Enantiomeric products formed via different mechanisms: asymmetric hydrogenation of an α,β-unsaturated carboxylic acid involving a Ru(CH3COO)2[(R)-binap] catalyst

Article abstract of DOI:10.1016/j.tet.2007.08.071

Hydrogenation of (Z)-3-phenyl-2-butenoic acid with a Ru(CH₃COO)₂[(R)-binap] (BINAP=2,2′-bis(diphenylphosphino)-1,1′-binaphthyl) catalyst in methanol gives (S)-3-phenyl-2-butanoic acid and its R enantiomer in a 97:3 (4 atm) to 94:6 (100 atm) ratio in quantitative yield. Both hydrogen gas and protic methanol participate in the saturation of the olefinic bond. Analysis of the products obtained using (Z)-3-phenyl-2-butenoic acid-3-¹³C and either H₂, a 1:1 H₂-D₂ mixture, or D₂ in CH₃OD indicates that several catalytic cycles are operative, showing different reactivity and stereoselectivity. The major S enantiomer was formed primarily by the standard Ru monohydride mechanism, whereas the minor R isomer is produced via more complicated routes.

Full text of DOI:10.1016/j.tet.2007.08.071

Tetrahedron 63 (2007) 11399–11409
Enantiomeric products formed via different mechanisms:
asymmetric hydrogenation of an a,b-unsaturated carboxylic
acid involving a Ru(CH3COO)2[(R)-binap] catalyst
Masahiro Yoshimura, Yoshitaka Ishibashi, Kengo Miyata, Yuhki Bessho,
*
Masaki Tsukamoto and Masato Kitamura
Research Center for Materials Science and the Department of Chemistry, Nagoya University, Chikusa, Nagoya 464-8602, Japan
Received 8 August 2007; revised 22 August 2007; accepted 22 August 2007
Available online 26 August 2007
Abstract—Hydrogenation of (Z)-3-phenyl-2-butenoic acid with a Ru(CH3COO)2[(R)-binap] (BINAP¼2,20-bis(diphenylphosphino)-1,10-bi-
naphthyl) catalyst in methanol gives (S)-3-phenyl-2-butanoic acid and its R enantiomer in a 97:3 (4 atm) to 94:6 (100 atm) ratio in quantitative
yield. Both hydrogen gas and protic methanol participate in the saturation of the olefinic bond. Analysis of the products obtained using (Z)-3-
phenyl-2-butenoic acid-3-13C and either H2, a 1:1 H2–D2 mixture, or D2 in CH3OD indicates that several catalytic cycles are operative, show-
ing different reactivity and stereoselectivity. The major S enantiomer was formed primarily by the standard Ru monohydride mechanism,
whereas the minor R isomer is produced via more complicated routes.
Ó 2007 Elsevier Ltd. All rights reserved.
1. Introduction
the minor enantiomers are derived more or less from iso-
meric substrates formed during the reaction.8 More gener-
ally, however, it is not unreasonable to assume that
a reaction system contains several potentially active species
that can catalyze reaction of the same substrate but with dif-
ferent reactivity and stereoselectivity, in either sense or ex-
tent. In a given catalytic system, these cycles are linked by
various reaction parameters and, therefore, the observed
enantioselectivity is the average of the contribution of these
different mechanistic cycles. The following fully details
a clear example of asymmetric catalysis giving enantiomers
from the same substrate by different mechanisms.9,10
Asymmetric catalysis involves the multistep conversion of
a prochiral substrate into an enantiomerically-enriched chi-
ral product. The enantioselectivity has generally been ana-
lyzed as being due to competing diastereomorphic
catalytic cycles utilizing a chiral catalyst, in which the ratio
of the enantiomeric products is determined by the relative
stabilities of the diastereomeric transition states of the first
irreversible step, according to the Arrhenius equation.1 In
most cases, however, the primary presupposition, namely
the operation of a single mechanism, has not been con-
firmed.2 We have long suspected that in some cases, two en-
antiomers may be produced via entirely different pathways,
instead of the diastereomorphically identical mechanism. A
notable example is the BINAP–Ru-catalyzed hydrogenation
of geraniol to citronellol.3,4 The minor enantiomer is derived
via initial isomerization of the allylic alcohol to a homoal-
lylic alcohol, g-geraniol, wherein the two olefinic isomers
exhibit opposite enantioselection.5 Certain (E)-a-(acylami-
no)acrylic esters undergo E/Z isomerization in the presence
of chiral phosphine catalysts in alcohols.6 Hydrogenation of
imines catalyzed by a chiral titanocene7 also involves prior
syn/anti isomerization of the substrates. Thus, in these cases,
2. Results and discussion
2.1. Asymmetric hydrogenation of (Z)-3-phenyl-2-bute-
noic acid catalyzed by (R)-BINAP–Ru diacetate
(Z)-3-Phenyl-2-butenoic acid [(Z)-2] was selected as sub-
strate and Ru(CH3COO)2[(R)-binap] [(R)-1] as catalyst.
The standard hydrogenation conditions were [(R)-
1]¼0.5 mM, [(Z)-2]¼100 mM (substrate/catalyst ratio (S/
C)¼200), solvent¼methanol, and temperature¼30 ꢀC. Re-
action at 4 atm H2 quantitatively gave (S)-3-phenylbutanoic
acid [(S)-4] in 94% enantiomeric excess (ee). The enantio-
selectivity was sensitive to the H2 pressure. The ee of
94%, obtained for the 1–4 atm range decreased to 92% ee
at 50 atm, and to 88% ee at 100 atm. As might be expected,
Keywords: Asymmetric hydrogenation; BINAP–Ru; a,b-Unsaturated
carboxylic acid; Minor enantiomeric product; Mechanism.
* Corresponding author. Tel.: +81 52 789 2957; fax: +81 52 789 2261;
e-mail: kitamura@os.rcms.nagoya-u.ac.jp
0040–4020/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2007.08.071

11400
M. Yoshimura et al. / Tetrahedron 63 (2007) 11399–11409
however, the reaction rate was enhanced with an increase in
hydrogen pressure.2,11–13
ratios, the 13C-labeled substrate, (Z)-3-phenyl-2-butenoic
acid-3-13C [(Z)-2-3-13C], was used for the hydrogenation.
The experiments were carried out in CH3OD using (R)-1 as
the catalyst, and varying the hydrogen pressure from 4 to
100 atm. The reactions using H2 and the H2–D2 mixture
were stopped upon 6–22% conversion to minimize complica-
tions caused by gas/solvent and gas/gas isotope exchange.2,20
Fortunately, isotope exchange is significantly retarded by the
presence of the substrate, and the extent of the observed
scrambling does not affect the mechanistic argument. The
hydrogenation product was a mixture of eight isotopomers,
(S)-4-h,h-3-13C, (S)-4-h,d-3-13C, (2R,3S)-4-d,h-3-13C,
(2R,3S)-4-d,d-3-13C, (R)-4-h,h-3-13C, (R)-4-h,d-3-13C,
(2S,3R)-4-d,h-3-13C, and (2S,3R)-4-d,d-3-13C.21 Deuterium
was not doubly incorporated at C(2). The S and R enantio-
mers of 4 were separated by preparativeHPLC on a chiral sta-
tionary phase and subjected to 13C{1H,2H}–1H correlation
NMR and 13C{1H,2H} NMR analyses to determine the
structure and the ratio of the four isotopomers, respectively.
The relative stereochemistry of 4-d,h-3-13C was determined
by 1H NMR analysis. Reference (2R*,3S*)-4-d,h and
(2S*,3S*)-4-d,h21 samples were prepared by diimide reduc-
tion of (Z)-2-2-d and (E)-2-2-d, respectively, and occurred
with complete cis stereochemistry.22 Both the product and
substrate were isotopically stable under the reaction condi-
tions. Table 1 summarizes the product analysis.
The substrate (Z)-2 is stable and does not undergo geometri-
cal or positional isomerization during the course of the hy-
drogenation.14 For comparison, hydrogenation of (E)-2 in
methanol at 4 atm and 10–25 ꢀC for 25 h gave (R)-4 in 5%
ee and 97% yield.15 Notably, the E isomer readily isomerizes
to 3.16 This b,g-unsaturated carboxylic acid was hydroge-
nated with (R)-1 in methanol under similar conditions (S/
C¼200, 1.5 atm, 10–25 ꢀC, 2 h) to afford (S)-4 in 74% ee.15
When (R)-1 was reacted with an excess of (Z)-2, ligand
exchange occurred smoothly to afford crystalline Ru[(Z)-
CH3(C6H5)C]CHCOO]2[(R)-binap] [(R)-5].17–19 The
molecular structure is described in Section 4. Over the
course of the hydrogenation of (Z)-2, no substantial change
1
was seen in the H and 31P NMR spectra of the reaction
mixture. Ru hydrides were undetectable when reacting (R)-1
and (Z)-2 under an H2 atmosphere.
