Part III. COD versus NBD precatalysts. Dramatic difference in the asymmetric hydrogenation of prochiral olefins with five-membered diphosphine Rh-hydrogenation catalysts

Article abstract of DOI:10.1016/S0022-328X(00)00756-7

Induction periods in the asymmetric hydrogenation of prochiral olefins with five-membered chelates of the type [Rh(PP)(diolefin)]BF₄ originate from the parallel-running hydrogenation of the prochiral substrate and the diolefin that enters the s

Full text of DOI:10.1016/S0022-328X(00)00756-7

www.elsevier.nl/locate/jorganchem
Journal of Organometallic Chemistry 621 (2001) 89–102
Part III. COD versus NBD precatalysts. Dramatic difference in
the asymmetric hydrogenation of prochiral olefins with
five-membered diphosphine Rh-hydrogenation catalystsꢀ
Hans-Joachim Drexler, Wolfgang Baumann, Anke Spannenberg, Christine Fischer,
Detlef Heller *
Institut fu¨r Organische Katalyseforschung an der Uni6ersita¨t Rostock e.V., Buchbinderstraße 5/6, D-18055 Rostock, Germany
Received 12 September 2000; accepted 15 September 2000
Dedicated to Professor Henri Brunner on the occasion of his 65th birthday
Abstract
Induction periods in the asymmetric hydrogenation of prochiral olefins with five-membered chelates of the type
[Rh(PP)(diolefin)]BF4 originate from the parallel-running hydrogenation of the prochiral substrate and the diolefin that enters the
system as a constituent of the precatalyst. Reactivities towards the most commonly used diolefins COD or NBD can differ by
three powers of ten. X-ray crystal structure analyses of [Rh((S,S)-Et-DuPHOS)(NBD)]BF4 and of [Rh((R,R)-Et-
DuPHOS)(COD)]BF4 imply that a recently discussed relation between the sense of rotation of the diolefin in the precatalyst
(clockwise or anticlockwise twist) is not very likely. © 2001 Elsevier Science B.V. All rights reserved.
model substrate (Z)-N-acetylaminocinnamic acid re-
ported, the catalyst was formed in situ, according to
Brunner et al. [1d].
Complexes of rhodium, ruthenium, and more re-
cently of iridium, have been successfully employed for
the catalytic asymmetric hydrogenation of prochiral
substrates [1]. The use of these complexes is in no way
limited to academia, as is powerfully demonstrated by
the large scale preparation of the grass herbicide ‘Meto-
lachlor’, where hydrogenation catalysed by an iridium
complex forms the key synthetic step [2]. Catalyst pre-
cursors are usually employed in organometallic chem-
istry since it is seldom possible to add the actual active
species of a catalytic cycle directly into the reaction.
These precursors normally contain stabilising ligands,
such as COD ((Z,Z)-cycloocta-1,5-diene), which allow
the complexes to be easily handled. The precatalyst is
usually formed in situ, for example through the reac-
tion of a chiral bisphosphine with [Rh(COD)2]BF4. In
about half of the asymmetric hydrogenations of the
In the literature it was generally accepted that the
well known hydrogenation of the diolefins COD or
NBD (norborna-2,5-diene) with cationic rhodium(I)
complexes [3] takes place before the more interesting
asymmetric hydrogenation of the prochiral olefin. How-
ever, quantitative investigations of diolefin hydrogena-
tion with seven-membered chelate complexes of the
type [Rh(PP)(diolefin)]BF4 (PP is a chelating bisphos-
phine) [4] showed that the hydrogenation of the dio-
lefins usually employed (i.e. COD) took considerably
longer than generally presumed. In Fig. 1 the hydrogen
uptake and the COD conversion (determined by gas
chromatography, in parallel experiments) are plotted as
a function of time for various catalyst systems [5]. The
investigations confirm that the hydrogenation of the
prochiral olefin takes place parallel to that of the
diolefin. In particular, when DIOP (2,3-O-isopropyli-
den-2,3-dihydroxy-1,4-bis(diphenylphosphino)butan) is
used as the ligand, it was shown that COD is still
present in the solution under normal reaction condi-
ꢀ Part II: Kinetic investigations of the hydrogenation of diolefin
ligands in catalyst precursors for the asymmetric reduction of prochi-
ral olefins [4b].
* Corresponding author. Tel.: +49-381-4669348; fax: +49-381-
4669324.
E-mail address: detlef.heller@ifok.uni-rostock.de (D. Heller).
0022-328X/01/$ - see front matter © 2001 Elsevier Science B.V. All rights reserved.
PII: S0022-328X(00)00756-7

90
H.-J. Drexler et al. / Journal of Organometallic Chemistry 621 (2001) 89–102
Recently the mechanism of the hydrogenation of
NBD with [Rh(NBD)(PPh3)2]BF4 in CH2Cl2 has been
investigated extensively [11]. The rate constants for the
hydrogenation of the diolefins COD and NBD with the
five-membered rhodium complexes have not been de-
scribed previously in the literature; first results are
presented in this paper.
The difficulties described in the literature concerning
the hydrogenation of COD from the [Rh(DIPAMP)-
(COD)]BF4 complex (DIPAMP: 1,2-bis-(o-methoxy-
phenyl-phenyl-phosphino)ethane) in MeOH [12], are
confirmed by corresponding in situ NMR investigations
under stationary conditions [6]. Induction periods also
pose a problem for the five-membered chelates of Rh(I).
In the case of the hydrogenation of (Z)-N-acety-
laminocinnamic acid with the COD precatalyst, which
contains the NORPHOS ligand (bis-(diphenylphos-
phino)-bicyclo[2,2,1]hept-5-ene) [13], gas chromato-
graphic analysis proves that 22% COD remains after
two consecutive hydrogenations, each of which was
conducted in the presence of a hundred-fold excess of
the prochiral olefin (Fig. 2). Thus the COD contained
in the precatalyst is not fully hydrogenated after 205
min.
tions, after the hydrogenation of the prochiral olefin is
complete (precatalyst:substrate 1:100). According to
NMR investigations under stationary conditions this
COD quantity remains unchanged as precatalyst [6].
The parallel hydrogenation of the diolefin and the
prochiral olefin leads to a time dependent ‘blocking’ of
part of the catalyst. The rate of the asymmetric hydro-
genation increases as more catalyst becomes available
due to the advancing hydrogenation of the diolefin.
This results in a typical induction period [7], which is
clearly visible as a maximum in the rate profile (see Fig.
4). The induction period itself depends on many fac-
tors, which are discussed at length in Ref. [4a]. As well
as the relationship between the hydrogenation rates for
the diolefin and the prochiral olefin, the relationship
between the stability constants of the corresponding
substrate complexes is also of considerable importance,
with the latter being largely determined by the solvent.
This induction period, which has been mentioned
qualitatively by other authors [8], prevents a kinetic
evaluation of the hydrogenation [9] and is also reput-
edly the cause of non-linear effects [10].
Five-membered rhodium chelates induce a high enan-
tioselectivity in asymmetric hydrogenations due to their
conformationally rigid ligand backbone. Although the
low activity of these five-ring chelates is a disadvantage,
this can be compensated for by raising the pressure of
hydrogen used during the hydrogenation. In addition,
due to this lower activity of the five-membered chelate
complexes, the diolefins should be reduced before the
prochiral olefin and thus an induction period does not
play a role [1b].
Analogous induction periods are seen with other
rhodium catalysts which have typical five-membered
chelating ligands, such as CHIRAPHOS (2,3-bis-
(diphenylphosphino)butane) [14] or PROPHOS (1,2-
bis(diphenylphosphino)propane) [14b,15] (Fig. 3).
In the case of the hydrogenation of a hundred-fold
excess of methyl N-acetylaminoacrylate with the [Rh-
(PROPHOS)(COD)]BF4 complex, the solution still con-
tains over 80% COD! Appropriate induction periods
Fig. 1. Hydrogenation curves as well as the proportion of non-hydrogenated COD present in the precatalyst, COD in %. (a) [Rh(Ph-b-glup-
OH)(COD)]BF4/dimethyl itaconate; 78% ee (R) (Ph-b-glup-OH=phenyl-2,3-bis-(O-diphenylphosphino)-b- -glycopyranosid). (b) [Rh((R)-2-
D
MHOP)(COD)]BF4/(Z)-methyl-N-acetylaminocinnamate; 41% ee (R) (2-MHOP=1,4-bis(diphenylphosphino)-2-hydroxy-butane). (c)
[Rh((R,R)-DIOP)(COD)]BF4/dimethyl itaconate; 30% ee (S)). (a) 0.02 mmol precatalyst, otherwise standard conditions, see Section 1).

