Adsorption and stability of chiral modifiers based on 1-(1-naphthyl)-ethylamine for Pt catalysed heterogeneous asymmetric hydrogenations

Article abstract of DOI:10.1039/c4cy01136h

Synthetic chiral modifiers suitable for modular build-up are highly desirable for tuning the efficiency and extending the versatility of asymmetric hydrogenations on chirally-modified metal catalysts. Adsorptive anchoring and structural stability of the simple chiral modifier (R)-1-(1-naphthyl)-ethylamine [(R)-NEA] and the upgraded, secondary amine chiral modifier (R,S)-pantoylnaphthylethylamine [(R,S)-PNEA] have been investigated under catalytic hydrogenation conditions. Using attenuated total reflection-infrared (ATR-IR) spectroscopy the adsorption modes of (R)-NEA and (R,S)-PNEA at the solid-liquid interface of a technical 5 wt% Pt/Al₂O₃ catalyst were investigated. In addition to the naphthalene group, (R,S)-PNEA is also anchored to Pt through its pantoyl moiety providing both enhanced anchoring and also a better defined chiral surface site for the asymmetric hydrogenation of ketopantolactone (KPL). Factors influencing the stability of NEA-based chiral modifiers are discussed. The recently discovered chiral fragmentation product of (R,S)-PNEA, (S)-amino-4,4-dimethyl-dihydrofuran-2-one [(S)-AF] is shown to play no role in conferring enantioselectivity in the asymmetric hydrogenation of KPL.

Full text of DOI:10.1039/c4cy01136h

Catalysis
Science &
Technology
View Article Online
View Journal
PAPER
Adsorption and stability of chiral modifiers based
on 1-IJ1-naphthyl)-ethylamine for Pt catalysed
heterogeneous asymmetric hydrogenations†
Cite this: DOI: 10.1039/c4cy01136h
Fabian Meemken, a Titian Steiger,a Mareike C. Holland,b Ryan Gilmour,b
*
a
a
*
Konrad Hungerbühler and Alfons Baiker
Synthetic chiral modifiers suitable for modular build-up are highly desirable for tuning the efficiency and
extending the versatility of asymmetric hydrogenations on chirally-modified metal catalysts. Adsorptive
anchoring and structural stability of the simple chiral modifier IJR)-1-IJ1-naphthyl)-ethylamine ij(R)-NEA] and
the upgraded, secondary amine chiral modifier IJR,S)-pantoylnaphthylethylamine ij(R,S)-PNEA] have been
investigated under catalytic hydrogenation conditions. Using attenuated total reflection-infrared (ATR-IR)
spectroscopy the adsorption modes of (R)-NEA and (R,S)-PNEA at the solid–liquid interface of a technical
5 wt% Pt/Al2O3 catalyst were investigated. In addition to the naphthalene group, (R,S)-PNEA is also anchored
to Pt through its pantoyl moiety providing both enhanced anchoring and also a better defined chiral
surface site for the asymmetric hydrogenation of ketopantolactone (KPL). Factors influencing the stability
of NEA-based chiral modifiers are discussed. The recently discovered chiral fragmentation product of
IJR,S)-PNEA, IJS)-amino-4,4-dimethyl-dihydrofuran-2-one [(S)-AF] is shown to play no role in conferring
enantioselectivity in the asymmetric hydrogenation of KPL.
Received 1st September 2014,
Accepted 19th September 2014
DOI: 10.1039/c4cy01136h
www.rsc.org/catalysis
of the catalytic system has been a serious obstacle for broad-
ening the application range and, despite significant effort to
Introduction
Chirally-modified metals have attracted great interest and are
currently among the most studied heterogeneous enantio-
selective catalysts due to their impressive performance in
some specific asymmetric hydrogenations.1 Prominent exam-
ples include the tartaric acid-modified Ni2 and cinchona-
modified Pt3 and Pd4 catalysts. Simple addition of small
amounts of a strongly adsorbing chiral compound, the
so-called chiral modifier, can induce chiral amplification on
the originally achiral metal catalyst, as exemplified by the
cinchonidine-modified Pt system which provides up to 98%
enantiomeric excess (ee) in the hydrogenation of pyruvates.3,5
Unfortunately, enantioselection is highly sensitive to the
molecular structures of the prochiral substrate and the chiral
modifier.1,6,7 In addition, the catalytic performance also strongly
depends on reaction parameters like temperature, nature of
the solvent,8,9 liquid phase concentrations of reactant ketone
and modifier, and hydrogen pressure.1,6,7 The high specificity
extend the scope of substrates and modifiers, due to the
limited versatility, chirally-modified metal catalysts currently
cannot compete with the very selective homogeneous coun-
terpart for industrial application.10
A variety of structurally modified cinchona alkaloids have
been synthesized and investigated with regard to their
suitability for enantiodifferentiation on a noble metal
catalyst.11–14 Catalytic and mechanistic studies revealed three
important requirements for efficient chiral modifiers:7,15 an
anchoring group to rivet the modifier molecule to the metal
catalyst, at least one chiral centre providing the stereo-
chemical information, and an interacting moiety to guide the
prochiral substrate to the modified site. For explaining the
enantioselective transformation current mechanistic under-
standing focuses on the formation of 1 : 1 diastereomeric sur-
face complexes involving co-adsorbed modifier and prochiral
substrate.7,15–18 The stereoselection at the chirally-modified
site is either controlled by the different kinetics of the path-
ways to (R)- and (S)-product (kinetic control) or the relative
stability of adsorbed pro-IJR) and pro-IJS) surface complexes
(thermodynamic control).19 Roughly, in case thermodynamics
control the reaction, a Gibbs free energy difference of ca.
2 kcal between the pro-IJR) and pro-IJS) surface complexes
would lead at room temperature to an ee > 90%.15 As a result
of the very high substrate specificity experienced with
a ETH Zürich, Department of Chemistry and Applied Biosciences, Institute for
Chemical and Bioengineering, Hönggerberg, HCI, 8093 Zürich, Switzerland.
E-mail: fabian.meemken@chem.ethz.ch, alfons.baiker@chem.ethz.ch;
Tel: +41 44 633 25 87; Tel: +41 44 632 31 53
b Organisch-Chemisches Institut, Westfälische Wilhelms-Universität Münster,
Corrensstrasse 40, 48149 Münster, Germany
† Electronic supplementary information (ESI) available. See DOI: 10.1039/
c4cy01136h
This journal is © The Royal Society of Chemistry 2014
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
cinchona alkaloids, synthetic modifiers suitable for a modu-
lar build-up would be favourable because they allow an easier
adjustment of the chiral site to the particular molecular
structure of the prochiral substrate. The ultimate goal for a
convenient and broad application of chirally-modified metal
catalysts would therefore be a modular tool box for tailor-
made, substrate-specific chiral modifiers.
