Rights / License: Research Collection In Copyright - Non ... · 1.1 Concepts of chirality ... 1.3...
Transcript of Rights / License: Research Collection In Copyright - Non ... · 1.1 Concepts of chirality ... 1.3...
Research Collection
Doctoral Thesis
Heterogeneous enantioselective hydrogenation of α-keto estersand fluorinated ketones over chirally modified platinum
Author(s): Diezi, Simon
Publication Date: 2006
Permanent Link: https://doi.org/10.3929/ethz-a-005164493
Rights / License: In Copyright - Non-Commercial Use Permitted
This page was generated automatically upon download from the ETH Zurich Research Collection. For moreinformation please consult the Terms of use.
ETH Library
Diss. ETH No 16541
0Heterogeneous Enantioselective
0Hydrogenation of α-Keto Esters and
0Fluorinated Ketones over
0Chirally Modified Platinum
A dissertation submitted to the
Swiss Federal Institute of Technology (ETH), Zurich
for the degree of Doctor of Natural Sciences
presented by
Simon Diezi
Dipl. Chem. ETH
born 12 March 1977
citizen of Thal (SG)
accepted on the recommendation of
Prof. Dr. A. Baiker, examiner
Prof. Dr. A. Togni, co-examiner
2006
für meine Eltern Huldi und Jürg†
in Liebe und Dankbarkeit
0
0Acknowledgment
Firstly, I would like to express my sincere gratitude to Prof. Dr. A. Baiker for his
support, both personally and scientifically, and the opportunity to complete
my doctoral studies in his group.
Moreover, I would like to thanks Prof. Dr. A. Togni for accepting the task
of co-examiner in this thesis.
I furthermore thank to Dr. T. Mallat for constructive discussions, his valu-
able suggestions and the big help during the past four years, Dr. A. Vargas for
his contributions to the modelling and calculations, and Dr. D. Ferri for the
spectroscopic part of this thesis.
Special thanks to M. Hess and S. Reimann for the valuable contributions
to this work during their diploma work.
Furthermore, I would like to thank for the contribution to the present
work: F. Bangerter for recording NMR spectra and his refreshing comments on
scientific research; U. Krebs for his help with fine mechanics and M. Wohl-
wend for electronic support.
A very big thank is dedicated to my office-mates Dr. M. Rohr, M. Ramin,
M. Caravati and Dr. M. Burgener for sharing many joyful moments in- and
outside the ETH, for the interesting talks and discussions, and all the personal
and scientific support which was essential to me.
vi
Additionally, I thank the whole Baiker-group for the unique atmosphere.
Specially I would like to mention the following members for sharing a lot of
good times and unforgettable moments: Dr. R. Hess, Dr. A. Gisler, Dr. L.
Schmid, Dr. C. Beck, Dr. M. S. Schneider, Dr. W.-R. Huck, F. Jutz and N. van
Vegten.
Financial support of this work by the Swiss National Science Foundation is
kindly acknowledged.
Finally, I would like to thank my family and Mandy for their love and sup-
port throughout all these years of my education.
ix
0
0Table of contents
Acknowledgment ......................................................................................................v
Table of contents.......................................................................................................ix
Summary .................................................................................................................. xv
Zusammenfassung .................................................................................................. xxi
1 Introduction ..............................................................................................................1
1.1 Concepts of chirality .......................................................................................1
1.2 Biological significance of chirality ................................................................3
1.3 Market of chiral pharmaceuticals ..................................................................5
1.4 Methods for the production of enantiopure compounds ........................5
1.4.1 Chiral pool and biosynthesis ..............................................................6
1.4.2 Starting from the racemate - resolutions ..........................................6
1.4.3 Asymmetric synthesis ...........................................................................9Biocatalysis .........................................................................................10Homogeneous enantioselective catalysis ............................................11Heterogeneous enantioselective catalysis ...........................................12
1.5 Pt/cinchona alkaloid system ........................................................................12
1.5.1 Modifiers and solvents .......................................................................13
1.5.2 Activated ketones as substrates .........................................................15
1.5.3 Catalysts ...............................................................................................18
1.6 Aim of the thesis ............................................................................................19
1.7 References ........................................................................................................19
x
2 Inversion of enantioselectivity in the hydrogenation of ketopantolactone on
platinum modified by ether derivatives of cinchonidine ...................................27
2.1 Summary .........................................................................................................27
2.2 Introduction ....................................................................................................27
2.3 Experimental ...................................................................................................29
2.3.1 Catalytic hydrogenation ....................................................................29
2.3.2 Synthesis of O-substituted cinchonidines ......................................30
O-Methyl-cinchonidine (MeOCD) .....................................................30O-Ethyl-cinchonidine (EtOCD) ..........................................................31O-Trimethylsilyl-cinchonidine (TMSOCD) .......................................31O-Phenyl-cinchonidine (PhOCD) .......................................................31
2.4 Results and discussion ...................................................................................32
2.5 Conclusions .....................................................................................................36
2.6 References ........................................................................................................37
3 Fine tuning the “chiral sites” on solid enantioselective catalysts ....................41
3.1 Summary .........................................................................................................41
3.2 Introduction ....................................................................................................42
3.3 Experimental ...................................................................................................46
3.3.1 Materials ...............................................................................................46
3.3.2 Synthesis of ether derivatives of CD ...............................................46
O-(1-Naphthyl)cinchonidine (NaphOCD) ..........................................46O-(3,5-Dimethylphenyl)cinchonidine (XylOCD) ..............................47O-[3,5-Bis(trifluoromethyl)phenyl]cinchonidine (HFXylOCD) ........47
3.3.3 Catalytic hydrogenations ..................................................................47
3.3.4 Spectroscopic methods ......................................................................48
3.4 Results and discussion ...................................................................................48
3.4.1 Inversion of enantioselectivity ........................................................48
3.4.2 Stability of ether derivatives of cinchonidine under reaction con-
ditions ...................................................................................................53
3.4.3 Nonlinear behavior of modifier mixtures ......................................57
3.4.4 UV-Vis study of modifier adsorption and hydrogenation ........61
3.4.5 Mechanistic considerations ...............................................................65
3.5 Conclusions .....................................................................................................67
3.6 References ........................................................................................................68
xi
4 An efficient synthetic chiral modifier for platinum ...........................................73
4.1 Summary .........................................................................................................73
4.2 Introduction ....................................................................................................73
4.3 Experimental ...................................................................................................75
4.3.1 Materials ...............................................................................................75
4.3.2 Ultrasonic pretreatment ....................................................................75
4.3.3 Catalytic hydrogenations ..................................................................75
4.4 Results and discussion ...................................................................................76
4.4.1 Chemoselectivity .................................................................................76
4.4.2 Solvent effect and comparison to CD ............................................77
4.4.3 Influence of catalyst pretreatments .................................................78
4.5 Conclusions .....................................................................................................81
4.6 References ........................................................................................................81
5 Chemo and enantioselective hydrogenation of fluorinated ketones on platinum
modified with (R)-1-(1-naphthyl)ethylamine derivatives ..................................85
5.1 Summary .........................................................................................................85
5.2 Introduction ....................................................................................................86
5.3 Experimental ...................................................................................................88
5.3.1 Materials ...............................................................................................88
5.3.2 Catalytic hydrogenations ..................................................................88
5.4 Results and discussion ...................................................................................89
5.4.1 Chemoselectivity .................................................................................89
5.4.2 High-throughput screening ..............................................................91
5.4.3 Solvent effect .......................................................................................94
5.4.4 Influence of catalyst pretreatments .................................................97
5.5 Conclusions .....................................................................................................98
5.6 References ........................................................................................................99
6 The origin of chemo- and enantioselectivity in the hydrogenation of diketones
on platinum ............................................................................................................103
6.1 Summary .......................................................................................................103
xii
6.2 Introduction ..................................................................................................104
6.3 Experimental .................................................................................................106
6.3.1 Materials .............................................................................................106
6.3.2 Catalytic hydrogenations ................................................................107
6.3.3 Spectroscopy ......................................................................................107
6.3.4 Computational methods .................................................................108
6.4 Results ............................................................................................................109
6.4.1 Catalytic experiments .......................................................................109
6.4.2 NMR study ........................................................................................114
6.4.3 FTIR spectroscopy ...........................................................................116
6.4.4 Adsorption on Pt ..............................................................................121
6.5 Discussion .....................................................................................................124
6.6 Conclusions ...................................................................................................128
6.7 References ......................................................................................................128
7 Steric and electronic effects in the enantioselective hydrogenation of activated
ketones on platinum: directing effect of ester group ......................................135
7.1 Summary .......................................................................................................135
7.2 Introduction ..................................................................................................136
7.3 Experimental .................................................................................................137
7.3.1 Materials .............................................................................................137
7.3.2 Catalytic hydrogenations ................................................................138
7.3.3 Analyses ..............................................................................................139
7.3.4 VCD spectroscopy ...........................................................................139
7.4 Results and discussion ................................................................................140
7.4.1 VCD spectroscopy ...........................................................................140
7.4.2 Steric effects of substrates and modifiers ......................................141
7.4.3 Electronic effects of aryl substituents ............................................143
7.4.4 The influence of acidic medium ...................................................144
7.4.5 The role of reaction conditions ......................................................146
7.4.6 Mechanistic considerations ............................................................147
7.5 Conclusions ...................................................................................................151
xiii
7.6 References ......................................................................................................151
Outlook ..................................................................................................................157
List of Publications ...............................................................................................161
Curriculum Vitae ..................................................................................................165
0
0Summary
Heterogeneous catalytic enantioselective hydrogenation represents an efficient
and promising strategy to deliver chiral compounds for the fine chemical
industry. Among the chirally modified metal catalysts, supported Pt in the
presence of cinchona alkaloids provides the highest enantioselectivity in the
transformation of various activated ketones to functionalized chiral alcohols.
Although the number of applications has been increasing in the past years, the
fundamental understanding of the nature of enantiodifferentiation is still at an
early stage of development, compared to the state of art in homogeneous enan-
tioselective catalysis.
The aim of the thesis was to investigate the behavior of platinum modified
by new synthetic modifiers and a series of ether derivatives of cinchonidine
using the hydrogenation of fluorinated ketones and α-keto esters as test reac-
tions. Further objectives were broadening of the application range of chirally
modified platinum, and analysis of the steric and electronic effects of the sub-
strates and modifiers at aiming the elucidation of the interaction between mod-
ifier und substrate.
The enantioselective hydrogenation of ketopantolactone, an industrially
relevant reaction, was investigated over Pt modified by cinchonidine (CD) and
O-alkylated and O-arylated derivatives thereof (Chapter 2). CD provides over
90% ee to (R)-pantolactone and methylation of its OH group has only minor
effect on the enantioselection, indicating that the OH function is not involved
in the substrate-modifier interaction as a hydrogen bond donor group. An
inversion of enantioselectivity was found with ether derivatives possessing the
bulky trimethylsilyl and phenyl substituents. The study supports the mechanis-
tic models assuming that in the modifier-substrate complex CD adsorbs via the
quinoline ring lying approximately parallel to the Pt surface (π-bonding).
xvi
Bulky ether groups hinder this adsorption geometry and partly occupy the
"chiral pocket" available for adsorption of the substrate, resulting in a dramatic
influence on the stereochemical outcome of the reaction. Differences in the
adsorption strength of the modifiers due to different adsorption geometries
were corroborated by a remarkable nonlinear behavior of CD and O-phenyl-
cinchonidine (PhOCD) mixtures.
To confirm these observations, the palette of ether derivatives of CD and
that of the ketones was extended. Beside ketopantolactone, also ethyl pyruvate,
4,4,4-trifluoroacetoacetate, and 1,1,1-trifluoro-2,4-pentanedione were exam-
ined under various experimental conditions (Chapter 3). All activated ketones
(independent of the kind of activation) behaved in the same way: increasing
bulkiness of the ether substituent of CD decreased gradually the ee or even the
opposite enantiomer formed in excess (up to 53% ee). UV-Vis spectroscopy
and the nonlinear behavior of modifier mixtures supported the assumption
that in a tilted position the modifier adsorbs weaker. The tilted adsorption
mode forced by the bulky ether groups was confirmed by the lower hydrogena-
tion rate of the quinoline ring of the ether derivatives, relative to that of the
parent alkaloid. Besides, the results illustrate the efficiency of developing new
catalyst systems by fine tuning the structure of the chiral modifier, an approach
which is well known in homogeneous catalysis but only scarcely applied in het-
erogeneous asymmetric catalysis.
After fine tuning the structure of CD, the focus has been changed to
Pt/Al2O3 modified by a new synthetic chiral modifier (R,R)-pantoyl-naphthyl-
ethylamine ((R,R)-PNEA). In the asymmetric hydrogenation of α-keto esters
(R,R)-PNEA proved to be less efficient than cinchona alkaloids but the relation
was the opposite in the hydrogenation of 1,1,1-trifluoro-2,4-pentanedione
(Chapter 4). In the latter reaction (R,R)-PNEA afforded 93% ee, which result
represents the first case that a synthetic modifier is superior to a naturally
occuring chiral compound. For comparison, the highest ee achieved with CD
and O-methyl-cinchonidine (MeOCD) were 35% and 85% ee, respectively. An
interesting feature in the hydrogenation of 1,1,1-trifluoro-2,4-pentanedione
was that in the absence of modifier the chemoselectivity was poor due to com-
peting hydrogenation of the non-activated keto group. However, addition of
even trace amounts of (R,R)-PNEA completely suppressed the side reactions
Summary xvii
and only one hydroxy ketone was formed by hydrogenation of the activated
keto group.
In order to broaden the application range of Pt modified by new synthetic
modifiers, various derivatives of 1-(1naphthyl)ethylamine were tested in the
hydrogenation of α,α,α-trifluoromethyl ketones (Chapter 5). All modifiers
possessed the same anchoring moiety (naphthalene ring) that allowed strong
adsorption on the Pt surface and only the surrounding of the basic N atom was
varied systematically. The highest ee (90%) in this series was observed in the
hydrogenation of 1,1,1-trifluoro-5,5-dimethyl-2,4-hexanedione when Pt/
Al2O3 was modified by (R,R)-PNEA. The study revealed that already small
structural changes in the substrate or the modifier could strongly alter the rate
and selectivity. The high substrate specificity of heterogeneous asymmetric
hydrogenation represents a real barrier to extension of the application range of
new catalyst systems, therefore the high-throughput approach applied in this
study is helpful to accelerate the research.
As mentioned previously, in the hydrogenation of the activated carbonyl
group of 1,1,1-trifluoro-2,4-pentandione the synthetic modifier (R,R)-PNEA
induced up to 93% ee and enhanced the chemoselectivity up to 100%. To clar-
ify the origin of improved chemo- and enantioselectivity, a combined catalytic,
spectroscopic (NMR and FTIR), and theoretical approach has been applied
(Chapter 6). The examination has revealed that the changes in chemo- and
enantioselectivity are coupled and the selectivity enhancement can be attrib-
uted to the formation of an ion pair involving the protonated chiral amine
modifier and the enolate form of the substrate. Calculations have shown that
the acid-base interaction in the liquid-phase is different from that on the metal
surface. In the calculated reaction intermediate only the carbonyl group next to
the CF3 function is anchored to the surface, which adsorption mode increases
the rate of hydrogenation of the activated and hinders the transformation of
the non-activated C=O group.
In the final part of the thesis the steric effects have been studied in the Pt-
catalyzed asymmetric hydrogenation of α-keto esters by variation of the bulki-
ness at the keto and ester side of the substrates, and by using CD, its 6'-meth-
oxy derivative quinine (QN), and PhOCD (Chapter 7). Independent of the
steric bulkiness of the α-keto ester always the (R)-enantiomer formed in good
xviii
to high ee (up to 96%) in the presence of CD. Only additional steric effects in
the modifier structure (QN, PhOCD) and replacement of the weakly interact-
ing toluene by acetic acid as solvent improved the sensitivity of the catalyst sys-
tem to steric effects in the substrates (ee = 0-94%). The most important
mechanistic consequence of the results is that CD-modified Pt favors the
adsorption of the α-keto ester on the si-side, which is actually directed by the
position of the ester group and not by the steric bulkiness on any side of the
keto-carbonyl group. Other activating groups, such as the trifluoromethyl
function have similar directing effects.
0
0Zusammenfassung
Die heterogen enantioselektive Hydrierung stellt eine leistungsfähige und viel
versprechende katalytische Synthesestrategie dar, um die steigende Nachfrage
der chemischen Industrie nach chiralen Feinchemikalien zu decken. Mit Cin-
chona Alkaloiden modifiziertem Platin wurden die höchsten Enantioselektivi-
täten (ee) in der Hydrierung von aktivierten Ketonen erreicht. Obgleich die
Anzahl Publikationen und Anwendungsbeispiele solcher Katalysatorsysteme
steigt, ist das grundlegende Verständnis der einzelnen Reaktionsschritte
während der Enantiodiskriminierung noch gering, verglichen mit dem der
homogenen enantioselektiven Katalyse. Ziel der vorliegenden Promotion war
es in der asymmetrischen Hydrierung von aktivierten Ketonen die Anwend-
ungsbreite von chiral modifizierten Pt-Katalysatoren zu erweitern, sowie die
sterischen und elektronischen Einflüsse, welche die Enantiodiskriminierung
entscheidend beeinflussen, zu analysieren. Als aktivierte Ketone wurden
hauptsächlich α,α,α-Trifluoromethylketone und α-Ketoester verwendet.
Die enantioselektive Hydrierung von Ketopantolacton ist eine relevante
Reaktion in der Herstellung von Pantothensäure, welche eine wichtige Kompo-
nente des Vitamin-B-Komplexes ist. Mit Cinchonidin (CD)-modifiziertem
Platin wurde in der Hydrierung von Ketopantolacton bereits ein Enantiomer-
enüberschuss von über 90% erreicht. Im folgenden wurde das Verhalten von
O-Alkyl und O-Arylderivaten von Cinchonidin auf Pt/Al2O3 in der Hydri-
erung dieses aktivierten Ketons untersucht (Kapitel 2). Für sterisch anspruchs-
volle O-Substituenten von CD (Trimethylsily- und Phenylsubstituent) konnte
eine Inversion der Enantioselektivität festgestellt werden. Dieses Ergebnis
erhärtet das bereits publizierte Modell, in welchem im enantiodiskriminier-
enden Schritt eine Interaktion zwischen der OH Gruppe des Modifikators und
dem Substrat ausgeschlossen wird, und die Ankergruppe des Modifikators
xxii
(Chinolinring) im diastereomeren ModifikatorSubstrat-Komplex parallel zur
Platin Oberfläche adsorbiert (π-Bindung). Das Verhindern dieser parallelen
Adsorptionsgeometrie durch sterisch anspruchsvolle O-Substituenten, führt zu
einer geneigten Adsorption der Ankergruppe an der Oberfläche und damit
schliesslich zur Reduktion der Enantioselektivität. Durch die gleichzeitige Ver-
wendung von zwei Modifikatoren, welche jeweils die Bildung verschiedener
Enantiomeren bevorzugen, konnten nichtlineare Effekte untersucht werden,
die Rückschlüsse auf die Adsorptionsstärken und Adsorptionsgeometrien der
einzelnen Modifikatoren zuliessen.
Um diese Ergebnisse zu verallgemeinern wurden die Palette an Etherderi-
vaten von CD erweitert und zusätzlich zu Ketopantolacton ausserdem Eth-
ylpyruvat, 4,4,4-Trifluoroacetoacetat und 1,1,1-Trifluoro-2,4-pentandion als
Substrate in der asymmetrischen Hydrierung über modifiziertem Platin unter
Verwendung verschiedener Lösungsmitteln und Drücke untersucht (Kapitel
3). Alle Ketone, unabhängig von der Art ihrer Aktivierung, zeigten ein identis-
ches Verhalten: mit zunehmender Grösse des O-Substituenten nahm die Enan-
tioselektivität ab oder es bildete sich gar das nicht zu erwartende Enantiomer
im Überschuss (bis zu 53% ee). UV-Vis-Spektroskopie Untersuchungen, sowie
Experimente zum nichtlinearen Verhalten von Modifikatormischungen stützen
die Aussage, dass der Modifikator schwächer adsorbiert wird, sobald eine paral-
lele Adsorption der Ankergruppe (π-Bindung) nicht mehr möglich ist. Unter-
schiedliche Hydrierungsgeschwindigkeit der Ankergruppe von CD, sowie von
O-Alkyl und O-Arylderivaten von CD bestätigen, das im letzteren Fall der Chi-
nolinring in einer geneigten Geometrie zur Oberfläche adsorbiert. Im weiteren
veranschaulichen die Resultate das Potenzial für die Entwicklung von neuen
Katalysatorsystemen durch Feinopti-mierung der Modifikatorstruktur.
Durch Verändern der Struktur des natürlichen Modifikators CD konnten
in den vorangehenden zwei Kapiteln wichtige Erkenntnisse über das Adsorp-
tionsverhalten von CD auf Platin in der Hydrierung von aktivierten Ketonen
gewonnen werden. Im folgenden Kapitel wurde das Potenzial eines neuen, syn-
thetischen Modifikators ((R,R)-Pantoyl-naphthylethylamin, PNEA) in der
asymmetrischen Hydrierung von 1,1,1-Trifluoro-2,4-pentandion auf Pt/
Al2O3 valuiert (Kapitel 4). Mit PNEA-modifiziertem Platin wurde bereits in
der enantioselektiven Hydrierung von α-Ketoestern hohe Selektivitäten erre-
Zusammenfassung xxiii
icht. In der racemischen Hydrierung von 1,1,1-Trifluoro-2,4-pentandion fin-
det in der Abwesenheit des Modifikators zusätzlich die Reduktion der nicht
aktivierten Carbonylgruppe statt. In Gegenwart bereits geringer Mengen an
PNEA wurde jedoch diese Nebenreaktionen unterdrückt und es erfolgt nur die
Hydrierung der aktivierten Carbonylgruppe. Nach optimieren der Reaktion-
sparameter wurden mit PNEA-modifiziertem Platin bis zu 93% ee in der
Hydrierung dieses fluorierten Ketons erreicht, bei einer Chemoselektivität von
100%. Dieses Ergebnis stellen den ersten Fall dar, in welchem mit einem syn-
thetischen Modifikator höhere ees erreicht wurden als mit natürlich vorkom-
menden Modifikatoren. Die höchsten mit CD und O-Methyl-cinchonidin
(MeOCD) erzielten Selektivitäten ergaben 35% und 85% ee.
Um die Einsatzbreite dieses Katalysatorsystems in der Hydrierung von
weiteren fluorierten Ketonen zu testen, wurde die Substratpalette durch andere
α,α,α-Trifluoromethylketone erweitert. Gleichzeitig wurde durch Verwenden
von PNEA-ähnlichen, chiralen Aminen die Variabilität von synthetischen
Modifikatoren analysiert (Kapitel 5). Alle neuen Modifikatoren besitzen die-
selbe Ankergruppe wie PNEA (Naphthalinring), welche für eine Adsorption
auf der Katalysatoroberfläche zuständig ist und nur die direkte Nachbarschaft
des Stickstoffatoms von PNEA wurde systematisch variiert. Die erfolgreichste
Substrat-Modifikator-Kombination ergab sich in der Hydrierung von 1,1,1-
Trifluoro-5,5-dimethyl-2,4-hexandion über PNEA-modifiziertem Platin, in
welcher eine Enantioselektivität von bis zu 90% erreicht wurde. Die Ergebnisse
zeigen, dass bereits kleine Veränderungen in der Substrat- oder Modifikator-
struktur die Ausbeute und Enantioselektiviät stark beeinträchtigten. Auf
Grund der limitierenden Substratselektivität von heterogen, asymmetrischen
Hydrierungssystemen vereinfachten Hochdurchsatz-Screening-Methoden die
Suche nach einer erfolgreichen Substrate-Modifikator-Kombination positiv.
Da in der enantioselektiven Hydrierung von 1,1,1-Trifluoro-2,4-pentan-
dion über Platin, sowohl mit synthetischen, also auch mit natürlichen Modi-
fikatoren hohe Enantioselektivitäten erreicht wurden, verwendete man in den
folgenden Untersuchungen katalytische Ergebnisse, NMR- und FTIR-Resul-
tate, sowie computergestützte Berechnungen um die erzielten Chemo- und
Enantioselektivitäten in Gegenwart dieser Modifikatoren besser verstehen zu
können (Kapitel 6). Die Auswertungen der Ergebnisse zeigten auf, dass
xxiv
Chemo- und Enantioselektivität stark voneinander abhängen. Des weiteren ist
die Verbesserung der Chemoselektivität in Gegenwart des Modifikators
gegenüber der unmodifizierten Reaktion, auf eine optimale Anordnung der
Enolatform des Substrates und des protonierten Modifikators auf der Ober-
fläche zurückzuführen. Diese besondere Interaktion zwischen Substrat und
Modifikator führt zur Beschleunigung der Hydrierung der aktivierten und zur
gleichzeitigen Blockierung der nicht aktivierten CO Gruppe. Die Berechnun-
gen zeigen aussderm, dass die Interaktion in der Flüssigphase nicht so entschei-
dend und spezifisch ist, wie die Wechselwirkung auf der Oberfläche.
Im abschliessenden Kapitel dieser Arbeit wurde der sterische Einfluss von
verschiedenen Keto- und Estersubstituenten in der asymmetrischen Hydri-
erung von α-Ketoestern über modifiziertem Platin untersucht (Kapitel 7). Als
Modifikatoren wurden CD, sowie das 6'-Methoxyderivat von CD Quinin
(QN) und PhOCD verwendet. Unabhängig vom räumlichen Anspruch der
Derivatisierung des α-Ketoesterns wurde mit CD-modifizertem Platin bis zu
96% ee erreicht wobei immer das (R)-Enantiomer dominierte. Nur durch
zusätzlich sterisch anspruchsvolle Substituenten in der Modifikatorstruktur
(QN, PhOCD), oder durch Wechseln des Lösungsmittel von Toluol zu Essig-
säure konnte die Empfindlichkeit des Katalysatorsystems hinsichtlich Ausbeute
und Enantioselektivität verbessert werden (ee = 0 - 94%). Das wichtigste mech-
anistische Ergebnis dieses Kapitels ist die Folgerung, dass die bevorzugte Seite
des α-Ketoesters, mit welcher er auf der Oberfläche adsorbiert (si-Seite), nicht
durch den sterischen Anspruch des Keto- und Estersubstituenten des Sub-
strates bestimmt wird, sondern durch die Position der Estergruppe in der 1:1
Wechselwirkung zwischen Substrat und Modifikator. Andere aktivierende
Gruppen, wie die Trifluoromethylgruppe, besitzen ähnlich dirigierende Effekte
welche für die Adsorption des Substrates entscheidend sind.
Chapter1
1Introduction
1.1 Concepts of chirality
In 1815 the French physicist Jean-Baptiste Biot reported first the chirality ofmolecules. He made the observation that α-quartz rotated the plane of polar-
ized light [1]. Louis Pasteur (Figure 1-1), a French chemist succeeded in 1848
in the first chiral separation. He was able to separate the mirror image crystals
of racemic sodium ammonium tartrate isomers by the use of magnifying glass
and tweezers and postulated that the enantiomers have different three-dimen-
sional arrangements [2]. They are mirror-images of each other on the micro-
scopic as on the macroscopic level.
Later in 1874, the French chemist Le Bel [3] and the Dutch chemist van´t
Hoff (Figure 1-1)[4], independently postulated that the four chemical bonds
Fig. 1-1: Portraits of the French chemist Louis Pasteur (left, 1822-1895) and the Dutchchemist Jacobus Henricus van’t Hoff (right, 1852-1911).
2 Chapter 1
that carbon atoms can form are directed to the corners of a tetrahedral struc-
ture. This discovery proved to be the cornerstone in the study of three-dimen-
sional structures in organic chemistry, and developed to what is commonly
referred to as stereochemistry.
Stereochemistry is today the study of the static and dynamic aspects of
three-dimensional shapes of molecules. It has long provided a foundation for
understanding structure and reactivity. Many chemists find this area of study
fascinating due simply to the aestetic beauty associated with chemical struc-
tures, and the challenging ability to combine the fields of geometry, topology,
and chemistry in the three-dimensional shapes. In general, carbon atoms that
have four different groups attached to them are not superimposable on their
mirror image are said to be asymmetric, or chiral (Greek: χειρ (cheir), meaning
hand) (Figure 1-2). According to Mislow and Siegel, this carbon atom is a ste-
reogenic centre of a molecule [5]. These mirror images are termed enantiomers.
Fig. 1-2: Molecules or objects that are non-identical with their mirror image are said to bechiral. Furthermore, nearly all amino acids in the human body are left-handed.
Introduction 3
The word enantiomer is derived from the Greek εναντιος (enantios), which
means opposite. With the exception of the optical rotation of polarized light,
both enantiomers show the same chemical and physical behavior in an achiral
environment (e.g. melting points, boiling points, densities). In a chiral envi-
ronment the enantiomers react differently.
1.2 Biological significance of chirality
The majority of important building blocks which make up the biological mac-
romolecules of living systems such as amino acids, sugars, proteins and nucleic
acids are chiral. In nature these molecules exist in only one enantiomeric form,
e.g. amino acids in the L-form and sugars in the D-form. Therefore, in a chiral
environment, for example that of living organisms, enantiomers can show dif-
ferent chemical behaviour due to different chiral discrimination and may lead
to different effects [6-8].
Most chemists are familiar with the role of chirality on odorants, such as
carvone and limonene, which provide classic examples of how enantiomerism
can lead to different interactions in the human body (Figure 1-3). The enantio-
mers of these odorants have different scents caused by a different 3-D fit on an
odour receptor and/or on different odour receptors [9,10]. One can imagine
Fig. 1-3: Enantiomers of carvone and limonene and their corresponding odours.
(S)-(+)-carvone (R)-(-)-carvone (R)-(-)-limonene (S)-(+)-limonene
caraway spearmint orange lemon
4 Chapter 1
the different pharmacological activities of enantiomers, as well as different
pharmacokinetic and pharmacodynamic effects, can even be more impressive
[11]. Thus, one isomer may produce the desired therapeutic activities, while
the other is inactive or, in worst case, produce unwanted effects. Consider as an
example (-)-propanolol, which was introduced in the 1960s as a β-blocker for
the treatment of heart disease (Figure 1-4). After some time the (+)-enantiomer
was identified to act as an effective contraceptive.
The same principles are important for herbicides and pesticides in agro-
chemistry containing stereogenic centers. For example, the (R)-(+)-enantiomer
of the herbicide dichlorprop is the active enantiomer in killing the weeds, while
the (S)-(-)-enantiomer is inactive as an herbicide (Figure 1-5). In the sense of
Fig. 1-4: The different pharmacological activities of propanolol enantiomers.
Fig. 1-5: The active and inactive enantiomers of the herbicide dichlorprop.
(-)-propanolol (+)-propanolol
β-blocker for the contraceptivetreatment of heart disease
(R)-(+)-dichlorprop (S)-(-)-dichlorprop
active in killing weeds inactive
Introduction 5
economical and ecological properties the inactive enantiomer must be regarded
as isomeric ballast and therefore the preparation of the enantiopure agent is
preferred.
1.3 Market of chiral pharmaceuticals
Although the concept of chirality was known over a century ago, awareness of
how this affects the pharmacological activity of drugs in pharmaceutical appli-
cations is much more recent. For many years it was common practice to sell
synthetic chiral drugs as racemates. Nowadays systematic investigation of the
pharmacological activity of the individual enantiomers became the rule for all
new racemic drugs and chiral considerations are an integral part of drug
research and development. In 1987 the U.S. Food and Drug Administration
issued a guideline that drug companies should examine individual enantiomers
in racemic mixtures preclinically and clinically, and for chiral drugs only its
therapeutically active isomer should be brought to market [13-15]. For the
approval of a racemate of chiral drugs rigorous justifications are required from
the market. However, a large number of chiral drugs are still sold as racemic
mixtures [16]. The main reasons for pharmaceutical industry to develop single
enantiomers are beside some ethical and economical reasons, the therapeutic
benefit (efficiency and safety) and, in several instances, extension of the life
cycle of the drug. Hence there has been a rapid development of enantioselective
synthetic technologies in the past two decades, which have reached a high
degree of diversity and complexity.
1.4 Methods for the production of enantiopure compounds
There are several basically different strategies to produce enantiopure com-
pounds. The most important approaches include [17,18]:
6 Chapter 1
- synthetic routes starting from a chiral synthon (chiral pool),
- resolution of racemates into pure enantiomers, and
- stereoselective reactions of prochiral substrates (asymmetric synthesis).
In the next few sections these three main approaches will be described shortly
and illustrated by several examples from literature.
1.4.1 Chiral pool and biosynthesis
Enantiopure compounds are either directly avaliable from natural source by
extration from plants and living sources such as animal and human beings, or
by synthetic routes starting from a synthon of the chiral pool. The chiral pool
consits of enatiomerically pure starting materials derived from natual resources,
such as amino acids, carbohydrates, hydroxy acids, alkaloids, and terpenes [18].
If the right source of a substrate is available from the chiral pool, then this
pathway is often the most cost-effective way of introducing asymmetry [19]. In
biotechnologie new strategies with genetically modified organism has received
increasing attention extending the number of processes and the variety of avail-
able compounds. The advantage of this method is that 100% purity in reson-
able quantities can be obtaind, also for highly complex molecules.
Disadvantages of biosynthetic pathways are, beside the low volume yields, the
difficulties which arise from the public acceptance of the use of genetically
modified organisms. Important examples from industry are the production of
the cardiovascular drug enalapril (Vastec®, Merck) whose chiral centres are
introduced by L-alanine and L-proline. Another example is the synthesis of the
herbicide (R)-flamprop-isopropyl starting from the well known chiral hydroxy
acid L-lactic acid [20] (Figure 1-6).
1.4.2 Starting from the racemate - resolutions
To obtain optically active compounds which are not derived from natural
sources crystallization is still one of the most important methods, despite its
‘low technique’-reputation. There are two different crystallization techniques:
direct crystallization and diasteromer crystallization.
Introduction 7
An efficient way to obtain pure enantiomers from enantiomeric mixtures
represents direct crystallization, but it is only possible if the mixture is a con-
glomerate. In a conglomerate the equimolar racemic mixture has a lower melt-
ing point than the pure enantiomers. Therefore a conglomerate and racemic
compounds possess different phase diagrams. By seeding a supersaturated race-
mic solutions by a small amount of the corresponding pure crystals the enanti-
omerically pure compound crystallize out. This technique was developed for
different industrial scale crystallization processes, such as the production of
menthol through fractional crystallization of benzoate ester intermediates
(Figure 1-7). Unfortunately, for 1:1 mixtures of racemic compounds, which
account for most of all racemates, this separation technique can not always be
applied successfully due to formation of crystals containing both enantiomers
[18]. Still, this method is helpfull in combination with other enantioselective
methods.
The most often used technique to separate racemates is the diastereomer
derivatisation (diastereomer salts) by enantiomerically pure resolving agent
which is normally a low cost compound from the chiral pool. The obtained
diastereomers have different physical properties that allows a simple crystalliza-
Fig. 1-6: Synthesis of the enantiopure herbicide (R)-(-)-flamprop-isopropyl from L-lacticacid.
(R)-(-)-flamprop-isopropyl
(S)-lactic acid
8 Chapter 1
tion, chromatography, or other physical manipulation. For instance, when the
solubilities of the two diastereomer salts are very different, one of the salts is
insoluable and can be filtered of the mixture, leaving the other in solution.
Treatment with an acid or base finally decompose the salt and the resolving
agent can be recovered. Two illustrative examples of classical resolution tech-
niques employed in industrial processes, are the diastereomeric crystallization
of D-phenylglycine from the racemate [21] and the resolution of racemic pan-
tolactone via diastereomer salt crystallization with quinine [22].
Beside the direct and diastereomer crystallization techniques, kinetic reso-
lution (via fermentation, bio- or chemo-catalyst) represents a further way
which is widely applied in industry as resolution of racemic mixtures. In the
kinetic resolution process one enantiomer reacts faster with a chiral auxiliary
and thus the enantiomeric excess of the residual substrate increases continu-
ously with conversion. This difference in rate is induced by chemical catalysts
or biocatalysts like enzymes. A representative example of enzymatic resolution
is involved in the preparation of benazepril (Figure 1-8), which is one of the
Fig. 1-7: Industrial production of menthol through fractional crystallization.
menthol
Introduction 9
most potential angiotensin converting enzyme (ACE) inhibitors. By using the
kidney acetone powders (KAPs) derived from different mammalian species it
was possible to resolve racemic N-acetyl HPA providing reasonable yields
(41%) and excellent enantiomeric excess (>99%) [23].