H.-J. Drexler et al. / Journal of Organometallic Chemistry 621 (2001) 89–102
91
Fig. 2. Hydrogenation of (Z)-N-acetylaminocinnamic acid with [Rh((R,R)-NORPHOS)(COD)]BF4 (standard conditions; a new charge of
substrate was added after the first hydrogenation was complete).
Fig. 3. Induction periods for asymmetric hydrogenations: (a) [Rh((R)-PROPHOS)(COD)]BF4/(Z)-methyl-N-benzoylaminocinnamate; (b)
[Rh((S,S)-CHIRAPHOS)(COD)]BF4/(Z)-N-acetylaminocinnamic acid; standard conditions.
are observed naturally also in other solvents as for
example THF or iso-PrOH.
ands of the BIPNOR type as well as BisP and its
relevant derivatives [21,22]. Due to the higher activity
of these systems, one would predict that the COD
introduced with the precatalyst could not be removed
completely before the asymmetric hydrogenation, and
indeed typical induction periods are observed, as is
expected.
In Fig. 4 various ways of conducting the hydrogena-
tion of (Z)-methyl-N-benzoylaminocinnamate using a
catalyst containing the Me-Duphos (1,2-bis(2,5-
dimethyl-phospholanyl)benzene) ligand are compared.
In the first case the catalyst is prepared in situ by the
A whole series of extremely active catalyst systems
containing chiral five-ring chelating ligands are known
in the literature for asymmetric hydrogenation. Exam-
ples are the bis-phospholane ligands of the BPE/
DuPHOS type [16], or the BASPHOS/ROPHOS type
[17], which are particularly notable since they can be
synthesised easily from cheap starting materials. Similar
ligands are based on phosphetanes (CnrPHOS), phosp-
holenes and phosphorous analogues of norbornane
(PennPHOS) [18–20]. Also worthy of mention are lig-

92
H.-J. Drexler et al. / Journal of Organometallic Chemistry 621 (2001) 89–102
reaction of [Rh(COD)2]BF4 with one equivalent of Me-
DuPHOS. In the second case the precatalyst is used
directly, in the form of [Rh(Me-DuPHOS)(COD)]BF4,
and in the final example the solvent complex [Rh(Me-
DuPHOS)(MeOH)2]BF4 is used (ca. 90 min pre-hydro-
genation of the precatalyst, NMR control). The hydrogen
uptake was recorded next to the characteristic rate profile
for the case of the catalyst prepared in situ. The results
from this experiment show that with equal enantioselec-
tivities in the first two cases, there was still free COD
present after completion of the asymmetric hydrogena-
tion.
Logically, there is more COD left after the completion
of asymmetric hydrogenation in the case where the
catalyst was prepared in situ. The example in Fig. 4
demonstrates very clearly that for asymmetric hydrogena-
tion under the usual reaction conditions only a part of
the usually expensive catalyst is required. It should be
stressed that the amounts of COD given in Fig. 4 are the
amounts left after the whole hydrogenation process. That
means during the asymmetric hydrogenation the concen-
tration of the COD complex is (time dependent) much
higher, and therefore the concentration of the active
catalyst is still lower as at the end. In addition the problem
of characterising the catalyst activity by means of the
‘half-life’ also becomes clear. Using the same catalysts,
the time for ‘half conversion’ of the substrate under
optimal conditions in the example is only about a quarter
of that for the reaction where the catalyst is formed in
situ.
[Rh((S,S)-Et-DuPHOS)(COD)]BF4
(Et-DuPHOS=
1,2-bis(2,5-diethyl-phospholanyl)benzene) was studied
in situ under stationary conditions using NMR spectro-
scopic methods. Since the reaction takes place too
quickly at 25°C for the use of the method described in
Ref. [6] (through improvements it is possible to follow
hydrogenation using a gas uptake rate of ca. 1 ml min−1
,
without diffusion having an appreciable influence), a re-
action temperature of ca. 13°C was chosen. The 31P-
NMR spectrum registered before exposure to hydrogen
(Fig. 5, bottom trace) exhibits besides the complex
[Rh(Et-DuPHOS)(COD)]BF4 (70.3 ppm, J(P,Rh)=148
Hz) minor amounts (less than 10%) of further, not un-
ambiguously identified (substrate complex) species. The
asymmetric hydrogenation starts with the introduction
of hydrogen. The new signals showing up on increasing
conversion from substrate to product are not due to the
major substrate complex (86.9 ppm, J(P,Rh)=163 Hz,
J(P,P)=34 Hz; 81.7 ppm, J(P,Rh)=153 Hz), and in the
end there are more species than the only expected solvent
complex [Rh(Et-DuPHOS)(CD3OD)2]BF4 (95.7 ppm,
J(P,Rh)=205 Hz) present. A stringent explanation for
these findings is not available at the moment [23]. Be-
sides dynamic (exchange) processes, the streaming hy
-drogen (gas bubbles) leads to a considerable broadening
of the resonance lines. Despite these difficulties it is
clearly visible that the concentration of the COD com-
plex decreases only slowly with the progress of the hy-
drogenation, and after its completion (63 min reaction
time with an enantioselectivity of 99.4% ee (S)) more
than half of the rhodium is still present as COD complex
[24]. This parallel-occurring hydrogenation of the
prochiral olefin and the diolefin corresponds qualita-
tively to the results presented in Figs. 2–4.
In order to prove that the hydrogenation of the diolefin
and the prochiral olefin also run parallel in the case of
the five-membered chelates, the asymmetric hydrogena-
tion of (Z)-methyl-N-benzoylaminocinnamate with
Fig. 4. Different methods for the asymmetric hydrogenation of (Z)-methyl-N-benzoylaminocinnamate with a catalyst containing the Me-DuPHOS
ligand; standard conditions for each experiment.

