Origin of the minor enantiomeric product in a noyori asymmetric hydrogenation: Evidence for pathways different to the major mechanism

Article abstract of DOI:10.1002/anie.200502463

(Chemical Equation Presented) Hidden pathways: The enantiomeric products of the catalytic hydrogenation of an isotopically labeled substrate were separated. The ratios of the isotopomers revealed that the major and minor enantiomers are formed mostly through different mechanisms. Hence, when analyzing the energy profile of an asymmetric catalytic reaction with the Arrhenius equation, two competing diastereomorphic catalytic cycles should be considered (see scheme).

Full text of DOI:10.1002/anie.200502463

Angewandte
Chemie
Reaction Mechanisms
DOI: 10.1002/anie.200502463
Origin of the Minor Enantiomeric Product in a
Noyori Asymmetric Hydrogenation: Evidence for
Pathways Different to the Major Mechanism**
Yoshitaka Ishibashi, Yuhki Bessho,
Masahiro Yoshimura, Masaki Tsukamoto, and
Masato Kitamura*
In a typical asymmetric catalytic process, a prochiral substrate
(Sub) is converted into a major (ProdS) and a minor (ProdR)
enantiomeric product via diastereomeric catalyst–substrate
(CatSub) and catalyst–product (CatProd) complexes
(Scheme 1). The two cycles are diastereomorphic to each
Scheme 1. Competition between diastereomorphic catalytic cycles.
other, but they take place through a single mechanism. As
such, the Arrhenius equation can be applied confidently to
correlate the ProdS/ProdR ratio to the relative stabilities of the
diastereomeric transition states at the first irreversible step
from CatSub to CatProd.[1] The energy profile thus obtained
can be utilized to estimate the structural difference between
the two transition states. In most cases, however, this premise
of a single mechanism has not been confirmed.[2] We have
long suspected that in some cases ProdS and ProdR may be
[*] Y. Ishibashi, Y. Bessho, M. Yoshimura, M. Tsukamoto,
Prof. Dr. M. Kitamura
Research Center for Materials Science and
Department of Chemistry, Nagoya University
Chikusa, Nagoya 464–8602 (Japan)
Fax: (+81)52-789-2261
E-mail: kitamura@os.rcms.nagoya-u.ac.jp
[**] This work was supported by a Grant-in-Aid for Scientific Research
(No. 14078121) from the Ministry of Education, Science, Sports,
and Culture, Japan. We are grateful to Professor R. Noyori for
valuable discussions and to T. Noda, K. Oyama, and Y. Maeda for
their technical support with reaction vessel production and
measurements of NMR spectra.
Supporting information for this article (asymmetric hydrogenation
and isotope-labeling experiments) is available on the WWW under
http://www.angewandte.org or from the author.
Angew. Chem. Int. Ed. 2005, 44, 7287 –7290
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
7287

