P-chiral tetraphosphine dirhodium complex as a catalyst for asymmetric hydrogenation: Synthesis, structure, enantioselectivity, and mechanism. Stereoselective formation of a dirhodium tetrahydride complex and its reaction with methyl (Z)-α-acetamidocinnamate

Article abstract of DOI:10.1021/om050759p

Optically active C₂-symmetric tetraphosphine 4 was prepared via the phosphine-borane methodology. Its dirhodium complex 5 was structurally characterized and probed as a catalyst in asymmetric hydrogenations of representative prochiral substrates, demonstrating high activity and good to excellent enantioselectivities. A mechanistic study revealed that 5 can be cleanly and stereoselectively converted to the tetrahydride species 6a, which is stable up to 0 °C and at higher temperatures slowly decomposes without the loss of hydrogen. The low-temperature (-80 °C) reaction of 6a with methyl (Z)-α-acetamidocinnamate (MAC) cleanly gave the tetrahydride complex 7 containing one molecule of the substrate coordinated only via the amidocarbonyl group, whereas the double bond of the substrate remained noncoordinated. Raising the temperature to -40 °C resulted in irreversible isomerization of complex 7 to 8, which differs from 7 only by the spatial arrangement of ligands. Migratory insertion proceeding simultaneously with the isomerization of 7 to 8 yields the trihydride complex 9, which is an analogue of the monohydride intermediates described previously. When the reaction of 6a with MAC was carried out in the presence of an excess of MAC, the released dirhodium complex was captured by the substrate to give the catalyst-substrate complex 10, which was characterized by multinuclear NMR. Substrate MAC is much more loosely bound in octahedral complexes 7 and 8 than in the square planar catalyst-substrate complex 10. This finding provides experimental support for the stereoselection during the association step of the Rh-catalyzed asymmetric hydrogenation.

Full text of DOI:10.1021/om050759p

908
Organometallics 2006, 25, 908-914
P-Chiral Tetraphosphine Dirhodium Complex as a Catalyst for
Asymmetric Hydrogenation: Synthesis, Structure,
Enantioselectivity, and Mechanism. Stereoselective Formation of a
Dirhodium Tetrahydride Complex and Its Reaction with Methyl
(Z)-r-Acetamidocinnamate
Tsuneo Imamoto,*,† Keiji Yashio,† Karen V. L. Cre´py,† Kosuke Katagiri,†
Hidetoshi Takahashi,† Mitsuhiro Kouchi,‡ and Ilya D. Gridnev*,‡
Department of Chemistry, Faculty of Science, Chiba UniVersity, Yayoi-cho, Inage-ku,
Chiba 263-8522, Japan, and COE Laboratory, Department of Chemistry, Graduate School of Science,
Tohoku UniVersity, Sendai 980-8578, Japan
ReceiVed September 2, 2005
Optically active C2-symmetric tetraphosphine 4 was prepared via the phosphine-borane methodology.
Its dirhodium complex 5 was structurally characterized and probed as a catalyst in asymmetric
hydrogenations of representative prochiral substrates, demonstrating high activity and good to excellent
enantioselectivities. A mechanistic study revealed that 5 can be cleanly and stereoselectively converted
to the tetrahydride species 6a, which is stable up to 0 °C and at higher temperatures slowly decomposes
without the loss of hydrogen. The low-temperature (-80 °C) reaction of 6a with methyl (Z)-R-
acetamidocinnamate (MAC) cleanly gave the tetrahydride complex 7 containing one molecule of the
substrate coordinated only via the amidocarbonyl group, whereas the double bond of the substrate remained
noncoordinated. Raising the temperature to -40 °C resulted in irreversible isomerization of complex 7
to 8, which differs from 7 only by the spatial arrangement of ligands. Migratory insertion proceeding
simultaneously with the isomerization of 7 to 8 yields the trihydride complex 9, which is an analogue of
the monohydride intermediates described previously. When the reaction of 6a with MAC was carried
out in the presence of an excess of MAC, the released dirhodium complex was captured by the substrate
to give the catalyst-substrate complex 10, which was characterized by multinuclear NMR. Substrate
MAC is much more loosely bound in octahedral complexes 7 and 8 than in the square planar catalyst-
substrate complex 10. This finding provides experimental support for the stereoselection during the
association step of the Rh-catalyzed asymmetric hydrogenation.
design based on detailed knowledge of the intricate catalytic
cycle and the mechanism of stereoselection.
Introduction
Advances in the Rh-catalyzed asymmetric hydrogenation of
prochiral activated olefins are well recognized academically1
and requested by the pharmaceutical industry.2 Further research
in this area will lead to the creation of still more effective and
enantioselective catalysts, and ultimately to rational ligand
In the past decade P-stereogenic diphosphine ligands have
played a pivotal role in the creation of a strictly defined
asymmetric environment around the metal center.3 On the other
hand, the almost exclusive importance of C2-symmetric diphos-
phines has been undermined by successful use of chiral
monophosphines4 or C1-symmetric diphosphines5 with only one
stereogenic center in the ligand. A further extension of the
structural variety of the rhodium-phosphine complexes would
be the synthesis of polymetallic complexes, since one could be
interested in studying the cooperative effects of several identical
* To whom correspondence should be addressed. (T.I.) Phone/fax: +81
(0)43 290 2791. E-mail: imamoto@faculty.chiba-u.jp. (I.D.G.) Phone: +81
(0)22 795 3585. Fax: +81 (0)22 795 6784. E-mail: igridnev@
mail.tains.tohoku.ac.jp.
† Chiba University.
‡ Tohoku University.
(1) For representative reviews, see: (a) Chi, Y.; Tang, W.; Zhang, X. In
Modern Rhodium-Catalyzed Organic Reactions; Evans, P. A., Ed.; Wiley-
VCH: Weinheim, 2005; pp 1-31. (b) Knowles, W. S. Angew. Chem., Int.
Ed. 2002, 41, 1998. (c) Noyori, R. Angew. Chem., Int. Ed. 2002, 41, 2008.
(d) Tang, W.; Zhang, X. Chem. ReV. 2003, 103, 3029. (e) Genet, J.-P. Acc.