O
Ru
O
Ph2
P
O
O
P
Ph2
(R)-1
HOOC
D
HOOC
CH3
D
CH3
HOOC
H
HOOC
CH3
H
HOOC
(Z)-2-2-d
(E)-2-2-d
CH3
HOOC
H
13
C
CH3
(Z)-2
(E)-2
3
H
H
(Z)-2-3-13
C
HOOC
H
HOOC
H
D
H
H
D
CH3
CH3
HOOC
H
HOOC
H
HOOC
H
HOOC
H
H
H
13
13
13
13
C
H
C
H
C
H
C
H
CH3
CH3
CH3
CH3
(S)-4
(R)-4
13
13
13
C
(S)-4-h,h-3-13
C
(2R,3S)-4-d,h-3-
C
(R)-4-h,h-3-
C
(2S,3R)-4-d,h-3-
CH3
D
H
D
H
HOOC
H
HOOC
H
HOOC
H
H
HOOC
13
13
13
13
C
C
D
C
D
C
D
O
Ru
O
Ph2
CH3
CH3
CH3
CH3
P
O
O
D
13
13
13
(S)-4-h,d-3- C (2R,3S)-4-d,d-3-
C
(R)-4-h,d-3-
C
(2S,3R)-4-d,d-3-13
C
P
Ph2
Of note, the saturation of the olefinic bond of (Z)-2-3-13C uti-
lizes both hydrogen gas and protic methanol. Thus, the reac-
tion using H2 in CH3OD at 4–100 atm gave 4 containing
98.5–95% protium at C(2), while the protium content at
C(3) ranges from 12% (4 atm) to 25% (10 atm) to 47%
(50 atm) and 60.5% (100 atm). This trend was also the
case when using an H2–D2 mixture, giving a 49–43% pro-
tium content at C(2) and 7% (4 atm), 23% (50 atm), and
26% (100 atm) protium content at C(3). An increase in hy-
drogen pressure enhanced the degree of contribution of gas-
eous hydrogen. The extent of methanol participation was
CH3
(R)-5
2.2. Isotopic labeling experiments
Comparison of the enantiomeric products obtained via reac-
tion using H2, D2, and a 1:1 mixture of H2 and D2 provides
useful information regarding the mechanism. To further in-
sure the accuracy of the measurement of the isotopomer

M. Yoshimura et al. / Tetrahedron 63 (2007) 11399–11409
11401
Table 1. Analysis of 4-3-13C obtained by isotopic labeling experiments using (Z)-2-3-13C and Ru(CH3COO)2[(R)-binap] [(R)-1]a in CH3ODb
Conditions
H2, atm
50
H2–D2, atm
50
D2, atm
4
4
10
100
4
100
Time
9 h
10 h
10 min
7 min
9 h
12 min
7 min
35.2 h
Gas analysis, %
H2
HD
98.0
2.0
0
98.7
1.3
0
99.7
0.3
0
99.8
0.2
0
49.0
2.3
48.7
44.0
0.3
55.7
48.6
2.0
49.4
0
2.0
98.0
D2
Solvent analysis, %
CH3OH
CH3OD
<1
>99
10
<1
>99
10
<1
>99
16
<1
>99
22
<1
>99
17
<1
>99
10
<1
>99
6
1
99
99.9
97:3
Conversion, %
(S)-4/(R)-4c
97:3
97:3
95.5:4.5
94:6
97:3
96:4
94:6
Protium incorporation, %
H in C(2)
H in C(3)
98.5
12
98.1
24.7
96.5
47
95
60.5
49
7
43
23
48
26
0
1
Product distribution, %
Major enantiomer
(S)-4-h,h-3-13C
(S)-4-h,d-3-13C
(2R,3S)-4-d,h-3-13C
(2R,3S)-4-d,d-3-13C
Minor enantiomer
(R)-4-h,h-3-13C
(R)-4-h,d-3-13C
(2S,3R)-4-d,h-3-13C
(2S,3R)-4-d,d-3-13C
10.3
88.7
0.0
22.7
75.8
0.5
44.5
52.4
1.0
57.4
37.8
1.1
3.1
45.9
3.1
9.3
12.8
35.1
11.7
40.4
0
0
1.1
98.9
32.1
11.9
46.7
1.0
1.0
2.1
3.7
47.9
53.3
30.0
16.7
0.0
56.7
26.7
16.6
0.0
72.9
15.5
11.1
0.5
82.2
9.3
7.8
0.7
23.3
20.1
13.3
43.3
24.4
15.6
17.8
42.2
30.0
16.7
18.3
35.0
0
0
0
>99.9
a
The detailed procedures for the reaction and analysis are described in Section 4.
CH3OD contained 0.5% of CH3OH. D2 gas contained 0.4% of HD.
The ratio was determined by HPLC analysis.
b
c
greater than for the hydrogenation of enamides reported
previously.2,13
37.8% and 57.4%, respectively, while the R enantiomer con-
tains largely 4-h,h-3-13C (82.2%) together with other iso-
topomers. The minor R enantiomer appears to utilize more
gaseous hydrogen than the major S isomer. Use of a 1:1
mixture of H2 and D2 led to different distributions of the
isotopomers, as discussed below in terms of mechanistic
models.
Reaction at 4 atm of H2 in CH3OD afforded (S)-4 in 94% ee,
consisting of (S)-4-h,h-3-13C, (S)-4-h,d-3-13C, (2R,3S)-4-
d,h-3-13C, and (2R,3S)-4-d,d-3-13C in a 10.3:88.7:0:1.0 ra-
tio. With 1:1 H2–D2 in CH3OD, the distribution was changed
to 3.1:45.9:3.1:47.9.
2.3. Possible catalytic cycles
At 50 atm, the major S product decreased to 95.5%. Increase
in hydrogen pressure enhanced the degree of contribution of
gaseous hydrogen. This trend is more significant at 100 atm.
In addition, the isotope incorporation pattern varied sig-
nificantly. The isotopomer ratio of (S)-4-h,h-3-13C, (S)-4-
h,d-3-13C, (2R,3S)-4-d,h-3-13C, and (2R,3S)-4-d,d-3-13C
becomes 44.5:52.4:1.0:2.1 (50 atm) and 57.4:37.8:1.1:3.7
(100 atm) using H2 in CH3OD, and further changes to
9.3:32.1:11.9:46.7 (50 atm) and 12.8:35.1:11.7:40.4 (100 atm)
for H2–D2 in CH3OD. The reaction using D2 and CH3OD
gave mostly 4-d,d-3-13C as expected.
Reaction of phosphine–Ru(II) complexes and hydrogen gen-
erates a variety of Ru hydrides depending on the reaction
conditions (solvent, acidity/basicity, H2 pressure, etc.),
which catalyze homogeneous hydrogenation of olefinic
substrates.23,24 Heterolytic cleavage of molecular hydrogen
results in Ru monohydrides but, upon reaction with
additional hydrogen molecules, Ru polyhydrides are also
produced. The reactivity and enantioselectivity in asymmet-
ric hydrogenation are highly influenced by hydrogen pres-
sure, among other parameters.2,11,13,23 This is due to either
a change of the major catalytic species, or alternation of
the mechanistic pathway, or both. The pressure effect is par-
ticularly significant in BINAP–Ru-catalyzed hydrogenation
of unsaturated carboxylic acids.23 This and earlier studies on
the BINAP–Ru-catalyzed hydrogenation suggest the opera-
tion of at least the three sets of cycles, as shown in Figure 1.