Communications
generated through entirely different catalytic cycles in which
several chain carriers produced in a given reaction system
convert the same substrate but with different reactivities and
enantioselectivities. In such a catalytic process, various
reaction parameters dynamically link the species, and there-
fore the observed enantioselectivity becomes the average
contribution of the mechanistically different cycles.[3] Herein,
we report for the first time a clear example of asymmetric
catalysis that yields ProdS and ProdR from the same substrate
but through different mechanisms.
This newcase was identified in the hydrogenation of ( Z)-
3-phenyl-2-butenoic acid ((Z)-2) in the presence of [Ru-
(CH3CO2)2{(R)-binap}] ((R)-1).[4] Under standard conditions
((R)-1 (0.5 mm), (Z)-2 (100 mm), Sub/Cat = 200:1, methanol,
308C, 1–4 atm), (S)-3-phenylbutanoic acid ((S)-3)
w as
obtained with 94% ee when a pressure of 1–4 atm was
applied. The value decreased to 92% ee at 50 atm and to
88% ee at 100 atm,[5] whereas the reaction rate was enhanced
when the pressure of the hydrogen was increased.[6] The
substrate (Z)-2 is stable and undergoes neither Z/E geo-
Scheme 2. Structures of the eight possible isotopomers from the
reduction of (Z)-[3-13C]2 in CH3OD under H2/D2 mixtures at different
pressures.
ꢀ
ꢀ
metrical nor a,b/b,g (C2 C3/C3 C4) positional isomerization
during the course of the hydrogenation.[5,7] To improve the
accuracy of the isotope-labeling experiments, the 13C-labeled
substrate (Z)-[3-13C]2 was used, and the (R)-1-catalyzed
hydrogenations were carried out in CH3OD by changing the
hydrogen pressure and the ratio of H2 and D2. All the
reactions were stopped at low conversions to minimize
complications caused by gas–solvent and gas–gas isotope
exchange.[2,5] The enantiomeric products, (S)-3 (major) and
(R)-3 (minor), were separated by chiral HPLC (CHIRAL-
CEL OD (20 mm ” 25 cm); eluent: hexane/2-PrOH/AcOH
1000:1:3),[5] and the ratio of the eight isotopomers (3S)-
2H,3H-[3-13C]3, (3S)-2H,3D-[3-13C]3, (2R,3S)-2D,3H-[3-
13C]3,[9]
(2R,3S)-2D,3D-[3-13C]3,[9]
(3R)-2H,3H-[3-13C]3,
(3R)-2H,3D-[3-13C]3, (2S,3R)-2D,3H-[3-13C]3,[9] and (2S,3R)-
2D,3D-[3-13C]3[9] was determined by 13C{1H,2H} NMR spec-
troscopic analysis of the S and R products (Scheme 2).[5,10]
Figure 1 represents the change in the distribution pattern
of the major and minor enantiomeric products in going from
the H2 to H2/D2 conditions. When the reaction was carried out
in CH3OD under a pressure of 4 atm of H2, a mixture of (S)-3
and (R)-3 was obtained in a ratio of 97:3. The major S product
consisted of 2H,3H, 2H,3D, 2D,3H, and 2D,3D isotopomers
(10:89:0:1). Under an atmosphere of H2/D2 (1:1) in CH3OD,
the ratio shifted to 3:46:3:48. This change in distribution can
be understood simply in terms of the Ru monohydride
mechanism proposed in the Ru–binap-catalyzed hydrogena-
tion of tiglic acid[11,12] or a-(acylamino)acrylic esters.[2,13] In
short, a RuH species, generated from (R)-1 and H2, delivers
hydride to the Si face at C2 of (Z)-2 to form the (3R)-C3–Ru
Figure 1. Change of the isotopomer ratios in the (R)-binap–Ru-cata-
lyzed hydrogenation of (Z)-[3-13C]2 in CH3OD under 4 atm of H2 or H2/
D2 (1:1). Top: (S)-[3-13C]3 obtained as the major product (97%).
Bottom: (R)-[3-13C]3 obtained as the minor product (3%). The colored
arrows indicate the reaction pathways and the estimated distribution
of the isotopomers when the conditions are switched from H2 to H2/
D2 (1:1). ! RuH/C3–Ru/H2 and/or RuH/C2–Ru/H2 (RuH/H2);
! RuH/C3–Ru/H+; ! RuH2; ! RuH/C2–Ru/H+.
mers will be derived from the 10 part, and ꢁ 44 for the (3S)-
2H,3D and (3S)-2D,3D isotopomers from the 89 part. Simple
calculation estimates the ratio to be 3:47:3:47, which is
consistent with the observed ratio.
ꢀ
intermediate. Cleavage of the C Ru bond utilizes both H2 gas
In contrast, the minor R enantiomer contained largely the
(3R)-2H,3H (53%) together with the (3R)-2H,3D (30%) and
(3R)-2D,3H (17%) isotopomers. More gaseous hydrogen
appears to be consumed in the formation of the minor
product than in that of the major S enantiomer, and formation
of the 2D,3H isotopomer is characteristic of the minor
product. The reaction carried out in CH3OD under a mixture
of H2/D2 (1:1) generated a 23:20:13:43 mixture of the four 3R
isotopomers. The ambiguous isotope-labeling patterns, which
(the RuH/C3–Ru/H2 route) and CH3OD (the RuH/C3–Ru/
H+ route), and leads to a mixture of (3S)-2H,3H and (3S)-
2H,3D (10:89).[14,15] Under a 1:1 mixture of H2/D2 gas, the
RuH/C3–Ru/H2 route should produce the four isotopomers in
a 1:1:1:1 ratio (blue line) and the RuH/C3–Ru/H+ route
should produce 2H,3D and 2D,3D in a 1:1 ratio (green line);
therefore, a relative proportion of ꢁ 3 each for the (3S)-
2H,3H, (3S)-2H,3D, (3S)-2D,3H, and (3S)-2D,3D isotopo-
7288
www.angewandte.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2005, 44, 7287 –7290

