Enantioselection on Heterogeneous Noble Metal Catalyst: Proline-Induced Asymmetry in the Hydrogenation of Isophorone on Pd Catalyst

Article abstract of DOI:10.1021/jacs.5b07904

In the (S)-proline-mediated asymmetric hydrogenation of isophorone (IP) on supported Pd catalyst, excellent enantioselectivity is achieved, with an enantiomeric excess of up to 99%. The role of the heterogeneous catalyst has been the subject of a controve

Full text of DOI:10.1021/jacs.5b07904

Subscriber access provided by UNIV OF CAMBRIDGE
Article
Enantioselection on Heterogeneous Noble Metal Catalyst: Proline-
Induced Asymmetry in the Hydrogenation of Isophorone on Pd Catalyst
Laura Rodríguez-García, Konrad Hungerbuehler, Alfons Baiker, and Fabian Meemken
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.5b07904 • Publication Date (Web): 02 Sep 2015
Downloaded from http://pubs.acs.org on September 2, 2015
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of the American Chemical Society is published by the American Chemical
Society. 1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 36
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Enantioselection on Heterogeneous Noble Metal Catalyst:
Proline-Induced Asymmetry in the Hydrogenation of
Isophorone on Pd Catalyst
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Laura Rodríguez-García, Konrad Hungerbühler, Alfons Baiker, Fabian Meemken*
Institute for Chemical and Bioengineering, Department of Chemistry and Applied
Biosciences, ETH Zürich, Hönggerberg, HCI, CH-8093 Zürich, Switzerland
* Corresponding author: E-mail fabian.meemken@chem.ethz.ch
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 36
1
2
3
4
Abstract
5
6
7
8
In the (S)-proline-mediated asymmetric hydrogenation of isophorone (IP) on supported Pd catalyst
excellent enantioselectivity with an enantiomeric excess of up to 99% is achieved. The role of the
heterogeneous catalyst has been subject to a controversial debate and the existing mechanistic
understanding cannot explain the observed enantioselectivity of this catalytic system. The lack of in
situ information about the role of the heterogeneous catalyst has prompted us to investigate the
surface processes occurring at the methanol-Pd catalyst interface using attenuated total reflection-
infrared spectroscopy. Time-resolved monitoring of the homogeneous solution and of the catalytic
solid-liquid interface coupled with catalytic data provides crucial information on the catalytically
relevant enantiodifferentiating processes. While the condensation of IP and the corresponding chiral
product 3,3,5-trimethyl cyclohexanone with the chiral amine is connected to the
enantiodifferentiation, it was found that the crucial enantioselectivity-controlling steps take place
on the metal surface and the reaction has to be classified as heterogeneous asymmetric
hydrogenation. The presented spectroscopic and catalytic results provide strong evidence for the
existence of two competing enantioselective processes leading to opposing enantioselection.
Depending on surface coverage of the Pd catalyst, the reaction is either controlled by kinetic
resolution ((S)-pathway) or by chiral catalysis ((R)-pathway). Steering the hydrogenation on the
(R)-reaction pathway requires sufficient concentration of IP-(S)-proline condensate as this chiral
reactive intermediate becomes the most abundant surface species inhibiting the competing kinetic
resolution. The unraveled (R)-reaction pathway emphasizes an intriguing strategy for inducing
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
chirality
in
heterogeneous
asymmetric
catalysis.
ACS Paragon Plus Environment

Page 3 of 36
Journal of the American Chemical Society
1
2
3
4
1 Introduction
5
6
7
8
9
Catalytic asymmetric hydrogenation is a key reaction in the production of optically active
intermediates and products, which find important application as pharmaceuticals, agrochemicals,
fragrances and flavors.1 While mostly homogenous catalysts are employed for these reactions, the
application of heterogeneous catalysts could be beneficial due to their inherent practical advantages
concerning catalyst stability, separation and reuse. Furthermore, heterogeneous catalytic reactions
are generally more amenable for continuous process operation. Prerequisite for the rational design
of solid catalysts is a proper understanding of the structure of the active sites as well as of the
mechanism of the chemical transformations occurring on the catalytic surface. Steering catalysis
toward a desired pathway requires sophisticated strategies for controlling the environment at the
catalytically active sites. Crucial for heterogeneous asymmetric catalysis is a chiral reaction
environment, which directs the stereochemical orientation of the prochiral reactant on the active
solid catalyst surface. This chiral reaction environment can be created by applying various
strategies, including grafting, tethering or encapsulation of active metal complexes with chiral
ligands2,3 or the strong adsorption of suitable chiral organic compounds.4-6 While all these methods
have been successfully applied, bestowing chirality to the surface by strongly adsorbing chiral
compounds has proven to be most practical.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Simple addition of a small amount of a chiral compound, the so-called chiral modifier, to the liquid
reaction solution is sufficient to induce strong stereochemical control in heterogeneous
hydrogenations, exemplified by efficient asymmetric hydrogenations of α-activated ketones on
cinchona-modified Pt and of α-functionalized olefins on cinchona-modified Pd.4-6 For cinchona-
modified Pt, the mechanistically best understood chirally-modified metal, the established
mechanistic models consider the active chiral modifier species to be strongly anchored to the Pt
surface.4-9 At such a chirally-modified hydrogenation site enantiodiscrimination originates from
attractive interaction involving hydrogen bonding between the prochiral substrate and the tertiary
amine moiety of the adsorbed modifier,10-12 leading to an enantiomeric excess (ee) of up to 98%.13
The crucial role of the hydrogen bonding between prochiral reactant and the chiral modifier for
enantiodifferentiation has recently been substantiated by attenuated total reflection-infrared (ATR-
IR) spectroscopic detection during the asymmetric hydrogenation.12
1
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 36
1
2
3
4
5
6
7
8
9
Scheme 1. Asymmetric hydrogenation of isophorone in the presence of (S)-Pr on Pd catalyst.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In the hydrogenation of the prochiral C=C bond in isophorone (IP) on supported Pd catalyst
addition of the chiral amino acid (S)-proline ((S)-Pr) to the liquid reaction solution also induces
remarkable enantiodiscrimination and an ee of up to 99% of the corresponding (S)-
trimethylcyclohexanone (TMCH) has been reported (Scheme 1).14 However, the observed behavior
of this catalytic system cannot be rationalized with the existing mechanisms proposed for chirally-
modified metals. The major difference is the large amount of chiral proline required to achieve high
ee. According to catalytic data reported in the literature, enantioselectivity is also highly sensitive to
other reaction parameters, e.g. the properties of the metal catalyst,15 the catalyst support,14,16 the
solvent14 and even the preparation of the reaction solution.17 The lack of in situ information about
the surface processes occurring at the catalytic solid-liquid interface as well as contradicting
experimental observations have led to a controversial debate on the origin of enantioselection in the
hydrogenation of IP in the presence of (S)-Pr.14,15,17-19 In 1989 Tungler et al. proposed
enantioselectivity to result from stereoselective hydrogenation of an IP-(S)-Pr condensation product
leading to the preferential formation of the (S)-enantiomer.20,21 More recently, the group of Lambert
discovered preferential condensation of the cyclohexanone product (R)-TMCH with (S)-Pr in
homogeneous solution and proposed kinetic resolution to be the sole origin of enantioselectivity.
They also investigated the competitive adsorption of (S)-Pr and IP in an electrochemical study, and
found the cyclohexenone to adsorb 105 times faster on a Pt electrode.22 The authors extrapolated
these findings attempting to explain the surface processes occurring at the methanol-Pd catalyst
interface, claiming that (S)-Pr is not able to translate chiral information to the surface reaction due
to its very weak adsorption strength. In agreement with significantly stronger adsorption of IP, (S)-
Pr was proposed to be exclusively present in the homogeneous reaction solution. Subsequent to the
heterogeneous hydrogenation of IP to racemic TMCH the chiral amino acid would then act as chiral
resolving agent by forming preferentially (R)-TMCH-(S)-Pr condensate in the liquid phase. As the
ee observed in the absence of Pd reaches only 8% (S)-TMCH, the role of the heterogeneous catalyst
was further assigned to the reduction of TMCH-(S)-Pr condensate, thereby shifting the equilibrium
to yield the unconsumed (S)-enantiomer in high excess.17 If this would be the underlying reaction
2
ACS Paragon Plus Environment

