Chiral ytterbium complex-catalyzed direct asymmetric aldol-tishchenko reaction: Synthesis of anti-1,3-diols

Article abstract of DOI:10.1002/chem.200600357

The asymmetric aldol-Tishchenko reaction of aromatic aldehydes with aliphatic and aromatic ketones has been developed as an efficient strategy for the synthesis of anti-1,3-diols in good yield with high diastereo-control and good levels of enantiose-lectivity. This domino-type reaction is catalyzed by a chiral ytterbium complex that promotes both the aldol reaction through enolization of the carbon yl compound and the Evans-Tishchenko reduction of the aldol intermediate. The stereochemistry of the resulting diols is also investigated and finally proved by using CD techniques.

Full text of DOI:10.1002/chem.200600357

DOI: 10.1002/chem.200600357
Chiral Ytterbium Complex-Catalyzed Direct Asymmetric Aldol-Tishchenko
Reaction: Synthesis of anti-1,3-Diols
Jacek Mlynarski,* Bartosz Rakiel, Maciej Stodulski, Agata Suszczy n´ ska, and
[
a]
Jadwiga Frelek
Abstract: The asymmetric aldol-Tish-
chenko reaction of aromatic aldehydes
with aliphatic and aromatic ketones
has been developed as an efficient
strategy for the synthesis of anti-1,3-
diols in good yield with high diastereo-
control and good levels of enantiose-
lectivity. This domino-type reaction is
catalyzed by a chiral ytterbium com-
plex that promotes both the aldol reac-
tion through enolization of the carbon-
yl compound and the Evans–Tishchen-
ko reduction of the aldol intermediate.
The stereochemistry of the resulting
diols is also investigated and finally
proved by using CD techniques.
Keywords:
diols
·
aldol
reaction
·
aldol-Tishchenko
reaction · asymmetric catalysis
Introduction
combine high chemo- and enantioselectivity with the power-
ful atom economy of the aldol reaction.
[5]
[6]
The development of catalytic asymmetric methodologies for
the construction of chiral, nonracemic molecules bearing se-
quences of stereocenters is currently a challenging research
Since the first artificial metal complex was documented to
be capable of promoting the direct asymmetric aldol reac-
[7]
[8]
tion, some other metal-based and purely organic mole-
[1]
[9]
area of interest. The development of a catalytic asymmet-
ric aldol reaction is one of the recent landmarks in this
cules were reported to activate unmodified donors under
direct catalytic conditions.
[2]
field.
The direct asymmetric aldol reaction between aldehydes
and methylene ketones should provide a powerful tool for
the formation of new carbon–carbon bonds and the con-
struction of two continuous chiral centers. In contrast to
The impressive achievements with regards to the asym-
metric aldol reaction made to date rely on the conversion of
the donor substrate into more reactive species, such as enol
silyl ethers or ketene silyl acetals (Mukaiyama-type aldol re-
[10]
well-documented transformations of methyl ketones their
methylene analogues remain, however, a formidable syn-
[3]
action), by using no less than stoichiometric amounts of re-
agents in separate steps. However, the most elegant and
most economically attractive way to introduce chirality into
a molecule is by using only a catalytic amount of a chiral
[11]
thetic challenge. Only some a-substituted methyl ketones,
particularly a-hydroxy ketones, work nicely in direct reac-
[12]
tions promoted by known polymetalic catalysts. Diaster-
eo- and enantioselective synthesis of aldols, starting from
methylene ketones (for example, 3-pentanone) by means of
the direct catalytic asymmetric aldol reaction is still imma-
[4]
controller without tedious preactivation of the nucleophile.
Only recently has tremendous effort been devoted to the de-
velopment of catalytic asymmetric methodologies which
[13]
ture.
Low reactivity of methylene ketones in the direct aldol re-
action could be explained by the strong tendency towards
fast retroaldol reaction of formed aldols. This unwelcome
tendency can be surpassed by coupling of an irreversible
Evans–Tishchenko reduction to a reversible aldol reaction
#
[
a] Dr. J. Mlynarski, B. Rakiel, M. Stodulski, A. Suszczy n´ ska,
Dr. J. Frelek
Institute of Organic Chemistry
Polish Academy of Sciences, Kasprzaka 44/52
[14]
[15]
in one tandem-type process.
In this regard, Shibasaki
[16]
and ourselves
have attempted the direct catalytic asym-
0
1-224 Warsaw (Poland)
metric aldol-Tishchenko reaction as one of the useful meth-
ods for overcoming the problem of the unreactivity of
higher ketones. In the one-step process, aldehydes reacted
Fax : (+48)22-632-6681
E-mail: mlynar@icho.edu.pl
#
[
] The author is a student of the Warsaw University of Technology.
8158
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 8158 – 8167

FULL PAPER
with methylene ketones to give 1,3-diol monoesters, which
were formed as a result of the bond reorganization in the
cyclic Evans-type intermediate (Scheme 1).
modified ketones is of particular interest not only because
of its synthetic potential but also because of its unexplored
mechanism with respect to the origin of enantioselectivity.
[17]
Results and Discussion
Catalyst development for the direct asymmetric aldol-Tish-
chenko reaction: Our discovery that ytterbium-triflate com-
[16]
[18a]
plexes with chiral diols and aminoesters
effectively cat-
alyzed the aldol-Tishchenko reaction between diethyl
ketone and aromatic aldehydes prompted us to test the cata-
lytic activity of a range of chiral ligands for this process.
Based on our previous observations, we expected that li-
gands containing both a hydroxy group and an amino func-
tion should be useful and active for the reaction. To test this
hypothesis, the condensation of benzaldehyde and 3-penta-
none to afford 2a/b was performed by using ytterbium tri-
flate and a representative group of chiral aminoalcohols
(Table 1). Our initial aim was to identify the best suited
chiral ligand in terms of both reaction conversion and enan-
Scheme 1. Condensation of ethyl ketones with aldehydes by means of the
aldol-Tishchenko reaction.
The aldol Tishchenko methodology, initially restricted to
[
14a,b]
homocoupling of three identical aldehydes,
was later de-
[
14c,d]
veloped by application of lithium ketone enolates
fur-
[21]
nishing cross aldol-Tishchenko products with an exceptional
level of stereoselectivity. Furthermore, the same reaction
tioselection. An initial study revealed that ephedrine-type
aminoalcohols 1 are the most promising candidates for fur-
[
14e,f]
[14g]
was performed by using silyl,
zinc,
and samarium
[14h]
enolates.
Simpura
Table 1. Solvent and ligand studies: reaction of 3-pentanone with benzal-
dehyde promoted by various ephedrine-type chiral ligands 1.
[14i,j]
[14k,l]
and Schneider
independently established
catalytic cross aldol-Tishchenko reactions of ketone aldols
as enol equivalents.
The first examples of direct catalytic aldol-Tishchenko re-
actions of unmodified ketones and aldehydes to yield 1,3-
[
14m]
diol monoesters were presented by Mahrwald
ken
and Mor-
[
14n]
using titanium complexes and metal alkoxides, re-
spectively.
Our development of the direct asymmetric aldol reaction
of ethyl ketones by the tandem aldol-Tishchenko reaction
provided a new venue for our continued interest in asym-
Entry
R
R’
Yb
A
H
R
U
G
3
Yield 2a+b
ee 2a+b
[%]
[
a]
[b]
[%]
[18]
metric aldol methodologies.
1
2
Me H (1a)
Me H,Me
0
0
–
–
In contrast to previously presented examples of the ster-
eoselective Tishchenko reactions of two different alde-
(
1b)
3
4
5
6
7
8
9
Me Me (1c)
Me Me (1c)
Me Me (1c)
Me Et (1d)
Me Pr (1e)
Me Bu (1 f)
Me
Me
Me
Ph
69
70
55
46
17
46
88
10
14
20
55
50
53
42
[19]
[20]
hydes
and ketone aldols,
three adjacent stereogenic
centers are created by the use of a ketone as the substrate
in the process, which makes this methodology very effective
in terms of chiral economy.
Despite the enormous potential of the aldol-Tishchenko
reaction of unmodified ketones with aldehydes leading di-
rectly to protected 1,3-diols, its enantioselective variant
presents unexplored problems, only preliminarily unraveled.
While the stereoselectivity problem was overcome for acti-
A
H
R
U
G
2
)
4
4
4
4
5
4
(
10
A
H
R
U
G
2
2
2
2
2
86
80
50
55
75
61
74
65
52
(
1g)
1
1
1
2
A
H
R
U
G
)
)
)
)
(
1g)
[15a]
vated aromatic donors,
there was still paucity for aliphat-
AHCTREUNG
ic substrates for which state-of-the-art methods provide con-
[16]
densation products in moderate ee values only.