H.-J. Drexler et al. / Journal of Organometallic Chemistry 621 (2001) 89–102
93
Fig. 5. 31P-NMR spectroscopically monitored hydrogenation of (Z)-methyl-N-benzoylaminocinnamate with [Rh((S,S)-Et-DuPHOS)(COD)]BF4 at
ca. 13°C and continuous hydrogen supply (0.01 mmol precatalyst, 1.0 mmol prochiral olefin). The first and the last spectrum were recorded
without streaming hydrogen, which leads to a smaller half-width and hence a larger signal amplitude. The conversion of the prochiral olefin was
determined from the methyl singlet (3.83 ppm) in the 1H-NMR spectra.
Fig. 6. Catalytic hydrogenations of NBD with five-membered chelates of the type [Rh(PP)(NBD)]BF4; because of the high rate only 0.00875 mmol
of catalyst with PP=Et-DuPHOS was employed, otherwise standard conditions.
In order to compare different precatalysts with re-
spect to the size of the induction periods or the time
required for quantitative removal of the diolefin, the
rate constants for the hydrogenation of the diolefins
(k2diolefin=k%2diolefin[H2]) according to the following reac-
tion sequence were determined:
the isobaric conditions, the thus determined values still
contain the hydrogen concentration in solution at a
partial pressure of 0.84 atm H2 (total pressure of 1.0
atm reduced by the solvent’s vapour pressure, 0.16 atm
for MeOH according to Ref. [25]).
Examples for the catalytic NBD hydrogenation with
several catalysts are collected in Fig. 6. As observed
with seven-membered chelates [4a,b], the hydrogenation
of NBD proceeds very selectively and in the saturation
region of Michaelis–Menten kinetics.
k
1diolefin
[Rh(PP)(MeOH)2]++diolefin _` [Rh(PP)diolefin]
k
−1diolefin
k
[H ]
2
2diolefin
+ ꢀꢀꢀꢀꢀꢀꢁ monoolefin+([Rh(PP)(MeOH)2]+)
Therefore, the catalytic hydrogenations of the dio-
lefins were followed in the saturation region of the
underlying Michaelis–Menten kinetics, as described in
detail earlier [4a,b]. Dividing the slope of the linear
curves of hydrogen uptake by the initial catalyst con-
centration yields the desired rate constant. Because of
Regarding the complex with the ligand CHI-
RAPHOS, the well known phenomenon is observed
that the hydrogenation of the monoolefin to the alkane
is faster than the hydrogenation of the diolefin to the
monoolefin. The reason of this uncommon behaviour
(cf. Ref. [4a] for seven-membered chelates) is obviously

94
H.-J. Drexler et al. / Journal of Organometallic Chemistry 621 (2001) 89–102
the great stability of the diolefin complex. The
monoolefin is not being hydrogenated before complete
consumption of the diolefin.
leads to the conclusion that the stationary concentra-
tion of [Rh(PP)(diolefin)]BF4 equals the weight-in of
catalyst. Pseudo-stationary conditions with respect to
hydrogen concentration can be approximated by a
sufficiently large volume of the gas phase. Under these
conditions, straight lines are obtained for the hydro-
genation of the first double bond, like under normal
pressure. Taking into account the initial catalyst con-
centration, the slopes of these lines deliver the desired
rated constants. Fig. 8 presents typical pressure–time
curves for such experiments. The upper trace provides a
reference, the autoclave was charged exclusively with
hydrogen to probe the tightness of the system.
Hydrogenations of COD with five-membered
chelates are distinctively slower. Results from such
reactions, employing catalysts with the ligand Et-
DuPHOS or with the achiral DCPE (1,2-bis(dicyclo-
hexylphosphino)ethane) are presented in Fig. 7. It is
remarkable here that the selectivity of the COD hydro-
genation with the Et-DuPHOS catalyst is possibly
lower than for the other examples. Because of the very
slow reaction at normal pressure, experimental causes
for the deviation cannot completely be excluded. The
determination of the constant k2diolefin is, however, pos-
sible without problems from the initial part of the curve
(up to 10 h).
The catalytic hydrogenations of COD for typical
five-membered chelates such as CHIRAPHOS or
PROPHOS can no longer be measured meaningfully
under normal pressure, because they run even slower.
The respective reactions were performed under elevated
pressure in an autoclave (isochoric conditions) to ob-
tain at least tentative values for the rate of diolefin
hydrogenation. The kinetics of product formation for
the above-mentioned reaction sequence is described as
follows:
But an exact statement on the rate constant k%
2diolefin
is difficult, because there are different values for the
solubility of hydrogen in MeOH at 25.0°C in the litera-
ture. At a partial pressure of 1.0 atm, the solubility is
0.00331 mol l−1 according to Ref. [26], 0.00288 mol l−1
according to Ref. [27]. To make the data obtained at
different pressures comparable, the values determined
from autoclave experiments were converted to normal
pressure. Additionally, some rate constants already
known from normal-pressure experiments were deter-
mined under elevated pressure to check the method.
Results obtained for hydrogenations of COD, catalysed
by complexes of the ligands DIOP or DPPB (1,4-bis-
(diphenylphosphino)butane), coincide very well with
published results [4a,b].
d[monoolefin]
=k%2diolefin[[Rh(PP)(diolefin)]BF4][H2]
dt
(k2diolefin=k%2diolefin[H2])
(1)
The results are collected in Table 1 which lists,
besides the rate constants for the diolefin hydrogena-
tion, the chemical shifts of the diolefin-rhodium com-
The assumption of a very large stability constant for
the diolefin complexes of five-membered chelates [3d]
Fig. 7. Hydrogenations of COD with five-membered rhodium chelates containing the ligands Et-DuPHOS or DCPE; standard conditions.