Angewandte
Chemie
are totally different from those of the major enantio-
mer, can be explained by assuming that the RuH2
route[5] (red line) and the RuH/C2–Ru/H+ route
(purple line) are involved as well as the above two
reaction pathways (blue and green lines) (Figure 1,
minor). The two hydrides on RuH2 are delivered from
the same Re,Si face of (Z)-2 to C2 and C3 in a pairwise
manner.[16] In the RuH/C2–Ru route, a conjugate-type
addition of the hydride of a RuH species onto the Si
face of C3 generates a (2R,3S)-C2–Ru intermedi-
ate,[17] which then undergoes hydrogenolysis (RuH/
C2–Ru/H2) or protonolysis (RuH/C2–Ru/H+). Thus,
the 2H,3H, the 2H,3D, and the 2D,3H isotopomers in
the H2/CH3OD reaction are produced through the
pathways RuH2 and RuH/H2,[18] the RuH/C3–Ru/H+
route, and the RuH/C2–Ru/H+ route, respectively. In
light of the 2H,3H:2H,3D:2D,3H:2D,3D distribution
coefficients (1:1:1:1 for RuH/C3–Ru/H2 and RuH/
C2–Ru/H2 (blue line), 0:1:0:1 for RuH/C3–Ru/H+
(green line), 1:0:0:1 for RuH2 (red line), and 0:0:1:1
for RuH/C2–Ru/H+ (purple line)), the isotopomer
ratio obtained by the replacement of H2 with a 1:1
mixture of H2/D2 gas is calculated to be 22:20:13:45.
Figure 2a shows the good agreement between the
values estimated from the H2/CH3OD results (colored
bars) and the experimental values (gray bars). The
colors correspond to those of the reaction pathways
shown in Figure 1.
Figure 2. Observed (gray) and estimated (colored) isotopomer ratios of the major S and
minor R products in the (R)-binap–Ru-catalyzed hydrogenation of (Z)-[3-13C]2 in CH3OD
under: a) 4 atm of H2/D2 (1:1), b) 100 atm of H2/D2 (1:1), and c) 50 atm of H2/D2
(44:56). Colored values are estimated from the experimental results of 4, 50, and 100 atm
+
&
&
H2/CH3OD conditions. RuH/C3–Ru/H ; RuH/C3–Ru/H2 and/or RuH/C2–Ru/H2;
+
&
&
&
RuH2; RuH/C2–Ru/H ; unknown mechanisms; x axis: isotopomers of 3; y axis:
yield [%].
RuH/C2–Ru/H+ and RuH2 pathways, which are hardly
involved in the formation of the major product (Scheme 3).
Asymmetric catalytic reactions convert a prochiral substrate
This tendency is also seen with changes in the pressure of
the hydrogen gas and the H2/D2 ratio (Figure 2b,c).[19] An
increase in hydrogen pressure from 4 to 100 atm in CH3OD
approximately tripled the formation of the minor enantiomer
(Figure 2a, minor vs. Figure 2b, minor), and enhanced the
contribution of gaseous hydrogen to both the major and
minor product pathways. The amount of major product
formed by the RuH/H2 pathway (blue bars) increased
approximately five times and was coupled with a significant
decrease in the involvement of the RuH/C3–Ru/H+ route
(green bars), whereas the contribution of the RuH2 pathway
(red bars) to the formation of the minor product doubled. In
the case of the reaction carried out in CH3OD under H2/D2
gas (44:56) at 50 atm, the S and R products (96:4) consisted of
the eight isotopomers in a 9:31:11:45:1.0:0.6:0.7:1.7 ratio, as
indicated by the gray bars in Figure 2c. A reasonable
agreement with the estimation from the results obtained
from the reaction in CH3OD under 50 atm of H2 was observed
after a 0.44/0.56 correction to the above distribution coef-
ficients.
Each asymmetric catalytic cycle proceeding through the
RuH/C3–Ru/H2, RuH/C3–Ru/H+, RuH/C2–Ru/H2, RuH/C2–
Ru/H+, or RuH2 routes, or other possible reaction pathways,
has its own energy diagram, according to which the overall
rate as well as the stereochemistry is determined.[20] Neither
the details of the kinetics nor the structures of the inter-
mediates are known at present, but a series of isotope-
labeling experiments indicated that this Ru–binap-catalyzed
asymmetric hydrogenation of an a,b-unsaturated carboxylic
acid involves several catalytic species and that a significant
amount of the minor R enantiomer is formed through the
Scheme 3. Possible catalytic cycles producing the major S and minor R
enantiomers from the same (Z)-2 substrate through different chain
carriers. C3 substituents of (Z)-2 and products have been omitted for
clarity.
into an enantiomer-enriched chiral product. The observed
enantioselectivity has generally been interpreted to result
from competing diastereomorphic catalytic cycles caused by a
single chiral catalyst. However, this supposition has not been
substantiated. Together with earlier examples,[2,11,12] our study
calls into question such a generalized treatment.
Received: July 14, 2005
Published online: October 18, 2005
Angew. Chem. Int. Ed. 2005, 44, 7287 –7290
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 7289

