Synthesis and evaluation of a new steroidal BINAP type phosphine

Article abstract of DOI:10.1016/S0957-4166(00)00127-0

The short and high yielding synthesis of a new cis-configured bissteroidal phosphine 7 is reported. The comparison of these new phosphines as ligands in ruthenium-based hydrogenation catalysts with the previously synthesized diastereomeric trans-configured phosphines 20 shows that the steroid backbone exerts only a minor influence on the enantioselection of the ruthenium catalysts and confirms that the bissteroidal phosphines behave as 'pseudo'-enantiomers in spite of their diastereomeric nature. Evidence is presented that the mode of catalyst preparation, i.e. catalyst structure, is the crucial reaction parameter which mainly determines the enantiomeric excess of the hydrogenation products. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0957-4166(00)00127-0

Tetrahedron: Asymmetry 11 (2000) 1767±1779
Synthesis and evaluation of a new steroidal BINAP
type phosphine
y
V. Enev, M. Harre, K. Nickisch, M. Schneider and J. T. Mohr*
Process Research, Schering AG-Berlin, MuÈllerstrasse 170±178, D-13342 Berlin, Germany
Received 18 February 2000; accepted 30 March 2000
Abstract
The short and high yielding synthesis of a new cis-con®gured bissteroidal phosphine 7 is reported. The
comparison of these new phosphines as ligands in ruthenium-based hydrogenation catalysts with the
previously synthesized diastereomeric trans-con®gured phosphines 20 shows that the steroid backbone
exerts only a minor in¯uence on the enantioselection of the ruthenium catalysts and con®rms that the
bissteroidal phosphines behave as `pseudo'-enantiomers in spite of their diastereomeric nature. Evidence is
presented that the mode of catalyst preparation, i.e. catalyst structure, is the crucial reaction parameter
which mainly determines the enantiomeric excess of the hydrogenation products. # 2000 Elsevier Science
Ltd. All rights reserved.
1
. Introduction
Recently, we reported the synthesis and application of a trans-con®gured bissteroidal phosphine
1
as an axially chiral hydrogenation ligand. In a ®rst substrate screening our phosphines displayed
higher activities relative to the benchmark phosphine BINAP in the hydrogenations of a-acet-
amidocinnamic and tiglic acid. Although the new ligand met our initial expectations, the success
was somewhat compromized by the high price of the starting material equilenine, which endangered
any further applications, for instance in hydrogenation catalysts.
We now report the economic synthesis of a new bissteroidal phosphine 7 and its application as
a chiral ligand in ruthenium-based asymmetric hydrogenation catalysts.
*
Corresponding author. E-mail: joergtorsten.mohr@schering.de
Postdoctoral Fellow on leave from the Bulgarian Academy of Science, So®a, Bulgaria. Current address: University
of Vienna, Austria.
y
0957-4166/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0957-4166(00)00127-0

1
768
V. Enev et al. / Tetrahedron: Asymmetry 11 (2000) 1767±1779
2
. Results and discussion
Except for our previously applied starting material equilenine no other steroidal monomer is
available which already has included an aromatic A- as well as an aromatic B-ring in its structure.
Therefore, our ®rst goal was to ®nd a simple and high yielding method for the synthesis of these
kinds of structures. We were intrigued by the idea of using estrone 1 as an economically attractive
starting material, because the dehydrogenation of estrone over palladium charcoal is a well-
2
known reaction in steroid chemistry. Two problems were anticipated with this strategy: the ®rst
was the moderate yield of only 48% stated in the literature as well as the long dehydrogenation
reaction time of 3 days which caused a bottleneck in the synthesis of the required larger amounts
of the monomer. This problem could be solved by replacing the original solvent triethyleneglycol
dimethylether with 1-methyl naphthalene which reduced the reaction time from 3 days to 3 h and
allowed the synthesis of isoequilenine 2 in 1 kg batch sizes and high yields (92%) in the pilot plant
(Scheme 1).
Scheme 1. Reagents and conditions: (a) cat. Pd/C (10%), 1-methyl naphthalene, 92%; (b) cat. CuCl(OH)±TMEDA,
i
CH
8
2
Cl
9%
2
, O
2
, 95%; (c) Tf O, Pr
2
2
NEt, crude; (d) cat. PtO
2
, H
2
, 81% (two steps); (e) cat. NiCl
2
dppe, Ph
2
PH, DABCO,

V. Enev et al. / Tetrahedron: Asymmetry 11 (2000) 1767±1779
1769
The second problem lay in the assessment of the isomerization of the proton at C-14 in the
course of the dehydrogenation reaction of estrone. As this isomerization means a change of the
C±D ring junction from trans to the thermodynamically more stable cis form we were confronted
with the question of whether this structural modi®cation would a€ect the properties of the ®nal
bissteroidal phosphine. Although this question could only be answered at the end of our studies,
our initial hypothesis was along the lines of not expecting large di€erences regarding the degree of
enantioselection which will be induced by the two phosphines, because the structural modi®cation
in the backbone is far away from the metal-binding site in the catalyst.
Subsequent to the dehydrogenation of estrone 1 we planned to proceed in the synthesis of 7 with
3
a Wol€±Kishner reduction of 2 followed by a copper-mediated coupling of the deoxygenated
isoequilenine (Scheme 1). Whereas the chemistry from isoequilenine 2 to the diastereomeric
intermediates 4 proceeded uneventfully, we were not able to separate the deoxygenated ligands
R ,S -4 by column chromatography on a preparative scale.
ax ax
Therefore, we decided to perform the coupling step prior to the reduction step and set out to
4
®
nd a racemization free method for the deoxygenation of 3. This goal was accomplished
according to the following transformations: (i) copper mediated coupling of 2 (95% yield) yielding
a 1:1 mixture of the diastereomeric compounds R -3 and S -3 which we were now able to
separate via preparative HPLC even on a kilogram scale; (ii) conversion of the separated ligands
ax
ax
R -3 and S -3 into R -5 and S -5 by reaction with tri¯ic acid anhydride in the presence of
ax
ax
ax
ax
diisopropylethylamine; (iii) selective hydrogenation of the crude tetra tri¯ates R -5 and S -5
ax
ax
5
over 10 mol% PtO at atmospheric pressure yielding bistri¯ates R -6 and S -6 in 81% com-
ax
2
ax
bined yield; and (iv) phosphinylation under previously established reaction conditions a€orded
the diastereomerically pure ligands R -7 and S -7 in 89% yield (Scheme 1).
1,6
ax
ax
Having the new phosphines in hand methyl acetoacetate 8 was chosen as the ®rst substrate to
1
test our new phosphines. To this end ligands 7 and 20 (previously prepared from equilenine)
7
were incorporated into the oligomeric ruthenium complexes [RuCl (R -7)(DMF)] 9a,
ax
2
n
8
[
RuCl (S -7)(DMF)] 9b, [RuCl (R -20)(DMF)] 10a and [RuCl (S -20)(DMF)] 10b. Under
2 ax n 2 ax n 2 ax n
identical reaction conditions 9a,b displayed similar activities as 10a,b and a€orded (R, resp. S)-3-
hydroxybutanoate 11 in quantitative yield (Eq. 1).
ꢀ1†
The asymmetric induction of catalysts 9a,b was only slightly lower (95% ee) than the induction
1
by catalysts 10a,b (98±99% ee ) implying that the structure of the steroid backbone in the
hydrogenation catalyst indeed a€ected the degree of enantioselection but only to a minor extent.
No di€erence concerning the asymmetric induction between the two diastereomers R -7 and S -7
ax
ax
was observed.