Chem. Res. 2003, 36, 908. (f) RajanBabu, T. V.; Yan, Y.-Y.; Shin, S. Curr.
Org. Chem. 2003, 7, 1759. (g) Ohkuma, T.; Kitamura, M.; Noyori, R. In
Catalytic Asymmetric Synthesis, 2nd ed.; Ojima, I., Ed.; Wiley-VCH: New
York, 2000; pp 1-110. (h) Brown, J. M. In ComprehensiVe Asymmetric
Catalysis; Jacobsen, E. N., Pfalz, A., Yamamoto, H., Eds.; Springer-
Verlag: Berlin, 1999; Vol. I, pp 121-182.
(2) For representative reviews, see: (a) Cobley, C. J.; Johnson, N. B.;
Lennon, I. C.; McCague, R.; Ramsden, J. A.; Zanotti-Gerosa, A. In
Asymmetric Catalysis on Industrial Scale; Blaser, H.-U., Schmidt, E., Eds.;
Wiley-VCH: Weinheim, 2004: pp 269-282. (b) Sumi, K.; Kumobayashi,
H. Top. Organomet. Chem. 2004, 6, 63.
(3) Reviews: (a) Cre´py, K. V. L.; Imamoto, T. Top. Curr. Chem. 2003,
229, 1. (b) Cre´py, K. V. L.; Imamoto, T. AdV. Synth. Catal. 2003, 345, 79.
(c) Pietrusiewicz, K. M.; Zablocka, M. Chem. ReV. 1994, 94, 1375.
(4) (a) Guillen, F.; Fiaud, J.-C. Tetrahedron Lett. 1999, 40, 2939. (b)
Claver, C.; Fernandez, E.; Gillon, A.; Heslop, K.; Hyett, D. J.; Martorell,
A.; Orpen, A. G.; Pringle, P. G. Chem. Commun. 2000, 961. (c) Reetz, M.
T.; Mehler, G. Angew. Chem., Int. Ed. 2000, 39, 3889. (d) van den Berg,
M.; Minnaard, A. J.; Schudde, E. P.; van Esch, J.; de Vries, A. H. M.; de
Vries, J. G.; Feringa, B. L. J. Am. Chem. Soc. 2000, 122, 11539. (e)
Komarov, I. V.; Bo¨rner, A. Angew. Chem., Int. Ed. 2001, 40, 1197. (f)
Ostermeier, M.; Priess, J.; Helmchen, G. Angew. Chem., Int. Ed. 2002, 41,
612. (g) Hu, A.-G.; Fu, Y.; Xie, J.-H.; Zhou, H.; Wang, L.-X.; Zhou, Q.-L.
Angew. Chem., Int. Ed. 2002, 41, 2348. (h) Jia, X.; Guo, R.; Li, X.; Yao,
X.; Chan, A. S. C. Tetrahedron Lett. 2002, 43, 5541. (i) Agbossou-
Niedercorn, F.; Suisse, I. Coord. Chem. ReV. 2003, 242, 145. (j) Jerphagnon,
T.; Renaud, J.-L.; Bruneau, C. Tetrahedron: Asymmetry 2004, 15, 2101.
10.1021/om050759p CCC: $33.50 © 2006 American Chemical Society
Publication on Web 01/17/2006

P-Chiral Tetraphosphine Dirhodium Complex
Organometallics, Vol. 25, No. 4, 2006 909
Scheme 1. Synthesis of Dirhodium Complex 5
catalytic centers in these highly effective asymmetric catalysts,6
which is important for the development of recoverable polymer-
supported and dendrimeric catalysts.7
Figure 1. Molecular structure (30% thermal ellipsoids) of complex
5. The two norbornadiene ligands and the two PF6 anions are
omitted for clarity.
-
Here we report the synthesis and crystal structure of the
dirhodium complex of a new P-stereogenic tetraphosphine
ligand combining two 1,2-bis(alkylmethylphosphino)ethane
(BisP*)-like units in one molecule. It is also reported that this
complex demonstrates excellent enantioselectivity of asymmetric
hydrogenation of representative prochiral substrates. The mecha-
nistic details of the asymmetric hydrogenation catalyzed by this
dirhodium complex revealed by an in-depth NMR study are
also described.
Figure 2. Two different quadrant diagrams conceivable for the
dirhodium complex 5.
tetraphosphine 4, which was subjected to complexation with 2
equiv of [Rh(nbd)2]PF6 to furnish Rh complex 5.
Results and Discussion
The X-ray structure of the dirhodium complex 5 shown in
Figure 1 gives important hints on the exact nature of the
asymmetric environment around the rhodium atoms in 5.
Formally, the bridging ethylene group might be regarded as a
small substituent, similar to a methyl group. In this case the
quadrant diagram for 5 consists of two independent units (Figure
2, A), each looking like a quadrant diagram for a t-Bu-BisP*
rhodium complex.12 On the other hand, the single-crystal X-ray
structure of 5 (Figure 1) suggests that two t-Bu groups and the
bridging ethylene group may provide sufficient hindrance to
preclude the formation of a chelate ring inside the molecule.
This kind of asymmetric environment corresponds to another
type of quadrant diagram (Figure 2, B) resembling that recently
reported for the “threechickenfootphos” ligand.5a Further ex-
perimental results (vide infra) suggest that the quadrant diagram
of type B (Figure 2) is operative in the Rh-catalyzed asymmetric
hydrogenation.
Asymmetric Hydrogenation Catalyzed by Dirhodium
Complex 5. The results of the asymmetric hydrogenation of
representative substrates using the dirhodium complex 5 as a
catalyst are summarized in Table 1. Excellent results comparable
to t-Bu-BisP* rhodium complex were obtained for R-dehy-
droamino acids (entries 1, 2),12a enamides (entries 5-7),13 (E)-
â-dehydroamino acid (entry 8),14 R,â-unsaturated phosphonate
Synthesis and Structure of the Dirhodium Complex. We
used the well-developed chemistry of air-stable phosphine-
boranes8 for the synthesis of a new tetraphosphine ligand
(Scheme 1). Thus, secondary phosphine-borane 19 was reacted
with 2 equiv of tosylate 210 to give a single enantiomer of the
tetraphosphine-borane 3 with a 65% yield. Deprotection of 3
by treatment with tetrafluoroboric acid11 afforded the desired
(5) (a) Hoge, G.; Wu, H.-P.; Kissel, W. S.; Pflum, D. A.; Greene, D. J.;
Bao, J. J. Am. Chem. Soc. 2004, 126, 5966. (b) Wu, H.-P.; Hoge, G. Org.