The carboxylates attached to Ru act as either monodentate or
bidentate ligands, while methanol may coordinate to the
metallic center to stabilize the complexes.25
Hydrogenation in CH3OH at 4–100 atm gave (S)-4 in
94–88% ee. Neither the reaction with D2 nor CH3OD dem-
onstrated any isotope effects on enantioselectivity. Most no-
tably, the major S enantiomer and the minor R enantiomer
displayed different isotopic labeling patterns. The hydrogen
pressure also affected the product distribution. Of note are
the distribution patterns in going from H2 to H2–D2 as shown
in Table 1—the major S enantiomer obtained with H2 at
4 atm in CH3OD was 4-h,d-3-13C (88.7%), which is fol-
lowed by 4-h,h-3-13C (10.3%) and 4-d,d-3-13C (1.0%). In
contrast, the R enantiomer consisted of 4-h,h-3-13C
(53.3%), 4-h,d-3-13C (30.0%), and 4-d,h-2-13C (16.7%). Hy-
drogenation at 100 atm resulted in the S product containing
4-h,d-3-13C and 4-h,h-3-13C in comparable amounts,
The cycles designated S1 and R1 are the standard mecha-
nism for BINAP–Ru-catalyzed hydrogenation of a,b-un-
saturated carboxylic acids, as elucidated by kinetic
studies, rate law analysis, and deuterium labeling experi-
ments using tiglic acid.18,26 The reaction begins with

11402
M. Yoshimura et al. / Tetrahedron 63 (2007) 11399–11409
Figure 1. Possible mechanisms for the (R)-BINAP–Ru-catalyzed hydrogenation of a,b-unsaturated carboxylic acids in methanol. The C(3)-substituents in the
substrates are omitted for clarity. Configuration of the Ru cycles are unspecified.
ligand exchange between the Ru diacetate (R)-1 (A) and
substrate (Z)-2, giving (R)-5 (B), which causes heterolysis
of H2 to afford catalytic Ru monohydride C. This com-
plex is in equilibrium with the chelate olefin complex
D, which undergoes intramolecular hydride transfer to
form the five-membered ring intermediate E. Protonolysis
of the Ru–C bond with (Z)-2 results in F. This step is me-
diated by methanol. Finally, F reacts with H2 directly or

M. Yoshimura et al. / Tetrahedron 63 (2007) 11399–11409
11403
Table 2. Theoretical isotope distribution factor for 1:1 H2–D2 in CH3OD,
assuming no kinetic isotope effect
through B formed by 4/(Z)-2 ligand exchange, giving the
Ru hydride C and the hydrogenation product 4, to com-
plete the S1-HD and R1-HD cycles. This mechanism
forms 4-h,d with the use of H2 and CH3OD. Cycles S1
and R1, giving enantiomeric 4, are diastereomorphic
with one another, and proceed via a series of diastereo-
meric intermediates. Alternatively, the Ru–C linkage in
E can be cleaved by H2 to afford G, which, upon ligand
exchange with (Z)-2, completes the S1-H2 and R1-H2 cy-
cles. Reaction of B or F with H2 limits the overall rate of
hydrogenation. This mechanism yields the non-deuterated
product 4-h,h, even in CH3OD.
Entry
Mechanism
4-h,h
4-h,d
4-d,h
4-d,d
1
2
3
4
5
S1-H2
S1-HD
S2-H2
S2-HD
S3
0.25
0
0.25
0
0.25
0.5
0.25
0
0.25
0
0.25
0.5
0
0.25
0.5
0.25
0.5
0.5
0
0.5
enantiomeric products. In addition to the mechanistic
models illustrated in Figure 1, there may exist other un-
known mechanisms involving different catalytic species.
The selectivity must be affected by kinetic isotope effects.
Nevertheless, comparison of the isotopomer ratios obtained
with H2/CH3OD and H2–D2/CH3OD allows for the estima-
tion of the relative significance of the three catalytic cycles
in the (R)-1-catalyzed asymmetric hydrogenation of (Z)-2
in methanol.
This monohydride mechanism is known to operate in the (R)-
1-catalyzed hydrogenation of enamides,2,27 in which a RuH/
enamide chelate complex undergoes RuH/C]C migratory
insertion via a kinetically favored ‘exo’ cyclization process.
In this reaction, however, the five-membered ring complex
E formed from D requires a spatially constrained ‘endo’ ad-
dition of the Ru–H linkage to the C(2)]C(3) bond.
Therefore, 2+2 Ru–H addition across the olefinic bond oc-
curs also via an additional non-chelate process in which
(Z)-2 acts simply as an olefinic substrate.27 Here the regio-
chemistry is determined electronically so as to form the
more stable Ru–C(2) bond.28 The resulting I is then cleaved
either by a proton (S2-HD and R2-HD cycles) or H2 (S2-H2
and R2-H2 cycles) to give 4. Each pathway gives 4-d,h or
4-h,h, respectively. In both S1 and S2 (or R1 and R2) path-
ways, the two newly incorporated hydrogens have different
origins—either H2 and methanol, or two H2 molecules.
To facilitate the understanding, the results of deuterium la-
beling experiments in Table 1 are rearranged in Figure 2,
particularly focusing on the H2/CH3OD and H2–D2/
CH3OD systems at 4, 50, and 100 atm. At each hydrogen
pressure, the left and right regions correspond to the (S)-4
major product and the minor enantiomeric product (R)-4.
The values shaded in yellow are the observed isotopomer ra-
tios for H2/CH3OD (top) and H2–D2/CH3OD (bottom). The
detailed process involved in the calculation of the estimated
result for H2–D2/CH3OD on the basis of the observed result
for H2/CH3OD and the theoretical isotopic distribution
(Table 2) is shown in the middle unshaded part, in which
the isotopomers distributed via the S1-HD, S2-HD, S1-H2
and S2-H2, and S3 cycles (Fig. 1) are indicated in green,
pink, blue, and red, respectively. The colors of the numerical
values also correspond to those of the reaction pathways. If
the vertical sum of the colored values is consistent with the
observed value in the yellow part (bottom), the degree of
contribution should be correct. The horizontal sum shows
the degree of each catalytic cycle. As accuracy to the tenth
place is meaningless in the discussion, all of the calculated
values are rounded to the nearest integer. Qualitatively,
a marked difference is easily seen both in the h,h/h,d/d,h/
d,d distribution patterns and the patterns of the enantiomeric
products, depending on hydrogen pressure and other para-
meters. Quantification can be performed as follows.
Furthermore, under certain conditions, the Ru complex can
form catalytic Ru ‘dihydrides’ possessing two hydridic
ligands or an h2-H2.29 Hydrogenation of (Z)-2 with such a
dihydride (S3 and R3 cycles) produces 4 in which two
hydrogen atoms are supplied from a single H2 molecule.
Reaction with H2 or D2 gives either 4-h,h or 4-d,d, respec-
tively, regardless of whether the reaction utilizes deuterated
or non-deuterated methanol.
Assuming the possibility of these three cycles, isotopic la-
beling experiments serve as a useful tool for investigating
the pathways leading to the chiral products. When H2 is
replaced by a 1:1 H2–D2 mixture, the isotopomers of 4
obtained with H2 and CH3OD are redistributed, depending
on the relative significance of each of these catalytic cycles.
Table 2 shows the theoretical isotopic distribution in each
possible catalytic cycle, where the H/D kinetic isotope
effects are ignored.
First, the results from the 4 atm reaction are analyzed
(Fig. 2a). The hydrogenation generates (S)- and (R)-4 in
a 97:3 ratio. The major S product consists of h,h, h,d, d,h,
and d,d isotopomers in a 10.3:88.7:0.0:1.0 ratio, and the ratio
shifts to 3.1:45.9:3.1:47.9 when the conditions are changed
to 1:1 H2–D2/CH3OD. The 88.7% h,d portion can be easily
understood as being the product of the S1-HD mechanism
involving Ru monohydride and a Ru–C(3) species (Fig. 1,
middle). S1-HD should distribute the 88.7% into the four
isotopomers in a 0:0.5:0:0.5 ratio under 1:1 H2–D2 condi-
tions (Table 2 entry 2), giving ca. 44 (S)-h,d and ca. 44
(S)-d,d. The 10.3% h,h portion is expected to be generated
via S1-H2, S2-H2, and/or S3. With the 1:1 H2–D2/CH3OD
system, S1-H2 and S2-H2 each give four isotopomers in
equal amounts (Table 2, entries 1 and 3), while the S3 route
generates them in a 0.5:0:0:0.5 ratio (entry 5). Note that h,h
2.4. Pathways leading to varying enantiomers
In asymmetric hydrogenation of unsaturated carboxylic
acids, the sense and extent of enantioselectivity as well as
the sensitivity to hydrogen pressure are highly dependent
upon the substitution pattern of the substrates.23 Earlier stud-
ies on the hydrogenation of tiglic acid in methanol contain-
ing (R)-1, giving S-enriched 2-butanoic acid, suggested the
operation of the Ru monohydride mechanism, S1-HD, as
the dominant pathway.18,26,30 However, the details are un-
known and, in particular, the origin of the minor R enantio-
mer remains totally unclear. Isotopic labeling experiments
provide a key to distinguish the origin of both the

11404
M. Yoshimura et al. / Tetrahedron 63 (2007) 11399–11409
Figure 2. Calculation process for obtaining the expected ratios in the 1:1 H2–D2/CH3OD system on the basis of the observed results for the H2/CH3OD system.
