Non-Linear Effect of Modifier Composition on Enantioselectivity in Asymmetric Hydrogenation over Platinum Metals

Article abstract of DOI:10.1002/adsc.200390020

Prominent non-linear behavior was observed when mixtures of cinchona alkaloids were applied as chiral modifiers in enantioselective hydrogenations over Pt/Al₂O₃ and Pd/TiO₂. The phenomenon is traced to differences in the strength and geometry of alkaloid adsorption on the metal surface. In ethyl pyruvate hydrogenation under close to ambient conditions the weaker adsorbing quinidine (QD) outperformed the generally preferred modifier cinchonidine (CD) and afforded the highest ee (96-98%) at 1-5 bar. In the partial hydrogenation of 4-methoxy-6-methyl-2-pyrone to the dihydro derivative 4 CD gave 73% ee to (S)-4 and QD provided 72% ee to (R)-4, and still the alkaloid mixture containing less than 5 mol % CD afforded 15% ee to (S)-4. This nonlinear behavior may be advantageous in asymmetric synthesis as low purity chiral compounds can be applied as efficient modifiers for Pt or Pd.

Full text of DOI:10.1002/adsc.200390020

FULL PAPERS
Non-Linear Effect of Modifier Composition on Enantioselectivity
in Asymmetric Hydrogenation over Platinum Metals
Wolf-R¸diger Huck, Tamas Mallat, Alfons Baiker*
Laboratory of Technical Chemistry, Swiss Federal Institute of Technology, ETH-Hˆnggerberg, 8093 Z¸rich, Switzerland
Phone: ( ‡ 41)-1-632-3153, fax: ( ‡ 41)-1-632-1163, e-mail: baiker@tech.chem.ethz.ch
Received: July 18, 2002; Accepted: September 9, 2002
Abstract: Prominent non-linear behavior was ob- methoxy-6-methyl-2-pyrone to the dihydro derivative
served when mixtures of cinchona alkaloids were 4 CD gave 73% ee to (S)-4 and QD provided 72% ee
applied as chiral modifiers in enantioselective hydro- to (R)-4, and still the alkaloid mixture containing less
genations over Pt/Al2O3 and Pd/TiO2. The phenom- than 5 mol % CD afforded 15% ee to (S)-4. This non-
enon is traced to differences in the strength and linear behavior may be advantageous in asymmetric
geometry of alkaloid adsorption on the metal surface. synthesis as low purity chiral compounds can be
In ethyl pyruvate hydrogenation under close to applied as efficient modifiers for Pt or Pd.
ambient conditions the weaker adsorbing quinidine
(QD) outperformed the generally preferred modifier Keywords: asymmetric hydrogenation; cinchona alka-
cinchonidine (CD) and afforded the highest ee (96
loids; ethyl pyruvate; 4-methoxy-6-methyl-2-pyrone;
98%) at 1 5 bar. In the partial hydrogenation of 4- palladium; platinum
Introduction
enantiomers in excess over Pt and Pd, are shown in
Figure 1.
From a synthetic point of view, a promising strategy in
An interesting and yet overlooked point is that the
heterogeneous asymmetric hydrogenation is the use of a purity of commercial CD is rather low: it contains a
conventional metal hydrogenation catalyst together considerable amount of QD (in our sample 7%) and also
with a strongly adsorbed chiral compound (™modifier∫) some QN (1%). As CD and QD afford the opposite
as the source of chirality.[1 11] Cinchona alkaloids, and in enantiomer in excess compared to that formed with QN,
particular cinchonidine (CD), are versatile chiral modi- it is at first sight astonishing that the CD-QN-QD
fiers of Pt and Pd: they provide the highest ee in the mixture can afford 97 98% ee on Pt [13,14,24] and 94% ee
hydrogenation of various a-functionalized (activated) on Pd.[23] It is even more startling when considering the
ketones,[12 20] a,b-unsaturated carboxylic acids [21,22] and inevitable racemic reaction, i.e., hydrogenation of the
various functionalized 2-pyrones.[23] The two groups of substrate without involvement of the chiral modifier.
important cinchona alkaloids, which afford the opposite Analysis of this phenomenon revealed an interesting
non-linear behavior of cinchona alkaloid mixtures,
which is in some aspects analogous to the well-known
non-linear effect (NLE) in homogeneous catalysis,[25 27]
although the physicochemical background is remark-
ably different.