1
770
V. Enev et al. / Tetrahedron: Asymmetry 11 (2000) 1767±1779
9
In order to investigate the in¯uence of the catalyst structure on the enantioselection of the
hydrogenation catalysts the following experiments were carried out in parallel runs of catalysts
and 10 and the dimeric catalysts [RuCl (R -7)] (NEt ) 12a, [RuCl (S -7)] (NEt ) 12b,
9
2
ax
2
3
2
ax
2
3
10
[
RuCl (R -20)] (NEt ) 13a and [RuCl (S -20)] (NEt ) 13b.
2 ax 2 3 2 ax 2 3
An in¯uence of the steroid backbone should be indicated by a di€erence in asymmetric induc-
tion of: (i) an S -con®gured vs an R -con®gured ligand; and (ii) ligands with a cis C±D ring
ax
ax
junction vs a trans C±D ring junction. Similarly, an in¯uence of the catalyst structure should be
easily detectable by a di€erence in asymmetric induction of the oligomeric vs the dimeric catalyst
with the same substrate.
Inspection of the results from the hydrogenation of compounds 14±16 with the di€erent catalysts
revealed again a small to moderate in¯uence of the steroid backbone (Table 1). We observed in
most cases a stereoreinforcing interaction of local and axial chirality in the R - and a non-
ax
reinforcing interaction in the S -diastereomers for both phosphines 7 and 20. For the oligomeric
ax
catalysts the greatest di€erence in enantiomeric excess (ee) was 7% for the CD cis-con®gured (9a
vs 9b, substrate 15) and 16% for the CD trans-con®gured phosphines (10a vs 10b, substrate 14).
The dimeric catalysts showed a maximum di€erence in ee of 1% for the cis-series (12a vs 12b, all
substrates) and of 6% for the trans-series (13a vs 13b, substrate 16). The fact that the extent of this
interaction was more pronounced in the case of the oligomeric catalysts 9 and 10 could be taken
as a ®rst hint of the importance of catalyst preparation for the success of the enantioselective
hydrogenation reaction.
Table 1
Comparison of (R ,S )-7 and (R ,S )-20
a
ax ax
ax ax
The in¯uence of the backbone con®guration as measured by the di€erence in enantioselection
between catalysts derived from the structurally di€erent phosphines R -7 vs R -20 [maximum
ax
ax
di€erence of 8% ee in the case of 9a vs 10a (substrate 16) and of 4% ee in the case of 12a vs 13a
substrate 16)] and S -7 vs S -20 [maximum di€erence of 10% ee in the case of 9b vs 10b
(
ax
ax

V. Enev et al. / Tetrahedron: Asymmetry 11 (2000) 1767±1779
1771
(substrate 14) and of 3% ee in the case of 12b vs 13b (substrate 16)] was of similar magnitude to
the di€erences between the R ,S -diastereomers of phosphines 7 and 20. These results led to the
ax ax
conclusion that the absolute con®guration of the steroid backbone was not the primary factor
in¯uencing the degree of enantioselection of the catalysts.
A comparison of the results clearly showed that the ees of compounds 17±19 were mainly
determined by the use of either a dimeric or an oligomeric catalyst. The e€ect brought about by
the variation of the catalyst structure was always larger than the in¯uence of the di€erent
steroidal structure of the ligands which were employed for the catalysts or the interaction e€ect of
axial and local chirality in the steroid backbone. In the case of tiglic acid 16 catalysts 9 and 10
displayed a signi®cantly higher asymmetric induction as well as a better conversion compared to
catalysts 12 and 13 which were obviously not suited for this type of substrate. In contrast, catalysts
1
2 and 13 displayed higher enantioselections as well as smaller ¯uctuations of the degree of
enantioselection compared to catalysts 9 and 10 in the hydrogenation of a-acetamidocinnamic
acid 14 and a-acetamidoacrylic acid 15. The most astonishing example of the in¯uence of the
catalyst structure came from the hydrogenation of 15 where catalysts 12 and 13 a€orded N-acetyl-
alanine 18 with ꢁ96% ee. To our knowledge, these are the highest ees which have ever been
reported for the enantioselective hydrogenation of a-acetamidoacrylic acid 15 with bisaryl or
bisnaphthyl derived phosphines.11
This in¯uence of the catalyst preparation is obviously not a singularity observed with our
steroidal phosphines, as a similar e€ect was noted while we were checking our catalyst preparations
with BINAP as a reference ligand: when using [RuCl (BINAP)(DMF)] in the hydrogenation of
2
n
1
an ee of 83%.
5 an ee of 77% was found for N-acetylalanine 18 while catalyst [RuCl (BINAP)] (NEt ) a€orded
2 2 3
12
The dimeric catalysts 12 and 13 are surprisingly stable as exempli®ed by using material, stored
ꢂ
for several month at ^20 C under argon, which a€orded N-acetylphenylalanine 17 with nearly the
same degree of enantioselectivity [79% ee (13a) and 84% ee (13b)] as freshly prepared catalysts 12
and 13 did. The in¯uence of the hydrogen pressure on the enantioselection of the hydrogenation
was only investigated in the case of tiglic acid 16 with catalysts 9a,b. Similar to the results
described for the hydrogenation of 16 with BINAP±Ru dicarboxylate complexes,13 an increase of
hydrogen pressure from 4 to 7 atm resulted in a dramatic decrease of enantioselection: when
using tiglic acid 16 as starting material and catalyst 9a at 7 atm 2-methyl butyric acid 19 was
obtained in only a 73% ee while tiglic acid 16 as starting material and catalyst 9b at 7 atm
a€orded compound 19 with a moderate ee of only 68%.
In conclusion, we have presented a short and e€ective synthesis of a new bissteroidal phosphine
which circumvents the previous synthetic limitations to this class of phosphines. The evaluation
of the new phosphine 7 and of our previously prepared phosphine 20 supported our initial
hypothesis that the in¯uence of the steroidal backbone on the enantioselectivity of the ruthenium-
based hydrogenation catalysts is only a minor one, thereby con®rming that our bissteroidal
phosphines behave as `pseudo'-enantiomers in spite of their diastereomeric nature. The catalyst
structure was identi®ed as the important reaction parameter which mainly determined the ee of
the hydrogenation products. The dimeric catalysts 12 and 13 displayed higher enantioselectivities
in the hydrogenation of the dehydroamino acids 14 and 15, while the oligomeric catalysts 9 and
1
from the investigation of catalysts 12 and 13 the CD cis-con®gured phosphines R ,S -7 and the
0 were more suited for the hydrogenation of tiglic acid 16. Based on the more consistent results
ax ax
trans-con®gured phosphines R ,S -20 have to be considered as ligands of similar activity.
ax ax
The observed activity di€erences between BINAP and our ligands in the hydrogenation of

