1-Naphthyl-1,2-ethanediol as a new chiral modifier of platinum in the enantioselective hydrogenation of activated ketones

Article abstract of DOI:10.1016/j.jcat.2003.10.008

1-Naphthyl-1,2-ethanediol (NED) is shown to be a useful modifier in the hydrogenation of ketopantolactone and ethyl-4,4,4-trifluoroacetoacetate under mild conditions. It represents the first effective chiral nonamine-type modifier of Pt for the enantioselective hydrogenation of activated ketones. The enantio-differentiation is attributed to substrate-modifier interactions involving hydrogen bonding between the keto-carbonyl O atom and one or two OH groups of NED. Prominent nonlinear behavior was observed when mixtures of (S)-NED and (R)-2-(1-pyrrolidinyl)-1-(1-naphthyl)ethanol {(R)-PNE} were applied as chiral modifiers. The phenomenon is traced to stronger adsorption of PNE on the metal surface, despite the identical "anchoring moiety" (naphthalene ring) of the two modifiers.

Full text of DOI:10.1016/j.jcat.2003.10.008

Journal of Catalysis 221 (2004) 666–669
www.elsevier.com/locate/jcat
Priority Communication
1-Naphthyl-1,2-ethanediol as a new chiral modifier of platinum
in the enantioselective hydrogenation of activated ketones
Alberto Marinas, Tamas Mallat, and Alfons Baiker ∗
Institute for Chemical and Bioengineering, Swiss Federal Institute of Technology, ETH Hönggerberg, CH-8093 Zurich, Switzerland
Received 3 October 2003; revised 24 October 2003; accepted 24 October 2003
Abstract
1-Naphthyl-1,2-ethanediol (NED) is shown to be a useful modifier in the hydrogenation of ketopantolactone and ethyl-4,4,4-trifluoro-
acetoacetate under mild conditions. It represents the first effective chiral nonamine-type modifier of Pt for the enantioselective hydrogenation
of activated ketones. The enantio-differentiation is attributed to substrate–modifier interactions involving hydrogen bonding between the
keto-carbonyl O atom and one or two OH groups of NED. Prominent nonlinear behavior was observed when mixtures of (S)-NED and
(R)-2-(1-pyrrolidinyl)-1-(1-naphthyl)ethanol {(R)-PNE} were applied as chiral modifiers. The phenomenon is traced to stronger adsorption
of PNE on the metal surface, despite the identical “anchoring moiety” (naphthalene ring) of the two modifiers.
2003 Elsevier Inc. All rights reserved.
Keywords: Asymmetric; Enantioselective; Hydrogenation; Pt/Al O ; 1-Naphthyl-1,2-ethanediol; 2-(1-Pyrrolidinyl)-1-(1-naphthyl)ethanol;
2
3
Ketopantolactone; Ethyl-4,4,4-trifluoroacetoacetate
1. Introduction
thryl- and phenanthryl-ethylamines [13], vinca alkaloid
derivatives [14], L-triptophane derivatives [15], and isocin-
chona alkaloids [16]. All these modifiers fulfill the require-
ments for an effective chiral modifier of Pt, extracted from
studies of cinchona derivatives as noted above. In other
words, all of them may be considered as analogous to the
cinchona alkaloids and not as truly new types of chiral modi-
fiers of Pt. This conclusion is supported by recent theoretical
calculations of the interaction of ethyl pyruvate with some
modifiers [17].
Interestingly, there is a report on 3% enantiomeric excess
(ee) achieved in pyruvate hydrogenation with quaternized
codeine (methylcodeinium iodide) [18] and 2% ee with N-
benzylcinchonidium chloride [19]. However, other authors
obtained reproducibly a racemic product with the latter cin-
chona derivative [20], and this observation is now commonly
accepted as a strong experimental evidence for the cru-
cial role of the basic N atom of cinchona alkaloids in the
reactant–modifier interaction(s).
The application of a metal hydrogenation catalyst to-
gether with a strongly adsorbing chiral molecule (“chiral
modifier”) is a simple and powerful method for the het-
erogeneous enantioselective hydrogenation of prochiral sub-
strates. Among the known catalyst systems, supported Pt
modified by small amounts of a cinchonaalkaloid (discov-
ered by Orito et al. [1]) is the best choice for the enantiose-
lective hydrogenation of activated ketones [2–5]. A system-
atic variation of the structure of cinchonidine [6] and some
other chiral amine modifiers [7–9] revealed that the crucial
structural parts of the modifier, beside the stereogenic cen-
ter(s), are the flat aromatic ring system for anchoring onto
the Pt surface and the basic aliphatic nitrogen function for
interacting with the substrate.