Lett. 2004, 6, 3645. (c) Blaser, H.; Brieden, W.; Pugin, B.; Spindler, F.;
Studer, M.; Togni, A. Top. Catal. 2002, 19, 3. (d) Ohashi, A.; Imamoto, T.
Org. Lett. 2001, 3, 373. (e) Ohashi, A.; Kikuchi, S.; Yasutake, M.; Imamoto,
T. Eur. J. Org. Chem. 2002, 15, 2535. (f) Matsumura, K.; Shimizu, H.;
Saito, T.; Kumobayashi, H. AdV. Synth. Catal. 2003, 345, 180.
(6) (a) Broussard, M. E.; Juma, B.; Train, S. G.; Peng, W. J.; Laneman,
S. A.; Stanley, G. G. Science 1993, 260, 1784. (b) Matthews, R. C.; Howell,
D. K.; Peng, W.-J.; Train, S. G.; Treleaven, W. D.; Stanley, G. G. Angew.
Chem., Int. Ed. Engl. 1996, 35, 2253. (c) Peng, W.-J.; Train, S. G.; Howell,
D. K.; Fronczek, F. R.; Stanley, G. G. Chem. Commun. 1996, 2607. (d)
Aubry, D. A.; Bridges, N. N.; Ezell, K.; Stanley, G. G. J. Am. Chem. Soc.
2003, 125, 11180. (e) Hunt, C., Jr.; Fronczek, F. R.; Billodeaux, D. R.;
Stanley, G. G. Inorg. Chem. 2001, 40, 5192. (f) Bowyer, P. K.; Cook, V.
C.; Gharib-Naseri, N.; Gugger, P. A.; Rae, A. D.; Swiegers, G. F.; Willis,
A. C.; Zank, J.; Zank, J.; Wilds, S. B. Proc. Natl. Acad. Sci. 2002, 99,
4877. (g) Bautista, M. T.; Earl, K. A.; Maltby, P. A.; Morris, R. H. J. Am.
Chem. Soc. 1988, 110, 4056.
(7) (a) Kollner, C.; Pugin, B.; Togni, A. J. Am. Chem. Soc. 1998, 120,
10274. (b) Kollner, C.; Togni, A. Can. J. Chem. 2001, 79, 1762. (c) Chiral
Catalyst Immobilization and Recycling; De Vos, D. E., Vankelecom, I. F.
J., Jacobs, P. A., Eds.; Wiley-VCH: Weinheim, 2000.
(8) (a) Imamoto, T.; Oshiki, T.; Onozawa, T.; Kusumoto, T.; Sato, K. J.
Am. Chem. Soc. 1990, 112, 5244. (b) Imamoto, T. Pure Appl. Chem. 1993,
65, 655.
(9) (a) Miura, T.; Yamada, H.; Kikuchi, S.; Imamoto, T. J. Org. Chem.
2000, 65, 1877. (b) Nagata, K.; Matsukawa, S.; Imamoto, T. J. Org. Chem.
2000, 65, 4185. (c) Cre´py, K. V. L.; Imamoto, T. Tetrahedron Lett. 2002,
43, 7735.
(11) (a) McKinstry, L.; Livinghouse, T. Tetrahedron Lett. 1994, 35, 9319.
(b) McKinstry, L.; Livinghouse, T. Tetrahedron 1994, 50, 6145.
(12) (a) Imamoto, T.; Watanabe, J.; Wada, Y.; Masuda, H.; Yamada,
H.; Tsuruta, H.; Matsukawa, S.; Yamaguchi, K. J. Am. Chem. Soc. 1998,
120, 1635. (b) Gridnev, I. D.; Higashi, N.; Asakura, K.; Imamoto, T. J.
Am. Chem. Soc. 2000, 122, 7183.
(13) Gridnev, I. D.; Yasutake, M.; Higashi, N.; Imamoto, T. J. Am. Chem.
Soc. 2001, 123, 5268.
(14) Yasutake, M.; Gridnev, I. D.; Higashi, N.; Imamoto, T. Org. Lett.
2001, 3, 1701.
(10) Ohashi, A.; Imamoto, T. Tetrahedron Lett. 2001, 42, 1099.

910 Organometallics, Vol. 25, No. 4, 2006
Imamoto et al.
Table 1. Asymmetric Hydrogenation of Representative
Prochiral Substrates Using Dirhodium Complex 5
interest was warmed by the dimeric structure of 5: the
asymmetric hydrogenation results shown in Table 1 implied that
the possibility of numerous isomers for each of the intermediates
does not affect strongly the excellent enantioselectivity that is
characteristic of the structurally similar P-stereogenic BisP*
ligands.
Similarly to all catalytic precursors studied previously, the
removal of the diene ligands from 5 was accomplished by
hydrogenation of a CD2Cl2 solution of 5 at -20 °C. However,
there were no sharp signals attributable to a solvate complex in
the NMR spectra of the resulting solution. Even if the sample
was thoroughly degassed in a vacuum, only a very broad signal
in the 31P spectrum was detected, whereas no hydride signals
were found in the 1H NMR spectrum. On the other hand, further
hydrogenation of this sample (10 min at -70 °C) led to the
quantitative and stereoselective formation of a dirhodium
tetrahydride species (6a) (Figure 3, Scheme 2), which follows
1
from the high symmetry of its H and 31P NMR spectra.
The tetrahydride species 6a is stable up to 0 °C (the retention
of hydrogen has been monitored by checking the 1:1 ratio of
-
the PF6 and R3PRh signals in the 31P NMR spectrum of 6a);
at higher temperatures 6a gradually decomposes without yielding
any definite product. The chemical shifts (-8.22 and -15.80
ppm) of the two couples of hydride protons correspond to their
axial and equatorial positions, respectively; the large coupling
2
constant JPH ) 95 Hz (confirmed by the heteronucleus
correlation experiment) of the lower field shifted hydrides attests
for their trans-disposition with respect to one of the phosphorus
atoms.