13
Me
Me
ACHTREUNG
(
1i)
Herein, we report an efficient application of a chiral ami-
noalcohol-based catalyst to the enantioselective aldol-Tish-
chenko reaction between aldehydes and aliphatic ketones.
This process provides good product yields and good ee
values leading to anti-1,3-diols in one single operation with-
out tedious preactivation of the donors. This reaction of un-
[
d]
1
4
A
H
R
U
G
À70
[c]
(
1g)
[
a] Overall isolated yield. [b] The ee values of esters and diols were deter-
mined by HPLC (Chiralpak AD-H column). [c] (1S,2R)-1g was used.
d] The use of +/À is only a convention to designate opposite enantiom-
ers.
[
Chem. Eur. J. 2006, 12, 8158 – 8167
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8159

J. Mlynarski et al.
ther optimization of the aldol-Tishchenko condensation.
While ligands containing an unprotected amino function,
that is, norephedrine (1a) and ephedrine (1b) were simply
unpromising. Application of ligands with protected amino
groups led to desired products 2. Thus, condensation of the
substrates in the presence of 20 mol% of the catalyst com-
with a reverse configuration (entry 14). In all cases both
regioisomeric esters 2a and 2b were detected in the reaction
mixture. An enantiomeric excess was determined independ-
ently for both esters and diol 3a. The reaction provides the
two esters 2a/b in similar enantiopurity, suggesting a nonse-
lective intramolecular acyl migration after formation of the
aldol-Tishchenko product (Scheme 2).
posed of Yb(OTf) and (1R,2S)-N-methylephedrine (1c) in
A
C
H
T
R
E
U
N
G
3
a 1:2 ratio in THF at room temperature led smoothly to de-
sired mixtures of esters 2 with excellent diastereoselectivity
but low enantiocontrol (Table 1, entry 3). Observed enantio-
selectivity was improved further when the reaction was car-
ried out in DME (DME=1,2-dimethoxyetane) in the pres-
ence of the catalyst composed of a 1:4 ratio of metal/ligand
(
entry 5).
A syn orientation of the hydroxy and amino substituents
in the ligand turned out to be essential for both the reactivi-
ty and selectivity of the catalyst composed of the ephedrine
[21]
ligand and ytterbium triflate. Thus, (1S,2S)-N-methylpseu-
doephedrine was not suitable as a ligand, due to the incon-
venient orientation of the substituents. Only trace of
amounts of product could be isolated from the reaction.
To improve the enantioselectivity in the reaction and pro-
vide a better understanding of the requirements for good
asymmetric induction, we sought a ligand design that would
met both reactivity and selectivity criteria.
Scheme 2. Aldol-Tishchenko condensation of 3-pentanone with benzalde-
A
H
R
U
G
3
hyde: a) Yb(OTf) /1g, THF, RT, 20 h; b) NaOMe, MeOH; c) DMP, ace-
tone, RT, 2 h. DMP=2,2-Dimethoxypropane.
As the phenyl substituent at C1 was expected to be indis-
pensable for catalyst reactivity, we decided to modify the C2
positions of the ligands and the type of protection used on
the amino group. In the first step, we decided to test the in-
fluence of the bulkiness of the amino function protecting
group on the catalytic efficiency of the ligand. To realize
this concept, we prepared a series of N,N-protected nore-
phedrine derivatives starting from commercial, optically
pure (1R,2S)-1-phenylo-2-amino-1-propanol ((1R,2S)-(À)-
norephedrine). The synthetic route to approach 1d–g was
reported previously by reaction of amine with alkyl io-
The structural assignment of the esters obtained was cor-
roborated by high-resolution NMR spectroscopic experi-
ments, and is in full agreement with previously published
[14h,m]
data.
The assigned 1,2-anti-1,3-anti stereochemistry of
2a/b was supported in both cases by the NMR spectroscopic
analysis of separately derived diols 3a and rigid O-isopropyl-
idene derivative 4. Characteristic signals of the isopropyl-
1
3
idene ring in the C NMR (d=23.6, 24.8, and 100.9 ppm)
are in full agreement with the rules presented by Rychnow-
[23]
ski for 1,3-diols. Such 1,3-anti-acetonides are expected to
adopt twist boat conformations in which both methyl groups
at the acetyl moiety are in nearly identical environments,
which results in similar chemical shifts of approximately
24 ppm.
[22]
dides.
As shown in Table 1, the nature of the protection used for
the amino group strongly affects the enantioselectivity of
the reaction. The enantioselectivity tends to increase as the
substituents vary from methyl to more bulky substituents
(
Table 1 entry 5 versus 6–8). The best result with respect to
Preparation of anti-1,3-diols: According to the optimized
reaction conditions discussed above, various methylene ke-
tones were treated with the aldehydes giving rise to a broad
range of 1,3-anti-diol monoesters 2 and after hydrolysis by
using NaOMe in MeOH, the corresponding diols 3. Experi-
ments to probe the scope of the methodology are summar-
ized in Table 2. Reactions were performed at room tempera-
ture, demonstrating the practical utility of the elaborated
catalytic system. Generally good yields were obtained rang-
ing between 60–85% for the two separated steps in which
15 mol% of the catalyst was used. To obtain optimal yields
and ee values 1.5–2.0 equivalents of the ketone were used. It
is noteworthy that this excess of the ketone donor in the re-
action mixture did not favor formation of aldols. In all cases,
aldol-Tishchenko esters were produced almost exclusively.
All tested donors, that is, diethyl- and dipropylketone, pro-
yield, diastereoselectivity, and enantioselectivity was, howev-
er, observed with a catalyst composed of ligand 1g. Combi-
nation of ytterbium triflate and (1R,2S)-1-phenyl-2-(1-pyrro-
lidynyl)-1-propanol (1g) in a 1:4 ratio gave rise to 2a/b with
8
0% isolated yield and 74% ee (entry 11). Optimal reaction
yield and selectivity were observed in DME at room tem-
perature.
To attain higher selectivity in the Yb(OTf) -catalyzed
A
H
R
U
G
3
aldol-Tishchenko condensation, we synthesized novel li-
gands 1h and 1i, which varied from 1b by a phenyl group
and six-membered ring, respectively. Modification of the
catalyst structure at C2 (entry 12) and expanding of the ni-
trogen-containing ring (entry 13) was unfortunately not pro-
gressive. It was found that an enantiomer of 1g with a
(
1S,2R) configuration favored formation of the products 2a/b
8160
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 8158 – 8167

Aldol-Tishchenko Reaction
FULL PAPER
Table 2. Substrate studies: condensation of ethyl and propyl ketones with various aldehydes promoted by piophenone, and 4-chloropro-
Yb
A
C
H
T
R
E
U
N
G
(OTf)
3
and (1R,2S)-1g (1:4).
piophenone can be used with-
out loss of reaction efficiency.
For dipropylketone
a lower
yield was observed, probably
because of steric hindrance (en-
tries 7 and 8). Activated aro-
matic ketones were more reac-
tive, but still selective substrates
(entries 9–14). All tested aro-
matic aldehydes gave useful
product yields. Besides benzal-
dehyde (entry 1) other aromatic
substrates reacted effectively to
give diols in good yields with
high anti selectivities beyond
[
a]
[b]
Entry Ar
R
R’
Product
Yield of 3 [%]
ee of 3 [%]
1
2
3
4
Ph
Et
Et
Et
Et
Me 3a
77
75
4-MeO-C
6
H
4
Me 3b
Me 3c
Me 3d
60
76
76
86
80
53
9
4%. Less electrophilic anisal-
dehyde delivered diols in
a
4-Me-C
6
H
4
slightly lower yield but with
higher enantiocontrol (~86%,
entry 2). The degree of enantio-
selectivity for the reaction
strongly depends on the elec-
tronic nature of the substituents
at the para position on the aro-
matic ring. The reactions with
6 4
4-Cl-C H
5
4-Br-C
6
H
4
Et
Me 3e
75
55
4
-methoxy- and 4-methylbenz-
aldehyde proceeded to give
enantioselectivities in the range
of 86 and 80%, respectively
6
7
2-naphthyl
Ph
Et
Pr
Me 3 f
Et 3g
67
77
70
65
(
entries 2 and 3). On the con-
trary, low enantioselectivity was
observed in the reaction of ben-
zaldehyde (75%) and 4-chloro-
(
53%) and 4-bromobenzalde-
8
9
1
4-MeO-C
6
H
4
Pr
Et 3h
Me 3i
Me 3j
25
85
45
80
73
75
hyde (55%). The last two alde-
hydes underwent condensation
with slightly worse diastereoiso-
meric control (about 91–93%).
To gather more information
about the influence of the elec-
tron nature of the aldehyde on
reaction enantioselectivity, we
decided to check the reactivity
of the 4-nitrobenzaldehyde, ex-
pecting isolation of 1,3-diol
with low enantioselectivity.