H.-J. Drexler et al. / Journal of Organometallic Chemistry 621 (2001) 89–102
95
Fig. 8. Decay of pressure for [Rh(PP)(COD)]BF4-catalysed hydrogenations of COD in an autoclave at 25.0°C plotted over time; isochoric
conditions, see text.
plexes in the 31P-NMR spectra and their absorption
maxima in the UV–Vis spectra. Some achiral diphos-
phane ligands forming differently sized chelates are
included for comparison.
such complexes ideal model systems for systematic in-
vestigations of factors governing the selectivity such as
the ratio of intermediates and the ratio of their reactiv-
ities [30].
Comparing the reactivites of the respective NBD and
COD complexes of five-membered chelates in diolefin
hydrogenation reveals clearly greater differences than
found for the seven-membered chelates. Introducing the
cyclohexyl residue instead of the phenyl group (DCPE
versus DPPE) obviously has the opposite effect, com-
pared to seven-membered chelates.
All results may be summarised in the following way:
Regardless of chelate ring size and of the diolefin, the
stepwise hydrogenation of the double bonds is charac-
teristic. The diolefin is hydrogenated first, with a high
selectivity, and after its consumption, the monoolefin.
The largest differences between COD and NBD hy-
drogenation for the formerly studied seven-membered
chelates were found for bis-phosphinite ligands (Me-a-
glup or Ph-b-glup-OH [4b]). The achiral bis-phosphinite
DPOE fits this pattern.
A comparison of the seven-membered chelates DPPB
[4b] and DCPB shows that replacing the phenyl by
cyclohexyl groups (which increases the steric require-
ments of the substituents as well as the electron density
at rhodium) increases the ratio of the rate constants
The influence of the diolefin ligand on the induction
period is more pronounced with five-membered than
with seven-membered chelates, because of the clear
differences in reactivity for the diolefin hydrogenation.
The choice of the diolefin within the precatalyst and the
manner of conducting the experiment (Fig. 4) allow us
to ‘play’ with the macroscopically observable activity.
This is illustrated in Fig. 9 by the hydrogenation of
(Z)-methyl-N-acetylaminocinnamate with an Et-
DuPHOS catalyst, either with a COD or NBD complex
or with the solvent complex. As expected, one still finds
unchanged COD at the end of the asymmetric hydro-
genation, therefore only a fraction of the catalyst has
been used to generate the chiral product. NBD and
solvent complex exhibit similar activities due to the fast
NBD hydrogenation in the precatalyst. Whereas the
same enantioselectivity is observed in all cases, the
apparent activity is clearly different, see Fig. 9. The
time required for conversion of half of the substrate
with the COD complex is six times as long as that
required with the NDB complex, and if we consider the
k2NBD/k2COD. The same effect is observed when compar-
ing JaPHOS to Cyc-JaPHOS, a new and interesting
ligand type investigated by Beller et al. [28]. The latter
exhibits furthermore a significantly greater difference
for the hydrogenation rates of COD and NBD than
known so far for seven-membered chelates. The differ-
ence in reactivity between the COD and the NBD
complex with the Cyc-JaPHOS ligand already ap-
proaches the difference in reactivity between the
diastereomeric substrate complexes in the hydrogena-
tion of (Z)-methyl-N-acetylaminocinnamate with
[Rh(DIPAMP)(MeOH)2]BF4 catalyst [29]. This makes

96
H.-J. Drexler et al. / Journal of Organometallic Chemistry 621 (2001) 89–102
Table 1
Rate constants k2diolefin for the hydrogenation of the complexes [Rh(PP)(diolefin)]BF4 in MeOH at 25.0°C and 1.0 atm total pressure, chemical
shifts and absorption maxima from UV–Vis spectra (k2diolefin values as peudo-constants still contain the hydrogen solubility!)

H.-J. Drexler et al. / Journal of Organometallic Chemistry 621 (2001) 89–102
97
Fig. 9. Asymmetric hydrogenations of (Z)-methyl-N-acetylaminocinnamate with catalysts bearing the (S,S)-Et-DuPHOS ligand (standard
conditions): (a) [Rh((S,S)-Et-DuPHOS)(MeOH)2]BF4, (b) [Rh((S,S)-Et-DuPHOS)(NBD)]BF4, (c) [Rh((S,S)-Et-DuPHOS)(COD)]BF4.
system (Z)-N-acetylaminocinnamic acid/PROPHOS
catalyst, the analogous factor is about 20!
Bond lengths and angles are similar to those of the
already published structure of [Rh((S,S)-Me-
DuPHOS)(COD)]SbF6 [16c]. The deviation from the
The reasons for these drastic differences in reactivity,
especially with five-membered chelates, are unclear, just
as they are unknown for the diastereomeric substrate
complexes involved in asymmetric hydrogenation. Ac-
cording to the major/minor concept developed by
Halpern–Landis and Brown, the minor substrate com-
plex dominates the selectivity, despite its lower concen-
tration, by its extraordinary reactivity. There are no
X-ray crystallographic studies on minor substrate com-
plexes, so we decided to take a closer look at the
structures of the diolefin complexes, which differ in
reactivity in a similar manner.
An important feature of all published X-ray crystallo-
gaphic data of diolefin complexes is that they do not
have an ideal square-planar coordination geometry
[16a,c,o,31]. Analogous investigations of seven-mem-
bered chelates revealed that the more reactive NBD
complexes show generally a less pronounced distortion
from the square-planar towards a tetrahedral ligand
sphere than the corresponding COD complexes do [32].
A detailed discussion is in preparation.
The following discussion concerns X-ray crystal struc-
ture analyses of the complexes [Rh((R,R)-Et-
DuPHOS)(COD)]BF4 and [Rh((S,S)-Et-DuPHOS)-
(NBD)]BF4, whose reactivity in diolefin hydrogenation
differs by three powers of ten. Relevant parameters are
given in Table 2, graphic representations in Figs. 10 and
11. The unit cell of the NBD complex contains two
independent molecules, the unit cell of the COD com-
plex only one.
square-planar towards
a tetrahedral coordination
sphere is characteristic. The dihedral angles between the
planes defined by P, Rh, P and CꢂC, Rh, CꢂC (the cen-
troids of the double bonds are used) are 6.1° and 15.0°
for the NBD complex, thus they are clearly smaller than
for the COD complex (21.1°, Fig. 12). An angle of 18°
has been described for the analogous complex
[Rh((S,S)-Me-DuPHOS)(COD)]SbF6, and 15.6° for
[Ir((R,R)-Me-DuPHOS)(COD)]BF4 [16c,33]; a further
discussion is given in Ref. [16o]. This is the most promi-
nent difference in the solid-state structure of the com-
plexes, whose hydrogenation activity towards COD and
NBD is distinguished by about three orders of magni-
tude. This shows that, at least for the diolefins under
consideration, a higher reactivity in hydrogenation for
five-membered chelates is also associated with a smaller
deviation from the square-planar ligand sphere [34].
The distortion of the expected square-planar structure
towards a tetrahedron has the consequence that the di-
olefin may be oriented either clockwise twist or anticlock-
wise twist (in this way chiral complexes may result even
with an achiral bisphosphane). Recently, the question
was raised whether a clockwise twist or an anticlockwise
twist arrangement of the diolefins COD or NBD in the
precatalyst would result in the formation of S or R
enantiomers, respectively, in asymmetric hydrogena-
tions [31f,33]. Regarding the structures of the complexes
[Rh(Et-DuPHOS)(diolefin)]BF4 one finds (Fig. 12) that
the direction of the twist is formally the same, for both