Page 5 of 36
Journal of the American Chemical Society
1
2
3
4
5
6
mechanism, the (S)-Pr-mediated asymmetric hydrogenation of IP on Pd catalyst cannot be
classified as a heterogeneous enantioselective hydrogenation.
7
8
9
Addressing these claims, the groups of Török and Tungler published catalytic data, which
emphasized the strong effect of different Pd catalysts on the ee.14,23 While they agreed on the
importance of the kinetic resolution occurring in the liquid phase, their published results highlight
the peculiar behavior of enantioselection during the course of the hydrogenation, which cannot be
rationalized without the existence of an additional enantioselective pathway involving the Pd
catalyst.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In addition to the debate on the underlying reaction mechanism, the synthetic aspect of the (S)-Pr-
mediated enantioselective hydrogenation of IP has also been addressed. With the goal of achieving
chiral modification of the heterogeneous catalyst thiol derivatives of proline were anchored to the
surface, but only very low ee (14%) was achieved.24 Other approaches focused more on the role of
the heterogeneous catalyst. Recently, catalytic systems using strongly basic support materials were
proposed to promote (S)-Pr adsorption by ionic interactions between the chiral auxiliary and the
catalyst support.14 Particularly, carbonate-supported catalysts were claimed to promote high (S)-Pr
coverage at the metal-support interface, which would inhibit racemic IP hydrogenation.14
Furthermore, the effect of Pd particle size on various supports was investigated by Shen and co-
workers and a pronounced effect on the obtained enantioselectivity was found.9 In another
comprehensive study on the effect of Pd particle size of Pd/MgO catalysts inversion of
enantioselectivity with increasing Pd particles was found.15
However, the current mechanistic understanding of the (S)-Pr-mediated hydrogenation of IP on Pd
catalyst fails to explain these catalytic results. Particularly, the controversy over the role of the
heterogeneous hydrogenation catalyst has hindered concerted research effort aiming at optimization
and application of this system. For a more rational development of this complex, but very intriguing
enantioselective catalytic system a better understanding of the origin of enantioselection is
indispensable. The lack of in situ information on the role of the heterogeneous catalyst has
prompted us to investigate the surface processes occurring at the catalytic solid-liquid interface
using attenuated total reflection-infrared (ATR-IR) spectroscopy. Time-resolved monitoring of the
homogeneous solution and of the methanol-Pd catalyst interface coupled with catalytic data
provides crucial information on the catalytically relevant enantiodifferentiating processes. While
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 36
1
2
3
4
5
6
7
8
9
the condensation of the cyclohexanone IP and the corresponding cyclohexanone TMCH with the
chiral amine is connected to the enantiodifferentiation, it will be shown that in fact the surface
coverage of the Pd catalyst controls the outcome of the asymmetric hydrogenation and the crucial
enantioselectivity-controlling steps take place on the metal surface. Based on our catalytic and
spectroscopic results, we derived a new reaction pathway and completed the early proposed
reaction scheme, which now can explain the existing published results. The unraveled reaction
pathway emphasizes an intriguing strategy for inducing chirality in heterogeneous asymmetric
catalysis.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2 Experimental
2.1 Materials
3,5,5-Trimethyl-2-cyclohexene-1-one (Isophorone) (Aldrich, 97%), 3,5,5-trimethyl cyclohexanone
(Aldrich, 98%), (S)-Pr (Sigma-Aldrich, ≥ 99%), N-methyl-(S)-proline (Sigma-Aldrich), methanol
(Sigma-Aldrich, 99.9%), 1-butanol (Merck, ≥ 99.5%) and propan-2-ol (Fisher, 99.95%) were used
as received. Hydrogen (PanGas, Hydrogen 5.0, 99.999%) was used as received. Pure IP-(S)-Pr
condensate (≥ 97% as determined by gas chromatography (GC)) was obtained by separation of an
equilibrated IP-(S)-Pr solution (≥ 60% IP conversion) using column chromatography.
The commercial 5 wt.% Pd/C (5000 Engelhard, Pd dispersion: 0.53) and 5 wt.% Pd/Al2O3 (76002
Fluka, Pd dispersion: 0.46) were used as received. Preparation of a 5 wt.% Pd/TiO2 catalyst (Pd
25
dispersion: 0.27) has been described elsewhere and prior to use this catalyst was reduced in a
fixed-bed tubular reactor under flowing hydrogen for 30 min at room temperature. Dispersion has
been determined by CO chemisorption.
2.2 Hydrogenations
Hydrogenations of IP were performed in a 50 mL three-necked glass reactor at 298 K (controlled
by a Mettler Toledo Easy Max 102) and at atmospheric pressure under constant flow of H2 (semi-
batch hydrogenation). The general reaction procedure was the following: the catalyst was
transferred to the reactor and reduced in situ in 6 ml solvent under constant H2 flow for 1 h. Then,
the reaction was initiated by addition of an IP, BuOH (internal standard) and (S)-Pr reaction
mixture in 10 mL solvent. The following reaction conditions were used for all hydrogenations
unless stated otherwise: 8.0 0.1 mg 5 wt.% Pd catalyst, 100 mM IP (1.6 mmol), 100 mM (S)-Pr
4
ACS Paragon Plus Environment