Indeed, aldol-Tishchenko con-
densation of 3-pentanone with
this substrate resulted in the
formation of the expected 1,2-
anti-1,3-anti-diol 5 in 28% yield
and with a low level of ee
Ph
Ph
Ph
0
4-MeO-C
6
H
4
1
1
1
1
2
3
4-Cl-C
6
H
4
Ph
Me 3k
Me 3l
Me 3m
70
72
89
60
70
72
2-naphthyl
Ph
Ph
4-Cl-C
6
H
H
4
4
(
Scheme 3).
Interestingly, the main prod-
1
4
4-MeO-C
6
H
4
4-Cl-C
6
Me 3n
92
78
uct was accompanied by its 1,2-
syn-1,3-anti isomer 6, isolated
[
a] Isolated yield. [b] The ee values of esters and diols were determined by HPLC (Chiralpak AD-H and AS-
H columns).
Chem. Eur. J. 2006, 12, 8158 – 8167
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8161

J. Mlynarski et al.
with a similar level of enantioselectivity (14%). Observed
loss of diastereoselectivity for substrates with strong elec-
tron-donating substituents at the para position suggested
nonselective formation of the six-membered ring in the
postulated active site (Schemes 1 and 3).
relation between the aldol product and the Tishchenko
product, as well as their stereocorrelation (Scheme 4). Inde-
pendently prepared racemic hydroxyketone 8 was subjected
Scheme 4. Evans–Tishchenko reduction of aldol 8 with benzaldehyde:
A
H
R
U
G
3
a) PhCHO, Yb(OTf) /1g (20 mol%), DME, RT, 3 h; b) NaOMe, MeOH.
to the Evans–Tishchenko catalytic reduction under condi-
tions elaborated previously. With ligand 1g, corresponding
diol 3i was obtained in 68% yield and with 70% ee. These
results strongly indicate that, with the catalyst composed of
ytterbium triflate and ligand 1g, the Tishchenko reduction
step is slower than the retroaldol reaction and it is the
stereo-determining step. It is noteworthy that starting from
syn-aldol 8, the 1,2-anti-1,3-anti-Tishchenko product was
formed exclusively. This observation confirms, as postulated
previously, the essential role of the Tishchenko reaction step
in controlling the stereoselectivity of the whole domino
Scheme 3. Aldol-Tishchenko condensation of 3-pentanone with p-nitro-
benzaldehyde: a) Yb
A
C
H
T
R
E
U
N
G
3
(OTf) /1g, DME, RT, 20 h; b) NaOMe, MeOH;
c) DMP, acetone, RT, 2 h.
Reaction mechanism: To gather more information about the
origin of the enantioselectivity, we examined the correlation
between the Hammett aromatic substituent constants (sp)
and the observed ee.
A linear Hammett plot with a negative rho value was ob-
served as shown in Figure 1. This data suggests that the re-
action proceeds by coordination of the aldehyde substrate
to chiral ytterbium complex. This also indicates that the co-
ordination step should be involved in at least the enantiode-
terminating step.
[15a]
process.
Based on our own and previous
[14h,15a]
observations, a
plausible mechanism for the aldol-Tishchenko reaction can
be described (Scheme 5).
The first step of the reaction most likely produces a mix-
ture of syn and anti-aldolates 10 as a result of a reversible
reaction of the metal enolate with aldehyde. Isomerization
Figure 1. Hammett plot for the enantioselective aldol-Tishchenko reac-
2
tion of substituted benzaldehydes (y=À0.7652xÀ0.0714; r =0.942).
To ascertain whether this reaction operates under similar
Curtin–Hammett conditions as postulated in Scheme 1 (in-
cluding initial formation of the aldol from an aldehyde and
ethyl ketone), we designed an experiment to illustrate the
Scheme 5. Proposed mechanism for the aldol-Tishchenko reaction.
8162
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 8158 – 8167

Aldol-Tishchenko Reaction
FULL PAPER
between both aldols (or corresponding metal aldolates) may
proceed by means of an enolization–protonation mechanism
via aldol enolate 9 under the control of a Lewis acid cata-
lyst. Metal aldolate 9 can form a hemiacetal metal alkoxide
with a second equivalent of aldehyde. High stereoselectivity
in the Tishchenko step arises from favored formation of the
transition structure 11 in which the ytterbium atom is chelat-
ed by the alkoxide and carbonyl oxygen atoms. All bulky
substituents in the postulated transition state occupy ener-
asymmetric induction was highly improbable, but not impos-
sible in the light of the different nature of aliphatic and aro-
matic ketones. Although the presence of (1R,3S)-configured
carbon atoms in 3a would be highly surprising from a syn-
thetic viewpoint, we could not rigorously exclude this possi-
bility, except by additional analysis.
As resolving of this problem requires a really reliable and
unambiguous technique we decided to apply circular dichro-
ism spectroscopy (CD). To determine the absolute configu-
ration at C1 and C3 carbon atoms in compound 3a, the di-
molybdenum method appeared to be a very convenient,
[17]
getically favorable equatorial positions.
The reaction is
completed by intramolecular hydride delivery towards a
proaxially oriented carbonyl group. A syn-aldolate 10 can
undergo a similar reaction with a slower rate through transi-
tion state 12 with the alkyl group in an axial position. The
energetically unfavorable structure of the transition state
can explain the minor formation of the 1,2-syn-1,3-anti-diols
[25]
straightforward, and versatile technique.
The method involves the in situ formation of chiral com-
plexes of optically active 1,3-diols with the achiral dimolyb-
denum tetraacetate [Mo2(OAc) ] acting as an auxiliary chro-
A
H
R
U
G
4
mophore. The resulting CD spectra are suitable for the as-
signment of absolute configuration, as the observed sign of
Cotton effects (CE) arising within the d–d absorption bands
of the metal cluster depends upon the chirality of the 1,3-
diol ligands. An additional advantage of the method is the
fact that in the chiral Mo complex an internal conformation-
al mobility of a flexible 1,3-diol molecule becomes substan-
tially restricted due to the steric requirements of the stock
complex. As a result, the molecule appears to exist only as a
single conformer in which both hydroxy groups adopt a syn-
periplanar orientation. Consequently, the determination of
the absolute configuration becomes possible on the basis of
the sector rule correlating the sign of the Cotton effect (CE)
occurring around 400 nm with the molecular structure of the
1,3-diol studied. The rule is formulated as follows provided
that in the chiral complex the conformation with synperipla-
nar oriented hydroxy groups is preferred: The sign of the
CE occurring around 400 nm in chiral Mo complexes with
1,3-diols is the same as the sign of the sector in which the
1
4. Thus, after fast isomerisation of a syn-aldol into anti-
aldol, the subsequent Tishchenko step proceeds via the
more favorable transition state to afford the observed 1,2-
anti-1,3-anti products 13.
Determination of the absolute configuration of 1,3-diols: To
get an insight into the sense of the asymmetric induction of
the aldol-Tishchenko reaction resulting from the elaborated
procedure we compared the optical rotation and results
from HPLC analysis of the obtained diols with published
data. The absolute configuration of anti-diol 3i was deter-
mined as (1S,3S) by a comparison of the optical rotation of
3
i
([a] =À13.0) with those of authentic compounds
D
[15a]
[
(1S,3S) enantiomer, [a] =À12.1, 84% ee].
The same
D
analysis in the case of p-chloro-substituted diol 3k con-
firmed the same (1S,2S,3S) orientation of asymmetric
[15a]
carbon atoms ([a] =+1.1, 70% ee; lit.
er, [a] =+1.3, 95% ee). Thus it is important to underline
that application of the (1R,2S)-1-phenyl-2-(1-pyrrolidynyl)-
(1S,3S) enantiom-
D
D
[25]
majority of the molecule is located.
1
-propanol (1g) as a chiral ligand resulted in the formation
of diols 3i and 3k with the same (1S,3S) configuration
Figure 2).
CD data of in situ formed Mo complexes of diol 3a and,
for comparison purposes, of compounds 3l (analogue of 3i
with two different chromophore groups) and 3k (the abso-
lute configuration of which is known), are collected in
Table 3 and Figure 3. For this purpose diol 3l was used in-
(
Table 3. CD data of in situ formed Mo complexes of compounds 3a, 3k,
and 3l recorded in DMSO (ligand-to-metal molar ratio 1:1) in the spec-
tral range 330–550 nm. Values are given as De’ [nm].
Figure 2. Deduced absolute configuration of compounds 3a, 3k, and 3i.
Diol
Band 1
Band 2
3
3
3a
k
l
À0.21 (373.0)
À0.10 (367.0)
À0.12 (369.5)
+0.08 (432.5)
+0.03 (434.0)
+0.04 (430.5)
Determination of the absolute configuration of the diol
3
a was, however, more puzzling. While its relative configu-
ration was proved as 1,2-anti-1,3-anti, the absolute configu-
rations at C1 (and linked configuration at C2, C3) was still
not clear. We expected the same configuration of the asym-
metric carbon atoms, that is, (1S,2S,3R) (Figure 2) as ob-
served for compounds 3i and 3k. Analysis of the literature
stead of compound 3i, which is not suitable for this analysis
because of the two identical chromophore groups in the
molecule.