In STM studies on the co-adsorption of NEA and activated
ketones24,31 a similar attractive interaction between the
amine moiety and the ketone was revealed. In addition to the
attractive N–H–O interaction, a second attractive interaction
between the aromatic C–H of NEA and the keto carbonyl of
the prochiral substrate was identified in these studies. In
contrast to the asymmetric hydrogenations on cinchona-
modified Pt, enantioselection on NEA- or PNEA-modified Pt
catalyst is poor in the aprotic solvent toluene. Good enantio-
discrimination is only achieved in an acidic reaction medium
and best results are generally obtained using acetic acid or
toluene in the presence of small amounts of trifluoroacetic
acid (TFA). The choice of solvent influences various properties
of the modifier, e.g. the solubility, the adsorption strength,32
the stability (aromatic ring saturation and fragmentation
reactions), the conformational behaviour as well as the
protonation of the basic amine moiety. Assuming that
enantiodiscrimination of NEA-derived modifiers also
originates from a N–H–O type hydrogen bond, the impor-
tance of an acidic medium was explained by the required
protonation enabling such an interaction.14 Interestingly,
in the hydrogenation of the trifluoroactivated diketone
trifluoropentandione (TFPD) on PNEA-modified Pt, enantio-
selection is optimal in aprotic solvents like toluene or
dichloromethane.33 It was proposed that the strong H-bond
donor ability of the enol/enolate of TFPD is sufficient to form
the hydrogen bond to the modifier34 and the advantages
connected to other solvent properties, e.g. solubility or
structural stability, have to prevail for the hydrogenation of
TFPD in aprotic solvents.
Recently, another interesting factor regarding the required
acidic medium was raised by McBreen and coworkers.20 On
Pt(111) under UHV conditions, fragmentation of PNEA yielding
chiral IJS)-amino-4,4-dimethyldihydrofuran-2-one ij(S)-AF] was
observed which formed multimolecular structures with KPL.
The formation of IJS)-AF was also found under catalytic reaction
conditions using acetic acid or a mixture of toluene and TFA
as solvents, but was merely detectable in pure toluene. As
PNEA only leads to efficient enantioselection in acidic reac-
tion medium the authors proposed that aminolactones may
play an active role as chiral modifiers in the hydrogenation
of α-ketoesters.20
Herein, we investigated factors which can affect the
performance of NEA-based chiral modifiers, including their
adsorption mode, structural stability and concentration.
Using attenuated total reflection-infrared (ATR-IR) spectros-
copy we studied the adsorption and structural stability of
(R)-NEA, (R,S)-PNEA and (S)-AF on a technical 5 wt% Pt/Al2O3
catalyst under catalytically relevant conditions. The spectro-
scopic results are compared to pertinent reaction data of
catalytic hydrogenations of KPL. In this context, we also
elucidate the role of the fragmentation product of PNEA,
IJS)-amino-4,4-dimethyl-dihydrofuran-2-one ij(S)-AF], previously
observed by McBreen and coworkers and provide an explana-
tion for the improved enantioselection of PNEA-modified
compared to NEA-modified Pt.
To date, the most efficient synthetic modifiers are based
on the simple chiral molecule 1-IJ1-naphthyl)-ethylamine
(NEA), which have been shown to afford good enantio-
selectivities in the asymmetric hydrogenations of activated
α-ketoesters such as ethyl pyruvate (EP) and ketopantolactone
(KPL). In the hydrogenation of EP the in situ reductive
alkylation of (R)-NEA by the substrate ketone yielding
IJ2R,1S)-N-ij10-(1-naphthyl)ethyl]-2-amino propionic acid ethyl
ester was observed, which was found to be an efficient modi-
fier affording an ee of 58% IJR)-ethyl lactate, independent of
the chirality at the C2.13 In the hydrogenation of the cyclic
α-ketoester KPL, reductive alkylation was not observed
under reaction conditions and an ee of 52% (R)-pantolactone
was achieved with (R)-NEA. However, synthetically obtained
(R,S)-pantoylnaphthylethylamine (PNEA) yields better
enantioselection with an ee of up to 79%14 and the secondary
amine upgraded with a substrate molecule seems to form
chirally better-defined surface sites, specifically for the hydro-
genation of KPL. Note that the stereochemistry of the effi-
cient chiral modifier PNEA was recently revisited and the
earlier assignment (R,R)-PNEA14 corrected to (R,S)-PNEA.20
Modification of Pt with cinchonidine (CD) even provides
91.5% ee in the hydrogenation of the same substrate.21 Due
to the better performance of naturally occurring cinchona
alkaloids the Pt-cinchona system has been most frequently
studied. However, a better mechanistic understanding of
NEA derived modifiers could be rewarding as the molecular
structure of these modifiers can be more systematically
altered.
Even though the enantioselectivities deviate, mechanistic
similarities for the chiral modification with the primary
(NEA), secondary (PNEA) and tertiary (CD) amine modifiers
are expected. Compared to the quinoline anchor of CD, NEA
and PNEA possess a naphthalene group and the adsorption
of NEA via the aromatic ring system has been evidenced on
single crystal Pt surfaces under UHV conditions22–24 and at
the solid–liquid interface from CCl4 solution.25,26 While the
enantiomer NEA has only one chiral centre, the diastereomer
PNEA has two and CD even five chiral centres, of which par-
ticularly the configuration at the C8-C9 chiral alcohol linker
has been shown to be crucial for enantiodifferentation.27
Regarding the enantiodifferentiating surface interaction
between co-adsorbed modifier and ketone, an intermolecular
N–H–O hydrogen bridge formed between the N atom of the
quinuclidine moiety and the O atom of the carbonyl was
identified for the cinchona system28 and later verified under
reaction conditions.29 In aprotic solvent chemisorbed hydrogen
was shown theoretically30 and experimentally29 to serve as the
proton linker for the intermolecular surface interaction.
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2014

View Article Online
Catalysis Science & Technology
Paper
solution stirred for 1 d. H2O was added in portions until no
more white precipitate was formed. The mixture was filtered
over a pad of celite with EtOH and concentrated in vacuo.
Purification by CC (1 : 20 → 1 : 10 EtOAc/cyclohexane) gave
the (R,S)-diastereomer as a yellow oil (373 mg, 13%).
Experimental
Materials
Dihydro-4,4-dimethyl-2,3-furandione (ketopantolactone), KPL
(Sigma-Aldrich, 97%), IJR)-1IJ1-naphthyl)-ethylamine, (R)-NEA
or NEA (Sigma-Aldrich, ≥99%), trifluoroacetic acid, TFA
(Fluka, 98%), toluene (Acros, 99.85%), acetic acid (Merck,
99–100%), ethanol (Fluka, ≥99.8%), and trifluoromethyl
benzene (Acros, 99 + %) were used as received.
Rf = 0.33 (EtOAc/cyclohexane 1 : 8); 1H NMR (300 MHz,
CDCl3): δ = 8.15 (1H, d, J = 8.1), 7.89 (1H, d, J = 7.5), 7.78
(1H, d, J = 8.1), 7.66 (1H, d, J = 7.0), 7.60–7.43 (3H, m), 4.89
(1H, q, J = 6.5), 3.89 (1H, d, J = 8.8), 3.70 (1H, d, J = 8.8), 3.23
(1H, s), 2.16 (1H, s), 1.51 (3H, d, J = 6.6), 1.11 (3H, s), and
1.09 (3H, s) ppm; HR-ESI-MS: m/z: 284.1648 ([M + H]+, calcd
for C18H22NO2+: 284.1645).
The
commercial
5
wt%
Pt/γ-Al2O3
catalyst15
(BASF-Engelhard 4759; platinum dispersion, 0.22, BET
surface area, 168 m2 g−1) was reduced in a fixed-bed tubular
reactor under flowing hydrogen for 1 h at 673 K.