98
H.-J. Drexler et al. / Journal of Organometallic Chemistry 621 (2001) 89–102
the NBD and the COD complex. But the complexes
differ in the chirality of the Et-DuPHOS ligand (S,S
with NBD, R,R with COD), so it is to be expected that
the twist would be opposite if the bisphosphine with the
same chirality were present.
results from a complex interdependence of single fac-
tors. The enantioselectivity, as a kinetic phenomenon, is
governed by the ratio of the reactivites of the
diastereomeric substrate complexes and by the concen-
tration ratio of these intermediates.
Another example showing that the direction of the
twist depends on the diolefin itself is the precatalyst
formed with the ligand SKEWPHOS (2,4-bis-
(diphenylphosphino)pentane, Fig. 13) [35]. Also in the
case of the complex with the ligand (S)-Cyc-PRO-
Summing up, it has to be stated that induction
periods are of relevance to the asymmetric hydrogena-
tion of prochiral olefins with five-membered chelates of
the type [Rh(PP)(diolefin)]BF4 as well [37]. These induc-
tion periods are recognised by a distinct increase in
activity during the asymmetric hydrogenation and orig-
inate from the parallel-running hydrogenation of the
prochiral substrate and the diolefin that enters the
system as a constituent of the precatalyst, as was al-
ready described for seven-membered chelates. A certain
amount of the expensive catalyst is thus unavailable for
the intended asymmetric hydrogenation, because it is
blocked by the diolefin. This characteristic feature
could be proven by in situ NMR spectroscopy under
stationary conditions for some examples. Surprisingly it
turned out that reactivities towards the most commonly
used diolefins COD or NBD might differ by three
powers of ten. This is primarily due to the very slow
hydrogenation of COD. The reason for these great
differences in reactivity remains unclear. The herein
presented X-ray crystal structure analyses of [Rh((S,S)-
PRAPHOS
((S)-2,3-O,N-bis(diphenylphosphino)-1-
(naphthoxy)-2-hydroxy-3-cyclohexylaminopropane) the
NBD complex shows a clockwise twist (2.2°), the COD
complex shows a anticlockwise twist (5.2°) [36].
All of these examples confirm that the chirality found
in the product of an asymmetric hydrogenation is
hardly to be correlated with the direction of the diolefin
twist. Instead, the macroscopically observed selectivity
Table 2
Crystallographic
and
refinement
data
for
[Rh((S,S)-Et-
DuPHOS)(NBD)]BF4 and [Rh((R,R)-Et-DuPHOS)(COD)]BF4
[Rh((S,S)-Et-Du-
[Rh((R,R)-Et-Du-
PHOS)(NBD)]+BF4− PHOS)(COD)]+BF−4
Empirical formula
Formula weight
Crystal system
C29H44BF4P2Rh
644.30
Monoclinic
P21
C30H48BF4P2Rh
660.34
Orthorhombic
P212121
Et-DuPHOS)(NBD)]BF4
and
of
[Rh((R,R)-Et-
DuPHOS)(COD)]BF4 reveal a significantly greater
deviation from square-planar coordination for the less
reactive COD complex. Whether it is possible to gener-
alise this trend remains yet unanswered. The crystal
structures now solved imply further that a recently
discussed relation between the sense of rotation of the
diolefin in the precatalyst (clockwise or anticlockwise
twist) is not very likely, for this direction may depend
on the diolefin itself, if the same chiral ligand backbone
is considered.
An important practical consequence of the induction
periods is that the frequently used COD complexes do
not imply optimal activity in the catalysis, at least for
five-membered chelates, and cannot therefore be re-
garded very economically.
Space group
Unit cell dimensions
,
a (A)
8.957(2)
19.566(4)
17.449(3)
90
101.96(3)
90
10.620(2)
16.882(3)
17.362(3)
90
90
90
,
b (A)
,
c (A)
h (°)
i (°)
k (°)
3
,
V (A )
2991.6(10)
1.431
3112.8(10)
1.409
Dcalc (Mg m−3
)
Z
4
0.72
1336
4
0.69
1376
v(Mo–Ka) (mm−1
F(000)
)
Crystal size (mm)
Temperature (K)
q Range for data
collection (°)
0.2×0.4×0.4
200
1.58–22.42
0.2×0.4×0.6
200
1.68–22.50
Index range (h,k,l)
−9/9, −20/18,
−18/18
−11/11, −18/18,
−18/7
Reflections collected 12973
11146
4079
1. Experimental
Independent
reflections
7285
Observed reflections 4565
3295
343
0.0428
0.0549
0.0967
0.966
1.1. General
Refined parameters
R1 (2|(I))
667
0.0452
Reactions were performed under argon using stan-
dard Schlenk techniques.
R1 (all data)
0.0772
wR2 (all data)
Goodness of fit
Flack x parameter
(esd)
0.0877
0.799
−0.0846 (0.0373)
0.0023 (0.0543)
1.2. Hydrogenations
Largest difference
0.65/−0.53
0.64/−0.46
The experiments under normal pressure and isobaric
conditions were carried out with an automatically regis-
trating gas measuring device described in Ref. [4c].
peak and hole
−1
,
(e A
)

H.-J. Drexler et al. / Journal of Organometallic Chemistry 621 (2001) 89–102
99
Fig. 10. Perspective view and numbering scheme of the two complexes [Rh((S,S)-Et-DuPHOS)(NBD)]BF4 in the asymmetric unit. All hydrogens
,
except for the asymmetric carbon atoms have been omitted for clarity. Interatomic distances (A): Rh1–C1=2.243(10), Rh1–C2=2.250(11),
Rh1–C4=2.211(11), Rh1–C5=2.198(11), Rh1–P1=2.297(3), Rh1–P2=2.276(3), Rh2–C31=2.254(11), Rh2–C32=2.222(12), Rh2–C34=
2.145(17), Rh2–C35=2.230(10), Rh2–P3=2.288(3), Rh2–P4=2.277(3). Intramolecular angles (°): P1–Rh1–P2=84.53(10), C1–Rh1–C2=
35.5(4), C4–Rh1–C5=35.7(4), P3–Rh2–P4=84.74(11), C31–Rh2–C32=35.6(4), C34–Rh2–C35=35.3(4).
Hydrogen (AGA 6.0) was used as received. The experi-
ments were carried out under standard conditions with
0.01 mmol catalyst, 1.0 mmol of substrate in 15.0 ml
solvent at 25.0°C. If it was necessary to hydrogenate
the precatalysts, the prochiral olefin in 1.0 ml of solvent
was fused in a glass ampulla under strictly anaerobic
conditions. To start the hydrogenation after completion
of the diolefin hydrogenation and thermic equilibration
(vapour pressure and H2 solubility), the ampulla was
destroyed by the magnetic stirrer.
(COD)]BF4} and Professor A. Bo¨rner {[Rh(2-
MHOP)(COD]BF4, [Rh(DIOP)(COD]BF4, [Rh(NOR-
PHOS)(COD]BF4, [Rh(DPOE)(COD]BF4}.
1.5. Analysis
GC was performed with a GC 5890 Serie II equipped
with FID, carrier gas argon (1 ml min−1). HPLC was
1.3. Substrates
(Z)-N-acetylaminocinnamic acid, methyl-N-acetyl-
aminoacrylate and dimethyl itaconate are commer-
cially available. (Z)-methyl-N-acetylaminocinnamate
and (Z)-methyl-N-benzoylaminocinnamate where pre-
pared by published procedures. (Z,Z)-Cycloocta-1,5-di-
ene (COD) and norborna-2,5-diene (NBD) were dried
with CaH2 and distilled under Ar.
1.4. Catalysts
The rhodium complexes of the commercially avail-
able ligands CHIRAPHOS, PROPHOS, Me-Duphos,
Et-Duphos, DPPE, DCPE, DCPB and DPPP, and of
the new ligands JaPHOS and Cyc-JaPHOS provided by
Professor M. Beller [28], were prepared by usual proce-
dures [38]. The other used complexes were kindly do-
nated by Professor R. Selke {[Rh(Ph-b-glup-OH)-
Fig. 11. Perspective view and numbering scheme of the complex
[Rh((R,R)-Et-DuPHOS)(COD)]BF4. All hydrogens except for the
asymmetric carbon atoms have been omitted for clarity. Interatomic
,
distances (A): Rh1–C1=2.236(7), Rh1–C2=2.297(6), Rh1–C5=
2.190(6), Rh1–C6=2.224(7), Rh1–P1=2.291(2), Rh1–P2=
2.262(2). Intramolecular angles (°): P1–Rh1–P2=85.32(6),
C1–Rh1–C2=34.3(3), C5–Rh1–C6=36.2(2).