1
772
V. Enev et al. / Tetrahedron: Asymmetry 11 (2000) 1767±1779
compounds 14 and 15 cannot be ascribed to the chirality of the steroid backbone of our compounds.
Di€erent electronic properties of our bissteroidal ligands and BINAP may be the reason for the
partly more e€ective asymmetric induction of their ruthenium complexes compared to the
BINAP analogue.14
3
. Experimental
3
.1. General methods
All reactions were carried out under an argon atmosphere with standard techniques for the
exclusion of air and moisture. Glassware used for moisture-sensitive reactions was oven-dried
ꢂ
(200 C). Tetrahydrofuran was distilled from sodium benzophenone ketyl, DMF and dimethyl
acetamide from calcium hydride. Methanol was deoxygenated by a freeze±thaw cycle in which the
solvent was ®rst frozen in liquid N and then warmed up in vacuo (0.1 mmHg). Flash column
2
chromatography was performed on Merck silica gel (grade 60, 230±400 mesh); TLC was
performed on Merck aluminium foils 60 F 254. Infrared (IR) spectra were recorded on a Nicolet
2
mass spectrometer. Optical rotations were recorded on a Perkin±Elmer 241 polarimeter. H (300
0 SBX spectrophotometer. Mass spectral analyses were recorded on a Micromass AutoSpec EQ
1
13
MHz) and C (75 MHz) NMR spectra were recorded in CDCl using TMS or the solvent as
3
internal standard. 31P (121 MHz) NMR spectra were recorded in CDCl using 80% phosphoric
3
acid (external) as standard. All melting points were uncorrected. CD spectra were taken on an
Instruments SA-CD6 spectrometer.
3
3
.2. Procedures
.2.1. Synthesis of R ,S -3
ax ax
To a suspension of isoequilenine 2 (10 g, 38 mmol) in CH Cl (300 ml) was added CuCl(OH)±
2
2
3
ꢂ
TMEDA (0.800 g, 0.35 mmol) and the solution was stirred at 0 C for 10±15 min while bubbling
oxygen through the mixture. The reaction was monitored by TLC. At the end of the reaction HCl
(10%, 5 ml) was added to the mixture and stirred for an additional 15 min (the colour changed
from blue to yellow). The organic layer was washed with brine and dried, and the solvent was
evaporated. The crude mixture (9.9 g) was separated by column chromatography using ethyl
ꢂ
acetate/toluene as eluent to give: compound R -3 (5.2 g, 52%), m.p. 293 C decomp.;
ax
1
[
(
ꢀ]365=+139.5 (c=2.4 in THF); H NMR: 8.17 (1H, d, J=9.3 Hz), 7.40 (1H, d, J=9.3 Hz), 7.17
13
1H, d, J=8.5 Hz), 7.08 (1H, d, J=8.5 Hz), 4.85 (1H, s, OH), 1.19 (3H, s); C NMR: 222.6 (s),
1
4
52.3 (s), 134.1 (s), 133.1 (s), 130.5 (s), 129.1 (d), 127.9 (s), 126.4 (d), 123.0 (d), 117.5 (d), 112.2 (s),
+
7.5 (s), 46.8 (d), 37.0 (t), 28.9 (t), 26.4 (t), 22.3 (t), 19.9 (q); MS-CI: 548 (100, M +1+NH ), 531
3
+
+
(
(
(
20, M +1); MS-EI: 530 (100, M ), 474 (15), 209 (20), 181 (25), 165 (20); HRMS: 530.246399
+
ꢂ
100, M ); calcd for C H O : 530.245709. Compound S -3 (4.3 g, 43%), m.p. 180 C decomp.
1
36
34
4
ax
hexane); [ꢀ]365=+154.0 (c=2.0 in THF); H NMR: 8.17 (1H, d, J=9.3 Hz), 7.42 (1H, d, J=9.3
13
Hz), 7.15 (1H, d, J=8.5 Hz), 7.05 (1H, d, J=8.5 Hz), 5.15 (1H, s, OH), 1.20 (3H, s); C NMR:
2
1
22.7 (s), 152.3 (s), 134.1 (s), 133.2 (s), 130.6 (s), 129.1 (d), 127.9 (s), 126.5 (d), 122.9 (d), 117.6 (d),
12.0 (s), 47.7 (d), 46.9 (s), 37.0 (t), 29.0 (t), 26.4 (t), 22.1 (t), 20.1 (q); MS-CI: 548 (100,
+
+
+
M +1+NH ), 531 (25, M +1); MS-EI: 530 (100, M ), 474 (15), 209 (25), 181 (65), 165 (20);
HRMS: 530.246399 (100, M ); calcd for C H O : 530.245709.
3
+
36
34
4