(S)-AF was prepared according to the literature procedure:35
To D-IJ−)-pantolactone (2.50 g, 19.21 mmol, 1.00 eq.) in
CH2Cl2 (10.00 ml) and pyridine (1.93 ml) was added Tf2O
(4.04 mL, 24.01 mmol, 1.25 eq.) at −78 °C under an
atmosphere of argon drop-wise over 10 min. After 10 min the
reaction was allowed to come to RT and stirred for 1 h to give
a thick pink solution (after 20 min CH2Cl2 (10.00 mL) was
added). The solvent was evaporated in vacuo and the residue
dissolved in a mixture of brine and Et2O. The organic
layer was washed twice with NaHCO3 and once with brine,
dried over MgSO4 and concentrated in vacuo to give
IJR)-4,4-dimethyl-2-oxotetrahydrofuran-3-yl trifluoromethane-
sulfonate as a yellow oil (4.86 g, 96%).
For the synthesis of (R,S)-PNEA and (S)-AF, all chemicals
were purchased as reagent grade and used without further
purification. Solvents for purification (extraction and chroma-
tography) were purchased as technical grade and distilled
on the rotary evaporator prior to use. All reactions using air
or moisture sensitive compounds were carried out in
flame-dried and evacuated glassware under an atmosphere of
argon. Solvents for these reactions were dried according to
standard procedures or were taken from the solvent drying
system. For column chromatography SiO2-60 (230–400 mesh
ASTM; Fluka) was used as stationary phase. Analytical thin
layer chromatography (TLC) was performed on glass plates
pre-coated with SiO2-60 F254 (Merck) and visualized with a
UV-lamp (254 nm) or by dipping in cerium ammonium
molybdate (CAM) stain or KMnO4 stain, followed by heating.
Flash column chromatography was carried out on Fluka silica
gel 60 (230-400 mesh). Concentration in vacuo was performed
at ~10 mbar and 40 °C, drying at ~10−2 mbar and RT.
NMR spectra were measured by the NMR service of the
Organisch-Chemisches Institut, Westfälische Wilhelms-
Universität Münster on a Bruker AV300 or an Agilent DD2 600
spectrometer at ambient temperature. 1H NMR spectra are
reported as follows: chemical shift δ in ppm (number of pro-
tons, multiplicity, coupling constant J in Hz). The deuterated
solvent residual peak was used as internal reference: CHCl3
(δH 7.260). The resonance multiplicity is abbreviated as: s
(singlet), d (doublet), q (quadruplet), m (multiplet) and b
(broad). Optical rotations were measured on a JASCO P-2000
polarimeter with 1 dm cell length and λ = 589 nm (Na D-line),
concentrations are given as mg mL−1. High-resolution
mass spectra (HR ESI MS) were measured by the MS
service of the Organisch-Chemisches Institut, Westfälische
Wilhelms-Universität Münster.
1
ijα]D20: 11.3 (c = 1.79, CH2Cl2); H NMR (300 MHz, CDCl3):
δ = 5.08 (1H, s), 4.14 (1H, d, J = 9.2), 4.07 (1H, dd, J = 9.2,
0.4), 1.31 (3H, s), and 1.22 (3H, s) ppm.
To IJR)-4,4-dimethyl-2-oxotetrahydrofuran-3-yl trifluoromethane-
sulfonate (1.80 g, 6.86 mmol, 1.00 eq.) in DMF (10 ml) was
added NaN3 (0.51 g, 7.89 mmol, 1.15 eq.) under an atmo-
sphere of argon at RT and stirred for 4.5 h. The reaction was
poured into H2O (50 mL) and extracted with Et2O (80 mL).
The organic layer was washed three times with brine (each
50 ml), dried over MgSO4 and concentrated in vacuo to
approx. 20 ml. MeOH (25 mL) was evaporated and the
remaining Et2O evaporated in vacuo. An atmosphere of argon
was created, HCl in MeOH (1.25 N, 6.60 ml, 8.22 mmol, 1.20 eq.)
and Pd/C (5%, 36 mg, 2 wt%) were added and then an
atmosphere of H2 was created. The reaction was stirred for
53 h, filtered over celite and concentrated in vacuo to give a
yellow oil. The oil was redissolved in a mixture of NaHCO3
and CH2Cl2, the aqueous layer extracted twice more with
CH2Cl2, the combined organic layers dried over MgSO4 and
concentrated in vacuo, the residue was filtered over a pad of
silica with 1 : 20 MeOH/CH2Cl2 to give the product as a yellow
oil (303 mg, 34%, two steps).
(R,S)-PNEA was prepared according to the literature
procedure:20
Rf = 0.31 IJMeOH/CH2Cl2 1 : 10); ijα]D20: −10.1 (c = 1.02,
1
To dihydro-4,4-dimethyl-2,3-furandione (KPL, 1.28 g,
10.0 mmol, 1.25 eq.) in TiIJOiPr)4 (4.00 ml) was added
IJ1R)-1-IJ1-naphthyl)ethanamine (1.37 g, 8.0 mmol, 1.00 eq.)
under an atmosphere of argon at RT resulting in a viscous
yellow oil. After 4 h stirring, NaBH3CN (0.34 g, 5.4 mmol,
0.68 eq.) and EtOH (12.00 mL) were added. After another
16 h, a second portion of NaBH3CN (0.34 g, 5.4 mmol,
0.68 eq.) and acetic acid (0.10 mL) were added and the
CH3OH); H NMR (300 MHz, CDCl3): δ = 3.99 (1H, d, J = 8.8),
3.90 (1H, d, J = 8.9), 3.31 (1H, s), 1.57 (2H, b), 1.18 (3H, s),
and 1.00 (3H, s) ppm; HR-ESI-MS: m/z: 130.0854 ([M + H]+,
calcd for C6H12NO2+: 130.0863).
Catalytic hydrogenations
The 5 wt% Pt/Al2O3 catalyst was pre-reduced in a fixed-bed
tubular reactor under flowing hydrogen for 1 h at 673 K,
This journal is © The Royal Society of Chemistry 2014
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
subsequently cooled down to room temperature for
30 min under flowing hydrogen and kept under flowing
nitrogen before transfer to the autoclave. The characteristics
for the as-pretreated catalyst are a platinum dispersion
of 0.22 and a mean platinum particle size of 4.5 nm.15
Hydrogenations of KPL were performed in a 50 mL
three-necked glass reactor at room temperature and at
atmospheric pressure under constant flow of H2 with a volu-
metric flow rate of 10 mL min−1 (semi-batch hydrogenation).