As can be seen, the sign of CE at around 430 nm is posi-
tive in all cases studied. Thus, the stereochemistry at C1 and
C3 carbon atoms in compounds 3a, 3k and 3l must be the
same. According to the sector rule, a positive CE at 430 nm
[24]
data [(1R,3S)-enantiomer: [a] =À7.6, 99% ee] suggested,
D
however, formation of the same isomer (1R,3S) of com-
pound 3a ([a] =À36, 75% ee). Such a reverse sense of
D
Chem. Eur. J. 2006, 12, 8158 – 8167
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8163

J. Mlynarski et al.
Figure 3. CD spectra of in situ formed Mo complexes of diols 3k (b), 3l (d) and 3a (c) recorded in DMSO (left) and projection of the sector
rule for diol 3a (right).
corresponds to a (1S,3R) absolute configuration in com-
pound 3a, whilst in the case of compounds 3k and 3l it cor-
responds to a (1S,3S) absolute configuration (Figure 3,
strated as a useful method to overcome the retroaldol reac-
tion problem of a direct aldol reaction of ethyl ketones. Ali-
phatic ketones were demonstrated to be convenient sub-
strates in the direct asymmetric aldol-Tishchenko reaction
for the first time. The presented methodology offers a
simple yet powerful way to prepare anti-1,3-diols in one step
from unmodified substrates with high diastereo- and good
enantiocontrol.
[26]
right).
To assure the configurational assignment at C1 in com-
pound 3a by the independent route we decided to apply the
[27]
benzene sector rule. According to the rule, the absolute
configuration at the stereogenic center contiguous to the
benzene ring can be established unambiguously on the basis
1
of the sign of the excitation attributed to the L benzene
b
transition occurring around 260 nm. In the case of com-
pound 3a, the rule predicts a positive sign of the CE at
around 260 nm for an (1S) isomer, which is in excellent
Experimental Section
General: Ytterbium
drich) and trifluoromethanesulfonic acid (Fluka) was dried for 24 h at
008C under vacuum. All reactions were carried out under argon. Opti-
A
H
R
U
G
A
H
R
N
[27]
agreement with the experimental data (Figure 4).
2
It is worth pointing out that the configurational assign-
ment to compound 3a has been achieved by using two dif-
ferent CD rules for the analysis of the chiroptical properties.
Therefore, this assignment can be considered as safe despite
its empirical character. Hence, the structure proposed for
compound 3a and its stereochemical assignment presented
previously in the literature clearly needs revision.
cal rotations were measured with a JASCO Dip-360 Digital Polarimeter
1
at room temperature. H NMR spectra were recorded on Varian-400 and
Bruker-500 spectrometers in CDCl with Me Si as the internal standard.
3
4
HRMS were taken on a Mariner PerSeptive Biosystems mass spectrome-
ter with time-of-flight (TOF) detector. IR spectra were taken with a
Perkin–Elmer FTIR-1600 spectrophotometer. Reactions were controlled
by using TLC on silica (Merck alu-plates (0.2 mm)). All reagents and sol-
vents were purified and dried according to common methods. All organic
2 4
solutions were dried over Na SO . Reaction products were purified by
flash chromatography by using Merckꢁs Kieselgel 60 (240–400 mesh).
HPLC analyses were performed on a Knauer-HPLC system equipped
with Daicel columns with a chiral stationary phase, detection at 254 nm.
Conclusion
CD analysis: UV measurements were made on a Cary 100 spectropho-
tometer in acetonitrile and DMSO (for UV-spectroscopy, Fluka). CD
spectra were measured at room temperature in acetonitrile (for UV-spec-
We have established a new catalyst for the enantioselective
aldol reaction between aldehydes and aliphatic ketones to
give aldol-Tishchenko products with a dramatic increase in
molecular complexity created in a single operation. 1,3-Diol
monoesters were formed in good to excellent yields, with
high anti diastereocontrol, and with up to 85% ee. More-
over, this direct catalytic aldol protocol has been demon-
À4
troscopy, Fluka) with solutions at concentrations 8”10 m on a Jasco 715
spectrophotometer by using cells with path length 0.1 to 1 cm (spectral
À6
À6
À1
band width 2 nm, sensitivity 5”10 or 10”10 [DA-unitnm ], where
DA = A is the difference in the absorbance. De is expressed in
L
À1
ÀA
R
À1
[
Lm cm ] units. For CD measurements with [Mo
chiral 1,3-diol (2.0–3.0 mg, approximately 0.0015mL ) was dissolved in a
2
A
H
R
U
G
4
], the solid
À1
Figure 4. CD (b) and UV (c) spectra of diol 3a recorded in acetonitrile in the range of 230–300 nm.
8164
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 8158 – 8167

Aldol-Tishchenko Reaction
FULL PAPER
3
stock solution of the dimolybdenum tetraacetate (4.4–4.7 mg, approxi-
mately 0.0015mL ) in DMSO so that the molar ratio of the stock com-
plex to ligand was about 1:1, in general. The spectra were measured im-
mediately after mixing of components and repeated after 1, 2, and 3 h,
for comparison purposes. Some of the De’ values were very small, but
nevertheless the signal-to-noise ratio in all cases was better than at least
1H), 2.34 (s, 3H), 2.44 (s, 2H; 2”OH), 3.72 (ddd,
J(H,H)=2.2, 4.6,
A
T
E
G
À1
3
3
ACHTREUNG
8.9 Hz, 1H), 4.67 (d,
J(H,H)=6.8 Hz, 1H), 7.16 (d, J(H,H)=7.7 Hz,
2H; Ar), 7.20–7.26 ppm (m, 2H; Ar); C NMR (100 MHz, CDCl3, 258C,
TMS): d=10.6, 11.4, 21.0, 26.6, 43.3, 74.0, 78.1, 126.1, 129.0, 137.0,
140.7 ppm; IR (film): n˜ =3351, 2966, 2936, 1514, 1458, 1379, 1103 cm
A
H
R
U
G
1
3
À1
;
+
HRMS (ESI): calcd for C13
HPLC (Chiralpak AD-H, hexane/iPrOH 9:1, flow rate=1 mLmin , l=
54 nm): t =8.4 min, t =9.1 min (major).
(1S,2S,3R)-1-(4-Chlorophenyl)-2-methylpentane-1,3-diol (3d): H NMR
20 2
H O : 231.1356 [M+Na] ; found: 231.1354;
À1
1
0:1.
2
1
2
General procedure for the aldol-Tishchenko reaction: Ytterbium(iii) tri-
A
H
R
U
G
1
flate (125 mg 0.20 mmol) was placed in an oven-dried flask with a mag-
netic stirring bar and the flask was heated at 2008C for 10 min in vacuo
and then flushed with argon. After the flask had been cooled down to
RT, a solution of ligand 1g (164 mg, 0.80 mmol) in DME (2 mL) was
added. The resulting solution was stirred for 30 min at RT under argon.
The catalyst 3-pentanone (100 mL, 0.95 mmol) and benzaldehyde (101 mL,
ACHTREUNG
3
(400 MHz, CDCl
3
, 258C, TMS): d=0.88 (d,
J(H,H)=7.1 Hz, 3H), 0.90
(H,H)=7.4 Hz, 3H), 1.39–1.49 (m, 1H), 1.50–1.58 (m, 1H), 1.85–
A
H
R
U
G
3
ACHTREUNG
(t,
J
3
ACHTREUNG
1.92 (m, 1H), 2.60 (s, 2H; 2”OH), 3.68 (ddd,
J
3
ACHTREUNG
1H), 4.69 (d,
J
1
3
3
1
.00 mmol) were then added successively to the solution. The resulting
À1
+
mixture was stirred for 20 h at RT and then poured onto a silica-gel
column and eluted with hexane/ethyl acetate 9:1 to afford separated
esters 2a/b as an oil. The first fraction contained ester 2a and the second
ester 2b. Esters 2a/b so obtained were dissolved in MeOH (2 mL) and
treated with NaOMe (5–10 mol%) overnight. The resulting mixture was
purified by column chromatography on silica gel (hexane/ethyl acetate
2
1491, 1380, 1090 cm ; HRMS (ESI): calcd for C12H17ClO : 251.0809 [M
+Na]; found: 251.0818; HPLC (Chiralpak AS-H, hexane/iPrOH 9:1,
À1
flow rate=1 mLmin , l=254 nm): t
1
=7.1 min, t
2
=10.4 min (major).