Meanwhile, several other chiral amines have been synthe-
sized and tested in the hydrogenation of activated ketones,
mainly of ethyl or methyl pyruvate. The useful modi-
fiers of Pt include 2-(1-pyrrolidinyl)-(1-naphthyl)ethanol
[10] and other similar chiral 1,2-aminoalcohols [9,11],
1-(1-naphthyl)ethylamine and their derivatives [8,12], an-
Here we report the enantioselective hydrogenation of
activated ketones over Pt/Al2O3 modified by 1-naphthyl-1,2-
ethanediol (NED, Scheme 1). To our knowledge, this is the
first example of an effective nonamine-type chiral modifier
of Pt and this unexpected finding may generate new opportu-
*
Corresponding author.
E-mail address: baiker@tech.chem.ethz.ch (A. Baiker).
0021-9517/$ – see front matter
doi:10.1016/j.jcat.2003.10.008
2003 Elsevier Inc. All rights reserved.

A. Marinas et al. / Journal of Catalysis 221 (2004) 666–669
667
(S)-1-naphthyl-1,2-ethanediol (NED), respectively. Calcu-
lation of ee’s in the modifiers was made on the basis of the
[α]D values and ee’s reported in Refs. [7,10,21].
2.2. Catalytic hydrogenations
A 5% Pt/Al2O3 catalyst (Engelhard 4759) was prere-
duced in flowing H2 for 90 min at 400 ◦C. The Pt dispersion
after heat treatment was 0.27 as determined by TEM mea-
surements.
The hydrogenationreactions were carried out in a 100-mL
stainless-steel autoclave equipped with a 50-mL glass liner
and PTFE cover. Pressure was held constant during reaction
with a Büchi BPC 9901 system. Under standard conditions,
20 mg prereduced catalyst, 1.84 mmol substrate, 12.7 µmol
modifier (corresponding to a substrate/modifier molar ra-
tio of 144), and 10 mL solvent (toluene) were used. Unless
otherwise stated, the reactions were carried out at room tem-
perature.
Scheme 1. Enantioselective hydrogenation of activated ketones over
Pt/Al O modified by 1-naphthyl-1,2-ethanediol. Models for interactions
between activated ketones and NED or PNE. For description of a, b, and c
2
3
Conversion and enantiomeric excess were determined
using an HP 6890 gas chromatograph (GC) and a Chirasil-
DEX CB (ChromPack) capillary column.
see text.
nities in the search for suitable chiral modifiers for platinum
metals.
3. Results and discussion
The new modifier 1-naphthyl-1,2-ethanediol was first
tested in the enantioselective hydrogenationof ketopantolac-
tone, an intermediate in the synthesis of pantothenic acid (vi-
tamin B5). In Fig. 1 the performance of NED is compared to
that of (R)-2-(1-pyrrolidinyl)-1-(1-naphthyl)ethanol {(R)-
PNE [10]}, the modifier of which possesses a similar struc-
ture but one of the OH functions is replaced by a 1-pyrroli-
dinyl group. The ee to (R)-pantolactone decreased at higher
pressures with both modifiers. The same correlation was ob-
served with (R)-PNE in the enantioselective hydrogenation
of ethyl pyruvate [10]. The loss of enantioselectivity to (R)-
lactate at high surface hydrogenconcentration was explained
by the hydrogenation of the naphthyl ring of PNE and the
2. Experimental
2.1. Chemicals
4,4-Dimethyldihydrofuran-2,3-dione (ketopantolactone,
Hoffmann-La-Roche, >99%), ethyl-4,4,4-trifluoroaceto-
acetate (Fluka, >98%), 2,2,2-trifluoroacetophenone (Fluka,
>98%), acetic acid (Fluka, >99%), toluene (J.T. Baker,
>99%), and 1,2-dichlorobenzene(Fluka, >99%) were used
as received. Ethyl pyruvate (Fluka, >97%) was carefully
distilled in vacuum before use.
(R)- and (S)-1-naphthyl-1,2-ethanediol and (R)-2-(1-
pyrrolidinyl)-1-(1-naphthyl)ethanol(PNE) were synthesized
from 1-vinylnaphthalene according to the method developed
in Pfaltz’s group [7,10]. 1-Vinylnaphthalene was freshly
synthesized from 1-naphthaldehyde (Acros, 95%) via a Wit-
tig reaction, because of its high tendency to polymerize.