1.4.3 Asymmetric synthesis
Another approach is the asymmetric synthesis where a pure enantiomer is pro-
duced under the influence of a chiral auxiliary. The chiral auxiliary interacts
with or is temporarily connected to the prochiral compound, or is part of the
catalyst [24]. The disadvantage of this approach is that up to one equivalent of
the chiral auxiliary has to be applied which can be expensive if the auxiliary is
not recyclable [25]. A similar approach, which is from an economic point of
view much more interesting is the use of small amounts of a chiral catalyst to
produce chiral materials in large quantities (chirality multiplication). This
approach is called asymmetric catalysis and depending on the type of catalyst
Fig. 1-8: Enzymatic resolution of (±)-N-acetyl-HPA with mammalian KAP in the synthe-sis towards benazepril.
10 Chapter 1
used it can be distinguished between biological processes, homogeneous enan-
tioselective catalysis, and heterogeneous enantioselective catalysis.
Biocatalysis
Biological processes are usually regulated by enzymes. Instead of an isolated
enzyme a whole microorganism can also be used [26]. Maybe the oldest
method for the synthesis of chiral compounds using this approach is the utili-
zation of microorganisms in fermentation processes. The big advantage of fer-
mentation processes lies in the potential to produce molecules of high
complexity such as in the production of Vitamine B12 or penicillin. These
reactions are normally running with very high selectivity (usually 100%) and in
high purity. Difficulties in finding an appropriate microorganism which is able
to produce the desired product can sometimes easily be solved by using gene
technology. According to this technology microorganisms are modified to pro-
duce different interesting products. Despite some negative features, such as the
often low volumetric productivity or the thermal instability of biological sys-
tems [27], many applications are established for industrial production [28]. An
important example from industry is the production of (S)-3,4-dihydroxyphe-
nylalanine (L-DOPA), which is utilized in the treatment of Parkinson’s disease
and which represents a precursor to the neurotransmitter dopamine used by the
enzyme tyrosine phenol-lyase (Tpl). In the industrial one-pot synthesis cate-
chol, pyruvic acid, and ammonia are combined in the presence of cells from the
Erwinia herbicola containing the Tpl-biocatalyst [29] (Figure 1-9).
Fig. 1-9: Production of L-DOPA using tyrosine phenol-lyase.
Introduction 11
Homogeneous enantioselective catalysis
For the enantioselective transformation of a prochiral substrate in homoge-
neous catalysis a chiral catalyst, typically in the form of an organic transition
metal complex is used. The chiral organic ligands must possess suitable func-
tionality, configuration, and conformational rigidity or flexybility to afford the
desired stereoselectivity. A selection of some typical ligands for the enantiose-
lective hydrogenation of ketones is shown in Figure 1-10 [25]. The substrate is
activated by coordination to the metal atom and the chiral environment in the
ligand sphere is responsible for the enantiodifferentiation.
Homogeneous asymmetric catalysis has emerged since 1968 to an impor-
tant production method for some commercially relevant intermediates. The
key reactions include the hydrogenation of C=O and C=C bonds, as well as the
epoxidation of olefins and allylic alcohols. This approach has received its due
recognition in the 2001 Nobel Prize to W. S. Knowles and R. Noyori for enan-
tioselective hydrogenation and to K. B. Sharpless for enantioselective oxidation
catalysis [30].
Fig. 1-10: Chiral ligands widely used in homogeneous enantioselective hydrogenation.
12 Chapter 1
Heterogeneous enantioselective catalysis
Compared to the number of highly selective homogeneous asymmetric cata-
lysts, heterogeneous asymmetric catalysts are still lagging behind as concerns
the performance, the mechanistic insight, and the scope of reaction. A compa-
rable development as in homogeneous asymmetric catalysis has not (yet) taken
place in chiral heterogeneous catalysis due to various reasons, the most impor-
tant being the difficulty to create well-defined catalytically active and stable
sites on a solid surface [31]. Compared to homogeneous catalysis, heteroge-
neous catalysis deals with insoluable catalysts, what offers them inherent practi-
cal advantages connected with separation, reuse, and stability of the catalyst.
Furthermore heterogeneous catalysts offers the opportunity for continuous
process operation.
The most promising strategies for creating a heterogenous asymmetric cat-
alysts are chirally modified metals, chiral polymers, and immobilised chiral
transition metal complexes [32]. In the field of hydrogenation the three most
important asymmetric catalysts are the (i) nickel catalysts modified with tar-
taric acid, useful for β-functionalized ketones with ees up to 98.6% [33-38],
(ii) platinum catalysts modified with cinchona alkaloids and related modifiers,
for α-functionalized ketones with ees up to 98% [39,40], and (iii) palladium
catalysts modified with cinchona alkaloids which provide ees up to 94% for
selected α-functionalized C=C bonds [41-45]. In these catalyst systems the
chiral information can be imparted to an achiral catalytically active metal sur-
face by adsorption of promising chiral organic molecules (chiral modifiers) that
interact with the co-adsorbed prochiral substrates in such a way that enantio-
differentiation occurs [46-51].
1.5 Pt/cinchona alkaloid system
The enantioselective hydrogenation of an α-keto ester (methyl and ethyl pyru-
vate) over platinum catalyst modified by cinchona alkaloids was first described
by Orito et al. in 1979 [52-55]. In Figure 1-11 the reaction scheme is illus-
trated with the well known modifier cinchonidine (CD). Since the discovery of
Introduction 13
this catalyst system several research groups have worked on this field and many
ongoing publications today show the up-to-dateness of this topic in catalysis.
The hydrogenation of ethyl pyruvate is the most studied heterogeneous enan-
tioselective hydrogenation and serves as a widely used model compound for
chirally modified Pt [56,57]. After optimization over several years, the highest
ee for ethyl pyruvate hydrogenations under mild conditions is 98% by quinine-
modified Pt/Al2O3 [58].
In the next chapters some important characteristic features of the plati-
num-cinchona system such as the choice of catalyst, modifier, substrate and
solvent, will be discussed.
1.5.1 Modifiers and solvents
Adsorption of trace amounts of a chiral modifier on the metal surface gen-
erates a ”chiral pocket” for adsorption and interaction of the substrate, and
thus introduce the “chiral information” to the achiral surface. Besides, the
modifier can remarkably accelerate the reaction compared to the reference reac-
tion on an unmodified surface. In α-keto ester hydrogenation already very
small amounts of cinchonidine (4-1000 ppm related to the substrate) are
enough to reach over 90% ee and increase the reaction rate by a factor of 10-
100 [59,60].
Fig. 1-11: Enantioselective hydrogenation of α-keto ester over platinum chirally modifiedby cinchonidine first published by Orito et al [53].
14 Chapter 1
Early studies concluded that the minimal requirements for an efficient
modifier for the hydrogenation of α-keto esters is the presence of a basic
nucleophilic nitrogen atom close to one or more stereogenic centers and con-
nected to an extended aromatic system (anchoring group) responsible for
adsorption on the surface [57,61]. Note that a basic nitrogen atom does not
seem to be absolute necessary due to the fact that (R)-1-napthyl-1,2-ethanediol
induced in the enantioselective hydrogenation of ketopantolactone up to 30%
ee under mild reaction conditions. The sense of chiral induction of cinchoni-
dine is mainly determined by the absolute configuration at C8 (Figure 1-12).
Cinchonidine and O-metyl-cinchondine (MeOCD) were found to be the best
modifiers for α-keto ester hydrogenation [59,62]. Several other natural [63-66]
and synthetic compounds [56,67-71] have been tested in α-keto ester hydroge-
nation, but only a few turned out to be successful. Synthetic modifiers such as
the structurally simple (R)-2-(1-pyrrolidinyl)-1-(1-napthyl)ethylalcohol (PNE,
Figure 1-12), and (R)-1-(1-naphthyl)ethylamine (NEA, Figure 1-12) and some
of their derivatives afforded up to 87% ee in the hydrogenation of ethyl pyru-
Fig. 1-12: Some natural and synthetic chiral modifiers for enantioselective hydrogenationson Pt.
Introduction 15
vate to ethyl lactate [67,68,72]. The presence of an OH group is not an essen-
tial requirement to an effective modifier as demonstrated by the performance
of the simple amine NEA demonstrates. Alkylation of the quinuclidine nitro-
gen of cinchonidine resulted in the complete loss of enantioselectivity. Further-
more, partial hydrogenation of the quinoline ring is a competing side reaction
and can also decrease the enantioselectivity. At lower pressures higher selectivi-
ties can be achieved by replacing the quinoline moiety by a 1-napthyl ring [73].
A single phenyl ring as anchoring group is not enough to enable the necessary
rigidity in the adsorption of the modifier on the catalyst surface, therefore an
extended aromatic ring system is necessary [67,74].
The solvent polarity and acidity can have a significant effect on the enan-
tioselectivity and rate. The solvent mainly influences the solubilty of hydrogen,
the adsorption of substrate and modifier on the Pt surface, and change further-
more the conformation of the cinchona alkaloid. The conformational behav-
iour of CD in solution is very important for understanding the possible
substrate-modifier interactions. It has been shown earlier by a combined NMR
and ab inito study that high population of the “open (3)” [75-77] or “anti-
open” [78] conformation of CD is beneficial for enantioselection.
The most effective modifier-solvent combinations are MeOCD in acetic
acid and CD in toluene. Most studies used as solvent toluene or acetic acid. An
acid additives (e.g. trifluoroacetic acid) has often been used in combination
with weakly polar solvents. A good negative correlation between the enanan-
tioselectivity and the solvent polarity was found for the hydrogenation of
α-keto esters, when the solvent polarity was increased the enantioselectivity
diminished [79]. A positive correlation was observed for the cinchona-modi-
fied hydrogenation of trifluoro β-keto esters [80]. The only deviation from the
negative correlation is the ee in acidic solvents, which value is higher than
expected from the solvent polarity. This deviation is due to the protonation of
CD in acids [79].
1.5.2 Activated ketones as substrates
One important limitation of chirally modified hydrogenation catalysts is their
high substrate specificity, i.e., only a few types of substrates are transformed
16 Chapter 1
with high selectivities. But in the past decade the number of substrates, which
could be hydrogenated by Pt with reasonable ee, increased remarkably. All sub-
Table 1-1: Best ees achieved in the hydrogenation of various α-ketoacid derivatives.
reactant modifier, solvent, conditions ee [%] ref
CD, AcOH, 40 bar, 25°C, R=Me
MeOHCD, AcOH, 10 bar, 25°C, R=Et
98
97
[39]
[84]
MeOHCD, AcOH, 10 bar, 25°C
HCD, AcOH, 5.8 bar, 17°C
96
94
[84]
[85]
HCD, AcOH/toluene, 25 bar, 0°C 98 [40]
MeOHCD, AcOH, 20 bar, 20°C 96 [86]
HCD, toluene, 60 bar, 25°C 86 [87]
CD, AcOH, 60 bar, 20°C 60 [88]
CD, toluene, 70 bar, -13°C 92 [89]
CD, toluene, 10 bar, 17°C 91 [89]
Introduction 17
strates possess a common feature, an electron withdrawing group (ketone, ester,
amide, carboxyl, or acetal) in α-position to the keto-carbonyl group being
reduced [81].
Various strategies were applied for increasing the ee (special colloids, use of
sonication, solvent mixtures, acids [82] and bases [83] as additives) but, not
surprisingly, the reaction conditions had to be optimized for each substrate.
The best examples with the highest ee of α-ketoacid derivatives hydrogenation
over cinchona-modified Pt are listed in Table 1-1. Most reactions are fast and
give full conversion and excellent selectivities up to 98% ee at ambient temper-
ature, within less than 1 h reaction time. The rapid hydrogenation of the vinyl
group of CD, leading to HCD, has practically no influence on the efficiency of
the alkaloid.
The trifluoromethyl group was identified as having a similar activating
property on the carbonyl group as an ester group [90]. In the asymmetric
Table 1-2: Best ee for various activated ketones.
reactant modifier, solvent, conditions ee [%] ref.
MeOCD, THF/TFA, 10 bar, 20°C,
R=CH2COOEt
CD, toluene, 10 bar, 0°C, R=Ph
96
92
[80]
[91]
MeOCD, AcOH, 60 bar, 25°C 98 [97]
CD, AcOH, 1bar, 25°C, R=Me
MeOCD, AcOH, 60 bar, 25°C,
R=(CH2)3
97
97
[98]
[100]
CD, CH2Cl2, 5 bar, 25°C, R=Ph
CD, toluene, 107 bar, 25°C, R=Me
94
90
[99]
[101]
18 Chapter 1
hydrogenation of methyl, ethyl and iso-propyl esters of 4,4,4-trifluoroacetoace-
tate ees of 96% were achieved in acetic acid or trifluoroacetic acid/THF mix-
tures. MeOCD turned out to be a significantly more efficient modifier than
CD [91,92,93]. Furthermore, various trifluoroacetophenone derivatives with
additional CF3 or N(Et)2-substituents on the aromatic ring gave ees between 36
and 92%, but ee and TOF were usually higher without any substituent at the
aromatic ring [94]. Other CF3-substituted ketones were also tested, but except
for 2-trifluoroacetylpyrrole (63% ee) none of them gave ees significantely above
20% [95,96]. Compared to α-keto esters, α,α,α-trifluoromethyl ketones can
easily form hemiketals, which can negatively influence the enantioselectivity
during the hydrogenation [92].
It has been shown that also keto, amide, carboxyl, ether, and acetal groups
in α-position can sufficiently activate ketones regarding asymmetric hydroge-
nation, and in Table 1-2 some results are summarized.
1.5.3 Catalysts
In the asymmetric hydrogenation of activated ketones supported Pt catalysts in
combination with cinchona alkaloids and related modifiers are preferred
[46,53,56,102]. Supported Pd is the metal of choice for the hydrogenation of
functionalized C=C bonds and for C-Cl hydrogenolysis [103-106]. Different
supports have been used, including Al2O3, SiO2, TiO2, various carbon sup-
ports and zeolites. Catalysts with metal particles less than 2 nm were reported
to be less selective and they also showed lower turnover frequencies [107]. In
general, high metal dispersions are detrimental for the enantioselection. Two
commercially available 5 wt% Pt/Al2O3 have emerged as ”standard” catalysts:
E4759 from Engelhard and JMC 94 from Johnson Matthey [62]. Both cata-
lysts have dispersions of ca. 0.2 - 0.3. While E4759 has rather small pores and
low pore volume, JMC 94 is a widepore catalyst with a large pore volume.
Despite of these differences, both catalysts offer high ees and reasonable rates
for a variety of α-functionalized ketones.
In many reactions the ee is doubled when the 5 wt% Pt/Al2O3 is pretreated
in hydrogen at elevated temperatures (300-400°C) [54,99,108]. The beneficial
effect of reductive catalyst pretreatment was attributed to an increase of particle
Introduction 19
size, a change of the particle shape, and cleaning of the surface from impurities.
Another way to improve the catalyst performance is an ultrasonic pretreatment
in solution in the presence of modifier and hydrogen [84,109,110]. It is
assumed that sonication restructures the catalyst and Pt-particle of the optimal
size and surface structure are obtained which also leads to an increased surface
density of the modifier.
1.6 Aim of the thesis
The aim of this thesis was to extend the application range of chirally-modified
platinum and to re-investigate the steric and electronic effects controlling the
stereochemical outcome of the heterogeneous enantioselective hydrogenation
of activated ketones. Two major classes of reactants were involved in the stud-
ies: α,α,α-trifluoromethyl ketones and α-keto esters. The new chiral modifiers
of platinum included ether derivatives of cinchonidine with varying bulkiness
of the ether group and a series of 1-(1-naphthyl)ethylamine derivatives where
the surroundings of the interacting N-atom was modified systematically.
Besides, the structure of the substrates was varied by introducing bulky and
electron-withdrawing or electron-releasing substituents. High-throughput cata-
lyst screening was used to collect sufficient amount of experimental data that
could support the mechanistic considerations.
1.7 References
[1] G. M. Loudon, Addison-Wesley Publishing Co., Massachusetts, 1984.
[2] L. Pasteur, Am. Chim., Phys. 1848, 24, 442.
[3] J. A. Le Bel, Bull. Soc. Chim. France 1874, 22, 337.
[4] J. H. van't Hoff, Bull. Soc. Chim. France 1875, 23, 295.
[5] K. Mislow, J. Siegel, J. Am. Chem. Soc. 1984, 106, 3319.
20 Chapter 1
[6] H. Frank, B. Holmstedt, B. Testa, Chirality and Biological Activity, H. R.
Liss Inc, New York, 1990.
[7] S. Ahuja, Chiral Separation by Liquid Chromatography, ed. S. Ahuja, Wash-
ington D.C., 1991.
[8] C. J. Welch, Advances in Chromatography, Vol. 35, Marcel Dekker, New
York, 1995, p. 171.
[9] E. Brenna, C. Fugati, S. Serra, Tetrahedron: Asymmetry 2003, 14, 1.
[10] M. H. Boelens, H. Boelens, L. J. Van Gemert, Sensory Properties of Optical
Isomers, Perfumer and Flavorist 1993, 18, 1.
[11] D. E. Drayer, Clin. Pharmacol. Ther. 1986, 40, 125.
[12] N. Y. Grigorieva, P. G. Tsiklauri, Russ. Chem. Rev. 2000, 69, 573.
[13] M. Strong, Food and Drug L. J. 1999, 54, 463.
[14] S. K. Branch, in: Chiral Separation Techniques: A Practical Approach, ed. G.
Subramanian, Wiley-VCH, Weinheim, 2001.
[15] S. Stinson, Chem. Eng. News 1995, 73, 49.
[16] B. C. Lin, X. F. Zhu, B. Koppenhoefer, U. Epperlein, LC GC-Mag. Sep.
Sci. 1997, 15, 40.
[17]M. Breuer, K. Dittrich, T. Habicher, B. Hauer, M. Kesseler, R. Stürmer, T.
Zelinski, Angew. Chem. Int. Ed. 2004, 43, 788.
[18] R. A. Sheldon, Chirotechnology: Industrial Synthesis of Optically Active
Compounds, M. Dekker, New York, 1993.
[19] A. N, Collins, G. N. Sheldrake, J. Crosby, Chirality in Industry II, Wiley,
New York, 1997.
[20] R. M. Scott, G. D. Armitage, DE Pat. 2, 946, 652, 1980.
[21] M. A. Wegman, M. H. A. Jannssen, F. Van Rantwijk, R. A. Sheldon, Adv.
Synth. Cat. 2001, 343, 559.
[22] S. A. Harris, K. Folkers, US Pat. 2319545, 1940.
[23] I. Regla, H. Luna, H. I. Pérez, P. Demare, I. Bustos-Jaimes, V. Zaldívar,
M. L. Calcagnac, Tetrahedron: Asymmetry 2004, 15, 1285.
[24] H. B. Kagan, J. C. Fiaud, Top. Stereochem. 1987, 10, 175.
[25] R. Noyori, Asymmetic Catalysis in Organic Reactions, Wiley, New York,
1994.
[26] K. Faber, Biotransformation in Organic Chemistry, Springer-Verlag, Berlin,
2004.
Introduction 21
[27] B. Testa, in: Chirality and Biological Activity, eds. H. Frank, B. Homstedt,
B. Testa, Symp. held in Tübingen in 1988, (1990).
[28] A. N. Collins, G. N. Sheldrake, J. Crosby, Chirality in Industry II Wiley,
New York, 1997.
[29] H.-U. Blaser, E. Schmidt, Asymmetric Catalysis on Industrial Scale - Chal-
lenges, Approaches and Solutions, Wiley, New York, 2004.
[30] For details see Nobel Lectures, Angew. Chem. Int. Ed. 2002, 41, 1998.
[31] A. Baiker, Catal. Today 2005, 100, 159.
[32]A. Tungler, T. Tarnai, L. Hegedüs, K. Fodor, T. Mathe, Platinum Met. Rev.
1998, 42, 108.
[33] T. Osawa, Y. Hayashi, A. Ozawa, T. Harada, O. Takayasu J. Mol. Catal. A:
Chem. 2001, 169, 289.
[34] T. Sugimura, Catal. Surv. Jpn. 1999, 3, 37.
[35] T. Osawa, T. Harada, O. Takayasu, Top. Catal. 2000, 13, 155.
[36] T. Osawa, T. Harada, A. Tai, Catal Today 1997, 37, 465.
[37] S. Nakagawa, T. Sugimura, A. Tai, Chem. Lett. 1997, 859.
[38] T. Sugimura, S. Nakagawa, A. Tai, Bull. Chem. Soc. Jpn. 2002, 75, 355.
[39] X. Zuo, H. Liu, D. Guo, X. Yang, Tetrahedron, 1999, 55, 7787.
[40] M. Sutyinszki, K. Szöri, K. Felföldi, M. Bartók, Catal. Commun. 2002, 3,
125.
[41] Y. Nitta, Top. Catal. 2000, 13, 179.
[42] W.-R. Huck, T. Mallat, A. Baiker, J. Catal. 2000, 193, 1.
[43] K. Borszeky, T. Bürgi, Z. Zhaohui, T. Mallat, A. Baiker, J. Catal. 1999,
187, 160.
[44] A. Tungler, T. Tarnai, L. Hegedüs, K. Fodor, T. Màthé, Plat. Met. Rev.
1998, 42, 108.
[45] W.-R. Huck, T. Mallat, A. Baiker, Catal. Lett. 2002, 80, 87.
[46] A. Baiker, H.-U. Blaser, in: Handbook of Heterogeneous Catalysis, eds. G.
Ertl, H. Knoezinger, J. Weitkamp, vol. 5, VCH, Weinheim, 1997, p. 2422.
[47] M. E. Davis, Top. Catal. 2003, 25, 3.
[48] T. Osawa, T. Harada, O. Takayasu, Top. Catal. 2000, 13, 155.
[49] T. Bein, Curr. Opin. in Solid State Mater. Sci. 1999, 4, 85.
[50] P. B. Wells, K. E. Simons, J. A. Slipszenko, S. P. Griffiths, D. F. Ewing, J.
Mol. Catal. A: Chem 1999, 146, 159.
22 Chapter 1
[51] C. J. Baddeley, Top. Catal. 2003, 25, 17.
[52] Y. Orito, S. Imai, S. Niwa, N. G. Hung, J. Synth. Org. Chem. 1979, 37,
173.
[53] Y. Orito, S. Imai, S. Niwa, Nippon Kagaku Kaishi 1979, 1118.
[54] Y. Orito, S. Imai, S. Niwa, J. Chem. Soc. Japan 1980, 670.
[55] Y. Orito, S. Imai, S. Niwa, Nippon Kagaku Kaishi 1982, 137.
[56] A. Baiker, J. Mol. Catal. A: Chem. 1997, 115, 473.
[57] H.-U. Blaser, H.-P. Jalett, M. Müller, M. Studer, Catal. Today 1997, 37,
441.
[58] W.-R. Huck, T. Mallat, A. Baiker, Adv. Synth. Catal. 2003, 345, 255.
[59] H.-U. Blaser, H. P. Jalett, J. Wiehl, J. Mol. Catal. A: Chem. 1991, 68, 215.
[60] J. T. Wehrli, A. Baiker, D. M. Monti, H.-U. Blaser, H. P. Jalett, J. Mol.
Catal. A: Chem. 1989, 57, 245.
[61] M. Studer, H.-U. Blaser, C. Exner, Adv. Synth. Catal. 2003, 345, 45.
[62] H.-U. Blaser, H. P. Jalett, D.M. Monti, A. Baiker, J. T. Wehrli, Stud. Surf.
Sci. Catal. 1991, 67, 147.
[63]P. B. Wells, K. E. Simons, J. A. Slipszenko, S. P. Griffiths, D. F. Ewing, J.
Mol. Catal A: Chem. 1999, 164, 159.
[64] G. Farkas, K. Fodor, A. Angler, T. Máthé, G. Tóth, R. A. Sheldon, J. Mol.
Catal. A: Chem. 1999, 138, 122.
[65] S. P. Griffiths, P. Johnston, W. A. H. Vermeer, P. B. Wells, Chem. Com-
mun. 1994 2431.
[66] M. Bartók, K. Felföldi, G. Szöllösi, T. Bartók J. Catal. 1998, 173, 187.
[67] B. Minder, M. Schürch, T. Mallat, A. Baiker, Catal. Lett. 1995, 31, 143.
[68] B. Minder, M. Schürch, T. Mallat, A. Baiker, T. Heinz, A. Pfalz, J. Catal.
1996, 160, 261.
[69] A. Pfalz, T. Heinz, Top. Catal. 1997, 4, 229.
[70] M. Schürch, T. Heinz, R. Aeschimann, T. Mallat, A. Pfalz, A. Baiker, J.
Catal. 1998, 173, 187.
[71] G. Szöllösi, C. Somlai, P. T. Szabo, M. Bartók, J. Mol. Catal. A: Chem
2001, 170, 165.
[72] T. Heinz, G. Wang, A. Pfaltz, B. Minder, M. Schürch, T. Mallat, A.
Baiker, Chem. Commun. 1995, 1421.
Introduction 23
[73] K. E. Simons, G. Z. Wang, T. Heinz, T. Giger, T. Mallat, A. Pfaltz, A.
Baiker, Tetrahedron: Asymmetry 1995, 6, 505.
[74] B. Minder, T. Mallat, A. Baiker, G. Wang, T. Heinz, A. Pfaltz, J. Catal.
1995, 154, 371.
[75] G. D. H. Dijkstra, R. M. Kellogg, H. Wynberg, J. Org. Chem. 1990, 55,
6121.
[76] G. D. H. Dijkstra, R. M. Kellogg, H. Wynberg, J. S. Svendsen, I. Marko,
K. B. Sharpless, J. Am. Chem. Soc. 1989, 111, 8069.
[77] T. Bürgi, A. Baiker, J. Am. Chem. Soc. 1998, 120, 161.
[78] M. Bartók, K. Felföldi, G. Szöllösi, T. Bartók, Catal. Lett. 1999, 612, 1.
[79] B. Minder, T. Mallat, P. Skrabal, A. Baiker, Catal. Lett. 1994, 29, 115.
[80] M. Von Arx, T. Mallat, A. Baiker, Catal. Lett. 2002, 78, 267.
[81] T. Mallat, M. Bodmer, A. Baiker, Catal. Lett. 1997, 44, 95.
[82] B. Török, K. Balázsik, K. Felföldi, M. Bartók, Stud. Surf. Sci. Catal. 2000,
130, 3381.
[83] J. L. Margitfalvi, E. Tálas, M. Hegedüs, Chem. Commun. 1999, 645.
[84] B. Török, K. Balàzsik, M. Török, G. Szöllösi, M. Bartók, Ultrason.
Sonochem. 2000, 7, 151,
[85] C. LeBlond, J. Wang, A. T. Andrews, Y.-K. Sun, Topics Catal. 2000, 13,
169.
[86] K. Balàzsik, K. Szöri, K. Felföldi, B. Török, M. Bartók, Chem. Commun.
2000, 555.
[87] M. Studer, S. Burkhard, A. F. Indolese, H.-U. Blaser, Chem. Commun.
2000, 1327.
[88] G.-Z. Wang, T. Mallat, A. Baiker, Tetrahedron: Asymmetry 1997, 8, 2133.
[89] N. Künzle, A. Szabo, M. Schürch, G. Z. Wang, T. Mallat, A. Baiker,
Chem. Commun. 1998, 1377.
[90] M. Bodmer, T. Mallat, A. Baiker, Chem. Ind. (Dekker) 1998, 75, 75.
[91] M. von Arx, T. Bürgi, T. Mallat, A. Baiker, Chem. Eur. J. 2002, 8, 1430.
[92] M. von Arx, T. Mallat, A. Baiker, J. Catal. 2001, 202, 169.
[93] M. von Arx, T. Mallat, A. Baiker, J. Catal. 2000, 193, 161.
[94] M. von Arx, T. Mallat, A. Baiker, Tetrahedron: Asymmetry 2001, 12, 3089.
[95] K. Balázsik, B. Török, K. Felföldi, G. Szakonyi, M. Bartók, Ultrasonics
Sonochemistry 1997, 4, 301.
24 Chapter 1
[96] M. von Arx, T. Mallat, A. Baiker, Spec. Pub. - Royal Chemical Society 2001,
266, 247.
[97] M. Studer, H.-U. Blaser, S. Burkhard, Adv. Synth. Catal. 2002, 344, 511.
[98] B. Török, K. Felföldi, K. Balàzsik, M. Bartók, Chem. Commun. 1999,
1725.
[99] E. Toukoniitty, P. Maki-Arvela, M. Kuzma, A. Villela, A. K. Neyestanaki,
T. Salmi, R. Sjoholm, R. Leino, E. Laine, D. Y. Murzin, Catal. Today 2001, 66,
411.
[100] M. Studer, S. Burkhard, H.-U. Blaser, Chem. Commun. 1999, 1727.
[101] M. Studer, V. Okafor, H.-U. Blaser, Chem. Commun. 1998, 1053.
[102] H.-U. Blaser, M. Müller, Stud. Surf. Sci. Catal. 1991, 59, 73.
[103] H.-U. Blaser, S. K. Boyer, U. Pittelkow, Tetrahedron: Asymmetry 1991, 2,
721.
[104] Y. Nitta, Y. Ueda, T. Imanaka, Chem. Lett. 1994, 1095.
[105] Y. Nitta, K. Kobiro, Chem. Lett. 1995, 165.
[106] Y. Nitta, K. Kobiro, Chem. Lett. 1996, 897.
[107] G. Webb, P. B. Wells, Catal. Today 1992, 12, 319.
[108] T. Mallat, S. Frauchiger, P. J. Kooyman, M. Schürch, A. Baiker, Catal.
Lett. 1999, 63, 121.
[109] B. Török, K. Balàzsik, M. Török, K. Felföldi, M. Bartók, Catal. Lett.
1999, 63, 121.
[110] B. Török, K. Balàzsik, G. Szöllösi, K. Felföldi, M. Bartók, Chirality
1999, 11, 470.
Chapter2
2Inversion of enantioselectivity in the hydro-genation of ketopantolactone on platinum
modified by ether derivatives of cinchonidine
2.1 Summary
Asymmetric hydrogenation of ketopantolactone was studied on a 5 wt%
Pt/Al2O3 catalyst in the presence of cinchonidine and its O-methyl, O-ethyl,
O-phenyl and O-trimethylsilyl derivatives. Inversion of enantioselectivity with
the latter two bulky substituents proved that in the enantio-differentiating step
cinchonidine adsorbs via the quinoline ring lying approximately parallel to the
Pt surface. The striking nonlinear behavior observed with cinchonidine – O-
phenyl-cinchonidine mixtures is attributed to differences in the adsorption
strength and geometry of the modifiers.
2.2 Introduction
The enantioselective hydrogenation of ketopantolactone (1) has been exten-
sively investigated as one enantiomer of the product, (R)-(-)-pantolactone (2,
Figure 2-1), is an intermediate in the synthesis of pantothenic acid (vitamin B
family) and a constituent of coenzyme A [1,2]. Various rhodium(I) complexes
have been tested in this reaction [3-7] that afforded up to 98.7% ee [8]. A het-
erogeneous catalyst, supported Pt chirally modified by cinchonidine (CD) [9-
11] was less selective: the best ee was only 91.6% to (R)-2 [12]. Practical advan-
tages of the latter process are the extremely low CD/1 molar ratio, only 4 ppm,
28 Chapter 2
and the high production rate achieved in a continuous flow fixed-bed reactor
(TOF = 1420 h-1) [13].
Supported by theoretical calculations, we proposed a model for the reac-
tant – modifier interaction on the Pt surface involving a H-bond between CD
and the half-hydrogenated state derived from 1 (Figure 2-2) [10]. However,
there is no direct evidence yet for the assumption implied to the model that 1
and the quinoline ring of CD are adsorbed parallel to the Pt surface during the
enantio-differentiating step. Various surface science techniques [14-22] and H/
D exchange [23,24] clarified some important details of the adsorption of CD
Fig. 2-1: Hydrogenation of ketopantolactone (1) to pantolactone (2) over chirally modi-fied Pt/Al2O3 and the structure of ether derivatives of cinchonidine used as modifiers.
Fig. 2-2: Top view of a calculated model for the CD - 1 interaction over the Pt surface,leading upon hydrogenation to (R)-2 [10]. The π-bonded quinoline ring of CD and the twocarbonyl groups of 1 lie approximately parallel to Pt.
ethylmethyl
PHOCDTMSOCDEtOCDMeOCDCD
modifier
phenyltrimethylsilyl
O
O
O
R
H
O
OHH
O*5 wt % Pt /Al2O3
H2, modif ier
1 2N
HORN
H
H
40 bar, toluene, rt
N
NH
OH
O
O
O
H
Inversion of enantioselectivity 29
on Pt but none of these methods allowed a real in situ investigation, i.e. in the
presence of the reactant. ATR-IR studies in H2-saturated CH2Cl2 revealed
three differently adsorbed species of CD: a π-bonded species, which adsorbed
via the aromatic ring nearly parallel to the Pt surface and two species, in which
the aromatic ring is tilted relative to the Pt surface [20,21] The former adsorp-
tion geometry would correspond to the assumption depicted in Figure 2-2.
In this chapter we present another approach to clarify the adsorption mode
of CD and its interaction with 1 in the enantio-differentiating complex: we
synthesized some ether derivatives of CD (Figure 2-1) and studied the influ-
ence of bulkiness of the functional group on the hydrogenation of 1. If CD
adsorbs via the quinoline ring lying approximately parallel to the Pt surface, the
increasing steric hindrance in the ether derivatives should successively diminish
the enantioselectivity, providing a truly in situ evidence for our model. This
approach led us also to some other interesting observations, which will be
reported in this chapter.
2.3 Experimental
2.3.1 Catalytic hydrogenation
Tetrahydrofuran (THF, 99.5%, J. T. Baker) was dried over Na before use. Tol-
uene (99.5%, J. T. Baker) and CD (92%, Fluka; impurities: 1% quinine, 7%
quinidine, determined by HPLC at Fluka) were used as received.
A 5 wt% Pt/Al2O3 catalyst (Engelhard 4759) was prereduced in flowing
H2 for 60 min at 400°C, cooled to room temperature in H2 in 30 min, and
flushed with nitrogen. The pretreated catalyst was used on the same day.
Hydrogenations were carried out at room temperature (ca. 20°C) in a stainless
steel autoclave equipped with a 50 ml glass liner and a PTFE cover, and ma-
gnetic stirring (1000 rpm). Total pressure (40 bar) and hydrogen uptake were
controlled by computerized constant-volume constant-pressure equipment
(Büchi BPC 9901). In a standard procedure, 42 mg catalyst in 5 ml solvent was
exposed to flowing H2 for 2 min. Then 6.8 µmol modifier or modifier mixture
30 Chapter 2
was added in 1 ml solvent. After a short preadsorption time of 1 min 236 mg
(1.84 mmol) ketopantolactone (1) was added and the reaction was started.
Conversion and enantioselectivity were determined by gas chromato-graphy
using a Chirasil-DEX CB column (Chrompack). No other product beside the
two enantiomers of pantolactone (2) could be detected.
2.3.2 Synthesis of O-substituted cinchonidines
Melting points were determined using Büchi-545 automatic melting point
apparatus. 1H-NMR spectra were recorded by Varian Unity Inova at 400 MHz
in CDCl3. All reactions were carried out under Ar. Starting materials were pur-
chased from Aldrich and used without further purification. DMSO, DMF and
THF were dried over molecular sieve before use.
O-Methyl-cinchonidine (MeOCD)
CD (10.0 g, 34 mmol) was added to a stirred suspension of NaH (2.1 g,
88 mmol) in dry DMF (50 ml) and the mixture was stirred for 30 min at rt,
then 1 h at 50°C. The solution was cooled below 5°C and iodomethane (2.2
ml, 15 mmol) was added. After 1 h stirring at 0 - 5°C the reaction mixture was
diluted with water (150 ml) and extracted with EtOAc (2 x 100 ml). The com-
bined organic layer was extracted with 2 M HCl, the aqueous solution was
washed with hexane (50 ml) and the pH of the solution was adjusted to alka-
line with solid NaHCO3. After extraction with EtOAc (2 x 100 ml), the
organic layer was dried over Na2SO4 and the solvent was evaporated to dryness
(6.9 g white crystals, a mixture of MeOCD and starting CD). Pure product
was obtained by chromatography on silica with hexane-acetone-TEA 40:18:1
followed by crystallization from hexane. Yield: 5.6 g (46%) white crystals. Mp
126.6 - 126.1°C; NMR (CDCl3) δ = 8.90 (d, 1H), 8.15 (d, 1H), 8.10 (d, 1H)
7.70 (t, 1H), 7.57 (t, 1H), 7.42 (d, 1H), 5.72 (m, 1H), 5.04 (d, 1H), 4.91-
4.88 (m, 2H), 3.40-3.30 (m, 1H), 3.28 (s, 3H), 3.10-3.03 (m, 2H), 2.75-2.55
(m, 2H), 2.13-2.10 (m, 1H), 1.80-1.70 (m, 3H), 1.60-1.50 (m, 2H); MS 309
(M+H+).