Page 7 of 36
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
(modifier : Pds, 850 : 1), 16 ml solvent, 298 K, hydrogen flow, mechanical stirring (500 rpm).
Studies on the kinetic resolution of TMCH were conducted according to the same procedure except
that TMCH was used instead of IP. Racemic hydrogenation of IP was performed with 5.0 0.1 mg
5 wt.% Pd catalyst, 100 mM IP (8.0 mmol), 80 ml solvent, room temperature, hydrogen flow,
mechanical stirring (900 rpm). The conversion and enantioselectivity were determined by GC
equipped with a flame ionization detector and a chiral capillary column (CP-Chirasil-Dex CB, 25 m
length, 0.25 mm internal diameter, 0.25 ꢀm film thickness). The conversion of IP was calculated
using butanol as internal standard.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2.3 Preparation of Pd/C thin films for ATR-IR
A slurry was prepared with Pd/C catalyst (2 mg) and deionized water (2 mL), which was ultra-
sonicated for 15 min to obtain a well-mixed suspension. Subsequently, the entire amount of slurry
was placed on a ZnSe internal reflection element (IRE, bevel of 45°, 72 mm x 10 mm x 6 mm,
Specac Ltd.) and water was evaporated at 353 K for 2 h. This procedure yielded a catalyst layer of
1.5 0.1 mg. After mounting the coated crystal in a flow through cell, the catalyst layer was
reduced by subjecting it to flowing solvent saturated with hydrogen at 298 K for 1 h prior to use.
The as-prepared catalyst layer adhered to the IRE such that after initial material loss of loosely
adhered Pd/C particles, no further ablation of catalyst was observed over the course of several hours
under flow-through conditions.
2.4 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 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 implementation into a recycle reactor has been described in
detail elsewhere.26
Modulation excitation spectroscopy (MES) experiments were carried out by changing between two
different solutions every 180 s. The cycles were repeated several times to obtain stable responses of
active species, and the last three cycles were integrated into one cycle to average the spectra and to
5
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 36
1
2
3
4
5
6
enhance the signal to noise ratio. Details of the experimental procedure are available in the
Supporting Information.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3 Results and discussion
3.1 Hydrogenation activity of IP in presence of (S)-Pr and (S)-Pr derivatives
Kinetic resolution of racemic TMCH by enantioselective condensation of (S)-Pr and (R)-TMCH in
the homogeneous phase is proposed to be responsible for the enantioselection in the hydrogenation
of IP and currently the accepted reaction mechanism in the literature.17 According to this
mechanism, IP has to be rapidly hydrogenated on Pd to racemic TMCH and subsequently the (R)-
enantiomer is enantioselectively consumed by condensation with (S)-Pr in the liquid phase. The
condensation, however, is limited by competing hydrolysis, but a consecutive reduction on the Pd
catalyst yielding the fully saturated (R)-TMCH-(S)-Pr adduct is proposed to complete the kinetic
resolution in the liquid phase with a maximum yield of 50% (S)-TMCH.
In preliminary experiments using freshly prepared reaction solutions (within 1 min prior to
initiation of hydrogenation) we also observed an ee of (S)-TMCH and obtained a turnover
frequency (TOF) of 214 h-1. Interestingly, in absence of (S)-Pr a significantly higher TOF of 1520
h-1 was obtained (Figure 1B). In contrast to the proposed displacement of (S)-Pr and strong
abundance of the cyclohexenone on the catalytic surface,22 such a pronounced drop in catalytic
activity does indicate strong interference of the chiral amino acid with the fast hydrogenation of IP.
Note that our Pd/C catalyst displays superior performance compared to the values reported in the
literature,9,27 but a similar effect on the catalytic activity was also observed with less active
catalysts.
To better understand the molecular interaction(s) of (S)-Pr with the reactant and with the Pd
catalyst, and to identify the decisive factors leading to the deterioration of the hydrogenation
activity we carried out IP hydrogenations in presence of proline derivatives. The effect of
6
ACS Paragon Plus Environment

Page 9 of 36
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
competitive liquid phase condensation on the TOF was investigated using the tertiary amine N-
methyl-(S)-proline ((S)-NMePr), which cannot form condensation products. Adding a
stoichiometric amount of (S)-NMePr to the hydrogenation of IP also led to a significant drop in
TOF (Figure 1C). Nevertheless, the tertiary amine did not induce any enantioselection indicating
the crucial role of the secondary amine for the enantioselection. For the hydrogenation in the
presence of pyrrolidine (Py) a TOF of 537 h-1 was achieved. However, TMCH formation
significantly decreased after 30 min, while the condensation of IP and Py led to ongoing
consumption of IP (Figure 1D).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. Yield of TMCH in the hydrogenation of IP on Pd/C catalyst in presence of a stoichiometric amount of (S)-Pr (A), in
absence of (S)-Pr, and in presence of a stoichiometric amount of (S)-NMePr (C) and of pyrrolidine (D). The formation of
condensation products is indicated in red.
From the catalytic activities presented in Figure 1 we infer that (S)-Pr does interfere with the
hydrogenation of IP due to competitive interaction in the liquid phase and on the catalytic surface,
7
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 36
1
2
3
4
5
6
7
8
9
which involves the adsorption of (S)-Pr and condensation products. While no condensates were
formed in the presence of (S)-NMePr, the nearly five times reduced hydrogenation rate indicates
that the acid can competitively adsorb on the Pd catalyst. In a reflection absorption infrared
spectroscopic study under ultra-high vacuum conditions Raval et al. have found strong adsorption
of (S)-Pr on Cu(110) via the carboxylate moiety and the nitrogen of the amino group.28 Considering
the observed effect of (S)-NMePr on catalytic activity sufficiently strong interaction via the
carboxylate with the Pd catalyst appears to also occur under hydrogenation conditions in presence
of methanol. Another crucial observation concerns the influence of IP-Py condensate formation and
its effect on the catalytic reaction (Figure 1D). From the ceasing TMCH formation rate we infer that
such condensate products are prone to accumulate on the catalytic surface, which then leads to
blocking of active IP hydrogenation sites.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3.2 Adsorption of IP and (S)-Pr on Pd catalyst in presence of methanol
For a more comprehensive understanding on the role of (S)-Pr in the hydrogenation of IP on Pd
catalyst in methanolic solution we attempted to monitor the catalytic solid-liquid interface under
reaction conditions by means of in situ ATR-IR spectroscopy. However, alumina supported
catalysts, normally preferred for such studies due to the strong adherence of alumina on the IRE,
proved to be unsuitable for studying this catalytic system. First of all, among the tested catalysts the
alumina supported catalysts performed poorly, independent of the catalyst pretreatment.
Furthermore, in ATR-IR adsorption experiments using Pd/Al2O3-catalyst layer no adsorption from
IP-containing methanol solution was discernible. In contrast, strong adsorption of (S)-Pr on this
catalyst was detected, but the control experiment with Al2O3 layer confirmed accumulation of the
amino acid on the support material. To obtain spectroscopic data relevant to the processes occurring
at the methanol-Pd interface, we developed a layer deposition technique suitable for making
sufficiently stable Pd/C-catalyst layer (see experimental section). The Pd/C catalyst not only
provided by far the best catalytic performance, but also the investigated adsorbates do not
dynamically adsorb and desorb on carbon within the time constant of the experiments, which
significantly facilitates spectral analysis of the surface processes occurring on Pd.
In Figure 2, time-resolved surface plots display the vibrational changes monitored on a blank (A)
and on a Pd/C-coated (B,C) internal reflection element (IRE) during switching from H2-saturated
methanol solution to H2-saturated methanol solution containing either (S)-Pr (A,B) or (S)-NMePr
8
ACS Paragon Plus Environment