The values shaded in yellow are the observed. The values in parentheses are for some unknown mechanism; (a) 4 atm (S)-4 and (R)-4 were obtained in a 97:3
ratio; (b) 50 atm (S)-4/(R)-4¼95.5:4.5. A 44:56 H2–D2 mixture was used, and this ratio is considered in the calculation of the distribution factors in Table 2; (c)
100 atm (S)-4/(R)-4¼94:6.
formation occurs in 3.1% yield in the 1:1 H2–D2/CH3OD
system. Where the entire 10.3 parts to be split equally among
h,h, h,d, d,h, and d,d, the yield would be 2.6%, which is
within the expanded range of experimental error expected
if the S3 mechanism is involved.
equally among h,h, h,d, d,h, and d,d (5:5:5:5) via the R1-H2
cycle, and the remaining ca. 34 to h,h and d,d (17:17) via
a RuH2-involved R3 route. The 30 parts of h,d generated
by R1-HD are divided equally between h,d and d,d, and
the 16.7 parts d,h generated via R2-HD are divided equally
between d,h and d,d. The vertical sum of these colored
values gives a ca. 23:20:13:43 ratio, which is, again, quite
consistent with the observed ratio. These results indicate
that 34%, 46%, and 20% of the minor product (R)-4 is
generated via a RuH2 route (R3), a RuH-protonolysis route
(R2-HD and/or R1-HD), and a RuH-hydrogenolysis route
(R1-H2 and/or R2-H2), respectively.
The d,h isotopomer, which should originate from the S2-HD
cycle via Ru–H and the Ru–C(2) species (Fig. 2a), can be di-
vided into d,h and d,d in the 1:1 H2–D2/CH3OD system. As
there is no d,h formed in the 4-atm experiment for the major
product, the d,h and d,d are also 0 under H2–D2 conditions.
The sum of the colored values in a column gives 3:47:3:47,
which is highly consistent with the observed ratio of
3.1:45.9:3.1:47.9. The origin of the 1.0 part of d,d is not
clear.31 Thus, the contribution of S1-HD and S1-H2/S2-H2
in the formation of (S)-4 is ca. 89% and 10%, respectively.
Good agreement between the values estimated from the H2/
CH3OD results and the experimentally observed values can
be seen both in the major and minor cases, even with changes
in hydrogen pressure and the H2–D2 ratio (Fig. 2b and c). In
all cases, an increase in the H2 pressure from 4 to 50 atm and
then to 100 atm in CH3OD enhances the contribution of gas-
eous hydrogen, giving larger amount of the h,h isotopomers
(major: 10.3/44.5/57.4; minor: 53.3/56.7/82.2) to-
gether with a significant decrease in the h,d isotopomers
(major: 88.7/52.4/37.8; minor: 30.0/26.7/9.3). A
similar detailed analysis of the 50 atm H2–D2/CH3OD
results reveals that the major S product formed by RuH
involves the S1-H2/S2-H2 cycle (45%) and S1-HD cycle
(52%), and that the involvement of S2-HD and S3 is
negligible (Fig. 2b). On the other hand, ca. 40% of the R
The minor product (R)-4 produces totally different patterns
from those of the major product (Fig. 2a, right). The ob-
served ratio of h,h, h,d, d,h, and d,d is 53.3:30.0:16.7:0.0,
which is characterized by a large amount of h,h. Gaseous hy-
drogen is more utilized in comparison with the major (S)-4
case. This phenomenon can be explained by the involvement
of the R3 cycle via a RuH2 species and the R2-H2 cycle via
RuH and Ru-C(2) species (Fig. 1, bottom and top). A change
to 1:1 H2–D2/CH3OD gives a 23.3:20.1:13.3:43.3 mixture.
Thus, taking the 23.3 parts h,d into consideration, ca. 20 of
the 53.3 parts from the H2/CH3OD system can be distributed

M. Yoshimura et al. / Tetrahedron 63 (2007) 11399–11409
11405
minor product is formed via R2-HD (17%) and R3 (23%)
cycles. The same tendency is observed with the 100 atm
H2–D2/CH3OD system (Fig. 2c): S1-H2/S2-H2, 57%; S1-HD,
38%; S2-HD, 1%; S3, 0% versus R1-H2/R2-H2, 48%; R1-
HD, 9%; R2-HD, 8%; R3, 34%. Although the R1-H2/R2-H2
ratio cannot be determined by the present isotope labeling
method, a significant contribution from R2-H2 is expected
in the formation of the R minor product by taking the
R2-HD/R1-HD ratio (ca. 1:1) into consideration.
DAICEL CHIRALCEL OD packed with cellulose tris(3,5-
dimethylphenylcarbamate) was used as a column. Eluted
compounds were detected at 254 nm.
All manipulations in the BINAP–Ru-catalyzed hydrogena-
tion were carried out using a standard Schlenk technique
on a dual manifold vacuum/Ar system. The structures of
(Z)-2 and (E)-2 were determined by comparing the spec-
tral data with reported values.32 Conversion of (Z)-2 and
1
(E)-2 was evaluated by comparison of the H NMR signal
These results indicate that the chelate RuH cycles S1 and R1
is S selective, but the non-chelate RuH cycles S2 and R2 and
the RuH2 cycles S3 and R3 are R-selective. Thus, the path-
way forming the minor enantiomer is complicated and is
not straightforward, while the major enantiomer is produced
primarily (99% at 4 atm and 95% at 100 atm) by the standard
monohydride mechanism.
intensity ratio at d 2.18 (CH3 of (Z)-2), d 2.61 (CH3 of
(E)-2), and d 1.31 (CH3 of 4). The ee values of 4
were determined by HPLC analysis (conditions: column,
CHIRALCEL OD (10 mmꢁ25 cm); eluent, a 1000:1:3
mixture of hexane/2-propanol/acetic acid; flow rate,
3 mL minꢂ1; temperature, 30 ꢀC; detector, 254 nm; retention
time (tR) of (R)-4, 27 min; tR of (S)-4, 43 min) The absolute
configuration of the hydrogenation product 4 was determined
by comparison of the sign of the optical rotation with the
reported value.33
3. Conclusion
Asymmetric catalysis converts a prochiral substrate into
an enantiomerically-enriched chiral product. The enantio-
selectivity has generally been interpreted as a result of
competing diastereomorphic catalytic cycles involving
a single chiral catalyst without substantiation of the pre-
supposition of the existence of the competing cycles.
This BINAP–Ru-catalyzed asymmetric hydrogenation of
an a,b-unsaturated carboxylic acid, together with earlier
examples18,26 suggests caution in making such an inter-
pretation. This study indicates that: (1) the hydrogenation
can involve several catalytic species, and (2) the major
and minor enantiomers can be produced by different
pathways.
4.2. Materials
CH3OH and CH3OD for hydrogenation were degassed at re-
flux in the presence of Mg (250 mg/100 mL) under argon for
6 h and distilled into Schlenk flasks. CH3OD contains 0.5%
OH species. The solvent was further degassed by three
freeze–thaw cycles before hydrogenation. CDCl3 and D2O
were purchased from Aldrich Chemical Co. and used with-
out further purification.
Ar gas was purified by passing it through a column of
BASF R3-11 catalyst at 80 ꢀC, and then through a column
of CaSO4 at room temperature. H2 gas of 99.99999% pu-
rity and D2 gas containing 0.4% HD were purchased from
Nippon Sanso, and HD gas containing 2% H2 and 2% D2
was obtained from Isotec. These gases were used for hy-
drogenation without purification. A 1:1 mixture of H2 and
D2 gases was prepared by mixing 50 atm H2 and 50 atm
D2 at ꢂ20 ꢀC in a stainless steel autoclave containing
a glass vessel. The H2/HD/D2 ratio was determined by
GC analysis. Ne and He for GC analysis of the H2/HD/
D2 ratio were obtained from Takachiho Kogyo and the
Japan Helium Center, respectively. He gas, purchased
from Nippon Sanso, was used for GC–MS analysis of the
CH3OH/CH3OD ratio. The purities of Ne and He were
99.999% and 99.9999%.
4. Experimental
4.1. General
Nuclear magnetic resonance (NMR) spectra were measured
on either a JEOL JNM-A400 equipped with a pulsed field
gradient unit and a triple-resonance probe, a JEOL JNM-
ECP500, or a Varian INOVA-500 instrument. The chemical
shifts are expressed either in parts per million (ppm) down-
field from Si(CH3)4 or in ppm relative to CHCl3 (d 7.26 in 1H
NMR and d 77.0 in 13C NMR). Signal patterns of 1H NMR as
well as 13C NMR are indicated as s, singlet; d, doublet; t,
triplet; q, quartet; m, multiplet; br, broad signal. GC–MS
analyses were performed using a Shimadzu QP-5000. All
melting points were determined with a Yanaco melting point
apparatus and are uncorrected.