100
H.-J. Drexler et al. / Journal of Organometallic Chemistry 621 (2001) 89–102
NMR spectra were recorded with a Bruker ARX 400
instrument (B0=9.4 T) in methanol-d4 solutions at 297
K, unless indicated otherwise. 31P chemical shifts are
given in ppm relative to 85% H3PO4 (ext.).
1.6. Crystal structure determinations for [Rh((S,S)-Et-
DuPHOS)(NBD)]BF4 and [Rh((R,R)-Et-
DuPHOS)(COD)]BF4
Crystals for the X-ray analyses were obtained by
slow diffusion of diethyl ether (NBD complex), tert-bu-
tyl-methyl ether (COD complex), respectively, into a
concentrated MeOH, THF, respectively, solution of the
respective rhodium complex. Crystal data and details of
the structure solution are summarised in Table 2.
For both compounds data were collected on a STOE-
IPDS diffractometer using graphite monochromated
Mo–Ka radiation. The structures were solved by direct
methods (SHELXS-97) [39] and refined by full-matrix
least-square techniques against F2 (SHELXL-97) [40]. XP
(Siemens Analytical X-ray Instruments, Inc.) was used
for structure representations.
Fig. 12. Elevation views of the X-ray crystal structures of the diolefin-
RhPP fragment of the [Rh(Et-DuPHOS)(diolefin)]BF4 complexes,
showing the dihedral angle between the P–Rh–P plane and the
centroid–Rh–centroid plane: (a) [Rh((S,S)-Et-DuPHOS)(NBD)]BF4
with the molecules A and B, anticlockwise twist; (b) [Rh((R,R)-Et-
DuPHOS)(COD)]BF4, anticlockwise twist.
The non-hydrogen atoms were refined anisotropi-
cally. The hydrogen atoms were placed in theoretical
positions and were refined by using the riding model.
The
weighting
schemes
are
ꢀ=1/[|2(F2o)+
(0.0290P)2+0.0000P] for [Rh((S,S)-Et-DuPHOS)-
(NBD)]BF4 and ꢀ=1/[|2(F2o)+(0.0562P)2+0.0000P]
for [Rh((R,R)-Et-DuPHOS)(COD)]BF4 with P=(F2o+
2F2c)/3.
2. Supplementary material
Crystallographic data (excluding structure factors)
for the structures reported in this paper have been
deposited at the Cambridge Crystallographic Data Cen-
tre as supplementary publication CCDC No. 147960
for [Rh((S,S)-Et-DuPHOS)(NBD)]BF4 and CCDC No.
147961 for [Rh((R,R)-Et-DuPHOS)(COD)]BF4. Copies
of the data can be obtained free of charge on applica-
tion to The Director, CCDC, 12 Union Road, Cam-
bridge, CB2 1EZ, UK (Fax: +44-1223-336033; e-mail:
deposit@ccdc.cam.ac.uk or www: http://www.ccdc.
cam.ac.uk).
Fig. 13. Parts of the structure of (a) [Rh((S,S)-SKEWPHOS)-
(COD)]ClO4, anticlockwise twist 17.6° and 3.4°, and (b) [Rh((S,S)-
SKEWPHOS)(NBD)]ClO4, clockwise twist 8.7° [35]. Crystallographic
data are from the Cambridge Crystallographic Data Centre (JEBRAJ
or JEBREN).
performed with a Liquid Chromatograph 1090 series II
equipped with DAD (Hewlett–Packard).
The conversion of COD and NBD was determined
by GC (HP 1, 50 m×0.2 mm ID, 85°C). (Z)-N-acetyl-
aminocinnamic acid was esterified with trimethylsilyl-
diazomethane before the GC measurements. Determi-
nation of ee were carried out by GC or HPLC on chiral
stationary phases. GC: (Z)-methyl-N-acetylaminocin-
Acknowledgements
namate (165°C, XE-60-
L
-valin-tert-butylamide, 10 m×
Financial support of the Deutsche Forschungsge-
meinschaft (BA 1911/1-1) and the Fonds der Chemis-
chen Industrie is greatefully acknowledged. We wish to
thank Professor A. Bo¨rner, Professor M. Beller and
Professor R. Selke for the donation of the catalysts or
ligands and Mrs C. Pribbenow and Mrs S. Buchholz
for skilled technical assistance.
0.2 mm ID); N-methyl-acetylaminoacrylate (120°C,
Chirasil-Val, 25 m×0.25 mm ID, Alltech); dimethyl
itaconate (120°C, FS-Lipodex E, 25 m×0.25 mm ID,
Machery-Nagel). HPLC: (Z)-methyl-N-benzoylamino-
cinnamate (column Chiralcel OD-H, 4.6×250 mm ID
(Merck), eluent n-hexane–ethanol 90:10).