The third opening of the reactor was sealed with a
septum, which allowed for sample taking and addition of
solutions containing modifier and ketone by the use of a
syringe. The stirring rate was set to 500 rpm (no effect on
the kinetics and enantioselectivities are experienced
under such conditions according to the comprehensive
study by Garland et al. covering potential internal and
external mass transfer limitations in the much faster enantio-
selective hydrogenation of ethyl pyruvate at 100 bar).36
The general reaction procedure was the following: the
pre-reduced catalyst was transferred to the reactor und
reduced again in situ in 5 ml solvent under constant H2
flow for 1 h. Then, the reaction was initiated by addition of
modifier and KPL premixed in 5 mL solvent. The samples
were analyzed by gas chromatography using a chiral capillary
column (CP-Chirasil-Dex CB, 25 m length, 0.25 mm internal
diameter, 0.25 μm film thickness). The following reaction
conditions were used for all hydrogenations unless stated
otherwise: 10 0.1 mg Pt/Al2O3, 50 mM (0.5 mmol) KPL,
0.056 mM (modifier : Pts 1 : 1) or 1 mM modifier (modifier : Pts
18 : 1), 10 ml solvent, room temperature, hydrogen flow,
mechanical stirring (500 rpm).
Results
Hydrogenation on (R)-NEA-modified Pt
To study the stability of the chirally-modified Pt catalyst with
regard to enantioselection the catalytic system was monitored
in 15 min intervals during the hydrogenation of KPL on
(R)-NEA-modified Pt/Al2O3 at different modifier concentra-
tions. In Fig. 1, conversion and ee for the hydrogenation of
KPL in acetic acid (A) and in toluene in the presence of
trifluoroacetic acid (TFA) (B) are shown. In acetic acid, the
best enantioselection with an ee of 35% was observed at low
KPL conversion and at a molar modifier ratios of 1 : 1 with
respect to platinum surface atoms (modifier : Pts ratio).
Higher NEA concentration had a positive effect on the
catalytic activity but the excess of chiral modifier did not
influence enantioselectivity. As the asymmetric hydro-
genation was less efficient (27% ee, 8% Δee throughout the
hydrogenation) when using molar excess of platinum surface
In situ ATR-IR spectroscopy
ATR-IR spectra were recorded on a Bruker 70v spectrometer
equipped with a liquid nitrogen cooled MCT detector. A
home-build stainless steel flow-through cell mounted on an
ATR-IR attachment (OPTISPEC) was used for all the adsorp-
tion–desorption measurements. Spectra were recorded at
4 cm−1 resolution. The measurement temperature (298 K) was
regulated by a thermostat (Julabo, F25).
The home-build flow-through cell and its implemen-
tation into a recycle reactor has been described in detail
elsewhere.37 Briefly, the spectroscopic reactor cell (0.115 mL
void volume) was implemented into a recycle loop (2.885 mL
volume) in which the consumed hydrogen was continu-
ously replaced mimicking a semi-batch reactor set-up.
The powder catalyst was deposited onto a ZnSe internal
reflection element according to a slurry-drop deposition
procedure creating
5
1.0 mg catalyst layers. The
reservoirs holding H2-saturated solutions were connected to
two-position-four-way valves (VICI, C22Z-2184EH) allowing
smooth exchange of the recycle reactor content without expo-
sure to air. Solutions were admitted by using a micro annular
gear pump (HNP, mzr-6335) installed within the recycle
reactor.
Fig. 1 Hydrogenation of KPL on NEA-modified Pt/Al2O3 in (A) acetic
acid and in (B) toluene in the presence of 20 eq. TFA (mol TFA/mol
NEA) at different initial NEA concentrations.
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2014

View Article Online
Catalysis Science & Technology
Paper
atoms with respect to the modifier concentration (0.1 : 1 NEA : Pts),
it appears that roughly equimolar modifier : Pts ratios are
required to achieve optimal enantioselection in the hydro-
genation of KPL at atmospheric hydrogen pressure. Never-
theless, for all three modifier concentrations the ee slightly
dropped with increasing ketone conversion. Interestingly,
when NEA has been added to the reactor 2 h prior to KPL,
the ee was significantly decreased. This is an indication that
NEA undergoes some changes and partially loses its enantio-
differentiating ability under reaction conditions.
Comparing the two different solvent systems, the ee is
approximately 20% higher in acetic acid (Fig. 1A) compared
to toluene/TFA (Fig. 1B). Exceeding a certain threshold
concentration (approximately 1 : 1 NEA : Pts), the ee on
NEA-modified Pt appears to be insensitive to the absolute
concentration of the modifier (modifier : Pts ratio) and to the
acidity of the reaction mixture (20 mM TFA compared to
1.1 mM TFA, pKa = 0.23). However, comparison of the
results obtained with simultaneous addition of KPL and
NEA and with KPL added to pre-modified Pt after 2 h indi-
cates that the modifier undergoes some structural changes
when exposed to hydrogenation conditions thereby losing
some of its ability of enantiodifferentiation.
Fig. 2 Adsorption of (R)-NEA onto Pt/Al2O3 (5.1
0.1 mg) from
H2-saturated toluene solutions containing 0.3 μmol (0.1 mM) or
3.0 μmol (1.0 mM) NEA and desorption under H2-saturated toluene
flow. Absorption is given in mA.
To gain deeper insight into the anchoring of NEA at the
catalytic solid–liquid interface we studied its adsorption on
Pt/Al2O3 from H2-saturated toluene solution using ATR-IR
spectroscopy. The ATR-IR flow-through reactor cell was
placed in a recycle reactor with a fixed total volume of 3 mL.
additional uptake from the liquid phase (Fig. 2, red and blue
spectra). At liquid concentrations of 0.1 mM and 1.0 mM the
modifier : Pts ratio is 1 : 1 and 10 : 1, respectively. Flushing the
catalyst with H2-saturated toluene for 50 min in flow-through
mode (Fig. 2, black spectrum) does not lead to significant
removal of NEA and mainly accumulated water, detected at
around 1640 cm−1, is desorbed. Therefore, it can be inferred
that the Pt/Al2O3 catalyst can strongly adsorb the modifier
exceeding a molar ratio of at least 1 : 1 (NEA : Pts). It is likely
that the acidic and basic sites of the alumina support also
interact with the amine. However, as evident from the cata-
lytic experiments a molar ratio of 1 : 1 is sufficient to obtain
optimal enantioselection.
After the in situ reduction of the catalyst (5.1
0.1 mg,
~0.3 μmol Pts), 0.3 μmol NEA dissolved in toluene (3 mL of
0.1 mM solution) were admitted to the reactor system and
recycled for 30 min. In Fig. 2, the purple spectra show the
adsorption of (R)-NEA on the catalyst surface within the first
10 min in 2 min intervals and after 30 min (from bottom to
top). The broad IR-absorption at 1500–1600 cm−1 is indicative
for the aromatic ring stretching modes of the naphthalene
ring interacting with Pt. Compared to the liquid phase spec-
trum with well-defined absorption modes for dipole moment
changes of the aromatic ring along its short axis (1597 cm−1
)
and along its long axis (1510 cm−1) the absorption detected
from NEA adsorbed on the 5 wt% Pt/Al2O3 catalyst is
considerably broadened indicating the detection of several
coexisting adsorption geometries with (slightly) varying
adsorption strength and rather vaguely defined orientation of
the naphthalene ring.