Inversion of enantioselectivity 31
O-Ethyl-cinchonidine (EtOCD)
EtOCD was synthesized according to the recipe for MeOCD, but iodoethane
was used instead of iodomethane. Yield: 55 %, colorless oil. 1H-NMR (400
MHz, CDCl3): δ = 8.90 (d, 1H), 8.12 and 8.10 (two overlapping d, 2H) 7.72
(t, 1H), 7.60 (t, 1H), 7.52 (d, 1H), 5.75 (m, 1H), 5.17 (d, 1H), 4.97-4.86 (m,
2H), 3.43 (qa, 2H) and 3.41 (s, 1H), 3.15-3.05 (m, 2H), 2.75-2.57 (m, 2H),
2.30-2.22 (m, 1H), 1.85-1.75 (m, 3H), 1.65-1.55 (m, 2H), 1.25 (t, 3H);
MS: 323 (M+H+).
O-Trimethylsilyl-cinchonidine (TMSOCD)
TMSOCD was prepared according to a recent procedure [38]. CD (2.0 g, 6.7
mmol) and triethylamine (TEA, 0.81 g, 8.0 mmol) were dissolved in THF.
The solution was cooled to 0 - 5°C and chlorotrimethylsilane (0.87 g,
8.0 mmol) dissolved in THF was added dropwise. The reaction mixture was
allowed to warm to room temperature and stirred at this temperature over-
night, then at 60°C for 2 h. The mixture was poured to ice-water (~50 ml) and
extracted with dichloromethane (2 x 50 ml), the combined organic layers were
washed with water and brine, dried over Na2SO4 and evaporated to dryness.
Column chromatography on silica with hexane-acetone-TEA 40:18:1 afforded
a white crystalline material; yield: 1.2 g, 49%. Mp 75.8 - 76.7°C; 1H-NMR
(CDCl3) δ = 8.84 (d, 1H), 8.14 and 8.10 (two overlapping d, 2H) 7.71 (t,
1H), 7.56 (t, 1H), 7.48 (d, 1H), 5.71 (m, 1H), 5.60 (d, 1H), 4.92-4.87 (m,
2H), 3.40-3.30 (m, 1H), 3.10-3.00 (m, 2H), 2.75-2.55 (m, 2H), 2.10-2.10
(m, 1H), 1.80-1.70 (m, 3H), 1.60-1.50 (m, 2H), 0.1 (s, 9H); MS 367
(M+H+).
O-Phenyl-cinchonidine (PhOCD)
This compound was prepared according to the literature procedure for the syn-
thesis of dihydroquinidyl aryl ethers [39]. CD (2.2 g, 7.5 mmol) was dissolved
in anhydrous DMSO (30 ml), NaH (0.40 g, 10 mmol) was added at room
temperature and the mixture was stirred for 1 h. Then abs. pyridine (1.2 ml,
15 mmol) and CuI (1.45 g, 7.5 mmol) were added and stirred for 30 min.
After addition of iodobenzene (0.85 ml, 7.5 mmol) the mixture was kept at
100°C for 72 hours. After cooling to room temperature, water (25 ml), dichlo-
32 Chapter 2
romethane (50 ml), ethylenediaminetetraacetic acid (0.5 g), and finally cc.
ammonia solution (5 ml) were added. The mixture was stirred at room temper-
ature for 1 h, the organic layer was separated, and the aqueous phase was
extracted with CH2Cl2 (2 x 15 ml). The combined organic layer was washed
with 5% ammonia solution (5 x 25 ml) until the aqueous phase remained col-
orless, then water (25 ml) and solvent were evaporated in vacuo and the residue
was dissolved in EtOAc (50 ml). It was extracted with 2M HCl solution
(50 ml), the acidic solution was washed with EtOAc (2 x 25 ml). Then the pH
was set to alkaline with solid NaHCO3 and extracted with EtOAc (2 x 30 ml).
The combined organic layer was washed with brine, dried over Na2SO4 and
evaporated to dryness. Crude product was purified over silica using hexane-ace-
tone-TEA 40:18:1 as eluent. After evaporation of the solvent the product was
crystallized from hexane (0.6 g, 1.6 mmol, white crystals). Yield: 22%. Mp
126.3 - 126.4°C; 1H-NMR (400 MHz, CDCl3): δ = 8.83 (d, 1H), 8.20 (d,
1H), 8.19 (d, 1H) 7.78 (t, 1H), 7.57 (m, 1H), 7.62 (m, 1H), 7.10 (m, 2H),
6.88 (m, 1H), 6.78 (m, 2H), 6.08 (d, 1H), 5.73 (m, 1H), 4.98-4.82 (m, 2H),
3.40-3.30 (m, 1H), 3.28 (s, 2H), 2.75-2.55 (m, 2H), 2.13-2.10 (m, 1H),
1.80-1.70 (m, 3H), 1.60-1.50 (m, 2H); MS 371 (M+H+).
2.4 Results and Discussion
At first we investigated the hydrogenation of 1 in toluene and THF over a
5 wt% Pt/Al2O3 catalyst (Figure 2-3). In both solvents CD afforded somewhat
higher ees to (R)-2 than MeOCD or EtOCD, though the efficiency of modifi-
ers depended also on the pressure in the range 1 - 40 bar. The latter two modi-
fiers possess small O-alkyl groups that do not substantially change the
adsorption geometry, compared to CD. The reasonably good ees achieved in
these experiments support the assumption that the OH function of CD is not
involved in the enantio-differentiating step (Figure 2-2). When using
TMSOCD or PhOCD as modifiers, the opposite enantiomer formed in excess.
The most striking effect was observed in toluene where CD afforded 79% ee to
(R)-2, whereas modification with PhOCD resulted in 52% ee to (S)-2.
Inversion of enantioselectivity 33
Decreasing the temperature to 0°C (under otherwise standard conditions)
improved the ee to 54.4% in the latter reaction.
Certainly, a complex between modifier and 1, as depicted in Figure 2-2,
cannot establish in the presence of the bulky trimethylsilyl and phenyl substitu-
ents in TMSOCD and PhOCD, respectively. In the original model calculated
for the CD-1 interaction, CD is bounded via hydrogen bonding to the halfhy-
drogenated state derived from 1 [10]. Besides, a steric repulsion excerted by the
quinoline ring ensures a fixed adsorption of 1 with dominantly one enantioface
on the Pt surface. Introduction of the bulky trimethylsilyl or phenyl substitu-
ents changes dramatically the chiral pocket available for the adsorption of 1
over the Pt surface and leads to the favored adsorption of 1 on the opposite
enantioface. A plausible explanation may be that TMSOCD and PhOCD do
not adsorb via the quinoline ring being approximately parallel to the Pt surface
(π-bonding) but rather in a tilted position (N-lone pair bonding). This change
in the adsorption geometry should result in a considerably weaker adsorption
Fig. 2-3: Enantioselectivities in the hydrogenation of 1 over a 5 wt% Pt/Al2O3 catalystwith two different solvents ( toluene: filled circles; THF: empty circles) under standard reac-tion conditions. The structure of the modifiers is shown in Figure 2-1.
80 (R)
60 (R)
40 (R)
20 (R)
0
20 (S)
40 (S)
60 (S)
0ee [%]
CD MeOCD EtOCD TMSOCD PhOCD
34 Chapter 2
of these modifiers compared to the adsorption of CD. Another possibility is
that TMSOCD and PhOCD still adsorb with the quinoline ring approxi-
mately parallel to the Pt surface but the bulky phenyl and trimethylsilyl groups
“lying” on the surface change the adsorption mode of 1.
To estimate the relative adsorption strength of the modifiers on Pt, the
hydrogenation of 1 was carried out with mixtures of CD and PhOCD. A
strong nonlinear behavior was observed as illustrated in Figure 2-4. The
“expected” or theoretical ee (dashed line) was calculated assuming that the
molar ratios of the modifiers in solution and on the Pt surface are identical, and
the reaction rates and ees are linear combinations of those measured with CD
and PhOCD alone. The average hydrogenation rates with the modifiers alone
Fig. 2-4: Hydrogenation of 1 over 5 wt% Pt/Al2O3 modified by CD - PhOCD mixtures;standard reaction conditions in toluene.
ee [%]
80 (R)
60 (R)
40 (R)
20 (R)
0
20 (S)
40 (S)
60 (S)
0
1.0 0.8 0.6 0.4 0.2 0
XPhOCD
0 0.2 0.4 0.6 0.8 1.0
XCD
eecalc.
Inversion of enantioselectivity 35
were similar: 183 and 140 mmol/h for CD and PhOCD, respectively. Still, CD
controlled the enantioselection in the whole range studied and even a modifier
mixture containing only 0.7 mol% CD afforded 33% ee to (R)-2. This nonlin-
ear behavior is even more striking when considering that the purity of commer-
cial CD was only 92% and contained 7% quinidine that alkaloid alone affords
(S)-2 in excess.
The well-known nonlinear behavior has originally been described for
homogeneous catalytic reactions carried out with enantiomerically impure
ligands [25-27]. This interesting phenomenon has recently been extended to
mixtures of two diastereomers and even to two chemically different chiral
ligands in homogeneous catalysis [28-32], and to similar but chemically differ-
ent modifiers in heterogeneous catalysis [33-38]. The basic requirement for the
extension is that the ligand or modifier pairs afford products of opposite con-
figuration.
As mentioned in the introduction, ATR-IR studies of CD adsorption on
Pt [20,21] allowed to select conditions nearest to those of catalytic hydrogena-
tion. Based on this study we propose that the nonlinear behavior shown in
Figure 2-4 is due to the different geometry and strength of the alkaloid modi-
fier adsorption on Pt. CD adsorbs stronger than PhOCD. The quinoline ring
of CD lying approximately parallel to the surface interacts strongly with Pt viaπ-bonding (Figure 2-5). In contrast, PhOCD adopts a tilted position; it is
anchored to Pt only weakly via the aromatic N atom and cannot efficiently
compete with CD for the active Pt sites. It is very likely that some steric effects
and electronic interactions also play a role though this contribution cannot be
reliably estimated yet. This type of interactions has been thoroughly investi-
gated in connection with the nonlinear behavior (ligand association) in homo-
geneous catalysis [25-27].
2.5 Conclusions
Hydrogenation of 1 over Pt modified by ether derivatives of CD provides
strong support to our structural model for the enantio-differentiating diaste-
36 Chapter 2
reomeric complex proposed earlier (Figure 2-2) [10]. Namely, the OH group of
CD is not involved in the modifier-1 interaction and in the modifier-1 com-plex CD adsorbs approximately parallel to the metal surface via the quino-line ring (π-bonding). The hydrogen bonding between the quinuclidine N and
the half-hydrogenated state of 1, together with the steric hindrance by the
quinoline ring, results in (R)-2 as the dominant product. Replacing the OH
group of CD by a bulky PhO or TMSO group prevents this adsorption mode
of 1 and affords (S)-2 as the major enantiomer. If the quinoline ring of CD
would adopt a tilted position relative to the Pt surface during interaction with
1 (compare to Figure 2-2 and Figure 2-5), replacement of the OH group by the
bulky PhO or TMSO groups should not significantly influence the adsorption
of 1 and thus the stereochemical outcome of the hydrogenation reaction.
A practically and scientifically important consequence of the observed
strong nonlinear behavior is that only carefully purified cinchona derivatives
provide genuine results concerning enantioselectivity. Trace amounts of CD
can outperform the enantiodifferentiation of its derivatives, as demonstrated
for PhOCD. This behavior is presumably more general and can be extended to
other types of chiral modifiers in heterogeneous catalysis. Furthermore due care
has to be taken when interpreting the behaviour of modifier mixtures screened
by high-throughput methods.
Fig. 2-5: Schematic illustration of the adsorption of CD and PhOCD on an idealized flatPt surface, illustrating the considerable steric hindrance by the phenyl group. The structureof PhOCD corresponds to the energetic minimum, as derived from preliminary calculationsusing Gaussian 98.
N N
H OH
O
N
H
N
CD PhOCD
Inversion of enantioselectivity 37
2.6 References
[1] R. Schmid, Chimia 1996, 50, 110.
[2] H. Brunner, T. Tracht, Tetrahedron: Asymmetry 1998, 9. 3773.
[3] A. Roucoux, F. Agbossou, A. Mortreux, F. Petit, Tetrahedron: Asymmetry
1993, 4, 2279.
[4] T. Morimoto, H. Takahashi, K. Achiwa, Chem. Pharm. Bull. 1994, 42,
481.
[5] F. Hapiot, F. Agbossou, C. Meliet, A. Mortreux, G. M. Rosair, A. J. Welch,
New J. Chem. 1997, 21, 1161.
[6] H. Brunner, A. Apfelbacher, M. Zabel, Eur. J. Inorg. Chem. 2001, 917.
[7] T. Sturm, W. Weissensteiner, F. Spindler, K. Mereiter, A. M. Lopez-Agenjo,
B. R. Manzano, F. A. Jalon, Organometallics 2002, 21, 1766.
[8] A. Roucoux, M. Devocelle, J.-F. Carpentier, F. Agbossou, A. Mortreux, Syn-
lett 1995, 358.
[9] O. Kanji, in: Agency of Ind. Sci. Technol., Japan, 1987.
[10] M. Schürch, O. Schwalm, T. Mallat, J. Weber, A. Baiker, J. Catal. 1997,
169, 275.
[11] T. Mallat, S. Frauchiger, P. J. Kooyman, M. Schürch, A. Baiker, Catal.
Lett. 1999, 63, 121.
[12] M. Schürch, N. Künzle, T. Mallat, A. Baiker, J. Catal. 1998, 176, 569.
[13] N. Künzle, R. Hess, T. Mallat, A. Baiker, J. Catal. 1999, 186, 239.
[14] T. Evans, A. P. Woodhead, A. Gutierrez-Sosa, G. Thornton, T. J. Hall, A.
A. Davis, N. A. Young, P. B. Wells, R. J. Oldman, O. Plashkevych, O. Vahtras,
H. Agren, V. Carraretta, Surf. Sci. 1999, 436, L691.
[15] T. E. Jones, C. J. Baddeley, Surf. Sci. 2002, 519, 237.
[16] T. Bürgi, F. Atamny, R. Schlögl, A. Baiker, J. Phys. Chem. B 2000, 104,
5953.
[17] A. F. Carley, M. K. Rajumon, M. W. Roberts, P. B. Wells, J. Chem. Soc.
Faraday Trans. 1995, 91, 2167.
[18] W. Chu, R. J. LeBlanc, C. T. Williams, Catal. Commun. 2002, 3, 547.
[19] I. Bakos, S. Szabó, M. Bartók, E. Kàlmàn, J. Electroanal. Chem. 2002,
532, 113.
[20] D. Ferri, T. Bürgi, A. Baiker, Chem. Commun. 2001, 1172.
38 Chapter 2
[21] D. Ferri, T. Bürgi, J. Am. Chem. Soc. 2001, 123, 12074.
[22] K. Kubota, F. Zaera, J. Am. Chem. Soc. 2001, 123, 11115.
[23] A. Solladié-Cavallo, F. Hoernel, M. Schmitt, F. Garin, Tetrahedron Lett.
2002, 43, 2671.
[24] G. Bond, P. B. Wells, J. Catal. 1994, 150, 329.
[25] D. Guillaneux, S. H. Zhao, O. Samuel, D. Rainford, H. B. Kagan, J. Am.
Chem. Soc. 1994, 116, 9430.
[26] C. Girard, H. B. Kagan, Angew. Chem., Int. Ed. 1998, 37, 2923.
[27] H. B. Kagan, Adv. Synth. Catal. 2001, 343, 227.
[28] K. B. Sharpless, W. Amberg, Y. L. Bennani, G. A. Crispino, J. Hartung,
K. S. Jeong, H. L. Kwong, K. Morikawa, Z. M. Wang, D. Q. Xu, X. Zhang, J.
Org. Chem. 1992, 57, 2768.
[29] S. Y. Zhang, C. Girard, H. B. Kagan, Tetrahedron: Asymmetry 1995, 6,
2637.
[30] M. Kitamura, S. Suga, M. Niwa, R. Noyori, J. Am. Chem. Soc. 1995, 117,
4832.
[31] K. Muniz, C. Bolm, Chem.-Eur. J. 2000, 6, 2309.
[32] H. B. Kagan, Synlett 2001, 888.
[33] K. E. Simons, P. A. Meheux, A. Ibbotson, P. B. Wells, Stud. Surf. Sci.
Catal. 1993, 75, 2317.
[34] A. Tungler, K. Fodor, T. Máthé, R. A. Sheldon, Stud. Surf. Sci. Catal.
1997, 108, 157.
[35] M. Schürch, T. Heinz, R. Aeschimann, T. Mallat, A. Pfaltz, A. Baiker, J.
Catal. 1998, 173, 187.
[36] Y. Nitta, A. Shibata, Chem. Lett. 1998, 161.
[37] W.-R. Huck, T. Mallat, A. Baiker, Adv. Synth. Catal. 2003, 345, 255.
[38] W.-R. Huck, T. Mallat, A. Baiker, Catal. Lett. 2003, 87, 241.
[39] A. Lindholm, P. Mäki-Arvela, E. Toukoniitty, T. A. Pakkanen, J. T. Hirvi,
T. Salmi, T. Yu. Murzin, R. Sjöholm, R. Leino, J. Chem. Soc. Perkin 1 2002,
23, 2605.
[40] W. Amberg, Y. L. Bennani, R. K. Chadha, G. A. Crispino, W. D. Davis, J.
Hartung, K.-S. Jeong, Y. Ogino, T. Shibata, K. B. Sharpless, J. Org. Chem.
1993, 58, 844.
Chapter3
3Fine tuning the “chiral sites” on solidenantioselective catalysts
3.1 Summary
A fundamental point in the mechanism of enantioselective hydrogenation over
chirally modified metals is the nature of “chiral sites” developed by adsorption
of the modifier on the metal surface. In spite of considerable effort towards
unravelling the adsorption mode of the modifier by surface science techniques,
most of these spectroscopic measurements were done at conditions relatively
far from those met under real reaction conditions. Here we applied a truly in
situ “synthetic” approach, the systematic variation of the structure of the chiral
modifier used for enantioselective hydrogenation over 5 wt% Pt/Al2O3. We
have synthesized various O-alkyl, O-aryl and O-silyl derivatives of cinchonidine
(CD) and tested them in the enantioselective hydrogenation of ethyl pyruvate,
ketopantolactone, 4,4,4-trifluoroacetoacetate and 1,1,1-trifluoro-2,4-diketo-
pentane. With increasing bulkiness of the ether group, the ee gradually
decreased or even the opposite enantiomer formed in excess (up to 53% ee).
We propose that the increasing bulkiness of the ether group prevents the
strong, π-bonded adsorption of the quinoline ring of CD close to parallel to
the Pt surface. In this tilted position the modifier adsorbs weaker via the quino-
line N and also the position of the interacting function, the quinuclidine N, is
shifted. This shift results in a different shape and size of the “chiral pocket”
available for adsorption of the activated ketone substrate. The weaker adsorp-
tion of the bulky ether derivatives was proved by UV-Vis spectroscopy and by
the non-linear behavior of modifier mixtures (NLE). The tilted adsorption
42 Chapter 3
mode was corroborated by the lower hydrogenation rate of the quinoline ring
of the ether derivatives, relative to that of CD.
3.2 Introduction
As already mentioned in Chapter 1.4.3 the most successful and commonly
applied strategy in heterogeneous enantioselective catalysis is the modification
of the active metal catalyst by a strongly adsorbing chiral organic compound
termed as modifier [1-6]. Supported Pt and Pd modified by cinchona alkaloids
provide the highest enantioselectivity in the hydrogenation of various activated
ketones [7-13], and functionalized olefins and heteroaromatic compounds
[14-19].
Although the number of applications is increasing, the fundamental under-
standing of the nature of enantiodifferentiation is still at an early stage of devel-
opment, compared to the state of art in homogeneous enantioselective catalysis
[20-22]. Presumably, the most important question to be answered is the nature
of “chiral sites” developed by adsorption of the chiral modifier on the metal
surface [23,24]. An example to the importance of adsorption geometry is
Augustine’s early model proposed for the hydrogenation of ethyl pyruvate on
cinchona-modified Pt [25,26]. According to the authors’ assumption, a change
of the adsorption mode of the modifier should result in the inversion of enatio-
selectivity. Despite of the impressive development in this field in the past years,
the adsorption mode of substrate and modifier in the enantio-differentiating
step is still the most speculative part of the mechanistic models.
Various surface science techniques, including near-edge absorption fine
structure spectroscopy (NEXAFS) [27], scanning tunneling microscopy (STM)
[28], X-ray photoelectron spectroscopy (XPS) [29,30], low-energy electron dif-
fraction (LEED) [30], electrochemical polarization [31], reflection-absorption
infrared spectroscopy (RAIRS) [32,33], and surface-enhanced Raman spectros-
copy (SERS) [34,35] revealed some important details of the adsorption behav-
ior of CD on Pt but the conditions were mostly far from those of catalytic
hydrogenation. A recent ATR-IR spectroscopic study on Pt/Al2O3 in the pres-
Fine tuning the “chiral sites” 43
ence of an organic solvent and hydrogen revealed three different adsorption
modes of cinchonidine (CD) [36,37]. These species differ from each other by
their adsorption strength and adsorption geometries: a π-bonded species,
which adsorbs via the aromatic ring nearly parallel to the metal surface and two
species, in which the aromatic ring is tilted relative to the Pt surface. Still, the
method could not answer the key question, namely, which of these species
interacts with the substrate in the enantio-differentiating step as the substrate
could generally not be involved in the investigation.
The aim of the present chapter is to explore how the adsorption mode of
the modifier interacting on the metal surface with the coadsorbed substrate
affects the enantiodifferentiation. The hydrogenation of some activated ketones
on cinchona-modified Pt has been chosen as model reactions. Several mecha-
nistic models have been developed in the past years for the hydrogenation of
Fig. 3-1: Schematic representation of the mechanistic models published by our group (a)[41], Margitfalvi et al. (b) [42], Sun et al. (c) [43], and McBreen et al. (d) [44] to explain theenantioselection in the hydrogenation of ethyl pyruvate on cinchona-modified Pt. The draw-ings represent top-side views over the Pt surface.
44 Chapter 3
ethyl pyruvate, the most investigated reaction in heterogeneous enantioselec-
tive catalysis. Beside the template model of Wells that has been withdrawn
[38], all other models postulate 1:1 type interactions between CD and ethyl
pyruvate. Augustine [25] and Bartók [39,40] proposed that not only the qui-
nuclidine N but also the OH function of CD would be involved in the interac-
tions but adsorption of the rigid structures on the metal surface is unlikely due
to steric restrictions [17]. Four other groups have rationalized the stereochemi-
cal outcome of the reaction by assuming a single interaction between CD and
ethyl pyruvate, as depicted in Figure 3-1 [8,41-44]. Despite of the striking dif-
ferences in the details, all these models agree in one point: the critical role of
the basic quinuclidine N atom of CD in the interaction with the substrate. The
quinoline ring of the alkaloid is commonly considered as the “anchoring” moi-
ety that provides fixed adsorption on the metal surface [41,45].
Our approach is the following: we tried to influence the adsorption mode
of CD, and thus the properties of the “chiral site” available for the adsorption
of the substrate, by replacing the OH group by O-alkyl, O-aryl and O-silyl
Fig. 3-2: Enantioselective hydrogenation of activated ketones over modified Pt/Al2O3.
*
3 4
1 2
5 6
*
*
*
7 8
5 wt % Pt /Al2O3
H2, modifier
Fine tuning the “chiral sites” 45
groups of increasing bulkiness. We assume that this modification does not
influence directly the modifier-substrate interaction as the OH function is rela-
tively far away from the quinuclidine N atom of CD. If the adsorption mode of
the modifier is really important in the enantioselection, than the increasing
bulkiness of the OR group in the close neighborhood of the quinoline moiety
should have a strong effect on the ee of the hydrogenation reactions (or should
not have any influence according to the model in Figure 3-1b). The structure
of the ketone substrates and those of CD derivatives used as modifiers are
shown in Figure 3-2 and Figure 3-3. Compared to Chapter 2 we broaden the
scope of substrates and modifiers and extended the range of experiments by
nonlinear behavior studies of modifier mixtures and modifier stability consi-
derations. The results will be discussed from a mechanistic point of view and
should confirm the model already mentioned in the previous chapter [46].
Fig. 3-3: Ether derivatives of cinchonidine which were used as chiral modifiers in theenantioselective hydrogenation of activated ketones over Pt/Al2O3.
PhOCD
TMSOCD
EtOCD
MeOCD
CD
modifier R
N
HORNH
H
NaphOCD
HFXylOCD
XylOCD
modifier R
46 Chapter 3
3.3 Experimental
3.3.1 Materials
Tetrahydrofuran (THF, 99.5%, J. T. Baker) was dried over Na before use. Tol-
uene (99.5%, J. T. Baker), acetic acid (AcOH, 99.8%, Fluka), trifluoroacetic
acid (TFA, 99%, Fluka), α,α,α-trifluorotoluene (99%, Sigma-Aldrich), anisole
(99%, Fluka), QN (Fluka), CD (92%, Fluka; impurities: 1% quinine, 7% qui-
nidine, determined by HPLC at Fluka) and ketopantolactone (3, Roche) were
used as received. Ethyl pyruvate (1, Fluka), 4,4,4-trifluoroacetoacetate (5,
ABCR) and 1,1,1-trifluoro-2,4-diketopentane (7, Acros) were carefully dis-
tilled in vacuum before use.
3.3.2 Synthesis of ether derivatives of CD
Melting points were determined using Büchi-545 automatic melting point
apparatus. 1H-NMR spectra were recorded by Varian Unity Inova at 400 MHz
in CDCl3. All reactions were carried out under Ar atmosphere. Starting mater-
ials were purchased from Aldrich and used without further purification.
DMSO, DMF and THF were dried over molecular sieves before use. The syn-
thesis of MeOCD, EtOCD, PhOCD and TMSOCD has already been
described in Chapter 2.3.2. TMSOCD was prepared using the literature proce-
dure of silylation of CD with chlorotrimethylsilane in anhydrous THF in the
presence of triethylamine [47]. For the synthesis of NaphOCD, XylOCD, and
HFXylOCD the general arylation procedure described in Chapter 2.3.2 was
used starting from CD and the corresponding aryl iodide.
O-(1-Naphthyl)cinchonidine (NaphOCD)
Yield: 29%; 1H-NMR d = 8.74 (d, 1H), 8.57 (d, 1H), 8.28-8.17 (m, 2H),
7.83-7.54 (m, 5H), 7.45 (d, 1H), 7.35 (d, 1H), 7.07 (t, 1H), 6.44-6.36 (m,
2H), 5.86-5.68 (m, 1H), 5.03-4.91 (m, 2H), 3.40-3.30 (m, 1H), 3.28 (s, 2H),
2.75-2.55 (m, 2H), 2.13-2.10 (m, 1H), 1.80-1.70 (m, 3H), 1.60-1.50 (m,
2H).
Fine tuning the “chiral sites” 47
O-(3,5-Dimethylphenyl)cinchonidine (XylOCD)
Yield: 25%; 1H-NMR d = 8.82 (d, 1H), 8.17 (d, 2H), 7.81-7.62 (m, 2H),
7.46 (d, 1H), 6.54 (s, 1H), 6.42 (s, 2H), 6.03 (d, 1H), 5.78-5.64 (m, 1H),
4.99-4.88 (m, 2H), 3.40-3.30 (m, 1H), 3.28 (s, 2H), 2.75-2.55 (m, 2H), 2.15
(s, 6H), 2.13-2.10 (m, 1H), 1.80-1.70 (m, 3H), 1.60-1.50 (m, 2H).
O-[3,5-Bis(trifluoromethyl)phenyl]cinchonidine (HFXylOCD)
Yield: 22%; 1H-NMR d = 8.86 (d, 1H), 8.20 (d, 2H), 7.86-7.69 (m, 1H),
7.65 (m, 1H), 7.46 (d, 1H), 7.39 (s, 1H), 7.24 (s, 2H), 6.03 (d, 1H), 5.82-
5.64 (m, 1H), 5.05-4.95 (m, 2H), 3.40-3.30 (m, 1H), 3.28 (s, 2H), 2.75-2.55
(m, 2H), 2.13-2.10 (m, 1H), 1.80-1.70 (m, 3H), 1.60-1.50 (m, 2H).
3.3.3 Catalytic hydrogenation
The 5 wt% Pt/Al2O3 catalyst (Engelhard 4759; BET surface area: 100 m2g-1;
metal dispersion after reductive heat treatment: 0.27, determined by TEM)
and Al2O3 (support of the catalyst Engelhard 4759) were pretreated as men-
tioned in Chapter 2.3.1. After prereduction the catalyst was immediately trans-
ferred to the reactor within the next 10 min.
Hydrogenation reactions were done in three different reactors. Reactions at
1 bar were carried out in a magnetically stirred 100 ml glass reactor. For higher
pressure reactions a stainless steel autoclave equipped with a 50 ml glass liner
and a PTFE cover, and magnetic stirring (1000 rpm) was used. Total pressure
and hydrogen uptake were controlled by computerized constant-volume con-
stant pressure equipment (Büchi BPC 9901). For screening of reaction condi-
tions or testing different modifiers a parallel pressure reactor system Endeavor
(Argonaut Technologies) was used. This multiple reactor system contains eight
mechanically stirred, 15 ml stainless steel pressure reactors equipped with glass
liners. The same standard conditions were used as in Chapter 2.3.1: 42 mg cat-
alyst, 1.84 mmol substrate, 6.8 µmol modifer and 5 ml solvent (1000 rpm,
room temperature). Deviations from these conditions are indicated in the text.
The conversion and enantioselectivity were determined by gas chromatog-
raphy using a Chirasil-DEX CB column (Chrompack). The actual or incre-
mental ee was calculated as ∆ee = (ee1y1-ee2y2)/(y1-y2), where y represents the
48 Chapter 3
yield to the hydrogenation product, and index 2 refers to a sample subsequent
to sample 1.
3.3.4 Spectroscopic methods1H-NMR spectra for the modifier stability in Figure 3-9 were measured at
500 MHz using a DPX 500 spectrometer from Bruker.
UV-Vis measurements were performed in transmission mode on a
CARY 400 spectrophotometer using a 1 cm path length quartz cuvette. For the
UV-Vis study the catalyst and support were pretreated as described for the
hydrogenation experiments.
3.4 Results and discussion
3.4.1 Inversion of enantioselectivity
The ether derivatives of CD as chiral modifiers of a 5 wt% Pt/Al2O3 catalyst
have been tested in the hydrogenation of activated ketones under various reac-
tion conditions. Illustrative examples on the influence of the O-alkyl, O-silyl,
and O-aryl substituents are shown in Figure 3-4 - Figure 3-7. A general ten-
dency is observable in all four test reactions, independent of the reaction condi-
tions: with increasing bulkiness of the ether groups the enantioselectivities
decreased close to zero or even the opposite enantiomer formed in excess. The
steric effects are particularly striking in the hydrogenation of ketopantolactone
(3) in α,α,α-trifluorotoluene (Figure 3-5): Unmodified CD afforded 73% ee
to (R)-4 and the bulky ether derivatives gave 37-53% ee to (S)-4. More details
of the reactions including kinetic data (conversions) are collected in Table 3-1.
In most cases the ee was determined at full conversion of the ketone. Hydroge-
nation of the trifluoromethylketones 5 and 7 was somewhat slower and after 2
h the conversion varied between 20% and 100%.
In the hydrogenation of 1 the influence of the small O-alkyl groups in
MeOCD and EtOCD was minor, in agreement with the former proposal that
Fine tuning the “chiral sites” 49
Fig. 3-4: Diminished enantioselectivity with increasing bulkiness of the ether function ofthe modifiers in the hydrogenation of 1 over Pt/Al2O3 (standard conditions, toluene, 1 bar).
Fig. 3-5: Inversion of enantioselectivity in the hydrogenation of 3 over Pt/Al2O3 chiral-ly modified by the ether derivatives of CD (standard conditions, α,α,α-trifluorotoluene,40 bar).
80 (R)
60 (R)
40 (R)
20 (R)
0
20 (S)
40 (S)
ee [%]
5 PhOCD4 TMSOCD3 EtOCD2 MeOCD1 CD
8 NaphOCD7 HFXylOCD6 XylOCD
1 2 3 4
5 67
8
80 (R)
60 (R)
40 (R)
20 (R)
0
20 (S)
40 (S)
60 (S)
ee [%]
1 2 34 5 6 7 8
5 PhOCD4 TMSOCD3 EtOCD2 MeOCD1 CD
8 NaphOCD7 HFXylOCD6 XylOCD
50 Chapter 3
Fig. 3-6: Gradual loss of ee and the subsequent inversion of the sense of enantioselectionby increasing bulkiness of the ether groups of the modifiers in the hydrogenation of 5 overPt/Al2O3 (standard conditions, THF, 40 bar).
Fig. 3-7: Gradual loss of enantioselectivity by increasing bulkiness of the ether groupsof the modifiers in the hydrogenation of 7 over Pt/Al2O3 (standard conditions, toluene,10 bar).
80 (S)
60 (S)
40 (S)
20 (S)
0
20 (R)
40 (R)
ee [%]
1 2 3 4 5
6 78
5 PhOCD4 TMSOCD3 EtOCD2 MeOCD1 CD
8 NaphOCD7 HFXylOCD6 XylOCD
80 (S)
60 (S)
40 (S)
20 (S)
0
20 (R)
40 (R)
ee [%]
5 PhOCD4 TMSOCD3 EtOCD2 MeOCD1 CD
8 NaphOCD7 HFXylOCD6 XylOCD
1 2 3 4 5
67
8
Fine tuning the “chiral sites” 51
the OH function of CD is not involved in the enantio-differentiating step
[48,49]. In case of 3 a small decrease for MeOCD and EtOCD compared to
CD was measured (Table 3-1). In the hydrogenation of 5 and 7 MeOCD [50]
and EtOCD are even more effective modifiers than CD, likely due to a com-
peting interaction of the free OH function of CD with the substrate.
The bulky trimethylsilyl and aryl substituents have more pronounced
effect on the enantioselection though no clear correlation could be found
between the (estimated) bulkiness of these groups and the ees. Obviously, spe-
cial interactions among the reaction partners including the solvent and the
competing adsorbent hydrogen (see pressure effects) are also important. For
example, in the hydrogenation of 1 TMSOCD gave (R)-2 and PhOCD (S)-2,
independent of the reaction conditions. In contrast, in the hydrogenation of 3
both modifiers afforded the (S)-product and in α,α,α-trifluorotoluene even the
extent of enantioselection was practically the same.
Another interesting example is the efficiency of XylOCD and the hexaflu-
orinated derivative HFXylOCD. In most cases they afford the opposite enanti-
omers in excess. This difference is attributed to steric and electronic effects of
the methyl and trifluoromethyl groups [51,52]. Note also the remarkably dif-
ferent enantioselectivities in toluene and α,α,α-trifluorotoluene (Table 3-1).
We assume that the bulky trifluoromethyl group prevents the adsorption of the
aromatic ring of the solvent molecule parallel to the Pt surface via π-bondingand thus weaken its adsorption.
The bulkiest modifier NaphOCD gave for all substrates low selectivities,
independent of the conditions (with only one exception, see Table 3-1). The
poor efficiency of this modifier in the hydrogenation of 1 has been reported
recently [48]. The naphthyl ring of NaphOCD has about the same size as the
“anchoring” moiety of CD, the quinoline ring. Interaction of the naphthyl
group with the metal surface may be significantly different from those of the
other bulky groups. Among the possible interpretations, a feasible explanation
for the low ees is that adsorption of NaphOCD requires too big ensembles of
surface Pt atoms. The smaller ensembles of surface sites, which cannot accom-
modate the modifier, afford a racemic mixture and diminish the ee.