We attempted to distinguish experimentally between the four
isomers 6a-d with a symmetry corresponding to the observed
1H and 31P NMR spectra (Scheme 3). However, the data
obtained in the low-temperature NOE experiments were incon-
clusive, probably due to the relayed NOEs. Hence, we chal-
lenged this problem computationally by optimization of the
structures of 6a-d at the B3LYP/SDD level of theory (Scheme
3). The isomers 6c,d have the axially disposed solvent molecule
coordinated in the inner side of the molecule. Their significant
relative instability with respect to 6a,b is explained by the
unavoidable close contacts of the solvent molecule and the
bridging ethylene group. On the other hand, 6a is slightly more
stable than 6b; apparently the equatorially coordinated molecule
of CH2Cl2 can better accommodate the bridging ethylene group
(in 6a) than the t-Bu group (in 6b). Therefore, the computational
results for the tetrahydride complexes 6a-d suggest that the
bridging ethylene group in 5 provides significantly stronger
hindrance than the t-Bu group to the axially oriented ligand,
whereas in the P-Rh-P plane the hindrance created by the
t-Bu group is slightly higher.
a Determined by chiral HPLC and GC.
(entry 9),15 and dimethyl itaconate (entry 11).12a Modest
enantioselectivities were observed only for some â,â-disubsti-
tuted R-dehydroamino acids (entries 3, 4)12a and for (Z)-â-
dehydroamino acid (entry 9).14
Mechanistic Studies. The mechanism of the rhodium-
catalyzed asymmetric hydrogenation has consistently attracted
the attention of researchers for over 30 years.1,16 Asymmetric
hydrogenation is the simplest possible enantioselective catalytic
reaction. This fact was expected to enable straightforward
conclusions on the details of the catalytic cycle as well as on
the intricate details of the mechanism of stereoselection.
However, the real situation has proved to be much more
complicated than the idealized expectations. There is a large
number of experimentally characterized intermediates of Rh-
catalyzed asymmetric hydrogenation. Nevertheless, neither of
the possible mechanistic routes for this reaction is so far
commonly accepted, whereas the new experimental results
demonstrating the limits of straightforward interpretations
continue to appear regularly.17
Thus, the hydrogenation of 5 yields quantitatively and
stereoselectively a highly symmetric tetrahydride species 6a
(Figure 4). This result prompted us to study the reaction of 6a
with a prochiral substrate of asymmetric hydrogenation.
Hence, we were interested in probing the mechanism of
asymmetric hydrogenation catalyzed by 5 in order to collect
the additional data on the intermediates and particular steps of
Rh-catalyzed asymmetric hydrogenation. Moreover, additional
Reaction of Tetrahydride 6a with Methyl (Z)-r-Acetami-
docinnamate. Addition of either 2 or 4 equiv of methyl (Z)-
R-acetamidocinnamate (MAC) to the solution of 6a in CD2Cl2
at -90 °C resulted in the immediate disappearance of the signals
of 6a from the NMR spectra. Instead, the clean formation of
another intermediate (7) was observed (Figures 5, 6). By
gradually raising the temperature of the sample and simulta-
(15) Gridnev, I. D.; Yasutake, M.; Imamoto, T.; Beletskaya, I. P. Proc.
Natl. Acad. Sci. U.S.A. 2004, 101, 5385.
(16) For a recent review, see: Gridnev, I. D.; Imamoto, T. Acc. Chem.
Res. 2004, 37, 633.
(17) (a) Evans, D. A.; Michael, F. E.; Tedrow, J. S.; Campos, K. R. J.
Am. Chem. Soc. 2003, 125, 3534. (b) Drexler, H.-J.; Baumann, W.; Schmidt,
T.; Zhang, S.; Sun, A.; Spannenberg, A.; Fischer, C.; Buschmann, H.-J.;
Heller, D. Angew. Chem., Int. Ed. 2005, 44, 1184. (c) Reetz, M. T.;
Meiswinkel, A.; Mehler, G.; Angermund, K.; Graf, M.; Thiel, W.; Mynott,
R.; Balckmond, D. G. J. Am. Chem. Soc. 2005, 127, 10305.
-
neously checking the ratio of the PF6 and R3PRh signals in
the 31P NMR spectrum, we found that the intermediate 7 is stable
up to approximately -40 °C. At -40 °C a slow reaction
transforming 7 to a different species, 8, begins; it is complete

P-Chiral Tetraphosphine Dirhodium Complex
Organometallics, Vol. 25, No. 4, 2006 911
1
Figure 3. Section plots of H NMR (300 MHz, CD2Cl2, -90 °C) (bottom) and 31P NMR (122 MHz, CD2Cl2, -90 °C) (top) of the
tetrahydride complex 6a. The sum intensity of the two doublets shown in the upper plot was equal to the intensity of the multiplet from two
hexafluorophosphate anions, which indicates the quantitative formation of 6a.
Scheme 2. Formation of Tetrahydride Complex 6a
Scheme 3. Structures of the Four C2-Symmetric
Tetrahydrides 6a-d and Their Relative Energies (optimized
Figure 4. Structure of the tetrahydride complex 6a optimized at
at the B3LYP/SDD level of theory)
the B3LYP/SDD level of theory. The calculations reproduce well
the weakening of the Rh-P bond trans to the hydride ligand. The
bond lengths are 2.31 Å for both Rh-Pcis bonds and 2.53 and 2.55
Å for the Rh-Ptrans bonds.
MAC was present in the reaction mixture, stereoselective
formation of the catalyst-substrate complex 10 was observed.