Communications
Keywords: enantioselectivity · homogeneous catalysis ·
.
hydrogenation · isotopic labeling · reaction mechanisms
[1] S. Arrhenius, Z. Phys. Chem. 1889, 4, 226 – 248.
[2] For the first report of the confirmation of the involvement of the
two diastereomeric cycles, see: M. Kitamura, M. Tsukamoto, Y.
Bessho, M. Yoshimura, U. Kobs, M. Widhalm, R. Noyori, J. Am.
Chem. Soc. 2002, 124, 6649 – 6667.
[3] For examples of the different pathways through initial isomer-
ization of substrates, see: a) Y. Sun, C. LeBlond, J. Wang, D. G.
Blackmond, J. Laquidara, J. R. Sowa, Jr., J. Am. Chem. Soc.
1995, 117, 12647 – 12648; b) C. Detellier, G. Gelbard, H. B.
Kagan, J. Am. Chem. Soc. 1978, 100, 7556 – 7561; c) K. E.
Koenig, W. S. Knowles, J. Am. Chem. Soc. 1978, 100, 7561 – 7564;
d) C. A. Willoughby, S. L. Buchwald, J. Am. Chem. Soc. 1994,
116, 11703 – 11714; e) D.-R. Hou, J. Reibenspies, T. J. Colacot,
K. Burgess, Chem. Eur. J. 2001, 7, 5391 – 5400; for a concept of
different pathways with the same chain carrier, see: f) M.
Sawamura, R. Kuwano, Y. Ito, J. Am. Chem. Soc. 1995, 117,
9602 – 9603.
[4] Binap = 2,2’-bis(diphenylphosphanyl)-1,1’-binaphthyl.
[5] For details, see Supporting Information.
[6] For an opposite effect, see: M. Kitamura, M. Yoshimura, M.
Tsukamoto, R. Noyori, Enantiomer 1996, 1, 281 – 303.
[7] T. Ohta, H. Takaya, M. Kitamura, K. Nagai, R. Noyori, J. Org.
Chem. 1987, 52, 3174 – 3176; hydrogenation of (E)-2 in methanol
at 4 atm and 10–258C for 25 h gave (R)-3 in 97% yield with
5% ee; notably, the E isomer readily isomerizes to b,g-unsatu-
rated carboxylic acid, which was hydrogenated with (R)-1 in
methanol under similar conditions (S/C = 200:1, 1.5 atm, 10–
258C, 2 h) to afford (S)-3 with 74% ee.[8]
[8] T. Uemura, X. Zhang, K. Matsumura, N. Sayo, H. Kumobayashi,
T. Ohta, K. Nozaki, H. Takaya, J. Org. Chem. 1996, 61, 5510 –
5516.
[9] The absolute configurations, 2R,3S and 2S,3R, are altered to
2S,3S and 2R,3R with 13C-unlabeled 2D,3H-3 and 2D,3D-3.
[10] The authentic (2R*,3S*)-2D,3H-3 and (2S*,3S*)-2D,3H-3 were
prepared by diimide reduction of (Z)-2D-2 and (E)-2D-2,
respectively.[9] The isotope stability of the product was con-
firmed.[5]
[11] M. T. Ashby, J. Halpern, J. Am. Chem. Soc. 1991, 113, 589 – 594.
[12] T. Ohta, H. Takaya, R. Noyori, Tetrahedron Lett. 1990, 31, 7189 –
7192.
[13] a) J. A. Wiles, S. H. Bergens, Organometallics 1998, 17, 2228 –
2240; b) J. A. Wiles, S. H. Bergens, Organometallics 1999, 18,
3709 – 3714.
[14] The hydrogenation proceeded with overall cis addition.[5]
[15] The mechanism for the generation of (S)-2D,3D in 1% yield is
unknown.
[16] R. Eisenberg, Acc. Chem. Res. 1991, 24, 110 – 116.
[17] D. W. Lubell, M. Kitamura, R. Noyori, Tetrahedron: Asymmetry
1991, 2, 543 – 554.
[18] In this experiment, RuH/C3–Ru/H2 and RuH/C2–Ru/H2 cannot
be discriminated.
[19] Some deviations from expectations are probably due to the gas/
gas and gas/solvent isotope exchange and to the isotope effect.
For the mechanism of the H/D exchange between RuH and
CH3OD and between RuH and D2, see: P. G. Jessop, R. H.
Morris, Coord. Chem. Rev. 1992, 92, 155 – 284.
[20] a) R. Noyori, M. Kitamura in Modern Synthetic Methods (Ed.:
R. Scheffold), Springer, Berlin, 1989, pp. 115 – 198; b) R. Noyori
in Asymmetric Catalysis in Organic Synthesis, Wiley, NewYork,
1994, chap. 2.
7290 www.angewandte.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2005, 44, 7287 –7290