We propose the following general interpretation of the results. Replace-
ment of the OH function of CD by alkoxy, silyloxy and aryloxy groups of
52 Chapter 3
Table 3-1: The efficiency of ether derivatives of cinchonidine (CD) used as chiral modifiersof Pt/Al2O3 in the hydrogenation of activated ketones.
substrate solvent pressure
[bar] ee (%) ee (%) ee (%) ee (%) ee (%) ee (%) ee (%) ee (%)[conv.] [conv.] [conv.] [conv.] [conv.] [conv.] [conv.] [conv.]
PhCH
THF
THF
10
40
1
1
10
3[37] [68] [74] [44] [34] [25] [23] [32]
PhCH3
[99] [99] [99] [96] [95] [94] [96] [90]
[100] [100] [100] [100] [100] [100] [100] [100]PhCF 103
[100] [100] [100] [100] [100] [100] [100] [100]
PhCH3
[100] [100] [100] [100] [100] [100] [100] [100]
[100] [100] [100] [100] [100] [100] [100] [100]
THF 1
1PhCH3
[100] [100] [100] [100] [100] [100] [100]
[100] [100] [100] [100] [100] [100] [100] [100]
PhOCDTMSOCDEtOCDMeOCDCD NaphOCDHFXylOCDXylOCD
1
3
5
7
40[100] [100] [100] [100] [100] [100] [100] [100]
PhCF3
THF 1[77] [91] [94] [70] [42]
Full conversion after 30 min.a
a
a
Fine tuning the “chiral sites” 53
increasing bulkiness results in increasing steric hindrance against the adsorp-
tion of the modifier close to parallel to the Pt surface via the quinoline ring.
This gradual shift of the position of the quinoline ring repositions the quinucli-
dine N atom, and thus the substrate that is adsorbed on the neighboring Pt
sites and interacts with the modifier during hydrogenation. In this new posi-
tion of the substrate the steric hindrance against the formation of the minor
enantiomer may be lost or even the formation of the opposite enantiomer is
favored. Formally, we can describe this situation as shifting and reshaping the
“chiral pocket” available for adsorption of the ketone substrate. In the follow-
ing, we shall present some additional observations that support our proposal.
3.4.2 Stability of ether derivatives of cinchonidine under reaction
conditions
A fundamental question in the interpretation of the results in Table 3-1 is the
possible distortion of enantioselectivity by transformation of the modifier dur-
ing the hydrogenation of ketone substrates. It has been shown earlier that even
under mild conditions CD is hydrogenated on Pt resulting in less efficient
modifiers [45,53]. Transformation of CD includes hydrogenation of the C=C
bond and (partial) hydrogenation of the quinoline ring. The former side reac-
tion has barely any effect on ee but the (partially) hydrogenated quinoline moi-
ety adsorbs much weaker on Pt [54]. Here, a further possibility is the
hydrogenolytic removal of the bulky ether groups [55].
We have examined the stability of the ether derivatives of CD under reac-
tion conditions in various solvents. The effect of acidic medium is shown in
Figure 3-8 by the example of the hydrogenation of 1 over Pt/Al2O3 modified
by PhOCD. In THF this modifier gave a small excess to (S)-2 but in AcOH
(pKa = 4.75 [56]) (R)-2 formed in higher than 60% ee. The latter value is close
to that achieved with CD under the same conditions. The likely explanation is
that hydrogenolysis of the ether C-O-C bond is catalyzed by acids and some
CD forms in the early stage of ketone hydrogenation and CD is the actual
modifier of Pt. To prove this assumption we repeated the experiment in THF
in the presence of the strong acid TFA (pKa = 0.3 [56]). The gradual shift of
the major enantiomer from (S)-2 to (R)-2 indicate the formation of increasing
54 Chapter 3
amount of CD. We have shown in Chapter 2.4 that CD adsorbs much stron-
ger on the Pt surface than PhOCD and as soon as sufficient amount of CD
formed, this modifier controls the enantioselection [46]. Note that hydro-
genolysis of TMSOCD in acidic medium was even faster than that of PhOCD.
Accordingly, no acidic solvents were used in testing of the modifiers
(Table 3-1).
Next, the resistance of PhOCD against hydrogenation was examined in
non-acidic medium. The reactions were carried out under standard reaction
conditions but in the absence of any substrate and the modifier concentration
was increased by a factor of ten (Figure 3-9). No hydrogenation of the phenyl
group or hydrogenolysis of the ether C-O-C bond could be detected by1H-NMR. Hydrogenation took place at the vinyl group and to a smaller extent
at the quinoline ring. Hydrogenation of the anchoring part of the modifier
(quinoline ring) leads to a weaker adsorption of the modifier [54]. The
destructed modifier molecule desorbs from the Pt surface and is replaced by an
Fig. 3-8: Stability of PhOCD in the presence of acids, as illustrated by the hydrogenationof 1 over Pt/Al2O3 modified by PhOCD or CD (standard conditions, 1 bar, TFA:PhOCD= 5 molar equiv.; x - conversion).
10080604002040
ee [%]
20
CD, AcOH, x = 100%
PhOCD, AcOH, x = 100%
PhOCD, THF, TFA, x = 58%
PhOCD, THF, TFA, x = 83%
PhOCD, THF, TFA, x = 100%
PhOCD, THF, x = 100%
(S) (R)
Fine tuning the “chiral sites” 55
intact modifier molecule from solution or from the alumina support. Hence,
the transformation shown in Figure 3-9 should not distort the output of
ketone hydrogenation reaction until there is sufficient amount of intact modi-
fier present in the system.
The good stability of the ether connectivity of PhOCD in non-acidic
medium can be explained by the hydrogenation of anisole, the simplest phenyl
ether. Under ambient conditions in THF, hydrogenolysis of the C-O bond was
very slow, affording less than 1% benzene in 2 h (Figure 3-10). Saturation of
Fig. 3-9: Transformation of PhOCD on Pt/Al2O3.
Fig. 3-10: Hydrogenation of anisole on Pt/Al2O3 (6.8 µmol in 5 ml solvent, otherwisestandard conditions).
N
HON
H
H
12 % hydrogenationof the quinoline ring
94 % hydrogenationof the vinyl group
No hydrogenation orhydrogenolysis of thephenyl group
5 wt % Pt/Al O
H , THF, rt, 2h2 3
2
PhOCD
5 wt % Pt/Al O
H , THF, rt, 2h2 3
2
56 Chapter 3
the phenyl ring was about 20-fold faster but the yield to methyl cyclohexyl
ether dropped rapidly with increasing pressure. Parallel to this change, benzene
formation was not detectable at 10 bar or above. The probable explanation to
this unusual reactivity is that at high surface hydrogen concentration (at high
pressure) the increased competition between hydrogen and anisole hinders the
p-bonded adsorption of the latter parallel to the metal surface and thus slows
down the saturation of the phenyl ring. Using this observation as an analogy we
can deduce that the absence of saturation of the phenyl ring in PhOCD is a
strong indication to the tilted adsorption of the phenyl ring of PhOCD relative
to the Pt surface. In this position the C-O-C fragment (ether bond) of the
modifier is far from the Pt surface and its hydrogenolysis is improbable.
In order to prove that transformation of the chiral modifiers is unimpor-
tant compared to the rate of hydrogenation of the ketone substrates, we made
some comparative study under standard conditions (Table 3-2). Three modifi-
ers have been chosen for this comparison: CD and its ether derivatives with a
small alkyl (MeOCD) and a bulky aryl (PhOCD) group. The rate of saturation
Table 3-2: Initial rate of conversion of substrates and modifiers followed by GC andUV-Vis, respectively (standard conditions, THF, 1 bar).
3680
substrate initial rate [µmol/h]modifier
substrate modifier
1 CD64503 CD91203 MeOCD
7105 CD8405 MeOCD7805 PhOCD4507 MeOCD
0.47CD0.32MeOCD0.11PhOCD
-------
---
---
Fine tuning the “chiral sites” 57
of the quinoline ring of the modifiers was determined by UV-Vis analysis in
the absence of substrate (for details see later). The reactivity order of modifiers
(CD > MeOCD > PhOCD) is the opposite to that of increasing bulkiness of
the OH < OMe < OPh groups. That is, increasing bulkiness of the OR groups
decreases the hydrogenation rate of the quinoline moiety of the chiral modifi-
ers. This observation is in agreement with our preliminary assumption that
increasing bulkiness of the ether group should prevent the adsorption of the
quinoline ring parallel to the Pt surface. It is expected that the more tilted the
position of the quinoline ring on the surface, the slower is the rate of its hydro-
genation.
The initial rates of hydrogenation of the ketone substrates (determined by
GC analysis) depended on the modifier but in all cases it was faster than that of
the corresponding modifier by a factor of 1400-28500 (Table 3-2). Under stan-
dard conditions the substrate/modifier ratio was 270. It can be deduced that
even in the worst case there was sufficient amount of chiral modifier present in
the reaction mixture to replace the hydrogenated, weakly adsorbing modifier
molecules on the Pt surface.
3.4.3 Nonlinear behavior of modifier mixtures
The non-linear effect (NLE) of enantiomerically impure ligands has been a
topic of great interest in homogeneous catalysis [57-59]. The concept can be
extended to two different ligands giving products of opposite configuration
[60,61], and also to chiral modifiers in heterogeneous asymmetric catalysis
[62,63]. Studying the non-linear behavior of mixtures of two modifiers is a
powerful tool in heterogeneous catalysis to characterize the relative adsorption
strength of modifiers under truly in situ conditions [54]. The non-linear behav-
ior is considered as a deviation from the expected ideal behavior assuming that
the molar ratios of the modifiers in solution and on the metal surface are iden-
tical, and the reaction rates and ees are linear combinations of those measured
by the two modifiers alone.
The simplest method to clarify the relative adsorption strength of modifi-
ers is to carry out the reaction in the presence of 1:1 mixtures of two modifiers
that give the opposite enantiomers in excess. Hydrogenation of 1 over Pt/Al2O3
58 Chapter 3
modified by MeOCD and PhOCD is shown in Figure 3-11 as a typical exam-
ple. The first two columns represent the result of the reaction with the modi-
fiers alone and the last column shows the enantioselectivity provided by the 1:1
mixture. The calculated ee based on linear combination is 5% to (R)-2 but the
measured ee was 68%, a value close to that given by MeOCD alone (79%). It
indicates that PhOCD adsorbs weaker on Pt and has only a minor influence on
the enantioselection when mixtures of the two modifiers are used.
In order to obtain a more general correlation, the experiments have been
extended to the hydrogenation of 3 and more cinchona derivatives have been
involved. On the basis of these experiments we propose the following order of
adsorption strength on Pt: CD > MeOCD > EtOCD > PhOCD = TMSOCD.
Apparently, the bulkier the ether group of the cinchona derivative, the weaker
is the adsorption on Pt/Al2O3. It has to be emphasized that this conclusion is
related to real in situ conditions, during transformation of 1 or 3. None of the
Fig. 3-11: Enantioselectivities in the hydrogenation of 1 over Pt/Al2O3 modified byMeOCD or PhOCD, or by an equimolar mixture of the two modifiers (standard condi-tions, THF, 1 bar).
1:1 mixture of
90 (R)
75 (R)
60 (R)
30 (R)
15 (S)
30 (S)
45 (R)
15 (R)
0
MeOCD PhOCDMeOCD and PhOCD
ee [%]
Fine tuning the “chiral sites” 59
known physico-chemical methods offers this possibility as we summarized it in
the introduction.
We applied also a transient method [54,62] to visualize the competition
between the various CD derivatives (Figure 3-12). Hydrogenation of 3 in THF
over Pt/Al2O3 afforded 50% ee to (R)-4 in the presence of CD and 26% ee to
(S)-4 when TMSOCD was applied alone as modifier; both values were mea-
sured at full conversion. In a transient experiment the reaction was started
with Pt/Al2O3 modified by CD and after 20 min one molar equivalent
TMSOCD related to CD was injected into the slurry. The enantioselectivity
was not affected by the addition of TMSOCD; the ee – time correlation
remained practically the same as measured with CD alone. In the control
experiment the reaction was started with Pt/Al2O3 modified by TMSOCD and
after 20 min one molar equivalent CD was added. The rapid shift of ee and
particularly that of the calculated differential ee (∆ee) demonstrate that within a
Fig. 3-12: Transient behavior of the hydrogenation of 3 over Pt/Al2O3 induced by additionof one molar equivalent of TMSOCD or CD to the reaction mixture containing CD or TM-SOCD, respectively. Standard conditions, THF, 1 bar; second modifier added in 1 mL THF;∆ee: differential ee.
+ TMSOCD60 (R)
50 (R)
30 (R)
10 (R)
10 (S)
20 (S)
30 (S)
40 (R)
20 (R)
0
0 10 20 30 40 50 60 70 80
time [min]
TMSOCD+ CD
CD ∆ee
ee, ∆ee [%] ee
ee
60 Chapter 3
few minutes CD replaced TMSOCD on the Pt surface resulting in the inver-
sion of enantioselectivity. Obviously, CD adsorbs much stronger on Pt than the
bulky ether derivative TMSOCD.
Stronger adsorption of the less bulky ether derivative is shown in
Figure 3-13 by the competition of MeOCD and PhOCD. Hydrogenation of 3
on Pt/Al2O3 modified by MeOCD afforded 40% ee to (R)-4 at full conversion,
and 21% ee to (S)-4 when PhOCD was applied. When the reaction was started
with MeOCD-modified Pt/Al2O3, addition of PhOCD after 20 min led to a
small decrease in enantioselectivity. In the reverse case, addition of MeOCD to
the reaction mixture containing PhOCD resulted in a bigger shift and even
inversion of the sense of enantioselection. The time-dependent changes of the
differential ee show that MeOCD controlled the enantioselection on the Pt sur-
face modified by the equimolar mixture, and the actual ee at full conversion
was about the same, independent of the order of addition of the two modifiers.
Fig. 3-13: Transient behavior in the hydrogenation of 3 on Pt/Al2O3 induced by additionof one molar equivalent of MeOCD or PhOCD to the reaction mixture containing PhOCDor MeOCD, respectively. Standard conditions, THF, 1 bar; second modifier added in 1 mLTHF; ∆ee: differential ee.
PhOCD
MeOCD
+ MeOCD
+ PhOCD∆ee
0 10 20 30 40 50 60 70 80
time [min]
40 (R)
30 (R)
10 (R)
10 (S)
20 (S)
30 (S)
20 (R)
0ee, ∆ee [%]
ee
ee
ee
Fine tuning the “chiral sites” 61
A comparison of Figure 3-12 and Figure 3-13 indicates that the difference in
adsorption strength between MeOCD and PhOCD is smaller than that
between CD and TMSOCD.
Overall, the transient experiments and the reactions using equimolar mix-
tures of modifiers indicated the same order of adsorption strength of CD deriv-
atives: CD adsorbs stronger on Pt than any of its ether derivatives, and the
ether derivatives with small alkyl groups adsorb stronger than the bulky trime-
thylsilyl or phenyl ethers (CD > MeOCD > EtOCD > PhOCD = TMSOCD).
These results are in line with the adsorption model we propose for the ether
derivatives of CD: the bulky ether groups prevent the adsorption of the quino-
line ring parallel to the Pt surface. As it was shown by ATR-IR measurements
[36, 37], in a tilted position of the quinoline ring relative to the Pt surface the
modifier adsorbs weaker (via the quinoline N atom) than in a close to parallel
position via the strong π-bonding between the quinoline ring and the Pt sur-
face.
3.4.4 UV-Vis study of modifier adsorption and hydrogenation
Interaction of the ether derivatives of CD with Pt/Al2O3 in the presence of
hydrogen but in the absence of ketone substrate was analyzed by an indirect
UV-Vis spectroscopic method (Figure 3-14). The disappearance of modifiers
from the THF solution was followed by the quinoline chromophore at 315 nm
[54]. A decrease of the modifier concentration detected by UV-Vis is attributed
to adsorption on Pt and on the Al2O3 support, and to partial or complete satu-
ration of the quinoline ring (Figure 3-9). It is commonly accepted that satura-
tion of the quinoline ring of the modifier leads to weaker adsorption on Pt and
the hydrogenated modifier is replaced by an intact modifier from solution
[12,53,54,64]. Adsorption on the support was taken into consideration by
repeating the analysis with the support alone. From the difference of the two
curves (Figure 3-14a) the consumption of each modifier due to adsorption and
hydrogenation on Pt was calculated (Figure 3-14b). The initial amounts
extrapolated to zero time represent the fraction of modifiers adsorbed on Pt
and the rate of hydrogenation of the quinoline ring is estimated from the slopes
of the lines. Clearly, under identical conditions remarkably more CD adsorbed
62 Chapter 3
Fig. 3-14: Kinetic analysis of the adsorption and hydrogenation of CD, MeOCD, andPhOCD (standard conditions but in the absence of substrate, 1 bar, THF). a) Filled symbolsrepresent adsorption and hydrogenation on Pt/Al2O3, open symbols show adsorption on thecatalyst support (Al2O3). The actual modifier concentration in solution was determined byUV-Vis spectroscopy from the absorbance at 315 nm. b) The dark columns represent the cal-culated ratio between the modifier molecules and surface Pt atoms (Pts). The light columnsshow the estimated initial rates of conversion of the modifiers.
0.20
0.18
0.16
0.14
0.12
0.10
0.08
0.06
0.04
0.02
0CD MeOCD PhOCD
0.5
0.4
0.3
0.2
0.1
0
Mod/Pts[mol/mol]
rate of hydrogenation[µmol/h]
7
6
5
4
30 50 100 150 200
CD
MeOCD
PhOCD
mod. in solution
time [min]a
b
[µmol]
Fine tuning the “chiral sites” 63
on the Pt surface than MeOCD or PhOCD, and the rate of their hydrogena-
tion followed the same order (CD > MeOCD > PhOCD). The order of the
adsorption strength of modifiers is the same as that concluded from the nonlin-
ear behavior of modifier mixtures in the presence of substrates. Note that the
stronger adsorption of CD relative to the ether derivatives on Al2O3 is likely
due to interaction between the OH function of CD and the basic O-atoms of
Al2O3.
In the former chapter the non-linear behavior of modifier mixtures in the
hydrogenation of 1 and 3 was attributed to the different adsorption strength of
the modifiers that results in considerably different concentrations on the Pt
surface. To prove this interpretation, we investigated the relative rate of the
hydrogenation of modifiers in equimolar mixtures by UV-Vis spectroscopy, in
the absence of ketone substrate. Since the UV-Vis spectra of CD and its ether
derivatives are the same, quinine (QN) was used as a reference modifier. In the
latter molecule the presence of the 6’-methoxy group at the quinoline ring
shifts the adsorption band from 315 nm to 338 nm [54].
At first the competitive hydrogenation of CD and QN was investigated
(Figure 3-15a). When the two alkaloids were hydrogenated alone, CD disap-
peared from the solution somewhat faster than QN. When the 1:1 mixture of
CD and QN was hydrogenated, the difference in their reactivity was much big-
ger. This change is attributed to the stronger adsorption of CD on Pt, i.e. in the
equimolar mixture the much faster consumption of CD is mainly due to its
higher surface concentration and partly to its higher reactivity. Next, the reac-
tivity and the adsorption strength of MeOCD (Figure 3-15b) and TMSOCD
(Figure 3-15c) were determined using QN as a reference. It is clear from the
very slow consumption of TMSOCD that this modifier adsorb much weaker
than QN and cannot efficiently compete for the Pt surface sites. The difference
between MeOCD and QN is smaller, as expected. The order of adsorption
strength obtained from these experiments (CD > MeOCD > TMSOCD) con-
firms that the bigger the ether group of the modifier, the weaker is the adsorp-
tion and slower the hydrogenation on Pt. This conclusion supports the
interpretation of the non-linear behavior of modifier mixtures measured in the
hydrogenation of ketone substrates (Figure 3-11 - Figure 3-13).
64 Chapter 3
Fig. 3-15: Kinetic analysis of the hydrogenation of modifier mixtures (standard conditionsbut in the absence of ketone substrate, THF, 1 bar). Top (a): hydrogenation of CD and QNalone and in a 1:1 mixture (index m). Middle (b): hydrogenation of QN alone and a 1:1 mix-ture of QN and MeOCD (index m). Bottom (c): hydrogenation of QN alone and a 1:1 mix-ture of QN and TMSOCD (index m).
QN
CD
CDm
QNm
7
6
5
4
3
7
6
5
4
3
mod. in solution
a
b
0 50 100 150 200
time [min]
7
6
5
4
3 c
m
m
m
m
[µmol]
mod. in solution [µmol]
mod. in solution [µmol]
Fine tuning the “chiral sites” 65
3.4.5 Mechanistic considerations
Various models have been proposed for the enantioselective hydrogenation of
activated ketones on cinchona-modified Pt [8,25,40-44]. These models assume
that in the enantio-differentiating step the quinoline ring of CD is adsorbed
either close to parallel or perpendicular to the Pt surface, or the modifier is in
an “upside down” position above the ketone substrate (“shielding” model). The
present study provides an interesting set of experimental data to test the feasi-
bility of these mechanistic ideas.
Our primary observation here is that replacement of the OH group of CD
by a relatively small methoxy or ethoxy group has only moderate effect on the
enantioselectivity but introduction of bulky silyloxy and aryloxy groups dimi-
nish the ee close to zero or even the opposite enantiomer is formed in a reason-
able excess. Obviously, the “shielding” model depicted in Figure 3-1b cannot
explain this dramatic shift in enantioselection. We mentioned in the introduc-
tion that two other models (by Augustine’s [25] and Bartók’s [39,40] groups)
that assume involvement of the O atom of CD in an electron-pair donor inter-
action are unlikely due to steric restrictions. The better than 50% ee achieved
with the bulky ether derivatives of CD (Table 3-1) provides further experimen-
tal evidence against the probability of these models. Involvement of the steri-
cally “hidden” ether O of the modifier in the enantiodifferentiating complex
can be excluded.
The second major result of the present study is that increasing bulkiness of
the ether groups of CD derivatives decreases the adsorption strength of the
modifier and diminishes also the rate of hydrogenation of the quinoline ring, a
well-known disturbing side reaction on cinchona-modified Pt and Pd. A fea-
sible interpretation of these observations is based on a recent ATR-IR spectro-
scopic study of CD adsorption under close to in situ conditions [36,37]. It has
been proved that in the strongest adsorption mode CD adopts a position in
which the π-bonded quinoline ring is close to parallel to the Pt surface, and a
change of the adsorption mode to a tilted position of the quinoline ring
(anchored to Pt via the N-atom) results in a considerable weakening of the
modifier-Pt interaction. The suggested adsorption geometries of CD and
PhOCD as examples are visualized in Figure 3-16. The bulky phenoxy group
66 Chapter 3
hinders the strong π-bonded adsorption of the quinoline ring. In the suggested
tilted adsorption mode the hydrogenation of the quinoline ring is slow and
hydrogenolysis of the ether C-O-C connectivity is not detectable (in non-
acidic medium, in which the modifiers were tested). Weaker adsorption of the
bulky ether derivatives of CD explains the strong non-linear effect observed in
the hydrogenation of 1 and 3 when mixtures of two modifiers were applied
(Figure 3-11 - Figure 3-13). The tilted position of the phenyl ring is in agree-
ment with its high stability against hydrogenation in non-acidic medium. For
comparison the aromatic ring of the model compound anisole is slowly satu-
rated at low pressures (Figure 3-10).
Two mechanistic models in Figure 3-1 (a and d) assume that in the enan-
tio-differentiating step the quinoline ring is adsorbed close to parallel to the Pt
Fig. 3-16: Schematic illustration of the adsorption of CD and PhOCD on an idealized flatPt surface via the aromatic rings (top-side views; the metal surface is not shown). The struc-tures of the molecules are optimized by HyperChem’s MM+ force field without any involve-ment of the solvent or Pt surface.
Fine tuning the “chiral sites” 67
surface, thus they are in agreement with the present results. The model
depicted in Figure 3-1c is similar to the above two structures concerning the
adsorption geometry of CD. This model, however, predicts that the enantiose-
lection is due to formation of a zwitterionic intermediate with a tetrahedral C
atom (from the keto-carbonyl group of the substrate) [43]. This structure is
sterically impossible with the cyclic ketone 3, in contrast to the good enantiose-
lectivities achieved in the hydrogenation of 3 using CD and some of its ether
derivatives as chiral modifiers of Pt (Table 3-1). Obviously, this model cannot
be generally valid for the hydrogenation of activated ketones as suggested by
the authors. It is also unlikely that the structurally related 1 and 3 (acyclic and
cyclic α-keto esters, respectively) would follow different reaction mechanisms.
3.5 Conclusions
Understanding the nature of substrate-modifier interaction and elimination of
the most speculative element of the available mechanistic models necessitates
clarification of the adsorption mode of the interacting molecules on the metal
surface under real in situ conditions. Since none of the spectroscopic or surface
science techniques could fulfill this requirement so far, we choose another
approach: the systematic variation of the structure of chiral modifier, a proven
strategy in homogeneous enantioselective catalysis.
In heterogeneous catalysis several surface metal atoms, which are located at
a certain distance to each other, interact with the substrate and the chiral mo-
difier. Due to adsorption on the metal surface, relatively small changes in the
structure of the chiral modifier far from the interacting function may result in
dramatic shifts in the enantioselectivity. Systematic variation of the modifier
structure confirmed that in the enantio-differentiating step during the hydro-
genation of activated ketones on cinchona-modified Pt the alkaloid adsorbs via
the quinoline ring being close to parallel to the Pt surface. A forced deviation
from this adsorption geometry diminishes the adsorption strength and the
enantioselectivity, and even the opposite enantiomer can form in excess.
68 Chapter 3
The results illustrate also the efficiency of developing new catalyst systems
by fine tuning the structure of the chiral modifier.
3.6 References
[1] A. Baiker, H.-U. Blaser, in: Handbook of Heterogeneous Catalysis, eds. G.
Ertl, H. Knoezinger, J. Weitkamp, vol. 5, VCH, Weinheim, 1997, p. 2422.
[2] M. E. Davis, Top. Catal. 2003, 25, 3.
[3] T. Osawa, T. Harada, O. Takayasu, Top. Catal. 2000, 13, 155.
[4] T. Bein, Curr. Opin. in Solid State Mater. Sci. 1999, 4, 85.
[5] P. B. Wells, K. E. Simons, J. A. Slipszenko, S. P. Griffiths, D. F. Ewing, J.
Mol. Catal. A: Chem 1999, 146, 159.
[6] C. J. Baddeley, Top. Catal. 2003, 25, 17.
[7] Y. Orito, S. Imai, S. Niwa, J. Chem. Soc. Japan 1979, 8, 1118.
[8] A. Baiker, J. Mol. Catal. A: Chem 1997, 115, 473.
[9] B. Török, K. Felföldi, K. Balàzsik, M. Bartók, Chem. Commun. 1999, 17,
1725.
[10] M. Schürch, N. Künzle, T. Mallat, A. Baiker, J. Catal. 1998, 176, 569.
[11] M. von Arx, T. Mallat, A. Baiker, Tetrahedron: Asymmetry 2001, 12, 3089.
[12] M. von Arx, T. Mallat, A. Baiker, Top. Catal. 2002, 19, 75.
[13] M. Studer, H.-U. Blaser, C. Exner, Adv. Synth. Catal. 2003, 345, 45.
[14] Y. Nitta, A. Shibata, Chem. Lett. 1998, 161.
[15] I. Kun, B. Török, K. Felföldi, M. Bartók, Appl. Catal. A 2000, 203, 71.
[16] W.-R. Huck, T. Mallat, A. Baiker, New J. Chem. 2002, 26, 6.
[17] K. Borszéky, T. Bürgi, Z. Zhaohui, T. Mallat, A. Baiker, J. Catal. 1999,
187, 160.
[18] M. Maris, W.-R. Huck, T. Mallat, A. Baiker, J. Catal. 2003, 219, 52.
[19] A. Tungler, G. Fogassy, J. Mol. Catal. A: Chem 2001, 176, 231.
[20] R. Noyori, Asymmetric Catalysis in Organic Synthesis, Wiley-VCH, New
York, 1994.
[21] B. E. Hanson, J. R. Zoeller, Catal. Today 1998, 42, 371.
[22] W. S. Knowles, Angew. Chem., Int. Ed. 2002, 41, 1999.
Fine tuning the “chiral sites” 69
[23] W. A. Hofer, V. Humblot, R. Raval, Surf. Sci. 2004, 554, 141.
[24] J. M. Bonello, F. J. Williams, R. M. Lambert, J. Am. Chem. Soc. 2003,
125, 2723.
[25] R. L. Augustine, S. K. Tanielyan, J. Mol. Catal. A: Chem 1996, 112, 93.
[26] R. L. Augustine, S. K. Taneilyan, L. K. Doyle, Tetrahedron: Asymmetry
1993, 4, 1803.
[27] T. Evans, A. P. Woodhead, A. Gutierrez-Sosa, G. Thornton, T. J. Hall, A.
A. Davis, N. A. Young, P. B. Wells, R. J. Oldman, O. Plashkevych, O. Vahtras,
H. Agren, V. Carravetta, Surf. Sci. 1999, 436, L691.
[28] T. E. Jones, C. J. Baddeley, Surf. Sci. 2002, 519, 237.
[29] T. Bürgi, F. Atamny, R. Schlögl, A. Baiker, J. Phys. Chem. B 2000, 104,
5953.
[30] A. F. Carley, M. K. Rajumon, M. W. Roberts, P. B. Wells, J. Chem. Soc.,
Faraday Trans. 1995, 91, 2167.
[31] I. Bakos, S. Szabó, M. Bartók, E. Kálmán, J. Electroanal. Chem. 2002,
532, 113.
[32] J. Kubota, F. Zaera, J. Am. Chem. Soc. 2001, 123, 11115.
[33] Z. Ma, J. Kubota, F. Zaera, J. Catal. 2003, 219, 404.
[34] W. Chu, R. J. LeBlanc, C. T. Williams, Catal. Commun. 2002, 3, 547.
[35] R. J. LeBlanc, W. Chu, C. T. Williams, J. Mol. Catal. A: Chem 2004, 212,
277.
[36] D. Ferri, T. Bürgi, A. Baiker, Chem. Commun. 2001, 1172.
[37] D. Ferri, T. Bürgi, J. Am. Chem. Soc. 2001, 123, 12074.
[38] G. Webb, P. B. Wells, Catal. Today 1992, 12, 319.
[39] M. B. Bartók, K. Felföldi, G. Szöllösi, T. Bartók, Catal. Lett. 1999, 61, 1.
[40] M. B. Bartók, F. Notheisz, React. Kinet. Catal. Lett. 2002, 77, 363.
[41] A. Baiker, J. Mol. Catal. A: Chem 2000, 163, 205.
[42] J. L. Margitfalvi, E. Tfirst, J. Mol. Catal. A: Chem 1999, 139, 81.
[43] G. Vayner, K. N. Houk, Y.-K. Sun, J. Am. Chem. Soc. 2004, 126, 199.
[44] S. Lavoie, M.-A. Laliberté, P. H. McBreen, J. Am. Chem. Soc. 2003, 125,
15756.
[45] J. T. Wehrli, A. Baiker, D. M. Monti, H.-U. Blaser, J. Mol. Catal. A: Chem
1989, 49, 195.
[46] S. Diezi, T. Mallat, A. Baiker, Tetrahedron: Asymmetry 2003, 14, 2573.
70 Chapter 3
[47] A. Lindholm, P. Mäki-Arvela, E. Toukoniitty, T. A. Pakkanen, J. T. Hirvi,
T. Salmi, D. Y. Murzin, R. Sjöholm, R. Leino, J. Chem. Soc. Perkin Trans. 1
2003, 23, 2605.
[48] H.-U. Blaser, H. P. Jalett, W. Lottenbach, M. Studer, J. Am. Chem. Soc.
2000, 122, 12675.
[49] M. Schürch, O. Schwalm, T. Mallat, J. Weber, A. Baiker, J. Catal. 1997,
169, 275.
[50] M. von Arx, T. Mallat, A. Baiker, J. Catal. 2000, 193, 161.
[51] K. Mikami, Y. Itoh, M. Yamanaka, Chemical Reviews 2004, 104, 1.
[52] B. E. Smart, J. Fluorine Chem. 2001, 109, 3.
[53] M. Bartók, G. Szöllösi, K. Balàzsik, T. Bartók, J. Mol. Catal. A: Chem
2002, 177, 299.
[54] W.-R. Huck, T. Bürgi, T. Mallat, A. Baiker, J. Catal. 2003, 216, 276.
[55] F. Notheisz, M. Bartók, Fine Chemicals through Heterogeneous Catalysis,
Wiley-VCH, Weinheim, 2001.
[56] M. von Arx, T. Bürgi, T. Mallat, A. Baiker, Chem. Eur. J. 2002, 8, 1430.
[57] M. Avalos, R. Babiano, P. Cintas, J. Jimenez, J. C. Palacios, Tetrahedron:
Asymmetry 1997, 8, 2997.
[58] C. Girard, H. B. Kagan, Angew. Chem., Int. Ed. 1998, 37, 2922.
[59] H. B. Kagan, Adv. Synth. Catal. 2001, 343, 227.
[60] M. Kitamura, S. Suga, M. Niwa, R. Noyori, J. Am. Chem. Soc. 1995, 117,
4832.
[61] H. B. Kagan, Synlett 2001, 888.
[62] M. Schürch, T. Heinz, R. Aeschimann, T. Mallat, A. Pfaltz, A. Baiker, J.
Catal. 1998, 173, 187.
[63] W.-R. Huck, T. Mallat, A. Baiker, Adv. Syn. Catal. 2003, 345, 255.
[64] C. LeBlond, J. Wang, A. T. Andrews, Y.-K. Sun, Topics Catal. 2000, 13,
169.
Chapter4
4An efficient synthetic chiral modifierfor platinum
4.1 Summary
A new chiral modifier (R,R)-pantoyl-naphthylethylamine (PNEA) was synthe-
sized by reductive alkylation of (R)-1-(1-naphthyl)ethylamine with ketopanto-
lactone. Platinum-on-alumina modified by PNEA afforded 93% ee and 100%
chemoselectivity in the hydrogenation of the activated carbonyl group of 1,1,1-
trifluoro-2,4-pentanedione. Reductive heat treatment and ultrasonication of
the catalyst, and the use of chlorinated solvents under mild conditions (10 bar,
10°C) enhanced the enantioselectivity. This is the first case in heterogeneous
catalysis that a synthetic modifier gives more than 90% ee, better than the com-
monly used modifier of natural origin (cinchonidine or O-methyl-cinchoni-
dine).
4.2 Introduction
Enantioselective hydrogenation of C=O and C=C bonds has been one of the
most intensively studied areas in asymmetric catalysis [1]. Compared to the
number of highly effective homogeneous asymmetric catalysts [2-4], the variety
and application range of heterogeneous asymmetric catalysts are still limited.
Nevertheless, heterogeneous catalysts have obvious advantages in handling and
separation and even continuous process operation is feasible [5-7]. Metal
hydrogenation catalysts modified by trace amounts of a strongly adsorbing
74 Chapter 4
chiral compound (modifier) offer a synthetically useful alternative to the
homogeneous catalytic hydrogenation of ketones and olefins [8-13]. The best
chiral modifiers are naturally occurring compounds. A thoroughly investigated
catalyst system is cinchonidine (CD)-modified Pt that provides higher than
90% ee in the hydrogenation of various activated ketones [14]. Any effort to
surpass the performance of cinchona alkaloids or their simple derivatives by
using synthetic organic compounds has failed [15-21].
Chiral fluorinated compounds have attracted great attention in the past
years in agro- and pharmaceutical chemistry [22,23]. A viable route to chiral
α,α,α-trifluoromethyl alcohols is the enantioselective hydrogenation of trifluo-
romethyl ketones. Some chiral transition metal catalysts afforded excellent
yields and ees up to 98% [24,25]. The efficiency of the platinum-cinchona sys-
tem in this reaction class is extremely substrate specific: at best 92–96% ee was
achieved under mild reaction conditions [26,27] but for many substrates the
enantioselectivities were medium to low or no reaction occurred at all [28-31].
Here we show that a new synthetic chiral modifier for Pt provides over 90% ee
in the hydrogenation of 1,1,1-trifluoro-2,4-pentanedione 1 (Figure 4-1) a reac-
tion for which CD is less effective.