V. Enev et al. / Tetrahedron: Asymmetry 11 (2000) 1767±1779
1773
3
.2.2. Synthesis of S -6
ax
To a solution of S -3 (2.2 g, 4.16 mmol) and N-ethyldiisopropylamine (4 ml, 23.5 mmol) in
ax
toluene (60 ml) was added over a period of 30 min dropwise tri¯ic anhydride (3.6 ml, 21.94
ꢂ
mmol). The mixture was stirred for 30 min at room temperature and then heated to 60 C and
kept at this temperature for 3 h. At the end of the reaction the mixture was cooled down and the
upper (toluene) layer was separated. The organic phase was washed with water, 2N HCl and
brine, dried with Na SO and the solvent was evaporated. The residue was ®ltered through SiO
2
2
4
(hexane:ethyl ether, 15:1, to give the crude S -5 (3.8 g, 3.59 mmol, 86%) which was dissolved in
ax
ethanol:methylene chloride (30 ml, 10:1) and exposed to hydrogen (1 atm) in the presence of PtO2
0.2 g) for 2 h. The mixture was ®ltered through a pad of Celite, the solvent was evaporated and
the crude product was dissolved in ethyl acetate. The solution was washed with NaHCO and
(
3
brine, and dried with Na SO . The solvent was evaporated and the crude product was puri®ed on
2
4
0
SiO to yield (S )-4,4 -bis(14-epi-3-tri¯uoromethylsulfonyloxyestra-1,3,5(10),6,8-pentaene) S -6:
ax
2.6 g, 3.39 mmol, 81.5%), m.p. 115±116 C; [ꢀ] =+128.8 (c=0.1 in THF); H NMR: 8.28 (1H,
2
a
ꢂ
1
(
D
d, J=10.0 Hz), 7.60 (1H, d, J=10.0 Hz), 7.12 (1H, d, J=9.0 Hz), 6.98 (1H, d, J=9.0 Hz), 1.10
(3H, s); C NMR: 144.8 (s), 139.2 (s), 132.0 (s), 131.4 (s), 130.7 (d), 130.4 (s), 126.6 (d), 124.5 (d),
13
1
7
7
20.3 (s), 118.5 (d), 50.7 (d), 40.7 (t), 39.3 (s), 35.4 (t), 31.4 (t), 25.7 (q), 23.2 (t), 22.7 (t); MS-CI:
+
+
84 (100, M +1+NH ); MS-EI: 766 (80, M ), 633 (17) 483 (100), 387 (21), 308 (17); HRMS:
3
66.185682; calcd for C H F O S : 766.185752.
38
36
6
6 2
3
.2.3. Synthesis of R -6
ax
The compound was synthesized according to the same procedure as for S -6: m.p. 204±
ax
ꢂ
1
2
Hz), 7.11 (1H, d, J=9.0 Hz), 7.01 (1H, d, J=9.0 Hz), 1.10 (3H, s); C NMR: 144.3 (s), 139.2 (s),
04.5 C; [ꢀ] =^23.9 (c=0.1 in THF); H NMR: 8.28 (1H, d, J=10.0 Hz), 7.60 (1H, d, J=10.0
D
13
1
3
32.0 (s), 131.4 (s), 130.8 (d), 130.5 (s), 127.0 (d), 124.6 (d), 124.9 (s), 118.5 (d), 50.6 (d), 40.6 (t),
+
9.6 (s), 35.4 (t), 31.5 (t), 25.7 (q), 23.3 (t), 22.8 (t); MS-CI: 784 (88, M +1+NH ); MS-EI: 766
3
+
(100, M ), 633 (17), 483 (75), 389 (26), 309 (7); HRMS: 766.185685; calcd for C H F O S :
66.185752.
38
36
6
6 2
7
3
.2.4. Synthesis of S -7
ax
To a solution of NiCl ±dppe (0.070 g, 0.13 mmol) in dimethyl acetamide (2 ml) was added
2
diphenylphosphine (0.13 ml, 0.139 g, 0.75 mmol) at room temperature, and the solution was
ꢂ
heated to 100 C. After 45 min a solution of tri¯ate S -6 (1 g, 1.3 mmol) and 1,4-diazabi-
ax
cyclo[2.2.2]octane (0.62 g, 5.5 mmol) in dimethyl acetamide (4 ml) was added at once, the result-
ꢂ
ing green solution was kept at 100 C and three additional portions of diphenylphosphine (0.13 ml
ꢂ
each) were added after 1, 3 and 5 h, respectively. The reaction was kept at 100 C for 6 days and
then the dark brown solution was diluted with MeOH. The desired product was ®ltered and the
®lter cake was washed with MeOH and dried under vacuum. The crude product (1.1 g) was
recrystallized from MeOH:toluene, 10:1, to give pure S -7 (0.97g, 1.156 mmol, 89%); the m.p. of
ax
1
the compound was not determined due to air sensitivity: [ꢀ] =^79.3 (c=0.13 in THF); H NMR:
D
7.98 (1H, d, J=10.0 Hz), 7.36 (1H, brd, J=10.0 Hz), 7.3±6.9 (10H, m), 6.52 (2H, s), 3.12 (1H, dt,
J=3.2, 21.0 Hz), 2.95 (1H, m), 2.52 (1H, t, J=7.5 Hz), 2.1 (2H, m), 1.7 (1H, m), 1.9±1.3 (5H, m),
1
3
0
(
.98 (3H, s); C NMR: 146.4 (s), 146.0 (s), 138.2 (s), 133.1 (s), 132.8 (s), 132.0 (s), 129.5 (s), 129.9
s), 130.3, 128.8, 128.3, 127.9, 127.4, 125.6, 123.2, 50.0 (each d), 40.3 (s), 38.7 (t), 35.8 (t), 25.0 (t),
31
+
24.6 (t), 20.8 (t), 16.4 (q); P NMR: ^15.2 (s); MS-HRFAB: 839.3927 (50, M +1); calcd for
C H P : 839.3936.
6
0
57 2