Page 11 of 36
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
(C). For all experiments, continuous switching between the two solutions in 3 min intervals
revealed that the detected species were captured in quasi-steady state, that is, appearance and
disappearance of absorption signals modulated between maximum and minimum intensities within
this time scale. To improve the signal-to-noise ratio of the low intensity absorption bands, the
quasi-steady state data were averaged over several runs, which in the literature is referred to as
concentration modulation excitation spectroscopy (MES).29,30
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2. Time-resolved surface spectra during concentration modulation between 1 and 0 mM (S)-Pr in methanol on a blank IRE
(A) and on a Pd/C-coated IRE (B), and between 1 and 0 mM (S)-NMePr solution on Pd/C-coated IRE (C) at 298 K. The modulation
period was set to 3 min and 3 modulation cycles were averaged.
On blank IRE, the liquid phase absorption of 1 mM (S)-Pr in methanol was detected. The broad and
intense vas(OCO-) of zwitterionic (S)-Pr peaked at 1641 cm-1. A second broad absorption band
peaking at around 1392 cm-1 was discernible, which corresponds to a combination of vs(OCO-) and
δ(C-H).31 On Pd/C-coated IRE (Figure 2B), the center of the vas(OCO-) was redshifted to 1633 cm-
1, and the vs(OCO-) was blueshifted to 1385 cm-1. Another spectral feature on Pd/C-coated IRE was
detected at 1595 cm-1, which we assign to the carbonyl bond of adsorbed formyl species originating
from partially dehydrogenated methanol. The additional MES experiment using He- and H2-
saturated methanol solutions corroborated this assignment as the signal grew in absence of H2
(compare Fig S1 in Supporting Information). Detection of formyl species upon admission of amino
acid as well as tailing of the νas(OCO-) band upon switching back to pure methanol revealed that
(S)-Pr interacts with Pd under reaction conditions. Furthermore, shifting of the carboxylate
absorption bands indicates interaction of the amino acid with the Pd catalyst via the carboxylate
moiety. Comparing the surface spectra of the secondary and tertiary amine in Figure 2B and Figure
2C, respectively, weaker adsorption of (S)-NMePr on Pd was found. While the νas(OCO-) was also
9
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 12 of 36
1
2
3
4
5
6
shifted upon interaction with the Pd catalyst, weaker band intensities and slower formation of
formyl species suggest weaker interaction of the tertiary amine.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3. Time-resolved surface spectra during concentration modulation between 1 and 0 mM IP in methanol (A), 1 and 0 mM IP in
a 1 mM (S)-Pr methanolic solution (B), and 1 and 0 mM IP in a 1 mM (S)-NMePr methanolic solution (C) on Pd/C-coated IRE. The
modulation period was set to 3 min and 3 modulation cycles were averaged.
In the concentration MES experiment shown in Figure 3A, adsorption of IP on Pd/C in absence of
(S)-Pr is depicted. Upon interaction of IP with the Pd catalyst, the ratio of C=O (1653 cm-1) to C=C
(1633 cm-1) stretch absorption was attenuated compared to the absorption of dissolved IP (compare
Fig S2 in Supporting Information) suggesting a rather parallel orientation of the ketone bond with
respect to the metal surface (Greenler`s surface selection rule32). However, similar adsorption
strength of IP was observed as in the case of (S)-Pr concentration MES, both accumulating only
rather weakly at the methanol-Pd interface.
Their competitive adsorption is emphasized in the surface spectra in Figure 3B, which displays the
spectral changes induced by switching the feed from H2-saturated methanol solution containing 1
mM (S)-Pr to H2-saturated methanol solution containing 1 mM (S)-Pr and 1 mM IP. Compared to
the surface spectra involving only IP (Figure 3A), the co-adsorption of (S)-Pr led to significant
broadening and stronger absorption in the C=O stretch region, which might originate from
intermolecular interactions as well as dynamic re-orientation of the reactants with respect to the
catalytic Pd surface. As no significant attenuation of the νas(OCO-) mode of (S)-Pr was discernible,
detection at the methanol-Pd interface reveals a rather cooperative than a competitive interaction
between the prochiral cyclohexenone and the amino acid. Instead of displacement of (S)-Pr, co-
adsorption of IP seems to be on the expense of adsorbed formyl species followed by the
compensated absorption at 1595 cm-1. Stronger adsorption of the chiral amino acid involving only
10
ACS Paragon Plus Environment