Ru(CH3COO)2[(R)-binap] [(R)-1] and Ru(CH3COO)2[(S)-
binap] [(S)-1] were prepared by the literature method.17b
(Z)-3-Phenyl-2-butenoic acid [(Z)-2] (mp 129.6–
130.8 ꢀC) was synthesized according to the known
method.32 The D-labeled substrates (Z)-2-2-d and (E)-2-
2-d were synthesized by addition of (CH3)2CuLi to phe-
nylpropionic acid, followed by treatment with 1 N DCl
in D2O solution, in a way similar to the established pro-
cedure for synthesis of the nonlabeled compound32
(mp¼130.6–131.8 ꢀC for (Z)-2-2-d and 95.0–95.7 ꢀC for
(E)-2-2-d). The 13C-labeled substrate, (Z)-2-3-13C, was
prepared from trimethyl phosphonoacetate and acetophe-
none-carbonyl-13C according to the literature method for
preparation of the unlabeled compound.34 The substrate
Analytical thin layer chromatography (TLC) was performed
using Merck 5715 indicating plates precoated with silica gel
60 F254 (layer thickness: 0.25 mm). Liquid chromatographic
purifications were performed by flash column chromato-
graphy, using glass columns packed with Merck 9385
(230–400 mesh). For analytical and preparative HPLC, a
Shimadzu LC-10AD instrument equipped with a SCL-10A
system controller, a DGU-4A degasser, a SIL-10AXL
autosampler, a GILSON MODEL 201 fraction collector,
and a Shimadzu SPD-10A UV detector were used. A

11406
M. Yoshimura et al. / Tetrahedron 63 (2007) 11399–11409
was purified by recrystallization from 3:1 hexane/ether at
4 ꢀC (mp 129.0–131.0 ꢀC).
4.3. Asymmetric hydrogenation of (Z)-3-phenyl-2-bute-
noic acid [(Z)-2]
The experiment with H2/CH3OH was conducted in an
80-mL Schlenk tube equipped with a Young’s tap for the
1-atm trial, and in an 80-mL glass autoclave for the 4-atm
trial. The high-pressure reaction was carried out in an
80-mL glass vessel placed in a stainless steel autoclave.
H2/CH3OH (1 atm): the substrate, (Z)-2 (58.2 mg,
0.36 mmol), and CH3OH (3.6 mL) were placed into a dry,
Ar-filled, 20-mL Schlenk tube. The solution was degassed
by three freeze–thaw cycles, and was transferred via a stain-
less canula to another 20-mL Schlenk tube containing (R)-1
(1.5 mg, 1.8 mmol). After the resulting yellow solution of the
substrate and catalyst was degassed by two freeze–thaw cy-
cles, it was introduced into the Ar-filled reaction vessel and
then stirred under an initial H2 pressure of 1 atm at 30 ꢀC for
116 h. After removal of the solvent, conversion of (Z)-2 was
determined by 1H NMR analysis to be 100%. The crude re-
action mixture was purified on a silica gel column (eluent,
1:1 hexane/ether) to give (S)-4 in 94% ee.
Figure 3. The X-ray crystal structure of Ru[(Z)-CH3(C6H5)C]
CHCOO]2[(R)-bianp] [(R)-5].
4.5. Isotopic labeling experiments
4.5.1. H2/HD/D2 and CH3OH/CH3OD analysis. The H2/
HD/D2 ratio was analyzed via low temperature column chro-
matography according to published methods.2,35 The
CH3OH/CH3OD ratio was determined by GC–MS analysis
on a Shimadzu GCMS-QP5000.2
H2/CH3OH (4 atm): 59.3 mg (100 mM) (Z)-2, [(R)-
1]¼0.5 mM; 4 atm H2; CH3OH; 30 ꢀC; 48 h. The conversion
and ee were 100% and 94%, respectively.
4.5.2. Structural determination of isotopomers. The
structures of the four isotopomers, 4-h,h, 4-h,d, 4-d,h,
and 4-d,d, were determined by analyses of the 13C{1H}
NMR spectra and 13C{1H,2H}–1H correlation spectra
taken in CDCl3 at 23 ꢀC using a JEOL JNM-A400. To
minimize the magnetic field instability associated with
no D-spin lock, the measuring time was shortened by
use of a high concentration of the sample (69 mg/
0.6 mL), an acquisition time of 0.550 s, a pulse delay of
2.000 s, a scan number of 16, and a measurement time
of 95 min. The a-carbons of the four isotopomers reso-
nate at d 42.58, 42.48, 42.26, and 42.16 as singlets and
were correlated with the a-proton signals at d 2.665
(dd, J¼7 and 15 Hz) and 2.567 (dd, J¼7 and 15 Hz),
2.656 (d, J¼15 Hz) and 2.560 (d, J¼15 Hz), 2.640 (d,
J¼9 Hz), and 2.641 (s), respectively. From analysis of
the coupling mode, the four a-carbon signals from low
to high field were assigned to 4-h,h, 4-h,d, 4-d,h, and
4-d,d. In a similar way, the C(3) carbon signals at
d 36.10, 36.04, 35.71, and 35.65 were assigned to 4-h,h,
4-d,h, 4-h,d, and 4-d,d, respectively. These assignments
were consistent with the empirical rule that a D-possess-
ing carbon atom resonates at higher field with a larger
number of D.36 (S)-4: 1H NMR (400 MHz, CDCl3)
d 2.560 (d, J¼15 Hz, C(2)HS of 4-h,d), 2.567 (dd, J¼7
and 15 Hz, C(2)HS of 4-h,h), 2.640 (d, J¼9 Hz, C(2)H
of 4-d,h), 2.641 (s, C(2)H of 4-d,d), 2.656 (d, J¼15 Hz,
C(2)HR of 4-h,d), 2.665 (dd, J¼7 and 15 Hz, C(2)HR of
4-h,h). 13C{1H,2H} NMR (100 MHz, CDCl3) d 35.65 (s,
C(3) of 4-d,d), 35.71 (s, C(3) of 4-h,d), 36.04 (s, C(3)
of 4-h,d), 36.10 (s, C(3) of 4-d,h), 42.16 (s, C(2) of
4-d,d), 42.26 (s, C(2) of 4-d,h), 42.48 (s, C(2) of 4-h,d),
42.58 (s, C(2) of 4-h,h). 13C{1H} NMR (100 MHz,
CDCl3) d 35.65 (t, J¼ 21 Hz, C(3) of 4-d,d), 35.71 (t,
J¼20 Hz, C(3) of 4-h,d), 36.04 (s, C(3) of 4-d,h), 36.10
(s, C(3) of 4-h,h), 42.16 (t, J¼19 Hz, C(2) of 4-d,d),
H2/CH3OH (50 atm): 99.2 mg (100 mM) (Z)-2, [(R)-
1]¼0.5 mM; 50 atm H2; CH3OH; 30 ꢀC; 24 h. The conver-
sion and ee were 100% and 92%, respectively.
H2/CH3OH (100 atm): 99.4 mg (100 mM) (Z)-2, [(R)-
1]¼0.5 mM; 100 atm H2; CH3OH; 30 ꢀC; 12 h. The conver-
1
sion and ee were 100% and 88%, respectively. (S)-4: H
NMR (400 MHz, CDCl3) d 1.31 (d, 3, J¼6.8 Hz, CH3), 2.57
(dd, 1, J¼16 Hz and 8.3 Hz, C(2)HS), 2.66 (dd, 1, J¼16 and
6.8 Hz, C(2)HR), 3.26 (ddd, 1, J¼8.3, 6.8 and 6.8 Hz,
C(3)H), 7.17–7.32 (m, 5, aromatic), 9.40–10.50 (br, 1, COOH).
4.4. X-ray crystallographic analyses for
Ru[(Z)-CH3(C6H5)C]CHCOO]2[(R)-binap]
Ru(CH3COO)2[(R)-binap] [(R)-1] (50.6 mg, 60.1 mmol) and
(Z)-2 (208 mg, 1.28 mmol) were placed in a dry 20-mL
Schlenk tube, and CH3OH (15 mL) was introduced under
an Ar stream. The solution was stirred for 1 h at room tem-
perature, during which time a yellow solid precipitated. The
supernatant liquid was removed via a canula covered with
filter paper at one end. The solid was washed with CH3OH
(10 mL) and dried in vacuo, affording yellow powdery crys-
tals of Ru[(Z)-CH3(C6H5)C]CHCOO]2[(R)-binap] [(R)-5]
(32.6 mg, 52% yield). Orange prismatic crystals were ob-
tained by recrystallization of (R)-5 (ca. 10 mg) using a liq-
uid–liquid diffusion of hexane (4 mL) into toluene (1 mL)
at 4 ꢀC. A single crystal (0.22 mmꢁ0.20 mmꢁ0.14 mm)
was mounted on the top of a quartz fiber with a small amount
of epoxy resin adhesive and transferred to a goniometer
head. The data was collected in a ꢂ100 ꢀC nitrogen
atmosphere. The molecular structure of (R)-5 is illustrated
in Figure 3.

