We now report that the catalysis over palladium surfaces
shows a more diverse chemistry than was hitherto suspected and
that, depending on the choice of solvent/substituent, each
alkaloid may give either enantiomer in excess.
is crucial in determining the stereochemical outcome of the
reaction.
The authors thank the Catalysis Initiative of the EPSRC
(J. A. I. and R. W.) and Johnson Matthey (P. B. W.) for financial
support and the provision of studentships (P. J. C., T. J. H and
J. A. S.).
Pt and Pd colloids stabilized with KD1, a proprietary
protecting agent, were prepared by a metal vapour synthesis
11
route § and used as catalysts in the enantioselective hydro-
genation of EP.1 ¶ The Pt colloids gave the expected
direction of enantioselectivity and rate enhancement on the
basis of the known chemistry of supported platinum catalysts,
i.e. CD and CN modification gave (R)- and (S)-lactate,
respectively (Table 1, entries 3–5). The KD1 stabilizing agent
was found to have only a marginal effect on the performance of
the catalysts (Table 1, compare entries 3 and 4 with 6 and 7).
Surprisingly however, when modified by addition of a butan-
2,13
Notes and References
†
‡
E-mail: iggo@liverpool.ac.uk
Current address: Johnson Matthey PLC, Orchard Road, Royston, Herts,
UK SG8 5HE
Colloidal metal solutions were prepared by metal vapour deposition in a
Torrovap reactor. Typically 2.5 g of KD1, a proprietary protecting agent
ICI), was melted and dispersed onto the walls of the 5 l spherical chamber,
§
(
27
after which the chamber was evacuated to 10 Torr and cooled in liquid
nitrogen. 11.3 mmol of Pd was then evaporated over 2 h and co-condensed
2
-one (MEK) solution of CD, the palladium colloids also gave
3
(R)-lactate with ees of up to 30% in the hydrogenation of EP
entry 8), i.e. in the opposite sense to that previously reported
with 145 cm of butan-2-one (MEK) which was simultaneously added to the
(
vessel as vapour. After deposition the chamber was warmed to room
temperature during which period a further 2.5 g KD1 in 60 cm of MEK was
6–8
3
for CD-modified Pd catalysts (e.g. entry 24). In contrast to
reactions using oxide-supported Pd catalysts these reactions
also appear to show rate enhancement, the rate of the
unmodified reaction being very slow (compare entries 8 and 9).
The enantioselectivity induced by CN was more muted, in
agreement with previous reports that CN is a less effective
modifier than CD, and was in favour of (S)-lactate (entry 10).
This is the first report of palladium emulating platinum in the
sense of the enantioselectivity induced in pyruvate ester
hydrogenation.
added. Pt colloids were prepared by an analogous procedure. High
resolution transmission electron microscopy showed the Pd colloids
typically to have a particle size distribution ranging from 1.75 to 4.75 nm,
with a maximum at 2.5 nm. After use in catalysis the particle size had
increased to 2 to 9 nm with a maximum at 4 nm.
¶
A solution of the colloid containing 1 mg of Pd, modifier (6 mg, 2.0 3
1025 mol), freshly distilled ethyl pyruvate (2.7 mg, 0.024 mol) (Aldrich)15
and solvent/substituent (total reaction volume: 20 ml) was added to a Parr
4592 autoclave. The autoclave was sealed, purged three times with H
bar and then pressurized to 70 bar with H and maintained at 25 ± 1 °C for
h.
2
to 50
2
1
Previous studies of oxide-supported palladium catalysed
reactions have used EtOH rather than MEK as the solvent6,8 or
∑ The reduced catalyst was first immersed in a solution of the alkaloid in the
appropriate solvent and the slurry was then stirred in air for about 1 h before
the catalyst was loaded into the high pressure reactor.
7
have used the methyl rather than the ethyl ester as reactant. In
order to investigate the possibility of unprecedented solvent
and/or substituent effects in Orito-type reactions, several oxide-
supported palladium catalysts were modified either in situ or by
1
P. B. Wells and A. G. Wilkinson, Top. Catal., in the press.
14
2 A. Baiker, J. Mol. Catal. A,, 1997, 115, 473.
the Orito procedure ∑ in EtOH, THF and MEK and used as
3
H-U. Blaser, H. P. Jalett, M. Muller and M. Studer, Catal. Today, 1997,
7, 441.
catalysts in the hydrogenation of ethyl and methyl pyruvates.
