Asymmetrie Hydrogenation. Dimerization of Solvate Complexes: Synthesis and Characterization of Dimeric [Rh(DIPAMP)]2+2, a Valuable Catalyst Precursor

Article abstract of DOI:10.1021/om801199q

The hydrogenation of [Rh(DIPAMP)(NBD)]BF₄ (DIPAMP= l,2-bis[(2-methoxyphenyl)(phenylphosphino)]ethane, NBD = 2,5-norbornadiene) in methanol leads not only to the formation of the expected solvate complex [Rh(DIPAMP)(MeOH)₂]BF₄

Full text of DOI:10.1021/om801199q

Organometallics 2009, 28, 3673–3677 3673
DOI: 10.1021/om801199q
Asymmetric Hydrogenation. Dimerization of Solvate Complexes:
Synthesis and Characterization of Dimeric [Rh(DIPAMP)]22+, a Valuable
Catalyst Precursor
Angelika Preetz,† Wolfgang Baumann,† Christian Fischer,† Hans-Joachim Drexler,†
Thomas Schmidt,† Richard Thede,‡ and Detlef Heller*,†
†Leibniz Institute for Catalysis at the University of Rostock, Albert-Einstein-Strasse 29A, D-18059 Rostock,
Germany, and ‡Institute for Biochemistry, Felix-Hausdorff-Strasse 4, D-17489 Greifswald, Germany
Received December 19, 2008
The hydrogenation of [Rh(DIPAMP)(NBD)]BF4 (DIPAMP=1,2-bis[(2-methoxyphenyl)(phenyl-
phosphino)]ethane, NBD=2,5-norbornadiene) in methanol leads not only to the formation of the
expected solvate complex [Rh(DIPAMP)(MeOH)2]BF4 but also to the arene-bridged dimeric species
[Rh(DIPAMP)]2(BF4)2. The dimer is characterized by X-ray analysis as well as by an extensive NMR
1
solution study including H, 31P, and 103Rh NMR. When used as a precursor for the asymmetric
hydrogenation of prochiral olefins, [Rh(DIPAMP)]2(BF4)2 operates without induction periods. It can
be used for the formation of catalyst-substrate complexes such as [Rh(DIPAMP)(MAC)]BF4
(MAC=methyl (Z)-R-acetamidocinnamate).
Introduction
be analyzed by X-ray. Dimeric Rh-arene complexes with
monodentate P-ligands are reported as well.6 This work
will present the first X-ray structure and a detailed NMR
solution study of the dimeric Rh-arene complex with the
chiral ligand DIPAMP (DIPAMP=1,2-bis[(2-methoxyphe-
nyl)(phenylphosphino)]ethane).
In want of superior precatalysts for homogeneous catalysis,
the existence of dimeric rhodium-arene complexes with dipho-
sphines has been proven several times. Halpern et al. were
the first to describe the dimerization of the solvate complex
[Rh(DPPE)(MeOH)2]BF4 (DPPE=1,2-bis(diphenylphosphi-
no)ethane) to yield [Rh(DPPE)]22+ (1), in which each Rh atom
is bonded to two P atoms of a DPPE ligand and through π-
arene coordination to a phenyl ring of the DPPE ligand of the
second Rh atom.1,2 Further investigations on the DPPE system
as precatalyst for homogeneous hydroacylation were carried
out by Bosnich et al.3 An extensive NMR solution study of the
CYCPHOS (CYCPHOS =1,2-bis(diphenylphosphino)-1-cy-
clohexylethane) system was elaborated by Riley.4 Other sys-
tems containing dimeric Rh-diphosphine species have been
observed;5 however, so far the DPPE ligand remains the only
diphosphine of which crystals of the dimeric Rh species could
The DIPAMP7 ligand is one example of the successful
application of homogeneous catalysis on an industrial scale8
that has also experienced wide mechanistic investigations
from an academic viewpoint.9 The transformation of diolefin
complexes [Rh(DIPAMP)(diolefin)]+ (diolefin = NBD (2,
(6) (a) PPh3: Rifat, A.; Patmore, N. J.; Mahon, M. F.; Weller, A. S.