Hydrogenation on IJR,S)-PNEA-modified Pt
As pointed out in the introduction, chiral modification of the
Pt catalyst with (R,S)-pantoylnaphthylethylamine (PNEA)
improves the ee considerably. The results from the catalytic
hydrogenations of KPL using molar PNEA : Pts-ratios of 18 : 1
and 1 : 1 are shown in Fig. 3. In the hydrogenation with acetic
acid as solvent (Fig. 3A) the ee decreased by 3–4% during the
reaction. The experiment on the PNEA-premodified catalyst
also indicates a slow decrease in enantiodifferentiation with
reaction time due to partial fragmentation of PNEA. Small
amounts of (S)-AF could be detected by GC analysis of the
liquid reaction solution. In toluene/TFA the catalytic behav-
iour appears to be more complex. Similar to the hydrogena-
tion on NEA-modified Pt, addition of TFA decreased the
catalytic activity for CO bond reduction. However, enantio-
selection decreased during the course of the reaction when
Additional, slightly broadened adsorption modes assigned
to the aromatic ring system such as the νipIJCC) at 1398 cm−1
,
δipIJCH) at 1260 cm−1 and δoopIJCH) at 1167 cm−1 corroborate
the broadly defined adsorption of the naphthalene ring. How-
ever, as the adsorption modes along the short and long axes
of the aromatic ring system are both visible the naphthalene
moiety of the majority of the accumulated species must be
tilted along both axes with respect to the metal surface
(Greenler's surface selection rule38).
Refilling the recycle reactor after 30 and 60 min with fresh
0.1 mM and 1.0 mM NEA solution, respectively, reveals
This journal is © The Royal Society of Chemistry 2014
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
Fig. 4 Adsorption of (R,S)-PNEA onto Pt/Al2O3 (6.5 0.1 mg) from
H2-saturated toluene solutions containing 0.3 μmol (0.1 mM) or
3.0 μmol (1.0 mM) PNEA and desorption under H2-saturated toluene
flow. Absorption is given in mA.
PNEA is not tilted along the short axis leading to the cancel-
ation of this dipole moment change.38
The orientation of the pantoyl group of adsorbed PNEA
can be deduced from the CO carbonyl stretch detected at
1765 cm−1. Opposed to the liquid phase spectrum only very
low carbonyl absorption is detected and the strongest absorp-
tion of adsorbed PNEA originates from the vibrational mode
assigned to the dipole moment changes of the naphthalene
ring along the short axis (~1595 cm−1). As the intensity of the
carbonyl stretch is significantly attenuated, this bond and
therefore, due to the rigidity of the ring, the pantoyl group
must be very close to parallel to the metal surface. The
detected surface species of PNEA also strongly resisted
solvent flushing (Fig. 4) and, again, mainly water could be
removed. While in the catalytic hydrogenations best results
were obtained in acetic reaction medium, the IR-studies on
the adsorption of the modifiers had to be carried out in tolu-
ene. Due to the strong IR-absorption acetic acid is not a suit-
able solvent for such studies. Instead, we further investigated
the co-adsorption of PNEA and TFA.
After the saturation of the Pt/Al2O3-catalyst with a H2-saturated
toluene solution containing 1.0 mM PNEA (red spectrum), the
recycle reactor content was exchanged with a solution containing
0.2 mM TFA. The co-adsorption of the carboxylic acid on the
PNEA-premodified catalyst is shown in Fig. 5 (green spectrum).
Besides absorption modes for the symmetric and asymmetric C–F
stretching modes (1205 cm−1 and 1146 cm−1), the accumulation
of TFA is evident by the formation of carboxylate species detected
by the asymmetric and symmetric carboxylate stretches at
1665 cm−1 and 1440 cm−1 , respectively. In a toluene solution
the carboxylate stretches are found at the same positions
Fig. 3 Hydrogenation of KPL on PNEA-modified Pt/Al2O3 in (A) acetic
acid and in (B) toluene in the presence of 20 eq. TFA (mol TFA/mol
PNEA) at two different initial PNEA concentrations.
using PNEA and the declining ee was even more pronounced
at 20 mM acid concentration. In control experiments with 1
and 10 eq. TFA a lower ee was observed after 2 h reaction
time (at PNEA : Pts = 3.6 : 1, 35 and 41% ee, respectively).
The observed performances of NEA- and PNEA-modified
Pt deviate in several aspects. The catalytic activity and
enantioselectivity was always better on the PNEA-modified
catalyst. Furthermore, the secondary amine was more resis-
tant to decomposition under hydrogenation conditions but
the sensitivity with respect to the acid concentration was
stronger.
Attempting to understand the origin of these differences
the adsorption on Pt/Al2O3 of 0.3 μmol PNEA dissolved in tol-
uene (3 mL of 0.1 mM solution) was monitored for 30 min
and the solution was refreshed twice (Fig. 4). In contrast to
the adsorption of the smaller molecule NEA, in the adsorp-
tion of PNEA the vibrational mode oriented along the long
axis of the naphthalene anchor (1510 cm−1) is significantly
attenuated. This is an indication that the aromatic ring of
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2014

View Article Online
Catalysis Science & Technology
Paper
Fig. 5 Adsorption of TFA onto PNEA-premodified Pt/Al2O3 (5.9 0.1 mg)
from H2-saturated toluene solution containing 0.6 μmol (0.2 mM) TFA and
desorption under H2-saturated toluene flow. Absorption is given in mA.
Fig. 6 Adsorption of (S)-AF onto Pt/Al2O3 (5.5
0.1 mg) from
confirming the acid–base interaction between TFA and pre-
adsorbed PNEA. Furthermore, the co-adsorption of TFA did
not seem to compete with adsorbed PNEA (compare also the
C–H stretching modes at 2960, 2930 and 2850 cm−1) under
the chosen conditions and only induced the removal of
adsorbed water (3390 cm−1). During recycling of the TFA
containing solution under hydrogenation conditions, no sig-
nificant spectral changes could be observed. Flushing of the
catalyst with H2-saturated toluene (blue spectrum) mainly led
to the removal of surface water (3390 and 1640 cm−1). Due to
the strong overlap of the water absorption at around
1640 cm−1, an unambiguous interpretation is cumbersome
but the detected absorption modes assigned to PNEA
(1565 cm−1) and to the carboxylate (1440 cm−1) remained nearly
constant indicating the detection of stable surface species.
H2-saturated toluene solutions containing 0.3 μmol (0.1 mM) or
3.0 μmol (1.0 mM) AL and desorption under H2-saturated toluene flow.
Absorption is given in mA.
was quicker reaching equilibrium coverage within almost
10 min. Furthermore, the orientation of the adsorbed amino-
lactone ring with respect to the metal surface appears to be
fundamentally different. Note that the comparison of the
vibrational spectra is based on the same molecular uptake
and deviations in intensity of the same vibrational mode
must arise due to different adsorption geometries (enhance-
ment/cancelation due to the paramagnetic metal surface38).
Comparing the spectra of the PNEA adsorption in Fig. 4 and
of the AF adsorption in Fig. 6 the different orientation of the
carbonyl group is evident. The smaller molecule AF adopts
an adsorption geometry with the ester carbonyl tilted with
respect to the Pt surface. Most likely its adsorption is con-
trolled by the primary amine functionality. This conjecture is
supported by the absence of the NH2 deformation modes in
the surface spectra which are centred at around 1650 cm−1 in
the spectra of dissolved AF. The rather strong adsorption of
the aminolactone on Pt/Al2O3 can be seen in the desorption
spectra in Fig. 6 which were acquired after solvent flushing
for 30 min.