Page 13 of 36
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
its carboxylate moiety is further supported by the surface spectra emphasizing the co-adsorption of
IP and (S)-NMePr. Upon addition of IP, no negative signals assignable to the tertiary amine are
detected, while the signal intensity of the ketone carbonyl is less pronounced (compare Figure 3A
and 3C).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In agreement with the catalytic results obtained with (S)-Pr and its derivatives, the spectroscopic
data provide evidence that (S)-Pr is able to interfere on the catalytic Pd surface with the
hydrogenation of IP in methanolic solution.
3.3 Homogeneous condensation reactions
As condensation in the liquid phase is proposed to play a crucial role for enantiodifferentiation we
studied methanolic reaction solutions in absence of Pd catalyst using ATR-IR spectroscopy and
chiral GC analysis. Upon mixing of IP and (S)-Pr (both 160 mM) in the reaction vessel a rapid
decrease of the IP carbonyl stretch at 1651 cm-1 was observed indicating the formation of
condensate (Figure 4A). The carbonyl signal decreased by 5% within less than 10 min and
remained constant thereafter. A sample was drawn after 30 min and 5% conversion of the
cyclohexenone corresponding to the formation of 8 mM IP-(S)-Pr condensate was confirmed by GC
analysis. For a stoichiometric mixture of racemic TMCH and (S)-Pr, the carbonyl band of the
cyclohexanone at 1705 cm-1 decreased by 25% within 30 min indicating the formation of 40 mM
TMCH-(S)-Pr condensate. Thereafter, the consumption of TMCH ceased and did not change
substantially during the following 60 min. Again, GC analysis confirmed TMCH consumption of
25%. However, the GC analysis also showed that 14.5% of (R)-TMCH and 10.0% of (S)-TMCH
were consumed revealing the poor enantioselection (6% ee (S)-TMCH) of the liquid-phase
condensation in its thermodynamic equilibrium. Considering the time constants of the catalytic
experiment (ca. 18 h) and of the rather rapid equilibration (ca. 30 min) kinetically controlled kinetic
resolution in the liquid phase can also be ruled out as the origin of the very high enantioselectivity.
Instead, enantioselection must be governed by a different mechanism.
11
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 14 of 36
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4. Formation of condensates in methanolic stoichiometric mixtures of IP and (S)-Pr as well as racemic TMCH and (S)-Pr (160
mM) followed within 60 min using in situ ATR-IR spectroscopy (A) and followed until complete equilibration using GC (B).
As the preparation of the liquid reaction solution has been reported to strongly influence the
outcome of the catalytic reaction we investigated the equilibrating liquid phase of stoichiometric
methanolic mixtures containing 160 mM ketone and (S)-Pr stored under air at room temperature
over several days (Figure 4B). Samples were drawn in regular intervals and the extent of
condensation was measured by GC using butanol as internal standard. Continuous consumption of
cyclohexenone was observed reaching the thermodynamic equilibrium at 67% conversion. At
337 K the thermodynamic equilibrium was reached significantly faster. It will be shown below that
forming IP-(S)-Pr condensate is the chiral reactive intermediate leading to chiral catalysis in the
asymmetric hydrogenation of IP. As also shown in Figure 4B, the kinetics of cyclohexanone
condensation is very different as the thermodynamic equilibrium is reached within the initial 30
min. The TMCH-(S)-Pr condensate even slowly hydrolyzed during the course of a month, which is
very likely caused by air moisture entering through the pierced GC vial, but the ee remained
constantly at 6 1%.
3.4 Heterogeneous hydrogenation of IP-(S)-Pr condensate
In order to grasp the importance of the IP-(S)-Pr condensate for the reaction mechanism, we
performed hydrogenations with pre-equilibrated reaction mixtures on Pd/C. Strikingly, inversion of
enantioselectivity was observed, that is, an ee of around 40% (R)-TMCH was obtained. As shown
in Figure 5, the ee remained almost constant for 5 h. Notice that the initial IP conversion – obtained
through homogeneous condensation – remained constant throughout the hydrogenation. Only after
12
ACS Paragon Plus Environment

Page 15 of 36
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
a reaction time of 24 h additional minor conversion (2%) of IP was observed. Therefore, we
propose that IP-(S)-Pr condensate strongly accumulates on the Pd surface blocking IP to access free
hydrogenation sites. In addition, the absence of fully saturated TMCH-(S)-Pr adducts in the reaction
mixture sampled during the experiment suggests that both (R)- and (S)-TMCH-(S)-Pr condensates
are rapidly replaced from the catalytic surface by stronger adsorbing IP-(S)-Pr condensate. Based
on the results presented in Figure 5 we conclude that the IP-(S)-Pr intermediate has to be the most
abundant surface species adsorbing significantly stronger on the Pd surface than its hydrogenation
products and the reactant IP.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 5. Conversion of IP, yield of TMCH and ee as a function of consumed IP-(S)-Pr condensate using equilibrated IP-(S)-Pr
solution and 8.0 mg Pd/C catalyst (A). Influence of the molar ratio of IP-(S)-Pr condensate to PdS on ee (determined after 60 min
hydrogenation) using pre-equilibrating reaction solutions (B). Solid circles denote the standard procedure described in the
experimental section, while the experiments corresponding to the open circles were performed with 4 and 2.2 mg catalyst,
respectively. The star indicates kinetic resolution starting with racemic TMCH.
Moreover, hydrogenations using standard reaction mixtures with varying content of IP-(S)-Pr
condensate showed an excellent correlation between obtained ee of (R)-TMCH and the used molar
ratio of IP-(S)-Pr condensate to Pd surface (PdS) (Figure 5B). Therefore, we propose that the crucial
enantioselectivity controlling steps do take place on the catalytic surface and chiral IP-(S)-Pr
intermediate is able to dictate the outcome of the catalytic reaction by inhibiting hydrogenation of
TMCH-(S)-Pr condensates. The significant contribution of forming IP-(S)-Pr condensate and the
competitive formation of (R)-TMCH even in the hydrogenation using fresh IP-(S)-Pr-reaction
solution is also highlighted in Figure 5B by the enantioselectivity obtained in the reaction of
racemic TMCH and (S)-Pr. In absence of IP, 10% higher (S)-excess is observed after 60 min
13
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 16 of 36
1
2
3
4
5
6
7
8
9
showing that initially fast formation of cyclohexene condensate (compare section 3.3) is able to
compete with the mechanism leading to the preferential formation of (S)-TMCH. In case the surface
is fully saturated with IP-(S)-Pr condensate, the best enantioselectivity to (R)-TMCH is achieved.
Using 2.2 mg instead of 8 mg catalyst we obtained the best ee of 52% (R)-TMCH after 240 min
corroborating the strong influence of high IP-(S)-Pr condensate surface coverage.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
To investigate the structure sensitivity of the Pd catalyst and the influence of the support material,
hydrogenations of equilibrated IP-(S)-Pr solutions were also performed on pre-reduced Pd/C,
Pd/Al2O3 and Pd/TiO2 catalysts (Table 1). Independent of the pre-treatment of the catalyst or the
used support material an ee of (R)-TMCH was always obtained with the equilibrated reaction
solution, whereas hydrogenations using fresh reaction solution always gave the (S)-enantiomer in
excess. Interestingly, under otherwise identical reaction conditions increasing the Pd particle sizes
of the Pd/C catalyst by reductive treatment at 400°C for 1 h significantly enhanced the ee from 40
to 52% (R)-TMCH and double the TOF. Using a Pd/Al2O3 catalyst the same ee as non-pre-treated
Pd/C but a higher activity was obtained. An exceptional activity for the IP-(S)-Pr condensate
hydrogenation was experienced with our Pd/TiO2 catalyst yielding also an enantioselectivity of
51% ee. The influence of particle size on enantioselectivity as well as the very high TOF found for
the titania supported catalyst demonstrate the potential for optimizing the performance of this
catalytic system.
Table 1. Catalytic results using several different Pd catalysts for fresh and for equilibrated reaction solutions. The TOF and ee were
determined after 15 and 60 min, respectively.
Catalyst
Dispersion
fresh solutions
equilibrated solutions
TOF (h-1)
(S)-ee (%)
TOF (h-1) (R)-ee (%)
Pd/C untreated
53%
13%
46%
27%
214
14
15
9
102
252
153
1144
40
Pd/C reduced
347
52
Pd/Al2O3 untreated
Pd/TiO2 untreated
60
39
321
9
33*
*ee started declining after 15 min (ee 51%) due to high conversion of IP-(S)-Pr condensate
14
ACS Paragon Plus Environment

