Ruthenium(II)-Catalyzed Asymmetric Transfer Hydrogenation of Carbonyl Compounds with 2-Propanol and Ephedrine-Type Ligands

Article abstract of DOI:10.1002/adsc.200390030

This account describes the development and application of Noyori's type catalysts based on ruthenium-arene complexes and simple chiral β-amino alcohols derived from ephedrine, for the asymmetric transfer hydrogenation of 2-propanol to carbonyl substrates. The influence of key parameters of the catalyst system has been studied systematically, resulting in particular in the design of the novel ligand (4-biphenylmethyl)norephedrine. Thanks to the latter, the catalytic precursors and true active species could be isolated for the first time, enabling a complete structural description of the catalytic cycle and of probable deactivation pathways. Highly effective applications of those catalysts systems, i.e., the asymmetric reductions of simple aryl ketones and aryl β-keto esters, the synthesis of chiral phthalides and syn-β,δ-dihydroxy esters, are described.

Full text of DOI:10.1002/adsc.200390030

REVIEWS
Ruthenium(II)-Catalyzed Asymmetric Transfer Hydrogenation of
Carbonyl Compounds with 2-Propanol and Ephedrine-Type
Ligands
a
a
b,
¬
Kathelyne Everaere, Andre Mortreux, Jean-FranÁois Carpentier *
a
Laboratoire de Catalyse de Lille, UPRESA 8010 CNRS, 59652 Villeneuve d×Ascq, France
Laboratoire Organometalliques et Catalyse, UMR 6509 CNRS Universite de Rennes 1, 35042 Rennes Cedex, France
Fax: ( ‡ 33)-223-236-939, e-mail: jean-francois.carpentier@univ-rennes1.fr
b
À
¬
¬
Received: July 15, 2002; Accepted: September 25, 2002
Abstract: This account describes the development
and application of Noyori×s type catalysts based on
ruthenium-arene complexes and simple chiral b-
1
2
Introduction
Development of [Ru(arene)(b-Amino Alcohol)]
Catalysts
amino alcohols derived from ephedrine, for the 2.1 Effect of the b-Amino Alcohol Ligand
asymmetric transfer hydrogenation of 2-propanol to 2.2 Effect of the Arene Ligand
carbonyl substrates. The influence of key parameters
of the catalyst system has been studied systematically,
resulting in particular in the design of the novel ligand
(4-biphenylmethyl)norephedrine. Thanks to the lat-
3
4
Isolation and Reactivity of Catalytic Intermedi-
ates
Applications of [Ru(arene)(b-Amino Alcohol)]-
Catalyzed Transfer Hydrogenation
ter, the catalytic precursors and true active species 4.1 Aromatic Ketones and Keto Esters
could be isolated for the first time, enabling a 4.2 Simple and Functionalized Aliphatic Ketones
complete structural description of the catalytic cycle
and Keto Esters
and of probable deactivation pathways. Highly effec- 4.3 Catalyst Deactivation Pathways
tive applications of those catalysts systems, i.e., the
asymmetric reductions of simple aryl ketones and aryl
5
Conclusions
b-keto esters, the synthesis of chiral phthalides and Keywords: amino alcohols; asymmetric catalysis;
syn-b,d-dihydroxy esters, are described.
chiral alcohols, ephedrine; hydrogen transfer; N,O
ligand; ruthenium
1 Introduction
taneously as a safe, inexpensive, easy to handle solvent
and reductant that is transformed into acetone, which
ˆ
Catalytic asymmetric reduction of C O bonds is a can be readily removed from the reaction mixture. The
pivotal transformation in organic chemistry to create resulting operational simplicity and high product yields
new stereocenters. Very successful developments in this in high enantiomeric excess for some specific ketones
field have been reported in recent years and a large make catalytic transfer hydrogenation a useful comple-
number of catalytic methods are now available to ment/alternative to hydrogenation using molecular
achieve this goal.[1] Among these, transfer hydrogena- hydrogen, particularly for small- to medium-scale
tion of 2-propanol to prochiral ketones has emerged as a reactions. After a rather long and quite primitive stage,
highly efficient technique (Scheme 1).[2]
several advances have been accomplished in the last
In this reversible process, referred to as the Meer- decade for designing efficient hydrogen transfer catalyst
wein Pondorf Verley reduction, 2-propanol acts simul- systems with some transition metal and lanthanide
complexes.[2] Undoubtedly, one of the most significant
breakthroughs, reported by Noyori et al., is the use of
chlororuthenium(II)arene precursors with chiral mono-
[Ru(arene)Cl2]2
O
OH
chiral ligand, 2-PrOK
arylsulfonylated-1,2-diamine or b-amino alcohol li-
gands (Scheme 1).[3,4] The structurally well-defined
[RuII(arene)(TsDPEN)] [TsDPEN ˆ (1R,2R)-N-(p-tol-
ylsulfonyl)-1,2-diphenylethylene-diamine, 1] system en-
ables the highly effective reduction of a variety of aryl
alkyl and related ketones in above 90% ee.[3]
R1 R2
R1 R2
OH
O
Scheme 1.
Adv. Synth. Catal. 2003, 345, No. 1+2
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim 1615-4150/03/3451+2-067 077 $ 17.50+.50/0
67

Kathelyne Evaraere et al.
REVIEWS
Kathelyne Everaere was born
Jean-FranÁois Carpentier was
born in Lille, France, in 1967.
He graduated from the Engi-
neering School of Chemistry
of Lille in 1989. He earned his
Ph. D. in chemistry from the
University of Lille in 1992,
working with Professors A.
Mortreux and F. Petit on
palladium-catalyzed carbony-
lation of organic halides with
¬
in Bethune, France, in 1973.
She studied chemistry at the
Graduate Engineering School
in Lille and earned her Ph. D.
degree in organic chemistry in
2000 from the University of
Lille, under the direction of
¬
Professor Andre Mortreux
and Dr. Jean-FranÁois Car-
pentier, with a thesis on the
topic of this account. After 15 months post-doctoral
research at the ATOFINA research center of King
of Prussia, USA, she recently joined CEREXAGRI
where she works on pesticides formulation.
alkyl formates. After a 12-month period of post-
doctoral research at the French Nuclear Agency in
Tours, he was appointed as a CNRS research fellow
at the University of Lille where he prepared his
habilitation working on a wide variety of homoge-
neously-catalyzed processes, mostly asymmetric
hydrogenation of carbonyl compounds and nucleo-
philic allylic substitution. In 1997, Carpentier was a
visiting scientist at the University of Iowa where he
worked with Professor R. F. Jordan on mechanistic
issues related to the interaction of olefins with alkyl
cationic group 4 metallocenes. Returning to Lille, he
pursued his studies on olefin polymerization using
organolanthanide catalysts. Carpentier joined the
University of Rennes in 2001 as Professor of
Chemistry. His current research interests include
the design of new ligands, organometallic chemistry
of early transition metals, and application of these
complexes as homogeneous single-site catalysts for
olefin polymerization and fine chemicals synthesis.