Fig. 4-1: Asymmetric hydrogenation of 1,1,1-trifluoro-2,4-pentanedione (1) over Pt/Al2O3 chirally modified by (R,R)-pantoyl-naphthylethylamine (PNEA).
5 wt% Pt/Al 2O3 H2 ,solvent1
2a
PNEA2b
2c
An efficient synthetic modifier 75
4.3 Experimental
4.3.1 Materials
1,1,1-Trifluoro-2,4-pentanedione 1 (Acros), cinchonidine (CD, Fluka), dichlo-
romethane (J. T. Baker), and acetic acid (Fluka) were used as received. Toluene
(J. T. Baker) was dried and stored over activated molecular sieve. The new
modifier (R,R)-pantoyl-naphthylethylamine (PNEA) was synthesized starting
from (R)-1-(1-naphthyl)ethylamine (NEA) and ketopantolactone via reductive
alkylation. The preferred method was formation of the imine in tita-
nium(IV)isopropoxide and reduction with NaBH3CN. The two diastereoiso-
mers were separated by flash chromatography [32]. The 5 wt% Pt/Al2O3
(E4759) catalyst was purchased from Engelhard. The metal dispersion was
0.32 and 0.20 before and after reductive heat treatment, respectively, as calcu-
lated from the average particle size determined by TEM [33].
4.3.2 Ultrasonic pretreatment
A multi-ultrasonic bath (Elma TI-H-5) was used for this catalyst pretreatment
at 20°C. The 50 mL glass liner of the autoclave was equipped with a gas inlet
and a rubber septum to enable the sonication under hydrogen. The slurry con-
taining the solvent, catalyst, and modifier was sonicated for the required time
(optimally 50 min); the substrate was injected to the reaction mixture only
after sonication.
4.3.3 Catalytic hydrogenations
The hydrogenation reactions were carried out in a mechanically stirred eight
parallel pressure reactor system (Argonaut Technologies) or in a magnetically
stirred stainless steel autoclave controlled by a computerized constant-volume
constant-pressure equipment (Büchi BPC9901). Optimally, the 5 wt% Pt/
Al2O3 catalyst was pre-reduced before use in a fixed-bed reactor by flushing
with N2 at 400°C for 30 min, followed by reductive treatment in H2 for
60 min at the same temperature. After cooling to room temperature in H2
76 Chapter 4
(30 min), the catalyst was used directly for hydrogenation or it was first soni-
cated before the hydrogenation reaction. Under standard conditions 42 mg cat-
alyst, 1.84 mmol substrate, 6.8 µmol modifier, and 5 ml solvent were stirred(1000 rpm) at 10 bar and room temperature (23-25°C) for 2 h.
Conversion and ee were determined by gas chromatography using a Chira-
sil-DEX CB column (Chrompack). The product was identified by GC/MS
(HP5973 mass spectrometer) and by 1H– and 13C–NMR (Bruker DPX 500
spectrometer). The enantiomers were identified by comparing the sign of their
optical rotation (Perkin Elmer 241 polarimeter) with literature data [34].
The reaction rate was characterized by the TOF value, that is the number
of moles of 1 converted by one mole of surface Pt atoms in one hour.
4.4 Results and discussion
4.4.1 Chemoselectivity
In the enantioselective hydrogenation of 1 over Pt/Al2O3 modified by PNEA
the (S)-α,α,α-trifluoromethyl alcohol 2a was produced in excess, similary to
the reactions where CD was used as a modifier (Figure 4-1). An interesting fea-
ture of the hydrogenation of this β-diketone is that in the absence of chiral
modifier the chemoselectivity is poor due to competing hydrogenation of the
non-activated keto group. For example, under standard conditions the target
molecule 2a formed with 79% chemoselectivity in toluene and with 25%
chemoselectivity in acetic acid. But addition of even trace amounts of PNEA
completely suppressed the side reactions and only the main product was
formed by hydrogenation of the activated carbonyl group. The same effect was
observed when CD was used as modifier [30]. The dramatic improvement in
chemoselectivity is probably due to interactions with the basic amine function
of the modifier. Another important effect may be the site blocking, i.e. the co-
verage of a considerable fraction of surface Pt sites by the strongly adsorbing
modifier. By decreasing the active site/reactant ratio the chemoselectivity may
An efficient synthetic modifier 77
improve, which is a general feature of heterogeneous catalytic hydrogenations
[35].
4.4.2 Solvent effect and comparison to CD
In the preliminary experiments several parameters were varied to find the
appropriate conditions. The solvent had a strong influence on the reaction rate
and ee (Table 4-1). The highest enantioselectivities were obtained in haloge-
nated solvents, in particular dichloromethane. In this solvent PNEA-modified
Pt/Al2O3 afforded 81% ee and 52% yield to 2a under standard conditions.
Similar ees but lower yields were achieved (in 2 h) in α,α,α-trifluorotoluene,
1,2-dichloromethane, and 1,2-dichlorobezene. Both yields and enantioselectiv-
ities were lower in other solvents and there was no correlation between solvent
polarity and ee. As illustrated in Table 4-1, PNEA was always a more effective
chiral modifier than CD, independent of the solvent. Note, however, that in
this reaction O-methyl-cinchonidine (MeOCD) is more efficient than CD and
gives 86% ee under optimized conditions [30].
The hydrogenation rate (TOF, Table 4-1) over the chirally modified cata-
lysts was either higher or lower than in the absence of modifier. The rate
depended on the solvent and the modifier (PNEA or CD). Clearly, rate acceler-
ation induced by the modifier is not a typical feature of this reaction, in con-
Table 4-1: Enantioselective hydrogenation of 1 over a 5 wt% Pt/Al2O3 catalyst in varioussolvents (2h, standard conditions, parallel reactor system).
yield eeyield eetoluene acetic acid
modifieryield ee
dichloromethane
PNEA12 0- 19 017 5723 60
13 052 81
31 16CD 28 35 20 21
TOF
10197118
TOF
11472130
TOF
15521984
[%] [%][%] [%] [%] [%][h ]-1 [h ]-1 [h ]-1
78 Chapter 4
trast to the hydrogenation of several activated ketones on cinchona-modified Pt
[14].
The influence of strong acid on the enantiodifferentiation was investigated
using small amounts of trifluoroacetic acid (TFA). No selectivity enhancement
could be achieved with TFA, in contrast to earlier studies with other α,α,α-tri-fluoromethyl ketones [27].
When discussing the solvent effect, we have to consider that the solvent has
an influence also on the keto-enol equilibrium as depicted in Figure 4-2. In
weakly polar solvents, such as toluene and dichloromethane, the β-diketonemainly exists in its enol form [30]. For ethyl 4,4,4-trifluoroacetoacetate it has
been proposed that the C=O bond of the keto form is hydrogenated on Pt and
not the C=C bond of the enol form [36]. For the β-diketone 1 the situation is
more complex because both the carbonyl group of the keto form and that of
the enol-2 form may be the reactive species. It is also not clear yet how the
keto-enol equilibrium is shifted by adsorption on Pt.
4.4.3 Influence of catalyst pretreatments
It was recognized early on that a reductive catalyst preconditioning at elevated
temperature enhanced considerably the enantioselectivity of cinchona-modi-
fied Pt/Al2O3 [37]. A similar but smaller effect was observed in our case. A
Fig. 4-2: Keto-enol equilibrium and the fraction of the keto form of 1 in different solvents,according to literature data [30].
keto enol-1 enol-2
slow fast
keto form (%)solvent
dichloromethanetoluene 2
7
An efficient synthetic modifier 79
comparison of entries 2 and 4 in Table 4-2 reveals that the heat pretreatment of
Pt/Al2O3 at 400°C in flowing H2 improved the ee by 18%. The enhanced
enantioselectivity is probably due to changes in the Pt particle size and mor-
phology, or to removal of surface impurities [38]. The importance of the latter
is supported by the strikingly higher rate of conversion of 1 (TOF). Similarly,
the yield to 2a was four-fold higher after heat treatment though the metal dis-
persion decreased from 0.32 to 0.20.
Sonochemical pretreatment of Pt/Al2O3 can also lead to restructuring (and
cleaning) of the metal particles and to higher enantioselectivity [39,40]. The
influence of sonication frequency on the reaction rate and enantioselectivity is
plotted in Figure 4-3. The positive effect of ultrasonication on the ee and yield
reached optima at 25-35 kHz. Besides, there was an optimum in sonication
time at 50 min (not shown). The best ee was achieved when ultrasonication
was applied to the catalyst previously reduced at 400°C (Table 2, entry 6). It
was important to carry out the sonication under hydrogen and in the presence
of modifier; otherwise even the rather small positive effect on ee (4-6%) was
lost. Ultrasonication of the catalyst was a far less effective pretreatment than
reduction at elevated temperature but – luckily – the two effects were additive.
Table 4-2: Enantioselective hydrogenation of 1 over the Pt/Al2O3-PNEA catalyst system:influence of catalyst pretreatment (2 h, 10°C, standard conditions, magnetically stirred auto-clave).
prereduction
Not determind.a
at 400°Cultrasonication
at r.t.modifier
a
yield ee[%] [%]
- 9 012 63-
12 93++ 10 87
15 67-+ 51 81
addedcatal. amount
(mg)solvent amount
(ml)
55
1010
55
4242
2121
4242
-+
++
++
--
+-
+-
12
65
34
entryTOF[h ]-1
3833
-42
-219
a
80 Chapter 4
Small variations in the catalyst and solvent amounts afforded 93% ee
though the yield (in 2 h) decreased (Table 4-2, entries 5 and 6). Figure 4-4
shows the time dependence of the yield and enantioselectivity under the best
conditions for achieving high ee. The monotonic decrease of ee after about 4 h
Fig. 4-3: Influence of the ultrasonic frequency during catalyst pretreatment on the yield to2a and enantioselectivity; reaction conditions according to Table 4-2, entry 6.
Fig. 4-4: Variation of ee and the yield to 2a with the reaction time in the hydrogenationof 1 over PNEA-modified Pt/Al2O3; reaction conditions according to Table 4-2, entry 6.
ee [%]
25 35 45
frequency [kHz]
100
80
60
40
20
0
yield [%]
25
20
15
10
5
0
30
silent
ee [%] yield [%]
reaction time [min]
100
80
60
40
20
0
100
80
60
40
20
04000 800 1200
An efficient synthetic modifier 81
reaction time is attributed to the competitive hydrogenation of the naphthyl
ring of the modifier and to the resulting weaker adsorption on Pt. NMR analy-
sis proved that after 2 h the degradation of the modifier was negligible except in
acidic medium [32]. For comparison, saturation of the quinoline ring of CD
during enantioselective hydrogenation and the resulting loss of ee due to
weaker adsorption of the partially hydrogenated alkaloid on Pt and Pd have
also been demonstrated [41-43].
4.5 Conclusions
The new chiral modifier for Pt, (R,R)-pantoyl-naphthylethylamine (PNEA)
provides 93% ee in the hydrogenation of 1,1,1-trifluoro-2,4-pentanedione.
This is the first case that a synthetic modifier surpasses the performance of a
naturally occurring compound and gives >90% ee. For comparison, the best
enantioselectivities achieved with CD and its simple derivative O-methyl-cin-
chonidine are 35 and 86% ee, respectively. The results confirm that proper tun-
ing of the structure of chiral modifiers plays a key role in improving the
efficiency and extending the application range of chirally modified metals in
asymmetric catalysis.
4.6 References
[1] E. N. Jacobsen, A. Pfaltz and H. Yamamoto, Comprehensive Asymmetric
Catalysis, Springer, Berlin, 1999, Vols. 1-3.
[2] R. Noyori, Adv. Synth. Catal. 2003, 345, 15.
[3] W. S. Knowles, Adv. Synth. Catal. 2003, 345, 3.
[4] H. Becker, K. B. Sharpless, in: Asymmetric Oxidation Reactions: A Practical
Approach in Chemistry, ed. T. Katsui, Oxford University Press, New York, 2001.
[5] N. Künzle, R. Hess, T. Mallat, A. Baiker, J. Catal. 1999, 186, 239.
82 Chapter 4
[6] E. Toukoniitty, P. Mäki-Arvela, A. K. Neyestanaki, T. Salmi, R. Sjöholm, R.
Leino, E. Laine, P. J. Kooyman, T. Ollonquist, J. Väyrynen, Appl. Catal. A-
Gen. 2001, 216, 73.
[7] Y. J. Zhao, F. Gao, L. Chen, M. Garland, J. Catal. 2004, 221, 274.
[8] D. Y. Murzin, P. Maki-Arvela, T. Salmi, Kinet. Catal. 2003, 44, 323.
[9] A. Tungler, G. Fogassy, J. Mol. Catal. A: Chem. 2001, 173, 231.
[10] T. Mallat, A. Baiker, in: Fine chemicals through heterogeneous catalysis, eds.
R. A. Sheldon and H. van Bekkum, Wiley-VCH, Weinheim, 2001.
[11] T. Osawa, T. Harada, O. Takayasu, Top. Catal. 2000, 13, 155.
[12] T. Sugimura, Catal. Surv. Jpn. 1999, 3, 37.
[13] A. Baiker, J. Mol. Catal. A: Chem. 1997, 115, 473.
[14] M. von Arx, T. Mallat, A. Baiker, Topics Catal. 2002, 19, 75.
[15] C. Exner, A. Pfaltz, M. Studer, H.-U. Blaser, Adv. Synth. Catal. 2003, 345,
1253.
[16] A. Solladier-Cavallo, C. Marsol, F. Garin, Tetrahedron Lett. 2002, 43,
4733.
[17] É. Sipos, A. Tungler, I. Bitter, J. Mol. Catal. A: Chem. 2003, 198, 167.
[18] G. Szöllösi, C. Somlai, P. T. S. Szabó, M. Bartók, J. Mol. Catal. A: Chem.
2001, 170, 165.
[19] A. Marinas, T. Mallat, A. Baiker, J. Catal. 2004, 221, 666.
[20] B. Minder, T. Mallat, A. Baiker, G. Wang, T. Heinz, A. Pfalz, J. Catal.
1995, 154, 371.
[21] B. Minder, M. Schurch, T. Mallat, A. Baiker, T. Heinz, A. Pfalz, J. Catal.
1996, 160, 261.
[22] M. Hudlicky, Chemistry of Organic Fluorine Compounds, Ellis Horwood
Ltd., Chichester, 1992.
[23] R. Filler, Y. Kobayashi, L. M. Yagupolskii, Organofluorine Compounds in
Medicinal Chemistry and Biomedical Applications, Elsevier, Amsterdam, 1993.
[24] Y. Kuroki, D. Asada, K. Iseki, Tetrahedron Lett. 2000, 41, 9853.
[25] Y. Kuroki, Y. Sakamaki, K. Iseki, Org. Lett. 2001, 3, 457.
[26] M. von Arx, T. Mallat, A. Baiker, Tetrahedron: Asymmetry 2001, 12, 3089.
[27] M. von Arx, T. Mallat, A. Baiker, Catal. Lett. 2002, 78, 267.
[28] T. Varga, K. Felföldi, P. Forgó, M. Bartók, J. Mol. Catal. A: Chem. 2004,
216, 181.
An efficient synthetic modifier 83
[29] K. Felföldi, T. Varga, P. Forgó, M. Bartók, Catal. Lett. 2004, 97, 65.
[30] R. Hess, S. Diezi, T. Mallat, A. Baiker, Tetrahedron: Asymmetry 2004, 15,
251.
[31] K. Balàzsik, B. Török, K. Felföldi, M. Bartók, Ultrasonics Sonochem. 1999,
5, 149.
[32] E. Orglmeister, T. Mallat, A. Baiker, Adv. Syn. Catal. 2005, 347, 78.
[33] R. Hess, F. Krumeich, T. Mallat, A. Baiker, Catal. Lett. 2004, 92, 141.
[34] A. Forni, I. Moretti, F. Prati, G. Torre, Tetrahedron 1994, 50, 11995.
[35] S. Nishimura, Handbook of Heterogeneous Catalytic Hydrogenation for
Organic Synthesis, Wiley, New York, 2001.
[36] M. von Arx, T. Mallat, A. Baiker, Angew. Chem. Int. Ed. 2001, 40, 2302.
[37] Y. Orito, S. Imai, S. Niwa, J. Chem. Soc. Jpn. 1980, 670.
[38] T. Mallat, S. Frauchiger, P. J. Kooyman, M. Schürch, A. Baiker, Catal.
Lett. 1999, 63, 121.
[39] B. Török, K. Balàzsik, M. Török, K. Felföldi, M. Bartók, Catal. Lett.
2002, 81, 55.
[40] S. C. Mhadgut, I. Bucsi, M. Török, B. Török, Chem. Commun. 2004,
984.
[41] M. Bartók, K. Balàzsik, G. Szöllösi, T. Bartók, J. Catal. 2002, 205, 168.
[42] V. Morawsky, U. Prüsse, L. Witte, K.-D. Vorlop, Catal. Commun. 2000,
1, 15.
[43] W.-R. Huck, T. Mallat, A. Baiker, Catal. Lett. 2003, 87, 241.
Chapter5
5Chemo and enantioselective hydrogenationof fluorinated ketones on platinum modified
with (R)-1-(1-naphthyl)ethylamine derivatives
5.1 Summary
The application range of platinum modified by new synthetic chiral amines
was investigated in the enantioselective hydrogenation of α,α,α-trifluorome-
thyl ketones to the corresponding chiral alcohols. High-throughput screening
was used to study the transformation of eight different substrates over a 5 wt%
Pt/Al2O3 catalyst chirally modified by eight derivatives of (R)-1-(1-naph-
thyl)ethylamine. The chiral modifiers possessed the same anchoring moiety
(naphthalene ring) that allows strong adsorption on the Pt surface and the sur-
roundings of the basic N atom was varied systematically. All modifiers
improved the chemoselectivity to hydrogenation of the activated carbonyl
group. The yield achieved in 2 h varied in a broad range (2-100%) and the
modifiers had no clear positive or negative influence on the rate of the unmod-
ified reaction. The best modifier (R,R)-pantoyl-naphthylethylamine (H)
afforded better than 99% chemoselectivity and 90% ee (at 5% conversion in 2
h) in the hydrogenation of 1,1,1-trifluoro-5,5-dimethyl-2,4-hexanedione (1).
Critical parameters of the reactions were the nature of solvent and catalyst pre-
treatments. This new catalyst system provided far better enantioselectivities
than the commonly used cinchonidine-modified Pt.
86 Chapter 5
5.2 Introduction
Significant progress in the area of asymmetric catalysis [1] has been achieved in
the past decades with homogeneous enantioselective catalysis, as reflected by
the Nobel prizes in 2001 awarded to K. B. Sharpless for enantioselective oxida-
tion and R. Noyori and W. S. Knowles for enantioselective hydrogenation [2-
4]. Compared to the number of highly selective homogeneous catalysts, the
application range of heterogeneous enantioselective catalysts is still limited.
In heterogeneous catalysis the Ni-tartaric acid [5-8], the Pt-cinchona [9-
12], and the Pd-cinchona alkaloid systems [13-16] afforded over 90% ee in the
hydrogenation of C=O and C=C bonds. Development of new catalyst systems
is commonly based on the empirical study of the modifier structure-selectivity
relationship [17,18]. This approach has been extensively used in the hydro-
genation of pyruvate esters on Pt [19-32].
The enantioselective hydrogenation of trifluoromethyl ketones represents a
direct and viable route to chiral trifluoromethyl alcohols that have attracted
increasing attention in chemical and biochemical industry. Soluble chiral tran-
sition metal catalysts afforded excellent ees (up to 98% [33,34]), whereas the
performance of cinchonidine (CD)-modified Pt is highly substrate specific.
Highest ee could be achieved in the hydrogenation of 2,2,2-trifluoroaceto-
phenone (92%, [35]) and ethyl-4,4,4-trifluoroacetoacetate (96%, [36]) but in
the hydrogenation of several other aliphatic fluorinated ketones the enantiose-
lectivity was poor [37-39].
(R,R)-Pantoyl-naphthylethylamine (H, Figure 5-1) has been shown to be a
promissing and efficient synthetic chiral modifier for Pt/Al2O3 in the enantio-
selective hydrogenation of ketopantolactone (79% ee [18]). In the enantioselec-
tive hydrogenation of 1,1,1-trifluoro-2,4-pentanedione in Chapter 4 the Pt/
Al2O3-H catalyst system even afforded up to 93% ee [40]. For this reason we
present in this chapter a structure-selectivity study aimed at extending the
application range of Pt modified by (R)-1-(1-naphthyl)ethylamine derivatives.
Pt/Al2O3 in combination with eight different modifiers A-H was tested in the
enantioselective hydrogenation of α,α,α-trifluoromethyl ketones 1-8 as shown
in Figure 5-1. All modifiers contain the same naphthalene ring as the "anchor-
ing" group, i.e. a fragment of the molecule responsible for the strong adsorp-
(R)-1-(1-naphthyl)ethylamine modifiers 87
tion on the metal surface, and a primary or secondary amino group for
interacting with the ketone substrate. The surroundings of the N atom of (R)-
1-(1-naphthyl)ethylamine was varied by introducing alkyl, cycloalkyl, hydroxy-
alkyl, and hydroxybenzyl groups or by an ester group in α-position to the N.
The fluorinated β-diketones 1 and 2 have already been hydrogenated on Pt
modified by cinchonidine and O-methyl-cinchonidine (MeOCD) and the ees
varied in the range 36-86% [41].
Fig. 5-1: Enantioselective hydrogenation of trifluoromethyl ketones 1-8 over Pt/Al2O3
chirally modified by naphthylethylamine derivatives A-H.
OH
O
OEt
HO
5 wt% Pt/Al 2O3 H2 ,solvent
X =
1 2 3 4
5 6 7 8
HN
Y
Y = H CH3
A B C D E
F G H
O
O
O
O
88 Chapter 5
5.3 Experimental
5.3.1 Materials
1,1,1-Trifluoro-5,5-dimethyl-2,4-hexanedione 1 (Acros, Lancaster), 1,1,1-trif-
luoro-2,4-pentanedione 2 (Acros), 2-(trifluoroacetyl)pyrrole 3 (Aldrich), 4,4,4-
trifluoro-1-(2-furyl)-1,3-butanedione 6 (ABCR, Acros), and 3-phenyl-1,1,1-
trifluoropropane-2-on 7 (Aldrich) were used as received. 2-(Trifluoroacetyl)-N-
methyl-pyrrole 4 [42], 2-trifluoroacetyl-5-methoxy-furan 5 [42], 1,1,1-triflu-
oro-3-pyridine-2-ylacetone 8 [43], and all modifiers [18,26] except A (Acros)
and B (Aldrich) were prepared according to published methods. Toluene (J. T.
Baker) was dried and stored over activated molecular sieve and THF (Fluka)
was dried over potassium before use.
The 5 wt% Pt/Al2O3 (E4759) catalyst was purchased from Engelhard.
After reductive heat treatment the metal dispersion decreased from 0.32 to
0.20, as calculated from the average particle size determined by TEM [44].
5.3.2 Catalytic hydrogenations
The high-throughput screening experiments were carried out in a parallel pres-
sure reactor system Endeavor (Argonaut Technologies), with eight mechani-
cally stirred stainless steel reactors equipped with glass liners, or in a
magnetically stirred stainless steel autoclave controlled by a computerized con-
stant-volume constant-pressure equipment (Büchi BPC 9901). According to
the standard procedure, the 5 wt% Pt/Al2O3 catalyst was prereduced before use
in a fixed-bed reactor by flushing with N2 at 400°C for 30 min, followed by
reductive treatment in flowing H2 for 60 min at the same temperature. After
cooling to room temperature in H2 (30 min), the catalyst was used directly for
hydrogenation or it was first sonicated before the hydrogenation reaction.
Under standard conditions 42 mg catalyst, 1.84 mmol substrate, 6.8 µmol
modifier, and 5 ml solvent were stirred (1000 rpm) at 10 bar and room tem-
perature (23-25°C). The reaction time was fixed to 2 h in all experiments to
obtain some qualitative data on the reaction rate.
(R)-1-(1-naphthyl)ethylamine modifiers 89
A multi-ultrasonic bath (Elma TI-H-5) was used for catalyst presonication
at 20°C. The 50 mL glass liner of the autoclave was equipped with a gas inlet
and a rubber septum to enable the sonication under hydrogen. The slurry con-
taining the solvent, catalyst, and modifier was sonicated for the required time
(optimally 50 min); the substrate was injected to the reaction mixture only
after sonication. Ultrasonic pretreatment was used for final optimization.
Conversion and enantioselectivity were determined by gas chromatography
using a Chirasil DEX CB capillary column (Chrompack). Some experiments
have been repeated several times and the reproducibility of the yields and ees
was always better than ±1%. Products were identified by GC/MS (HP 5973
mass spectrometer) and by 1H- and 13C-NMR spectra (Bruker DPX 500 spec-
trometer). The enantiomers of 1, 2, 5 and 6 were identified by comparing the
sign of their specific rotation (Perkin Elmer 241 Polarimeter) with literature
data [45].
5.4 Results and discussion
5.4.1 Chemoselectivity
Hydrogenation of the fluorinated ketones 1-8 in toluene or acetic acid over
unmodified Pt/Al2O3 was moderately selective. The chemoselectivity was
diminished by reduction of the second (non-activated) keto group or by satura-
tion of the (hetero)aromatic ring (Figure 5-2). An exception was the pyrrole
derivative 3; in this reaction quantitative transformation to the corresponding
alcohol was achieved in toluene within 2 h. In general, the chemoselectivity
decreased in acetic acid because the rate of the side reactions increased and the
rate of the target reaction decreased (with the exception of 6).
Interestingly, the β-diketone 1 possessing a bulky tert-butyl group was
more reactive than the analogous methyl derivative 2. A comparison of the
hydrogenation of 3 and 4 reveals that N-methylation decreased the rate of
ketone reduction, probably due to steric effects, and diminished the chemose-
lectivity in acidic medium. In toluene, substitution of the phenyl ring in 7 by a
90 Chapter 5
2-pyridyl ring in 8 resulted in a drop in the yield of the trifluoromethyl alcohol
from 99% to only 5%. The likely explanation is that 8 is mainly present in the
enol form due to stabilization via an internal hydrogen bridge (Figure 5-3).
Fig. 5-2: Products and yields (%) in the hydrogenation of 1-8 on Pt/Al2O3 in the absenceof chiral modifier in toluene under standard conditions. Values in acetic acid are presentedin brackets.
1
2
3
4
5
6
7
8
(R)-1-(1-naphthyl)ethylamine modifiers 91
This stable H-bond and the resulting rigid structure is expected to strongly
influence the adsorption and reactivity of this compound.
When the hydrogenations were carried out on Pt/Al2O3 modified by the
(R)-1-(1-naphthyl)ethylamine derivatives A-H, the fraction of byproducts was
suppressed to below 2% for all substrates. No byproducts could be detected in
the hydrogenation of 1 and 2 when Pt was modified by H. Cinchonidine had a
similar effect on the chemoselectivity in the hydrogenation of some fluorinated
ketones [41]. We attribute the dramatic improvement in chemoselectivity to
the basic amine function of the modifiers, and partly to the site blocking effect,
that is, the coverage of a considerable fraction of surface Pt sites by the strongly
adsorbing modifiers. It is frequently observed in heterogeneous catalytic hydro-
genations that chemoselectivity is improved by decreasing the active site/sub-
strate ratio [46].
5.4.2 High-throughput screening
It has been shown earlier [41] that in the hydrogenation of trifluoromethyl
ketones (and also other activated ketones), the most important parameters that
control the performance of a catalyst system are the solvent and catalyst pre-
treatment methods, assuming that there is sufficient amount of chiral modifier
Fig. 5-3: Possible keto-enol structures for the trifluoromethyl ketones 7 and 8.
7enolketo
8enolketo
92 Chapter 5
in the system to cover a large fraction of Pt surface sites. Accordingly, at the first
stage of screening of the substrates and modifiers only the solvent was varied
and the other parameters were fixed at values frequently applied in Pt-catalyzed
hydrogenation of ketones. The reaction time was also fixed to 2 h to obtain
some information on the rate of the hydrogenation reactions.
The yields and enantioselectivities achieved in the weakly polar and aprotic
toluene (Table 5-1), and the strongly polar and protic acetic acid (Table 5-2)
Table 5-1: The efficiency of (R)-1-(1-naphthyl)ethylamine derivatives used as chiral modi-fiers of Pt/Al2O3 in the hydrogenation of fluorinated ketones in toluene (standard condi-tions).
substrateee (%) ee (%) ee (%) ee (%) ee (%) ee (%) ee (%) ee (%)[yield] [yield] [yield] [yield] [yield] [yield] [yield] [yield]
[100] [99] [100] [100] [100] [100] [100] [100]
[-] [-] [-] [-] [-] [-] [9] [8]
[8] [6] [12] [4] [4] [44] [9] [10]
[-] [-] [-] [-] [-] [-] [10] [12]
[89] [87] [98] [91] [85] [99] [99]
[46] [53] [35] [17] [29] [34] [30] [26]
O
OO
OOHO
OEtHO
HN
Y
H CH3
A B C D E F G H
1
2
3
4
5
6
7
8
[98]
[10] [8] [13] [2] [5] [4] [5] [5]
[20] [19] [8] [6] [15] [7] [8] [4]
(R)-1-(1-naphthyl)ethylamine modifiers 93
revealed strong structural effects that are difficult to generalize. For example, in
the hydrogenation of 1 in toluene the yields varied between 4% and 44%
depending on the modifier. Since the unmodified Pt/Al2O3 gave 19% yield in
this reaction (Figure 5-2), apparently some modifiers induced rate acceleration
while others reduced the catalyst activity considerably. Note that an unknown
fraction of the Pt surface is covered by the modifier and thus a real evaluation
of the rate acceleration or deceleration is not possible.
Table 5-2: The efficiency of (R)-1-(1-naphthyl)ethylamine derivatives used as chiral modi-fiers of Pt/Al2O3 in the hydrogenation of fluorinated ketones in acetic acid (standard condi-tions).
substrateee (%) ee (%) ee (%) ee (%) ee (%) ee (%) ee (%) ee (%)[yield] [yield] [yield] [yield] [yield] [yield] [yield] [yield]
[99] [100] [97] [98] [99] [97] [100] [100]
[12] [8] [7] [6] [4] [5] [6] [14]
[10] [12] [17] [16] [13] [14] [17]
[9] [7] [9] [8] [9] [-] [10] [10]
OHO
OEt
HO
HN
Y
H CH3
A B C D E F G H
1
3
4
5
6
7
[17]
[12] [6] [8] [5] [5] [3] [7] [1]
[21] [12] [5] [5] [12] [-] [9] [7]
[-] [-] [-] [-] [-] [-] [13] [9]
[-] [-] [-] [-] [-] [-] [12] [10]2
8
O
O
O
O
94 Chapter 5
Alkylation of (R)-1-(1-naphthyl)ethylamine (A) with increasing bulkiness
of the substituent (B and C) resulted in a significant improvement in enantio-
selectivity in the hydrogenation of 1 and 6 in toluene but this transformation
of the modifier had no or even negative effect in some other reactions in tolu-
ene and acetic acid. In most reactions the modifiers possessing a phenolic OH
(E) or an ester group (F-H) outperformed the other derivatives. The latter
group was particularly effective in acidic medium which difference is attributed
to the formation of an intramolecular H-bond between the protonated N atom
and the ester carbonyl O atom, leading to a more rigid conformation of the
modifier [47].
An analysis of the role of substrate structure reveals that the best selecti-
vities were obtained in the hydrogenation of the β-diketones 1, 2, and 6, and
the furan derivative 5. The lower ees in the hydrogenation of the pyrrole deriv-
atives 3 and 4 may be due to additional (undesired) interactions involving the
basic N atom. None of the modifiers were effective in the hydrogenation of the
"benzylic" ketones 7 and 8. The poor ees in the hydrogenation of 7 can be
explained by the methylene unit that separates the carbonyl group and the aro-
matic ring and allows additional flexibility during adsorption of the substrate.
Note the similarity to the Pt-cinchona system that affords high ee in the hydro-
genation of acetophenone and derivatives thereof but only very low ee in the
hydrogenation of 7 [35,39,48]. Interestingly, the latter reaction was very fast
and (almost) quantitative transformation was achieved with all modifiers in
both solvents. Unfortunately, those substrates that could be hydrogenated with
reasonable enantioselectivity were much less reactive.
To sum up, the best modifier in this series was the α-amino ester type
modifier H in acidic medium and the highest enantioselectivities were
obtained in the hydrogenation of 1 (63% ee) and 2 (60% ee).
5.4.3 Solvent effect
Next, the role of solvent was investigated in the hydrogenation of 1 and 5,
using only the most promising catalyst system: Pt/Al2O3 modified by H. It is
clear from Table 5-3 and Table 5-4 that there is no correlation between the sol-
vent polarity, characterized by the relative permittivity (εr) and the empirical
(R)-1-(1-naphthyl)ethylamine modifiers 95
solvent parameter (ENT) [49], and the enantioselectivity. In both hydrogena-
tion reactions the highest enantioselectivities were obtained in halogenated sol-
Table 5-3: Solvent effect in the hydrogenation of 1 using Pt/Al2O3 modified by H (standardconditions). εr relative permittivity; EN
T, empirical solvent parameter [49].
Table 5-4: Effect of solvent and trifluoroacetic acid (TFA) on the enantioselection in the hy-drogenation of 5 using Pt/Al2O3 modified by H (10 mole equivalent of TFA compared tomodifier, standard conditions).
Eε
10.378.93
2.38
6.177.58
6.02
35.94
9.93
toluene
acetic acidTHF
ethyl acetate
acetonitrile
1,2-dichlorobenzene1,2-dichloroethanedichloromethane
2-propanol
α,α,α-trifluorotoluene
19.92
-
ee [%]yield [%]
2332
10
142
11
33
13
18
6
solvent r TN
without TFA
ee [%]yield [%]
2119
8
104
5
53
10
toluene
acetic acidTHF
ethyl acetate
acetonitrile
1,2-dichlorobenzene1,2-dichloroethanedichloromethane
2-propanol
DMF
19
2
with TFA
ee [%]yield [%]
11
7
123
Eε
10.378.93
2.38
6.177.58
6.02
35.94
9.93
19.92
-
r TNsolvent
-
-
-
-
-
-
-
-
-
-
-
-
96 Chapter 5
vents, particularly in dichloromethane. The reactions were the fastest in
acetonitrile, without any significant enantioselection in this medium.
Excluding the halogenated solvents, the best ees were achieved in acetic
acid which observation is not unusual in the enantioselective hydrogenation of
α-fluorinated ketones [35,36,38,50,51]. The influence of acid on the hydroge-
nation of 1 and 5 has been further investigated by repeating some reactions in
the presence of 10 molar equivalent of trifluoroacetic acid (TFA) related to the
amount of modifier. Slightly better enantioselectivities were achieved in the
hydrogenation of 5 in toluene, THF, and dichloromethane but the yields
decreased (Table 5-4). No improvement of ee could be achieved in the hydroge-
nation of 1. In the hydrogenations of α-keto esters, α-keto lactones, and pyrro-
lidine triones the effect of improved enantioselectivities by using acetic acid as
solvent or small amounts of an acid (TFA) in toluene were attributed mainly to
protonation of the basic N atom of cinchonidine [52-55] and H [18].
In the above screening study hydrogenation of 1 in dichloromethane over
Pt/Al2O3 modified by H provided the highest ee (85%, Table 3). The perfor-
mance of H is compared to that of cinchonidine in Table 5-5. Clearly, in this
reaction H is a far more effective chiral modifier of Pt than cinchonidine,
except in the apolar solvent toluene. It is also interesting that rate acceleration
related to the unmodified reaction is not a typical feature of the hydrogenation
of 1, independent of the modifier.
Table 5-5: Enantioselective hydrogenation of 1 over a 5 wt% Pt/Al2O3 catalyst in varioussolvents (standard conditions, parallel reactor system).
yield [%] ee [%]yield [%] ee [%]toluene acetic acid
H 6 639 --
10 2719 -
substrate modifier
1
yield [%] ee [%]dichloromethane
32 8520 -
3 2CD 22 19 9 10
(R)-1-(1-naphthyl)ethylamine modifiers 97
5.4.4 Influence of catalyst pretreatments
In the final stage of this limited optimization study, we investigated the effect
of various catalyst pretreatment methods on the activity and selectivity of
H-modified Pt/Al2O3 in the hydrogenation of 1. It is well known since the
early work of Orito et al. that a reductive catalyst preconditioning at elevated
temperatures enhances considerably the enantioselectivity of cinchona-modi-
fied Pt/Al2O3 [56]. The same effect was observed in our study. A comparison
of entries 2 and 4 in Table 5-6 shows that the heat treatment of Pt/Al2O3 at
400 °C in flowing H2 improved the ee by 38%. This improvement in enantio-
selectivity can probably be ascribed to changes in the Pt particle size and mor-
phology, or to removal of surface impurities [57]. Also significant is the more
than three-fold higher reaction rate after heat treatment although the metal dis-
persion decreased from 0.32 to 0.20.