Results and Discussion
At first we investigated the hydrogenation of ethyl
pyruvate (1, Scheme 1), a commonly used test reaction
for the Pt-cinchona system. Under ambient conditions
in AcOH, QD was more effective than (commercial)
CD: it afforded 3% higher ee when it was used alone.
Still, CD controlled the enantioselection when mixtures
of the two alkaloids were used as modifiers (Figure 2a).
Even with a QD-CD molar ratio of 20, (R)-2 formed as
Figure 1. Some cinchona alkaloids and their abbreviations.
the major enantiomer at 11% ee. An explanation based
Adv. Synth. Catal. 2003, 345, No. 1+2
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim 1615-4150/03/3451+2-255 260 $ 17.50+.50/0
255

Wolf-R¸diger Huck et al.
FULL PAPERS
Scheme 2. Partial hydrogenation of a 4-methoxy-6-methyl-2-
pyrone (3) over cinchona-modified 5 wt % Pd/TiO2. a:
method D; b: optimized value.[23]
ior is even more impressive when the reaction rates are
also considered: in the presence of QD alone the rate
was 2.5 times faster than the rate of hydrogenation with
CD as the sole modifier of Pd (Figure 2b). Clearly, the
dominance of CD in enantioselection cannot be as-
signed to kinetic effects.
Let us assume an ideal case where (i) the two
modifiers adsorb on the metal surface with the same
adsorption free energy, (ii) they do not interact with
each other in solution or on the metal surface, and (iii)
both modifiers catalyze the enantioselective hydro-
genation of the reactant with the same rate and ee that
are characteristic for each modifier when they are used
alone. For this ideal case an expected ee can be
calculated based on the linear combination of specific
rates and ees determined independently. When the
measured ee deviates considerably from the calculated
value, this is an indication that the above assumptions
are not valid for that case. This situation is termed here
as ™non-linear behavior∫.
Figure 2. Influence of change in composition of a binary
modifier mixture (CD-QD) on the hydrogenation of 1 to 2
over Pt/Al2O3 (Scheme 1, method C), and 3 to 4 over Pd/TiO2
(Scheme 2, method D). (a) Enantioselectivities to 2 and 4; (b)
conversion of 1 and 3 after 30 min reaction time.
Wells and coworkers[28] have already observed small
deviations from the expected linear correlation in the
hydrogenation of 1 using mixtures of the diastereomers
CD and CN, or QN and QD. They attributed the
deviations between the observed and calculated ees to
differences in the adsorption strength of the alkaloids.
Tungler et al.[29] compared the efficiency of vinca and
cinchona alkaloid mixtures in the Pt-catalyzed hydro-
genation of 1 and the Pd-catalyzed hydrogenation of
isophorone. The strong non-linear behavior in the
second reaction is particularly interesting as either the
vinca or the cinchona alkaloid controlled the stereo-
chemical outcome of the reaction, depending on the
catalyst applied (Pd or Pd/C). This is a clear indication
for the crucial importance of adsorption on the metal
surface. Note that when modifying Pd by a vinca or a
Scheme 1. Hydrogenation of ethyl pyruvate (1) to ethyl
lactate (2) over cinchona-modified 5 wt % Pt/Al2O3. a:
method A (1 bar); b: method B (5 bar).
on different reaction rates can be ruled out, as the cinchona alkaloid, not only the adsorption strength and
conversion was almost independent of the alkaloid thus their surface concentration may differ, but also the
composition (Figure 2b).
nature of reactant-modifier interaction may be differ-
The non-linear behavior was slightly more pro- ent,[30] which is not the case for mixtures of cinchona
nounced in the hydrogenation of 3 to the corresponding alkaloids.[31] Recently, we have found a considerable
dihydropyrone 4 over 5 wt % Pd/TiO2 (Scheme 2 and non-linearity in the hydrogenation of 1 using mixtures of
Figure 2a). Again, CD controlled the enantiodifferen- CN and a synthetic modifier (R)-NPE (1-{1-naphthyl}-
tiation in the presence of CD-QD mixtures though the 2-{pyrrolidin-1-yl}ethanol).[32] Deviation from the ex-
two alkaloids alone afforded very similar ees to the pected linear correlation was rationalized by rapid
opposite enantiomers. The observed non-linear behav- saturation of the naphthyl rings of NPE in the early
256
Adv. Synth. Catal. 2003, 345, 255 260

Non-Linear Effect of Modifier Composition on Enantioselectivity in Asymmetric Hydrogenation
FULL PAPERS
(i) a p-bonded species adsorbed via the aromatic rings
nearly parallel to the metal surface (schematically
shown in Figure 3),
(ii) an N-lone pair-bonded species in which the aro-
matic rings are tilted relative to the Pt surface, and
(iii) a species where the aromatic rings are tilted
relative to the surface and the hydrogen atom in
a-position to the quinoline N has been abstracted
(a-quinolyl-like).[38,39]
Figure 3. Schematic representation of CD-1 interaction over
an ideally flat Pt surface in the enantiodifferentiating step.[31]
The strongly adsorbed p-bonded species (Figure 3) was
The p-bonded quinoline rings of CD and the two carbonyl inferred to be the crucial species for catalysis, in
groups of 1 lie approximately parallel to Pt; the quinuclidine
part is above the surface.