¬
Andre Mortreux was born in
Libercourt, France, in 1943.
He received his Ph. D. from
the University of Poitiers
working with Professor M.
Blanchard on catalytic reac-
tions of alkynes, where he
discovered in 1973 the first
homogeneous catalytic sys-
tem for alkyne metathesis.
After post-doctoral research
in 1976 with Professor F. G. A. Stone in Bristol,
U. K., he moved to the University of Lille to join
Professor F. Petit who had just set up a research
team devoted to homogeneous catalysis. He was
promoted to Professor of Chemistry in 1983 and
Professor at the Institut Universitaire de France in
2001. In the same year, he received the Clavel-
Lespiau award from the French Academy of
Sciences. He his now head of the homogeneous
catalysis team of the Laboratoire de Catalyse of
Lille University. His current research interests
include the development of new catalysts for
transition metal-catalyzed processes involving
asymmetric catalysis, carbonylation, oligomeriza-
tion and polymerization reactions. He is the author
and co-author of more than 200 research papers and
holds 30 patents.
this time for RuII-catalyzed transfer hydrogenation.[4]
Since chiral diamines are notoriously complicated to
synthesize in contrast to the large variety and ready
availability of chiral amino alcohols, this step-change
has led to intense exploration of [RuII(arene)(b-amino
alcohol)] systems for rationalizing the enantioselectiv-
ity, designing more efficient ligands, improving catalyst
activity and broadening the scope of asymmetric hydro-
gen transfer of 2-propanol to functionalized carbonyl
compounds.[5 7]
In this account we summarize our own efforts toward
these objectives using ephedrine-type ligands.[6] Three
essential parameters of these asymmetric hydrogen
transfer systems, i.e., (i) structure of the b-amino alcohol
ligand, (ii) structure of the arene ligand in the ruthenium
precursor, and (iii) influence of functional groups of
ketones to be reduced, have been studied systematically.
NH2
NHSO2Ar
1 (Ar = 4-CH3C6H4)
Initial systems designed by Noyori et al. that employ The complementary information thus collected, togeth-
simple b-amino alcohols, e.g., ephedrine, gave also very er with reactivity data of unique isolated [RuII(arene)
good results in terms of enantioselectivity and catalytic (b-amino alcohol)] complexes, have led to a better
activities, the latter being among the highest reported at understanding of the process with a complete structural
68
Adv. Synth. Catal. 2003, 345, 67 77

Ruthenium(II)-Catalyzed Asymmetric Transfer Hydrogenation of Carbonyl Compounds
REVIEWS
description of the intermediates involved
in the catalytic cycle. Valuable synthetic
applications that have been developed
from these catalysts will be described.
R2
Me
NHR3
HO
NHR3
HO
3a R2 = H
3b R2 = H
3c R2 = Et
3d R2 = Et
3e R2 = Bn
3f R2 = Bn
3g R2 = Ph
2a R3 = H
2b R3 = Me
2c R3 = iPr
2d R3 = CH2Bn
2e R3 = Bn
R3 = Me
R3 = Bn
R3 = Me
R3 = CH2C6H4-(4-Ph)
R3 = Me
2 Development of [Ru(arene)-
(b-Amino Alcohol)]
Catalysts
2f R3 = CH2C6H4-(4-Ph)
2g R3 = CO2Et, COMe, CO2Me, Ts
R3 = CH2C6H4-(4-Ph)
R3 = Bn
2.1 Effect of the b-Amino Alcohol
Ligand
The effect of the b-amino alcohol struc-
ture on the performance of in situ gen-
R2
R1
HO
Ph
Ph
Me
erated catalysts was addressed by varying
NHR3
the nature of substituents R1, R2, and R3
HO
NHR
in R1CH(OH)CHR2NHR3 ligands. No
4a R1 = Me
4b R1 = H
R2 = H
R2 = Me R3 = CH2C6H4-(4-Ph)
R3 = Me
R3 = Me
ligand having a tertiary amino group
was investigated since the presence of
an N-H moiety was shown previously to
5a R = H
4c R1 = 4-MeOC H R2 = Bn
5b R = Bn
have a crucial importance for catalytic
activity.[4] About 50 different chiral ligands, related to
ephedrine, were prepared using various synthetic pro-
cedures and evaluated systematically in the reduction of
ketone substrates. Noteworthy, similar trends in reac-
tion rates and enantioselectivities* were observed with
different [Ru(arene)Cl2]2 precursors and ketones, thus
allowing us to draw some general conclusions. Repre-
sentative results for the transfer hydrogenation of model
substrate tert-butyl acetoacetate (K1) using [Ru(benze-
ne)Cl2]2 (6a) as catalyst precursor are summarized in
Table 1.
Table 1. Influence of the ligand structure on the asymmetric
transfer hydrogenation of tert-butyl acetoacetate (K1).[a]
Entry
Ligand
Time[b]
[h]
Con-
TOF50
[hÀ1][d]
ee[e]
[%, (S)]
version[c]
[%]
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
1
17
16
1
16
16
5
2.5
16 50
5
24
6
14
4.5
16
14
5
13
4.5
22
21
>99
98
>99
>99
>99
>99
ꢀ
9
4
32
44
29
36
68
67
2a
2b
2c
2d
2e
2f
2g
3a
3b
3c
3d
3e
3f
190
12
12
22
46
ꢀ
28
0
23
10
25
8
10
23
11
25
ꢀ
The influence of the R3 substituent proved to be the
most important and was addressed from N-substituted
(1S,2R)-norephedrine amino alcohols (2a h). The
highest apparent reaction rate was observed with
ephedrine (2b). Use of the corresponding primary
amine (2a) or of any other N-alkyl-substituted ligand
(e.g., 2c, d), or even worse of the N-carbamate, amide
and sulfonamide derivatives (2g), resulted in reduced
reaction rates and ees. The sole but remarkable ex-
ception was found with benzyl-type substituents (2e, f)
that caused the enantioselectivity to increase up to 68%.