Page 17 of 36
Journal of the American Chemical Society
1
2
3
4
5
6
3.5 Adsorption of condensates on Pd/C
7
8
9
Finally, we investigated the adsorption of condensates at the methanol-Pd catalyst interface. Due to
the dynamic nature of the condensation equilibria in liquid solution and competing hydrolysis of
condensates (compare section 3.3), we decided to implement the ATR-IR flow-through cell into a
recycle reactor set-up (3 mL volume)26 to follow adsorption at the solid-liquid interface from a
liquid reaction solution in dynamic equilibrium. After reduction of the deposited Pd/C-catalyst
layer for 1 h using H2-saturated methanol, a stable background was obtained and condensate
containing solutions were fed to the recycle reactor.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
For investigating the adsorption of IP-(S)-Pr condensate, purified inlet solution was prepared
employing column chromatography. Within 27 s at a flow rate of 11 mL/min, the recycle reactor
content was quickly exchanged with H2-saturated methanol solution containing 3.7 mM IP-(S)-Pr
condensate and approximately 0.4 mM IP as well as 0.4 mM (S)-Pr. As seen in the surface spectra
presented in Figure 6A, broad absorption bands centered at 1366, 1549 and 1609 cm-1 grew
continuously during recycling, which is highlighted for the latter absorption mode in the inset. After
33 min (red spectrum), accumulation on the catalyst had ceased and flushing of the recycle reactor
with H2-saturated methanol revealed the detection of species strongly bound to Pd/C (blue
spectrum). In comparison to the adsorption experiments using pure IP and (S)-Pr (compare section
3.2), these absorption bands were also significantly broadened and appeared at shifted positions
evidencing high surface coverage by another species, namely IP-(S)-Pr condensate adsorbed to
Pd/C. However, among the different IP-(S)-Pr condensate species (imine, diene, ketone) proposed
in the literature20,23 no unambiguous distinction was possible due to the broad nature of the
absorption bands.
Nevertheless, to confirm stronger adsorption of IP-(S)-Pr condensate at the methanol-Pd interface
we also investigated the adsorption of TMCH-(S)-Pr condensate. The fast hydrolysis makes it very
difficult to isolate this condensate in methanol. Therefore, an inlet solution containing 10 mM
racemic TMCH and 10 mM (S)-Pr was equilibrated under H2 atmosphere for 60 min yielding 2.5
mM TMCH-(S)-Pr condensate. As seen in Figure 6B, quick exchange of the recycle reactor content
induced increasing absorption at 1365, 1630 and 1705 cm-1. The latter absorption mode is
assignable to the carbonyl band of TMCH, which did not change after 2 min. The other two
15
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 18 of 36
1
2
3
4
5
6
7
8
9
absorption modes obeyed slightly slower kinetics reaching saturation within 3 min. The broad
absorption centered at 1630 cm-1 can originate from residual (S)-Pr and from TMCH-(S)-Pr
condensate accumulating on Pd, but no isolated peak indicating only the adsorption of the latter was
discernible. After saturation of the catalyst (red spectrum), the recycle reactor was again flushed
with H2-saturated methanol for 4 min (blue spectrum). Compared to the adsorption of IP-(S)-Pr
condensate, the detected species were removed significantly faster revealing their relatively weak
adsorption. Note that also adsorbed formyl species have to contribute to the remaining absorption
centered at 1585 cm-1.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 6. Adsorption on Pd/C from 3 mL H2-saturated methanol solution containing 3.7 mM IP-(S)-Pr condensate (A) and 2.5 mM
TMCH-(S)-Pr condensate (B), respectively. Due to the competing hydrolysis in liquid phase, 0.4 mM IP and 0.4 mM (S)-Pr (A) as
well as 7.5 mM TMCH and 7.5 mM (S)-Pr were also present in the inlet solutions, respectively.
In agreement to the catalytic results presented in section 3.4, the surface spectra provide evidence
for the strong accumulation of IP-(S)-Pr condensate on Pd. In contrast, IP, (S)-Pr, TMCH as well as
TMCH-(S)-Pr condensates seem to only weakly interact with the catalyst. While the active surface
species leading to enantiodifferentiation toward the (R)-enantiomer could not be identified, the
crucial role of IP-(S)-Pr condensate for controlling access to the heterogeneous catalyst was
corroborated by the spectroscopic analysis.
3.6 Mechanistic implications and proposed reaction scheme
The presented catalytic and spectroscopic results provide strong evidence for the existence of two
competing enantioselective processes leading to opposing enantioselection in the Pr-mediated
16
ACS Paragon Plus Environment

Page 19 of 36
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
asymmetric hydrogenation of IP on Pd-supported catalyst. Furthermore, we propose that the crucial
enantioselectivity-controlling steps occur on the heterogeneous catalyst and even the earlier
discovered kinetic resolution is governed by a stereoselective hydrogenation on Pd. Depending on
surface coverage of the Pd catalyst, enantioselection is either controlled by kinetic resolution ((S)-
pathway) or by chiral catalysis ((R)-pathway). The two competing reaction pathways presented in
Scheme 2 can explain the outcome of the catalytic hydrogenation, and the decisive reaction
parameter determining the sense of enantioselection is the concentration of IP-(S)-Pr condensate
formed in the first equilibrium involving IP and (S)-Pr.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 2. Enantioselectively competing reaction pathways involving kinetic resolution ((S)-pathway) and chiral catalysis ((R)-
pathway) for the (S)-Pr-mediated asymmetric hydrogenation of IP on supported Pd catalyst.
Following the (S)-pathway in Scheme 2, hydrogenation of IP in presence of (S)-Pr without pre-
equilibration favors kinetic resolution and the amino acid will mainly act as chiral resolving agent.
While competitive adsorption of (S)-Pr and continuous IP-(S)-Pr condensate formation interfere
with the efficient catalysis governed by this reaction pathway, the cyclohexenone IP is still
hydrogenated on Pd to racemic TMCH. In the liquid phase, the condensation between (S)-Pr and
both enantiomers of the cyclohexanone rapidly reaches equilibrium, consuming slightly more (R)-
17
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 20 of 36
1
2
3
4
5
6
7
8
9
TMCH leading to an enantiomeric ratio 150 : 133 in favor of (S)-TMCH. However, the crucial
enantiodifferentiating step in the kinetic resolution is the diastereoselective hydrogenation of the
(R)-TMCH-(S)-Pr condensate on the Pd catalyst. Fast hydrogenation of the (R)-(S)-diastereomer on
Pd catalyst drives the equilibrium toward full consumption of (R)-TMCH and the significantly
slower hydrogenation of the (S)-TMCH-(S)-Pr condensate (r(R) >> r(S) in Scheme 2) enables
efficient kinetic resolution leading to very high enantioselectivity of (S)-TMCH. The most efficient
kinetic resolution would be achieved by addition of (S)-Pr after the fast racemic hydrogenation of
IP. Deterioration of activity due to the competitive adsorption of the chiral resolving agent on active
hydrogenation sites would be avoided and lower yield due to the preferential formation of the other
(R)-enantiomer would be prevented in the absence of any IP-(S)-Pr condensate. Nevertheless,
undesired consumption of 50% of the prochiral starting material IP as well as an equivalent molar
amount of resolving agent (S)-Pr limit the maximum yield to 50% (S)-TMCH, which is the main
drawback of this pathway.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
From an economic and environmental point of view, chiral catalysis is a more elegant approach to
produce only one enantiomer of a chiral compound, because a small amount of chiral catalyst is
able to induce optical activity in a large amount of product and formation of the undesired
enantiomer is avoided. Following the (R)-pathway in Scheme 2, true heterogeneous asymmetric
catalysis is achieved in the hydrogenation of IP on supported Pd using (S)-Pr as chiral modifier.
The crucial reaction parameter for steering the hydrogenation on this pathway is the concentration
of the chiral reactive intermediate IP-(S)-Pr, which can control the availability of active
hydrogenation sites. If the molar ratio of IP-(S)-Pr condensate to PdS is sufficiently high (IP-(S)-Pr
condensate to PdS exceeding roughly 700 for Pd/C) the chiral reactive intermediate becomes the
most abundant surface species, blocking surface access for any other species present in the liquid
reaction solution. While it remains an open question whether the catalytically active IP-(S)-Pr
condensate surface species is the imine or the diene species, its hydrogenation on Pd catalyst leads
to the preferential formation of (R)-TMCH-(S)-Pr condensate. Following the stereoselective
hydrogenation both diastereomers, the major (R)- and the minor (S)-TMCH-(S)-Pr condensate, are
quickly replaced by stronger adsorbing IP-(S)-Pr enantiomer preventing the consecutive kinetic
resolution. Fast hydrolysis of the cyclohexane condensates in liquid solution yields the ee of (R)-
TMCH and also the chiral modifier (S)-Pr is recovered. Steering the hydrogenation on the (R)-
pathway, we obtained an ee of 52% (R)-TMCH. However, optimization with respect to the reaction
18
ACS Paragon Plus Environment