If, in the presence of the excessive amount of MAC, the
temperature of the sample was sharply raised to 20 °C, the
formation of additional intermediate 9 was detected.The struc-
tures of the intermediates 7-10 were elucidated from their
spectral data and chemical properties. The intermediates 7 and
8 are slightly unsymmetrical tetrahydrides containing coordi-
nated MAC. The manner of MAC coordination follows from
the 13C NMR spectra of 7 and 8. The position and multiplicity
in 30 min at -20 °C. The NMR spectra of 7 and 8 are very
of the CH carbon of the double bond were checked by using
â-13C-labeled MAC: this carbon resonates at δ ) 136.6 and
129.7 in 7 and 8, respectively, and is not coupled to either
rhodium or phosphorus. Thus, the double bond in 7 and 8
remains noncoordinated. The striking difference in the 1H
chemical shifts of the amide proton of 7 (δ ) 9.50) and
noncoordinated MAC (δ ) 5.08) gives reason to conclude that
it is the amidocarbonyl that is coordinated to Rh. In the 13C
NMR spectrum of 7 the coordinated carbonyl resonates at δ )
178. Reversible dynamic effects due to the chemical exchange
are observed in the 1H, 13C, and 31P NMR spectra of the
intermediate 7. When the temperature was gradually raised from
similar; the main difference is slightly further separated signals
1
of the hydride protons in the H NMR as well as of the low-
field phosphorus atoms in the 31P NMR. The monitoring by
multinuclear NMR showed that the whole transformation of 7
to 8 can be carried out without the appearance of appreciable
amounts of the hydrogenation product (e.g., at -40 °C);
however, at -20 °C the isomerization 7 f 8 is accompanied
by the acquisition of the hydrogenation product. If an equivalent
amount of MAC (2 equiv of MAC for 1 equiv of 6a) has been
used, the signals of 8 in the 31P NMR spectrum were replaced
by a very broad signal due to the solvate complex. If excessive

912 Organometallics, Vol. 25, No. 4, 2006
Imamoto et al.
Scheme 4. Intermediates Intercepted in the Reaction of
Tetrahydride 6 with MAC
Figure 5. Evolution of the 31P NMR spectrum (122 MHz, CD2-
Cl2) of the sample prepared by addition of a 4-fold excess of MAC
to the solution of 6a in CD2Cl2: (a) immediately after mixing, -90
°C; (b) at -20 °C; immediately after raising the temperature; (c)
at -20 °C after 10 min; (d) at 20 °C, immediately after raising the
temperature; (e) at 20 °C after 20 min.
assigned the structure shown in the Scheme 4 for 7. Only one
molecule of MAC is coordinated to the dirhodium tetrahydride
via its amidocarbonyl, whereas the double bond remains
noncoordinated as follows from the ratio of coordinated MAC
to free MAC in solution.
The spectral properties of the intermediate 8 closely resemble
those of 7. Hence, we conclude that 8 must be an isomer of 7
with the same mode of substrate coordination, but different
spatial arrangement of the ligands.
The intermediate 9 is a trihydride complex with two hydride
ligands in the cis-position to both phosphorus atoms and one
hydride trans to phosphorus. Apparently, this is an analogue of
the well-known asymmetric hydrogenation monohydride inter-
mediates.
The catalyst-substrate complex 10 forms in the reaction
mixture if an excessive amount of MAC is present. It has a
symmetric structure with two MAC molecules coordinated to
both rhodium atoms, as follows from the symmetry of its 31P
NMR spectrum (closely resembling that of 5) and also from
1
Figure 6. Evolution of the H NMR spectrum (300 MHz, CD2-
Cl2) of the sample prepared by addition of a 4-fold excess of MAC
to the solution of 6a in CD2Cl2: (a) immediately after mixing, -90
°C; (b) at -20 °C; immediately after raising the temperature; (c)
at -20 °C after 10 min; (d) at 20 °C, immediately after raising the
temperature; (e) at 20 °C after 2 h. The enantiomeric excess of the
hydrogenation product obtained after quenching of this sample was
95% ee (R).
1
the number of signals in the H and 13C NMR spectra. The
structure shown in Scheme 4 was assigned by the high-field
shift of one of the t-Bu groups (δ ) 0.61) that according to the
previous studies12b,13 indicates its vicinity to the phenyl group
of the coordinated substrate. The two carbon atoms of the
coordinated double bond in 10 exhibited expected values of
chemical shifts and coupling constants.
The results of the study of the low-temperature reaction of
6a with MAC allow several interesting conclusions to be made.
First, quite unexpectedly and in contrast to previously studied
reactions, the simultaneous presence of the Rh(III) dihydride
moiety (two of them) and the substrate in the reaction mixture
did not result in immediate migratory insertion. The reason for
-90 to -50 °C, the reversible broadening of the signals in the
31P NMR spectrum resulted in their collapse, indicating the
occurrence of a degenerate rearrangement that makes the four
phosphorus atoms of 7 equivalent in pairs. In the 13C NMR
spectrum of the labeled MAC the exchange of the CHPh groups
of 7 and free MAC occurred with approximately the same rate
as the exchange in the 31P NMR. On the basis of these data we

P-Chiral Tetraphosphine Dirhodium Complex
Organometallics, Vol. 25, No. 4, 2006 913
this is that proper coordination of the double bond of MAC is
impossible. The relatively small molecule of solvent acting as
a ligand in 6 can more conveniently accommodate itself in the
inner part of the molecule, avoiding the vicinity of the t-Bu
groups (see above). This makes the tetrahydride 6a more stable
than 6b, and as a result the exclusive formation of 6a is
observed. However, a quite large molecule of MAC cannot
coordinate in the inner part of the molecule; therefore the
interaction of 6a with MAC gives the nonchelating complex 7
(Scheme 4). The dynamic effects in the spectra of 7 are
explained by reversible dissociation of this weak complex and
subsequent association of the MAC molecule to another rhodium
atom of the same molecule of the catalyst (Scheme 4).
At -40 °C the isomerization of 6a via pseudorotation18
becomes possible. When nothing except for 6a is present in
the reaction mixture, this rearrangement is not observed
experimentally since 6a is the most stable isomer among the
other tetrahydrides. However, in the presence of MAC the initial
binding of the amidocarbonyl in 8 or 8a can be accompanied
by the chelating coordination of the double bond across the less
hindered quadrant in 11, which results in immediate migratory
insertion leading to 9.
Experimental Section
General Procedures. All reactions and manipulations were
performed under a dry argon atmosphere using standard Schlenk-
type techniques. Substrates for acymmetric hydrogenation were
prepared according to the procedures described in the literature
(methyl (Z)-R-Acetamidocinnamate,19 â,â-disubstituted dehydroam-
ino acids,20 enamides,21 â-dehydroamino acids,22 and R,â-unsatur-
ated phosphonate23) or were commercially available (dimethyl
itaconate). Melting points were determined in open capillaries using
a Yamato micro melting point apparatus and were not corrected.