The influence of additional substituents at the benzyl
ring proved to be sensitive in terms of reaction rate but
with minimal effects on the selectivity. While ortho-
substituted ligands [N-CH2(2-XC6H4); X ˆ Me, OMe,
Cl, Ph] led to very low reduction rates, most likely
8
10
7
39
22
>99
>1
>99
>99
99
44
64
47
34
36
35
43
45
nd
98
3g
4a
4b
4c
5b
>99
>99
99
98
3
[a]
Reactions were carried out at 20 8C using 2.0 mmol of K1
in a 0.1 M 2-propanol solution with [K1]/[i-PrOK]/[chiral
ligand]/ [Ru(benzene)] ˆ 100:6:2:1.
* The enantioselectivities of the catalytic reactions per-
formed with the systems described in this paper were found
to be constant in time. Hence, these values are representative
of the intrinsic discriminating abilities of the catalytic species,
which was confirmed from the reactions carried out with
isolated catalytic intermediates.
[b]
[c]
[d]
Reaction time not optimized.
Conversions of K1 determined by GLC.
Catalyst turnover frequency at half-reaction determined by
GLC monitoring, in mole of product/(mole of Ru ¥ h).
Determined by chiral GLC analysis.
[e]
Adv. Synth. Catal. 2003, 345, 67 77
69

Kathelyne Evaraere et al.
REVIEWS
because of steric reasons, ligands having para-substitut- R1 ˆ Ph,Ph) turned out to be detrimental in terms of
ed benzyl groups [N-CH2(4-XC6H4); X ˆ Me, OEt, Ph] selectivity and reaction rate, respectively. Also, systems
gave rise to slightly improved reaction rates. As will be based on pseudoephedrine derivatives led to lower
exemplified below, these results implicated the system selectivities than those based on the corresponding
derived from (4-biphenylmethyl)norephedrine (2f) to ephedrine analogues, highlighting the importance of
be the most efficient for the transfer hydrogenation of relative 1S,2R/1R,2S stereoselectivity.
many non-aromatic ketones.
The effect of the substituent at the 2-position was
studied using b-amino alcohols 3a g and was shown to 2.2 Effect of the Arene Ligand
be moderate in most cases. Thus, replacing of the methyl
R2 group of ephedrine by an ethyl, a benzyl or a phenyl Most of the work reported in the field of Ru-catalyzed
group (3c g) had almost no effect on the reaction rate. asymmetric transfer hydrogenation has focused on
Also, in most cases, the selectivity abilities of catalytic commercially available [Ru(arene)Cl2]2 precursors
species in this series are not affected by the 2-substitu- such as the para-cymene and benzene complexes.
ent. However, two exceptions were found to these Preliminary observations by Noyori et al. revealed,
trends: (i) surprisingly, changing R2 from a methyl to an however, that substituents on the Ru-arene ring also
H atom (3a, b) caused a dramatic decrease in the play a significant role in the performances of the catalyst
reaction rate. (ii) Systems derived from ligands bearing system.[4] To get a better insight in this effect, we have
phenyl rings at both the R2 and R3 substituents, gave rise evaluated in situ combinations generated from a set of
to lower ees than their analogues with R2 ˆ Me, (e.g., chloroRuII(arene) precursors (6a i) in the transfer
compare 3g, 34% ee, vs. 2e, 68%); computed molecular reduction of K1 (Table 2). These chloroRuII(arene)
models suggested that such decreased selectivity ability dimers were prepared via reaction of RuCl3 ¥ 3 H2O with
may originate from p-p-stacking between the aryl 1,4-dienes,[7] which are available through the Birch
groups at C-2 and the NH-benzyl moieties.
reduction of the corresponding arene.
The influence of the R1 substituent was addressed with
a variety of b-amino alcohols and proved also to be
variable. Systems based on ligands containing the
relatively more bulky para-methoxyphenyl group in-
stead of a phenyl group at the 1-position gave nearly
identical performances (R2 ˆ Me, Bn; 4c vs. 3e). Also,
changing the 1-phenyl group by a methyl R1 substituent
in the case R2 ˆ H (4a vs. 3a) almost did not affect the
results. Hence, a small group at C-1 apparently suffices
to reach the maximal enantioselectivity. On the other
hand, non-substitution (4b, R1 ˆ H) or disubstitution (5,
R3
R2
R4
R1
6a-i
R5
R6
Cl
Ru
Cl
2
O
OMe
OEt
a
e
c
b
d
Table 2. Influence of the arene structure on the asymmetric
transfer hydrogenation of tert-butyl acetoacetate (K1).[a]
Entry Ru(II)- Ligand Time[b] Con-
TOF50 ee[e]
arene
[h]
version[c] [hÀ1][d] [%, (S)]
[%]
g
f
h
i
1
20
21
22
23
24
25
26
27
6
6a
6b
6c
6d
6e
6f
6g
6h
6i
6a
6c
6e
6f
2b
2b
2b
2b
2b
2b
2b
2b
2b
2e
2e
2e
2e
2e
1
48
2
24
3
14
14
48
24
5
98
>99
>99
0
>99
>99
>99
70
190
5
250
0
33
12
20
3
ꢀ
22
18
6
44
18
42
As anticipated, introduction of alkyl substituents on
the arene ring logically resulted in a concomitant
decrease of the apparent reaction rate with increased
steric hindrance; thus, the RuII(benzene) (6a) and the
RuII(hexamethylbenzene) (6i) systems feature extreme
positions, the former system being the most active among
those investigated. On the other hand, the ephedrine-
RuII(anisole) system (6c) led to similar kinetics to the one
based on RuII(benzene) (6a), but no activity was ob-
served with the RuII(benzoate) system (6d). This indi-
catesthat, besides steric considerations, electronic factors
contribute also to the reaction outcome and that a
minimal electron density on the arene ring may be
requiredfor thereactiontoproceed. Itis, however, stillan
5[f]
15
17
12[f]
3[f]
68
5
>99
>99
98
>99
>99
28
29
30
31
6
65
30
51
46
23
14
16
20
10
6g
[a e]
[f]
See Table 1.
The configuration of the major isomer of A1 is R-( ).
70
Adv. Synth. Catal. 2003, 345, 67 77

Ruthenium(II)-Catalyzed Asymmetric Transfer Hydrogenation of Carbonyl Compounds
REVIEWS
open question whether these steric and electronic factors
affect the intrinsic activityofthe activerutheniumspecies
in the rate-determining-step of the process or the
base
Cl
Ru
NH
6e + 2f
efficiency of its in situ generation from the catalyst
7
– HCl
O
R
precursors, i.e., the amount of active species (vide infra).