Hydrogenation on (S)-AF-modified Pt
The formation of (S)-AF due to the C–N bond scission of
PNEA on Pt(111) has been previously observed by STM20
under UHV conditions and an active role of aminolactones as
chiral modifiers in the asymmetric hydrogenation of KPL has
been proposed.
For proper identification of the active chiral site on PNEA-
modified Pt the role of (S)-AF also needs to be addressed.
The adsorption of (S)-AF from H2-saturated toluene solution
is shown in Fig. 6. The purple spectra show the uptake of
(S)-AF during the first 10 min in 2 min intervals and after
20 min (from bottom to top). Following the adsorption by the
carbonyl absorption band at 1765 cm−1, additional exposure
of 0.3 μmol (S)-AF during the second recycling nearly doubled
the intensity (red spectrum). Compared to the adsorption of
the NEA-based modifiers the adsorption of the aminolactone
According to McBreen and coworkers, the C–N bond scis-
sion occurs under reaction conditions (10 bar H2, Pt/Al2O3)
in acetic acid and also in toluene, but only in the presence of
TFA.20 We also found (S)-AF in the liquid reaction solution
(compare results to Fig. 3) but could not observe the decom-
position of PNEA nor the accumulation of AF in the surface
spectra (compare Fig. 5). Therefore, the co-adsorption of
PNEA and AF was investigated.
This journal is © The Royal Society of Chemistry 2014
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
In Fig. 7a, the saturation of the Pt/Al2O3-catalyst with
with the 1.0 mM PNEA solution, the recycle reactor was filled
with a toluene solution containing 1.0 mM PNEA and
0.1 mM AL, corresponding to a PNEA conversion of 9%.
In Fig. 7b, the surface spectra show the spectral changes
induced by the presence of the lactone.
1.0 mM PNEA solution is shown. Refreshing the recycle
reactor content for the third time did not lead to any further
increase of the absorption modes assigned to PNEA (red and
blue spectra in Fig. 7a). For the sake of comparison, the
spectrum of AF adsorbed onto a fresh catalyst in a separate
experiment is also shown (green dotted spectrum). Due to
the similar molecular structures the characteristic vibrations
of PNEA and AF feature nearly at the same wavenumbers.
Unfortunately, AF also shows a vibrational feature at 1595
and 1260 cm−1 when adsorbed onto the catalyst and merely
the absorption at 1565 cm−1 of the adsorbed naphthalene
ring of PNEA is not overlapping entirely. Therefore, the analy-
sis of the competitive adsorption requires careful comparison
of absorption intensities, particularly at the intensive car-
bonyl stretch at 1765 cm−1. After the saturation of the catalyst
The removal of PNEA could not be observed as absorption
bands only increased. However, particularly the increasing
absorption at 1765 cm−1 indicates additional uptake of AF.
Quick exchange of the recycle reactor content with a solution
containing an equimolar amount of AF (1 mM PNEA and
1 mM AF) led to further increase of the carbonyl absorption
as well as the absorption at 1595 cm−1, which can only origi-
nate from additional AF molecules adsorbing onto the cata-
lyst. However, continuous monitoring of the competitive
adsorption of the two chiral molecules revealed information
about the adsorption dynamics. Following the admission of
the equimolar solution absorption of the carbonyl mode
reached its maximum after 10 min (black spectrum) to
decrease thereafter. After an additional recycling for 20 min
(blue spectrum), the broad band assigned to adsorbed water
(~1640 cm−1) had grown on the expense of the species associ-
ated with the carbonyl absorption band.
Opening the recycle reactor and flushing with H2-saturated
toluene for 30 min (green spectrum in Fig. 7c) led to the
gradual decrease of the carbonyl absorption reaching nearly
the same level as before the admission of the equimolar mix-
tures (orange spectrum) but the absorption of the adsorbed
naphthalene ring is discernible at 1565 cm−1.
The surface spectra of the competitive adsorption seem to
indicate the following scenario: pre-adsorbed bulky PNEA is
strongly bound to the catalyst with its naphthalene and
pantoyl moieties occupying rather large and specific Pt
sites. The smaller chiral molecule AF (nearly the same struc-
ture as the substrate ketone KPL and the alcohol product
pantolactone) is able to occupy remaining surface sites and
might also cover chiral surface sites imparted by adsorbed
PNEA. But, comparing the competitive adsorption with the
adsorption of pure substances, the surface coverage of AF
was significantly lower, particularly obvious after solvent
flushing.
The co-adsorption dynamics monitored during the 30 min
recycling also revealed weaker adsorption of AF on the
PNEA-pre-modified catalyst, as a gradual accumulation of
water led to the removal of AF. In the adsorption experiment
of pure AF (Fig. 6) no such removal was observed.
Nevertheless, to unambiguously answer the question
about the role of AF in the hydrogenation of KPL, catalytic
experiments with addition of AF were performed. As the high
sensitivity of NEA-based modifiers on the nature of the
solvent is known14 the hydrogenations on AF-modified
catalyst were carried out in ethanol, toluene, acetic acid and
trifluoromethyl benzene. Independent of the solvent
used, addition of 1 mM AF (AF : Pts = 18 : 1) or of 0.056 mM
(AF : Pts = 1 : 1) did not induce any enantioselection in the
hydrogenation of KPL at atmospheric hydrogen pressure. The
KPL conversions shown in Fig. 8 even indicate that AF acts as
Fig. 7 (a) Adsorption of (S)-AF on (R,S)-PNEA modified Pt/Al2O3 (4.8
0.1 mg) from H2-saturated toluene solutions containing 3.0 μmol
(1.0 mM) PNEA and co-adsorption adding (b) 0.1 mM AF and
(c) 1.0 mM AF to the feed. For the sake of comparison the spectrum of
adsorbed AF is included in (a). Absorption is given in mA.
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2014

View Article Online
Catalysis Science & Technology
Paper
Protic solvents can protonate the quinuclidine N atom of
cinchona alkaloids.15 High hydrogen pressure can cause satu-
ration of the aromatic quinoline ring of cinchona alkaloids
that is crucial for anchoring of the cinchona modifier on the
metal surface.41 Similar effects can also be expected for syn-
thetically prepared modifiers due to their closely related
structure. While some studies uncovered factors, which
impair the proper functioning of cinchona alkaloids as chiral
modifiers, such investigations are lacking for synthetic modi-
fiers. In the present study we have investigated factors
influencing the structural stability of the most relevant syn-
thetic modifiers NEA and PNEA, and the role of the recently
discovered fragmentation product of the latter, AF. The
ATR-IR investigation revealed significant differences in the
adsorption behaviour of the three related chiral molecules.
Catalytic tests using the asymmetric hydrogenation of KPL
showed how differences in the structural properties affect the
performance of the modifiers, and revealed that (S)-AF does
not contribute to enantiodifferentiation.