Organometallics 2002, 21, 2856–2865. (b) Marcazzan, P.; Ezhova, M. B.;
Patrick, B. O.; James, B. R. C. R. Chim. 2002, 5, 373–378. (c) BINAP-
HTHOXO: Gridnev, I. D.; Fan, C.; Pringle, P. G. Chem. Commun.
2007, 1319–1321. (d) For redox-switchable ligand, see for example:
Singewald, E. T.; Mirkin, C. A.; Stern, C. L. Angew. Chem., Int. Ed.
Engl. 1995, 35, 1624–1627.
(7) Vineyard, B. D.; Knowles, W. S.; Sabacky, M. J.; Bachman, G. L.;
Weinkauff, D. J. J. Am. Chem. Soc. 1977, 99, 5946–5952.
*Corresponding author. E-mail: detlef.heller@catalysis.de.
(1) (a) Halpern, J.; Riley, D. P.; Chan, A. S.; Pluth, J. J. J. Am. Chem.
Soc. 1977, 99, 8055–8057. (b) For the influence of arene complexes on the
catalytic activity of asymmetric hydrogenations see: Heller, D.; Drexler,
H.-J.; Spannenberg, A.; Heller, B.; You, J.; Baumann, W. Angew.
Chem., Int. Ed. 2002, 41, 777–780.
(2) Halpern’s X-ray structure of the DPPE dimer was not deposited in
the Cambridge database. A new X-ray crystal structure of the DPPE
dimer1 is included in the SupportingInformation, Figure S1, ofthis work.
(3) (a) Barnhart, R. W.; Wang, X.; Noheda, P.; Bergens, S. H.;
Whelan, J.; Bosnich, B. Tetrahedron 1994, 50, 4335–4346. (b) Fairlie,
D. P.; Bosnich, B. Organometallics 1988, 7, 946-954, and on the DPPP,
DPPB, and CHIRAPHOS.(c) Bergens, S. H.; Noheda, P.; Whelan, J.;
Bosnich, B. J. Am. Chem. Soc. 1992, 114, 2121–2128. (d) Fairlie, D. P.;
Bosnich, B. Organometallics 1988, 7, 936–945.
(8) Knowles, W. S. Asymmetric Hydrogenation-The Monsanto
L-Dopa Process. In Asymmetric Catalysis on Industrial Scale; Blaser, H.
U., Schmidt, E., Eds.;Wiley-VCH:NewYork, 2004;Chapter 1.1, pp 23-53.
(9) (a) Schmidt, T.; Drexler; Sun, A.; Dai, Z.; Baumann; Preetz, A.
Adv. Synth. Catal. 2009, 351, 750–754. (b) Preetz, A.; Baumann, W.;
Drexler, H.-J.; Fischer, C.; Sun, J.; Spannenberg, A.; Zimmer, O.; Hell,
W.; Heller, D. Chem. Asian J. 2008, 3, 1979–1982. (c) Schmidt, T.; Dai,
€
Z.; Drexler, H.-J.; Baumann, W.; Jager, C.; Pfeifer, D.; Heller, D.
Chem.;Eur. J. 2008, 14, 4469–4471. (d) Gridnev, I. D.; Imamoto, T.;
Hoge, G.; Kouchi, M.; Takahashi, H. J. Am. Chem. Soc. 2008, 130,
2560–2572. (e) Drexler, H.-J.; Baumann, W.; Schmidt, T.; Zhang, S.;
Sun, A.; Spannenberg, A.; Fischer, C.; Buschmann, H.; Heller, D.
Angew. Chem., Int. Ed. 2005, 44, 1184–1188. (f) Wada, Y.; Imamoto,
T.; Tsuruta, H.; Yamaguchi, K.; Gridnev, I. D. Adv. Synth. Catal. 2004,
346, 777–788. (g) Schmidt, T.; Baumann, W.; Drexler, H.-J.; Arrieta, A.;
Buschmann, H.; Heller, D. Organometallics 2005, 24, 3842–3848. (h)
McCulloch, B.; Halpern, J.; Thomas, M. R.; Landis, C. R. Organome-
tallics 1990, 9, 1392–1395. (i) Landis, C. R.; Halpern, J. J. Am. Chem.