H.-J. Drexler et al. / Journal of Organometallic Chemistry 621 (2001) 89–102
101
References
[13] (a) H. Brunner, W. Pieronczyk, Angew. Chem. 91 (1979) 655;
Angew. Chem. Int. Ed. Engl. 18 (1979) 620. (b) H. Brunner, W.
Pieronczyk, B. Schonhammer, K. Streng, I. Bernal, J. Korp,
Chem. Ber. 114 (1981) 1137. For problems encountered with
RENORHOS refer to [31b]. A hydroxy derivative of NORPHOS
is described in J. Ward, A. Bo¨rner, H.B. Kagan, Tetrahedron:
Asymm. 3 (1992) 849.
[14] (a) M.D. Fryzuk, B. Bosnich, J. Am. Chem. Soc. 99 (1977) 6262.
(b) M.D. Fryzuk, B. Bosnich, J. Am. Chem. Soc. 100 (1978)
5491.
[15] Since the partial reaction order for hydrogen, for the asymmetric
reaction as well as for the diolefin hydrogenation, is one [4c]
analogous results will also be obtained with other pressures.
[16] (a) M.J. Burk, J.E. Feaster, R.L. Harlow, Organometallics 9
(1990) 2653. (b) M.J. Burk, J. Am. Chem. Soc. 113 (1991) 8518.
(c) M.J. Burk, J.E. Feaster, W.A. Nugent, R.L. Harlow, J. Am.
Chem. Soc. 115 (1993) 10125. (d) M.J. Burk, S. Feng, M.F.
Gross, W. Tumas, J. Am. Chem. Soc. 117 (1995) 8277. (e) M.J.
Burk, M.F. Gross, J.P. Martinez, J. Am. Chem. Soc. 117 (1995)
9375. (f) M.J. Burk, Y.M. Wang, J.R. Lee, J. Am. Chem. Soc.
118 (1996) 5142. (g) M.J. Burk, M.F. Gross, T.G.P. Harper, C.S.
Kalberg, J.R. Lee, J.P. Martinez, Pure Appl. Chem. 68 (1996)
37. (h) M.J. Burk, J.G. Allen, W.F. Kiesmann, J. Am. Chem.
Soc. 120 (1998) 657. (i) J. Albrecht, U. Nagel, Angew. Chem. 108
(1996) 444; Angew. Chem. Int. Ed. Engl. 35 (1996) 407. (j) M.J.
Burk, C.S. Kalberg, A. Pizzano, J. Am. Chem. Soc. 120 (1998)
4345. (k) S.D. Debenham, J. Cossrow, E.J. Toone, J. Org.
Chem. 64 (1999) 9153. (l) M.J. Burk, K.M. Bedingfield, W.F.
Kiesman, J.G. Allen, Tetrahedron Lett. 40 (1999) 3093. (m) T.A.
Stammers, M.J. Burk, Tetrahedron Lett. 40 (1999) 3325. (n)
M.J. Burk, N.B. Johnson, J.R. Lee, Tetrahedron Lett. 40 (1999)
6685. (o) M.J. Burk, A. Pizzano, J.A. Martin, Organometallics
19 (2000) 250.
[17] (a) J. Holz, D. Heller, R. Stiirmer, A. Bo¨rner, Tetrahedron Lett.
40 (1999) 7059. (b) J. Holz, M. Quirmbach, U. Schmidt, D.
Heller, R. Stu¨rmer, A. Bo¨rner, J. Org. Chem. 63 (1998) 8031. (c)
W. Li, Z. Zhang, D. Xiao, X. Zhang, Tetrahedron Lett. 40
(1999) 6701. (d) Y.-Y. Yan, T.V. RajanBabu, J. Org. Chem. 65
(2000) 900. (e) W. Li, Z. Zhang, D. Xiao, X. Zhang, J. Org.
Chem. 65 (2000) 3489. (f) Y.-Y. Yan, T.V. RajanBabu, Org.
Lett. 2 (2000) 199.
[18] (a) A. Marinetti, J.P. Geneˆt, S. Jus, D. Blanc, V. Ratovelo-
manana-Vidal, Chem. Eur. J. 5 (1999) 1160. (b) A. Marinetti, F.
Labrue, J.P. Geneˆt, Synlett. 12 (1999) 1975. (c) A. Marinetti, S.
Jus, J.P. Geneˆt, Tetrahedron Lett. 40 (1999) 8365. (d) A.
Marinetti, S. Jus, J.P. Geneˆt, L. Ricard, Tetrahedron 56 (2000)
95.
[1] (a) J.M. Brown, Hydrogenation of functionalized carbon–car-
bon double bonds. In: E.N. Jacobsen, A. Pfaltz, H. Yamamoto
(Eds.), Comprehensive Asymmetric Catalysis, Springer, Berlin,
1999, Chapter 5.1, pp. 121–182. (b) U. Nagel, J. Albrecht,
Topics Catal. 5 (1998) 3. (c) M.J. Burk, F. Bienewald, Unnatural
a-amino acids, via asymmetric hydrogenation of enamides. In:
M. Beller, C. Bolm (Eds.), Transition Metals for Organic Syn-
thesis, vol. I, Wiley-VCH, Weinheim, 1998, Chapter 1.1.2, pp.
13–25. (d) H. Brunner, Hydrogenation. In: B. Cornils, W.A.
Herrmann (Eds.), Applied Homogeneous Catalysis with
Organometallic Compounds, vols. I and II, VCH, Weinheim,
1996, Chapter 2.2, pp. 201–219. (e) R. Noyori, Asymmetric
Catalysis in Organic Synthesis, Wiley, New York, 1994, Chapter
2, pp. 16–94. (f) P.A. Chaloner, M.E. Esteruelas, F. Joo, L.A.
Oro, Homogeneous Hydrogenation, Kluwer Academic Publish-
ers, Dordrecht, 1994. (g) H. Takaya, T. Ohta, R. Noyori,
Asymmetric hydrogenation. In: I. Ojima (Ed.), Catalytic Asym-
metric Synthesis, VCH, New York, 1993, Chapter 1, pp. 1–39.
[2] (a) F. Spindler, H.U. Blaser, Enantiomer 4 (1999) 557. (b) H.U.
Blaser, M. Studer, Appl. Catal. A 189 (1999) 191. (c) H.U.
Blaser, H.P. Buser, H.P. Jalett, B. Pugin, F. Spindler, Synlett. S1
(1999) 867. (d) H.U. Blaser, H.P. Buser, K. Coers, R. Hanreich,
H.P. Jalett, E. Jelsch, B. Pugin, H.D. Schneider, F. Spindler, A.
Wegmann, Chimia 53 (1999) 275. (e) H.-U. Blaser, F. Spindler,
Topics Catal. 4 (1997) 275.
[3] (a) J.R. Shapley, R.R. Schrock, J.A. Osborn, J. Am. Chem. Soc.
91 (1969) 2816. (b) R.R. Schrock, J.A. Osborn, J. Am. Chem.
Soc. 93 (1971) 3089. (c) R.R. Schrock, J.A. Osborn, J. Am.
Chem. Soc. 98 (1976) 2134. (d) R.R. Schrock, J.A. Osborn, J.
Am. Chem. Soc. 98 (1976) 4450.
[4] (a) D. Heller, K. Kortus, R. Selke, Liebigs Ann. (1995) 575. (b)
D. Heller, S. Borns, W. Baumann, R. Selke, Chem. Ber. 129
(1996) 85. (c) D. Heller, J. Holz, S. Borns, A. Spannenberg, R.
Kempe, U. Schmidt, A. Borner, Tetrahedron: Asymm. 8 (1997)
213.
[5] cis-Cyclo-octene (COE) and/or cyclo-octane (COA) as well as
norbornene (NBE) and/or norbornane (NBA) resulted as hydro-
genation product when COD or NBD was used.
[6] W. Baumann, S. Mansel, D. Heller, S. Borns, Magn. Reson.
Chem. 35 (1997) 701.
[7] Substrate inhibition as the cause for the observed induction
period can be ruled out. The induction periods also cannot be
caused by neglecting to add additional hydrogen. For a sub-
strate/precatalyst ratio of 100, there is an error of 2% expected
for the complete hydrogenation of the olefin to the saturated
product distributed over the entire gas uptake curve.
[19] F. Bienewald, L. Ricard, F. Mercier, F. Mathey, Tetrahedron:
Asymm. 10 (1999) 4701.
[8] (a) U. Nagel, T. Krink, Angew. Chem. 105 (1993) 1099; Angew.
Chem. Int. Ed. Engl. 32 (1993) 1052. (b) U. Nagel, A. Bublewitz,
Chem. Ber. 125 (1992) 1061. (c) U. Nagel, E. Kinzel, J. Andrade,
G. Prescher, Chem. Ber. 119 (1986) 3326.
[9] D. Heller, R. Thede, D. Haberland, J. Mol. Catal. A 115 (1996)
273.
[10] D.G. Blackmond, T. Rosner, T. Neugebauer, M. Reetz, Angew.
Chem. 111 (1999) 2333; Angew. Chem. Int. Ed. Engl. 38 (1999)
2196. 1:1 mixtures of the diastereomeric catalysts led to unex-
pectedly high enantioselectivities for the hydrogenation of
dimethyl itaconate. These are the results of the various induction
periods for the diastereomeric COD employed catalysts, so that
the end effect is that the hydrogenation at first runs via a
diastereomeric catalyst.
[20] Q. Jiang, Y. Jiang, D. Xiao, P. Cao, X. Zhang, Angew. Chem.
110 (1998) 1203; Angew. Chem. Int. Ed. Engl. 37 (1998) 1100.
[21] (a) F. Robin, F. Mercier, L. Ricard, F. Mathey, M. Spagnol,
Chem. Eur. J. 3 (1997) 1365. (b) F. Mathey, F. Mercier, F.
Robin, L. Ricard, J. Organomet. Chem. 577 (1998) 117.
[22] (a) T. Imamoto, J. Watanabe, Y. Wada, H. Masuda, H. Ya-
mada, H. Tsuruta, S. Matsukawa, K. Yamaguchi, J. Am. Chem.
Soc. 120 (1998) 1635. (b) T. Miura, T. Imamoto, Tetrahedron
Lett. 40 (1999) 4833.
[23] Very recently Gridnev and Imamoto have described product
complexes for similar systems. I.D. Gridnev, N. Higashi, K.
Asakura, T. Imamoto, J. Am. Chem. Soc. 122 (2000) 7183.
[24] The results obtained in the NMR tube correspond approxi-
mately to those determined in the standard hydrogenation vessel
at 15.0°C (total reaction time 45 min, 99.5% ee (5) 48% COD).
[25] T.E. Jordan, Vapor Pressure of Organic Compounds, Inter-
science, New York, 1954.
[11] M.E. Esterulas, J. Herrero, M. Martin, L.A. Oro, V.M. Real, J.
Organomet. Chem. 599 (2000) 178.
[12] (a) J.M. Brown, R.A. Chaloner, Tetrahedron Lett. 21 (1978)
1877. (b) J.M. Brown, P.A. Chaloner, J. Am. Chem. Soc. 102
[26] P.G.T. Fogg, W. Gerrard, Solubility of Gases in Liquids, Wiley,
Chichester, 1991.
(1980) 3040.
.