Chiral modification of the catalyst with the bulkier sec-
ondary amine PNEA generally leads to higher ee. The adsorp-
tion of PNEA seems to be directed by both the naphthalene
and the pantoyl group, and a distinct, strongly adsorbed sur-
face species was observed in the IR spectra (Scheme 1A). In
contrast, the naphthalene group of adsorbed NEA was more
tilted on the surface. The higher enantioselectivity of PNEA
could be traced back to the different anchoring of the two
chiral modifiers. Adsorbed to Pt the naphthalene ring of
PNEA is closer to the surface similar to the adsorption of
CD.42 The presence of the pantoyl group seems to stabilize
the aromatic ring in a nearly parallel position and supports
the strong anchoring of the chiral molecule. Interestingly,
the naphthalene group also helps to stabilize the nearly
Fig. 8 Hydrogenation of KPL on Pt/Al2O3 in different solvents in
presence of (A) 0.0546 mM and (B) 1.0 mM (S)-AF.
a catalyst poison for the CO bond reduction after a certain
threshold concentration. Following the hydrogenation in tol-
uene without addition of AF as a reference, addition of very
small amounts of the aminolactone (Fig. 8A) led to slightly
faster conversion of KPL but at higher aminolactone concen-
tration (Fig. 8B) the catalytic activity generally decreased in
all solvents used.
Discussion
Suitable adsorptive anchoring as well as structural stability
are necessary prerequisites for the proper functioning of
chiral modifiers in asymmetric hydrogenations on chirally-
modified metal catalysts. Various factors such as choice of
solvent and hydrogen pressure can lead to unfavourable
structural changes as uncovered for the well-investigated
cinchona modifiers. Depending on the reaction conditions
conformational changes of CD can be induced which
affect the enantiodifferentiation of cinchona alkaloids.9,39,40
Scheme 1 Proposed adsorption of (R,S)-PNEA on Pt (A) and the
structure of the pro-IJR) KPL-PNEA surface complex (B).
This journal is © The Royal Society of Chemistry 2014
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
parallel adsorption of the pantoyl group as the aminolactone
ring alone adsorbs in a more upright position. Despite the
improved anchoring, in such an adsorption geometry the chi-
ral pocket formed by PNEA is also chirally better defined for
the enantiodiscrimination toward the (R)-alcohol. Docking of
KPL with the pro-IJS) adsorption face is sterically and electron-
ically hindered by the proximity of the carbonyl group of the
adsorbed pantoyl moiety (Scheme 1B).
role. As indicated by the arrows in Scheme 2 the performance
of a chiral modifier hinges on the careful tuning of the inter-
dependent properties: stability, protonation, conformation,
and anchoring, and an isolated consideration of a single
property of the chiral modifier might be misleading. The
extended molecular structure of PNEA compared to NEA
improves the anchoring of the modifier on the surface and
introduction of the pantoyl group sterically and electronically
limits the stereochemical flexibility of the chirally-modified
site. Regarding the stability the altered adsorption geometry
of the naphthalene anchor also seems to stabilize the modi-
fier against structural changes induced by saturation or par-
tial fragmentation providing better turnover-numbers at the
intact chirally-modified site. A drawback of the secondary
amine PNEA is its susceptibility for fragmentation leading
eventually to active site blocking by AF formation.
The necessary protonation requiring an additional acid
source is another challenging disadvantage. Faster saturation
of the aromatic ring anchor of CD was observed in acetic acid
as compared to in toluene,41 which is also expected for a
naphthalene group. Furthermore, the fragmentation of PNEA
is more prominent at higher acidity. In the asymmetric
hydrogenation of TFPD, which can itself function as a proton
donor, beneficial effects of aprotic solvents like toluene or
dichloromethane emerge for the Pt-PNEA system.33,34 In the
hydrogenation on CD-modified catalyst, the tertiary amine
moiety, quinuclidine, is able to pick up chemisorbed hydrogen
and efficiently form the N–H–O hydrogen bridge to ketones
even in aprotic solvents.29,30 Higher basicity of the interacting
moiety could be valuable for enabling strong interaction to
the prochiral ketone, even in a mild reaction environment.
The last feature, the conformational behaviour of the chi-
ral modifier, has not been studied for NEA-based modifiers
so far, but its strong influence on the enantiodifferentiating
ability of the structurally more complex CD has been
evidenced in an earlier study.9 The conformation which can
be altered by the choice of the solvent (polarity, aprotic,
protic) and the temperature may sterically influence the
adsorption geometry as exemplarily shown for CD for which
open conformers are favoured for the interaction between
adsorbed CD and prochiral substrates.15,39
Regarding the sensitivity with respect to the reaction con-
ditions, up to a certain concentration higher acidity of the
reaction medium had a detrimental effect on the enantio-
selection, which was clearly observed in the hydrogenation in
toluene/TFA (compare Fig. 1B and 3B). The effect of the acid
on PNEA-modified catalyst seems to be two-fold. On the one
hand, protonation of the N atom of the amine is required to
enable the N+–H–O interaction to the carbonyl of the ketone.
In the surface spectra of the co-adsorption of PNEA and TFA,
in small concentrations the acid does not compete with the
adsorption of PNEA and protonation of the pre-adsorbed
amine was observed. On the other hand, the fragmentation
of the secondary amine leading to the formation of (S)-AF is
favoured in acidic reaction medium. The C–N bond scission
of adsorbed PNEA and the formation of AF were not observed
in the surface spectra. Probably the reaction conditions (low
acid concentration, only atmospheric hydrogen pressure) are
too mild for conversion of a detectable quantity. In the com-
petitive adsorption experiment, however, the significantly
stronger adsorption of (R,S)-PNEA is evident and co-adsorbed
(S)-AF seems to act merely as a “gap filler” of smaller adsorp-
tion sites left uncovered by the bulky PNEA. In absence of
PNEA, the aminolactone alone is not able to induce enantio-
selection. Furthermore, decreasing catalytic activity at higher
aminolactone concentration indicates the competitive
adsorption of AF and KPL at the active hydrogenation sites.
The successful upgrade of (R)-NEA by the pantoyl group
can be discussed in the context of Scheme 2, which high-
lights important features of chiral modifiers and their inter-
relation. Tuning the efficiency of chirally-modified metals the
molecular structure of the modifier is the key factor, but the
correct choice of the reaction environment (solvent, acid
additive, hydrogen surface coverage) also plays a decisive
Conclusions
The adsorptive anchoring and structural stability of NEA-based
chiral modifiers has been studied by means of ATR-IR
spectroscopy and catalytic experiments. In the Pt-catalysed
asymmetric hydrogenation of KPL, the improved enantio-
differentiation of PNEA- compared to NEA-modified catalyst
could be traced back to the adsorption of the chiral modifiers
on Pt/Al2O3. At the catalytic solid–liquid interface a rather
vaguely defined adsorption mode of the primary amine
(R)-NEA with the naphthalene group tilted along both of
its axes of symmetry was observed. The secondary amine
(R,S)-PNEA additionally anchors to the Pt surface via its
pantoyl group, thus forcing the naphthalene group in a more
Scheme
2 Important features of a chiral modifier influencing
enantioselection during heterogeneous hydrogenation.
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2014

View Article Online
Catalysis Science & Technology
Paper
parallel adsorption mode with respect to the catalyst surface.
In this adsorption geometry PNEA sterically and electronically
limits the stereochemical flexibility around its interacting N
atom and co-adsorbing KPL is more likely forced to the Pt
surface with its pro-IJR) configuration. Further improvement
of the chiral modifier PNEA and related NEA-based modifiers
is feasible if a deeper understanding of the governing reac-
tion mechanism involving these chiral modifiers is achieved.