The final products were purified by flash column chro-
matography and then recrystallized several times in an
ether/hexane (1/1) mixture until clean 1H NMR spectra
1
were obtained. H NMR (CDCl3, 500 MHz): NED 8.04
(m, 1H), 7.87 (m, 1H), 7.79 (d, 1H), 7.68 (d, 1H), 7.48
(m, 3H), 5.61 (dd, 1H), 3.96 (dd, 1H), 3.78 (dd, 1H),
2.88 (d, 1H, OH), 2.45 (m, 1H, OH); PNE 8.05 (d, 1H),
7.87 (m, 1H), 7.77 (m, 2H), 7.47 (m, 3H), 5.53 (dd, 1H),
4.75–3.50 (br s, 1H, OH), 2.80 (m, 4H), 2.62 (m, 2H), 1.85
(m, 4H). The measured [α]D values at 20 ◦C in CHCl3 and
calculated ee values (in brackets) are −116◦ (>99% ee),
−91◦ (>98% ee) and +90◦ (>97% ee) for (R)-2-(1-
pyrrolidinyl)-1-(1-naphthyl)ethanol (PNE) and (R)- and
Fig. 1. Pressure dependence of enantioselectivity in the hydrogenation of
ketopantolactone on 5% Pt/Al O modified by NED and PNE under stan-
2
3
dard conditions.

668
A. Marinas et al. / Journal of Catalysis 221 (2004) 666–669
Fig. 3. Hydrogenation of various activated ketones on Pt/Al O modified by
2
3
Fig. 2. Nonlinear behavior of mixtures of (R)-PNE and (S)-NED in the
enantioselective hydrogenation of ketopantolactone on 5% Pt/Al O ; stan-
dard conditions, 10 bar.
either PNE or NED. The ee’s are compared at 100% conversion. Standard
reaction conditions, 10 mL toluene (dichlorobenzene for 3), 20 mg catalyst
(40 mg for 3 and 4), and 10 bar. The reaction time necessary to achieve full
conversion varied in the range 0.3–1.5 h.
2
3
weaker adsorption of the (partially) saturated modifier. Note
that saturation of the quinoline ring of cinchonidine during
enantioselective hydrogenation is a well-documented side
reaction [22,23]. Hence, we assume that the better efficiency
of NED at high pressures is due to the slower hydrogenation
of the naphthalene ring of NED, compared to that of PNE.
The initial TOF values at 10 bar were 2.4, 2.9, and 3.6 s−1
in the absence of modifier, in the presence of NED {either
(R) or (S)}, and (R)-PNE, respectively. Thus both NED and
PNE induce a small rate acceleration compared to the un-
modified reaction.
group metals and the different basicities of the amino and
alcoholic OH functions.
The stronger adsorption of PNE on Pt leads to surface en-
richment of PNE in PNE–NED mixtures, the effect of which
seems to be responsible for the nonlinear behavior in Fig. 2,
and presumably also for the faster hydrogenation of the aro-
matic ring of PNE and its lower efficiency at high pressures
(Fig. 1).
Beside the different adsorption strength of NED and PNE,
the various electronic and steric effects among the adsorbed
species may also contribute to the nonlinear behavior of their
mixture but these effects are largely unknown.
The two chiral modifiers NED and PNE have been tested
in the hydrogenation of some other activated ketones, such
as ethyl pyruvate (2), ethyl-4,4,4-trifluoroacetoacetate (3),
and 2,2,2-trifluoroacetophenone (4) (Fig. 3). Though the
conditions are not optimized for any of the reactions, we can
conclude that PNE is effective in all four reactions as a chi-
ral modifier of Pt, whereas NED can only be used for 3 and
ketopantolactone (1). The reason for this unusual structural
effect is yet unclear. Besides, the ee’s obtained with PNE are
always higher than those achieved with NED.
An intriguing question is the nature of substrate–modifier
interactions on the Pt surface resulting in enantioselection.
The model suggested for PNE in the hydrogenation of ethyl
pyruvate [17] is depicted in Scheme 1a. The attractive inter-
action, represented by the N–H–O-type hydrogen bond, is
completed by a repulsion exerted by the aromatic ring of the
modifier (not shown in Scheme 1).