Soc. 1987, 109, 1746–1754. (j) Brown, J. M.; Parker, D. J. Org. Chem.
1982, 47, 2722–2730. (k) Brown, J. M.; Chaloner, P. A. J. Chem. Soc.,
Chem. Commun. 1980, 344–346. (l) Brown, J. M.; Chaloner, P. A. J. Am.
Chem. Soc. 1980, 102, 3040–3048. (m) Brown, J. M.; Chaloner, P. A.
Tetrahedron Lett. 1978, 21, 1877–1881.
(4) Riley, D. P. J. Organomet. Chem. 1982, 234, 85–97.
(5) BINAP: (a) Miyashity, A.; Takaya, H.; Souchi, T.; Noyori, R.
Tetrahedron 1984, 8, 1245–1253. (b) Miyashity, A.; Yasuda, A.; Takaya, H.;
Toriumi, K.; Noyori, R. J. Am. Chem. Soc. 1980, 102, 7932–7934. (c) DIPH:
Allen, D. G.; Wild, S. B.; Wood, D. L. Organometallics 1986, 5, 1009–1015.
(d) SKEWPHOS: Mikami, K.; Yusa, Y.; Hatano, M.; Wakabayashi, K.;
Aikawa, K. Chem. Commun. 2004, 98-99.(e) BIPHEP: Mikami, K.;
Kataoka, S.; Yusa, Y.; Aikawa, K. Org. Lett. 2004, 6, 3699–3701.
r
2009 American Chemical Society
Published on Web 06/17/2009
pubs.acs.org/Organometallics

3674 Organometallics, Vol. 28, No. 13, 2009
Preetz et al.
Table 1. Pseudo-First-Order Rate Constants k0NBD for the
NBD Hydrogenation of [Rh(DIPAMP)(NBD)]BF4 in Different
Solvents
solvent
k0NBD in min-1
MeOH
EtOH
i-PrOH
THF
trifluoroethanol
propylene carbonate
ca. 9
13.2
5.9
5.0
3.9
4.8
5-norbornadiene), COD (1,5-cyclooctadiene)) into the solvate
complex [Rh(DIPAMP)(MeOH)2]+ has been investigated
qualitatively by the groups of Halpern and Brown.9i,9l,9m
Results and Discussion
The determination of pseudo-first-order rate constants
for the hydrogenation of several diolefin complexes of the
type [Rh(PP)(diolefin)]+ (PP=diphosphine) to the respec-
tive solvate complexes under isobaric conditions has been
described in detail.10
Figure 1. UV/vis reaction spectrum for the stoichiometric hy-
drogenation of [Rh(DIPAMP)(COD)]BF4 (0.008 mmol) in
15.0 mL of MeOH at 1.0 bar of total pressure (cycle time
10.0 min, overall time 800 min).
Applying the hydrogenation technique elaborately de-
scribed in ref 11, pseudo-first-order rate constants (k0NBD
kNBD ꢀ [H2]) result for the hydrogenation of [Rh(DIPAMP)
(NBD)]BF4 in MeOH, EtOH, THF, i-PrOH, trifluoroetha-
nol, and propylene carbonate as summarized in Table 1; see
also Supporting Information, Figures S2, S3.
very clear due to the very long reaction time of 800 min.
Analysis as a pseudo-first-order reaction with the program
SPECFIT (see Supporting Information, Figures S10, S11
contain extinction diagrams and the first-order fit for various
wavelengths) gives a pseudo rate constant of 0.0028 min-1
(0.0030 min-1 for a second measurement), which is even
smaller than for the diolefin complex with the achiral
ligand DPPE.10a
From the pseudo-first-order rate constant it followsthatthe
complete removal of COD of [Rh(DIPAMP)(COD)]BF4 in
MeOH at 25 °C and 1 bar of hydrogen pressure takes ca. 27 h,
which confirms the difficulties described in refs 9l and 9m.