1
A
H
R
N
(1S,2S,3R)-1-(4-Bromophenyl)-2-methylpentane-1,3-diol (3e): H NMR
3
(
(
400 MHz, CDCl
3
, 258C, TMS) d=0.89 (d,
J(H,H)=7.4 Hz, 3H), 1.37–1.47 (m, 1H), 1.48–1.59 (m, 1H), 1.84–
1.91 (m, 1H), 2.46 (brs, 1H; OH), 3.51 (brs, 1H; OH), 3.65 (ddd,
J(H,H)=7.1 Hz, 3H), 0.90
A
H
R
U
3
t,
A
H
R
U
G
3
:2) to afford the diol 3a. Based on this general procedure all other diols
3
3
ACHTREUNG
were isolated.
J
A
H
R
U
G
J(H,H)=6.4 Hz, 1H), 7.20–7.27
1
3
A
C
H
T
R
E
U
N
G
(1S,2R,3R)-1-Hydroxy-2-methyl-1-phenylpent-3-yl benzoate (2a) and
3
,
[
14h,m]
(
2
(
(
(
1S,2S,3R)-3-hydroxy-2-methyl-1-phenylpentyl benzoate (2b):
a: Yield: 42%; oil; [a] =+3.3 (c=1.00 in CH Cl
400 MHz, CDCl , 258C, TMS): d=0.75 (d,
(H,H)=J 7.4 Hz, 3H), 1.55–1.76 (m, 1H), 1.81–2.12 (m, 2H), 3.71
Ester
258C, TMS): d=10.5, 11.3, 26.7, 43.1, 74.0, 77.6, 121.0, 127.9, 131.3,
142.9 ppm; IR (film): n˜ =3339, 2966, 2936, 1592, 1487, 1380, 1098 cm
HRMS (ESI): calcd for C H17BrO : 295.0304 [M+Na] ; found:
12 2
295.0307; HPLC (Chiralpak AS-H, hexane/iPrOH 9:1, flow rate=
1
À1
, 72% ee); H NMR
;
D
2
2
3
+
3
J
A
H
R
U
G
3
t,
J
J
A
C
H
T
R
E
U
N
G
3
3
À1
d,
A
C
H
T
R
E
U
N
G
(H,H)=J=3.8 Hz, 1H; OH), 4.19 (dd,
J
A
H
R
U
G
1 mLmin , l=254 nm): t
1
=6.9 min, t
2
=10.1 min (major).
(3 f):
(H,H)=7.4 Hz, 3H), 0.90
3
1
5
8
1
1
.62 (ddd,
J
A
C
H
T
R
E
U
N
G
A
H
R
U
G
H NMR
1
3
3
ACHTREUNG
.10 ppm (m, 2H; Ar); C NMR (100 MHz, CDCl
3
, 258C, TMS): d=9.9,
J
3
3
0.5, 25.7, 44.3, 75.7, 75.8, 127.0, 127.6, 128.3, 128.4, 129.7, 130.2, 133.2,
42.8, 167.6 ppm; HPLC (Chiralpak AD-H, hexane/iPrOH 9:1, flow
(d,
A
H
R
U
G
3
J(H,H)=
A
C
H
T
R
E
U
N
G
À1
rate=1 mLmin , l=254 nm): t
2
1
=12.3 min, t
2
=21.6 min (major). Ester
1
13
b: Yield: 39%; [a]
D
=À8.5 (c=0.75 in CH
2
Cl
2
, 73% ee); H NMR
3
3
(
(
400 MHz, CDCl
t,
3
, 258C, TMS): d=0.75 (d,
J
A
H
R
U
G
3
À1
J
A
C
H
T
R
E
U
N
G
(H,H)=7.3 Hz, 3H), 1.36–1.48 (m, 1H), 1.56–1.67 (m, 1H), 2.02–
.20 (m, 1H), 2.52 (brs, 1H; OH), 3.75 (ddd,
;
3
+
2
1
2
4
1
1
J
A
H
R
U
G
16 20 2
HRMS (ESI): calcd for C H O : 267.1355 [M+Na] ; found: 267.1368;
HPLC (Chiralpak AS-H, hexane/iPrOH 9:1, flow rate=1 mLmin , l=
254 nm): t =8.7 min, t =11.7 min (major).
3
À1
H), 5.95 (d, J
A
C
H
T
R
E
U
N
G
1
3
3
1
2
3.0, 71.2, 78.8, 127.4, 128.1, 128.3, 128.4, 129.7, 129.9, 133.1, 139.4,
1
A
H
R
U
G
66.5 ppm; HPLC (Chiralpak AD-H, hexane/iPrOH 97:3, flow rate
CDCl
3
, 258C, TMS): d=0.84 (t, J
.4 Hz, 3H), 1.15–1.70 (m, 7H), 2.88 (d, J
(H,H)=4.6 Hz, 1H; OH), 3.62–3.85 (m, 1H), 4.89 (t, J
A
H
R
U
(H,H)=6.9 Hz, 3H), 0.93 (t, J
A
C
H
T
R
E
U
N
G
(H,H)=
(H,H)=4.9, 1H; OH), 3.47 (d,
À1
mLmin , l=254 nm): t
1
=23.2 min (major), t
2
=24.3 min; (Chiralpak
3
ACHTREUNG
7
À1
OD-H, hexane/iPrOH 97:3, flow rate=1 mLmin
.1 min, t =8.7 min (major).
(1S,2S,3R)-2-Methyl-1-phenylpentane-1,3-diol (3a):
,
l=254 nm):
t
1
=
3
3
ACHTREUNG
J
A
H
R
U
G
8
2
13
1
3
[
14h,m, 28]
A
C
H
T
R
E
U
N
G
[a]
, 258C, TMS):
(H,H)=7.4 Hz, 3H), 1.39–
D
=À36.2
TMS): d=12.5, 13.9, 18.4, 19.4, 35.6, 50.3, 71.7, 75.7, 126, 127.1, 128.2,
143.9 ppm; IR (film): n˜ =3307, 2960, 2874, 1455, 1198, 1117, 1013,
1
(
c=0.60 in CH
2
Cl
2 3
, 75% ee); H NMR (500 MHz, CDCl
3
3
À1
+
d=0.87 (d,
J
A
C
H
T
R
E
U
N
G
(H,H)=7.1 Hz, 3H), 0.91 (t,
.48 (m, 1H), 1.51–1.60 (m, 1H), 1.94 (dq,
J
A
T
E
N
701 cm ; HRMS (EI): calcd for C H O : 204.1514 [MÀH O] ; found:
1
4
22
2
2
3
1
2
3
1
J
A
H
R
U
G
204.1508; HPLC (Chiralpak AS-H, hexane/iPrOH 9:1, flow rate=
3
3
À1
.46 (d, J
A
C
H
T
R
E
U
N
G
(H,H)=4.9 Hz, 1H; OH), 3.09 (d, J
A
H
R
U
G
1 mLmin , l=254 nm): t =5.1 min, t =5.7 min (major).
1
2
3
.70–3.75 (m, 1H), 4.72 (dd,
J
A
H
R
U
G
1
A
H
R
U
G
1
3
H; Ar), 7.35 ppm (d, 4H; Ar); C NMR (125 MHz, CDCl
3
, 258C,
3
ACHTREUNG
(
(
(
TMS): d=10.6, 11.3, 26.7, 43.4, 74.0, 78.3, 126.3, 127.4, 128.4, 143.9 ppm;
HPLC (Chiralpak AD-H, hexane/iPrOH 9:1, flow rate=1 mLmin , l=
2
3
ACHTREUNG
t,
J
À1
3
J
J
ACHTREUNG
54 nm): t
1
=7.8 min, t
2
=10.1 min (major).
3
ACHTREUNG
3
1
H), 6.88 (d,
J
ACHTREUNG
1
13
A
C
H
T
R
E
U
N
G
(1S,2S,3R)-1-(4-Methoxyphenyl)-2-methylpentane-1,3-diol (3b): H NMR
3
3
(
400 MHz, CDCl
3
, 258C, TMS) d=0.83 (d,
(H,H)=7.2 Hz, 3H), 1.40–1.50 (m, 1H), 1.52–1.60 (m, 1H), 1.92 (m,
H), 2.30 (brs, 2H; 2”OH), 3.72–3.76 (m, 1H), 3.81 (s, 3H), 4.66 (d,
J
A
H
R
U
G
3
À1
(
t, J
A
C
H
T
R
E
U
N
G
+
1
C H O : 275.1620 [M+Na] ; found: 275.1621; HPLC (Chiralpak AD-H,
hexane/iPrOH 9:1, flow rate=1 mLmin
(major), t =9.4 min.