1
774
V. Enev et al. / Tetrahedron: Asymmetry 11 (2000) 1767±1779
3
.2.5. Synthesis of R -7
ax
The compound was synthesized according to the same procedure as for S -7; the m.p. of the
ax
1
compound was not determined due to air sensitivity: [ꢀ] =+160.8 (c=0.16 in THF); H NMR:
D
8
.08 (1H, d, J=10.0 Hz), 7.46 (1H, brd, J=10.0 Hz), 7.3±7.0 (10H, m), 6.62 (2H, ABq, J=12Hz),
13
3
NMR: 146.4 (s), 146.0 (s), 138.2 (s), 133.1 (s), 132.6 (s), 132.0 (s), 130.1 (s), 129.9 (s), 130.3, 128.8,
.12 (2H, m), 2.62 (1H, t, 7.5 Hz), 2.1 (2H, m), 1.7 (1H, m), 1.9±1.3 (5H, m), 1.10 (3H, s);
C
1
2
28.3, 127.9, 127.4, 125.6, 123.2, 50.0 (each d), 40.8 (t), 39.2 (s), 35.3 (t), 31.6 (t), 25.7 (q), 23.2 (t),
2.8 (t); 31P NMR: ^15.2 (s); MS-HRFAB: 839.3919 (50, M +1); calcd for C H P : 839.3936.
+
60
57 2
3
.2.6. Hydrogenation of methyl acetoacetate 8 in the presence of RuCl (S -7)(DMF) 9b
2 ax n
1
A dry 50 ml Schlenk tube containing a Te¯on coated stirring bar was charged with [RuCl -
2
(benzene)] (0.018 g, 0.036 mmol), S -7 (0.06 g, 0.072 mmol) and DMF (1 ml). The resulting
brown suspension was heated at 100 C under argon for 30 min to give a clear reddish-brown
2
ax
ꢂ
ꢂ
solution. The reaction mixture was allowed to cool at 50 C during which it was concentrated at 1
mmHg and then at 0.05 mmHg for 1 h to give RuCl (S -7)(DMF) 9b. To the resulting reddish-
2
ax
n
brown solid of 9b was added under argon a solution of methyl acetoacetate 8 (7.3 g, 63 mmol) in
degassed methanol (40 ml) and the mixture was stirred for 5 min. Then the solution was trans-
ꢂ
ferred to a 125 ml stainless steel autoclave and kept for 1 h at 100 C under hydrogen (100 atm)
followed by 10 h at room temperature at 100 atm. After the excess hydrogen had been blown o€,
ꢂ
the apparatus was disassembled. The content was concentrated. Distillation (110 C, 46 mmHg)
a€orded methyl (S)-3-hydroxybutanoate 11 (7.2 g, 98%): [ꢀ] =+49.4 (c=1.5, in CHCl ), in 95%
D
3
ee assayed as MTPA ester; lit. (Fluka), [ꢀ] =+50.5 (c=1.4 in CHCl ).
D
3
3
.2.7. Hydrogenation of methyl acetoacetate 8 in the presence of RuCl (R -7)(DMF) 9a
2 ax n
According to the same procedure hydrogenation of 8 (7.3 g) with 9a (from [RuCl (benzene)]2
2
(0.018 g, 0,036 mmol) and R -7 (0.06 g, 0.072 mmol)) a€orded (R)-3-hydroxybutanoate 11 (7.20
ax
g, 98%): [ꢀ] =^49.6 (c=1.5, in CHCl ), in 95% ee assayed as MTPA ester.
D
3
3
.2.8. Hydrogenation of ꢀ-acetamidocinnamic acid 14 in the presence of 9b
To the reddish-brown solid of 9b (from [RuCl (benzene)] (0.018 g, 0.036 mmol), S -7 (0.06 g,
2
2
ax
0
.072 mmol)) was added a solution of a-acetamidocinnamic acid 14 (3.6 g, 17.6 mmol) in degas-
sed methanol (40 ml) and the mixture was stirred for 5 min. Then the solution was transferred to
a 125 ml stainless steel autoclave and kept for 48 h at room temperature under hydrogen (7 atm).
After the excess hydrogen had been blown o€, the apparatus was disassembled. The content was
concentrated to give 3.5 g crude product. A small amount of the crude product (0.05 g) was
treated with an excess of diazomethane (solution in ethyl ether) to yield the corresponding methyl
ꢂ
ester for determination of the ee. GC analysis (25 m Chiralsil-DEX, 150 isoterm, 1 bar hydrogen,
ꢂ
ꢂ
ꢂ
ꢂ
t detector=280 C, t column=280 C) of this ester shows an ee of 79.5% and no trace of the
starting material. The rest of the product was dissolved in hot water (200 ml) and washed with
toluene (2Â30 ml). The water was evaporated and the residue was dried under vacuum to give 3.5
g (97%) N-acetyl-(S)-phenylalanine 17: [ꢀ] =+30.5 (c=1.2, in methanol).
D
3
.2.9. Hydrogenation of ꢀ-acetamidocinnamic acid 14 in the presence 9a
According to the same procedure, hydrogenation of 14 (3.0 g, 14.6 mmol) with 9a (from
[RuCl (benzene)] (0.01 g, 0.020 mmol), R -7 (0.035 g, 0.042 mmol)) gave 3 g crude product. A
2 2 ax
small amount of the crude product (0.05 g) was treated with an excess of diazomethane (solution

V. Enev et al. / Tetrahedron: Asymmetry 11 (2000) 1767±1779
1775
in ethyl ether) to yield the corresponding methyl ester for determination of the ee. GC analysis
ꢂ ꢂ ꢂ ꢂ ꢂ
(25 m Chiralsil-DEX, 150 isoterm, 1 bar hydrogen, t detector=280 C, t column=280 C) of
this ester shows an ee of 82% and no trace of the starting material. The rest of the product was
dissolved in hot water (200 ml) and washed twice with toluene (30 ml each). The water was
evaporated and the residue was dried under vacuum to give 2.9 g (96%) N-acetyl-(R)-phenylalanine
1
7: [ꢀ] =^32.1 (c=1.0, in methanol).
D
3
.2.10. Hydrogenation of ꢀ-acetamidoacrylic acid 15 with RuCl (R -20)(DMF) 10a
2
ax
n
To the solution of a-acetamidoacrylic acid 15 (1.4 g, 10.8 mmol) in degassed methanol (40 ml)
was added 10a (from [RuCl (benzene)] (0.007 g, 0.014 mmol), R -20 (0.02 g, 0.024 mmol)) and
2
2
ax
the mixture was stirred for 5 min. Then the solution was transferred to a 125 ml stainless steel
autoclave and kept for 48 h under hydrogen (7 atm). After the excess hydrogen had been blown
o€, the apparatus was disassembled. The content was concentrated to give 1.3 g crude product
1
8. A small amount of the crude product (0.05 g) was treated with an excess of diazomethane
solution in ethyl ether) to yield the corresponding methyl ester for determination of the ee. GC
(
analysis (50 m FS-LIPODEX-E N 211, 145 isoterm, 1 bar hydrogen, t detector=280 C, t
column=280 C) of this ester shows an ee of 72% and >98% conversion.
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
3
.2.11. Hydrogenation of ꢀ-acetamidoacrylic acid 15 with RuCl (S -20)(DMF) 10b
2 ax n
According to the same procedure, hydrogenation of 15 (1.25 g, 9.7 mmol) in degassed methanol
50 ml) with 10b (from [RuCl (benzene)] (0.007 g, 0.014 mmol), S -20 (0.02 g, 0.024 mmol))
gave 1.2 g crude product 18. A small amount of the crude product (0.05 g) was treated with an
(
2
2
ax
excess of diazomethane (solution in ethyl ether) to yield the corresponding methyl ester for
ꢂ
ꢂ
determination of the ee. GC analysis (50 m FS-LIPODEX-E N 211, 145 isoterm, 1 bar hydro-
ꢂ
ꢂ
ꢂ
ꢂ
gen, t detector=280 C, t column=280 C) of this ester shows an ee of 66% and >98%
conversion.
3
.2.12. Hydrogenation of ꢀ-acetamidoacrylic acid 15 with 9a
According to the same procedure, hydrogenation of 15 (1.4 g, 10.8 mmol) in degassed methanol
(50 ml) with 9a (from [RuCl (benzene)] (0.012 g, 0.024 mmol), R -7 (0.035 g, 0.042 mmol)) gave
2 2 ax
1
.2 g crude product 18. A small amount of the crude product (0.05 g) was treated with an excess
of diazomethane (solution in ethyl ether) to yield the corresponding methyl ester for determina-
ꢂ
ꢂ
ꢂ
tion of the ee. GC analysis (50 m FS-LIPODEX-E N 211, 145 isoterm, 1 bar hydrogen, t
ꢂ
ꢂ
ꢂ
detector=280 C, t column=280 C) of this ester shows an ee of 75% and >98% conversion.
3
.2.13. Hydrogenation of ꢀ-acetamidoacrylic acid 15 with 9b
According to the same procedure, hydrogenation of 15 (1.2 g, 9.3 mmol) in degassed methanol
40 ml) with 9b (from [RuCl (benzene)] (0.014 g, 0.028 mmol), S -7 (0.04 g, 0.048 mmol)) gave
(
2
2
ax
1
of diazomethane (solution in ethyl ether) to yield the corresponding methyl ester for determina-
.15 g crude product 18. A small amount of the crude product (0.05 g) was treated with an excess
ꢂ
ꢂ
ꢂ
tion of the ee. GC analysis (50 m FS-LIPODEX-E N 211, 145 isoterm, 1 bar hydrogen, t
ꢂ
ꢂ
ꢂ
detector=280 C, t column=280 C) of this ester shows an ee of 68% and >98% conversion.
3
.2.14. Hydrogenation of tiglic acid 16 in the presence of 9a
To the reddish-brown solid of 9a (from [RuCl (benzene)] (0.015 g, 0.030 mmol), R -7 (0.045
g, 0.054 mmol)) was added a solution of tiglic acid 16 (0.83 g, 8.3 mmol) in degassed methanol (35
2
2
ax