Hydrogenation of either [Rh(DIPAMP)(NBD)]BF4 or
[Rh(DIPAMP)(COD)]BF4 in MeOH should result in the
solvate complex [Rh(DIPAMP)(MeOH)2]BF4 (2). The 31P
NMR spectrum of a solution of [Rh(DIPAMP)(MeOH)2]-
BF4 gained from the respective NBD complex, however, not
only exhibits the doublet expected for such a C2-symmetrical
ligand but shows further signals (see Supporting Informa-
tion, Figure S4).12
=
The time for complete removal of NBD in the precatalyst
can be calculated from the 7-fold half-life t1/2 for the pseudo-
first-order reaction (t1/2 = ln 2/k0), which corresponds to ca.
99.2% conversion of NBD. After this time practically no
more signals of diolefin complex [Rh(DIPAMP)(NBD)]BF4
should be detectable in the 31P NMR spectrum. To verify the
results, [Rh(DIPAMP)(NBD)]BF4 was hydrogenated in the
respective solvents for the time calculated from k0 values. In
all the cases, signals of the diolefin complexes were no longer
detectable; see Supporting Information, Figures S4-S9.
Independent of the solvent, the stoichiometric hydrogena-
tion of NBD in [Rh(DIPAMP)(NBD)]+ is remarkably faster
than the hydrogenation of COD in [Rh(DIPAMP)(COD)]
BF4 in MeOH, a fact that has been known for some time.9l,9m
A different approach to determine pseudo rate constants is
opened up by UV/vis spectroscopy. While for the cases
shown above catalytic hydrogenations were monitored by
hydrogen consumption, here stoichiometric hydrogenations
can be followed. This method is especially suitable for very
slow hydrogenations. In ref 10a application of UV/vis
spectroscopy was introduced for the determination of the
pseudo rate constant k0 for the hydrogenation of [Rh(DPPE)
(COD)]BF4 in MeOH (k0 = 0.0033 min-1).
From a highly concentrated solution (0.09 mol/L) of [Rh
((S,S)-DIPAMP)(MeOH)2]BF4 orange crystals of [Rh((S,
S)-DIPAMP)]22+ (3a) could be isolated and were analyzed by
X-ray, Figure 2, proving a dimeric structure.
This is in analogy with the DPPE dimer 1, which also
precipitates from a highly concentrated solution of the
solvate complex. As previously described in the literature
for DPPE and other ligands,1 the structure of 3a exhibits two
rhodium atoms that each coordinate to the chelate ligand
and to a phenyl ring of the DIPAMP ligand of the other
rhodium atom. The bite angle P-Rh-P (84.2°) does not
differ notably from that of the respective [Rh(DIPAMP)
(NBD)]BF4 (83.6-84.4°) and [Rh(DIPAMP)(COD)]BF4
(82.7°) complexes, whereas the Rh-P bond lengths are
The reaction spectrum for the stoichiometric hydrogena-
tion of COD in [Rh(DIPAMP)(COD)]BF4 in MeOH is
exhibited in Figure 1. In principle, two isosbestic points are
visible at 416 and 459 nm. However, the one at 416 nm is not
€
(10) (a) Preetz, A.; Drexler, H.-J.; Fischer, C.; Dai, Z.; Borner, A.;
˚
˚
slightly shortened (2.228 A versus 2.290-2.312 A (NBD)
Baumann, W.; Spannenberg, A.; Thede, R.; Heller, D. Chem.;Eur. J.
2008, 14, 1445-1451, and references therein. (b) Heller, D.; de Vries,
A. H. M.; de Vries, J. G. Catalyst Inhibition and Deactivation in
Homogeneous Hydrogenation. In Handbook of Homogeneous Hydro-
genation; de Vries, H. G., Elsevier, C., Eds.; Wiley-VCH: New York,
2007; Chapter 14, pp 1483-1516.