Page 21 of 36
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
conditions, the properties of the heterogeneous hydrogenation catalyst, the support material and the
molecular structure of the chiral modifier provide plenty of opportunity to maximize the
enantioselection. In addition, sophisticated exploitation of both reaction pathways allows for
combining the benefits of chiral catalysis and kinetic resolution. While chiral catalysis using (S)-Pr
as chiral modifier produces an enantioenriched reaction solution with higher efficiency, subsequent
kinetic resolution employing (R)-Pr as resolving agent allows for obtaining (R)-TMCH with optical
purity.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 3. Proposed strategy for translating chiral information to an originally racemic heterogeneous hydrogenation of a prochiral
substrate.
Furthermore, the discovered pathway for the (S)-Pr mediated asymmetric hydrogenation of IP
employing the amino acid as chiral modifier emphasizes an intriguing strategy for translating chiral
information to an originally racemic heterogeneous hydrogenation of a prochiral substrate. The
general strategy for imprinting a chiral selector onto the prochiral substrate to achieve
heterogeneous asymmetric catalysis is summarized in Scheme 3): In the initial step, a chiral, amine-
based modifier rapidly condensates with the prochiral α,β-unsaturated ketone to form a chiral
reactive intermediate (step I). The obtained enantiomer preferentially adsorbs with one of its two
enantiofaces on the Pd catalyst (step II), which represents the enantiodifferentiating step of the
catalytic cycle. In step III, hydrogenation of the chiral alkene leads to the preferential formation of
one of the two diastereomeric alkanes, which upon desorption (step IV) hydrolyze in the liquid
19
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 22 of 36
1
2
3
4
5
6
7
8
9
phase to yield the chiral product alkane and to recover the chiral modifier for the next
enantioselective catalytic cycle (step V). Tuning electronic and steric properties of the chiral
modifier is one of the crucial parameters to allow for an efficient enantioselective catalytic
turnover. However, the lack of information about the exact molecular structure of the catalytically
active IP-(S)-Pr condensate does not allow for distinguishing between electronic and steric effects
responsible for controlling the enantiospecific adsorption of the condensate intermediate, and
further mechanistic understanding is required to optimize such stereoselective hydrogenations in a
more rational way.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4 Conclusions
We have investigated the reaction steps occurring in the bulk liquid phase and on the metal surface
during the (S)-Pr-mediated hydrogenation of IP over supported palladium catalysts. The combined
catalytic and spectroscopic studies revealed the existence of two competing reaction pathways:
kinetic resolution resulting in the production of (S)-TMCH, and palladium catalyzed stereoselective
hydrogenation of IP-(S)-Pr condensate leading to (R)-TMCH. The revisited reaction mechanism not
only explains the existing controversial interpretation of the origin of enantioselectivity, but also
answers the long-standing debate about the role of the Pd catalyst for enantioselection. The slowly
forming condensate between IP and (S)-Pr is a key reaction intermediate in the reaction scheme
owing to its ability to block active hydrogenation sites. Furthermore, it is stereoselectively
hydrogenated on Pd catalyst. The preferentially formed (R)-TMCH-(S)-Pr condensate hydrolyses
rapidly in the liquid phase with residual water and thus leads to the observed ee of (R)-TMCH.
An intriguing general strategy for translating chiral information to a heterogeneously catalyzed
reaction can be drawn from the discovered reaction pathway. In contrast to chirally-modified metal
catalysts, chiral information is imprinted onto the substrate by simple condensation of a α,β-
unsaturated carbonyl group and a secondary amine carrying the chiral information. The resulting
unsaturated chiral condensate is prone to adsorb enantiospecifically on the heterogeneous catalyst
and controls access to the catalytically active sites. After the asymmetric transformation of the
chiral reactive intermediate, scission of the chiral modifier by hydrolysis yields the chiral product.
5 Associated Content
20
ACS Paragon Plus Environment