1
5
24
3
3
À1
J(H,H) =7.1 Hz, 1H), 6.87–6.90 (m, 2H; Ar), 7.25–7.28 ppm (m, 2H; Ar);
,
l=254 nm): t1 =8.1 min
1
3
C NMR (100 MHz, CDCl
3
, 258C, TMS): d=10.7, 11.5, 26.5, 43.5, 55.2,
2
7
1
4.2, 77.7, 113.7, 127.4, 135.9, 158.9 ppm; IR (film): n˜ =3361, 2964, 2936,
[15a,28]
A
H
R
U
G
[a]
D
=À13.0
À1
612, 1513, 1248, 1175 cm ; HRMS (ESI): calcd for C13
H
20
O
3
: 247.1305
[15a]
(
8
=À12.1 (c=1.0 in CH
2 2
Cl ,
+
[
M+Na] ; found: 247.1319; HPLC (Chiralpak AD-H, hexane/iPrOH 9:1,
3
, 258C, TMS): d=0.71 (d, J
A
C
H
T
R
E
U
N
G
(H,H)=
À1
flow rate=1 mLmin , l=254 nm): t
1
=11.3 min, t
2
=12.7 min (major).
7
.1 Hz, 3H), 2.14 (m, 1H), 3.05 (d, J(H,H)=3.4 Hz, 1H; OH), 3.15 (d,
J(H,H)=3.4 Hz, 1H; OH), 4.64 (dd, J(H,H)=3.6, 6.6 Hz, 1H), 4.96 (t,
1
3
A
C
H
T
R
E
U
N
G
(1S,2S,3R)-2-Methyl-1-(4-methylphenyl)pentane-1,3-diol (3c): H NMR
3
3
13
(
400 MHz, CDCl
3
, 258C, TMS) d=0.85 (d,
J
A
H
R
U
G
C NMR
3
(
t,
J
A
C
H
T
R
E
U
N
G
(H,H)=7.4 Hz, 3H), 1.38–48 (m, 1H), 1.49–1.61 (m, 1H), 1.92 (m,
Chem. Eur. J. 2006, 12, 8158 – 8167
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8165

J. Mlynarski et al.
1
26.9, 127.5, 127.9, 128.3, 142.5, 143.4 ppm; HPLC (Chiralpak AD-H,
CDCl
3
, 258C, TMS): d=10.6, 11.3, 23.6, 24.0, 24.8, 41.7, 71.1, 77.6, 100.9,
À1
hexane/iPrOH 9:1, flow rate=1 mLmin , l=254 nm): t
1
1
=11.4 min, t
2
=
126.9, 127.6, 128.4, 142.0 ppm; IR (film): n˜ =2984, 2966, 2937, 2878, 1496,
+
1455, 1378, 1223 cm ; HRMS (EI): calcd for C H O : 234.1619 [M] ;
15 22 2
À1
4.5 min (major).
A
C
H
T
R
E
U
N
G
(1S,2S,3S)-1-(4-Methoxyphenyl)-2-methyl-3-phenylpropane-1,3-diol
found: 234.1618; HPLC (Chiralpak AS-H, hexane/iPrOH 99:1, flow
[
29,30]
1
3
À1
(
3j):
H NMR (200 MHz, CDCl
3
, 258C, TMS): d=0.66 (d, J
A
H
R
U
G
rate=1 mLmin , l=254 nm): t
1
2
=3.3 min (major), t =3.6 min.
1
7
3
1
.1 Hz, 3H), 2.05–2.20 (m, 1H), 3.35 (brs, 1H, OH), 3.57 (brs, 1H; OH),
A
H
R
U
G
H NMR
3
3
3
.78 (s, 3H), 4.57 (d,
J
A
H
R
U
G
J
A
T
E
N
(H,H)=2.2 Hz,
(400 MHz, CDCl , 258C, TMS): d=0.87 (t,
(
(H,H)=7.4 Hz, 3H), 0.97
J(H,H)=7.1 Hz, 3H), 1.39–1.49 (m, 1H), 1.50–1.60 (m, 1H), 1.89–
3
ACHTREUNG
H), 6.84–6.88 (d,
J
1
3
Ar); C NMR (50 MHz, CDCl
7
n˜ =3368, 2968, 1611, 1513, 1451, 1248, 1175, 1032, 830 cm ; HRMS (EI):
calcd for C17
3
1
.97 (m, 1H), 2.40 (brs, 1H; OH), 3.60–3.64 (m, 1H), 4.07 (brs, 1H;
3
OH), 4.83 (d,
8
J(H,H)=6.0 Hz, 1H), 7.52–7.54 (m, 2H; Ar), 8.19–
A
H
R
U
G
À1
13
3
.22 ppm (m, 2H; Ar); C NMR (100 MHz, CDCl , 258C, TMS): d=
+
H
20
O
3
: 272.1412 [M] ; found: 272.1404; HPLC (Chiralpak
1
0.6, 11.4, 27.1, 43.1, 74.2, 77.8, 123.7, 127.2, 147.3, 151.8 ppm; IR (KBr):
À1
AD-H, hexane/iPrOH 9:1, flow rate=1 mLmin
6.2 min, t =20.8 min (major).
(1S,2S,3S)-1-(4-Chlorophenyl)-2-methyl-3-phenylpropane-1,3-diol
,
l=254 nm):
t
1
=
À1
n˜ =3524, 3393, 2919, 1606, 1523, 1457, 1348, 1090 cm ; HRMS (ESI):
calcd for C12
alpak AD-H, hexane/iPrOH 9:1, flow rate=1 mLmin , l=254 nm): t =
1
2
+
H
17NO
4
: 262.1049 [M+Na] ; found: 262.1057; HPLC (Chir-
À1
A
C
H
T
R
E
U
N
G
1
[
15a]
[15a]
(
3k):
[a]
D
=+1.1 (c=0.50 in CH
2
Cl
2
, 70% ee) (lit.
[a]
D
=+1.3 (c=
12.9 min, t =13.7 min (major).
2
1
1
0
1
6
.75 in CH
2
Cl
(H,H)=7.1 Hz, 3H), 2.18 (m, 1H), 3.00 (d,
2
, 95% ee)); H NMR (400 MHz, CDCl
3
, 258C, TMS) d=
1
A
H
R
U
G
H NMR
3
3
ACHTREUNG
.75 (d,
H; OH), 3.30 (d,
.6 Hz, 1H), 5.00 (t,
J
A
C
H
T
R
E
U
N
G
J
(H,H)=3.6 Hz,
3
ACHTREUNG
(
(
(
(H,H)=7.1 Hz, 3H), 1.02
J(H,H)=7.4 Hz, 3H), 1.67–1.80 (m, 2H), 1.84–1.92 (m, 1H), 2.27
3
3
ACHTREUNG
J
A
C
H
T
R
E
U
N
G
(H,H)=4.1 Hz, 1H; OH), 4.67 (dd,
J
3
J
A
C
H
T
R
E
U
N
G
J(H,H)=5.2, 11.7 Hz, 1H), 3.74 (brs, 1H;
1
3
3
OH), 5.28 (s, 1H), 7.51 (d, J
2
4
1
27.1, 127.5, 128.0, 128.4, 133.0, 141.9, 142.1 ppm; HPLC (Chiralpak AD-
À1
H, hexane/iPrOH 95:5, flow rate=1 mLmin , l=254 nm): t
t
1
=22.8 min,
2
=28.8 min (major).
C
: 262.1049 [M+Na] ; found: 262.1050; HPLC (Chiralpak AD-
A
C
H
T
R
E
U
N
G
(1S,2S,3S)-2-Methyl-1-(2-naphthyl)-3-phenylpropane-1,3-diol
(3l):
H, hexane/iPrOH 9:1, flow rate=1 mLmin , l=254 nm): t
=11.7 min,
1
3
ACHTREUNG
H NMR (200 MHz, CDCl
H), 2.14–2.34 (m, 1H), 3.48 (d,
3
, 258C, TMS): d=0.73 (d,
J
2
t =13.6 min (major). Based on the procedure described for 4 diol, 6 was
transformed into (1S,2R,3R)-1,3-O-isopropylidene-2-methyl-1-(4-nitro-
phenyl)pentane-1,3-diol (7).
3
ACHTREUNG
3
J
3
3
ACHTREUNG
J
A
C
H
T
R
E
U
N
G
(H,H)=3.8 Hz, 1H; OH), 4.78 (dd,
J(H,H)=3.3, 6.1 Hz, 1H), 4.96
(
(
brs, 1H), 7.20–7.31 (m, 5H; Ar), 7.38–7.50 (m, 3H; Ar), 7.75–7.83 ppm
1
3
A
H
R
U
G
m, 4H; Ar); C NMR (50 MHz, CDCl
4.4, 77.8, 124.1, 125.1, 125.8, 125.9, 126.1, 126.9, 127.6, 127.9, 128.2,
32.8, 133.1, 140.8, 142.4 ppm; IR (film): n˜ =3339, 3027, 2975, 2882, 1451,
3
, 258C, TMS): d=11.4, 45.4,
1
3
ACHTREUNG
diol (7): H NMR (500 MHz, CDCl
6
1.55–1.67 (m, 2H), 1.97–2.04 (m, 1H), 3.29 (dt,
1
7
1
7
3
1
À1
+
56 cm ; HRMS (ESI): calcd for C20
20 2
H O : 315.1355 [M+Na] ; found:
15.1345; HPLC (Chiralpak AD-H, hexane/iPrOH 9:1, flow rate=
À1
3
8.20 ppm (m, 2H; Ar); C NMR (125 MHz, CDCl , 258C, TMS): d=
mLmin , l=254 nm): t
1
=16.8 min, t
2
=21.7 min (major).