(11) Drexler, H.-J.; Preetz, A.; Schmidt, T.; Heller, D. Kinetics of
Homogeneous Hydrogenation: Measurement and Interpretation. In
Handbook of Homogeneous Hydrogenation; de Vries, H. G., Elsevier,
C., Eds.; Wiley-VCH: New York, 2007; Chapter 10, pp 1483-1516.
(12) A similar spectrum is gained in EtOH, i-PrOH, and propylene
carbonate, whereas in THF and trifluoroethanol there are no more
signals of a solvate complex observable, but only the unknown signals;
see Supporting Information, Figures S5-S9. In the 31P NMR spectrum
of [Rh(DIPAMP)(MeOH)2]BF4, with the widely investigated dimethyl
itaconate (Rh:substrate=1:1), the same signals are visible, which were
then left unidentified; see Supporting Information, Figure S12 and
ref 9c.

Article
Organometallics, Vol. 28, No. 13, 2009 3675
that five protons of a phenyl bridge and four protons of an
o-anisyl bridge can be found with high-field shifts of up to
4.3 ppm; see Supporting Information, Figures S18, S19.
Thus, the second species in solution is the unsymmetrical
dimeric species in which one phenyl and one o-anisyl ring
establish the arene bridge (3b). Dissolving single crystals of
3a in CD3OD at -30 °C revealed that already at this low
temperature the equilibrium between the two dimeric species
3a and 3b was established.14
Further analysis of the NMR spectra was accomplished
by shift correlation experiments including both 31P-103Rh
and 1H-103Rh HMQC NMR (Figure 3) as well as
1H-31P HMQC NMR (Figure S20). We find three different
rhodium environments. Correlation with the 31P NMR of
3a reveals that the signal at -837 ppm belongs to symme-
trical dimer 3a, Figure 3 (top). The other two signals at -822
and -766 ppm15 are both assigned to unsymmetrical 3b, thus
confirming the interpretation of the 1H NMR. Figure 3
(bottom) clearly shows the correlation of three of the arene
bridging protons to the respective rhodium cores.
Figure 2. X-ray structure of the cation in [Rh((S,S)-
DIPAMP)]2(BF4)2 MeOH (3a) (ORTEP, 30% probability
ellipsoids). Hydrogen atoms and the MeOH molecule are
3
˚
omitted for clarity. Selected distances [A] and angles [deg]:
Rh1-Rh1a=4.144, Rh1-P1=2.221(2), Rh1-P2=2.228(2),
Rh1-C=2.258-2.346(8), Rh1-centroid of phenyl rings =
1.843, P1-Rh1-P2=84.2(1).
As low-temperature NMR experiments have shown in
CD2Cl2, the two dimeric species 3a and 3b are in equilibrium;
that is, they can interconvert. In MeOH equilibration can
additionally be obtained via the solvate complex 2; dissolu-
tion of single crystals of 3a leads to a mixture of 2, 3a, and 3b.
These experimental findings lead to the reaction sequence
shown in Scheme 1.
At ambient temperature equilibration between dimers
3a and 3b is fast; already 1 min after dissolution of 3a in
CD2Cl2 the UV/vis spectrum is constant. In MeOH-d4 in the
31P NMR spectrum both dimers 3a and 3b can be detected
besides the solvate complex 2. The ratio of dimers 3a and
3b is constant, independent of the initial concentration
(dilution) of 3a.
13
˚
and 2.282 A (COD)). Redissolving the crystals of 3a in
CD3OD reproduces the 31P NMR spectrum gained by
hydrogenation of [Rh(DIPAMP)(NBD)]BF4 in CD3OD.
From the respective 1H NMR spectrum one can deduce that
there must be at least one other species in solution beside the
solvate complex and 3a. Both 3a and the third species exhibit
high-field protons characteristic for arene bridges.1,5,6 With
the structure of the DIPAMP ligand in mind other structural
motifs can be considered to be in equilibrium with the
phenyl-phenyl-bridged one (3a) that was crystallized. First,
an o-anisyl-o-anisyl-bridged symmetric species is possible,
and second, an unsymmetrical species in which one phenyl
and one o-anisyl ring establish the arene bridge is possible.