The results in Fig. 1 indicate that the adsorption strength
of NED and PNE on Pt may be a critical parameter in de-
termining their efficiency. To clarify this point, the nonlinear
behavior of mixtures of the two modifiers was studied in the
hydrogenationof ketopantolactone (Fig. 2). This method has
been shown to be a useful tool for in situ investigation of
modifier adsorption on Pt and Pd [24–26]. The calculated ee
in Fig. 2 corresponds to an ideal case in which both modifiers
adsorb on the metal surface with the same adsorption free en-
ergy, they do not interact with each other in solution or on
the metal surface, and catalyze the enantioselective hydro-
genation of ketopantolactone with the same rate and ee that
are characteristic for each modifier when used alone [27].
The experimental values in Fig. 2 demonstrate that PNE
governs the enantioselection when using mixtures of (R)-
PNE and (S)-NED as modifiers. Even when Pt was modified
by a 10 mol% PNE–90 mol% NED mixture, the ee was
the same as obtained for pure PNE. The remarkable nonlin-
ear behavior is attributed mainly to the stronger adsorption
of PNE on Pt. This conclusion may be astonishing at first
sight, as both modifiers possess the same anchoring moiety:
a naphthalene ring. We assume that the stronger adsorption
of PNE on Pt is due to the presence of the basic N atom.
Ranade et al. [28,29] have shown recently that an amino
function is a much better directing group in diastereoselec-
tive hydrogenation of aromatic compounds than a hydroxyl
function, due to the stronger electronic attraction of the
amino group to the metal surface. The same conclusion may
be drawn when considering the soft acid character of Pt-
We propose two possible structures for the attractive in-
teraction between NED and the activated ketone substrate,
involving hydrogen bonding between the oxygen atom of
the keto-carbonyl group of the substrate and one or two
hydroxyl groups of the modifier (Schemes 1b and 1c). Us-
ing ab initio calculations, Klein [30] found two analogous
structures as the most stable interactions between vic-diols
(ethane-1,2-diol, propane-1,2-diol, or butane-2,3-diol) and
water. Middleton et al. [31] described structures similar to
Scheme 1c for fluorinated gem-diols with hydrogen-bond

A. Marinas et al. / Journal of Catalysis 221 (2004) 666–669
669
acceptors. The presence of fluorine bonded directly to the
References
α-carbon atom confers a strongly acidic character to the hy-
droxyl groups and an unusually high stability to the complex.
In our case the double interaction cannot be so strong due to
the weak acidity of the diol (pka = 13.4 [32]). On the other
hand, the single O–H–O-type interaction in Scheme 1b is
expected to be considerably weaker than the N–H–O inter-
action in Scheme 1a.
[1] Y. Orito, S. Imai, S. Niwa, J. Chem. Soc. Jpn. (1979) 1118.
[2] M. Studer, H.U. Blaser, C. Exner, Adv. Synth. Catal. 345 (2003) 45.
[3] P.B. Wells, A.G. Wilkinson, Top. Catal. 5 (1998) 39.
[4] A. Baiker, J. Mol. Catal. A 163 (2000) 205.
[5] M. Von Arx, T. Mallat, A. Baiker, Top. Catal. 19 (2002) 75.
[6] H.U. Blaser, H.P. Jalett, D.M. Monti, A. Baiker, J.T. Wehrli, Stud.
Surf. Sci. Catal. 67 (1991) 147.
[7] K.E. Simons, G. Wang, T. Heinz, A. Pfaltz, A. Baiker, Tetrahedron:
Asymmetry 6 (1995) 505.
[8] B. Minder, M. Schürch, T. Mallat, A. Baiker, T. Heinz, A. Pfaltz, J. Ca-
tal. 160 (1996) 261.
[9] M. Schürch, T. Heinz, R. Aeschimann, T. Mallat, A. Pfaltz, A. Baiker,
J. Catal. 173 (1998) 187.
[10] G. Wang, T. Heinz, A. Pfaltz, B. Minder, T. Mallat, A. Baiker, J. Chem.
Soc., Chem. Commun. (1994) 2047.
An attempt to confirm the structures in Scheme 1b or 1c
by NMR failed. Note that there is no NMR evidence yet for
the N–H–O-type hydrogen bond between the known chiral
modifiers and any of the activated ketones. However, very
recently this hydrogen bonding type could be evidenced for
the cinchonidine–ketopantolactone interaction using ATR
infrared concentration modulation spectroscopy [33].
[11] A. Solladié-Cavallo, C. Marsol, F. Garin, Tetrahedron Lett. 43 (2002)
4733.