M. Yoshimura et al. / Tetrahedron 63 (2007) 11399–11409
11407
42.26 (t, J¼19 Hz, C(2) of 4-d,h), 42.48 (s, C(2) of
4-h,d), 42.58 (s, C(2) of 4-h,h).
effected only during the acquisition time. The values of re-
laxation time, T1, obtained by the inversion recovery method
were <4.5 s for both C(2) and C(3) carbons and <5 s for all
protons. These values satisfied the necessary conditions,
PD>5ꢁT1 for protons and AT+PD> 5ꢁT1 for carbons.
4.5.3. Determination of the relative stereochemistry of
4-d,h. The relative stereochemistry of 4-d,h was determined
by comparing the 1H NMR data with those of reference iso-
topomers, namely a mixture of (2R*,3S*)-4-d,h and
(2S*,3S*)-4-d,h prepared by reduction of (Z)-2-2-d and
(E)-2-2-d, respectively, with diimide (Section 4.5.4), which
is known to proceed via cis addition of hydrogen.21 1H
NMR of (2R*,3S*)-4-d,h (500 MHz, CDCl3) d 2.53 (1, br
d, J¼8.3 Hz, C(2)H). 1H NMR of (2S*,3S*)-4-d,h
(500 MHz, CDCl3) d 2.64 (1, dr s, C(2)H). Because of the
priority rule, (2R,3S)-4-d,h-3-13C and (2S,3R)-4-d,h-3-13C
correspond to (2S*,3S*)-4-d,h.
4.5.7. H2/CH3OD conditions. The experiments with H2/
CH3OD under 4 or 10 atm were conducted in a 400-mL glass
autoclave, whereas the high-pressure reactions under 50 or
100 atm of H2 were carried out in a 200-mL glass vessel
placed in a stainless steel autoclave.
H2/CH3OD (4 atm): reaction scale, 202 mg (100 mM) (Z)-2-
3-13C; [(R)-1]¼0.5 mM; 4 atm H2; CH3OD; 30 ꢀC. After
9 h, the gas in the reaction vessel was transferred to an 80-
mL vacuumed Schlenk flask, and the solvent was recovered
by distillation. The H2/HD/D2 and CH3OH/CH3OD ratios,
as analyzed by the methods above, were 98.0:2.0:0 and
<1:>99, respectively. The residue was subjected to 1H
NMR and HPLC analyses, which determined the conversion
and ee to be 10% and 94%, respectively. The crude reaction
mixture was purified by silica gel column chromatography
(eluent, 1:1 hexane/ether) to remove the Ru complex. The
(Z)-2-3-13C and 4-3-13C mixture was separated by pre-
parative HPLC (conditions, see Section 4.5.5) to give
(Z)-2-3-13C, (S)-4-3-13C containing (S)-4-h,h-3-13C, (S)-4-
h,d-3-13C, (2R,3S)-4-d,h-3-13C, and (2R,3S)-4-d,d-3-13C
in a 10.3:88.7:0.0:1.0 ratio and (R)-4-3-13C containing (R)-
4-h,h-3-13C, (R)-4-h,d-3-13C, (2S,3R)-4-d,h-3-13C, and
(2S,3R)-4-d,d-3-13C, in a 53.3:30.0:16.7:0.0 ratio. The ratio
of the 1H NMR signal intensity at d 2.62–2.71(m) and
d 2.53–2.62(m) was 100:99 for the major enantiomer and
100:83 for the minor enantiomer.
4.5.4. Synthesis of reference isotopomers. The 3S-enriched
reference isotopomers 4-h,h, 4-d,d and mixtures of 4-h,h and
4-h,d, and 4-d,h and 4-d,d were prepared by the reduction of
(Z)-2 under the following conditions. For 4-h,h: (Z)-2
(50.0 mg, 0.31 mmol); (S)-1 (1.0 mg, 1.2 mmol); CH3OH
(3.1 mL); 4 atm H2; 30 ꢀC; 48 h. For 4-d,d: (Z)-2
(50.0 mg, 0.31 mmol); (S)-1 (1.0 mg, 1.2 mmol); CH3OD
(3.1 mL); 1 atm D2; 30 ꢀC; 216 h. For a 77:23 mixture of
4-h,h and 4-d,h: (Z)-2 (100 mg, 0.62 mmol); (S)-1 (3.0 mg,
3.6 mmol); CH3OD (6.2 mL); 4 atm HD; 30 ꢀC; 338 h. For
a 24:76 mixture of 4-h,d and 4-d,d: (Z)-2 (100 mg,
0.62 mmol); (S)-1 (3.0 mg, 3.6 mmol); CH3OD (6.2 mL);
4 atm HD; 30 ꢀC; 168 h. All conversions were 100%. Prod-
ucts were purified by silica gel column chromatography
(eluent, 1:1 hexane/ether).
The reference isotopomers (2R*,3S*)-4-d,h and (2S*,3S*)-
4-d,h were prepared by the reduction of (Z)-2-2-d and (E)-
2-2-d with diimide under the following conditions.
H2/CH3OD (10 atm): reaction scale, 202 mg (100 mM)
(Z)-2-3-13C; [(R)-1]¼0.5 mM; 10 atm H2; CH3OD; 30 ꢀC;
10 h. These conditions converted 10% of the (Z)-2-3-13C
to give (S)-4-3-13C in 94% ee. The H2/HD/D2 and
CH3OH/CH3OD ratios were 98.7:1.3:0 and <1:>99. The
(S)-4-h,h-3-13C/(S)-4-h,d-3-13C/(2R,3S)-4-d,h-3-13C/(2R,3S)-
4-d,d-3-13C ratio and the (R)-4-h,h-3-13C/(R)-4-h,d-3-
13C/(2S,3R)-4-d,h-3-13C/(2S,3R)-4-d,d-3-13C ratio were
22.7:75.8:0.5:1.0 and 56.7:26.7:16.6:0.0. The ratio of the
1H NMR signal intensity at d 2.62–2.71(m) and d 2.53–
2.62(m) was 100:98 for the major enantiomer and 100:83
for the minor enantiomer.
(2R*,3S*)-4-d,h:
(Z)-2-2-d
(22.0 mg,
0.14 mmol);
NH2NH2$H2O (40 mg, 0.80 mmol); CuSO4$5H2O
(1.4 mg, 5.6 mmol); C2H5OH (3.0 mL); 22 h; 10% conver-
sion. For (2S*,3S*)-4-d,h: (E)-2-2-d (22.0 mg, 0.14 mmol);
NH2NH2$H2O (40 mg, 0.80 mmol); CuSO4$5H2O
(1.4 mg, 5.6 mmol); C2H5OH (3.0 mL); 22 h; 35% conver-
sion. No E/Z isomerization was observed in the reduction
of (E)-2-2-d, whereas the unreacted substrate contained
about 10% of the E isomer in the reduction of (Z)-2-2-d.
4.5.5. Separation of the enantiomers. The major (S)-4 and
minor (R)-4 products obtained in the BINAP–Ru-catalyzed
hydrogenation with either H2/CH3OD, H2–D2/CH3OD, or
D2/CH3OD were separated under the following conditions:
a mixture of (Z)-2 and 4, 20–25 mg; column, CHIRALCEL
OD (20 mmꢁ250 mm); eluent, 1000:1:3 hexane/2-propa-
nol/acetic acid; flow rate, 10.0 mL minꢂ1; temperature,
30 ꢀC; detector, 254 nm; tR of (R)-4, 45 min; tR of (S)-4,
85 min. There is no overlap of the compounds.
H2/CH3OD (50 atm): reaction scale, 202 mg (100 mM) (Z)-
2-3-13C; [(R)-1]¼0.5 mM; 50 atm H2; CH3OD; 30 ꢀC;
10 min. These conditions converted 16% of the (Z)-2-
3-13C to give (S)-4-3-13C in 91% ee. The H2/HD/D2
and CH3OH/CH3OD ratios were 99.7:0.3:0 and <1:>99.
The (S)-4-h,h-3-13C/(S)-4-h,d-3-13C/(2R,3S)-4-d,h-3-13C/
(2R,3S)-4-d,d-3-13C ratio and the (R)-4-h,h-3-13C/(R)-
4-h,d-3-13C/(2S,3R)-4-d,h-3-13C/(2S,3R)-4-d,d-3-13C ratio
were 44.5:52.4:1.0:2.1 and 72.9:15.5:11.1:0.5. The ratio of
the 1H NMR signal intensity at d 2.62–2.71(m) and
d 2.53–2.62(m) was 100:98 for the major enantiomer and
100:89 for the minor enantiomer.