3
6–8
All these catalysts gave the expected enantioselectivity, i.e.
in the opposite sense to that observed for platinum catalysts, CD
giving (S)- and CN (R)- product, and showed a reduced rate in
MP hydrogenation, (entries 11–17). However, when used in
ethyl pyruvate hydrogenation, CD modified catalysts gave
4
5
6
R. L. Augustine and S. K. Tanielyan, J. Mol. Catal. A, 1997, 118, 79.
G. Bond and P. B. Wells, J. Catal., 1994, 150, 329.
H-U. Blaser, H. P. Jallett, D. M. Monti, J. F. Reber and J. T. Wehrli,
Stud. Surf. Sci. Catal., 1988, 41, 153.
7 H-U. Blaser, H. P. Jallett, M. Muller and M. Studer, Catal. Today, 1997,
37, 441.
(
R)-lactate in substantial ee and CN modified catalysts
S)-lactate, (entries 18–23), i.e. in the opposite sense to that
8
9
0
T. J. Hall, P. Johnston, W. A. H. Vermeer, S. R. Watson and P. B. Wells,
Stud. Surf. Sci. Catal., 1996, 101, 221.
T. Mallat, S. Szabo, M. Schurch, U. W. Gobel and A. Baiker, Catal.
Lett., 1997, 47, 221.
(
6–8
previously reported. These results originate in independent
studies in Liverpool (entries 18 and 19) and Hull (entries
2
0–23). Thus, this work is not only the first report of Pd
1
1
I. M. Sutherland, A. Ibbotson, R. B. Moyes and P. B. Wells, J. Catal.,
following Pt in the sense of the observed enantioselectivity, but
is also the only report, for either platinum or palladium, of the
direction of the enantioselectivity being solvent and/or sub-
stituent dependent.
1
990, 125, 77.
1 R. W. Devenish, T. Goulding, B. T. Heaton and R. Whyman, J. Chem.
Soc., Dalton Trans., 1996, 673.
12 J. U. Kohler and J. S. Bradley, Catal. Lett., 1997, 45, 203.
13 H. Bonnemann and G. A. Braun, Angew. Chem., Int. Ed. Engl., 1996,
The mechanistic pathway of the palladium-catalysed reaction
is not well-understood. Deuterium labelling experiments have
shown that, in contrast to the case with Pt, over Pd the main
product forming route is via the enol and carbon–carbon double
3
5, 1992; H. Bonnemann and G. A. Braun, Chem. Eur. J., 1997, 3,
1
200.
1
1
4 Y. Orito, S. Imai and S. Niwa, Nippon Kagaku Kaishi, 1979, 1118.
5 Impurities in the pyruvate reactant can have an effect on both the rate
and ee in these reactions. Ethyl pyruvate was therefore distilled daily.
The same batch of ethyl pyruvate was used in the comparative study.
Ethyl pyruvate supplied by Aldrich has been reported as suffering
fewest problems in this regard. H-U. Blaser, H. P. Jalett and F. Spindler,
J. Mol. Catal. A, 1996, 107, 85; T. Mallat, Z. Bodmer, B. Minder, K.
Borszelay and A. Baiker, J. Catal., 1997, 168, 183.
8
bond hydrogenation. Deuterium tracer experiments are in
progress to test whether the palladium catalysed reaction in
MEK also follows this mechanistic pathway.
These observations reveal that the palladium catalysed
enantioselective hydrogenation of pyruvate esters is a much
more complicated system than the analogous platinum-cata-
lysed reaction and that coadsorption of solvent/substituent
molecules on the Pd surface both during the modification
procedure and thereafter during enantioselective hydrogenation
Received in Cambridge, UK, 9th February 1998; revised manuscript
received, 18th May 1998; 8/03901A
1452
Chem. Commun., 1998
Collier, Paul J.
