) and enhanced reaction
rates by a factor of approximately 5–7 in the ethyl pyruvate
hydrogenation. Comparison with the cinchonidine on Pt
system was conducted for qualitative explanation of the ob-
served enantioselectivity and enhancement of the reaction rate.
Because of the low enantiomeric excess induced over diop
modified Pt nanoclusters this system can not be considered as
H.-U. Blaser and E. Schmidt, Asymmetric Catalysis on Industrial
Scale Challenges Approaches and Solution, Wiley-VCH, Weinheim,
2
004.
A. M. Thayer, Chem. Eng. News, 2005, 83, 40–48.
H. Bonnemann and R. Richards, in Encyclopedia of Catalysis, ed.
I. Horvath, Wiley-VCH, New York, 2003.
6 H. Bonnemann and R. M. Richards, Eur. J. Inorg. Chem., 2001,
455–2480.
4
5
¨
¨
2
7
E. Sulman, V. Matveeva, A. Usanov, Y. Kosivtsov, G. Demiden-
ko, L. Bronstein, D. Chernyshov and P. Valetsky, J. Mol. Catal. A:
Chem., 1999, 146, 265–269.
This journal is ꢀc the Owner Societies 2007
Phys. Chem. Chem. Phys., 2007, 9, 884–890 | 889

View Article Online
8
9
X. B. Zuo, H. F. Liu, D. W. Guo and X. Z. Yang, Tetrahedron,
999, 55, 7787–7804.
A. Behr, N. Doring, S. Durowiczheil, B. Ellenberg, C. Kozik, C.
28 S. Toros, L. Kollar, B. Heil and L. Marko, J. Organomet. Chem.,
1982, 232, C17–C18.
29 T. Hayashi, K. Kanehira and M. Kumada, Tetrahedron Lett.,
1981, 22, 4417–4420.
30 P. Yuan, D. Q. Wu, H. P. He and Z. Y. Lin, Appl. Surf. Sci., 2004,
227, 30–39.
31 T. O. Ely, C. Pan, C. Amiens, B. Chaudret, F. Dassenoy, P.
Lecante, M. J. Casanove, A. Mosset, M. Respaud and J. M. Broto,
J. Phys. Chem. B, 2000, 104, 695–702.
1
Lohr and H. Schmidke, Fett. Wiss. Technol., 1993, 95, 2–11.
0 R. A. T. M. Vanbenthem, H. Hiemstra, P. W. N. M. Vanleeuwen,
1
J. W. Geus and W. N. Speckamp, Angew. Chem., Int. Ed. Engl.,
1995, 34, 457–460.
1
1
1
1
1 M. Beller, H. Fischer, K. Kuhlein, C. P. Reisinger and W. A.
Herrmann, J. Organomet. Chem., 1996, 520, 257–259.
2 H. Bo
996, 35, 1992–1995.
3 H. Bonnemann and G. A. Braun, Chem.–Eur. J., 1997, 3,
200–1202.
4 S. Jansat, M. Gomez, K. Philippot, G. Muller, E. Guiu, C. Claver,
S. Castillon and B. Chaudret, J. Am. Chem. Soc., 2004, 126,
¨
nnemann and G. A. Braun, Angew. Chem., Int. Ed. Engl.,
32 K. Moseley and P. M. Maitlis, J. Chem. Soc. D, 1971, 982–983.
33 M. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A.
Robb, J. R. Cheeseman, V. G. Zakrzewski, J. A. Montgomery, Jr,
R. E. Sratmann, J. C. Burant, S. Dapprich, J. M. Millam, A. D.
Daniels, R. Cammi, B. Mennucci, C. Pomelli, C. Adamo, S.
Clifford, J. Ochterski, G. A. Petersson, P. Y. Ayala, Q. Ciu, K.
Morokuma, D. K. Malick, A. D. Rabuck, K. Raghavachari, J. B.
Foresman, J. P. Cioslowski, I. Komaromi, R. Gompets, R. L.
Martin, D. J. Fox, T. Keith, M. Al-Laham, P. M. W. Gill, B.
Johnson, W. Chen, M. W. Wong, J. L. Andres, C. Gonzalez, M.
Head-Gordon, E. S. Replogle and J. A. Pople, GAUSSIAN 03,
Gaussian Inc., Pittsburgh, 2003.
1
¨
1
1
592–1593.
5 X. B. Zuo, H. F. Liu and J. F. Tian, J. Mol. Catal. A: Chem., 2000,
57, 217–224.
6 J. Le Bars, U. Specht, J. S. Bradley and D. G. Blackmond,
Langmuir, 1999, 15, 7621–7625.
1
1
1
1
7 A. Kraynov, A. Suchopar, L. D’Souza and R. Richards, Phys.
Chem. Chem. Phys., 2006, 8, 1321–1328.
34 Swanson and Tatge, Natl. Bur. Stand. Circ. (U. S.), 1953, 539, 31.
35 P. Scherrer, Go¨ttingen Nachrichten, 1918, 2, 98.
36 G. Schmid, Endeavour, 1990, 14, 172–178.
1
1
8 A. Baiker, J. Mol. Catal. A: Chem., 2000, 163, 205–220.
9 D. Y. Murzin, P. Maki-Arvela, E. Toukoniitty and T. Salmi,
Catal. Rev. Sci. Eng., 2005, 47, 175–256.
37 R. G. Greenler, J. Chem. Phys., 1966, 44, 310.
2
2
2
0 A. Urakawa, T. Burgi and A. Baiker, Chimia, 2006, 60, 231–233.
1 M. Bieri and T. Burgi, Phys. Chem. Chem. Phys., 2006, 8, 513–520.
2 A. Urakawa, T. Burgi, H. P. Schlapfer and A. Baiker, J. Chem.
Phys., 2006, 124.
38 S. Haq and D. A. King, J. Phys. Chem., 1996, 100, 16957–16965.
39 A. Kraynov and R. Richards, Appl. Catal., A, 2006, 314, 1–8.
40 P. B. Wells and R. P. K. Wells, in Chiral Catalyst Immobilization
and Recycling, ed. D. E. De Vos, I. F. J. Vankelecom and P. A.
Jacobs, Wiley-VCH, Weinheim, 2000, pp. 123–138.
41 G. J. Hutchings, Chem. Commun., 1999, 4, 301–306.
42 A. Blaser, H. Halett, M. Mueller and M. Struder, Catal. Today,
1997, 37, 441.
23 M. Bieri and T. Burgi, J. Phys. Chem. B, 2005, 109, 22476–22485.
24 D. Ferri and T. Burgi, J. Am. Chem. Soc., 2001, 123, 12074–12084.
25 D. Ferri, T. Burgi and A. Baiker, J. Catal., 2002, 210, 160–170.
26 W. Chu, R. J. LeBlanc, C. T. Williams, J. Kubota and F. Zaera, J.
Phys. Chem. B, 2003, 107, 14365–14373.
¨
43 A. Vargas, T. Burgi, M. Von Arx, H. Reto and A. Baiker, J. Catal.,
27 T. P. Dang and H. B. Kagan, J. Chem. Soc. D, 1971, 481–481.
2002, 209, 489–500.
8
90 | Phys. Chem. Chem. Phys., 2007, 9, 884–890
This journal is ꢀc the Owner Societies 2007