Significant and subtle influence of the arene ligand on
enantioselectivity was also observed. Contrary to reac-
tion rates, but not surprisingly, the relative efficiency of
RuII(arene) precursors in this matter depends strictly on
the nature of the ketone to be reduced. In the case of tert-
butyl acetoacetate (K1), the use of the benzene pre-
cursor (6a) in combination with ephedrine (2b) or N-
benzylnorephedrine (2e) leads to improved enantiose-
lectivities with respect to corresponding combinations
based on the p-cymene precursor (6e). A reversal in the
configuration of the major reduction product A1 is
observed with ephedrine as the ligand [44% (S) and 5%
(R), respectively] in contrast with N-benzylnorephe-
drine [68% (S) and 30% (S)]; however, both variations
base
– HCl
R
OH
HO
8
Ru
R'
R
N
O
O
R
R'
of ees correspond to the same difference in the Gibbs
H
energies [D(DG=)6a/6e ˆ 0.62 kcal ¥ molÀ1). In the same
way, the 1,2,4-trimethylbenzene systems (6g) are less
selective than the corresponding systems based on 6a,
giving rise to nearly identical energy differences
Ru
N
C
O
H
H
Ru
O
NH
O
R
R
TS-10
9
[D(DG=)6a/6g ˆ 0.35 kcal ¥ molÀ1 with 2b; D(DG=)6a/6g
ˆ
0.39 kcal ¥ molÀ1 with 2e]. The same trend was found
R
for the tetrahydronaphthalene systems [D(DG=)6a/6f
ˆ
O
R'
0.38 kcal ¥ molÀ1 with 2b; D(DG=)6a/6f ˆ 0.32 kcal ¥ molÀ1
with 2e]. As pointed out above, it is obvious that steric
interactions between substituents on the arene ring and
the substrate do exist in the transition state. Nonethe-
less, considering the similarity in the differences in the
Scheme 2.
The formally 18-electron catalyst precursor 7 was
D(DG=) values going from the ™naked∫ RuII(benzene) prepared readily by reacting the RuII-p-cymene com-
precursor to arene-substituted precursors, despite the plex (6e) and ligand (2f) with 2 equivalents of NEt3 in
difference in the steric bulk of 2b and 2e (R3 ˆ Me vs. refluxing 2-propanol. Complex 7 was fully characterized
Bn), there is apparently no significant steric interaction by elemental analysis, ESI-MS, and an X-ray diffraction
between substituents on the arene ring and the R3 study of its methanol adduct, which revealed a distorted
1
substituent on the ligand.
™piano stool∫ structure in the solid state. The H and
13C NMR data confirmed that the amino alcohol
chelates to Ru in CDCl3 and C6D6 solutions and that 7
exists as a single diastereomer. Complex 7 catalyzes the
asymmetric transfer hydrogenation of tert-butyl acetoa-
cetate (K1) in 2-propanol only upon addition of i-PrOK,
3 Isolation and Reactivity of Catalytic
Intermediates
Since Noyori et al. established the identity of the catalyst with the same enantioselectivity and activity as those
precursors and true active species involved in the observed with the complex formed in situ.
transfer hydrogenation process promoted by [Ru(are-
Upon treatment with one equivalent of KOH, com-
ne)(TsDPEN)] catalysts,[3c] the nature of the corre- plex 7 undergoes facile elimination of HCl to give the
sponding [Ru(arene)(b-amino alcohol)] intermediates formally 16-electron complex 8 as a very air-sensitive
was speculated to be analogous.[3g] Repeated attempts to deep purple solid. This compound is most easily
isolate intermediates from the [Ru(p-cymene)Cl2]2/ prepared in >80% yield via the direct reaction of a 1:1
ephedrine system using a synthetic methodology similar mixture of 6e/2e with excess KOH (7 equivalents) in
to that reported by Noyori for TsDPEN led, however, to CH2Cl2. It reacts rapidly at room temperature in chloro-
mixtures of products. Fortunately, thanks to the new form solution through capture of HCl to give back 7 in
ligand (1S,2R)-(4-biphenylmethyl)norephedrine (2f), quantitative yield. The identity of 8 was unambiguously
we were able to prepare for the first time the three established from NMR, IR spectroscopy and ESI-MS. In
1
speculated intermediates and to establish formally the particular, the H NMR resonances in 8 around the
mechanism pathway (Scheme 2).[6d]
oxygen atom are unchanged from those of 7, whilst those
Adv. Synth. Catal. 2003, 345, 67 77
71

Kathelyne Evaraere et al.
REVIEWS
Table 3. Asymmetric transfer hydrogenation of acetophenone (K2) and 2-acetylpyridine (K3).[a]
Entry
Ru(II)-arene
Ligand
Ketone
Time[b]
[h]
Conversion[c] [%]
TOF50
[hÀ1][d]
ee[e]
[%, (S)]
32
33
34
35
36
37
38
39
6a
6e
6e
6e
6a
6e
6e
6a
2e
2b
2e/2f
2h[f]
2b
2b
2e/2f
1
K2
K2
K2
K2
K2
K2
K3
K3
2
0.5
2
>99
>99
>99
>99
>99
>99
>99
40
250
120
120
120
450
375
7
82
91
91
95
68
83
89
88[g]
2
0.5
0.5
16
21
ꢀ
[a e]
See Table 1.
[f]
2h ˆ (1S,2R)-N-(cyclohexylmethyl)-norephedrine.
[g]
The configuration of the major isomer of A3 is R-(‡).
Table 4. Asymmetric transfer hydrogenation of 2-acylarylcarboxylates K4.[a]
Entry
Ketone
Catalyst
Temperature
[8C]
Time[b]
[h]
Conversion[c]
[%]
Selectivity A4[d]
[%]
ee[e]
[%]
40
41
42
43
44
45
46
47
48
49
50
K4a
K4a
K4a
K4a
K4a
K4b
K4b
K4b
K4c
K4c
K4c
Ru-1 in situ
20
™
18
23
3.5
16
3
2
1
3
92
93
96
65
96
65
100
99
80
91
99
32[f]
99
>99
41[f]
>99
96
>99
>99
98
97 (S)
97 (S)
94 (S)
84 (S)
83 (S)
94 (À)
96 (À)
23 (À)
92 (S)
80 (S)
72 (S)
Ru-1 preformed
Ru-1 preformed
Ru-2f in situ
Ru-2f preformed (8)
Ru-1 in situ
Ru-1 preformed
Ru-2f preformed (8)
Ru-1 preformed
Ru-1 preformed
Ru-2f preformed (8)
50
20
20
50
™
20
™
50
20
18
22
18
98
>99
[a c]
See Table 1.