agreement with the previous model suggestions. The
two tilted species adsorb more weakly and are consid-
ered as spectator species since they are not involved in
the enantiodifferentiating step. We assume that upon
applying CD-QD mixtures CD will dominantly adopt
stage of the reaction, resulting in a weak adsorption and the adsorption mode with the quinoline rings lying
replacement of saturated NPE by the more stable and nearly parallel to Pt, while the weaker adsorbing QD
strongly adsorbing cinchona alkaloid.
will prevalently adsorb as a tilted species. This arrange-
An explanation for the non-linear behavior shown in ment minimizes the overall adsorption free energy and
Figure 2, which would be based on the different allows the control of enantiodifferentiation by the
resistance of the alkaloids against hydrogenation, can stronger adsorbing alkaloid, CD.
be excluded. In a control experiment we repeated the
Compared to adsorption on Pt, CD is more weakly
hydrogenation on Pt/Al2O3 in the absence of 1 and in the bound on Pd under hydrogenation conditions, as
presence of CD or QD alone (method A with half evidenced very recently by a comparative ATR-IR
amount of alkaloid). A complete conversion of the study of CD adsorption on both metals.[40] Also, the
quinoline moiety was observed by UV-Vis and NMR relative stability of the p- and N-lone pair-bonded
analysis after about 45 min for both alkaloids. Satura- species is different for the two metals, with the p-bonded
tion of the vinyl groups was much faster but this species being relatively more stable on Pt. The weaker
transformation barely influences the enantioselec- adsorption of CD in the p-bonded mode on Pd than on
tion.[33,34]
Pt may explain why typically a higher modifier to
Preliminary studies of the adsorption of CD and QD reactant ratio is needed in Pd-catalyzed enantioselective
on Pt/Al2O3 by UV-Vis analysis indicated that CD hydrogenations to minimize the faster non-modified
adsorbs stronger on Pt by a factor of approximately 2
reaction.[21 23,41] As concerns the mechanism of the
2.5, though this value is distorted by the slow hydro- hydrogenation of 3 (Scheme 2), we assume that in the
genation of the alkaloids during the in situ adsorption enantiodifferentiating step the cinchona alkaloid and 3
study. At first sight, the relatively small difference in adsorb via p-bonding of the aromatic rings approxi-
adsorption strength cannot account for the striking non- mately parallel to the Pd surface, but the real nature of
linearity shown in Figure 2. To find a feasible explan- substrate-modifier interaction(s) is not clarified yet.[23,42]
ation we have to consider the possible adsorption modes
Beside the different adsorption strength of cinchona
of substrate and modifier on the metal surface and their alkaloids, a further point to be addressed is the various
interaction during hydrogen uptake. According to the electronic and steric effects among the adsorbed species.
mechanistic models, developed for the hydrogenation of An illustration to the importance of these interactions is
[35]
1 over Pt by Wells et al.
and our group,[31,36] the shown in Figure 4. Hydrogenation of 3 over Pd/TiO2 was
cinchona alkaloid adsorbs via the quinoline rings lying carried out with CD-CN mixtures. Only small deviations
parallel to the Pt surface (p-bonded ™anchoring moi- related to the calculated ee were observed and, inter-
ˆ
ety∫). The substrate 1 adsorbs also flat via the C O estingly, in this reaction CN dominated in the mixture.
bonds parallel to the Pt surface (Figure 3). In this The ™expected∫ ee was calculated assuming that the
position the quinuclidine N-atom interacts with the molar ratios of alkaloids in solution and on the Pd
carbonyl O-atom of 1 via an H-bond. In AcOH this surface are identical, and the reaction rates and ees are
interaction corresponds to the 1-protonated CD inter- linear combinations of those measured with CD or CN
action,[37] in apolar medium the complex resembles the alone. In fact, a minimum in rate was observed in the
half-hydrogenated state of 1.[35]
reactions where alkaloid mixtures were applied. A
At least three differently adsorbed species could be plausible explanation may be that mixtures of CD and
identified by in situ ATR-studies of CD adsorption on CN occupy more surface Pd sites than any of them alone
platinum:
and this effect reduces the overall reaction rate. Besides,
Adv. Synth. Catal. 2003, 345, 255 260
257

Wolf-R¸diger Huck et al.