4.5.6. Determination of the isotopomer ratios. The iso-
topomer ratios of (S)-4-3-13C and (R)-4-3-13C were deter-
mined by measurement of the C(3) carbon signal area in
the 13C{1H,2H} NMR spectra, which were taken using
a nanoprobe under the following conditions: flip angle,
ꢀ
48.0 ; acquisition time (AT), 2.621 s; pulse delay (PD),
30.000 s; sample concentration, 1.0 mg/40 mL; measure-
ment time, 256 scans (2.3 h). H and H decoupling was
H2/CH3OD (100 atm): reaction scale, 202 mg (110 mM) (Z)-
2-3-13C; [(R)-1]¼0.5 mM; 100 atm H2; CH3OD; 30 ꢀC;
7 min. These conditions converted 22% of the (Z)-2-3-13C
1
2

11408
M. Yoshimura et al. / Tetrahedron 63 (2007) 11399–11409
to give (S)-4-3-13C in 88% ee. The H2/HD/D2 and CH3OH/
CH3OD ratios were 99.8:0.2:0 and <1:>99. The (S)-4-h,h-
3-13C/(S)-4-h,d-3-13C/(2R,3S)-4-d,h-3-13C/(2R,3S)-4-d,d-3-
13C ratio and the (R)-4-h,h-3-13C/(R)-4-h,d-3-13C/(2S,3R)-4-
d,h-3-13C/(2S,3R)-4-d,d-3-13C ratio were 57.4:37.8:1.1:3.7
and 82.2:9.3:7.8:0.7. The ratio of the 1H NMR signal inten-
sity at d 2.62–2.71(m) and d 2.53–2.62(m) was 100:95 for
the major enantiomer and 100:92 for the minor enantiomer.
Under all conditions, the ratio of the 1H NMR signal intensi-
ties were quite consistent with the values calculated from the
13C signal intensities, and the C(2)H/C(3)CH3 proton signal
ratio of the recovered (Z)-2-3-13C were determined to be
D2/CH3OD (4 atm): reaction scale, 123 mg (100 mM)
(Z)-2-3-13C; [(R)-1]¼0.5 mM; 4 atm D2; CH3OD; 30 ꢀC;
35.2 h. These conditions converted 99.9% of (Z)-2-3-13C
to give (S)-4-3-13C in 94% ee. The H2/HD/D2 and
CH3OH/CH3OD ratios were 0:2.0:98.0 and 1:99. The
(S)-4-h,h-3-13C/(S)-4-h,d-3-13C/(2R,3S)-4-d,h-3-13C/(2R,3S)-
4-d,d-3-13C ratio and the (R)-4-h,h-3-13C/(R)-4-h,d-3-13C/
(2S,3R)-4-d,h-3-13C/(2S,3R)-4-d,d-3-13C ratio were 0:0:1.1:98.9
and 0:0:0:>99.9. The proton signal ratio of the vinylic
proton and the CH3 of the recovered (Z)-2-3-13C was
determined to be 1.0:3.0 by 1H NMR analysis.
1
1.0:3.0 by H NMR analysis (flip angle, 45ꢀ; repetition
4.5.10. Isotope stability of (2R,3R)-4-d,d. The hydrogena-
tion of tiglic acid (13.7 mg, 0.137 mmol) was carried out
in CH3OH (2.7 mL) containing (2R,3R)-4-d,d with an ee
of 94% (22.3 mg, 0.137 mmol) under 4 atm of H2 at 30 ꢀC
for 48 h. Substrate of 100% was converted to 2-methylbuta-
noic acid. After removal of all the volatiles in vacuo, a por-
tion of the residue was chromatographed on a TLC plate
(20 cmꢁ20 cmꢁ1 mm; eluent, a 4:1 CHCl3/acetone) to
give 4 (4.2 mg). The recovered 4 dissolved in CDCl3 was
time, 10 s; measurement time, 128 scans).
4.5.8. H2–D2/CH3OD conditions. The procedure for the
H2–D2/CH3OD system was the same as that for the H2/
CH3OD system.
H2–D2/CH3OD (4 atm): reaction scale, 202 mg (100 mM)
(Z)-2-3-13C; [(R)-1]¼0.5 mM; 4 atm H2–D2; CH3OD;
30 ꢀC; 9 h. These conditions converted 17% of the (Z)-2-
3-13C to give (S)-4-3-13C in 94% ee. The H2/HD/D2 and
CH3OH/CH3OD ratios were 49.0:2.3:48.7 and <1:>99.
The (S)-4-h,h-3-13C/(S)-4-h,d-3-13C/(2R,3S)-4-d,h-3-13C/
(2R,3S)-4-d,d-3-13C ratio and the (R)-4-h,h-3-13C/(R)-
4-h,d-3-13C/(2S,3R)-4-d,h-3-13C/(2S,3R)-4-d,d-3-13C ratio
were 3.1:45.9:3.1:47.9 and 23.3:20.1:13.3:43.3. The inten-
1
subjected to H NMR analysis. The intensities of the C(2)
and C(3) protons and ee of the recovered 4 were virtually
unchanged, indicating that (2R,3R)-4-d,d is isotopically
stable.
Acknowledgements
1
sity ratios of the H NMR signals at d 2.62–2.71(m) and
d 2.53–2.62(m) were 100:48 for the major enantiomer and
100:43 for the minor enantiomer, respectively.
This work was supported by a Grant-in-Aid for Scientific
Research (Nos. 07CE2004, 05234101, 06226234,
08240101, 00000548, 60000491, and 9900481) from the
Ministry of Education, Science, Sports and Culture, Japan.
The authors would like to thank Professor R. Noyori for
valuable discussions and Messrs. T. Noda, K. Oyama, and
Y. Maeda for their technical support for reaction vessel
production and NMR measurements.
H2–D2/CH3OD (50 atm): reaction scale, 202 mg (100 mM)
(Z)-2-3-13C; [(R)-1]¼0.5 mM; 50 atm H2–D2; CH3OD;
30 ꢀC; 12 min. These conditions converted 10% of (Z)-2-
3-13C to give (S)-3-3-13C in 92% ee. The H2/HD/D2 and
CH3OH/CH3OD ratios were 44.0:0.3:55.7 and <1:>99.
The (S)-4-h,h-3-13C/(S)-4-h,d-3-13C/(2R,3S)-4-d,h-3-13C/
(2R,3S)-4-d,d-3-13C ratio and the (R)-4-h,h-3-13C/(R)-
4-h,d-3-13C/(2S,3R)-4-d,h-3-13C/(2S,3R)-4-d,d-3-13C ratio
were 9.3:32.1:11.9:46.7 and 24.4:15.6:17.8:42.2. The ratio
of the 1H NMR signal intensity at d 2.62–2.71(m) and
d 2.53–2.62(m) was 100:40 for the major enantiomer and
100:38 for the minor enantiomer.
Supplementary data
ORTEP drawings, tables of the atomic parameters, selected
bond distances and bond angles, selected torsion angles, and
anisotropic temperature factors for Ru[(Z)-CH3(C6H5)C]
CHCOO]2[(R)-binap] are available (12 Pages). Supple-
mentary data associated with this article can be found in
the online version, at doi:10.1016/j.tet.2007.08.071.
H2–D2/CH3OD (100 atm): reaction scale, 202 mg (100 mM)
(Z)-2-3-13C; [(R)-1]¼0.5 mM; 100 atm H2–D2; CH3OD;
30 ꢀC; 7 min. These conditions converted 6% of (Z)-2-
3-13C to give (S)-4-3-13C in 88% ee. The H2/HD/D2 and
CH3OH/CH3OD ratios were 48.6:2.0:49.4 and <1:>99.
The (S)-4-h,h-3-13C/(S)-4-h,d-3-13C/(2R,3S)-4-d,h-3-13C/
(2R,3S)-4-d,d-3-13C ratio and the (R)-4-h,h-3-13C/(R)-
4-h,d-3-13C/(2S,3R)-4-d,h-3-13C/(2S,3R)-4-d,d-3-13C ratio
were 12.8:35.1:11.7:40.4 and 30.0:16.7:18.3:35.0. The ratio
of the 1H NMR signal intensity at d 2.62–2.71(m) and d 2.53–
2.62(m) was 100:45 for the major enantiomer and 100:45 for
the minor enantiomer. At all conditions, the proton signal ra-
tio of the vinylic proton and the CH3 of the recovered (Z)-2-
3-13C was determined to be 1.0:3.0 by 1H NMR analysis.
References and notes
1. Arrhenius, S. Z. Phys. Chem. 1889, 4, 226–248.
2. The only exception: Kitamura, M.; Tsukamoto, M.; Bessho, Y.;
Yoshimura, M.; Kobs, U.; Widhalm, M.; Noyori, R. J. Am.
Chem. Soc. 2002, 124, 6649–6667.