To identify the signals of the isolated crystals with the
phenyl-phenyl structure 3a, single crystals were dissolved at
-80 °C to freeze out any equilibrium to other possible
dimeric species.9e The solvent was changed to relatively
noncoordinating dichloromethane (CD2Cl2) due to the low
solubility of 3a in MeOH at low temperature and owing to
the low concentration of 3a and the additional species in
MeOH-d4 (equilibrium to the solvate complex in MeOH).
Accordingly, low-temperature NMR of 3a in CD2Cl2
afforded spectra of pure 3a. The 1H NMR spectrum in
CD2Cl2 at 213 K shows three characteristic high-field pro-
tons at 5.4, 6.1, and 6.5 ppm, which can be attributed to
protons of the phenyl ring of the DIPAMP ligand that
interact with a rhodium atom; see Supporting Information,
Figures S13-S17. In the 31P NMR of 3a at 213 K two signal
sets are visible: the one at 83 ppm showing four lines, the one
at 73 ppm broadened but probably four lines as well. Gentle
heating of the sample resulted in a slight upfield shift and
resolution of a complicated fine structure. Due to the higher
order of the spin system consisting of four 31P and two 103Rh
nuclei (AA0BB0XX0), there are clearly more than the eight
lines one would expect for a mononuclear C1-symmetric
rhodium complex (ABX system). Not until a temperature
higher than 0 °C is reached do additional signals appear.
1H NMR proves an additional species in solution that must
be unsymmetrical (four instead of two methoxy resonances).
1H-1H COSY NMR of the equilibrated solution confirms
Thus, from the NMR spectra equilibrium constants for the
formation of the dimeric species can be determined. The
gross stability constant as the sum of the individual stability
constants is calculated according to eq 1.
½3aꢁ þ ½3bꢁ
Kgross
¼
¼ K3a þ K3b
ð1Þ
2
½2ꢁ
The ratio of individual stability constants (K3a/K3b) di-
rectly corresponds to the concentration ratio of the two
dimers [3a]/[3b].
For the determination of concentrations of the solvate
complex and the dimers, 1H NMR data (signals of the
methoxy groups of the ligand) were used. In addition, the
signals of the protons establishing the arene bridge can be
used for verification of the ratio of [3a]/[3b], Supporting
Information, Figure S22.16
Four measurements result in an average gross stability
constant of 52 L/mol and stability constants of 34 L/mol for
3a and 18 L/mol for 3b. Hence, the phenyl-phenyl-bridged
dimer is twice as stable as the phenyl-o-anisyl-bridged one.
Those stability constants are consistent with values reported
by Halpern et al. for [Rh(DPPE)(benzene)]+ (18 L/mol) and
[Rh(DPPE)(toluene)]+ (97 L/mol).1a
(14) This temperature was chosen owing to the high viscosity of
MeOH and to the low solubility of 1 in MeOH at lower temperatures.
(15) The chemical shifts are in the range of known values for rhodium
DIPAMP arene complexes, e.g., [Rh(DIPAMP(p-xylene)]BF4 with -
956 ppm and [Rh(DIPAMP)(benzene)]BF4 with -1006 ppm; see ref 1b.
(16) 31P NMR data are not suitable due to the fact that signals of all
species overlap.
(13) Drexler, H.-J.; Zhang, S.; Sun, A.; Spannenberg, A.; Arrieta, A.;
Preetz, A.; Heller, D. Tetrahedron: Asymmetry 2004, 15, 2139–2150.

3676 Organometallics, Vol. 28, No. 13, 2009
Preetz et al.
Figure 3. 31P-103Rh{1H} (top) and 1H-103Rh{31P} (bottom) HMQC NMR spectra of [Rh(DIPAMP)]2(BF4)2 (3) in CD2Cl2 (298 K):
mixture of 3a [δ(103Rh) -837 ppm] and 3b [δ(103Rh) -822 and -766 ppm]. The signals encircled are due to η6-coordinated arenes.
Further details on acquisition and interpretation of these spectra are provided in the Supporting Information, Figure S21.