FULL PAPERS
(70 100 bar) CD afforded the highest ee of 97.5%.[13,14]
The likely explanation is that among the cinchona
alkaloids CD adsorbs strongest and can best compete
with hydrogen for the surface Pt sites. Repeating the
hydrogenation at different pressures with all four
alkaloids (Scheme 1) we found that by lowering the
hydrogen pressure the efficiency of CD decreased,
whereas that of QD and QN increased. At 5 bar QN
afforded 98% ee to (R)-2 and at 1 bar 96% (Scheme 1).
At atmospheric pressure even QD is more effective than
CD (94% vs. 91% ee) an interesting fact that has been
overlooked as yet.
Another consequence of the observed non-linearity is
that due care is necessary when applying cinchona
derivatives as chiral modifiers. If traces of the unreacted
alkaloid are still present and they adsorb stronger than
the derivative, this impurity may distort the ee and lead
to incorrect mechanistic conclusions. For the same
reason, interpretation of high-throughput catalyst
screening experiments may be flawed when several
modifiers are applied simultaneously, as the adsorption
strength on the metal may differ considerably.
Finally, we have to emphasize the differences between
the classical non-linear effect (NLE) thoroughly inves-
tigated by Kagan and others[25 27] in homogeneous
asymmetric catalysis and the non-linearity discussed
here. The asymmetric amplification or depletion in
homogeneous catalysis is attributed to molecular inter-
actions (associations) between two enantiomers of the
auxiliary or ligand. In contrast, we assume that in
heterogeneous catalysis the non-linear behavior of
modifier mixtures (i.e., diastereomers or even chemi-
cally different compounds) is mainly due to differences
in the strength and geometry of their adsorption on Pt or
Pd.
Figure 4. Hydrogenation of 4-methoxy-6-methyl-2-pyrone (3)
and cyclohexene (5) over Pd/TiO2 in the presence of CD-CN
mixtures. (a): calculated (based on linear combination) and
measured enantioselectivity in the hydrogenation of 3; (b):
relative rate of substrate conversion in the hydrogenation of 3
(method D) and 5 (method E), related to the rate with CN
alone.
Conclusions
Different adsorption strength and adsorption mode
(geometry) of chiral modifiers may lead to strong non-
linear behavior in heterogeneous asymmetric hydro-
genations that are relevant from both synthetic and
mechanistic points of view. CD is considered as the most
effective chiral modifier for Pt, and also for Pd in several
reactions. However, our study revealed that QN and QD
might outperform CD when their weaker adsorption on
the metal surface is taken into account by an appropriate
selection of reaction conditions.
the modifier-modifier interactions may also affect the
substrate-modifier interaction and thus the ee. To
support this hypothesis we repeated the reactions and
replaced 3 by cyclohexene (5, Figure 4b). In contrast to
the enantioselective hydrogenation of 3, no special
alkaloid-substrate interaction is expected in the hydro-
genation of 5. The effect of using alkaloid mixtures on
the relative rate was smaller but a clear minimum was
observed at intermediate alkaloid concentrations. This
is a confirmation for the assumption that mixtures of the
two diastereomers CD and CN adsorb differently on Pd
than the diastereomers alone.
Experimental Section
A practically useful consequence of the present study
is that the different adsorption strength of cinchona
alkaloids has to be taken into account when selecting the
appropriate conditions for the reaction. For example, in
Materials
The commercially available cinchona alkaloids contained high
the hydrogenation of 1 under high pressure conditions amounts of impurities: CD (Fluka, 92%; impurities: 1% QN,
258
Adv. Synth. Catal. 2003, 345, 255 260

Non-Linear Effect of Modifier Composition on Enantioselectivity in Asymmetric Hydrogenation
FULL PAPERS
7% QD, determined by HPLCat Fluka), CN (Fluka, 85%;
impurity: 15% 10,11-dihydro-CN, determined by HPLC at
Fluka), QD (Fluka, >90%; impurity: <10% 10,11-dihydro-
QD, determined by HPLCat Fluka), and QN (Fluka, >95%;
impurity: <5% 10,11-dihydro-QN, determined by HPLCat
Fluka). Ethyl pyruvate (Fluka) was distilled before use. 4-
Methoxy-6-methyl-2-pyrone 3 was synthesized as described
earlier[43] and purified by sublimation in vacuum (410 K,
0.1 mbar), followed by double recrystallization from hexane.