1H, 13C, 31P, and 11B NMR spectra were recorded using JEOL AL-
300, LA-400, and LA-500 instruments. Optical rotations were
measured with a JASCO DIP-370 digital polarimeter with a 1-dm-
long cell. Enantiomeric excesses were determined by HPLC analysis
performed on a Shimadzu LC-10AD pump, Shimadzu CTO-10AC
VP column oven, and Shimadzu SPD-10A VP UV detector with
an appropriate chiral column. Hydrogen of 99.9999% purity
(Nippon Sanso) was used for the mechanistic studies.
Preparation of Tetraphosphine-borane 3. n-Butyllithium (1.58
M solution) (15.4 mL, 24.4 mmol) was slowly added to a stirred,
cooled (-78 °C) solution of secondary diphosphine-borane 1 (2.6
g, 11.1 mmol) in THF (50 mL) under an Ar atmosphere. The
solution was stirred for 2 h and then added to a stirred solution of
(R)-tert-butyl(methyl)(2-(4-methylphenylsulfonyloxy)ethylphosphine-
borane 2 (7.7 g, 24.4 mmol) in THF (50 mL). The reaction mixture
was stirred for 1 h at -78 °C and for 5 h at room temperature. The
reaction was quenched by addition of 1 M aqueous hydrochloric
acid (100 mL). The organic layer was separated, and the aqueous
layer was extracted with ethyl acetale (3 × 100 mL). The combined
extracts were washed with aqueous NaHCO3 and brine and dried
over Na2SO4. The solvent was removed on an evaporator to leave
a white residue, which was recrystallized from hot toluene-hexane
Conclusion
Thus, we can see that the binding of the substrate in the
dihydride species directly preceding the migratory insertion step
(e.g., 7 or 8) is much weaker than the binding of the substrate
in the catalyst-substrate complex (e.g., 11). This conclusion is
important, since it provides experimental support for the idea
of the stereoselection during the association step of asymmetric
hydrogenation that we have recently proposed in several
publications.12b,13,15 Indeed, even if some step before the
irreversible migratory insertion is stereoselective, the weak
binding of the substrate in octahedral dihydride intermediates
destroys the results of this strereoselection via the dissociation
of the double bond. On the other hand, the fast equilibrium
between all possible isomers of dihydride intermediates is
facilitated by the easy dissociation of the substrate, and the
whole system can swoop down the energy minimum via the
single isomer of the dihydride intermediate characterized by the
lowest barrier of migratory insertion. The manner in which the
reaction of 6a with MAC proceeds demonstrates how the
tetrahydride species (corresponding to the dihydride intermedi-
ates in the case of monorhodium complexes) can be nonreactive
if the arrangement of the ligands appropriate for the occurrence
of migratory insertion is not achieved, and how they can easily
interconvert in the presence of substrate, thereby opening the
way for the proper coordination mode.
Straightforward interpretation of our experimental data sug-
gests that only one diphosphine-rhodium unit out of two is
participating in the hydrogenation catalytic cycle. This is a rather
unexpected result, and although it might be considered as bad
news from the point of view of possible future construction of
polynuclear catalytic systems, it reveals the important structural
restrictions for the synchronous operation of two adjacent metal
centers. Apparently, when one of the rhodium atoms undergoes
oxidative addition followed by coordination of the substrate,
the resulting octahedral moiety creates enough hindrance to
prevent coordination of the second molecule of the substrate.
This must be taken into account when either polymeric or
multifunctional catalytic systems are designed.
(2:1) to give pure 3 (65% yield) as colorless needles: mp 180-
1
182 °C; [R]24 -4.9 (c 1.0, CHCl3); H NMR (400 MHz, CDCl3)
D
δ 0.15-0.88 (m, 12H), 1.17 (d, JHP ) 13.8 Hz, 18H), 1.20 (d, JHP
) 13.8 Hz, 18H), 1.24 (d, JHP ) 9.7 Hz, 6H), 1.61-1.83 (m, 6H),
1.85-2.08 (m, 6H); 31P NMR (202 MHz, CDCl3) δ 29.6 (d, JPB
)
62 Hz), 38.1 (br s); 11B NMR (160 MHz, CDCl3) δ -62.9 (br s),
-61.0 (br d, JPB ) 62 Hz); IR (KBr) 2970, 2360, 2340 cm-1; FAB-
MS (rel intensity) 521 (M+ - H, 27), 491 (M+ - 2BH3 - 3H,
47), 479 (M+ - 3BH3 - H, 100), 423 (M+ - C4H9 - 3BH3, 38),
57 (C4H9, 74).
Deboranation of Phosphine-borane 3. Tetrafluoroboric acid
(solution in diethyl ether, 50 wt %) (0.98 mL, 6.9 mmol) was slowly
added to a stirred, cooled (0 °C) solution of tetraphosphine-borane
3 (150 mg, 0.29 mmol) in dichloromethane (10 mL) under an Ar
atmosphere. After 30 min the cooling bath was removed and the
reaction mixture was stirred at room temperature until the disap-
pearance of 3 (TLC control). Then the reaction mixture was again
cooled to 0 °C, and saturated aqueous NaHCO3 (15 mL) was slowly
added. After stirring for 30 min at 0 °C and for 3 h at room
temperature the reaction mixture was extracted with diethyl ether
(3 × 30 mL). The combined extracts were dried over Na2SO4, and
the solution was passed through a column of basic alumina. The
solvent was removed in a vacuum to give 128 mg (95% yield) of
practically pure tetraphosphine 4 (white solid).
Preparation of Dirhodium Tetraphosphine Complex 5. A
solution of tetraphosphine 4 (128 mg, 0.27 mmol) in freshly distilled
dichloromethane (2 mL) was added to a stirred solution of [Rh-
(19) Vineyard, B. D.; Knowles, W. S.; Sabacky, G. L.; Bachman, G. L.;
Weinkauff, D. J. J. Am. Chem. Soc. 1977, 99, 5946.
