144
C. de Bellefon et al. / Journal of Organometallic Chemistry 567 (1998) 143–150
catalysts and chiral ligands are often prepared at the
laboratory scale in quantities required for the estimation
of activities and enantioselectivity with some typical
substrates. Whenever more tests are needed, then more
catalyst is synthesised with possible risks of non repro-
ducible results and lost of time and money.
The picture is not the same in homogeneous catalysis
where liquid catalysts are most often used. How to design
a reactor that possess a liquid fixed bed? Liquid/liquid
and gas/liquid columns are well known and widely used
in industry [6]. However, the volume of such apparatus
is still too large (\2000 ml), even at the laboratory scale
for applications with expensive catalysts. This is mainly
due to the size effects. The droplets of the dispersed phase
are millimetric in size which will drive strong wall effects
when using small bore (B0.04 m) columns. In a previous
paper, we have reported that the centrifugal partition
chromatograph (CPC) is a convenient liquid/liquid plug
flow continuous reactor for the study of homogenous
catalysis [7].
In some cases, when catalysts (ligands) are readily
prepared or available, the chemist is still facing a
problem: which catalyst for which reaction ? Amazingly,
that question comes from one of the well recognised
advantages of homogeneous catalysis, i.e. its very high
selectivity. In homogeneous catalysis, a very high selec-
tivity (enantioselectivity) is generally obtained with a
specific ligand for a specific substrate or class of sub-
strate. This may be called the catalyst (ligand) to sub-
strate specificity. The reverse of the coin is the lack of
‘universal’ catalysts. When a catalyst is to be chosen for
an industrial production, the latter conclusion calls for
numerous catalytic tests for the selection of the ‘right’
catalyst. Here again, investment in time and money may
become prohibitive with the expensive catalysts that are
used.
Along with this paper, technical problems linked with
liquid/liquid catalysis, results obtained with the CPC and
future applications of the CPC for catalysts selection will
be presented and discussed.
2. Results and discussion
2.1. One (liquid) phase batch operations
Quantitative investigation of kinetics with homoge-
neous catalysts is also a field to be developped. Most of
the chemists working at the synthesis of homogeneous
catalysis and ligands have been educated and trained in
organic chemistry. In evaluating catalysts, they generally
pay more attention to the selectivity rather than to the
activity or possible deactivation. For example, in the field
of enantioselective catalysis, except for few examples [4],
the activity is seldom given as a result of a kinetic study
but rather as ‘a time to reach quantitative conversion’
and catalysts deactivation is almost never investigated.
These four general remarks, which mostly apply to
homogeneous enantioselective catalysis, point to the
need for new methods and apparatus for the study and
the selection of catalysts.
When dealing with monophasic catalytic reactions,
mixing during the reaction is not of tremendous impor-
tance (except for fast reactions with tRB10 s). Most of
homogeneous catalytic reactions are slow enough to
consider that diffusion is not the limiting process so that
the chemistry is the actual process under investigation.
In such cases, only mixing of the reaction mixture at the
early beginning of the reaction is important. Thus, a gas
phase chromatography vial may be used as a reactor
which both allows kinetic investigations and the full
automation of the sampling.
This technique has been used with a gas chro-
matograph equipped with an automatic sampler for
the kinetic investigation of the enantioselective reduc-
tion of acetophenone to 2-phenylethanol by isopro-
panol catalysed by a Rh/chiral diamine system (Eq. 1)
[8].
Looking to the methods that are used in the field of
gas/solid heterogeneous catalysis may provide some
guidelines to develop new methods and apparatus in
Thus, out of ca. 50 tests involving more than 1000
samples, 30 have been used for the kinetic study. Com-
plex kinetic laws have been confirmed (Eq. 1). The results
have not yet been published but suffice to say that the
observed complex kinetic order in acetophenone and the
inhibition by water and acetone confirms the mechanism
proposed for this reaction, [9] all being done with B0.12
g of the expensive diamine ligand L*n .[10].
liquid, liquid/liquid and gas/liquid/liquid homogeneous
catalysis. Fixed bed plug flow like G/S continuous
tubular reactors are largely used in the steady-state mode
for the kinetic and deactivation studies of heterogeneous
catalysts. Non steady-state or pulsed modes have also
been described in the case of chromatographic reactors
[5].