The 5 wt % Pt/Al2O3 catalyst (Engelhardt 4759) was treated
with flowing H2 for 60 min at 400 8Cand cooled to room
temperature in H2 for 30 min. After flushing with Ar, the
catalyst was transferred to the reactor within 10 min. A 5 wt %
Pd/TiO2 was prepared by precipitation of Pd hydroxide onto
suspended TiO2.[44] Small portions of the dry catalyst were
reduced in an H2 flow at room temperature for 30 min and
stored in air until use.
[7] P. B. Wells, A. G. Wilkinson, Top. Catal. 1998, 5, 39 50.
[8] T. Sugimura, T. Catal. Surv. Jpn. 1999, 3, 37 42.
[9] T. Osawa, T. Harada, O. Takayasu, Top. Catal. 2000, 13,
155 168.
[10] A. Tungler, G. Fogassy, J. Mol. Catal. A 2001, 173, 231
247.
[11] M. von Arx, T. Mallat, A. Baiker, Topics Catal. 2002, 19,
75 87.
[12] G.-Z. Wang, T. Mallat, A. Baiker, Tetrahedron: Asym-
metry 1997, 8, 2133 2140.
[13] X. Zuo, H. Liu, M. Liu, M. Tetrahedron Lett. 1998, 39,
1941 1944.
[14] B. Tˆrˆk, K. Felfˆldi, G. Szakonyi, K. Balazsik, M.
Bartok, Catal. Lett. 1998, 52, 81 84.
[15] M. Sch¸rch, N. K¸nzle, T. Mallat, A. Baiker, J. Catal.
1998, 176, 569 571.
[16] N. K¸nzle, A. Szabo, M. Sch¸rch, G. Z. Wang, T. Mallat,
A. Baiker, Chem. Commun. 1998, 1377 1378.
[17] M. Studer, S. Burkhardt, H.-U. Blaser, Chem. Commun.
1999, 1727 1728.
Catalytic Hydrogenations
The reactions were carried out in a magnetically stirred 100 mL
glass reactor. A magnetically stirred stainless steel autoclave
equipped with a 100 mL glass liner and a PTFE cap was used
for the high-pressure experiments. In a standard procedure, the
catalyst was prereduced in the solvent in flowing H2 for 5 min,
at 1 bar and room temperature. Then the modifiers were added
in 1 mL solvent. After 5 min preadsorption the reaction was
started by introducing the reactant.
¡
¬
[18] B. Tˆrˆk, K. Felfˆldi, K. Balazsik, M. Bartok, Chem.
Commun. 1999, 1725 1726.
[19] M. von Arx, T. Mallat, A. Baiker, Tetrahedron: Asym-
metry 2001, 12, 3089 3094.
[20] M. von Arx, T. Mallat, A. Baiker, Catalysis Lett. 2002, 78,
267 271.
[21] Y. Nitta, K. Kobiro, K. Chem. Lett. 1996, 897 898.
[22] K. Borszeky, T. Mallat, A. Baiker, Tetrahedron: Asym-
metry 1997, 8, 3745 3753.
Method A: 0.5 mL (4.1 mmol) 1, 20 mg 5 wt % Pt/Al2O3,
3.4 mmol alkaloid or alkaloid mixture, 10 mL AcOH, 1 bar,
25 8C.
[23] W. R. Huck, T. Mallat, A. Baiker, New J. Chem. 2002, 26,
Method B: As method A but at 5 bar and 15 8C.
6
8.
Method C: As method A but with 2 mL (16.5 mmol) 1.
Method D: 50 mg (0.35 mmol) 3, 20 mg 5 wt % Pd/TiO2, 10 mL
2-propanol, 3.4 mmol alkaloid or alkaloid mixture, 25 8C, 1 bar.
Method E: As method D but with 1.2 mmol cyclohexene (5).
Conversion and enantioselectivity were determined by an HP
6890 gas chromatograph using a Chirasil-DEX CB column
(Chrompack). An HP-5 column was applied for the analysis of
cyclohexene/cyclohexane mixtures.