(20) Burk, M. J.; Gross, M. F.; Martinez, J. P. J. Am. Chem. Soc. 1995,
117, 9375.
(21) Burk, M. J.; Wang, Y. M.; Lee, J. R. J. Am. Chem. Soc. 1996, 118,
5142.
(18) (a) Berry, R. S. J. Chem. Phys. 1960, 32, 933. (b) Feldgus, S.;
Landis, C. R. J. Am. Chem. Soc. 2000, 122, 12714.
(22) Zhu, G.; Chen, Z.; Zhang, X. J. Org. Chem. 1999, 64, 6907.
(23) Burk, M. J.; Stammers, T. A.; Straub, J. A. Org. Lett. 1994, 1, 387.

914 Organometallics, Vol. 25, No. 4, 2006
Imamoto et al.
3
3
(nbd)2]PF6 (235 mg, 054 mmol) in dichloromethane (2 mL) under
an Ar atmosphere. The solution was stirred at room temperature
for 3 h. Then the reaction mixture was filtered, and the solvent
was removed in a vacuum. The residual solid was washed with
diethyl ether to give an orange powder, which was dried in a
vacuum. Recrystallization from methanol afforded 144 mg (46%
yield) of 5 as red cubes: 1H NMR (300 MHz, 293 K, CD3OD) δ
1.05 (d, JPH ) 14 Hz, 18H), 1.11 (d, JPH ) 14 Hz, 18H), 1.38 (d,
JPH ) 8 Hz, 6H), 1.61 (m, 6H), 1.86 (m, 4H), 2.0-2.7 (m, 6H),
1.26 (d, JHP ) 16 Hz, 18H; 2But); 1.68 (d, JHP ) 10 Hz, 6H;
2Me); 1.8-2.8 (m, 12H; 6CH2); 31P NMR (122 MHz, CD2Cl2, -30
1
1
°C) δ ) 82.2 (d, JPRh) ) 107 Hz), 114.1 (d, JPRh ) 142 Hz).
Intermediate 7: 1H NMR (hydride signals, 300 MHz, CD2Cl2,
-20 °C) δ -16.48 (m, 2H; axial hydrides); -7.96 (ddd, 2H,
2
2
1
trans
cis
equatorial hydrides, JHP ) 82 Hz, JHP JHRh ) 23, 45 Hz);
,
31P NMR (122 MHz, CD2Cl2, -20 °C) δ 80.1 (d, 1JPRh ) 103 Hz);
82.3 (d, 1JPRh ) 99 Hz); 101.0 (d, 1JPRh ) 136 Hz); 101.2 (d, 1JPRh
) 137 Hz).
4.10 (br. S, 2H), 4.16 (br. S, 2H), 5.62 (m, 2H), 5.84 (m, 6H); 13
C
Intermediate 8: 1H NMR (hydride signals, 300 MHz, CD2Cl2,
-20 °C) δ -15.90 (m, 2H; axial hydrides); -10.02 (ddd, 1H,
NMR (75 MHz, 293 K, CD3OD) δ 6.14 (d, JCP ) 23 Hz), 21.79
(m), 24.00 (m), 26.82 (d, JCP ) 26 Hz), 26.95 (d, JCP ) 26 Hz),
27.58 (m), 33.57 (m), 71.65, 73.36, 82.81 (t, JCP ) 7 Hz), 86.14
(m), 91.82 (m), 92.83 (t, JCP ) 9 Hz); 31P NMR (122 MHz, 293 K,
CD3OD) δ -146.0 (heptet, JPF ) 709 Hz), 63.4 (dm, JPRh ) 150
Hz), 76.3 (dm, JPRh ) 163 Hz); HR-MS (MALDI) 1252.2157 (M
+ 2CH3OH + K + 3H), calcd for C40H81O2F12P6Rh2K 1252.221,
proper isotopic pattern.
2
2
1
cis
trans
equatorial hydride, JHP ) 104 Hz, JHP JHRh ) 25, 47 Hz);
,
2
1
-9.26 (ddd, 1H, equatorial hydride, 2JHP ) 108 Hz, JHP JHRh
trans
cis
,
) 21, 43 Hz); 31P NMR (122 MHz, CD2Cl2, -30 °C) δ ) 82.2 (d,
1JPRh) ) 107 Hz), 114.1 (d, 1JPRh ) 142 Hz); 31P NMR (122 MHz,
CD2Cl2, -20 °C) δ 72.3 (d, 1JPRh ) 107 Hz); 79.6 (d, 1JPRh ) 102
1
1
Hz); 92.5 (d, JPRh ) 116 Hz); 98.8 (d, JPRh ) 119 Hz).
Intermediate 9: 1H NMR (hydride signals, 300 MHz, CD2Cl2,
20 °C) δ -24.11 (m, 1H; axial hydride); -16.00 (m, 1H; axial
X-ray Data Collection. A crystal appropriate for the X-ray
analysis was obtained by recrystallization of complex 5 from
methanol. The intensity data were collected on a Rigaku RAXIS-
II imaging plate diffractometer with graphite-monochromated Mo
KR radiation. The structure was solved by direct methods and
expanded using Fourier techniques. The non-hydrogen atoms were
refined anisotropically. Hydrogen atoms were included but not
refined. The final cycle of full-matrix was based on 3169 observed
reflections (I > 2.00 σ|I|) and 541 variable parameters. Atomic
coordinates, bond lengths and angles, and thermal parameters have
been deposited at the Cambridge Crystallographic Data Centre.
2
trans
hydride); -7.46 (ddd, 1H, equatorial hydride, JHP ) 134 Hz,
2
1
JHP JHRh ) 16, 27 Hz); 31P NMR (122 MHz, CD2Cl2, 30 °C) δ
cis
,
67.1 (dd, 1JPRh ) 150 Hz, 2JPRh) ) 26 Hz); 75.9 (ddd, 1JPRh ) 160
Hz, 2JPRh ) 37 Hz, 2JPRh ) 26 Hz); 84.7 (dd, 1JPRh ) 203 Hz, 2JPRh
) 24 Hz); 111.4 (ddd, 1JPRh ) 200 Hz, 2JPRh ) 37 Hz, 2JPRh ) 24
Hz).