Sonochemical pretreatment of Pt/Al2O3 can also have a beneficial effect on
the enantioselectivity presumably due to restructuring (and cleaning) of the
metal particles [58,59]. Similarly to the earlier findings, the positive effect of
ultrasonication on the ee and yield reached its optima at 25-35 kHz and at a
sonication time of 50 min. Furthermore, it was important to carry out the so-
nication in a solution containing the modifier and under hydrogen. Even
under these conditions the impact of sonication was minor compared to that of
Table 5-6: Enantioselective hydrogenation of 1 over the Pt/Al2O3 modified by H: influenceof catalyst pretreatment (standard conditions, magnetically stirred autoclave).
prereductionat 400°C
ultrasonicationat r.t.
modifier yield ee[%] [%]
- 16 06 47-
5 90++ 5 88
9 49-+ 22 85
addedcatal. amount
(mg)solvent amount
(ml)
55
1010
55
4242
2121
4242
-+
++
++
--
+-
+-
12
65
34
entry
98 Chapter 5
reductive heat treatment. Fortunately, the two positive effects were additive
(Table 5-6, entry 6). Some small changes in the catalyst and solvent amount
finally afforded 90% ee, albeit on the expense of the reaction rate.
5.5 Conclusions
With a few exceptions, the known chirally modified metal catalysts are highly
substrate specific and determination of the application range of a new catalyst
system is troublesome. A further difficulty is that some critical parameters, such
as the chemical nature of the solvent or the catalyst preconditioning are so-
called "qualitative" parameters that cannot easily be optimized with the usual
techniques. High-throughput screening is a helpful approach to accelerate such
structure-performance studies.
The present work revealed that already small structural changes in the sub-
strate or the chiral modifier of Pt can strongly alter the reaction rate and enan-
tioselectivity. Though the yields and enantioselectivities were moderate in most
cases, a limited optimization of a few parameters afforded 90% ee in the hydro-
genation of 1 over Pt/Al2O3 in the presence of the new modifier (R,R)-pantoyl-
naphthylethylamine H. A further advantage is that addition of H suppresses
the side reactions on Pt and improves the chemoselectivity above 99%.
Hydrogenation of 1 is already the second highly enantioselective reaction
with the Pt/Al2O3 - H catalyst system. In the study described in Chapter 4 we
have reached 93% ee in the hydrogenation of 2 under partly similar conditions
[40]. It seems likely that in both reactions the enantioselectivities can be further
improved by applying a more extended optimization strategy. Final assessment
of the synthetic potential of the Pt/Al2O3 - H catalyst system will need further
extension of the scope of substrates.
(R)-1-(1-naphthyl)ethylamine modifiers 99
5.6 References
[1] E. N. Jacobsen, A. Pfaltz, H. Yamamoto, Comprehensive Asymmetric Cataly-
sis, Springer, Berlin, 1999, Vols. 1-3.
[2] R. Noyori, Adv. Synth. Catal. 2003, 345, 15.
[3] W. S. Knowles, Adv. Synth. Catal. 2003, 345, 3.
[4] H. Becker, K. B. Sharpless, in: Asymmetric Oxidation Reactions: A Practical
Approach in Chemistry, ed. T. Katsui, Oxford University Press, New York, 2001.
[5] T. Osawa, Y. Hayashi, A. Ozawa, T. Harada, O. Takayasu, J. Mol. Catal. A:
Chem 2001, 169, 289.
[6] T. Osawa, T. Harada, O. Takayasu, Top. Catal. 2000, 13, 155.
[7] T. Sugimura, Catal. Surv. Jpn. 1999, 3, 37.
[8] A. Tai, T. Harada, in: Asymmetrically Modified Ni Catalysts. Tailored Metal
Catalysts, ed. Y. Iwasawa, Reidel, Dordrecht, 1986, p. 265.
[9] Y. Orito, S. Imai, S. Niwa, Nippon Kagaku Kaishi 1979, 1118.
[10] M. von Arx, T. Mallat, A. Baiker, Top. Catal. 2002, 19, 75.
[11] M. Studer, H.-U. Blaser, C. Exner, Adv. Synth. Catal. 2003, 345, 45.
[12] P. B. Wells, A. G. Wilkinson, Top. Catal. 1998, 5, 39.
[13] Y. Nitta, Top. Catal. 2000, 13, 179.
[14] W.-R. Huck, T. Mallat, A. Baiker, J. Catal. 2000, 193, 1.
[15] K. Borszeky, T. Bürgi, Z. Zhaohui, T. Mallat, A. Baiker, J. Catal. 1999,
187, 160.
[16] A. Tungler, G. Fogassy, J. Mol. Catal. A: Chem 2001, 173, 231.
[17] C. Exner, A. Pfaltz, M. Studer, H.-U. Blaser, Adv. Synth. Catal. 2003, 345,
1253.
[18] E. Orglmeister, T. Mallat, A. Baiker, Adv. Synth. Catal. 2005, 347, 78.
[19] W.-R. Huck, T. Mallat, A. Baiker, Adv. Synth. Catal. 2003, 345, 255.
[20] M. Bartók, M. Sutyinszki, K. Felföldi, G. Szöllösi, Chem. Commun. 2002,
10, 1130.
[21] H.-U. Blaser, H. P. Jalett, W. Lottenbach, M. Studer, J. Am. Chem. Soc.
2000, 122, 12675.
[22] M. Bartók, K. Felföldi, G. Szöllösi, T. Bartók, Catal. Lett. 1999, 61, 1.
[23] T. Bürgi, Z. Zhou, N. Künzle, T. Mallat, A. Baiker, J. Catal. 1999, 183,
405.
100 Chapter 5
[24] H.-U. Blaser, H. P. Jalett, D. M. Monti, A. Baiker, J. T. Wehrli, Stud. Surf.
Sci. Catal. 1991, 67, 147.
[25] T. Heinz, G. Z. Wang, A. Pfaltz, B. Minder, M. Schürch, T. Mallat, A.
Baiker, J. Chem. Soc. Chem. Commun. 1995, 1421.
[26] B. Minder, M. Schürch, T. Mallat, A. Baiker, T. Heinz, A. Pfaltz, J. Catal.
1996, 160, 261.
[27] E. Toukoniitty, P. Maeki-Arvela, M. Kuzma, A. Villela, A. K. Neyestanaki,
T. Salmi, R. Sjoeholm, R. Leino, E. Laine, D. Y. Murzin, J. Catal. 2001, 204,
281.
[28] K. E. Simons, G. Wang, T. Heinz, T. Giger, T. Mallat, A. Pfaltz, A. Baiker,
Tetrahedron: Asymmetry 1995, 6, 505.
[29] A. Pfaltz, T. Heinz, Top. Catal. 1997, 4, 229.
[30] A. Solladié-Cavallo, C. Marsol, F. Garin, Tetrahedron Lett. 2002, 43,
4733.
[31] A. Solladié-Cavallo, C. Marsol, C. Suteu, F. Garin, Enantiomer 2001, 6
245.
[32] K. E. Simons, P. A. Meheux, S. P. Griffiths, I. M. Sutherland, P. Johnston,
P. B. Wells, A. F. Carley, M. K. Rajumon, M. W. Roberts, A. Ibbotson, Recl.
Trav. Chim. Pays-Bas 1994, 113, 465.
[33] Y. Kuroki, D. Asada, K. Iseki, Tetrahedron Lett. 2000, 41, 9853.
[34] Y. Kuroki, Y. Sakamaki, K. Iseki, Org. Lett. 2001, 3, 457.
[35] M. von Arx, T. Mallat, A. Baiker, Tetrahedron: Asymmetry 2001, 12, 3089.
[36] M. von Arx, T. Mallat, A. Baiker, Catal. Lett. 2002, 78, 267.
[37] T. Varga, K. Felföldi, P. Forgó, M. Bartók, J. Mol. Catal. A: Chem. 2004,
216, 181.
[38] K. Felföldi, T. Varga, P. Forgó, M. Bartók, Catal. Lett. 2004, 97, 65.
[39] K. Balàzsik, B. Török, K. Felföldi, M. Bartók, Ultrasonics Sonochem. 1999,
5, 149.
[40] S. Diezi, M. Hess, E. Orglmeister, T. Mallat, A. Baiker, Catal. Lett. 2005,
102, 121.
[41] R. Hess, S. Diezi, T. Mallat, A. Baiker, Tetrahedron: Asymmetry 2004, 15,
251.
[42] A. D. Allen, J. M. Kwong-Chip, W. C. Lin, P. Nguyen, T. T. Tidwell,
Can. J. Chem. 1990, 68, 1709.
(R)-1-(1-naphthyl)ethylamine modifiers 101
[43] M. Kawase, M. Teshima, S. Saito, S. Tani, Heterocycles 1998, 48, 2103.
[44] R. Hess, F. Krumeich, T. Mallat, A. Baiker, Catal. Lett. 2004, 92, 141.
[45] A. Forni, I. Moretti, F. Prati, G. Torre, Tetrahedron 1994, 50, 11995.
[46] S. Nishimura, Handbook of Heterogeneous Catalytic Hydrogenation for
Organic Synthesis, Wiley, New York, 2001.
[47] E. Orglmeister, T. Bürgi, T. Mallat, A. Baiker, J. Catal. 2005, 232, 137.
[48] T. Mallat, M. Bodmer, A. Baiker, Catal. Lett. 1997, 44, 95.
[49] C. Reichardt, Solvents and Solvent Effects in Organic Chemistry, Wiley,
2003.
[50] M. von Arx, T. Bürgi, T. Mallat, A. Baiker, Chem. Eur. J. 2002, 8, 1430.
[51] M. Von Arx, T. Mallat, A. Baiker, J. Catal. 2000, 193, 161.
[52] B. Minder, T. Mallat, P. Skrabal, A. Baiker, Catal. Lett. 1994, 29, 115.
[53] M. Schürch, O. Schwalm, T. Mallat, J. Weber, A. Baiker, J. Catal. 1997,
169, 275.
[54] H.-U. Blaser, H. P. Jalett, J. Wiehl, J. Mol. Catal. A: Chem 1991, 68, 215.
[55] A. Szabo, N. Künzle, T. Mallat, A. Baiker, Tetrahedron: Asymmetry 1999,
10, 61.
[56] Y. Orito, S. Imai, S. Niwa, J. Chem. Soc. Jpn. 1980, 670.
[57] T. Mallat, S. Frauchiger, P. J. Kooyman, M. Schürch, A. Baiker, Catal.
Lett. 1999, 63, 121.
[58] B. Török, K. Balàzsik, M. Török, K. Felföldi, M. Bartók, Catal. Lett.
2002, 81, 55.
[59] S. C. Mhadgut, I. Bucsi, M. Török, B. Török, Chem. Commun. 2004,
984.
Chapter6
6The origin of chemo- and enantioselectivityin the hydrogenation of diketones on platinum
6.1 Summary
In the Pt-catalyzed hydrogenation of 1,1,1-trifluoro-2,4-diketones addition of
trace amounts of cinchonidine, O-methyl-cinchonidine, or (R,R)-pan-
toyl-naphthylethylamine induces up to 93% ee and enhances the chemoselec-
tivity up to 100% in the hydrogenation of the activated carbonyl group to an
OH function. A combined catalytic, NMR and FTIR spectroscopic, and theo-
retical study revealed that the two phenomena are coupled, offering the unique
possibility for understanding the substrate-modifier-metal interactions. The
high chemo- and enantioselectivities are attributed to the formation of an ion-
pair involving the protonated amine function of the chiral modifier and the
enolate form of the substrate. DFT calculations including the simulation of the
interaction of a protonated amine with the enolate adsorbed on a Pt 31 cluster
revealed that only the C-O bond next to the CF3 group of the substrate is in
direct contact with the Pt and can be hydrogenated. The present study illus-
trates the fundamental role played by the metal surface and indicates that also
the enol form can be the reactive species in the hydrogenation of the activated
ketone on chirally modified Pt.
104 Chapter 6
6.2 Introduction
Various strategies have been developed for the creation of chiral surface struc-
tures that combine high catalytic activity and stereochemical control [1-5].
Among them imparting chirality to a catalytic active metal surface by adsorp-
tion of chiral organic molecules (chiral modifiers) seems to be one of the most
promising [6-8]. This heterogeneous catalytic approach even allows continuous
process operation [9-11]. In the hydrogenation of C=O and C=C functions
chirally modified supported metal catalysts represent a successful approach
with synthetic potential [12-20].
An emerging field is the Pt-catalyzed enantioselective hydrogenation of
α,α,α-trifluoromethyl ketones to chiral fluorinated alcohols [21-24], which
compounds have attracted great attention in the past years in agro- and phar-
maceutical chemistry [25,26]. Although the ees are frequently lower than those
achieved in homogeneous catalysis [27,28], the supported metal catalyst (usu-
ally Pt/Al2O3) and the cinchona alkaloids used as modifiers are cheap and com-
mercially available, and the technical handling is not demanding. Thoroughly
investigated substrate groups are aromatic trifluoromethyl ketones [29,30] and
aliphatic ketones possessing an additional (non-activated) carbonyl group in
β-position [31-32] as already seen in Chapter 4 and 5 [33]. At best 92% ee was
achieved in the hydrogenation of 2,2,2-trifluoroacetophenone [30] and 96% ee
in the hydrogenation of ethyl-4,4,4-trifluoroacetoacetate [34]. The best chiral
modifiers for Pt are the naturally occurring cinchonidine (CD), its O-methyl-
derivative (MeOCD), and the synthetic modifier (R,R)-pantoyl-naphthylethy-
lamine (PNEA). In Chapter 4 the latter modifier afforded better than 99%
chemoselectivity and 93% ee in the hydrogenation of 1,1,1-trifluoro-2,4-pen-
tanedione (1, Figure 6-1) to the corresponding alcohol [35].
Despite of the impressive development in this field in the past years, the
mechanistic details at the base of the reaction are still under debate. Several
models have been developed and the conflicting opinions reflect the great sci-
entific interest of this topic [23,24,36,37]. There is some agreement on the for-
mation of chiral sites on the metal surface upon adsorption of the modifier, and
also on 1:1 interactions between adsorbed modifier and substrate [38]. How-
ever, contrasting hypotheses have been formulated concerning the nature of
Enol form as reactive species 105
substrate-modifier interactions that lead to selectivity during hydrogenation of
activated ketones on Pt. Since the basic N atom of the modifier is in most cases
crucial for enantiodifferentiation (with only one known exception [39]), inter-
action models have clustered around the two main roles that can be played by
this function: i) hydrogen bond donor to the keto-carbonyl oxygen after proto-
nation [38,40], and ii) nucleophile to the keto-carbonyl carbon as free amine
[23,37,41]. Steric effects are generally considered crucial for selectivity, after
formation of one of the before mentioned interactions. It has also been pro-
posed that the ketone might interact via hydrogen bonding with the adsorbed
aromatic anchoring group [42]. In this proposal the anchoring group would
then become a second docking site, but the relevance of such interactions still
has to be proven. On the other hand, catalytic and basic organic chemistry
studies questioned recently the role of quinuclidine as a nucleophile [43,44]
whereas the 1:1 model operating via hydrogen bond seems to become an
increasingly durable concept. With this background any experimental observa-
tion that can shed some light on the nature of the 1:1 interactions between the
ketone substrate and the chiral amine modifier is of great interest for rational-
ization of the catalyst action. A major limitation of heterogeneous enantioselec-
Fig. 6-1: Hydrogenation of (fluorinated) β-diketones over Pt/Al2O3 modified by Me3N,Et3N, quinuclidine, quinoline, PNEA, CD, and O-methyl-CD.
PNEA CD
Pt/Al 2O3 H2 ,
modifier
106 Chapter 6
tive catalysis is that truly in situ spectroscopic studies [45,46] supporting the
mechanistic models are barely available due to technical difficulties.
Hydrogenation of fluorinated β-diketones in Chapter 4 and 5 revealed that
addition of the chiral amine or amino alcohol type modifier not only induced
enantioselection but also enhanced the low chemoselectivity to 99-100% in the
hydrogenation of the activated keto function [32,33,35]. Similar observations
have been reported for the hydrogenation of α,γ-diketo esters [47] and no
explanation for the excellent chemoselectivity has been found yet. We assumed
that chemo- and enantioselectivities in the hydrogenation of fluorinated β-
diketones are coupled phenomena and understanding how the modifier
enhances the chemoselectivity can lead us to the mechanism of enantioselec-
tion. The following mechanistic study based on catalytic experiments, NMR
and FTIR spectroscopic data, and high level calculations, indicates that the
acid-base type substrate-modifier interaction is strongly affected by the adsorp-
tion on the metal surface.
6.3 Experimental
6.3.1 Materials
1,1,1-Trifluoro-2,4-pentanedione 1 (Acros), acetylacetone 2 (Fluka), and
3-penten-2-one 4 (Alfa Aesar) were distilled before use. Toluene (J. T. Baker)
was dried and stored over activated molecular sieve. Trimethylamine (Me3N,
Fluka), triethylamine (Et3N, Fluka), quinoline (Fluka), quinuclidine (Aldrich),
trifluoroacetic acid (TFA, Fluka), cinchonidine (CD, Fluka) and 1,1,1,5,5,5-
hexafluoroacetylacetone 3 (Acros) were used as received. O-Methyl-cinchoni-
dine (MeOCD) [48] and (R,R)-pantoyl-naphthylethylamine (PNEA,
Figure 6-1) [49] were synthesized by known procedures. The 5 wt% Pt/Al2O3
(E4759) catalyst was purchased from Engelhard.
Enol form as reactive species 107
6.3.2 Catalytic hydrogenations
The hydrogenation reactions were carried out in a mechanically stirred parallel
pressure reactor system (Argonaut Technologies) or in a magnetically stirred
stainless steel autoclave controlled by a computerized constant-volume con-
stant-pressure equipment (Büchi BPC 9901). Optimally, the 5 wt% Pt/Al2O3
catalyst was pre-reduced before use in a fixed-bed reactor by flushing with N2
at 400 °C for 30 min, followed by reductive treatment in H2 for 60 min at the
same temperature. After cooling to room temperature in H2 (30 min), the cat-
alyst was directly used for hydrogenation. The metal dispersion was 0.32 and
0.20 before and after reductive heat treatment, respectively, as calculated from
the average particle size determined by TEM [50]. According to standard con-
ditions 42 mg catalyst, 1.84 mmol substrate, 6.8 µmol modifier, and 5 ml sol-
vent were stirred (1000 rpm) at 10 bar and room temperature (23-25 °C) for 2
h. Deviation from the standard conditions are specified in the text.
Conversion and enantiomeric excess (ee) were determined by gas chroma-
tography using a Chirasil-DEX CB column (Chrompack). The products were
identified by GC/MS (HP 5973 mass spectrometer) and by NMR. The enanti-
omers were verified by comparing the sign of their optical rotation (Perkin
Elmer 241 polarimeter) with literature data [51,52]. In the hydrogenation of 1
PNEA, CD, and MeOCD always afforded the (S)-enantiomer in excess, and in
the hydrogenation of 2 the (R)-enantiomer formed in slight excess.
6.3.3 Spectroscopy
Protonation of quinuclidine by TFA and the influence of quinuclidine on the
keto-enol ratio of 1 have been investigated by 1H– and 13C–NMR. All NMR
data were recorded on a Bruker Avance 500 with TMS as internal standard.
The concentration of 1 mimics the conditions in the general reaction proce-
dure (0.37 mol/l).
Transmission infrared spectra were acquired with an IFS66 spectrometer
(Bruker Optics) equipped with a DTGS detector. Spectra of quinuclidine/1
solutions in CH2Cl2 with increasing molar ratios were recorded using a KBr
cell with 1 mm path length. Similar experiments were carried out with 2 and 3.
For comparison CD was also used.
108 Chapter 6
6.3.4 Computational methods
All calculations of theoretical infrared spectra were performed with the Gauss-
ian 98 set of programs [53], using Density Functional Theory (DFT) with a
hybrid functional (B3LYP) and a 6-31Gdp basis set.
Adsorption studies have been performed using the Amsterdam Density
Functional (ADF) program package [54], and the surface was simulated using a
Pt 31 cluster for which the spin state was optimized (α-β=8). A frozen core
approximation was used for the description of the inner core of the atoms. The
orbitals up to 1s were kept frozen for the second row elements, while orbitals
up to 4f were kept frozen for platinum. Decreasing the Pt frozen core to 4d,
which implies the explicit calculation of 14 additional electrons per platinum
atom, has been shown to increase the adsorption energy by only about 5 kJ/
mol for the adsorption of benzene [55]. The importance of relativistic effects
has been shown for calculations involving platinum [56,57], therefore the core
was modeled using a relativistically corrected core potential created with the
DIRAC utility of the ADF program. The DIRAC calculations imply the local
Density Functional in its simple X-α approximation without any gradient cor-
rection, but the fully relativistic Hamiltonian is used, including spin-orbit cou-
pling. Furthermore the relativistic scalar approximation (mass-velocity and
Darwin corrections) was used for the Hamiltonian, with the Zero Order Regu-
lar Approximation (ZORA) formalism [58-62], where spin-orbit coupling is
included already in zero order. The first order Pauli formalism [63] was shown
to have theoretical deficiencies due to the behavior of the Pauli Hamiltonian at
the nucleus, which lead to variational collapse for increasing basis set size [64].
It was shown that the scalar relativistic correction could account for up to 70%
of the total energy in the adsorption of carbon monoxide on platinum, and
that also the calculated adsorption site was influenced by the use of relativistic
corrections [56]. The ZORA formalism requires a special basis set, to include
much steeper core-like functions, that are implemented in the code. Within
this basis set the double-ζ (DZ) basis functions were used for platinum, and
double-ζ plus polarization (DZP) basis functions for the second row elements.
The local part of the exchange and correlation functional was modeled using a
Vosko, Wilk, Nuisar parametrization of the electron gas [65]. The non-local
Enol form as reactive species 109
part of the functional was modeled using the Becke correction for the exchange
[66] and the Perdew correction for the correlation [67]. Energies were com-
pared by setting as reference the most stable structure, since only relative ener-
gies were of interest for this work, and its values have been shown to be more
reliable (less variation with the method used) than absolute ones [55]. All cal-
culations were run unrestricted. In order to partially account for the surface
reaction to the adsorbates, the central seven platinum atoms of the cluster were
set free to optimize, while the other bond distances for platinum were fixed to
the experimental value of 2.775 Angstrom for bulk metal [68]. Molden [69]
was used as graphical interface.
6.4 Results
6.4.1 Catalytic experiments
In the hydrogenation of 1 the chemoselectivity to 1b (Figure 6-2) was only
25% in acetic acid and 79% in toluene. Addition of small amounts of achiral
or chiral amines (modifiers) enhanced the chemoselectivity up to 100%; the
results in the weakly polar solvent toluene are shown in Table 6-1. Compared
to the unmodified hydrogenation, both chiral modifiers CD and PNEA
(Figure 6-1) enhanced the reaction rate even at low concentration (modifier/1
= 0.0037 mol/mol). Interestingly, CD almost doubled the yield to 1b, though
quinuclidine (the function of CD that interacts with the substrate) barely
affected the reaction and quinoline (the anchoring moiety of CD) diminished
the yield remarkably when they were applied alone in the same concentration.
The highest rate of the formation of 1b was observed in the presence of
MeOCD which derivative is a more effective modifier of Pt in this reaction
than the parent alkaloid (Figure 6-3).
Considering only the achiral amines, there is a positive correlation between
the pKa values of the amines and the rates of the target reaction. At tenfold
higher concentration all amines enhanced the hydrogenation rate relative to the
unmodified reaction. Note that tertiary amines can induce rate acceleration
110 Chapter 6
also in the Pt-catalyzed hydrogenation of ethyl pyruvate [70-72] though the
reliability of those observations has been questioned recently [73].
A more detailed analysis revealed that the effect of amines was strongly
concentration-dependent, as illustrated by some examples in Figure 6-3 and
Fig. 6-2: Reaction scheme for the hydrogenation of 1.
Table 6-1: Hydrogenation of 1 in the presence of achiral and chiral amines (standard con-ditions, toluene).
1
1a
1b
1c
modifier conv. [%]1a 1b 1c
pKamod/1
ratio [%]
not detectable (<0.2%)
yields [%]
11 11
33 1 29 3
60 1 56 3
18 1 15 213 n. d. 13
38 1 33 4
Me N19 2 15 2-
9.80
9.80
10.72
10.72
3
3
-0.37
3.7
0.37
3.73.7
0.37
-
0.37
3.7
- 0.37- 0.37
Et N3
Me NEt N3
quinuclidine
quinuclidine
quinoline
quinoline
PNEACD
11.29
11.29
4.94
4.94
28 n. d. 28 n. d.23 n. d. 23 n. d.4 n. d. 4
37 1 34 2
n. d. n. d.n. d.
n. d.
- 0.37MeOCD 66 n. d. 66 n. d.
Enol form as reactive species 111
Figure 6-4. In all cases the reaction rate characterized by the yield to 1b went
through a maximum and at sufficiently high concentration the yield dropped
below 15% achieved in the reference (unmodified) reaction. This apparent cat-
alyst poisoning is attributed to a too high coverage of the active Pt sites by the
amines that leads to a reduced number of free sites available for the target reac-
tion. Compared to CD and MeOCD, the rate acceleration was smaller with
Et3N and quinoline and the maximum of the reaction rate was shifted to lower
amine/substrate ratios. The chemoselectivity reached 100% at low chiral amine
concentrations (CD, MeOCD, and PNEA, Table 6-1), while it increased
monotonously in the presence of achiral amines up to relatively high amine/
Fig. 6-3: The influence of the concentration of CD and MeOCD on the yield and enan-tioselectivity in the hydrogenation of 1 to 1b on Pt/Al2O3; standard conditions, in toluene.
1b
1b
112 Chapter 6
substrate ratios (Figure 6-4). The modifier concentration influenced also the
enantioselectivity; the highest ees in these series were 57% for CD and 76% for
MeOCD.
To sum up, amines can accelerate the hydrogenation of the activated and
suppress the transformation of the non-activated keto group of 1 (Figure 6-2).
The achiral amines Me3N, Et3N, quinuclidine, and quinoline mimic the influ-
ence of CD, MeOCD, and PNEA though they are less efficient than the chiral
modifiers.
Fig. 6-4: The influence of the concentration of Et3N and quinoline on the yield andchemoselectivity to 1b in the hydrogenation of 1 on Pt/Al2O3; standard conditions, in tolu-ene.
1b
1b
Enol form as reactive species 113
To clarify the role of the CF3 group in the reactivity of 1, acetylacetone (2)
was also hydrogenated in the presence and absence of CD (Table 6-2). In con-
trast to the hydrogenation of 1, the reaction rate decreased by the addition of
CD irrespective of the solvent. The small but well reproducible ee reveals that
the position of the quinuclidine N of CD relative to that of 2 adsorbed on the
Pt surface is not symmetric. A possible explanation may be that not the keto
but the enol form of 2 interacts with CD in the enantio-differentiating step.
It was assumed earlier that in the enantioselective hydrogenation of the
β-keto ester ethyl-4,4,4-trifluoroacetoacetate the keto carbonyl group rather
than the C=C bond of the enol form was the reactive species on cinchona-
modified Pt [22]. This assumption is in agreement with all mechanistic models
developed for the enantioselective hydrogenation of activated ketones on Pt
though no experimental evidence could be found yet to support this hypothesis
[6]. To estimate the reactivity of the C=C and C=O bonds in the enol forms of
2, the hydrogenation of an analogous compound 3-penten-2-one (4) was
investigated under standard conditions (Table 6-3). The C=C bond was com-
pletely saturated in 2 h while the hydrogenation of the carbonyl group barely
took place. Obviously, the reactivity of the C=C bond is much higher than that
of the (deactivated) carbonyl group of 4 and thus the reactivity of one of the
enol forms of 1 (Figure 6-5) on Pt cannot be excluded.
Table 6-2: Enantioselective hydrogenation of 2 on Pt/Al2O3 modified with CD; standardconditions.
modifier conv. [%]2a
ee [%]
2b29 17 - 128 7 8 (R) 1
21 16 - 57 6 4 (R) 1
-
solvent
tolueneCD toluene- dichloromethane
CD dichloromethane
O H2O OH O OH OH
yield [%] yield [%]
114 Chapter 6
6.4.2 NMR study
β-diketones can be present in a keto- and two enol-forms (Figure 6-5), and
the keto-enol equilibrium depends on the solvent [74,75]. In weak to medium
polar solvents 1 exists mainly in its enol form (98% in toluene and 93% in
dichloromethane) [32]. The amount of enol-1 and enol-2 in solution can be
quantified in CDCl3 using the formula
pMe = (208-δMe)/33 (eq. 1)
where δMe represents the chemical shift of the non-activated carbonyl group in
the 13C–NMR and pMe indicates the amount of enol form on the side of the
methyl group (enol-2) [76]. Using this equation, we found that enol-1 was the
dominant species (59%) in CDCl3. The slight shift of the equilibrium towards
enol-1 is due to the presence of the electron withdrawing CF3 group [77].
Table 6-3: Unmodified and CD-modified hydrogenation of 4 on Pt/Al2O3 (CH2Cl2 sol-vent and standard conditions).
Fig. 6-5: Keto-enol equilibrium for 1.
modifier conv. [%]4a 4b
100 97 3100 99 1
-CD
O O OHH2
yields [%]
enol-1 enol-2keto
slow fast
Enol form as reactive species 115
Addition of an amine to 1 in solution leads to the formation of a proto-
nated amine-enolate salt [78-80] 1H–NMR investigation on the protonation
of quinuclidine by 1 and trifluoroacetic acid (TFA) is shown in Figure 6-6. At
least 4 to 6 mol equivalents of TFA were necessary for complete protonation of
quinuclidine in CDCl3. Compared to the shifts in the TFA/quinuclidine mix-
ture, a strong acidity of the enolic proton of 1 can be estimated in this weakly
polar medium.
Furthermore, the 1H–NMR spectrum of 1 in CDCl3 shows a broad signal
at 14.15 ppm, which belongs to the enolic proton. After the addition of 1 mole
equivalent of quinuclidine, the signal of the enolic proton disappeared and a
new signal appeared at 9.76 ppm (protonated nitrogen).
These NMR results confirm the strong acid-base interaction between 1
and quinuclidine. There was no signal in the NMR spectra giving evidence for
a nucleophilic interaction between the quinuclidine N and the electrophilic
carbonyl C atom and no zwitterion formation could be detected, in contrast to
former reports on the interaction of α,α,α-trifluoro-ketones and quinuclidine
[23,37].
Fig. 6-6: 1H-NMR investigation of the protonation of quinuclidine by 1 and TFA inCDCl3. The arrows mark the chemical shifts for a 1:1 mixture of 1 and quinuclidine. Theconcentration of 1 mimics the conditions in the general reaction procedure (0.37 mol/l).
116 Chapter 6
The 13C-NMR signals of the two carbonyl carbons of 1 were found at
194.76 ppm (-COCH3) and at 176.05 ppm (-COCF3). After addition of 1
mole equivalent quinuclidine the signal of the activated carbonyl group shifted
upfield to 168.84 ppm, whereas the signal of the non-activated carbonyl group
remained constant at 194.82 ppm. Further increase of the amount of quinucli-
dine to 3 equivalents had no influence on the chemical shifts, indicating a 1:1
interaction.
6.4.3 FTIR spectroscopy
The infrared spectrum of 1 in CH2Cl2 (Figure 6-7) displays bands at ca. 1645,
1601, 1206, 1157, and 1109 cm-1, together with a very weak envelope at
around 1790 cm-1. This last feature is given by the n(C=O) mode of the free
diketone species which however account only for about 7% of the species in
solution [32]. The spectrum of compound 1 is predominantly a combination
of the spectra of the two enol forms, whose calculated vibrational spectra are
reported in Figure 6-7(c) and Figure 6-7(d) scaled by 0.967. The spectra agree
well with those already available in the literature [81] The linear combination
of the calculated vibrational spectra of the two enol forms in the ratio 41:59 in
favor of enol-1, as suggested by the 1H-NMR data, matches the experimental
spectrum of 1. According to the calculated vibrational spectra, the signals at
1645 and 1601 cm-1 correspond to the asymmetric and symmetric stretching
motions of the (C=O) and (C=C) groups, whereas the set of signals between
1250 and 1100 cm-1 is assigned to the vibrational modes associated with the
CF3 and CH groups.
The infrared spectrum of 1 exhibited striking changes upon titration of a
solution of 1 with a solution of quinuclidine (Figure 6-8(a)). In the high fre-
quency region (left panel) a very broad band centered at 2640 cm-1 increased
with increasing quinuclidine concentration. The changes observed in the mid-
dle panel of Figure 6-8(a) suggest the immediate disappearance of the free dike-
tone species (1790 cm-1) and the replacement of the bands at 1645 and 1601
cm-1 by signals at 1640 and 1530 cm-1, respectively, at 1 mole equivalent of
base. In the right panel of Figure 6-8(a), the signals at 1157 and 1109 cm-1 dis-
appeared almost completely upon addition of quinuclidine, whereas new sig-
Enol form as reactive species 117
nals were found at 1175 and 1128 cm-1. The signal at 1206 cm-1 is strongly
attenuated. Two isosbestic points can be clearly observed at 1167 and 1113
cm-1 that indicate the transition from the species related to the spectrum before
addition of quinuclidine (dotted) to new species formed after addition of the
base (bold spectrum).
Fig. 6-7: Experimental and calculated vibrational spectra of 1. (a) Transmission IR spec-trum of a solution of 1 in CH2Cl2 (5 mM). Calculated vibrational spectra (scaled by 0.967)of (b) a mixture of the enol forms 59:41, (c) enol-2 and (d) enol-1 forms. (b) Vibrational spec-trum obtained by combining the calculated vibrational spectra of the two enol forms in theratio 59:41. The dotted spectrum in (a) corresponds to neat 1.
118 Chapter 6
Figure 6-8(b) shows that the interaction between CD and 1 is identical to
that between quinuclidine and 1, with only minor differences in the position of
the new signals. The additional signals at 1593, 1572, and 1510 cm-1 (quino-
line ring modes) and at 3594 cm-1 (ν(O–H)) are characteristic of the alkaloid
[82]. The presence of the mode of the free OH group of CD indicates that this
function is not involved in the interaction with 1.
The spectral changes observed upon titration of 1 by quinuclidine (and
CD) are associated with the formation of the ion pair composed of the quater-
nary ammonium ion of quinuclidine (2600 cm-1) and the enolate anion of 1
(1640, 1530, 1204, 1175, and 1128 cm-1) [79,82,83]. The two major bands at
1640 and 1530 cm-1 belong to groups displaying C=O [79] and C=C charac-
ter. Although deprotonation of 1 does not allow the localization of the two
bonds (formation of the enolate ion), the presence of molecular asymmetry in
Fig. 6-8: Transmission infrared spectra of a solution of (a) 1 with 0 (dotted), 0.1, 0.2, 0.4,1, and 2 (bold) equivalents of quinuclidine and (b) 1 with one equivalent of CD. Arrows in-dicate the direction of the changes upon base addition. The calculated vibrational spectrumof the complex 1:Me3N is shown in (c) (scaling factor 0.967). Spectra (b) and (c) are offset.C1 = CCD = 5 mM, Camine = 0.5, 1, 2, 5, and 10 mM, ambient temperature, in CH2Cl2.
Enol form as reactive species 119
1 induced by the CF3 group suggests that the charge distribution between the
O-atoms is asymmetric.