[d]
[e]
[f]
1
Selectivities for A4 determined by GLC and H NMR analysis; corresponding compound B4 accounts for the balance.
Determined by chiral GLC analysis.
Compound B4 accounts for the balance.
O
OH
around the nitrogen atom are shifted, as expected
from the transformation of the amino to amido func-
tionality.
X
X
2-PrOH
[Ru*]
Treatment of the purple complex 8 with a large excess
of 2-propanol at room temperature gave the subsequent
intermediate, the true catalytically active hydrido species
9, as a brown red solid. The variable temperature Scheme 3.
1H NMR monitoring of this reaction showed the ap-
pearance at À 25 8C of one set of resonances, including a
single Ru-H resonance at d ˆÀ 5.20, that remained
unique (x95%) till 0 8C. In light of Noyori×s results,[3c]
this is consistent with the kinetically controlled forma-
tion of a major diastereomer.
K2 X = CH
K3 X = N
A2 X = CH
A3 X = N
R1
2-PrOH
[Ru*]
O
O
X
X
O
R1
A4
OR2
Both complexes 8 and 9 catalyze the asymmetric
transfer hydrogenation of b-keto ester K1 and aceto-
phenone (K2) in 2-propanol without addition of a base to
give the corresponding alcohols in the same stereo-
selectivity and catalytic activity as those observed with
the complex formed in situ. As shown in the next section,
the possibility to achieve hydrogen transfer under
neutral conditions turned to be of major interest for
X
R1 O-i-Pr
O
2-PrOH
[OH-]
O
O
K4
a: X = CH, R1 = Me, R2 = Me
b: X = N, R1 = Me, R2 = i-Pr
c: X = CH, R1 = n-Bu, R2 = Me
B4
the reduction of sensitive ketone substrates.
Scheme 4.
72
Adv. Synth. Catal. 2003, 345, 67 77

Ruthenium(II)-Catalyzed Asymmetric Transfer Hydrogenation of Carbonyl Compounds
REVIEWS
Table 5. Asymmetric transfer hydrogenation of 2-aroylacetates K5.[a]
Time[b] [h]
Conversion[c]
[%]
TOF50
[hÀ1]
ee[e]
[%]
[d]
Entry
Ketone
Ru(II)-arene
51
52
53
54
55
56
57
58
59
K5a
K5a
K5b
K5c
K5d
K5e
K5f
K5g
K5h
6a
6e
6e
6e
6e
6e
6e
6e
6e
2.5
15
99
>99
>99
95
>99
>99
>99
>99
>99
136
10
120
10
>100
nd
40 (S) (À)
94 (S) (À)
89 (S) (À)
75 (À)
1
2[f]
1
82 (À)
16
82 (S) (À)
72 (À)
14[f]
1[f]
1[f]
nd
25
30
81 (À)
84 (À)
[a e]
See Table 1 except Tˆ 50 8C ; ligand used: 2b.
[K5]/[Ru] ˆ 20.
[f]
OH
O
O
O
4 Applications of [Ru(arene)(b-Amino
OR'
OR'
R' =
i-PrOH
[Ru*]
Alcohol)]-Catalyzed Transfer
Hydrogenation
R
R
K5
A5
a
b
c
d
R =
H
e
f
R =
R' =
Et
Et
4-OMe
4.1 Aromatic Ketones and Keto Esters
H
t-Bu
Et
2-Cl
4-Cl
4-F
Et
Et
Et
Carbonyl compounds having one sp2-carbon atom a to
the keto function belong to the most favorable class of
2-Me
g
h
2-OMe
Et
substrates for asymmetric transfer hydrogenation with Scheme 5.
[Ru(arene)(b-amino alcohol)] catalysts. High enantio-
selectivities have been readily achieved for a number of
aromatic ketones and keto esters.
ligands, in the presence of a base promoter (6 equiv. i-
Acetophenone (K2), the usual model substrate, and 2- PrOK vs. Ru), gave the expected phthalides A4 in up to
acetylpyridine (K3) are reduced with high turnover 91 97% ee (Scheme 4, Table 4).[6e] The synthetic effi-
frequencies and ees up 91% with N-benzylnorephe- ciency of this procedure is, however, hampered by the
drine-type ligands such as 2e and 2f (Scheme 3, Table 3). formation of large amounts of side-products B4, which
The use of N-(cyclohexylmethyl)-norephedrine (2h), arise from lactonization of the hemiacetal of K4 formed
prepared readily from norephedrine and cyclohexylcar- under basic conditions. This could be efficiently pre-
baldehyde, afforded A3 in up to 95% ee.[6d, g] Consistent vented by performing asymmetric transfer hydrogena-
with all the other ketone substrates, a decrease in tion in the absence ofbase, under neutral conditions, with
reactivity was observed upon increasing the steric bulk the presynthesized true catalysts [Ru(p-cymene)(b-
on the arene ligand. Nonetheless, the best enantiose- amino alcohol)] (8) or its analogue [Ru(p-cyme-
lectivities for the reduction of K2 and K3 were obtained ne)(TsDPEN)] (Ru-1). Thus, 3-alkylphthalides were
with catalytic combinations based on the para-cymene recovered in virtually quantitative yields and the same
precursor (6e). With respect to acetophenone, the enantioselectivity. As a general trend, the b-amino
pyridyl ring caused a systematic decrease in enantiose- alcohol catalyst 8 was found to be more active, although
lectivity with corresponding differences in D(DG=)K3/K2 significantly less enantioselective than the TsDPEN
in the range 0.1 0.45 kcal ¥ molÀ1, as well as in the catalyst Ru-1, especially for the pyridyl derivative
reaction rate.
Catalytic asymmetric reduction of 2-acylbenzoates
K4b.
Another valuable class of carbonyl substrates is b-
(K4) is a process of high interest to prepare chiral keto esters. Benzoylacetate esters and aryl-substituted
phthalides (A4).[8] Such molecules, almost all of them analogues K5 are efficiently reduced in 2-propanol using
having an S configured chiral center, feature an impres- the simple catalytic combination of (1S,2R)-ephedrine
sive list of applications due to their pharmacological (2b) with [RuCl2(p-cymene)]2 (6e) to give the corre-
effects and are also key intermediates for the synthesis sponding b-hydroxy esters A5 in high yields and
of several alkaloids.[9] The transfer hydrogenation of 2- valuable enantioselectivities up to 94% ee (Scheme 5,
acylbenzoates and 3-acetylpyridine-2-carboxylate, e.g., Table 5).[6c] Introduction of a variety of ortho substitu-
K4a c, using in situ combinations of [RuCl2(p-cym- ents on the phenyl ring systematically resulted in lower
ene)]2 with either TsDPEN (1) or b-amino alcohols 2 as ees, probably reflecting steric factors. The reduction of
Adv. Synth. Catal. 2003, 345, 67 77
73

Kathelyne Evaraere et al.