1
0.4, 12.9, 23.7, 25.1, 27.6, 41.4, 70.3, 77.2, 101.2, 123.2, 126.6, 146.7,
A
C
H
T
R
E
U
N
G
(1S,2R,3S)-1-(4-Chlorophenyl)-2-methyl-3-phenylpropane-1,3-diol (3m):
1
3
148.1 ppm.
H NMR (200 MHz, CDCl
H), 1.99–2.24 (m, 1H), 3.51 (d,
3
, 258C, TMS): d=0.67 (d,
J
ACHTREUNG
3
3
J
A
H
R
U
G
Ligand synthesis: general procedure for the synthesis of N,N-dialkylnor-
3
3
[22]
J
A
C
H
T
R
E
U
N
G
(H,H)=3.9 Hz, 1H; OH), 4.59 (dd, J
A
H
R
U
G
ephedrines:
A mixture of (1S,2R)-(+)- or (1R,2S)-(À)-norephedrine
1
3
1
H), 7.00–7.40 ppm (m, 9H; Ar); C NMR (50 MHz, CDCl
3
, 258C,
(10 mmol), alkyl iodide (20 mmol),
K
2
CO (20 mmol), and CH CN
3
3
TMS): d=11.2, 45.6, 73.6, 77.7, 74.0, 126.1, 127.3, 127.6, 128.0, 128.4,
(10 mL) was refluxed for 2–24 h. After this time, the reaction mixture
was cooled to room temperature and filtered. The filtrate was concentrat-
ed under reduced pressure. The residue was purified by column chroma-
1
1
32.4, 141.0, 143.1 ppm; IR (KBr): n˜ =3364, 2981, 2927, 1491, 1454, 1091,
+
À1
011 cm ; HRMS (EI): calcd for C16
H
17ClO
2
: 258.0811 [MÀH
2
O] ;
found: 258.0807; HPLC (Chiralpak AD-H, hexane/iPrOH 9:1, flow
rate=1 mLmin , l=254 nm): t
tography on silica gel (EtOAc).
À1
=9.9 min, t
2
=15.0 min (major).
[22]
1
A
H
R
U
G
[a]
=À22.7 (c=
D
1
A
C
H
T
R
E
U
N
G
(1S,2R,3S)-1-(4-Chlorophenyl)-3-(4-methoxyphenyl)-2-methylpropane-
0.30 in CHCl ); H NMR (200 MHz, CDCl , 258C, TMS): d=0.87 (d,
(H,H)=7.1 Hz, 6H), 2.46 (q,
7.1 Hz, 4H), 2.95–3.08 (m, 1H), 4.22 (brs, 1H; OH), 4.67 (d,
[
29]
1
3
3
1
,3-diol (3n):
H NMR (200 MHz, CDCl
3
, 258C, TMS): d=0.67 (d,
J
J
A
C
H
T
R
E
U
N
G
(H,H)=
3
3
J
A
C
H
T
R
E
U
N
G
(H,H)=7.2 Hz, 3H), 2.03–2.20 (m, 1H), 3.00 (brs, 1H; OH), 3.50 (brs,
ACHTREUNG
(H,H)=
3
3
ACHTREUNG
1
2
H; OH), 3.80 (s, 3H), 4.59 (d, J
.2 Hz, 1H), 6.86–6.88 (d, J
A
H
R
U
G
(H,H)=
4.7 Hz, 1H), 7.08–7.20 ppm (m, 5H; Ar).
3
A
H
R
U
G
[22]
A
H
R
U
G
[a]
D
=À42.0
1
3
Ar); C NMR (50 MHz, CDCl
3
1
(
, 258C, TMS): d=0.80 (t,
(H,H)=7.0 Hz, 3H), 1.12–1.49 (m, 4H),
7
3
À1
n˜ =3368, 2971, 2935, 1611, 1512, 1249, 1175 cm ; HRMS (EI): calcd for
C
hexane/iPrOH 4:1, flow rate=1 mLmin , l=254 nm): t
2
.68–2.40 (m, 4H), 2.94–3.07 (m, 1H), 4.13 (brs, 1H; OH), 4.65 (d,
+
17
H
19ClO
3
: 306.1022 [M] ; found: 306.1016; HPLC (Chiralpak AD-H,
3
J(H,H)=5.0 Hz, 1H), 7.08–7.20 ppm (m, 5H; Ar).
A
H
R
U
G
À1
1
=8.1 min, t
2
=
[
22]
A
H
R
U
G
[a]
D
=À20.1
1
0.4 min (major).
1
(
, 258C, TMS): d=0.87 (t,
(H,H)=7.0 Hz, 4H), 1.34–1.53 (m, 8H),
ACHTREUNG
(1S,2S,3R)-1,3-O-Isopropylidene-2-methyl-1-phenylpentane-1,3-diol (4):
3
Camphorosulfonic acid (small crystal) was added to a solution of the 1,3-
diol 3a (97 mg, 0.5 mmol) in acetone and 2,2-dimethoxypropane (5 mL,
2
.12–2.41 (m, 4H), 2.94–3.07 (m, 1H), 4.16 (brs, 1H; OH), 4.66 (d,
3
J(H,H)=5.2 Hz, 1H), 7.08–7.18 ppm (m, 5H; Ar).
A
H
R
U
G
4
:1) at RT. The mixture was stirred at ambient temperature for 1 h, then
quenched with one drop of Et N, concentrated under reduced pressure,
and purified by flash chromatography (hexane/ethyl acetate 95:5) to
yield diacetonide 4 as an oil (113 mg, 93%); [a]
=À40.1 (c=0.65 in
, 258C, TMS): d=0.88 (d,
(H,H)=7.3 Hz, 3H), 1.43, 1.45 (2 s, 2”
H; OiPr), 1.39–1.46 (m, 1H), 1.49–1.58 (m, 1H), 2.00–2.07 (m, 1H),
A
H
R
U
G
D
=+90.9
3
1
(c=1.00 in CHCl
1.95 (m, 4H), 2.50–2.82 (m, 4H), 3.28 (d,
(brs, 1H; OH), 5.22 (d, J
7.15 ppm (m, 6H); C NMR (50 MHz): d=23.5, 52.9, 74.0, 76.8, 126.0,
D
1
CH
2
Cl
2
, 43% ee); H NMR (500 MHz, CDCl
3
3
3
J
A
C
H
T
R
E
U
N
G
(H,H)=6.8 Hz, 3H), 0.97 (t,
J
A
H
R
U
G
3
126.6, 127.0, 127.1, 127.4, 129.2, 137.4, 140.6 ppm; IR (film): n˜ =3465,
3
À1
3
.73 (ddd,
J
A
C
H
T
R
E
U
N
G
(H,H)=2.1, 4.6, 6.8 Hz, 1H), 3.95 (m, 1H), 4.24 (d,
3034, 2968, 2799, 1453 cm ; HRMS (ESI): calcd for C18
[M] ; found: 268.1693.
H
21NO: 268.1695
3
13
+
J
A
C
H
T
R
E
U
N
G
(H,H)=8.3 Hz, 1H), 7.26–7.42 ppm (m, 5H; Ar); C NMR (125 MHz,
8166
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 8158 – 8167

Aldol-Tishchenko Reaction
FULL PAPER
[
31]
A
C
H
T
R
E
U
N
G
(1S,2R)-2-Piperydynyl-1-phenylpropan-1-ol (1i):
[a]
258C, TMS): d=0.82 (d,
(H,H)=7.0 Hz, 3H), 1.35–1.62 (m, 6H), 2.41–2.57 (m, 4H), 2.63–2.75
D
=À1.2 (c=0.17
Bodnar, J. T. Shaw, K.A. Woerpel, J. Org. Chem. 1997, 62, 5674–
5675; e) C. Delas, O. Blacque, C. Moꢂse, J. Chem. Soc. Perkin Trans.
1 2000, 2265–2270; f) C. Delas, C. Moꢂse, Synthesis 2000, 251–254;
g) Y. Horiuchi, M. Taniguchi, K. Oshima, K. Utimoto, Tetrahedron
Lett. 1995, 36, 5353–5356; h) L. Lu, H.-Y. Chang, J.-M. Fang, J. Org.