3. (a) Sun, Y.; LeBlond, C.; Wang, J.; Blackmond, D. G.;
Laquidara, J.; Sowa, J. R., Jr. J. Am. Chem. Soc. 1995, 117,
12647–12648; (b) Sun, Y.; Wang, J.; LeBlond, C.; Reamer,
R. A.; Laquidara, J.; Sowa, J. R., Jr.; Blackmond, D. G.
J. Organomet. Chem. 1997, 548, 65–72.
4.5.9. D2/CH3OD condition. The procedure was the same as
that for the H2/CH3OD system except for the size of the
reactor (250-mL Schlenk tube) and the purification process.
4. BINAP¼2,20-bis(diphenylphosphino)-1,10-binaphthyl.
5. A similar type of double bond isomerization is also seen in the
hydrogenation of arylalkenes mediated by a chiral phosphine/Ir

M. Yoshimura et al. / Tetrahedron 63 (2007) 11399–11409
11409
catalyst: Hou, D.-R.; Reibenspies, J.; Colacot, T. J.; Burgess, K.
Chem.—Eur. J. 2001, 7, 5391–5400.
6. (a) Detellier, C.; Gelbard, G.; Kagan, H. B. J. Am. Chem. Soc.
1978, 100, 7556–7561; (b) Koenig, K. E.; Knowles, W. S.
J. Am. Chem. Soc. 1978, 100, 7561–7564.
7. (a) Willoughby, C. A.; Buchwald, S. L. J. Am. Chem. Soc. 1994,
116, 8952–8965; (b) Willoughby, C. A.; Buchwald, S. L. J. Am.
Chem. Soc. 1994, 116, 11703–11714.
23. (a) Noyori, R.; Kitamura, M. Modern Synthetic Methods;
Scheffold, R., Ed.; Springer: Berlin, 1989; pp 115–198; (b)
Noyori, R. Asymmetric Catalysis in Organic Synthesis; John
Wiley: New York, NY, 1994; Chapter 2.
24. Clapham, S. E.; Hadzovic, A.; Morris, R. H. Coord. Chem. Rev.
2004, 248, 2201–2237.
25. For a molecular structure of Ru(CH3COO)2[(R)-binap]-
(CH3OH) in catalysis: see Ref. 2.
8. Certain chiral substrates undergo racemization during reaction,
allowing dynamic kinetic resolution. See: (a) Kitamura, M.;
Tokunaga, M.; Noyori, R. J. Am. Chem. Soc. 1993, 115, 144–
152; (b) Noyori, R.; Tokunaga, M.; Kitamura, M. Bull.
Chem. Soc. Jpn. 1995, 68, 36–56.
9. Ishibashi, Y.; Bessho, Y.; Yoshimura, M.; Tsukamoto, M.;
Kitamura, M. Angew. Chem., Int. Ed. 2005, 44, 7287–7290.
10. A report on a concept of the different pathways via same chain
carrier: Sawamura, M.; Kuwano, R.; Ito, Y. J. Am. Chem. Soc.
1995, 117, 9602–9603.
11. Landis, C. R.; Halpern, J. J. Am. Chem. Soc. 1987, 109, 1746–
1754.
12. This is not always the case, see Ref. 13.
13. Kitamura, M.; Yoshimura, M.; Tsukamoto, M.; Noyori, R.
Enantiomer 1996, 1, 281–303.
14. Ohta, T.; Takaya, H.; Kitamura, M.; Nagai, K.; Noyori, R.
J. Org. Chem. 1987, 52, 3174–3176.
15. Uemura, T.; Zhang, X.; Matsumura, K.; Sayo, N.;
Kumobayashi, H.; Ohta, T.; Nozaki, K.; Takaya, H. J. Org.
Chem. 1996, 61, 5510–5516.
16. For a similar double bond isomerization, see Refs. 2 and 6. The
simplest type 2 substrate (Z)-3-methyl-d3-2-butenoic acid also
suffers from isomerization from a,b- to b,g-unsaturated
carboxylic acid.
17. (a) Ohta, T.; Takaya, H.; Noyori, R. Inorg. Chem. 1988, 27,
566–569; (b) Kitamura, M.; Tokunaga, M.; Noyori, R.
J. Org. Chem. 1992, 57, 4053–4054.
26. Ohta, T.; Takaya, H.; Noyori, R. Tetrahedron Lett. 1990, 31,
7189–7192.
27. Enamides: (a) Wiles, J. A.; Bergens, S. H. Organometallics
1998, 17, 2228–2240; (b) Wiles, J. A.; Bergens, S. H.
Organometallics 1999, 18, 3709–3714; Other alkenes: (c)
Ohta, T.; Ikegami, H.; Miyake, T.; Takaya, H. J. Organomet.
Chem. 1995, 502, 169–176; (d) Schmid, R.; Broger, E. A.;
Cereghetti, M.; Crameri, Y.; Foricher, J.; Lalonde, M.;
M€uller, R. K.; Scalone, M.; Schoettel, G.; Zutter, U. Pure
Appl. Chem. 1996, 68, 131–138; (e) Bell, S.; W€ustenberg, B.;
Kaiser, S.; Menges, F.; Netscher, T.; Pfaltz, A. Science 2006,
311, 642–644.
28. Lubell, D. W.; Kitamura, M.; Noyori, R. Tetrahedron:
Asymmetry 1991, 2, 543–554.
29. Dihydride h2-H2: (a) Yu, J.; Spencer, J. B. J. Am. Chem. Soc.
1997, 119, 5257–5258; (b) Yu, J.; Spencer, J. B. J. Org.
Chem. 1997, 62, 8618–8619; (c) Fong, T. P.; Lough, A. J.;
Morris, R. H.; Mezzetti, A.; Rocchini, E.; Rigo, P. J. Chem.
Soc., Dalton Trans. 1998, 2111–2113.
30. Mechanistic study of asymmetric hydrogenation of a,b-unsat-
urated acids with diphosphine–Ru complexes: (a) Chan,
A. S. C.; Chen, C. C.; Yang, T. K.; Huang, J. H.; Lin, Y. C.
Inorg. Chim. Acta 1995, 234, 95–100; (b) Shaharuzzaman,
M.; Chickos, J.; Tam, C. N.; Keiderling, T. A. Tetrahedron:
Asymmetry 1995, 6, 2929–2932; (c) Hardick, D. J.;
Blagbrough, I. S.; Potter, B. V. L. J. Am. Chem. Soc. 1996,
118, 5897–5903; (d) Shaharuzzaman, M.; Braddock-Wiking,
J.; Chickos, J. S.; Tam, C. N.; Silva, R. A. G. D.; Keiderling,
T. A. Tetrahedron: Asymmetry 1998, 9, 1111–1114.
31. The 4-d,d isotopomer may be derived via the RuD species
formed by the H/D exchange between RuH and CH3OD,
see: Bakos, J.; Karaivanov, R.; Laghmari, M.; Sinou, D.
Organometallics 1994, 13, 2951–2956.
18. Ashby, M. T.; Halpern, J. J. Am. Chem. Soc. 1991, 113, 589–
594.
19. (a) Ashby, M. T.; Khan, M. A.; Halpern, J. Organometallics
1991, 10, 2011–2015; (b) Chen, C.-C.; Huang, T.-T.; Lin,
C.-W.; Cao, R.; Chan, A. S. C.; Wong, W. T. Inorg. Chim.
Acta 1998, 270, 247–251.
20. For the mechanism of the H/D exchange between RuH and
CH3OD and between RuH and D2, see: (a) Jessop, P. G.;
Morris, R. H. Coord. Chem. Rev. 1992, 121, 155–284; (b)
Heinekey, D. M.; Mellows, H.; Pratum, T. J. Am. Chem. Soc.
2000, 122, 6498–6499.
32. (a) Klein,J.;Turkel,R.M.J. Am. Chem. Soc. 1969,91, 6186–6187;
(b) Klein, J.; Aminadav, N. J. Chem. Soc. C 1970, 1380–1385.
33. Almy, J.; Cram, D. J. J. Am. Chem. Soc. 1969, 91, 4459–4468.
34. Takahashi, H.; Fujiwara, K.; Ohta, M. Bull. Chem. Soc. Jpn.
1962, 35, 1498–1500.
21. The absolute configurations, 2R,3S and 2S,3R, are altered to
2S,3S and 2R,3R with the 13C-nonlabeled 4-d,h and 4-d,d.
22. Corey, E. J.; Pasto, D. J.; Mock, W. L. J. Am. Chem. Soc. 1961,
83, 2957–2958.
35. Kudo, H.; Fujie, M.; Tanase, M.; Kato, M.; Kurosawa, K.;
Sugai, H.; Umezawa, H.; Matsuzaki, T.; Nagamine, K. Appl.
Radiat. Isot. 1992, 43, 577–583.
36. Hansen, P. E. Annu. Rep. NMR Spectrosc. 1984, 15, 105–234.