¬
[24] M. Sutyinszki, K. Szˆri, K. Felfˆldi, M. Bartok, M. Catal.
Commun. 2002, 3, 125 127.
[25] D. Guillaneux, S. H. Zhao, O. Samuel, D. Rainford, H. B.
Kagan, J. Am. Chem. Soc. 1994, 116, 9430 9439.
[26] M. Avalos, R. Babiano, P. Cintas, J. L. Jimenez, J. C.
Palacios, Tetrahedron: Asymmetry 1997, 8, 2997 3017.
[27] C. Girard, H. B. Kagan, Angew. Chem.-Int. Edit. 1998,
37, 2923 2959.
[28] K. E. Simons, P. A. Meheux, A. Ibbotson, P. B. Wells,
Stud. Surf. Sci. Catal. 1993, 75, 2317 2320.
Acknowledgements
¬
¬
[29] A. Tungler, K. Fodor, T. Mathe, R. A. Sheldon, Stud.
Surf. Sci. Catal. 1997, 108, 157 165.
[30] A. Tungler, Y. Nitta, K. Fodor, G. Farkas, T. Mathe, J.
Financial support by the Swiss National Science Foundation is
greatly acknowledged. Thanks are also due to T. B¸rgi for
valuable discussions.
¡
¬
Mol. Catal. A 1999, 149, 135 140.
[31] A. Baiker, J. Mol. Catal. A. 2000, 163, 205 220.
[32] M. Sch¸rch, T. Heinz, R. Aeschimann, T. Mallat, A.
Pfaltz, A. Baiker, J. Catal. 1998, 173, 187 195.
References
[33] H.-U. Blaser, H. P. Jalett, D. M. Monti, A. Baiker, J. T.
Wehrli, Stud. Surf. Sci. Catal. 1991, 67, 147 155.
[1] Y. Orito, S. Imai, S. Niwa, J. Chem. Soc. Jpn. 1979, 1118
1120.
[2] Y. Izumi, Adv. Catal. 1983, 32, 215 271.
[3] A. Tai, T. Harada, in Tailored Metal Catalysts, (Ed.:
Y.Iwasawa), D. Reidel, Dordrecht, 1986.
[4] A. Baiker, J. Mol. Catal. A: Chem. 1997, 115, 473 493.
[5] H.-U. Blaser, H. P. Jalett, M. M¸ller, M. Studer, Catal.
Today 1997, 37, 441 463.
¬
¡
¬
[34] M. Bartok, G. Szˆllˆsi, K. Balazsik, T. Bartok, J. Mol.
Catal. A 2002, 177, 299 305.
[35] K. E. Simons, P. A. Meheux, S. P. Griffiths, I. M. Suther-
land, P. Johnston, P. B. Wells, A. F. Carley, M. K.
Rajumon, M. W. Roberts, A. Ibbotson, Recl. Trav.
Chim. Pays-Bas 1994, 113, 465 474.
[36] O. Schwalm, J. Weber, J. Margitfalvi, A. Baiker, J. Mol.
Structr. 1993, 297, 285 293.
[6] A. Pfaltz, T. Heinz, Top. Catal. 1997, 4, 229 239.
Adv. Synth. Catal. 2003, 345, 255 260
259

Wolf-R¸diger Huck et al.
FULL PAPERS
[37] O. Schwalm, B. Minder, J. Weber, A. Baiker, Catal. Lett.
1994, 23, 271 279.
[38] D. Ferri, T. B¸rgi, A. Baiker, Chem. Commun. 2001,
[41] T. Tarnai, A. Tungler, T. Mathe, J. Petro, R. A. Sheldon,
G. Toth, G. J. Mol. Catal. A. 1995, 102, 41 47.
[42] W.-R. Huck, T. Mallat, A. Baiker, A. Catal. Lett. 2002,
87 92.
1172 1173.
ƒ
[43] P. De March, M. Morena-Manas, R. Pi, I. Ripoll, F.
[39] D. Ferri, T. B¸rgi, J. Am. Chem. Soc. 2001, 123, 12074
¬
Sanchez-Ferrando, J. Heterocyclic. Chem. 1985, 22,
12084.
1537 1542.
[40] D. Ferri, T. B¸rgi, A. Baiker, J. Catal. 2002, 210, 160
[44] W. R. Huck, T. B¸rgi, T. Mallat, A. Baiker, J. Catal. 2001,
200, 171 180.
170.
260
Adv. Synth. Catal. 2003, 345, 255 260