REVIEWS
Table 6. Asymmetric transfer hydrogenation of b-keto esters and methoxyacetone (K10).[a]
Entry
Ketone
Ru(II)-arene
Ligand
Temperature
[8C]
Time[b]
[h]
Conversion[c]
[%]
TOF50
[hÀ1][d]
ee[e]
[%, (S)]
1
6
K1
K1
K6
K6
K9
K9
K10
K10
K10
K10
6a
6a
6a
6a
6a
6e
6a
6e
6a
6a
2b
2e
2b
2e
2b
2b
2b
2b
2i/2o
1
20
20
50
1
5
0.5
0.5
1
98
>99
>99
>99
>99
>99
>99
>99
>99
39
190
22
300
300
200
44
68
36
56
25
50
54
36
66
nd
60
61
62
63
64
65
66
67
50
50
20
20
20
20
5
32
0.33
0.33
0.33
16
>600
>600
>600
ꢀ
[a e]
See Table 1 except for temperatures.
4.2 Simple and Functionalized Aliphatic Ketones and
Keto Esters
O
O
OH
O
2-PrOH
[Ru*]
OR2
OR2
R1
R1
K1
K6
K7
K8
K9
R1 = CH3
R1 = CH3
R1 = CH2Cl
R1 = CF3
R2 = t-Bu
R2 = Et
In view of the literature published over the last decade,
asymmetric transfer hydrogenation of keto compounds
in which the carbonyl function is surrounded by two a
sp3-carbon atoms, i.e., aliphatic ketones, has received
much less attention than the previous class of sub-
strates.[2 7] Obviously, this can be explained by the
difficulty to achieve highly selective reduction of these
carbonyl derivatives.
Transfer hydrogenation of simple b-keto esters such as
K1 and K6 with catalysts generated in situ, in the
presence of a base promoter, proceeds with high
chemoselectivity but rather poor enantioselectivity.
Best but still modest ees up to 67% could be reached
A1
A6
A7
A8
A9
R2 = Et
R2 = Et
R2 = t-Bu
R1 = CH2OCH3
O
OH
2-PrOH
[Ru*]
O
O
K10
A10
Scheme 6.
tert-butyl esters, e.g., K5b, proceeded somewhat less using N-benzylnorephedrine-type ligands such as 2e and
enantioselectively but much faster than that of the ethyl 2f (Scheme 6, Table 6). Interestingly, the apparent
equivalents. This increase in the apparent rate going catalytic activity of a given catalyst system varies with
from ethyl to tert-butyl esters was related to the the steric bulk of the alkoxycarbonyl group of the b-keto
bulkiness of the tert-butoxy group, which possibly ester; the bulkier the COOR moiety, the higher the
prevents the formation of less or non reactive species apparent catalyst turnover frequency. As a general
(vide infra). Another relevant information is the kinetic trend, most significant improvements were observed
resolution that was observed in the transfer hydro- with tert-butyl esters (vide supra). The increase in the
genation of a 1:1 mixture of ethyl benzoylacetate (K5a) enantioselectivity going from ethyl to tert-butyl esters,
and of its ortho-methyl derivative (K5c). In fact, though though noticeable, is much more limited.
transfer hydrogenation of K5a and K5c proceeded in
Ethyl 4-chloroacetoacetate (K7) and ethyl 4,4,4-
individual experiments both with the same apparent trifluoroacetoacetate (K8) failed to be reduced with
turnover frequency (TOF50 ˆ 10 hÀ1 at 50 8C), the re- various catalytic systems, even under drastic conditions.
duction of K5c in the mixture proceeded after that of Cross-experiments established that those functionalized
K5a was almost completed (kK5a/kK5c ˆ 40 at 80% b-keto esters and not their reduction products (A7 and
conversion of K5a; kK5a/kK5c ˆ 10 at 95% conversion of A8) deactivate the catalytic species. On the other hand,
K5a).[6c] In light of the mechanism depicted in Scheme 2, we found that 4-methoxy-acetoacetate esters are re-
this kinetic resolution may correspond to the difference ducible, especially the tert-butyl ester K9, but apparent
in easiness of approach of the ketone [one preferred reaction rates were significantly decreased compared to
enantioface] toward hydrido complex 9 to form the six- those for the reduction of K1. The presence of the 4-
membered transition state 10 for hydrogen transfer.
methoxy substituent affected also strongly the selectiv-
ity and only poor ees up to 50% were obtained for A9.
The effect ofthe methoxy groupwas probed individually
using methoxyacetone (K10) (Scheme 6, Table 6).
Quite surprisingly, this ketone is reduced in very high
74
Adv. Synth. Catal. 2003, 345, 67 77

Ruthenium(II)-Catalyzed Asymmetric Transfer Hydrogenation of Carbonyl Compounds
REVIEWS
Table 7. Asymmetric transfer hydrogenation of chiral aldols K11.[a]
Entry
Aldol
Ru(II)-arene
Ligand
Temperature
[8C]
Time[b]
[h]
Conversion[c] K11
[%]
de A11[c]
[%]
68
69
70
71
72
73
74
75
(S)-K11a
(S)-K11a
(S)-K11a
(S)-K11a
(S)-K11a
(S)-K11a
(S)-K11a
(R)-K11b
6a
6a
6a
6e
6e
6e
(1R,2S)-2a
(1S,2R)-2a
(1S,2R)-2a
(1S,2R)-2a
(1S,2R)-2a
(1S,2R)-2f
50
50
20
50
20
50
20
30
0.33
0.16
20
1
30
1
99
75
85
96
79
29
86
50
25 anti
66 syn
70 syn
71 syn
79 syn
80 syn
75 syn
72 syn
(1S,2R)-Ru 8[d]
1.5
2
6e
(1S,2R)-2a
[a]
The reaction was carried out using 2.0 mmol of K11 and 0.02 mmol of Ru in a 0.1 M 2-propanol solution as in Table 1.
Optimized reaction time for >98% chemoselectivity to A11.
Conversion and selectivities determined by GLC and NMR analysis.