102
H.-J. Drexler et al. / Journal of Organometallic Chemistry 621 (2001) 89–102
[27] P. Lu¨hring, A. Schumpe, J. Chem. Eng. Data (1989) 250.
[28] M. Beller, R. Jackstell, unpublished results.
[29] C.R. Landis, J. Halpern, J. Am. Chem. Soc. 109 (1987) 1746.
[30] H. Neumann, K. Wannowius, W. Baumann, R. Thede, D.
Heller, submitted for publication.
[31] (a) R.G. Ball, N.C. Payne, Inorg. Chem. 16 (1977) 1187. (b) E.P.
Kyba, R.E. Davis, P.N. Juri, K.R. Shirley, Inorg. Chem. 20
(1981) 3616. (c) M.P. Anderson, L.H. Pignolet, Inorg. Chem. 20
(1981) 4101. (d) J. Oliver, D.P. Riley, Organometallics 2 (1983)
1032. (e) M.J. Burk, J.E. Feaster, R.L. Harlow, Tetrahedron:
Asymm. 2 (1991) 569. (f) S.K. Armstrong, J.M. Brown, M.J.
Burk, Tetrahedron Lett. 34 (1993) 879. Reviews on structures of
chiral rhodium complexes are found in J.M. Brown, P.L. Evans,
Tetrahedron 44 (1988) 4905 or H. Brunner, A. Winter, J. Breu,
J. Organomet. Chem. 553 (1998) 285.
phino)ethane) are still the more reactive species but also exhibit
greater distortions from the square-planar arrangement (R.
Kempe, A. Spannenberg, D. Heller, Z. Kristallogr. (in press). R.
Kempe, A. Spannenberg, H.-J. Drexler, D. Heller, Z. Kristal-
logr. (in press)). The role of the cyclohexyl ligand in such
situations is still unexplored.
[35] J. Bakos, I. Toth, B. Heil, G. Szalontai, L. Parkanyi, V. Fu¨lu¨p,
J. Organomet. Chem. 370 (1989) 263.
[36] (a) R. Kempe, A. Spannenberg, D. Heller, Z. Kristallogr. 213
(1998) 636. (b) A. Spannenberg, H.-J. Drexler, D. Heller, Z.
Kristallogr. (in press).
[37] The problem of induction periods is by no means restricted to
asymmetric hydrogenation, but is a general phenomenon in
homogeneous catalysis whenever precatalysts are used. This has
been shown for the photocatalysed [2+2+2] cycloaddition of
alkynes with nitriles, for instance: B. Heller, D. Heller, G.
Oehme, J. Mol. Catal. A 110 (1996) 211.
[32] D. Heller, Habilitationsschrift, Emst-Moritz-Arndt-Universita¨t,
Greifswald, 1998.
[33] B.F.M. Kimmich, E. Somsook, C.R. Landis, J. Am. Chem. Soc.
120 (1998) 10115.
[34] It must be mentioned here that the NBD complexes of ligands
with cyclohexyl instead of phenyl residues (DCPB 1,4-bis(dicy-
clohexylphosphino)butane, DCPE 1,2 bis(dicyclohexylphos-
[38] R.R. Schrock, J.A. Osborn, J. Am. Chem. Soc. 93 (1971) 2397.
[39] G.M. Sheldrick, SHELXS-97, University of Go¨ttingen, Germany,
1997.
[40] G.M. Sheldrick, SHELXL-97, University of Go¨ttingen, Germany,
1997.
.