1
776
V. Enev et al. / Tetrahedron: Asymmetry 11 (2000) 1767±1779
ml) and the mixture was stirred for 5 min. Then the solution was transferred to a 125 ml stainless
steel autoclave and kept for 24 h at room temperature under hydrogen (4 atm). After the excess
hydrogen had been blown o€, the apparatus was disassembled. The content was concentrated to
give 0.8 g crude product. A small amount of the crude product (0.05 g) was treated with an excess
ꢂ
of diazomethane (solution in ethyl ether). GC analysis (25 m Chiralsil-DEX, 150 isoterm, 1 bar
ꢂ
ꢂ
ꢂ
ꢂ
hydrogen, t detector=280 C, t column=280 C) of this solution shows 82% ee and no trace of
ꢂ
the starting material. Distillation of the crude product (78 C, 24 mmHg) a€orded (R)-2-methyl-
butyric acid 19 (0.7 g, 84%, [ꢀ] =^13.9 (neat)).
D
3
.2.15. Hydrogenation of tiglic acid 16 in the presence of 9b
According to the same procedure, hydrogenation of tiglic acid 16 (1.0 g, 10 mmol) in degassed
methanol (40 ml) with 9b (from [RuCl (benzene)] (0.018 g, 0.036 mmol), S -7 (0.055 g, 0.066
ax
2
2
mmol)) gave 1.0 g crude product. A small amount of the crude product (0.05 g) was treated with
ꢂ
an excess of diazomethane (solution in ethyl ether). GC analysis (25 m Chiralsil-DEX, 150 isoterm,
ꢂ
ꢂ
ꢂ
ꢂ
1
no trace of the starting material. Distillation of the crude product (78 C, 24 mmHg) a€orded
bar hydrogen, t detector=280 C, t column=280 C) of this solution shows an ee of 85% and
ꢂ
(
S)-2-methylbutyric acid 19 (0.9 g, 90%, [ꢀ] =+12.9 (neat)).
D
3
.2.16. General procedure for the synthesis of [RuCl (ligand)] TEA
2 2
To a suspension of phosphine (1 g, 1.2 mmol) and [RuCl COD] (0.33 g, 1.2 mmol) in degassed
2
2
ꢂ
xylene (25 ml) was added TEA (2.8 ml, 20 mmol) and the mixture was heated at 140 C in a closed
vessel. After 4 h the apparatus was allowed to cool to room temperature and the solvent was
evaporated under vacuum. The crude red product was recrystallized from hot MTB (10 ml) to
give the desired catalyst as orange crystals (0.28 g). A crystalline product (0.42 g) was isolated
additionally from the solution after cooling. The product was identical (from 31P NMR) with the
®
rst one.
[
RuCl (R -20)] TEA 13a: 31P NMR, two isomers in a 3:1 ratio: 54.3 (d, J=42.5 Hz), 50.8 (d,
J=42.5 Hz), 45.1 (d, J=42 Hz), 40.0 (d, J=42 Hz).
2
ax
2
31
[
[
RuCl (S -20)] TEA 13b: P NMR, one isomer: 54.8 (d, J=42.5 Hz), 51.3 (d, J=42.5 Hz).
2 ax 2
31
RuCl (R -7)] TEA 12a: (0.8 g) P NMR, one isomer: 52.8 (d, J=42.5 Hz), 50.3 (d, J=42.5
2
2
ax
Hz).
[
RuCl (S -7)] TEA 12b: (0.7 g) 31P NMR, shows three isomers in a 2:1:0.5 ratio: 54.5 (d,
J=42.0 Hz), 51.1 (d, J=42.0 Hz), 49.3 (d, J=42.5 Hz), 44.9 (d, J=42.5 Hz), 44.9 (d, J=40.0 Hz),
2 ax 2
3
9.7 (d, J=40.0 Hz).
3
.2.17. Hydrogenation of ꢀ-acetamidocinnamic acid 14 with 13a
To the solution of a-acetamidocinnamic acid 14 (1.0 g, 4.87 mmol) in degassed EtOH:THF, 1:1
50 ml), was added 13a (0.05 g, 0.024 mmol) and the mixture was stirred for 5 min. Then the
solution was transferred to a 125 ml stainless steel autoclave and kept for 24 h at 35 C under
hydrogen (2 atm). After the excess hydrogen had been blown o€, the apparatus was disassembled.
The content was concentrated to give 0.98 g crude product. A small amount of the crude product
(
ꢂ
(0.05 g) was treated with an excess of diazomethane (solution in ethyl ether) to yield the corresponding
methyl ester for determination of the ee. GC analysis (25 m Chiralsil-DEX, 150 isoterm, 1 bar
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
hydrogen, t detector=280 C, t column=280 C) of this ester shows an ee of 84% and no trace
of the starting material. The rest of the product was dissolved in hot water (200 ml) and washed