The possible fragmentation of (R,S)-PNEA under catalytic
conditions and the active role of formed (S)-AF was also
investigated. The chiral aminolactone adopts an upright
adsorption geometry with the ester carbonyl tilted with
respect to the Pt surface. In the competitive adsorption bulky
adsorbed PNEA suppresses accumulation of AF on the sur-
face. In the absence of PNEA the chiral aminolactone does
not induce enantioselection but rather competes for active
hydrogenation sites leading to decreased catalytic activity.
19 T. Bürgi and A. Baiker, Acc. Chem. Res., 2004, 37, 909–917.
20 G. Goubert, V. Demers-Carpentier, R. P. Loach, R. Lafleur-
Lambert, J. C. Lemay, J. Boukouvalas and P. H. McBreen,
ACS Catal., 2013, 3, 2677–2683.
21 M. Schürch, N. Künzle, T. Mallat and A. Baiker, J. Catal.,
1998, 176, 569–571.
22 J. M. Bonello, E. C. H. Sykes, R. Lindsay, F. J. Williams,
A. K. Santra and R. M. Lambert, Surf. Sci., 2001, 482, 207–214.
23 J. M. Bonello, F. J. Williams and R. M. Lambert, J. Am. Chem.
Soc., 2003, 125, 2723–2729.
24 V. Demers-Carpentier, G. Goubert, F. Masini, R. Lafleur-
Lambert, Y. Dong, S. Lavoie, G. Mahieu, J. Boukouvalas,
H. L. Gao, A. M. H. Rasmussen, L. Ferrighi, Y. X. Pan,
B. Hammer and P. H. McBreen, Science, 2011, 334, 776–780.
25 I. Lee, Z. Ma, S. Kaneko and F. Zaera, J. Am. Chem. Soc.,
2008, 130, 14597–14604.
26 A. D. Gordon and F. Zaera, Angew. Chem., Int. Ed., 2013, 52,
3453–3456.
Acknowledgements
27 E. Schmidt, C. Bucher, G. Santarossa, T. Mallat, R. Gilmour
and A. Baiker, J. Catal., 2012, 289, 238–248.
28 N. Bonalumi, T. Bürgi and A. Baiker, J. Am. Chem. Soc.,
2003, 125, 13342–13343.
29 F. Meemken, K. Hungerbühler and A. Baiker, Angew. Chem.,
Int. Ed., 2014, 53(33), 8640–8644.
30 A. Vargas, G. Santarossa and A. Baiker, J. Phys. Chem. C,
2011, 115, 1969–1977.
31 V. Demers-Carpentier, A. M. H. Rasmussen, G. Goubert,
L. Ferrighi, Y. Dong, J. C. Lemay, F. Masini, Y. Zeng,
B. Hammer and P. H. McBreen, J. Am. Chem. Soc., 2013, 135,
9999–10002.
32 Z. Ma, I. Lee and F. Zaera, J. Am. Chem. Soc., 2007, 129,
16083–16090.
Financial support by the foundation Claude & Giuliana is
kindly acknowledged. We thank the ETH Zürich and WWU
Münster for supporting this research.
Notes and references
1 T. Mallat, E. Orglmeister and A. Baiker, Chem. Rev.,
2007, 107, 4863–4890.
2 Y. Izumi, M. Imaida, H. Fukawa and S. Akabori, Bull. Chem.
Soc. Jpn., 1963, 36, 21–25.
3 Y. Orito, S. Imai, S. Niwa and J. Nguyengiahung, Yuki Gosei
Kagaku Kyokaishi, 1979, 37, 173–174.
4 Y. Nitta, Y. Ueda and T. Imanaka, Chem. Lett., 1994, 1095–1098.
5 M. Sutyinszki, K. Szori, K. Felföldi and M. Bartók, Catal.
Commun., 2002, 3, 125–127.
33 S. Diezi, M. Hess, E. Orglmeister, T. Mallat and A. Baiker,
Catal. Lett., 2005, 102, 121–125.
6 M. Bartók, Curr. Org. Chem., 2006, 10, 1533–1567.
7 H. U. Blaser, H. P. Jalett, M. Müller and M. Studer, Catal.
Today, 1997, 37, 441–463.
8 Z. Ma and F. Zaera, J. Phys. Chem. B, 2005, 109, 406–414.
9 T. Bürgi and A. Baiker, J. Am. Chem. Soc., 1998, 120,
12920–12926.
34 S. Diezi, D. Ferri, A. Vargas, T. Mallat and A. Baiker, J. Am.
Chem. Soc., 2006, 128, 4048–4057.
35 A. G. Roach, C. E. A. Maillol, F. X. Wilson, G. M. Wynne,
J. Raufer, J. Vile, N. Thomas, R. Storer, P. Middleton and
O. De Moor, GB2467560-A, GB2467560-A 11 Aug 2010 A61K-
031/341 201056.
10 H. U. Blaser, B. Pugin and F. Spindler, J. Mol. Catal. A:
Chem., 2005, 231, 1–20.
11 H. U. Blaser, H. P. Jalett, W. Lottenbach and M. Studer,
J. Am. Chem. Soc., 2000, 122, 12675–12682.
36 M. Garland, H. P. Jalett and H. U. Blaser, in Stud. Surf. Sci.
Catal., ed. M. Guisnet, J. Barbier, J. Barrault, C. Bouchoule,
D. Dupréz, G. Perot, C. Montassier, Elsevier, 1991, vol. 59,
pp. 177–184.
12 M. Bartók, M. Sutyinszki, I. Bucsi, K. Felföldi, G. Szöllösi,
F. Bartha and T. Bartók, J. Catal., 2005, 231, 33–40.
13 B. Minder, M. Schürch, T. Mallat, A. Baiker, T. Heinz and
A. Pfaltz, J. Catal., 1996, 160, 261–268.
37 F. Meemken, P. Müller, K. Hungerbühler and A. Baiker,
Rev. Sci. Instrum., 2014, 85, 084101.
38 R. G. Greenler, D. R. Snider, D. Witt and R. S. Sorbello,
Surf. Sci., 1982, 118, 415–428.
14 E. Orglmeister, T. Mallat and A. Baiker, Adv. Synth. Catal.,
2005, 347, 78–86.
39 A. Urakawa, D. M. Meier, H. Rugger and A. Baiker, J. Phys.
Chem. A, 2008, 112, 7250–7255.
15 A. Baiker, J. Mol. Catal. A: Chem., 1997, 115, 473–493.
16 S. Lavoie, M. A. Laliberte, I. Temprano and P. H. McBreen,
J. Am. Chem. Soc., 2006, 128, 7588–7593.
17 F. Zaera, J. Phys. Chem. C, 2008, 112, 16196–16203.
18 O. Schwalm, B. Minder, J. Weber and A. Baiker, Catal. Lett.,
1994, 23, 271–279.
40 R. A. Olsen, D. Borchardt, L. Mink, A. Agarwal, L. J. Mueller
and F. Zaera, J. Am. Chem. Soc., 2006, 128, 15594–15595.
41 E. Schmidt, T. Mallat and A. Baiker, J. Catal., 2010, 272,
140–150.
42 D. Ferri and T. Bürgi, J. Am. Chem. Soc., 2001, 123,
12074–12084.
This journal is © The Royal Society of Chemistry 2014
Catal. Sci. Technol.