This last observation becomes particularly important for the discussion on
how the chemoselectivity is controlled upon addition of the chiral amine dur-
ing the enantioselective hydrogenation of 1. Therefore, interpretation of the
changes observed in Figure 6-8(a) (and Figure 6-11(b), see later) required the
aid of vibrational analysis to resolve the structure of the enolate ion and clarify
which C-O bond has a partial C=O character in the ion pair that information
is missing from the available literature. Theoretical calculations were carried
out enabling the relaxation of the proton located on the amine towards one of
the two O-atoms of the enolate. The structure is shown in the inset of
Figure 6-8 together with the calculated spectrum of the acid-base pair 1:Me3N,
which matches well the experimental spectrum in Figure 6-8(a). The frequency
of the sharp signal at 1951 cm-1 (calcd. 2018 cm-1) for the ν(N–H+) mode is
underestimated compared to the other vibrations probably due to the fact that
the calculations localize the proton on the N-atom. Interestingly, the vibra-
tional analysis suggests that the C–O on the side of CF3 rather than that on the
side of CH3 has a partial C=O character. The signals at 1640 and 1530 cm-1
are then assigned to ν(CCF3–O)+ν(C–CCH3) and ν(C–CCH3)+ν( CCF3–O)
modes of the enolate of 1, respectively.
Figure 6-9 shows the FTIR spectra of solutions of the diketones 1, 2, and 3
before and after addition of 1 mole equivalent of amine. The least reactive sub-
strate towards deprotonation, i.e. the least acidic, is 2, in agreement with the
effect induced by substitution of a CH3 group by a CF3 group. In the FTIR
spectra of the neat substrates, this substitution results in the splitting of the sig-
nals assigned to the asymmetric and symmetric stretching of the O=C-C=C-O
group [84] and in the complete disappearance of the free diketone species in
the case of 3 [85].
Addition of quinuclidine does not affect the spectrum of 2 but substan-
tially alters the spectra of 1 and 3. As in the case of the ion pair 3:quinuclidine,
the signals at 1666 and 1549 cm-1 in Figure 6-9(f ) are attributed to the forma-
tion of the enolate of 3. Additional signals are also found between 3000 and
2000 cm-1 (broad, ν(N–H+)) and at 1192, 1144, and 1136 cm-1. The enolate
ions of 2 and 3 exist only in a single form due to molecular symmetry. The
120 Chapter 6
enolate of 2 in aqueous solution absorbs at frequencies below 1600 cm-1
[86,87], i.e. more than 40 cm-1 below the frequency observed for the enolate of
1. The enolate of 3 exhibits the ν(C–O) at 1666 cm-1 and shows a ∆ν with
respect to the enolate of 1 of only 26 cm-1. The difference in shift suggests that
the enolate of 1 resembles more the enolate of 3 and displays some carbonyl
character on the C–O bond near the CF3 group, in agreement with the usually
observed blue-shift of the ν(C=O) band of ketones upon introduction of the
electron-withdrawing CF3 group, thus further supporting the validity of the
vibrational analysis of the ion pair.
Fig. 6-9: FTIR spectra of solutions of (a) 2, (c) 1 and (e) 3. Addition of one equivalent ofquinuclidine afforded spectra (b), (d), and (f), respectively. C = 5 mM, ambient temperature,in CH2Cl2.
Enol form as reactive species 121
6.4.4 Adsorption on Pt
Adsorption of 1 was studied using a Pt 31 cluster to model the surface in order
to better understand the catalytic behavior. Figure 6-10(a) shows the adsorp-
tion mode of 1 in the diketo-form, chemisorbed η2 via the activated keto-car-
bonyl group (i.e. the keto-carbonyl in α-position to the CF3 group). The other
possibility, the adsorption of the diketo-form chemisorbed η2 via the non-acti-
vated keto-carbonyl group is presented in Figure 6-10(b). Consistently with
previous results [88], the former was more strongly adsorbed by ca. 3 kcal/mol
than the latter. This result was rationalized by the energy lowering of the fron-
tier orbitals of the activated keto-carbonyl, that can better interact with the
filled states of the metal in back-bonding interactions [88,89]. This chemisorp-
tion mode of C=O double bonds is also called di-σ to evidence the double σbinding of the keto carbonyl moiety to the metal atoms. Also η1 adsorption
modes are theoretically possible, but it has been shown that their adsorption
energy is much smaller than the η2 modes, and in particular, in such adsorp-
tion modes the keto-carbonyl moiety is distant to the surface and cannot reach
surface hydrogen, therefore they are less interesting for our discussion [88]. On
the other hand, the η2 modes show rehybridization of the keto-carbonyl group
and have the right geometry for hydrogen uptake. Another adsorption mode of
the diketo-form is shown in Figure 6-10(c), where the diketone is adsorbed
with both carbonyl groups close to the surface. Since both carbonyls cannot
simultaneously find an optimal interaction with the surface Pt atoms, the most
strongly interacting moiety dominates the adsorption. This mode is energeti-
cally disfavored compared to the previous two structures; nevertheless we
observe that the result of this interaction is the rehybridization of the activated
keto-carbonyl only, although also the non-activated carbonyl is close to the
metal.
The adsorption of the enol forms of 1 (Figure 6-10(d) and (e)) were the
most energetically favored, with a preference for the adsorption of enol-2
(Figure 6-10(d)), where the optimal chemisorption of the activated carbonyl is
possible. It should be noted that the adsorption of enol-1 (Figure 6-10(e)) does
not lead to the expected η2 interaction of the non-activated carbonyl with the
metal, but rather with the formation of a σ-bond between the carbon neigh-
122 Chapter 6
boring the CF3 group and a Pt atom. Also in this case the competition for sur-
face sites is in favor of the activated carbonyl group, and the result is that the
non-activated carbonyl undergoes no evident structural alteration induced by
the metal. Also to be noted is that both adsorbed enol forms are stabilized by
internal hydrogen bonds. However, while in enol-2 the H-bond length is 1.35
Å, it elongates to 1.71 Å in enol-1, showing the stronger hydrogen-acceptor
character of the adsorbed activated carbonyl group.
The structures of the diketo and enol forms without surface interaction
were also calculated using the same level of theory. The two conformations of
Fig. 6-10: Calculated adsorption modes of 1 on a Pt cluster. Structures (a), (b), and (c)show the adsorption of the diketo-form, (d) and (e) the adsorption of the enol forms, and (f)the adsorption of the enolate. The energy of the most stable structure (d) has been set to zeroand taken as a reference for the others. The numbers in brackets are the energy differences(kcal/mol) between (f) and the other (less stable) structures.
Enol form as reactive species 123
the diketo-form (not shown) are about 10 and 15 kcal/mol, respectively, less
stable than the lowest energy enol-2 at this level of theory. For our purposes it is
important to note that after chemisorption the difference in energy between
the enol and keto forms of 1 is greatly enhanced with respect to the molecules
which are not interacting with the metal. Binding to the metal increases the
difference of stability by a factor two, raising it to 20 and 30 kcal/mol in favor
of the enol forms.
Figure 6-10(f ) shows the adsorption mode of the enolate, the deproto-
nated form of the enol, thus a charged species. Unlike the enols, the enolate has
a more symmetric adsorption mode where both oxygen atoms are bound to a
metal atom, and with a third σ binding site due to C(3) that is also bound to a
metal atom and rehybridized. Note that the relative energy is not given, since
this structure contains one hydrogen less than the others and its energy cannot
be simply compared. The negative charge of the enolate is delocalized on the
metal upon adsorption.
It has been discussed previously that introducing an amine in the reaction
system leads to an almost complete chemoselectivity in the saturation of the
activated keto-carbonyl group of 1. It has also been shown that in solution the
amine deprotonates the enol and generates an ammonium ion and the enolate.
Therefore, the interaction between the trimethylammonium ion and the eno-
late was modeled in presence of the Pt surface. Figure 6-11 shows three steps of
this simulation: Figure 6-11(a) shows the starting geometry whereby the (trim-
ethyl)ammonium ion is positioned close to the adsorbed enolate ion, such that
the distance between its acid hydrogen and the oxygen atoms of the enolate is
equal (2.5 Angstroms). During minimization the ammonium ion moves
towards the surface and comes closer to the non-activated CO, as shown in
Figure 6-11(b). In this intermediate position the hydrogen bond formed with
the non-activated moiety is 1.8 Å, while that formed with the activated moiety
is 2.3 Å. As the optimization continues this preferential coordination leads to a
displacement of the non-activated keto-carbonyl from the surface, and when
convergence is reached the ammonium ion coordinates the activated CO bond
and changes its adsorption mode to η2 (Figure 6-11(c)).
124 Chapter 6
6.5 Discussion
A big effort has been made in the past years to understand the functioning of
chirally modified Pt in the enantioselective hydrogenation of ketones. As dis-
cussed in the introduction, several mechanistic models have been proposed, but
only one of these models is supported by in situ experimental observations
[45]. A major difficulty is the high reactivity of activated ketones in the pres-
ence of amine-type modifiers that complicates or even hinders the study of sub-
Fig. 6-11: Simulation of the interaction between Me3NH+ and adsorbed enolate. Structure(a) is the starting geometry, in which the acidic proton is equidistant from the two oxygenatoms. Structure (b) is an intermediate point of the optimization, where the Me3NH+ is closerto the non-activated moiety, and removes it from the surface. Structure (c) is the convergedgeometry, where the Me3NH+ is coordinated to the di-σ adsorbed CO.
Enol form as reactive species 125
strate-modifier interactions in solution [90]. The present study is the first
example where the nature of the amine-ketone interaction in solution could be
refined by spectroscopic methods, and further modeled by state-of-the-art cal-
culations.
The hydrogenation experiments have shown that both carbonyl groups are
reactive on Pt that may lead to poor chemoselectivity, but only the activated
carbonyl group is reduced to an alcoholic OH function in the presence of the
chiral amine and amino alcohol type modifiers CD, MeOCD, and PNEA
(Table 6-1, Figure 6-3). The remarkable enhancement in chemoselectivity is
due to acceleration of the hydrogenation of the activated and deceleration of
the hydrogenation of the non-activated carbonyl functions by the basic amine.
It has also been proved by catalytic and spectroscopic methods that simple
achiral amines (e.g. Me3N, Et3N, quinuclidine) could improve the chemoselec-
tivity, though to a smaller extent (Figure 6-4), and the OH function of CD was
not involved in the substrate-modifier interaction (Figure 6-4 and Figure 6-5).
The better efficiency of the chiral modifiers compared to simple amines at
identical (low) solution concentrations is attributed to the stronger adsorption
of the modifiers on the Pt surface. Note that strong adsorption leading to high
surface concentration is one of the key characteristics of effective chiral modi-
fiers.
NMR and FTIR spectroscopy has revealed that, in contrast to former
reports [91,92] 1 is about as strong proton donor as TFA (Figure 6-6) and for-
mation of the enolate-ammonium ion pair, where the quaternary ammonium
ion coordinates to the enolate, is responsible for enantioselection. The struc-
ture assigned to this ion pair partially localized the proton between the N-atom
of the amine and the O-atom of the non-activated keto-carbonyl of 1. For
comparison, the small but significant ee observed in the hydrogenation of 2
(Table 6-2) indicated that the position of the N atom relative to the O atoms in
the modifier-substrate complex is non-symmetric even in the absence of the
electronic effect of fluorine atoms.
A striking result of this study is that the substrate-modifier interaction
identified in solution by spectroscopic methods cannot interpret the chemose-
lectivity observed in the hydrogenation of 1. The real origin of enhanced
126 Chapter 6
chemoselectivity induced by addition of amines could be clarified only by con-
sidering the adsorption on the metal surface.
The theoretical study of the adsorption of the different (equilibrated) spe-
cies of 1 on a model Pt cluster has shown that the enol forms are more stable
also on the surface with respect to the diketo-form and adsorption even
increased the difference in their stability (Figure 6-10). More interestingly, the
simulation of the formation of the ion pair on the metal surface has revealed
that coordination of the ammonium ion to the adsorbed enolate results in the
displacement of the non-activated CO from the metal surface and at the same
time the promotion of an η2 adsorption mode of the activated CO. In the
equilibrium geometry shown in Figure 6-11(c) the ammonium ion coordinates
to the activated CO moiety adsorbed η2. It has already been proposed that η2
adsorption modes of ketones are more activated towards hydrogenation, and
the results shown in Figure 6-11 seem to be in very good agreement with this
proposal [88].
The apparent mismatch between the spectroscopic results obtained in solu-
tion and the adsorption study correctly interpreting the observed changes in
chemoselectivity may be explained as follows. As mentioned, the ion pair
formed with the ammonium ion and the enolate in solution is characterized by
closer interaction between the acidic proton and the non-activated CO. This
interaction is also present when the ion pair is interacting on the metal surface,
but only in the first phase of the interaction (Figure 6-11b). Incontrast, the
activated CO which is di-σ adsorbed dominates the interaction with the acidic
proton at equilibrium, consistently with the electron transfer to the metal mak-
ing this group a better hydrogen bond acceptor. Comparison of these results to
the catalytic experiments strongly suggest that the ammonium ion promotes
the hydrogenation of the activated CO by selective coordination, with a funda-
mental contribution from the metal surface enhancing its proton affinity that
connot be neglected. On the other hand, the calculated geometry of the enolate
explains why hydrogenation of the non-activated carbonyl lying well above the
metal surface is hindered in the presence of an amine. This observation high-
lights the influence of the metal surface on the substrate-modifier interaction.
The unambiguous relation between chemoselectivity and metal–substrate–
modifier interaction is very valuable to assess the debated model of the interac-
Enol form as reactive species 127
tion between cinchona alkaloids and activated ketones during enantioselective
hydrogenation on cinchona-modified platinum. In fact, a similar N-H-O type
interaction is expected between the protonated quinuclidine moiety of the
alkaloid and an adsorbed activated ketone substrate. This can occur also in a
non-acidic medium (such as toluene). It has been shown recently that the alka-
loid can be protonated by uptake of activated hydrogen from the Pt surface
[44,45].
The hydrogen bonding interaction has already been interpreted as the trig-
ger for rate acceleration [93,94] which is observed also in the experiments dis-
cussed here. The enantioselectivity in the presence of the adsorbed modifier is
to be interpreted as the result of different interaction energies that the prochiral
faces of the ketone substrate show within a chiral environment generated by the
alkaloid on the surface. As already proposed [93,94] enantiodiscrimination
should arise from two combined effects, both related to the strength of the
hydrogen bond interaction: on the one hand, the best fitting prochiral face has
the lowest energy and therefore a higher fractional coverage; on the other hand,
the stronger hydrogen bond interaction should show the largest rate enhance-
ment, generating kinetic resolution. A quantification of the two effects is only
possible via modeling of the interaction between the alkaloid and the substrate
both adsorbed on the metal, which is the subject of ongoing investigation in
our laboratory. The better performances obtained using cinchona alkaloids as
carriers for the amino-group show the unique feature of these molecules that
provide the reactive site (basic nitrogen) in proximity of the metal, still allow-
ing the presence of free metal sites. The structural features of the cinchona alka-
loids [95] allow reshaping of the surface upon adsorption and generation of
new surface sites without annihilation of the availability of the metal for the
contact with the substrate, unlike quinoline (the ‘anchoring group’ of the alka-
loid) which is a catalyst poison when used alone (Table 6-1, Figure 6-4). There-
fore, the cinchona alkaloids are properly called surface chiral modifiers, whereas
a simple alkyl amine is a reaction additive.
128 Chapter 6
6.6 Conclusions
Catalytic experiments combined with spectroscopic and computational struc-
tural analysis have been applied to understand the coupled phenomena: the
enantioselection and the enhancement of chemoselectivity in the hydrogena-
tion of the activated keto-carbonyl group of 1. The study revealed that the
structures of ammonium ion – enolate type ion pairs formed between the mod-
ifier and 1 are different in solution and on the Pt surface. The outstanding
chemoselectivity is attributed to the selective coordination of the protonated
amine (modifier) to the adsorbed activated keto-carbonyl group and preven-
tion of the interaction of the non-activated carbonyl group in this position
with the metal surface. The fundamental differences in the structures of the
substrate-modifier ion-pair in solution and on the Pt surface underline that the
reaction mechanism cannot be understood without considering the adsorption
on the metal surface.
The presented model shows important similarities to those suggested for
the enantioselective hydrogenation of other activated ketones on Pt modified
by chiral amines [14,36,49]. In all cases an N-H-O type hydrogen bond
between the amine modifier and the keto-carbonyl group of the adsorbed sub-
strate is critical for rationalizing the outcome of the reaction, concerning both
the rate enhancement and the enantioselectivity.
6.7 References
[1] A. Baiker, Curr. Opin. Solid State Mater. Sci. 1998, 3, 86.
[2] T. Bein, Curr. Opin. Solid State Mater. Sci. 1999, 4, 85.
[3] D. C. Sherrington, Catal. Today 2000, 57, 87.
[4] M. E. Davis, Topics Catal. 2003, 25, 3.
[5] P. McMorn, G. J. Hutchings, Chem. Soc. Rev. 2004, 33, 108.
(6] T. Bürgi, A. Baiker, Acc. Chem. Res. 2004, 37, 909.
[7] D. Y. Murzin, P. Mäki-Arvela, E. Toukoniitty, T. Salmi, Catal. Rev. - Sci.
Eng. 2005, 47, 175.
Enol form as reactive species 129
[8] A. Baiker, Catal. Today 2005, 100, 159.
[9] N. Künzle, R. Hess, T. Mallat, A. Baiker, J. Catal. 1999, 186, 239.
[10] A. Baiker, N. Künzle, T. Mallat, Hoffmann-La Roche AG: Switzerland,
2000.
[11] E. Toukoniitty, P. Mäki-Arvela, A. Kalantar Neyestanaki, T. Salmi, D. Y.
Murzin, Appl. Catal. A: General 2002, 235, 125.
[12] P. B. Wells, A. G. Wilkinson, Topics Catal. 1998, 5, 39.
[13] T. Sugimura, Catal. Surv. Jpn. 1999, 3, 37.
[14] A. Baiker, J. Mol. Catal. A: Chem. 2000, 163, 205.
[15] T. Osawa, T. Harada, O. Takayashu, Topics Catal. 2000, 13, 155.
[16] B. Török, K. Balázsik, K. Felföldi, M. Bartók, Ultrason. Sonochem. 2001,
8, 191.
[17] M. von Arx, T. Mallat, A. Baiker, Topics Catal. 2002, 19, 75.
[18] C. J. Baddeley, Topics Catal. 2003, 25, 17.
[19] M. Studer, H.-U. Blaser, C. Exner, Adv. Synth. Catal. 2003, 345, 45.
[20] E. Sipos, A. Tungler, G. J. Fogassy, Mol. Catal. A: Chem. 2004, 216, 171.
[21] K. Balázsik, B. Török, K. Felföldi, M. Bartók, Ultrasonics Sonochem. 1999,
5, 149.
[22] M. von Arx, T. Mallat, A. Baiker, Angew. Chem. Int. Ed. 2001, 40, 2302.
[23] T. Varga, K. Felföldi, P. Forgo, M. Bartok, J. Mol. Catal. A: Chem. 2004,
216, 181.
[24] K. Felföldi, T. Varga, P. Forgo, M. Bartók, Catal. Lett. 2004, 97, 65.
[25] M. Hudlicky, Chemistry of Organic Fluorine Compounds; Ellis Horwood
Ltd.: Chichester, 1992.
[26] R. Filler, Y. Kobayashi, L. M. Yagupolskii, Organofluorine Compounds in
Medicinal Chemistry and Biomedical Applications, Elsevier: Amsterdam, 1993.
[27] Y. Kuroki, D. Asada, K. Iseki, Tetrahedron Lett. 2000, 41, 9853.
[28] Y. Kuroki, Y. Sakamaki, K. Iseki, Org. Lett. 2001, 3, 457.
[29] T. Mallat, M. Bodmer, A. Baiker, Catal. Lett. 1997, 44, 95.
[30] M. von Arx, T. Mallat, A. Baiker, Tetrahedron: Asymmetry 2001, 12, 3089.
[31] M. von Arx, T. Mallat, A. Baiker, J. Catal. 2000, 193, 161.
[32] R. Hess, S. Diezi, T. Mallat, A. Baiker, Tetrahedron: Asymmetry 2004, 15,
251.
130 Chapter 6
[33] S. Diezi, M. Hess, E. Orglmeister, T. Mallat, A. Baiker, J. Mol. Catal. A:
Chem. 2005, 239, 49.
[34] M. von Arx, T. Mallat, A. Baiker, Catal. Lett. 2002, 78, 267.
[35] S. Diezi, M. Hess, E. Orglmeister, T. Mallat, A. Baiker, Catal. Lett. 2005,
102, 121.
[36] M. Bodmer, T. Mallat, A. Baiker, in: Catalysis of Organic Reactions, ed. F.
E. Herkes, M. Dekker: New York, 1998, pp 75.
[37] G. Vayner, K. N. Houk, Y. K. Sun, J. Am. Chem. Soc. 2004, 126, 199.
[38] O. Schwalm, B. Minder, J. Weber, A. Baiker, Catal. Lett. 1994, 23, 271.
[39] E. Orglmeister, T. Mallat, A. Baiker, J. Catal. 2005, 233, 333.
[40] H.-U. Blaser, H. P. Jalett, D. M. Monti, A. Baiker, J. T. Wehrli, Stud. Surf.
Sci. Catal. 1991, 67, 147.
[41] R. L. Augustine, S. K. Tanielyan, L. K. Doyle, Tetrahedron: Asymmetry
1993, 4, 1803.
[42] S. Lavoie, M. A. Laliberté, P. H. McBreen, J. Am. Chem. Soc. 2003, 125,
15756.
[43] E. Orglmeister, T. Mallat, A. Baiker, J. Catal. 2005, 234, 242.
[44] A. Vargas, D. Ferri, A. Baiker, J. Catal. 2005, 236, 1.
[45] N. Bonalumi, T. Bürgi, A. Baiker, J. Am. Chem. Soc. 2003, 125, 13342.
[46] M. S. Schneider, A. Urakawa, J.-D. Grunwaldt, T. Bürgi, A. Baiker, Chem.
Comm. 2004, 744.
[47] M. Studer, S. Burkhardt, A. F. Indolese, H.-U. Blaser, Chem. Commun.
2000, 1327.
[48] K. Borszeky, T. Bürgi, Z. Zhaoui, T. Mallat, A. Baiker, J. Catal. 1999,
187, 160.
[49] E. Orglmeister, T. Mallat, A. Baiker, Adv. Synth. Catal. 2005, 347, 78.
[50] R. Hess, F. Krumeich, T. Mallat, A. Baiker, Catal. Lett. 2004, 92, 141.
[51] A. Forni, I. Moretti, F. Prati, G. Torre, Tetrahedron 1994, 50, 11995.
[52] P. Besse, S. Ciblat, J. L. Canet, Y. Troin, H. Veschambre, Tetrahedron:
Asymmetry 2000, 11, 2045.
[53] M. J. Frisch, et al. GAUSSIAN98, A.7 ed., Gaussian Inc.: Pittsburgh, PA,
1998.
Enol form as reactive species 131
[54] E. J. Baerends, et al. ADF - Amsterdam Density Functional, Release 2004-
01 ed.; Scientific Computing and Modelling NV - Vrije Universiteit, Theoreti-
cal Chemistry: Amsterdam.
[55] M. Saeys, M.-F. Reyniers, G. B. Marin, M. J. Neurock, J. Phys. Chem. B
2002, 106, 7489.
[56] P. H. T. Philipsen, E. vanLenthe, J. G. Snijders, E. Baerends, J. Phys. Rev.
B 1997, 56, 13556.
[57] G. Pacchioni, S. C. Chung, S. Kruger, N. Rosch, Surf. Sci. 1997, 392,
173.
[58] E. van Lenthe, E. J. Baerends, J. G. Snijders, J. Chem. Phys. 1993, 99,
4597.
[59] E. van Lenthe, E. J. Baerends, J. G. Snijders, J. Chem. Phys. 1994, 101,
9783.
[60] E. van Lenthe, E. J. Baerends, J. G. Snijders, J. Chem. Phys. 1996, 105,
6505.
[61] E. van Lenthe, R. van Leeuwen, E. J. Baerends, J. G. Snijders, Int. J.
Quantum Chem. 1996, 57, 281.
[62] E. van Lenthe, E. J. Baerends, J. G. Snijders, J. Chem. Phys. 1999, 110,
8943.
[63] J. G. Snijders, E. J. Baerends, P. Ros, Mol. Phys. 1979, 38, 1909.
[64] G. teVelde, F. M. Bickelhaupt, E. J. Baerends, C. F. Guerra, S. J. A. van
Gisbergen, J. G. Snijders, T. J. Ziegler, Comput. Chem. 2001, 22, 931.
[65] S. H. Vosko, L. Wilk, M. Nusair, Can. J. Phys. 1980, 58, 1200.
[66] A. D. Becke, Phys. Rev. A 1988, 38, 3098.
[67] J. P. Perdew, Phys. Rev. B 1986, 33, 8822.
[68] G. A. Somorjai, Introduction to surface chemistry and catalysis, Wiley: New
York, 1994.
[69] G. Schaftenaar, J. H. Noordik, J. Comput. - Aided Mol. Design 2000, 14,
123.
[70] J. L. Margitfalvi, M. Hegedüs, J. Catal. 1995, 156, 175.
[71] H.-U. Blaser, H. P. Jalett, D. M. Monti, J. F. Reber, J. T. Wehrli, Stud.
Surf. Sci. Catal. 1988, 41, 153.
[72] G. Bond, P. A. Meheux, A. Ibbotson, P. B. Wells, Catal. Today 1991, 10,
371.
132 Chapter 6
[73] E. Toukoniitty, D. Y. Murzin, Catal. Lett. 2004, 93, 171.
[74] J. Emsley, Structure and Bonding, Comlex Chemistry: the Composition,
Structure and Hydrogen bonding of the β-Diketones, Springer Verlag: Berlin,
1984.
[75] M. T. Rogers, J. L. Burdett, Can. J. Chem. 1965, 43, 1516.
[76] A. I. Koltsov, J. Mol. Struct. 1998, 444, 1.
[77] J. Niwa, M. Yamazaki, T. Takeuchi, Chem. Lett. 1975, 4, 707.
[78] A. I. Koltsov, N. S. Golubev, I. S. Milevskaya, Zh. Obshch. Khim. 1980,
50, 2504.
[79] K. I. Pashkevich, V. I. Filyakova, Bull. Acad. Sci. USSR 1984, 33, 575.
[80] V. I. Filyakova, V. G. Ratner, N. S. Karpenko, K. I. Pashkevich, Russian
Chem. Bull. 1996, 45, 2163.
[81] Y. Minoura, N. Nagashima, S. Kudoh, M. Nakata, J. Phys. Chem. A 2004,
108, 2353.
[82] D. Ferri, T. Bürgi, A. Baiker, J. Chem. Soc., Perkin Trans. 2 1999, 1305.
[83] L. J. Bellamy, The Infrared Spectra of Complex Molecules. Advances in Infra-
red Group Frequencies, Chapmann and Hall: London, 1980, Vol. 2.
[84] H. Ogoshi, K. Nakamoto, J. Chem. Phys. 1966, 45, 3113.
[85] S. L. Wallen, C. R. Yonker, C. L. Phelps, C. M. Wai, J. Chem. Soc., Fara-
day Trans. 1997, 93, 2391.
[86] E. E. Ernstbrunner, J. Chem. Soc. A 1970, 1558.
[87] S. A. Dickie, A. J. McQuillan, Langmuir 2004, 20, 11630.
[88] A. Vargas, T. Bürgi, A. Baiker, J. Catal. 2004, 222, 439.
[89] B. Hammer, J. K. Nørskov, Adv. Catal. 2000, 45, 71.
[90] D. Ferri, T. Bürgi, K. Borszeky, T. Mallat, A. Baiker, J. Catal. 2000, 193,
139.
[91] K. I. Pashkevich, V. I. Saloutin, T. Y. Postovskii, Russian Chem. Rev. 1981,
50, 180.
[92] J. P. Shukla, R. S. Sharma, Electrochim. Acta 1986, 31, 1449.
[93] A. Vargas, T. Bürgi, A. Baiker, New J. Chem. 2002, 26, 807.
[94] A. Vargas, T. Bürgi, M. von Arx, R. Hess, A. Baiker, J. Catal. 2002, 209,
489.
[95] A. Baiker, J. Mol. Catal. A: Chem. 1997, 115, 473.
Chapter7
7Steric and electronic effects in the enantio-selective hydrogenation of activated ketoneson platinum: directing effect of ester group
7.1 Summary
Steric effects in the Pt-catalyzed asymmetric hydrogenation of nine different
α-keto esters have been studied by variation of the bulkiness at the keto and
ester side of the substrates, and by using cinchonidine (CD), its 6’-methoxy
derivative quinine, and O-phenyl derivative PhOCD as chiral modifiers. In the
presence of CD always the (R)-enantiomer formed in good to high ee (up to
96%), independent of the steric bulkiness of the α-keto ester. None of the
mechanistic models developed for ketone hydrogenation on Pt are conform to
the observations. Only additional steric effects in the modifiers and replace-
ment of toluene by acetic acid as reaction medium enhanced the sensitivity of
the catalyst system to steric effects in the substrates (ee = 0 - 94%).
An important mechanistic consequence of the observations is that on CD-
modified Pt preferred adsorption of the α-keto ester on the si-side is directed
by the position of the ester group relative to the modifier, independent of the
steric bulkiness on any side of the keto-carbonyl group. Ester, carboxyl, amido,
carbonyl, acetal, and trifluoromethyl functions have similar directing effects,
but when both trifluoromethyl and an ester or carbonyl groups are present in
the molecule, the latter function is dominant. The directing effect of the elec-
tron withdrawing (activating) function on the adsorption of the ketone is obvi-
ously related to the electronic environment provided by the chiral modifier.
The critical role of electronic interactions is supported by the remarkable influ-
ence of aryl substituents in the hydrogenation of ethyl benzoylformates.
136 Chapter 7
7.2 Introduction
Optically active α-hydroxy esters and α-hydroxy acids are important building
blocks for the synthesis of biologically active natural products and analogues
thereof [1]. A viable option for their synthesis is the heterogeneous enantiose-
lective hydrogenation of α-keto esters on cinchona-modified Pt. Since the first
description of the route by Orito et al. in 1979 [2,3], several research groups
have been fascinated by the potential of this catalyst system (for recent reviews
see refs. [4-13]). Hydrogenation of ethyl pyruvate (1, Figure 7-1) is the most
studied heterogeneous enantioselective hydrogenation reaction and it serves
now as a standard model reaction for chirally modified Pt. After years of opti-
mization, 97-98% ee has been achieved in the synthesis of α-hydroxy esters by
using CD [14-16] or QN [17] as chiral modifiers of supported and colloidal
Pt.
The only systematic study on the structural effects in the hydrogenation of
α-keto esters revealed that an increase in the size of the alkyl group (from
methyl to t-butyl) in the ester group of pyruvate barely decreased the ee [18].
Introduction of a t-butyl group in α-position to the keto group reduced the ee
to 81%, probably due to hindered adsorption of the keto group. The electronic
effects of substituents on the performance of chirally modified Pt have been
studied only in two series of acetophenone and trifluoroacetophenone deriva-
tives [19-21].
Several mechanistic models have been developed and refined in the past
years to rationalize the stereochemical outcome of the hydrogenation of α-keto
esters (for a recent overview see ref. [22]). Most models postulate a 1:1 type
interaction between substrate and modifier, involving an N-H-O type hydro-
gen bond interaction [23-27] or a nucleophylic attack of the basic N atom of
the modifier on the keto C atom [28-33]. A common limitation of these mod-
els is that only transformation of the simplest substrates, usually methyl or
ethyl pyruvate, are described and steric effects relevant in the hydrogenation of
bulkier molecules have not been considered.
The aim of the study presented in this chapter is to analyze the steric and
electronic effects on the Pt-catalyzed enantioselective hydrogenation of α-keto
esters. Nine different substrates were hydrogenated in the presence of CD, its
Directing effect of ester group 137
6’-methoxy derivative QN, and O-phenyl derivative PhOCD (Figure 7-1).
That is, not only the bulkiness of α-keto esters at the keto and the ester side
was varied, but also additional steric effects were introduced in the modifier. As
seen in Chapter 2 and 3 for the modifier PhOCD, additional functions of CD
influence the adsorption mode and strength of the modifier, [17,31,34-36],
and thus the shape of the chiral pocket available for adsorption of the α-keto
ester while interacting with the alkaloid during hydrogen uptake.
7.3 Experimental
7.3.1 Materials
Ethyl pyruvate 1 (Fluka) and ethyl benzoylformate 3 (Aldrich) were carefully
distilled in vacuum before use. t-Butyl benzoylformate 4 was synthesized via
α-oxobenzeneacetyl chloride by treating the acid chloride with t-butanol in
basic medium [37,38]. α-Oxobenzeneacetyl chloride was foregoing synthesized
from benzoylformic acid by chlorination with dichloromethyl methyl ether
Fig. 7-1: Enantioselective hydrogenation of α-keto esters 1-9 over Pt/Al2O3 modifiedby CD, QN, and PhOCD.
*5 wt% Pt/Al2O3
H2, modif ier
modifier
138 Chapter 7
[37,38]. Ethyl t-butylglyoxylate 2, ethyl 3,5-dimethylbenzoylformate 5, ethyl
3,5-difluorobenzoylformate 6, ethyl 3,5-di(trifluoromethyl)benzoylformate 7,
ethyl 3,5-dimethoxybenzoylformate 8, and ethyl naphthylglyoxylate 9 were
prepared by reaction of the corresponding Grignard reagents with diethyl
oxalate according to a known method [39]. All synthesized substrates were
flash chromatographed and their purity (>99%) was confirmed by GC, HPLC,
and NMR analysis. Acetic acid (AcOH, 99.8%, Fluka) was used as received
and toluene (99.5%, J. T. Baker) was dried and stored over activated molecular
sieve. Cinchonidine (CD, 92%, Fluka; impurities: 1% quinine, 7% quinidine,
determined by HPLC at Fluka) and quinine (QN, 99%, Fluka) were used
without further purification. O-Phenyl-cinchonidine (PhOCD) was synthe-
sized as described earlier in Chapter 2.3.2 [34]. The 5 wt% Pt/Al2O3 (E4759)
catalyst was purchased from Engelhard.
7.3.2 Catalytic hydrogenation
The hydrogenation reactions were carried out in a mechanically stirred eight
parallel pressure reactor system (Argonaut Technologies) or in a magnetically
stirred stainless steel autoclave controlled by a computerized constant-volume
constant-pressure equipment (Büchi BPC 9901). Optimally, the 5 wt% Pt/
Al2O3 catalyst was pre-reduced before use in a fixed-bed reactor by flushing
with N2 at 400 °C for 30 min, followed by reductive treatment in H2 for
60 min at the same temperature. After cooling to room temperature in H2
(30 min), the catalyst was used directly for hydrogenation. Under standard
conditions 42 mg catalyst, 1.84 mmol substrate, 6.8 µmol modifier, and 5 ml
solvent were stirred (1000 rpm) at 10 bar and room temperature (23-25 °C) for
2 h. In the hydrogenation of 1-9 over Pt/Al2O3 modified by CD and QN
(almost) always the (R)-α-hydroxy ester was produced in excess, whereas the
(S)-α-hydroxy ester was the major enantiomer in the presence of PhOCD.
7.3.3 Analyses
Conversion and ee were determined by a Thermo Finigan Trace gas chromato-
graph using a Chirasil-DEX CB (25 m × 0.25 mm × 0.25 µm) capillary co-
Directing effect of ester group 139
lumn for the hydrogenation products of 1, 2, 4, 6 and 7, or by HPLC using a
MERCK LaChrom system with a CHIRACEL OD (4.6 mm internal diame-
ter, 240 mm length, 10 µm paricle size) chiral column for the hydrogenation
products of 3, 5, 8, and 9. The HPLC analysis was carried out at 10 °C with a
liquid flow rate of 0.5 ml/min. The UV detector was set at 210 nm. For all sub-
strates a mixture of n-hexane / isopropanol 90% / 10% was used as eluent.
Products were identified by GC/MS (HP-6890 coupled with a HP-5973 mass
spectrometer) and by 1H– and 13C–NMR. All NMR data were recorded on a
Bruker Avance 500 with TMS as internal standard. The enantiomers of 1 was
verified by GC analysis of the commercially available products and those of 3
[2] and 4 [40] by comparing the sign of their optical rotation (Jasco DIP-1000
polarimeter) with literature data. In the hydrogenation of 2 and 5-9 the prod-
ucts were identified by assuming the analogous chromatographic separation of
the products. The conversion values could be reproduced within +/- 1% by
repeating the experiments. The reproducibility of ee was at best +/- 0.5% in
case of GC analysis, but the error increased with HPLC analysis (to +/-1%)
and particulary at ee values below 5% ee, as expected.