[b]
[c]
[d]
Reaction carried out with isolated catalyst under neutral conditions.
alternative to traditional reduction with borane reagents
OH
O
OH OH
to prepare syn-3,5-dihydroxyesters A11 (Scheme 7,
Table 7). syn-3,5-Dihydroxy esters A11 are advanced
building blocks for the preparation of HMG-CoA
reductase inhibitors of industrial interest.[10] Preliminary
investigations on model substrate (S)-K11a indicated
that chirality of the catalyst are necessary to control the
diastereoselectivity of the reduction. The best compro-
mise between activity and stereoselectivity for the syn-
2-PrOH
[Ru*]
CO2-t-Bu
CO2-t-Bu
R
R
K11a R = H
K11b R = BnO
K11c R = Cl
A11a - c
Scheme 7.
reaction rates with most of [RuII(arene)(b-amino alco- diol was obtained with the 6e/(1S,2R)-2b combination.
hol)] systems, which suggests a limited impact of the High yields and promising diastereoselectivity up to
methoxy group as a coordinating functionality. Enan- 80% were obtained under smooth conditions. Usual
tioselectivity of the transfer hydrogenation of K10 systems based on TsDPEN (1) did not allow control of
follows the same trends as those observed for K1; the the stereoselectivity, irrespective of the ligand config-
combination of the RuII(benzene) precursor 6a with the uration. The reduction of tert-butyl (R)-6-benzyloxy-5-
N-benzylnorephedrine-type ligands 2e and 2f afforded hydroxy-3-oxohexanoate (K11b), an adequately substi-
the highest enantioselectivity so far (66%). Noteworthy, tuted molecule for further functionalization, proceeded
Ru-TsDPEN systems proved totally inefficient in re- in similar de for the syn-diol but was much faster and
ducing K10.
also more sensitive in terms of chemoselectivity. No
We have shown recently that asymmetric transfer reduction took place with tert-butyl (R)-6-chloro-5-
hydrogenation of 2-propanol to chiral 5-hydroxy-3- hydroxy-3-oxohexanoate (1c), most likely because of
oxohexanoates K11 provides a promising catalytic catalyst poisoning by the chloro group as with other
chloro-containing substrates.
OH
O
4.3 Catalyst Deactivation Pathways
CF3
OEt
Cl
Cl
Ru
Ru
The above examples show that the efficiency of [ruthe-
nium-{amino alcohol}] systems for transfer hydrogena-
tion is sometimes annihilated by the presence of some
functional groups in the ketone substrate. To get a better
understanding of these catalyst deactivation processes,
model reactions between the isolated catalytic inter-
mediates and ™reluctant∫ ketone substrates were under-
taken. Thus, it was shown that Ru-chloro complex 7, a
non-reducing analogue of Ru-hydride 9, reacts rapidly
with b-keto ester K8 to yield a new b-diketonato
complex 11, with concomitant release of the b-amino
alcohol 2f (Scheme 8). Consistent with the mechanism
depicted in Scheme 2, which implies an active role of the
O
NHCH2(4-PhC6H4)
O
O
2f
CF3
OEt
Ph
7
11
H
H
Ru
Ru
O
O
O
X
O
R
OR'
R'O
12
13
Scheme 8.
Adv. Synth. Catal. 2003, 345, 67 77
75

Kathelyne Evaraere et al.
REVIEWS
amino alcohol N-H moiety, b-diketonato complex 11 References and Notes
showed no activity as catalyst precursor in the transfer
[1] a) R. Noyori, in Asymmetric Catalysis in Organic Syn-
thesis, Wiley, New York, 1994, Ch. 2; b) Houben-Weyl,
Methods of Organic Chemistry, Stereoselective Synthesis,
Vol. E21d, 4th edn. (Eds.: G. J. Helmchen, E. Schau-
mann), Georg Thieme Verlag, Stuttgart, New York,
1995, pp. 3945 4464; c) T. Ohkuma, R. Noyori, H.
Nishiyama, S. Itsuno, in Comprehensive Asymmetric
Catalysis, Vol. 1. (Eds.: E. N. Jacobsen, A. Pfaltz, H.
Yamamoto), Springer, Berlin, 1999, Ch. 6, pp. 199 317.
[2] Reviews: a) G. Zassinovich, G. Mestroni, S. Gladiali,
Chem. Rev. 1992, 92, 1051 1069; b) C. F. de Graauw,
J. A. Peters, H. van Bekkum, J. Huskens, Synthesis 1994,
1007 1017; c) R. Noyori, S. Hashiguchi, Acc. Chem. Res.
1997, 30, 97 102; d) M. J. Palmer, M. Wills, Tetrahedron:
Asymmetry 1999, 10, 2045 2061.
[3] a) S. Hashiguchi, A. Fujii, J. Takehara, T. Ikariya, R.
Noyori, J. Am. Chem. Soc. 1995, 117, 7562 7563; b) A.
Fujii, S. Hashiguchi, N. Uematsu, T. Ikariya, R. Noyori, J.
Am. Chem. Soc. 1996, 118, 2521 2522; c) K.-J. Haack, S.
Hashiguchi, A. Fujii, T. Ikariya, R. Noyori, Angew.
Chem. Int. Ed. 1997, 36, 285 288; d) K. Matsumura, S.
Hashiguchi, T. Ikariya, R. Noyori, J. Am. Chem. Soc.
1997, 119, 8738 8739; e) K. Murata, K. Okano, M.
Miyagi, H. Iwane, R. Noyori, T. Ikariya, Org. Lett. 1999,
1, 1119 1121; f) I. Yamada, R. Noyori, Org. Lett. 2000,
2, 3425 3427; g) M. Yamakawa, H. Ito, R. Noyori, J.
Am. Chem. Soc. 2000, 122, 1466 1478.
[4] J. Takehara, S. Hashiguchi, A. Fujii, I. Shin-ichi, T.
Ikariya, R. Noyori, Chem. Commun. 1996, 233 234.
[5] a) M. J. Palmer, J. A. Kenny, T. Walsgrove, A. M. Kawa-
moto, M. Wills J. Chem. Soc. Perkin Trans. 1 2002, 416
427 and references cited therein; b) S. J. M. Nordin, P.
Roth, T. Tarnai, D. A. Alonso, P. Brandt, P. G. Ander-
sson, Chem. Eur. J. 2001, 7, 1431 1436 and references
cited therein; c) A. J. Sandee, D. G. I. Petra, J. N. H.
Reek, P. C. J. Kamer, P. W. M. N. van Leeuwen, Chem.
Eur. J. 2001, 7, 1202 1208 and references cited therein;
d) L. Schwink, T. Ireland, K. P¸ntener, P. Knochel,
Tetrahedron: Asymmetry 1998, 9, 1143 1163; e) C. G.
Frost, P. MendoÁa, Tetrahedron: Asymmetry 2000, 11,
1845 1848; f) M. Henning, K. P¸ntener, M. Scalone,
Tetrahedron: Asymmetry 2000, 11, 1849 1858.