[12] B. Minder, M. Schürch, T. Mallat, A. Baiker, Catal. Lett. 31 (1995)
143.
[13] A. Solladié-Cavallo, C. Marsol, C. Suteu, F. Garin, Enantiomer 6
(2001) 245.
4. Conclusions
A new chiral modifier of Pt, 1-naphthyl-1,2-ethanediol
(NED) has been found for the enantioselective hydrogena-
tion of activated ketones. At best, the NED–Pt/Al2O3 cata-
lyst system afforded 30% ee to (R)-(−)-pantolactone in the
hydrogenation of ketopantolactone at 5 bar and 0 ◦C. A re-
markable nonlinear effect was observed when mixtures of
(S)-NED and (R)-2-(1-pyrrolidinyl)-1-(1-naphthyl)ethanol
((R)-PNE) were applied as modifiers. This behavior is traced
to the stronger adsorption of PNE on Pt, likely due to the
stronger electronic attraction of the amino group to the Pt
surface. Two feasible models have been proposed for the
NED–substrate interaction involving O–H–O-type hydrogen
bonds.
Until now it was generally believed [2,4] that a cru-
cial feature of an effective chiral modifier of Pt is a basic,
aliphatic nitrogen atom for interacting with the activated
ketone substrate. Here we provide strong experimental evi-
dence against this hypothesis. Understanding the functioning
of the Pt–NED catalyst system may initiate new approaches
for tailor-made modifiers and broaden the scope of hetero-
geneous asymmetric catalysis.
[14] A. Tungler, T. Mathe, K. Fodor, R.A. Sheldon, P. Gallezot, J. Mol.
Catal. A 108 (1996) 145.
[15] G. Szöllösi, C. Somlai, P.T. Szabó, M. Bartók, J. Mol. Catal. A 170
(2001) 165.
[16] M. Bartók, K. Felföldi, B. Török, T. Bartók, Chem. Commun. (1998)
2605.
[17] A. Vargas, T. Bürgi, A. Baiker, J. Catal. 197 (2001) 378.
[18] S.P. Griffiths, P.B. Wells, K.G. Griffin, P. Johnston, in: F.E. Herkes
(Ed.), Catalysis of Organic Reactions, Dekker, New York, 1998, p. 89.
[19] P.B. Wells, R.P.K. Wells, in: D.E. De Vos, I.F.J. Vankelecom, P.A.
Jacobs (Eds.), Chiral Catalyst Immobilization and Recycling, Wiley–
VCH, Weinheim, 2000, p. 123.
[20] H.U. Blaser, H.P. Jalett, D.M. Monti, A. Baiker, J.T. Wehrli, Stud.
Surf. Sci. Catal. 67 (1991) 147.
[21] T. Heinz, PhD dissertation, University of Basel, 1997.
[22] M. Bartók, G. Szöllösi, K. Balázsik, T. Bartók, J. Mol. Catal. A 177
(2002) 299.
[23] W.R. Huck, T. Mallat, A. Baiker, Catal. Lett. 87 (2003) 241.
[24] K.E. Simons, P.A. Meheux, A. Ibbotson, P.B. Wells, Stud. Surf. Sci.
Catal. 75 (1993) 2317.
[25] A. Tungler, K. Fodor, T. Mathe, R.A. Sheldon, Stud. Surf. Sci.
Catal. 108 (1997) 157.
[26] W.R. Huck, T. Bürgi, T. Mallat, A. Baiker, J. Catal. 216 (2003) 276.
[27] W.R. Huck, T. Mallat, A. Baiker, Adv. Synth. Catal. 345 (2003) 255.
[28] V.A. Ranade, G. Consiglio, R. Prins, J. Org. Chem. 64 (1999) 8862.
[29] V.A. Ranade, G. Consiglio, R. Prins, J. Org. Chem. 65 (2000) 1132.
[30] R.A. Klein, J. Comput. Chem. 23 (2002) 585.
[31] W.J. Middleton, R.V. Lindsey Jr., J. Am. Chem. Soc. 86 (1964) 4948.
[32] Data taken from Scifinder Scholar 2001. American Chemical Society,
calculated using Advanced Chemistry Development (ACD) Software
Solaris V4.67.
Acknowledgments
Financial support by the Swiss National Science Founda-
tion is gratefully acknowledged. A. Marinas is thankful to
Fundación Ramón Areces for a postdoctoral grant.
[33] N. Bonalumi, T. Bürgi, A. Baiker, J. Am. Chem. Soc. 125 (2003)
13342.