V. Enev et al. / Tetrahedron: Asymmetry 11 (2000) 1767±1779
1777
twice with toluene (30 ml each). The water was evaporated and the residue was dried under
vacuum to give 0.9 g (90%) N-acetyl-(R)-phenylalanine 17: [ꢀ] =^31.8 (c=1.0, in methanol).
D
3
.2.18. Hydrogenation of ꢀ-acetamidocinnamic acid 14 with 13b
According to the same procedure, hydrogenation of 14 (0.9 g, 4.39 mmol) in degassed
EtOH:THF, 1:1 (50 ml), with 13b (0.05 g, 0.024 mmol) gave 0.88 g crude product. A small
amount of the crude product (0.05 g) was treated with an excess of diazomethane (solution in
ethyl ether) to yield the corresponding methyl ester for determination of the ee. GC analysis (25
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
m Chiralsil-DEX, 150 isoterm, 1 bar hydrogen, t detector=280 C, t column=280 C) of this
ester shows an ee of 87% and >98% conversion. The rest of the product was dissolved in hot
water (200 ml) and washed twice with toluene (30 ml each). The water was evaporated and the
residue was dried under vacuum to give 0.8 g (88%) N-acetyl-(S)-phenylalanine 17: [ꢀ] =+33.5
D
(c=1.0, in methanol).
3
.2.19. Hydrogenation of ꢀ-acetamidocinnamic acid 14 with 12a
According to the same procedure, hydrogenation of 14 (0.9 g, 4.39 mmol) in degassed
EtOH:THF, 1:1 (50 ml), with 12a (0.05 g, 0.024 mmol) gave 0.88 g crude product. A small
amount of the crude product (0.05 g) was treated with an excess of diazomethane (solution in
ethyl ether) to yield the corresponding methyl ester for determination of the ee. GC analysis (25
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
m Chiralsil-DEX, 150 isoterm, 1 bar hydrogen, t detector=280 C, t column=280 C) of this
ester shows an ee of 87% and >98% conversion. The rest of the product was dissolved in hot
water (200 ml) and washed with toluene (2Â30 ml). The water was evaporated and the residue
was dried under vacuum to give 0.8 g (88%) N-acetyl-(R)-phenylalanine 17: [ꢀ] =^33.9 (c=1.0,
D
in methanol).
3
.2.20. Hydrogenation of ꢀ-acetamidocinnamic acid 14 with 12b
According to the same procedure, hydrogenation of 14 (0.9 g, 4.39 mmol) in degassed
EtOH:THF, 1:1 (50 ml), with 12b (0.05 g, 0.024 mmol) gave 0.88 g crude product. A small
amount of the crude product (0.05 g) was treated with an excess of diazomethane (solution in
ethyl ether) to yield the corresponding methyl ester for determination of the ee. GC analysis (25
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
m Chiralsil-DEX, 150 isoterm, 1 bar hydrogen, t detector=280 C, t column=280 C) of this
ester shows an ee of 86% and >98% conversion. The rest of the product was dissolved in hot
water (200 ml) and extracted twice with toluene (30 ml each). The water was evaporated and the
residue was dried under vacuum to give 0.8 g (88%) N-acetyl-(S)-phenylalanine 17: [ꢀ] =+33.2
D
(c=1.0, in methanol).
3
.2.21. Hydrogenation of ꢀ-acetamidoacrylic acid 15 with 13a
To the solution of a-acetamidoacrylic acid 15 (1.0 g, 7.7 mmol) in degassed EtOH:THF, 1:1 (50
ml), was added 13a (0.089 g, 0.042 mmol) and the mixture was stirred for 5 min. Then the solution
ꢂ
was transferred to a 125 ml stainless steel autoclave and kept for 24 h at 35 C under hydrogen (2
atm). After the excess hydrogen had been blown o€, the apparatus was disassembled. The content
was concentrated to give 0.98 g crude product 18. A small amount of the crude product (0.05 g)
was treated with an excess of diazomethane (solution in ethyl ether) to yield the corresponding
ꢂ
ꢂ
methyl ester for determination of the ee. GC analysis (50 m FS-LIPODEX-E N 211, 145 isoterm,
ꢂ
ꢂ
ꢂ
ꢂ
1
>
bar hydrogen, t detector=280 C, t column=280 C) of this ester shows an ee of 97% and
98% conversion.