7.3.4 VCD spectroscopy
Vibrational Circular Dichroism (VCD) spectra were measured using a Bruker
PMA 37 accessory coupled to a VECTOR 33 Fourier transform infrared spec-
trometer. A photoelastic modulator (Hinds PEM 90) set at λ/4 retardation was
used to modulate from right to left circular polarized light at a frequency of 50
kHz. The demodulation was performed by a lock-in amplifier (Stanford
Research System SR830 DSP). The sensitivity of the lock-in amplifier was set
at 1 mV for the samples with a dynamic reserve of 40 dB. In order to enhance
the signal-to-noise ratio an optical low-pass filter (<1800 cm-1) was set before
the photoelastic modulator. The samples were analyzed in a transmission cell
equipped with KBr windows and a 0.2 mm Teflon spacer was used. A single
beam spectrum of the neat solvent served as the reference for the absorption
spectrum and a spectrum of the neat solvent recorded in VCD mode was sub-
tracted from the VCD spectrum of the dissolved molecules.
140 Chapter 7
Calculations were performed using the Gaussian 98 program package [41].
All internal coordinates of the molecules were optimized at the density func-
tional B3LYP level of theory, and using a 6-31G (dp) basis set. For the mini-
mized structures a normal coordinate analysis was performed. Rotational and
dipole strengths associated with the normal modes were calculated in order to
simulate adsorption and VCD spectra [42].
7.4 Results and discussion
Hydrogenation of 1-3 (Table 7-1) on CD-modified Pt has already been
reported in literature [2,3,18,43] and partly in Chapter 3 [35]. In this study
slightly modified reaction conditions and a broader range of modifiers were
applied and these substrates are used for comparison.
Although a detailed kinetic analysis was out of the scope of this study, a
comparison of the conversions achieved in toluene in 2 h in the high-through-
put screening enables a qualitative assessment of the reactivity of α-keto esters
1-9 in the presence and absence of the modifiers CD, QN, and PhOCD
(Table 7-1). Hydrogenation of 2, 4, and 9 indicates that rate acceleration due
to the presence of a cinchona alkaloid (CD or QN) is not a general feature of
α-keto esters hydrogenation, in agreement with recent observations also in
pyruvate hydrogenation [43-45]. Note, however, that a reliable kinetic analysis
of α-keto ester hydrogenation is complicated by numerous side reactions cata-
lyzed by the amine-type modifier and the Pt surface [8,46-49].
7.4.1 VCD spectroscopy
In the hydrogenation of 2 and 5-9 the products were identified by assuming
the analogous chromatographic separation of the products. This assumption
was confirmed by VCD spectroscopy for the most critical substrates 2, 7, and 8
as the biggest effect of the functional groups on the adsorption on the chiral
chromatographic column was expected in these cases. The absolute configura-
tion of the major enantiomers was determined by comparing the calculated
Directing effect of ester group 141
VCD spectrum of one enantiomer with the experimental VCD spectrum of
the product of the hydrogenation reaction on CD-modified Pt/Al2O3. As a
representative example, Figure 7-2a shows the theoretical VCD spectrum and
Figure 7-2b the experimental VCD spectrum of the product for substrate 8.
The matching of the spectra allows assigning the absolute configuration of the
main product of hydrogenation as R.
7.4.2 Steric effects of substrates and modifiers
Variations in the rate and ee in the hydrogenation of 1-9 (Table 7-1) may be
attributed to steric and electronic effects of R1 and R2 (Figure 7-1). In some
cases the steric effects are dominant. For example, the lower reactivity of the
α-keto ester due to bulkiness of the ester group is shown by the replacement of
the ethyl group (3) by a t-butyl group (4). Interestingly, the bulky ester group
had a negative effect only on the reactivity of 4; the enantioselectivities were the
highest in this series with all three modifiers. In contrast, replacement of the
Fig. 7-2: (a) Theoretical VCD spectra; (b) experimental VCD spectra of the mixture ofenantiomers obtained by the hydrogenation of 8 in the presence of CD (0,2 M in methylenechloride-d2).
142 Chapter 7
methyl group at the keto carbonyl function in 1 by a t-butyl group in 2
decreased both the rate and the enantioselectivity (except with PhOCD).
Another example on the steric effects is the lower rate and ee in the hydrogena-
tion of 9 compared to 3, due to replacement of the phenyl group by a naphtha-
lene ring next to the keto-carbonyl group.
The data in Table 7-1 demonstrate that even stronger steric effects can be
induced by variations in the modifier structure. Replacement of CD by its
Table 7-1: Enantioselective hydrogenation of 1-9 in toluene (stand. cond.). (as an example:entry 1, R1 –CH3, R
2 –CH2CH3)
substrate
ee (%) ee (%) ee (%)[conv.] [conv.] [conv.]
[100] [100] [100]
CD QN PhOCD
[91] [61] [98]
[100] [100] [100]
[94] [43] [52]
[100] [97] [100]
[100] [99] [99]
[94] [90] [94]
[95] [100] [100]
[100] [32] [31]
[conv.]
[100]
no
[100]
[100]
[61]
[100]
[93]
[38]
[100]
[48]
Directing effect of ester group 143
6'-methoxy derivative QN caused a general negative effect on the reaction rate
and ee. The probable explanation is that the aromatic methoxy function occu-
pies a part of the chiral pocket available for adsorption of the substrate. This
observation is in line with the mechanistic models assuming that in the enan-
tio-differentiating complex the quinoline ring of the alkaloid, being in the so-
called “open-3” conformation [50], is "flipped" towards the α-keto ester
adsorbed in the neighborhood of the modifier [22,51]. Replacement of the
OH function of CD by a phenoxy group in PhOCD inverted the major enan-
tiomer, in agreement with earlier measurements on the hydrogenation of some
other activated ketones in Chapter 2 and 3 [35-37]. The inversion is probably
due to the steric bulkiness of the phenoxy group relative to that of OH func-
tion, and to a change in the adsorption geometry of the alkaloid, resulting in a
shift in the position of the interacting function, the quinuclidine N atom.
A general observation is that additional steric effects in the chiral modifiers
increase the "sensitivity" of the catalyst system to steric effects in the substrates,
as indicated by the greater differences in enantioselectivities when CD was
replaced by QN or PhOCD in the hydrogenation of 1-9.
7.4.3 Electronic effects of aryl substituents
The α-keto esters 3 and 5-8 possess the same ester group but differ at the aryl
substituents on the keto side. The electron withdrawing and releasing proper-
ties of the functional groups at the phenyl ring, characterized by the Hammett
σ-constants [52], correlate reasonably well with the enantioselectivities
achieved in toluene. For a better visualization of this relationship, the impor-
tant data in Table 7-1 are plotted in Figure 7-3. In all reactions the conversion
was higher than 90%, hence the ees represent the final, integral values. The
advantage of this modification is that by aryl substitution the electron density
at the carbonyl group can be tuned without changing significantly the environ-
ment in the close neighborhood of the carbonyl group. Besides, in the weakly
polar toluene the distorting effect of solvent-substrate interactions is expected
to be minor.
For all three modifiers the highest ee was achieved in the hydrogenation of
8, which substrate contains the strongest electron releasing substituents, two
144 Chapter 7
methoxy groups (Figure 7-1). On the other hand, the ee was the lowest in the
hydrogenation of 7, which substrate contains the strongest electron withdraw-
ing substituents, two CF3 groups. The enantioselectivity is between the two
extremes in the hydrogenation of ethyl benzoylformate 3 that substrate has an
unsubstituted aromatic ring. Considering all five α-keto esters, the correlation
between the electron releasing or withdrawing effects of the substituents and
the ee is the best for CD-modified Pt and deviations become more important
with increasing steric effects in the modifiers.
7.4.4 The influence of acidic medium
In the commonly used model reaction, the hydrogenation of ethyl pyruvate
(1), the enantioselectivities in toluene and acetic acid are similar [53]. An
extension of the study to structurally more demanding substrates reveals
remarkable differences between weakly polar and acidic medium (Table 7-1
Fig. 7-3: The influence of electron withdrawing and releasing aryl substituents of ethylbenzoylformate (3) on the enantiomeric excess (standard conditions, in toluene). The(R)-enantiomer is formed with CD and QN, and the (S)-enantiomer with PhOCD.
Directing effect of ester group 145
and Table 7-2). Since hydrogenolysis of the C-O-C (ether) bond of PhOCD is
catalyzed by acids [53], only CD and QN could be used in acidic medium.
The hydrogenation rates in toluene and acetic acid are rather similar. The
biggest drop in conversion was measured for the hydrogenation of 9 on CD-
modified Pt when changing to acidic medium. This behavior is traced to the
strong, competing adsorption of acetic acid.
In acetic acid the highest ee of 94% was achieved with CD-modified Pt in
the hydrogenation of 8. Introduction of the 6'-methoxy group in QN mostly
Table 7-2: Enantioselective hydrogenation of 1-9 in acetic acid (stand. cond.).
[100] [100]
[99] [99]
[100] [100]
[73] [64]
[100] [100]
[100] [100]
[98] [98]
[94] [91]
ee (%) ee (%)[conv.] [conv.]
CD QN
[31] [7]
substrate
146 Chapter 7
diminished the ee in this medium, with the important of the hydrogenation of
ethyl pyruvate (1) in which reaction up to 98% ee was obtained [17].
Interestingly, in the hydrogenation of 2 a small ee to the opposite enanti-
omer was observed. Considering the estimated error of the analytical method at
low ee, it is better to evaluate the result as a loss of enantioselection. Still, the
effect of replacement of the methyl group at the keto side in 1 by a t-butyl
group in 2 is dramatic. Hydrogenation of 3 and 9 was also non-selective.
Clearly, the negligible difference between acidic and non-acidic solvents in
the hydrogenation of pyruvates is an exception, rather than the rule.
7.4.5 The role of reaction conditions
The influence of some reaction parameters was investigated further for the two
most selective reactions, the hydrogenation of 4 and 8 in toluene, in the pres-
ence of CD (Table 7-3). Working at higher pressure (i.e. at higher surface
hydrogen concentration) had no influence (4) or even diminished the ee (8). A
similar negative effect of high surface hydrogen concentration was found in the
Pt catalyzed enantioselective hydrogenation of some other ketones, including
acetophenone [54], 2,2,2-trifluoroacetophenone [19], ring-substituted ace-
tophenones [21], and fluorinated β-diketones [55]. This correlation is oppo-
site to the typical behavior of the Pt-cinchona system in the hydrogenation of
α-keto esters [56].
Reaction temperatures below room temperature increased the ee at the
expense of conversion. Changes in the amounts of catalyst and solvent did not
significantly improve the enantioselectivity. Application of a mixture of toluene
and acetic acid (1:1), which medium was favorable in the hydrogenation of 3
over CD modified Pt [43], diminished the ee in the hydrogenation of 4 and 8.
Ultrasonication of the catalyst slurry in the presence of CD had no positive
effect either. Finally, the best ees in the hydrogenation of 4 and 8 were 96% and
95%, respectively; only marginally better than those achieved in the prelimi-
nary screening.
Directing effect of ester group 147
7.4.6 Mechanistic considerations
Hydrogenation of α-ketoesters 1-9 on Pt/Al2O3 modified by CD demonstrates
that always the (R)-enantiomer forms in good to excellent ee, independent of
the steric bulkiness of the ester group and the size and electronic structure of
the alkyl and functionalized aryl group on the other side of the keto function
(Table 7-1). The lowest ee of 56% in the hydrogenation of 2 is probably due to
sterically hindered adsorption and hydrogenation of the keto-carbonyl group
by the neighboring t-butyl group as suggested earlier [18]. Note also that a lim-
ited optimization of the reaction conditions resulted in an improvement to
81% ee in the same reaction [18].
The steric effects in the substrates are more pronounced when additional
steric effects in the alkaloid are introduced: the methoxy group in QN and the
phenoxy group in PhOCD (Table 7-1). We assume that both modifications
change the “chiral pocket” available for adsorption of the ketone while interact-
ing with the modifier during hydrogen uptake and thus modify the efficiency
of Pt. It is expected that these results will be valuable in refining the mechanis-
tic concepts; nevertheless, the following discussion will be limited to the com-
monly used and most understood catalyst, CD-modified Pt.
Table 7-3: Influence of temperature, pressure and acid additive on the hydrogenation of 4and 8 over CD-modified Pt/Al2O3 (stand. cond.).
pressure temperature[bar]
solvent conv. ee[°C] (ml) [%] [(R), %]
25 2525 1025 0
toluene (5)toluene (5) 99toluene (5) 40toluene (5) 25
959695
substrate
25 25toluene (5)toluene (5) 95toluene (5)
91
toluene (5)
25
25
148 Chapter 7
The mechanistic models developed mainly for pyruvate hydrogenation
assume two interactions between the amine type modifier and the ketone: an
N-H-O or N-C type attractive interaction (as discussed in the introduction),
and a second attractive [26] or repulsive [22] interaction to direct the adsorp-
tion of the ketone on Pt. (Note that some models describe only the nature of
the attractive interaction and in fact do not explain the origin of enantioselec-
tion.) A careful supervision of the data in Table 7-1 shows that none of the
models can account for the observed enantioselectivities. For example, in case
of ethyl pyruvate (1) the ester group is the bulky and electron-rich substituent.
The situation is different for 8 or 9: the ester group is less bulky than the aro-
matic function but still the (R)-enantiomer forms with excellent ee. Obviously,
the model operating with a second, repulsive interaction [22] based on the dif-
ferent bulkiness on the two sides of the keto group needs refinement. Similarly,
the proposed second attractive interaction [26]) involving a H-bond between
the small methyl group of ethyl pyruvate (R1 in 1) and the quinoline ring of
CD cannot be valid with substrates 3-9.
It seems to be more likely that the ester group serves not only as an activat-
ing group for carbonyl hydrogenation but also as a directing group which is
mainly responsible for favoring the adsorption on one side of the substrate. In
this respect it is interesting to compare the dominant adsorption mode of vari-
ous activated ketones on CD-modified Pt. The drawings in Figure 7-4 are
based on the stereochemical outcome of the hydrogenation reactions. The
model for the CD-α-keto ester interaction (Figure 7-4a) was published a few
years ago [57] and the feasibility of the uptake of a proton by CD from the Pt
surface in aprotic medium has been shown recently [58]. Assuming the H-
uptake from the Pt surface (from “below”), a similar adsorption on the si-side is
expected for α-keto acids [59], α-keto amides [60], ketopantolactone [61],
pyrrolidine-2,3-5-triones [62], α-keto acetals [63,64], and α-diketones [65-
67].
Besides α-keto esters, α,α,α-trifluoromethyl ketones are the most studied
substrates on cinchona-modified Pt. A probable model for their hydrogenation,
the bifurcated H-bond interaction of the quinuclidine N atom of CD with the
O and one F atom of the ketones is depicted in Figure 7-4b. (Note that a bifur-
cated H-bond involving additionally one F atom is also feasible.) In the hydro-
Directing effect of ester group 149
genation of simple alkyl and aryl trifluoromethyl ketones CD favors the
formation of the (R)-alcohol in weakly polar reaction medium where solvent
effects are less important [68-72].
Fig. 7-4: Schematic illustration of the adsorption of α-keto esters (a), trifluoromethyl ke-tones (b), and a trifluoromethyl β-keto ester (c), and a trifluoromethyl β-diketone (d) on Ptduring interaction with CD. These adsorption modes lead to the formation of the majorenantiomers, assuming hydrogen-uptake from the metal surface.
F
FF
FF
F
F
FF
OHH
150 Chapter 7
An interesting case is the hydrogenation of a trifluoromethyl β-keto esters
(Figure 7-4c) where the (S)-alcohol is the major enantiomer on CD-modified
Pt, indicating adsorption on the re-side of the activated carbonyl [73]. The trif-
luoromethyl group in α-position activates the carbonyl, whereas the ester
group in β-position seems to be responsible for adsorption on the re-side of the
substrate. It is likely that the activated ketone and the ester interact with the
modifier via a bifurcated H-bond, similarly to the model for α-keto esters
(Figure 7-4a). Apparently, the ester group is a stronger directing function than
the trifluoromethyl group and inverts the adsorption mode of the substrate rel-
ative to that of simple α,α,α-trifluoromethyl ketones. A similar case is shown
in Figure 7-4d, the hydrogenation of a trifluoromethyl-β-diketone that affords
the (S)-enantiomer in excess in weakly polar solvents [55] though in the
absence of the directing effect of the second carbonyl group the (R)-trifluorom-
ethyl alcohol is expected as the major enantiomer.
Analysis of the hydrogenation of aryl-substituted α-keto esters 3 and 5 - 8
(Figure 7-3) on CD-modified Pt provides further information on the directing
effect of the ester group. In these reactions electron withdrawing substituents
(F, CF3) decreased the ee whereas electron releasing substituents (CH3, OCH3)
had a minor positive effect. Since the nature of the substrate-modifier interac-
tion is expected to be the same for all substrates, the observed differences in ee
are (mainly) attributed to variations in the electron density at the carbonyl
group. In α-keto esters the carbonyl group is already activated by the ester
group. The results in Figure 7-3 show that additional activation on the other
side of the keto group via the phenyl ring (6, 7) is disadvantageous to the enan-
tioselectivity. It seems that activation should come only from one side of the
carbonyl group, indicating the critical role of the electronic effect in enantio-
differentiation.
The influence of aryl substituents was the opposite in the hydrogenation of
acetophenone derivatives [21]. Electron withdrawing substituents on the phe-
nyl ring (CF3, F, ester group) increased the hydrogenation rate and the ee (from
17 to 60%), while the electron releasing methoxy function diminished both
the rate and the ee. The inverse effect of aryl substituents is probably connected
with the absence of activating function on the other side of the keto group of
acetophenones. In other words, in the hydrogenation of ketones on cinchona-
Directing effect of ester group 151
modified Pt the highest enantioselectivity may be expected for substrates where
the electronic effect of functional groups points in the same direction.
7.5 Conclusions
The unexpected steric and electronic effects observed in the hydrogenation of
α-keto esters over Pt/Al2O3 modified by CD, QN, and PhOCD cannot be
explained by the existing mechanistic models. We propose that the position of
the ester group determines the direction and extent of enantioselection, and the
steric bulkiness on any side of the keto-group and additional electronic effects
of substituents play only a secondary role. Qualitatively, the directing effects of
other activating functions including carboxyl, amido, carbonyl, acetal, and trif-
luoromethyl groups are the same as that of the ester function.
Hydrogenation of α-keto esters 1-9 on Pt/Al2O3 modified by CD, QN,
and PhOCD has uncovered that development of a concept for the hydrogena-
tion of ethyl or methyl pyruvate is not sufficient to understand the complex
steric and electronic effects in ketone hydrogenation exerted by structural vari-
ations in the substrate and the modifier. We hope that the present results will
provide a useful database for testing the feasibility of future mechanistic mo-
dels developed for the hydrogenation of α-keto esters and other related
ketones.
Some years ago our group suggested - in contrast to the general opinion at
that time - that the application range of chirally modified Pt covers all those
ketones that are activated by an electron withdrawing group in α-position to
the keto-carbonyl group [68]. The present study may bring us closer to under-
stand the critical role of the electron withdrawing group and thus the potential
and limitations of Pt modified by chiral amines and amino alcohols.
152 Chapter 7
7.6 References
[1] G. M. Coppola, H. F. Schuster, Chiral alpha-Hydroxyacids in Enantioselec-
tive Synthesis, Wiley-VCH, Weinheim, 2000, p. 1.
[2] Y. Orito, S. Imai, S. Niwa, N. G. Hung, J. Synth. Org. Chem. 1979, 37,
173.
[3] Y. Orito, S. Imai, S. Niwa, J. Chem. Soc. Japan 1979, 1118.
[4] G. V. Smith, F. Notheisz, Heterogeneous Catalysis in Organic Chemistry, Aca-
demic Press, San Diego, 1999.
[5] P. B. Wells, R. P. K. Wells, in: Chiral Catalyst Immobilization and Recycling,
eds. D. E. De Vos, I. F. J. Vankelecom, P. A. Jacobs , Wiley-VCH, Weinheim,
2000, p. 123.
[6] B. Török, K. Balàzsik, K. Felföldi, M. Bartók, Ultrasonics Sonochem. 2001,
8, 191.
[7] G. J. Hutchings, Catal. Lett. 2001, 75, 1.
[8] M. von Arx, T. Mallat, A. Baiker, Topics Catal. 2002, 19, 75.
[9] K. A. Avery, R. Mann, M. Norton, D. J. Willock, Top. Catal. 2003, 25, 89.
[10] M. Studer, H.-U. Blaser, C. Exner, Adv. Synth. Catal. 2003, 345, 45.
[11] A. Tungler, E. Sipos, V. Hadac, Arkivoc 2004, 223.
[12] D. Y. Murzin, P. Maki-Arvela, E. Toukoniitty, T. Salmi, Catal. Rev.-Sci.
Eng. 2005, 47, 175.
[13] A. Baiker, Catal. Today 2005, 100, 159.
[14] X. Zuo, H. Liu, M. Liu, Tetrahedron Lett. 1998, 39, 1941.
[15] B. Török, K. Balàzsik, M. Török, G. Szöllösi, M. Bartók, Ultrason.
Sonochem. 2000, 7, 151.
[16] M. Sutyinszki, K. Szöri, K. Felföldi, M. Bartók, Catal. Commun. 2002, 3,
125.
[17] W.-R. Huck, T. Mallat, A. Baiker, Adv. Synth. Catal. 2003, 345, 255.
[18] K. Szöri, M. Sutyinszki, K. Felföldi, M. Bartók, Appl. Catal. A-Gen.
2002, 237, 275.
[19] M. von Arx, T. Mallat, A. Baiker, Tetrahedron: Asym. 2001, 12, 3089.
[20] A. Vargas, T. Bürgi, M. von Arx, R. Hess, A. Baiker, J. Catal 2002, 209,
489.
[21] R. Hess, T. Mallat, A. Baiker, J. Catal. 2003, 218, 453.
Directing effect of ester group 153
[22] T. Bürgi, A. Baiker, Acc. Chem. Res. 2004, 37, 909.
[23] O. Schwalm, J. Weber, J. Margitfalvi, A. Baiker, J. Mol. Structr. 1993,
297, 285.
[24] K. E. Simons, P. A. Meheux, S. P. Griffiths, I. M. Sutherland, P. Johnston,
P. B. Wells, A. F. Carley, M. K. Rajumon, M. W. Roberts, A. Ibbotson, Recl.
Trav. Chim. Pays-Bas 1994, 113, 465.
[25] X. Zuo, H. Liu, D. Guo, X. Yang, Tetrahedron 1999, 55, 7787.
[26] S. Lavoie, M. A. Laliberté, P. H. McBreen, J. Am. Chem. Soc. 2003, 125,
15756.
[27] A. Taskinen, E. Toukoniitty, Catal. Today 2005, 100, 373.
[28] R. L. Augustine, S. K. Tanielyan, J. Mol. Catal. A-Chem. 1996, 112, 93.
[29] J. L. Margitfalvi, E. Tfirst, J. Mol. Catal. A-Chem. 1999, 139, 81.
[30] G. Vayner, K. N. Houk, Y. K. Sun, J. Am. Chem. Soc. 2004, 126, 99.
[31] M. Bartók, M. Sutyinszki, I. Bucsi, K. Felföldi, G. Szöllösi, F. Bartha, T.
Bartók, J. Catal. 2005, 231, 33.
[32] E. Orglmeister, T. Mallat, A. Baiker, J. Catal. 2005, 233, 333.
[33] J. W. D. Carneiro, C. D. B. de Oliveira, F. B. Passos, D. A. G. Aranda, P.
R. N. de Souza, O. A. C. Antunes, J. Mol. Catal. A: Chem. 2005, 226, 221.
[34] S. Diezi, A. Szabo, T. Mallat, A. Baiker, Tetrahedron: Asymmetry 2003, 14,
2573.
[35] S. Diezi, T. Mallat, A. Szabo, A. Baiker, J. Catal. 2004, 228, 162.
[36] N. Bonalumi, A. Vargas, D. Ferri, T. Bürgi, T. Mallat, A. Baiker, J. Am.
Chem. Soc. 2005, 127, 8467.
[37] W. K. Chan, W. Y. Yu, C. M. Che, M. K. Wong, J. Org. Chem. 2003, 68,
6576.
[38] F. Heaney, J. Fenlon, P. McArdle, D. Cunningham, Org. Biomol. Chem.
2003, 1, 1122.
[39] X. Creary, J. Org. Chem. 1987, 52, 5026.
[40] H. C. Brown, G. G. Pai, J. Org. Chem. 1985, 50, 1384.
[41] M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J.
R. Cheeseman, V. G. Zakrzewski, J. A. Montgomery, R. E. Stratmann, J. C.
Burant, S. Dapprich, J. M. Millam, A. D. Daniels, K. N. Kudin, M. C. Strain,
O. Farkas, J. Tomasi, V. Barone, M. Cossi, R. Cammi, B. Mennucci, C.
Pomelli, C. Adamo, S. Clifford, J. Ochterski, G. A. Petersson, P. Y. Ayala, Q.
154 Chapter 7
Cui, K. Morokuma, D. K. Malick, A. D. Rabuck, K. Raghavachari, J. B. Fores-
man, J. Cioslowski, J. V. Ortiz, A. G. Baboul, B. B. Stefanov, G. Liu, A. Liash-
enko, P. Piskorz, I. Komaromi, R. Gomperts, R. L. Martin, D. J. Fox, T. Keith,
M. A. Al-Laham, C. Y. Peng, A. Nanayakkara, C. Gonzalez, M. Challacombe,
P. M. W. Gill, B. Johnson, W. Chen, M. W. Wong, J. L. Andres, M. Head-Gor-
don, E. S. Replogle, J. A. Pople, in GAUSSIAN98 Pittsburgh, PA (1998).
[42] J. A. Schellman, Chem. Rev. 1975, 75, 323.
[43] M. Sutyinszki, K. Szori, K. Felföldi, M. Bartok, Catal. Lett. 2002, 81,
281.
[44] M. von Arx, G. J. Hutchings, New J. Chem. 2003, 27, 1367.
[45] E. Toukoniitty, P. Maki-Arvela, N. Kumar, T. Salmi, D. Y. Murzin, Catal.
Lett. 2004, 95, 179.
[46] T. Mallat, Z. Bodnar, B. Minder, K. Borszéky, A. Baiker, J. Catal. 1997,
168, 183.
[47] D. Ferri, T. Bürgi, K. Borszéky, T. Mallat, A. Baiker, J. Catal. 2000, 193,
139.
[48] J. M. Bonello, R. M. Lambert, N. Künzle, A. Baiker, J. Am. Chem. Soc.
2000, 122, 9864.
[49] D. Ferri, T. Bürgi, A. Baiker, J. Phys. Chem. B 2004, 108, 14384.
[50] G. D. H. Dijkstra, R. M. Kellogg, H. Wynberg, J. Org. Chem. 1990, 55,
6121.
[51] P. B. Wells, A. G. Wilkinson, Top. Catal. 1998, 5, 39.
[52] C. Hansch, A. Leo, R. W. Taft, Chem. Rev. 1991, 91, 165.
[53] H.-U. Blaser, H. P. Jalett, J. Wiehl, J. Mol. Catal. 1991, 68, 215.
[54] A. Perosa, P. Tundo, M. Selva, J. Mol. Catal. A: Chem. 2002, 180, 169.
[55] R. Hess, S. Diezi, T. Mallat, A. Baiker, Tetrahedron: Asymmetry 2004, 15,
251.
[56] A. Baiker, H.-U. Blaser, in: Handbook of Heterogeneous Catalysis, eds. G.
Ertl, H. Knoezinger, J. Weitkamp, vol. 5, VCH, Weinheim, 1997, p. 2422.
[57] T. Bürgi, A. Baiker, J. Catal. 2000, 194, 445.
[58] A. Vargas, D. Ferri, A. Baiker, J. Catal. 2005, 236, 1.
[59] H.-U. Blaser, H. P. Jalett, Stud. Surf. Sci. Catal. 1993, 78, 139.
[60] G.-Z. Wang, T. Mallat, A. Baiker, Tetrahedron: Asymmetry 1997, 8, 2133.
[61] M. Schürch, N. Künzle, T. Mallat, A. Baiker, J. Catal. 1998, 176, 569.
Directing effect of ester group 155
[62] A. Szabo, N. Künzle, T. Mallat, A. Baiker, Tetrahedron: Asymmetry 1999,
10, 61.
[63] B. Török, K. Felföldi, K. Balàzsik, M. Bartók, Chem. Commun. 1999,
1725.
[64] M. Studer, S. Burkhardt, H.-U. Blaser, Chem. Commun. 1999, 1727.
[65] W. A. H. Vermeer, A. Fulford, P. Johnston, P. B. Wells, J. Chem. Soc.
Chem. Commun. 1993, 1053.
[66] M. Studer, V. Okafor, H.-U. Blaser, Chem. Commun. 1998, 1053.
[67] E. Toukoniitty, P. Maki-Arvela, J. Warna, T. Salmi, Catal. Today 2001, 66,
411.
[68] T. Mallat, M. Bodmer, A. Baiker, Catal. Lett. 1997, 44, 95.
[69] K. Balàzsik, B. Török, K. Felföldi, M. Bartók, Ultrason. Sonochem. 1999,
5, 149.
[70] J. Zhang, X.P . Yan, H. F. Liu, J. Mol. Catal. A: Chem. 2001, 175, 125.
[71] K. Felföldi, T. Varga, P. Forgó, M. Bartók, Catal. Lett. 2004, 97, 65.
[72] T. Varga, K. Felföldi, P. Forgó, M. Bartók, J. Mol. Catal. A: Chem. 2004,
216, 181.
[73] M. von Arx, T. Mallat, A. Baiker, J. Catal. 2000, 193, 161.
7
7Outlook
The results presented in this work indicate that already small changes in the
modifier structure far from the interacting functional group can have a dramatic
influence on the enantioselectivity, due to changes in the adsorption geometry
of the modifier (Chapter 2 & 3). Fine tuning of the modifier structures may
extend the application range of chirally modified metal catalyst systems and it is
essential to support and refine existing mechanistic models developed for heter-
ogeneous catalysis. The high-throughput screening proved to be a suitable and
really helpful approach to accelerate such structure-performance studies.
Substantial work is furthermore necessary in the field of structurally new,
synthesized modifiers. The most promising synthetic modifier so far is (R,R)-
pantoyl-naphthylethylamine (Chapter 4 & 5) affording over 90% enantiomeric
excess (ee) in the asymmetric hydrogenation of fluorinated β-diketones. This
result represents the first case in heterogeneous catalysis that a synthetic modi-
fier provides higher ee than the commonly used modifier of natural origin (cin-
chona alkaloids and their simple derivatives). Although the activity of this new
synthetic catalyst system is moderate and the stability needs to be improved,
first success with this modifier-Pt catalyst system can help to create well
designed modifiers and extend the application range of heterogeneous enanti-
oselective catalysts.
Catalytic experiments combined with spectroscopic and computational
structural analysis have been applied in the hydrogenation of 1,1,1-trifluoro-
2,4-diketopentane to understand the coupled phenomena of enantioselection
and chemoselectivity (Chapter 6). The study revealed the powerfull combina-
tion of experimental, spectroscopic and computational methods which are
truly helpful in mechanistic studies, and in understanding the functioning of
chirally modified metal catalysts.
158
Important challenges for the future remain a better fundamental under-
standing of these complex catalytic systems and the extension of the aquired
knowledge to other type of reactions.
0
0List of Publications
The following list summarizes publications, which are based on this thesis. The
pertinent chapters of this thesis, which are the source of the publications, are
given in brackets.
“Inversion of Enantioselectivity in the Hydrogenation of Ketopantolactone on
Platinum Modified by Ether Derivatives of Cinchonidine”
S. Diezi, T. Mallat, A. Baiker, Tetrahedron: Asymmetry 14 (2003) 2573.
(Chapter 2)
“Fine Tuning the "Chiral Sites" on Solid Enantioselective Catalysts”
S. Diezi, T. Mallat, A. Szabo, A. Baiker, J. Catal. 228 (2004) 162.
(Chapter 3)
“An Efficient Synthetic Chiral Modifier for Platinum”
S. Diezi, M. Hess, E. Orglmeister, T. Mallat, A. Baiker, Catal. Lett. 102 (2005)
121.
(Chapter 4)
“Chemo and Enantioselective Hydrogenation of Fluorinated Ketones on Plati-
num Modified with (R)-1-(1-naphthyl)ethylamine Derivatives”
S. Diezi, M. Hess, E. Orglmeister, T. Mallat, A. Baiker, J. Mol. Catal. A: Chem.
239 (2005) 49.
(Chapter 5)
162
“The Origin of Chemo- and Enantioselectivity in the Hydrogenation of Dike-
tones on Platinum”
S. Diezi, D. Ferri, A. Vargas, T. Mallat, A. Baiker, J. Am. Chem. Soc. 128
(2006) 4048.
(Chapter 6)
“Steric and Electronic Effects in the Enantioselective Hydrogenation of Acti-
vated Ketones on Platinum: Directing Effect of Ester Group”
S. Diezi, S. Reimann, N. Bonalumi, T. Mallat, A. Baiker, J. Catal. 239 (2006)
255.
(Chapter 7)
List of other publications not included in the thesis:
“Alumina-catalysed Degradation of Ethyl Pyruvate during Enantioselective
Hydrogenation over Pt/Alumina and its Inhibition by Acetic Acid”
D. Ferri, S. Diezi, M. Maciejewski, A. Baiker, Applied Catalysis A: General 297
(2006) 165.
“Chemo- and Enantioselective Hydrogenation of the Activated Keto Group of
Fluorinated β-Diketones”
R. Hess, S. Diezi, T. Mallat, A. Baiker, Tetrahedron: Asymmetry 15 (2004) 251.
List of contributions to conferences:
"A Real In Situ Study of the Adsorption Mode of Cinchona Alkaloids on Pt
using O-alkyl-cinchonidines as Chiral Modifiers"
S. Diezi, T. Mallat, and A. Baiker, EuropaCat VI, Innsbruck, Austria, 2003,
poster.
List of Publications 163
"Asymmetric Hydrogenation of an α-functionalized Ketone on Pt/Al2O3 Mo-
dified by Ether Derivatives of Cinchonidine"
S. Diezi, T. Mallat, and A. Baiker, Swiss Chemical Society, Fall Meeting, Lau-
sanne, Switzerland, 2003, poster.
"Enantioselektive Hydrogenation on Pt/Al2O3 Modified by O-alkylated and
arylated Cinchonidines"
S. Diezi, T. Mallat, and A. Baiker, Jahrestreffen Deutscher Katalytiker, Weimar,
Germany, 2004, poster.
"The Critical Role of Cinchonidine Adsorption on Pt/Al2O3 in the Enantio-
selective Hydrogenation of Ketones"
S. Diezi, T. Mallat, and A. Baiker, Swiss Chemical Society, Fall Meeting, Zürich,
Switzerland, 2004, poster.
"A New Synthetic Modifier for the Pt-Catalyzed Enantioselective Hydrogen-
ation of Fluorinated Ketones"
S. Diezi, E. Orglmeister, T. Bürgi, T. Mallat and A. Baiker, 19th North American
Catalysis Society Meeting, Philadelphia, PA, USA, 2005, oral presentation.
"A New Synthetic Modifier for the Pt-Catalyzed Enantioselective Hydrogen-
ation of Fluorinated Ketones"
S. Diezi, E. Orglmeister, T. Mallat, and A. Baiker, Swiss Chemical Society, Fall
Meeting, Lausanne, Switzerland, 2005, poster.
"A New Synthetic Modifier for the Pt-Catalyzed Enantioselective Hydrogen-
ation of Fluorinated Ketones"
S. Diezi, E. Orglmeister, T. Mallat, and A. Baiker, 7th International Symposium
on Catalysis to Fine Chemicals, Bingen/Mainz, Germany, 2005, oral presenta-
tion.
0
0Curriculum Vitae
Name Simon Diezi
Date of Birth 12 March 1977
City Thal (SG)
Citizen of Thal (SG)
Nationality Swiss
Education
1992-1997 Kantonsschule Heerbrugg
Graduation with Matura Type C
1997-2002 ETH Zürich, Chemistry Department
Chemistry Studies
Graduation as Dipl. Chem. ETH
2002-2006 ETH Zürich, Institute for Chemical and Bioengineering
Doctor Thesis under the Supervision of
Prof. Dr. A. Baiker