[6] a) K. Everaere, J.-F. Carpentier, A. Mortreux, M.
Bulliard, Tetrahedron: Asymmetry 1998, 9, 2971 2974;
b) K. Everaere, J.-F. Carpentier, A. Mortreux, M.
Bulliard, Tetrahedron: Asymmetry 1999, 10, 4083 4086;
c) K. Everaere, J.-F. Carpentier, A. Mortreux, M. Bul-
liard, Tetrahedron: Asymmetry 1999, 10, 4663 4666;
d) K. Everaere, A. Mortreux, M. Bulliard, J. Brussee,
A. van der Gen, G. Nowogrocki, J.-F. Carpentier, Eur. J.
Org. Chem. 2001, 275 291; e) K. Everaere, J.-L. Schleff-
er, A. Mortreux, J.-F. Carpentier, Tetrahedron Lett. 2001,
42, 1899 1901; f) N. Franceschini, K. Everaere, A.
Mortreux, J.-F. Carpentier, Tetrahedron Lett. 2002, 43,
2569 2571; g) K. Everaere, PhD thesis, University of
Lille (France), 2000; h) K. Everaere, J.-F. Carpentier, A.
hydrogenation of K1. The 16-electron true catalyst 8 and
the reducing species 9 react also rapidly with enol K8,
with disappearance of the hydride resonances in the case
of 9; although the exact nature of the products could not
be established, formation of analogous b-diketonato
species was strongly suspected. On the basis of these
elementary reactions and of the results of catalytic tests,
we have proposed that catalyst deactivation proceeds
via removal of the amino group from the Ru coordina-
tion sphere to form inactive species such as 12 and 13
(X ˆ Cl) (Scheme 8).[11]
An important consequence of the deactivation proc-
esses is that the apparent reaction rates do not neces-
sarily reflect the intrinsic activity of the catalytic species,
but rather the amount of active species. Therefore, it is
questionable in which terms the influence of the
structure of the chiral ligand and of the ruthenium
precursor on reaction rate should be discussed: enhanc-
ing the intrinsic activity of the active species and/or
preventing the formation of non-productive species?
5 Conclusions
Factors that govern the activity and enantioselective
outcome of asymmetric hydrogen transfer of 2-propanol
to ketone substrates promoted by ruthenium catalysts
based on simple ephedrine-type ligands reaction are
nowadays reasonably well-understood. Thanks to a
modified norephedrine ligand, the two catalytic key
intermediates involved in the reduction process could be
isolated, which has allowed confirmation of the mech-
anistic pathway so far envisioned from intuitive sense
and computational studies. Concerning catalytic appli-
cations, high performances have been reached for the
reduction of a variety of aryl ketone substrates while
aliphatic functionalized ketones still remain more
problematic. In particular, the deactivation of catalytic
species by b-dicarbonyl compounds constitutes an
intrinsic limitation of the Ru-catalyzed transfer hydro-
genation process, which so far cannot compete, in this
case, with classical hydrogenation.
Acknowledgements
We thank PPG-SIPSYand the Centre National de la Recherche
Scientifique (CNRS) for financial support of this research (PhD
grant for K.E.) and Dr. Michel Bulliard for valuable discus-
sions.
76
Adv. Synth. Catal. 2003, 345, 67 77

Ruthenium(II)-Catalyzed Asymmetric Transfer Hydrogenation of Carbonyl Compounds
REVIEWS
Mortreux, M. Bulliard, (PPG-SIPSY), PCT Int. Appl.
WO 0109077, 2001; Chem. Abst. 2001, 134, 163196.
[7] M. A. Bennett, A. K. Smith, J. Chem. Soc. Dalton Trans.
1974, 233 241.
[8] For the synthesis of optically active phthalides through
ruthenium-catalyzed asymmetric hydrogenation of o-
acylbenzoic esters and acids, see: a) T. Okhuma, M.
Kitamura, R. Noyori, Tetrahedron Lett. 1990, 31, 5509
5512; b) M. Kitamura, T. Okhuma, S. Inoue, N. Sayo, H.
Kumobayashi, S. Akutagawa, H. Takaya, R. Noyori, J.
Am. Chem. Soc. 1988, 110, 629 631.
[9] a) G. Q. Zheng, P. M. Kenney, J. Zhang, L. K. Lam,Nutr.
Cancer 1993, 19, 77 86; b) S. Kobayashi, Y. Mimura, K.
Notaya, I. Kimura, M. Kimura, Jpn. J. Pharmacol. 1992,
60, 397 401; c) H. Sato, S. Yamaoka, Biomed. Res. 1993,
14, 385 390; d) Y. Ogawa, K. Hosaka, K. Kubota, M.
Chin, Jap. Pat. 0477,480, 1992; Chem. Abst. 1992, 117,
69721n; e) M. Elander, K. Leander, B. Luning, Acta
Chem. Scand. 1969, 23, 2177 2180; f) T. V. Hung, B. A.
Mooney, R. H. Prager, J. M. Tippett, Aust. J. Chem. 1981,
34, 383 395.
[10] a) G. Beck, H. Jendrella, K. Kesseler, Synthesis 1995,
1014 1018; b) A. Endo, M. Kuroda, Y. Tsujita, J.
Antibiot. 1976, 29, 1346 1348; c) A. O. Alberts, J.
Chen, G. Kuron, V. Hunt, C. Hoffman, J. Rothrock, M.
Lopez, H. Joshua, E. Harris, A. Patchett, R. Monaghan,
S. Currie, E. Stapley, G. Albers-Schonberg, O. Hensens,
J. Hirshfild, K. Hoogsteen, J. Liesch, J. Springer, Proc.
Natl. Acad. Sci. USA 1980, 77, 3957 3961.
[11] Formation of inactive b-enolate chelate complexes has
been also suggested in Meerwein Pondorf Verley re-
ductions (Al-alkoxides) and in rhodium- and iridium-
catalyzed hydrogen transfer, see: a) C. F. de Graauw,
J. A. Peters, H. van Bekkum, J. Huskens, Synthesis 1994,
1007 1017; b) R. Sariego, I. Carkovic, M. Martinez, M.
Valderrama, J. Mol. Catal. 1986, 35, 161 167; c) F.
Martinelli, G. Mestroni, A. Camus, G. Zassinovich, J.
Organomet. Chem. 1981, 220, 383 392; d) P. Kvintovics,
B. Heil, J. Organomet. Chem. 1989, 361, 117 122.
Adv. Synth. Catal. 2003, 345, 67 77
77