1
778
V. Enev et al. / Tetrahedron: Asymmetry 11 (2000) 1767±1779
3
.2.22. Hydrogenation ꢀ-acetamidoacrylic acid 15 with 13b
According to the same procedure, hydrogenation of 15 (0.8 g, 6.2 mmol) in degassed
EtOH:THF, 1:1 (50 ml), with 13b (0.062 g, 0.029 mmol) gave 0.8 g crude product 18. A small
amount of the crude product (0.05 g) was treated with an excess of diazomethane (solution in
ethyl ether) to yield the corresponding methyl ester for determination of the ee. GC analysis (50 m
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
FS-LIPODEX-E N 211, 145 isoterm, 1 bar hydrogen, t detector=280 C, t column=280 C)
of this ester shows an ee of 96% and >98% conversion.
3
.2.23. Hydrogenation of ꢀ-acetamidoacrylic acid 15 with 12a
According to the same procedure, hydrogenation of 15 (0.8 g, 6.2 mmol) in degassed
EtOH:THF, 1:1 (50 ml), with 12a (0.068 g, 0.032 mmol) gave 0.8 g crude product 18. A small
amount of the crude product (0.05 g) was treated with an excess of diazomethane (solution in
ethyl ether) to yield the corresponding methyl ester for determination of the ee. GC analysis (50 m
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
FS-LIPODEX-E N 211, 145 isoterm, 1 bar hydrogen, t detector=280 C, t column=280 C)
of this ester shows an ee of 97% and >98% conversion.
3
.2.24. Hydrogenation of ꢀ-acetamidoacrylic acid 15 with 12b
According to the same procedure, hydrogenation of 15 (0.8 g, 6.2 mmol) in degassed
EtOH:THF, 1:1 (50 ml), with 12b (0.064 g, 0.030 mmol) gave 0.8 g crude product 18. A small
amount of the crude product (0.05 g) was treated with an excess of diazomethane (solution in
ethyl ether) to yield the corresponding methyl ester for determination of the ee. GC analysis (50 m
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
FS-LIPODEX-E N 211, 145 isoterm, 1 bar hydrogen, t detector=280 C, t column=280 C)
of this ester shows an ee of 96% and >98% conversion.
3
.2.25. Hydrogenation of tiglic acid 16 with 13a
To the solution of tiglic acid 16 (1.0 g, 10 mmol) in degassed EtOH:THF, 1:1 (50 ml), was
added 13a (0.089 g, 0.042 mmol) and the mixture was stirred for 5 min. Then the solution was
ꢂ
transferred to a 125 ml stainless steel autoclave and kept for 24 h at 35 C under hydrogen (2 atm).
After the excess hydrogen had been blown o€, the apparatus was disassembled. The content was
concentrated to give 1.0 g crude product 19. A small amount of the crude product (0.05 g) was
treated with excess of diazomethane (solution in ethyl ether) to yield the corresponding methyl
ꢂ
ester for determination of the ee. GC analysis (25 m Chiralsil-DEX, 150 isoterm, 1 bar hydrogen,
ꢂ
ꢂ
ꢂ
ꢂ
t detector=280 C, t column=280 C) of this ester shows an ee of 65% and about 40% con-
version.
3
.2.26. Hydrogenation of tiglic acid 16 with 13b
According to the same procedure, hydrogenation of 16 (1.0 g, 10 mmol) in degassed
EtOH:THF, 1:1 (50 ml), with 13b (0.070 g, 0.033 mmol) gave 1 g of crude product 19. A small
amount of the crude product (0.05 g) was treated with an excess of diazomethane (solution in
ethyl ether) to yield the corresponding methyl ester for determination of the ee. GC analysis (25
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
m Chiralsil-DEX, 150 isoterm, 1 bar hydrogen, t detector=280 C, t column=280 C) of this
ester shows an ee of 71% and about 40% conversion.
3
.2.27. Hydrogenation of tiglic acid 16 with 12a
According to the same procedure, hydrogenation of 16 (1.18 g, 11.8 mmol) in degassed
EtOH:THF, 1:1 (50 ml), with 12a (0.082 g, 0.039 mmol) gave 1.1 crude product 19. A small

V. Enev et al. / Tetrahedron: Asymmetry 11 (2000) 1767±1779
1779
amount of the crude product (0.05 g) was treated with an excess of diazomethane (solution in
ethyl ether) to yield the corresponding methyl ester for determination of the ee. GC analysis (25
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
m Chiralsil-DEX, 150 isoterm, 1 bar hydrogen, t detector=280 C, t column=280 C) of this
ester shows an ee of 69% and about 40% conversion.
3
.2.28. Hydrogenation of tiglic acid 16 with 12b
According to the same procedure, hydrogenation of tiglic acid 16 (1.2 g, 12 mmol) in degassed
EtOH:THF, 1:1 (50 ml), with 12b (0.089 g, 0.042 mmol) gave 1.1 g of crude product 19. A small
amount of the crude product (0.05 g) was treated with excess of diazomethane (solution in ethyl
ether) to yield the corresponding methyl ester for determination of the ee. GC analysis (25 m
ꢂ
ꢂ
ꢂ
ꢂ
ꢂ
Chiralsil-DEX, 150 isoterm, 1 bar hydrogen, t detector=280 C, t column=280 C) of this ester
shows an ee of 68% and about 40% conversion.
Acknowledgements
Financial support for V.E. by Schering's Unternehmenforschungsprogramm is gratefully
acknowledged. We thank Th. Hilsmann, C. Pieper and M. Stoppa for experimental assistance
and H. Igogeit for GC measurements.
References
1
2
3
4
. Enev, V.; Ewers, Ch. L. J.; Harre, M.; Nickisch, K.; Mohr, J. T. J. Org. Chem. 1997, 62, 7092.
. Pelletier, S. H.; Ichinohe, Y.; Herald, D. L. Tetrahedron Lett. 1971, 4179.
. Noji, M.; Nakajima, M.; Koga, K. Tetrahedron Lett. 1994, 43, 7983.
. The exposure of pure diastereomers R ,S -3 as well as methoxy methyl-protected R ,S -3 to the conditions of a
ax ax
ax ax
Wol€±Kishner reduction a€orded a fully racemized starting material as well as a fully racemized product.
. Jigajinni, V. B.; Wightman, R. H. Tetrahedron Lett. 1982, 23, 117.
. (a) Cai, D.; Payack, J. F.; Verhoeven, T. R. US Patent 5,399,771. (b) Cai, D.; Payack, J. F.; Bender, D. R.;
Hughes, D. L.; Verhoeven, T. R.; Reider, P. J. J. Org. Chem. 1994, 59, 7180.
. Ager, D. J.; Laneman, S. A. Tetrahedron: Asymmetry 1997, 8, 3327.
5
6
7
8
9
. Kitamura, M.; Tokunaga, M.; Ohkuma, T.; Noyori, R. Tetrahedron Lett. 1991, 32, 4163.
. Catalyst structure refers to the di€erent modes of catalyst preparation, as the structures of the active catalysts are
not known. 31P NMR spectroscopic investigations of [RuCl
single species, while compound mixtures were found for Rax-20 and Sax-7.
2
2 3
(ligand)] (NEt ) with Sax-20 and Rax-7 hinted at a
1
0. (a) King, A.; Thompson, A.; Verhoeven, T. J. Org. Chem. 1992, 57, 6689. (b) King, S. A.; DiMichele, L. In
Catalysis of Organic Reactions; Dekker: New York, 1995; pp. 157±166. This assignment was arbitrarily chosen as we
could not distinguish between the possible catalyst structures [RuCl (ligand)]
2
2
(NEt ) and [Ru
3
2
Cl (ligand)
5
2
](NH Et
2
2
)
1
due to overlapping signals in the H NMR.
1
1
1. Genet, J.-P. In Advanced Asymmetric Synthesis; Stephenson, G. R., Ed.; Chapman & Hall, 1996.
2. see:Ikariya, T. et. al. Chem. Commun. 1985, 922: Ikariya reported 86% ee for the hydrogenation of 14 and 76% ee
for the hydrogenation of 15 with [RuCl (BINAP)] (NEt ).
3. Ohta, T.; Takaya, H.; Kitamura, M.; Nagai, K.; Noyori, R. J. Org. Chem. 1987, 52, 3176.
4. See, for instance, the comparison of BINAP and H -BINAP in: Takaya et al. J. Chem. Soc., Perkin Trans. 1 1994,
309.
2
2
3
1
1
8
2