To demonstrate the applicability of the dimers 1 and 3 as
precatalysts, they were used in the asymmetric hydrogena-
tion of (Z)-R-acetamidocinnamic acid and methyl (Z)-
R-acetamidocinnamate (MAC). In Figure 4 the hydrogen
consumption is shown in comparison to that in which the
solvate complex [Rh(PP)(MeOH)2]BF4 was prepared in situ
and used as catalyst. Within the range of reproducibility the
hydrogenations result in the same activity (and enantioselec-
tivity for the chiral ligand DIPAMP, 96% ee) under identical
conditions and in that agree well with the literature data of
ref 9i. Obviously, the catalyst-substrate complexes are
rapidly formed from the respective dimer and the substrate
in solution.
The hydrogenation of MAC with the Rh/DIPAMP sys-
tem is one of the most widely investigated enantioselective
hydrogenations.9i Although the diastereomeric catalyst-
substrate complexes exhibit a high thermodynamic stability,
a catalyst-substrate complex could not yet be characterized
by X-ray analysis. Apparently, the substrate complexes are
too soluble in the applied solvent methanol.
Here, the dimeric species 3a offers an alternative approach.
Due to the fact that 3a already precipitates during the

Article
Organometallics, Vol. 28, No. 13, 2009 3677
with the expected coordination mode already observed for
other dehydroamino acid derivatives9e,9g,9h and dimethyl
itaconate9c with DIPAMP as ligand.
The addition of hydrogen from the rhodium site to the
coordinating substrate would formally lead to the (R) pro-
duct. The enantioselective hydrogenation of MAC in MeOH
at 25 °C under normal pressure with the (R,R)-DIPAMP
catalyst, however, yields the (S)-product in enantiomeric
excess (ee = 96%). Obviously, the isolated crystal corre-
sponds to the major substrate complex, which does not lead
to the main enantiomer. Hence, this result is a supplementary
confirmation of the milestone work of Landis and Halpern,
who derived this well-known major-minor-principle mainly
from kinetic investigations.9i
Experimental Section
All experiments were carried out under an argon atmosphere
using standard Schlenk techniques. NMR spectra were recorded
with a Bruker AV 400 NMR spectrometer. UV/vis spectra were
measured with a Lambda 19 (Perkin-Elmer).
The general procedure for the catalytic hydrogenation of
diolefins and prochiral olefins is described in refs 10 and 11.
[Rh(DIPAMP)(solvent)2]BF4. 31P NMR data: CD3OD, d,
δ 81.3 ppm, JP-Rh 208.1 Hz; EtOH (nondeuterated), d, δ 79.2,
Figure 4. Comparison of the asymmetric hydrogenation of (Z)-
R-acetamidocinnamic acid with [Rh(DPPE)(MeOH)2]BF4 (red)
and 1 (gray) and of MAC with [Rh((S,S)-DIPAMP)(MeOH)2]-
BF4 (green) and (S,S)-3a (blue). Conditions: 0.01 mmol of Rh;
1.0 mmol of prochiral olefin; 15.0 mL of MeOH; 25.0 °C; 1 bar
of total pressure.
J
P-Rh 209.0 Hz; i-PrOH-d8, d, δ 77.6, JP-Rh 210.8 Hz; propylene
carbonate (nondeuterated), d, δ 74.1, JP-Rh 207.2 Hz.
[Rh(DIPAMP)]2(BF4)2 (3). [Rh(DIPAMP)(NBD)]BF4 (200 mg,
0.27 mmol) was dissolved in 4 mL of MeOH. The flask was purged
with hydrogen and the solution stirred. After 20 min the hydrogen
was substituted by argon by freezing the solution and subsequent
evacuation (3 times). The volume of MeOH was concentrated to
approximately 1 mL. An orange precipitate resulted, which was
dissolved in the heat. Slow cooling to ambient temperature yielded
orange to red crystals that were washed with diethyl ether and dried
in vacuo (yield of 3: 134 mg, 76% isolated yield).