Page 23 of 36
Journal of the American Chemical Society
1
2
3
4
Supporting Information
5
6
7
8
Experimental details, time-resolved surface spectra during modulation between H2- and He-
saturated methanol, liquid phase IR-spectrum of a methanolic IP solution, CO chemisorption
procedure and TEM images of Pd/C and Pd/TiO2 catalysts.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
6 Author Information
Corresponding Author
* fabian.meemken@chem.ethz.ch
Notes
The authors declare no competing financial interest.
7 Acknowledgment
We thank Dr. Robert Büchel for the CO chemisorption measurements and Dr. Frank Krumeich for
the TEM measurements. Financial support by the Foundation Claude & Giuliana is kindly
acknowledged.
8 References
(1)
(2)
(3)
Ager, D. J.; de Vries, A. H. M.; de Vries, J. G. Chem. Soc. Rev. 2012, 41, 3340.
McMorn, P.; Hutchings, G. J. Chem. Soc. Rev. 2004, 33, 108.
De Vos, D. E.; Vankelecom, I. F. J.; Jacobs, P. A. Chiral catalyst immobilization and recycling;
Weinheim : Wiley-VCH, 2000.
(4)
(5)
(6)
(7)
(8)
(9)
(10)
Mallat, T.; Orglmeister, E.; Baiker, A. Chem. Rev. 2007, 107, 4863.
Studer, M.; Blaser, H. U.; Exner, C. Adv. Synth. Catal. 2003, 345, 45.
Bartók, M. Curr. Org. Chem. 2006, 10, 1533.
Lavoie, S.; Laliberte, M. A.; Temprano, I.; McBreen, P. H. J. Am. Chem. Soc. 2006, 128, 7588.
Zaera, F. J. Phys. Chem. C 2008, 112, 16196.
Zhan, E. S.; Li, S.; Xu, Y. D.; Shen, W. J. Catal. Commun. 2007, 8, 1239.
Bonalumi, N.; Bürgi, T.; Baiker, A. J. Am. Chem. Soc. 2003, 125, 13342.
21
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 24 of 36
1
2
3
4
5
6
(11)
Demers-Carpentier, V.; Goubert, G.; Masini, F.; Lafleur-Lambert, R.; Dong, Y.; Lavoie, S.; Mahieu,
G.; Boukouvalas, J.; Gao, H.; Rasmussen, A. M. H.; Ferrighi, L.; Pan, Y.; Hammer, B.; McBreen, P.
H. Science 2011, 334, 776.
7
8
9
(12)
(13)
(14)
(15)
(16)
(17)
(18)
(19)
Meemken, F.; Hungerbühler, K.; Baiker, A. Angew. Chem. Int. Ed. 2014, 53, 8640.
Sutyinszki, M.; Szöri, K.; Felföldi, K.; Bartók, M. Catal. Commun. 2002, 3, 125.
Mhadgut, S. C.; Török, M.; Esquibel, J.; Török, B. J. Catal. 2006, 238, 441.
Li, S.; Chen, C. H.; Zhan, E. S.; Liu, S. B.; Shen, W. J. J. Mol. Catal. A-Chem. 2009, 304, 88.
Schäfer, C.; Mhadgut, S. C.; Kugyela, N.; Török, M.; Török, B. Catal. Sci. Technol. 2015, 5, 716.
McIntosh, A. I.; Watson, D. J.; Burton, J. W.; Lambert, R. M. J. Am. Chem. Soc. 2006, 128, 7329.
Kyriakou, G.; Beaumont, S. K.; Lambert, R. M. Langmuir 2011, 27, 9687.
Szabados, E.; Györffy, N.; Tungler, A.; Balla, J.; Könczöl, L. React. Kinet. Mech. Catal. 2014, 111,
107.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(20)
(21)
(22)
(23)
(24)
Tungler, A.; Kajtar, M.; Máthé, T.; Toth, G.; Fogassy, E.; Petró, J. Catal. Today 1989, 5, 159.
Tungler, A.; Máthé, T.; Petró, J.; Tarnai, T. J. Mol. Catal. 1990, 61, 259.
McIntosh, A. I.; Watson, D. J.; Lambert, R. M. Langmuir 2007, 23, 6113.
Györffy, N.; Tungler, A.; Fodor, M. J. Catal. 2010, 270, 2.
Watson, D. J.; Jesudason, R.; Beaumont, S. K.; Kyriakou, G.; Burton, J. W.; Lambert, R. M. J. Am.
Chem. Soc. 2009, 131, 14584.
(25)
(26)
(27)
(28)
(29)
(30)
(31)
(32)
Huck, W. R.; Bürgi, T.; Mallat, T.; Baiker, A. J. Catal. 2001, 200, 171.
Meemken, F.; Müller, P.; Hungerbühler, K.; Baiker, A. Rev. Sci. Instruments 2014, 85, 17.
Mhadgut, S.; Török, M.; Dasgupta, S.; Török, B. Catal. Lett. 2008, 123, 156.
Mateo Marti, E.; Barlow, S. M.; Haq, S.; Raval, R. Surf. Sci. 2002, 501, 191.
Baurecht, D.; Fringeli, U. P. Rev. Sci. Instruments 2001, 72, 3782.
Urakawa, A.; Bürgi, T.; Baiker, A. Chem. Eng. Sci. 2008, 63, 4902.
Bieri, M.; Bürgi, T. ChemPhysChem 2006, 7, 514.
Greenler, R. G.; Snider, D. R.; Witt, D.; Sorbello, R. S. Surf. Sci. 1982, 118, 415.
22
ACS Paragon Plus Environment

Page 25 of 36
Journal of the American Chemical Society
1
2
3
4
TOC
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
23
ACS Paragon Plus Environment

80
70
60
50
40
30
20
10
B
A
Journal of the American Chemical Society
Page 26 of 36
TOF: 214 h-1
TOF: 1520 h-1
+
1
2
3
4
5
6
7
8
9
10
11
12
13
TMCH (IP/PdS = 800)
condensates
TMCH (IP/PdS = 800)
TMCH (IP/PdS = 4600)
0
80
C
D
14
70
15
16
TOF: 342 h-1
TOF: 537 h-1
+
+
60
17
18
50
19
40
20
21
TMCH (IP/PdS = 800)
TMCH (IP/PdS = 1200)
condensates
30
22
23
20
24
10
25
26
27
28
29
ACS Paragon Plus Environment
40 50 60
10
0
0
10
20
30
0
20
30
40
50
60
time / min
time / min

1.5
1.0
0.5
0
6
A
blank
B
Pd/C
C
Pd/C
Page 27 of 36
Journal of the American Chemical Society
4
1
2
3
4
5
6
7
8
9
2
ACS Paragon Plus Environment
0
-0.5
1700 1600 1500 1400 1300
1700 1600 1500 1400 1300 1700 1600 1500 1400 1300
-1
~
ν / cm

6
4
2
0
Pd/C
Pd/C
A
B
C
Pd/C
0.8
Journal of the American Chemical Society
Page 28 of 36
0.4
1
2
3
4
5
6
7
8
9
0
-0.4
-0.8
ACS Paragon Plus Environment
1700 1600 1500 1400 1300
1700 1600 1500 1400 1300 1700 1600 1500 1400 1300
-1
~
ν / cm

Page 29 of 36
Journal of the American Chemical Society
A
B
80
30
25
1
60
2
3
20
IP at 298 K
IP at 337 K
4
5
6
7
8
9
10
11
12
13
IP at 298 K
TMCHat 298 K
TMCHat 298 K
15
40
20
0
10
5
0
ACS Paragon Plus Environment
0
10
20
30
40
50
60
70
0
200
400
600
800
1000
1200
t / min
t / h