Catalyst-substrate complex 10: 1H NMR (300 MHz, CD2Cl2)
3
3
δ 0.61 (d, JHP ) 15 Hz; 6CH3); 1.05 (d, JHP ) 14 Hz; 6CH3);
2
1.38 (d, JHP ) 9 Hz; 2CH3); 2.00 (s; 2CH3CON); 1.9-2.5 (m;
6CH2), 3.72 (s; 2CH3O), 5.91 (m; 2CHd), 6.9-7.5 (m; 2C6H5);
13C NMR (75 MHz, CD2Cl2, 30 °C) δ 8.4 (d, 1JCP ) 27 Hz; 2Me);
Asymmetric Hydrogenation (General Procedure). A 50 mL
Fisher-Porter tube was charged with the substrate (0.6 mmol) and
the catalyst precursor (0.003 mmol). The tube was connected to
the hydrogen tank via stainless steel tubing. The vessel was
evacuated and filled with hydrogen gas (Nippon Sanso, 99.9999%)
to a pressure of about 2 atm. This operation was repeated, and the
bottle was immersed in a dry ice-ethanol bath. The upper cock of
the bottle was opened, and anhydrous and degassed methanol was
added quickly using a syringe. After four vacuum/H2 cycles, the
tube was pressurized to 2 atm. The solution or suspension was
magnetically stirred at ambient temperature. After completion of
hydrogenation (30 min) the resulting solution was passed through
silica gel using ethyl acetate as the eluent, and the filtrate was
submitted to direct analysis for the ee value by HPLC.
1
2
1
20.6 (dd, JCP ) 24 Hz, JCP ) 7 Hz; 2CH2); 22.6 (d, JCP ) 24
2
Hz, 2CH2); 23.9 (s; CH3CONH), 24.4 (m; 2CH2); 26.6 (d, JCP
)
3 Hz, 6CH3); 29.1 (d, 2JCP ) 3 Hz, 6CH3); 31.4 (dd, 1JCP ) 18 Hz,
2JCRh ) 3 Hz; 2Ctert); 31.8 (dd, 1JCP ) 16 Hz, 2JCRh ) 4 Hz; 2Ctert);
54.0 (s, OMe); 81.9 (d, 2JCP ) 13 Hz; Câ); 82.3 (dd, 2JCP ) 13 Hz,
1JCRh ) 8 Hz); 129.3, 130.4, 130.9 (3 CH arom.); 138.2 (d, 3JCP
)
2 Hz); 168.4 (d, 2JCRh ) 4 Hz; O-CdO); 185.8 (d, 2JCRh ) 3 Hz;
N-CdO); 31P NMR (122 MHz, CD2Cl2, 30 °C) δ 73.7 (dm, 1JPRh
1
) 163 Hz); 79.4 (dm, JPRh ) 166 Hz).
Computational Details. Geometries of all stationary points were
optimized using analytical energy gradients of self-consistent-field24
and density functional theory (DFT).25 The latter utilized Becke’s
three-parameter exchange-correlation functional26 including the
nonlocal gradient corrections described by Lee-Yang-Parr (LYP),27
as implemented in the Gaussian 03 program package.28 All
geometry optimizations were performed using the SDD basis set.29
NMR Detection of Dirhodium Tetrahydride Species. Complex
5 (16 mg) was dissolved in 0.6 mL of CD2Cl2 under argon in a 5
mm NMR tube. The tube was connected to the Ar vacuum line
and to the lecture bottle with H2 via different stopcocks. The sample
was cooled to -20 °C, and after several degassing cycles 2 atm of
hydrogen was admitted. The solution in the tube was manually
stirred at -20 °C under 2 atm of hydrogen for 30 min. Then the
hydrogen was removed in a vacuum, and the tube was filled with
argon. The NMR spectra of the thus obtained sample were recorded
Acknowledgment. This research was supported by a Grant-
in-Aid for Scientific Research from the Ministry of Education,
Culture, Sports, Science and Technology and the Chemistry
COE program of Tohoku University. The computational results
in this research were obtained using supercomputing resources
at Information Synergy Center, Tohoku University.
1
in the temperature interval from -100 to -20 °C. The H NMR
spectrum showed that the norbornadienyl ligands were completely
hydrogenated, but no signals of Rh hydrides were detected in the
spectral region from 0 to -40 ppm. In the 31P NMR spectrum a
Supporting Information Available: Cartesian coordinates of
the optimized structures 6a-d; full ref 28. This material is available
free of charge via the Internet at http://pubs.acs.org.
broad signal was observed with the center at 132 ppm (LW1/2
)
2500 Hz); its integral intensity was approximately double of that
OM050759P
-
of the heptet belonging to the two PF6 anions (δ ) -148.7).
(24) Pulay, P. In Modern Theoretical Chemistry; Schaefer, H. F., Ed.;
Plenum: New York, 1977; Vol. 4, p 153.
The sample was returned to the vacuum-gas line, and argon was
removed by degassing. The sample was cooled to -70 °C, and 2
atm of H2 was applied for 30 min. During this time the sample
was stirred manually to avoid mass transfer problems. Then the
sample was placed into a precooled probe of the NMR spectrometer.
(25) Parr, R. G.; Yang, W. Density Functional Theory of Atoms and
Molecules; Oxford University Press: New York, 1989.
(26) Becke, A. D. J. Chem. Phys. 1993, 98, 5648.
(27) Lee, C.; Yang, W.; Parr, R. G. Phys. ReV. B 1988, 37, 785.
(28) Frisch, M. J.; et al. Gaussian 03, Revision B.05; Gaussian, Inc.:
Pittsburgh, PA, 2003.
Tetrahydride complex 6a: 1H NMR (300 MHz, CD2Cl2, -30
2
trans
°C) δ -15.80 (m, 2H; axial hydrides); -8.22 (ddd, JHP ) 95
(29) Leininger, T.; Nicklass, A.; Stoll, H.; Dolg, M.; Schwerdtfeger, P.
J. Chem. Phys. 1996, 105, 1052.
1
Hz, 2JHP JHRh ) 22, 46 Hz); 1.19 (d, 3JHP ) 17 Hz, 18H; 2But);
cis
,