3a. 1H NMR (219 K, in CD2Cl2): 7.68-7.51 (10H, m); 7.39-
7.34 (2H, m); 7.32-7.29 (8H, m); 7.23 (2H, t, J=6.5 Hz); 7.15
(2H, t, J=7.6 Hz); 7.09 (2H, dd, J=8.3 Hz, 3.9 Hz); 6.92 (2H, dd,
J=8.6 Hz, 2.9 Hz); 6.79 (2H, t, J=7.3 Hz); 6.49 (2H, t, J=6.4
Hz); 6.08 (2H, t, J = 6.4 Hz); 5.43 (2H, t, J = 5.6 Hz); 3.76 (6H,
s); 3.63 (6H, s); 3.01-2.89 (2H, m); 2.41-2.27 (4H, m); 2.00-
1.88 (2H, m); characteristic proton signals at 298 K, 6.48 (2H, t,
J=7.1 Hz, #5); 6.09 (2H, t, J = 6.4 Hz, #2); 5.67 (2H, t,
J =6.4 Hz, #3); 3.71 (6H, s, OCH3); 3.65 (6H, s, OCH3).
31P NMR (298 K, in CD2Cl2): 83.2 (JP-Rh = 218 Hz, #P1);
72.7 (JP-Rh = 195 Hz, #P2); labeling refers to Figure S16. 103Rh
NMR (298 K, in CD2Cl2): -837.
Figure 5. X-ray structure of the cation in [Rh((R,R)-DIPAMP)
(MAC)]BF4, 4 (ORTEP, 30% probability ellipsoids). Hydro-
gen atoms except on C2 and N1 are omitted for clarity. Selec-
ted distances [A] and angles [deg]: Rh1-O3=2.092(5); Rh-C1=
2.164(5); Rh-C2=2.162(6); Rh-P1=2.230(2); Rh-P2=2.300-
(2); P1-Rh-P2 = 82.7(1).
˚
Scheme 1. Reaction Sequence for the Equilibria between Species
2, 3a, and 3b
3b. 1H NMR (in CD2Cl2): characteristic proton signals at
298 K, 6.60 (1H, t, J=13.0 Hz, 6.6 Hz, #8); 6.39 (1H, t, J=
6.9 Hz, #6); 6.01 (1H, t, J=6.2 Hz, #3); 4.98 (1H, t, J=6.6 Hz,
#4); 4.25 (1H, t, J=5.7 Hz, #7); 3.88 (3H, s, OCH3); 3.65 (3H, s,
OCH3); 3.61 (3H, s, OCH3); 3.51 (3H, s, OCH3) (other signals
under those of phenyl-phenyl-bridged dimer). 31P NMR (298 K,
in CD2Cl2): 81.6 (JP-Rh=217 Hz,); 74.7 (JP-Rh=204 Hz,); 74.6
(JP-Rh =199 Hz,); 73.6 (JP-Rh = 222 Hz,); labeling refers to
Figure S19. 103Rh NMR (298 K, in CD2Cl2): -822/-766.
[Rh(DIPAMP)(MAC)]BF4 (4). [Rh(DIPAMP)]2(BF4)2
(0.01 mmol) and MAC (0.1 mmol) were dissolved in 1 mL
of i-PrOH, and the solution was subsequently frozen. Diethyl
ether (5 mL) was allowed to diffuse into the slowly melting
solution, yielding dark red crystals of 4 overnight.
hydrogenation of [Rh(DIPAMP)(NBD)]BF4 in i-PrOH owing
to its very low solubility, the isolation of catalyst-substrate
complexes seemed possible. Indeed, single crystals suitable for
X-ray crystallography were grown by layering a solution of 3a
and MAC in i-PrOH (Rh:substrate=1:5) with diethyl ether,
yielding [Rh((R,R)-DIPAMP)(MAC)]BF4 4, Figure 5.
Supporting Information Available: Crystal structure of
[Rh(DPPE)]2(BF4)2 (1), crystallographic data of 1, 3a, and
4, supportive NMR spectra, extinction diagrams. This
material is available free of charge via the Internet at http://
pubs.acs.org.
The substratecoordinatesviathe C-C doublebond andthe
oxygen of the acetyl group to the Rh. This is in accordance