Chem. 1999, 64, 843–853; i) I. Simpura, V. Navalainen, Tetrahedron
Lett. 2001, 42, 3905–3907; j) I. Simpura, V. Navalainen, Tetrahedron
1
3 3
in CHCl ); H NMR (200 MHz, CDCl ,
3
ACHTREUNG
J
3
(
7
6
m, 1H), 4.10 (brs, 1H; OH), 4.81 (d,
J(H,H)=4.2 Hz, 1H), 7.17–
.32 ppm (m, 5H; Ar); C NMR (50 MHz): d=10.2, 24.5, 26.5, 51.7,
A
H
R
U
G
1
3
4.5, 72.1, 125.9, 126.6, 127.8, 142.2 ppm.
2
2
003, 59, 7535–7546; k) C. Schneider, M. Hansch, Chem. Commun.
001, 1218–1219; l) C. Schneider, M. Hansch, T. Weide, Chem. Eur.
Acknowledgements
J. 2005, 11, 3010–3021; m) R. Mahrwald, B. Cosisella, Synthesis
996, 1087–1089; n) C. M. Mascarenhas, M. O. Duffey, S.-Y. Liu,
1
J.M. thanks the Alexander von Humboldt Foundation for the generous
donation of a HPLC system. Financial support by the Polish State Com-
mittee for Scientific Research (KBN Grant 3 T09A 126 27) is gratefully
acknowledged.
J. P. Morken Org. Lett. 1999, 1, 1427–1429; o) D. P. Curran, R. L.
Wolin, Synlett 1991, 317–318; p) F. Abu-Hasanayan, A. Streitwieser,
J. Org. Chem. 1998, 63, 2954–2960; q) for a review on the aldol-
Tishchenko reaction see: R. Mahrwald, Curr. Org. Chem. 2003, 7,
1
713–1723.
[
15] a) V. Gnanadesikan, Y. Horiuchi, T. Ohshima, M. Shibasaki, J. Am.
Chem. Soc. 2004, 126, 7782–7783; b) Y. Horiuchi, V. Gnanadesikan,
T. Ohshima, H. Masu, K. Katagiri, Y. Sei, K. Yamaguchi, M. Shiba-
saki, Chem. Eur. J. 2005, 11, 5195–5204.
[
1] a) Stereoselective Synthesis (Houben–Weyl), Vol. 1–10 (Eds.: G.
Helmchen, R. W. Hoffmann, J. Mulzer), Thieme, Stuttgart, 1996;
b) Comprehensive Asymmetric Catalysis, Vol. 1–3 (Eds.: E. N. Jacob-
sen, A. Pfaltz, H. Yamamoto), Springer, Berlin, 1999; c) Compre-
hensive Asymmetric Catalysis, Suppl. 1 (Eds.: E. N. Jacobsen, A.
Pfaltz, H. Yamamoto), Springer, Berlin, 2004.
2] Modern Aldol Reactions, Vol. 1–2 (Ed.: R. Mahrwald), Wiley-VCH,
Weinheim, 2004.
3] a) T. Mukaiyama, K. Narasaka, K. Banno, Chem. Lett. 1973, 1011–
[
[
16] J. Mlynarski, M. Mitura, Tetrahedron Lett. 2004, 45, 7549–7552.
17] a) D. A. Evans, A. H. Hoveyda, J. Am. Chem. Soc. 1990, 112, 6447–
6
449; b) E. R. Burkhardt, R. G. Bergman, C. H. Heathcock, Organo-
[
metallics 1990, 9, 30–44.
[
18] a) J. Mlynarski, J. Jankowska, B. Rakiel, Tetrahedron: Asymmetry
[
2
2
005, 16, 1521–1526; b) J. Jankowska, J. Mlynarski, J. Org. Chem.
006, 71, 1317–1321.
1
014; b) T. Mukaiyama, Org. React. 1982, 28, 203–232; c) T. Mu-
kaiyama, J. Matsuo in Modern Aldol Reactions (Ed.: R. Mahrwald),
Wiley-VCH, Weinheim, 2004, pp. 127–160.
4] a) M. Shibasaki, H. Sasai, T. Arai, Angew. Chem. 1997, 109, 1290–
[
19] a) O. Loog, U. Mꢃeorg, Tetrahedron: Asymmetry 1999, 10, 2411–
415; b) C. M. Mascarenhas, S. P. Miller, P. S. White, J. P. Morken,
Angew. Chem. 2001, 113, 621–623; Angew. Chem. Int. Ed. 2001, 40,
01–603.
2
[
1
311; Angew. Chem. Int. Ed. Engl. 1997, 36, 1236–1256; b) S. Matsu-
6
naga, T. Ohshima, M. Shibasaki, Adv. Synth. Catal. 2002, 344, 3–15;
c) M. Shibasaki, N. Yoshikawa, Chem. Rev. 2002, 102, 2187–2209;
d) S. Saito, H. Yamamoto, Acc. Chem. Res. 2004, 37, 570–579; e) W.
Notz, F. Tanaka, C. F. Barbas III, Acc. Chem. Res. 2004, 37, 580–
[
20] a) C. Schneider, M. Hansch, Synlett 2003, 837–840; b) recently
Mahrwald presented an interesting example of the aldol-Tishchenko
reaction of aldols: K. Rohr, R. Herre, R. Mahrwald, Org. Lett. 2005,
7
, 4499–4501.
5
91.
5] For a discussion on atom economy see: a) B. Trost, Science 1991,
54, 1471–1477; b) B. Trost, Angew. Chem. 1995, 107, 285–307;
[
[
[
[
[
21] For a preliminary communication see: J. Mlynarski, J. Jankowska, B.
[
Rakiel, Chem. Commun. 2005, 4854–4856.
2
22] K. Soai, S. Yokoyama, T. Hayasaka, J. Org. Chem. 1991, 56, 4264–
Angew. Chem. Int. Ed. Engl. 1995, 34, 259–281.
6] a) B. Alcaide, P. Almendros, Eur. J. Org. Chem. 2002, 1595–1601;
b) C. Palomo, M. Oiarbide, J. M. García, Chem. Soc. Rev. 2004, 33,
4
268.
23] S. D. Rychnovsky, B. Rogers, G. Yang, J. Org. Chem. 1993, 58, 3511–
515.
[
3
6
1
5–75; c) T. D. Machajewski, C.-H. Wong, Angew. Chem. 2000, 112,
406–1430; Angew. Chem. Int. Ed. 2000, 39, 1352–1374.
24] J. L. Vicario, D. Badía, E. Domínguez, M. Rodríguez, L. Carrillo, J.
Org. Chem. 2000, 65, 3754–3754.
25] a) J. Frelek, G. Snatzke, W. J. Szczepek, Fresenius J. Anal. Chem.
[
7] Y. M.A. Yamada, N. Yoshikawa, H. Sasai, M. Shibasaki, Angew.
Chem. 1997, 109, 1942–1944; Angew. Chem. Int. Ed. Engl. 1997, 36,
1
993, 345, 683–687; b) J. Frelek, W. J. Szczepek, W. Voelter, J. Prakt.
1
871–1873.
Chem./Chem.-Ztg. 1997, 339, 135–139; c) J. Frelek, M. Geiger, W.
Voelter, Curr. Org. Chem. 1999, 3, 145–174; d) J. Frelek, A. Klimek,
[
8] a) Y. M.A. Yamada, M. Shibasaki, Tetrahedron Lett. 1998, 39, 5561–
5
1
2
564; b) B. M. Trost, H. Ito, J. Am. Chem. Soc. 2000, 122, 12003–
2004; c) B. M. Trost, E. R. Silcoff, H. Ito, Org. Lett. 2001, 3, 2497–
500; d) N. Yoshikawa, Y. M.A. Yamada, J. Das, H. Sasai, M. Shiba-
´
P. Ruskowska, Curr. Org. Chem. 2003, 7, 1081–1104.
[
26] All hydroxy groups are located on the same side; the fact that the
centers at C3 are S-configured in 3k and R-configured in 3a simply
reflects the formalism of the Cahn–Ingold–Prelog nomenclature.
27] a) H. E. Smith, Application of the benzene sector and benzene chiral-
ity rules in Circular Dichroism Principles and Applications (Eds.: K.
Nakanishi, N. Berova, R. W. Woody), Wiley-VCH, New York, 2000,
pp. 397–429; b) L. P. Fontana, H. E. Smith, J. Org. Chem. 1987, 52,
saki, J. Am. Chem. Soc. 1999, 121, 4168–4178.
9] A. Berkessel, H. Groger, Asymmetric Organocatalysis, Wiley-VCH,
Weinheim, 2005.
[
[
[
10] M. Shibasaki, S. Matsunaga, N. Kumagai in Modern Aldol Reactions
Ed.: R. Mahrwald), Wiley-VCH, Weinheim, 2004, pp. 197–228.
(
[
[
11] N. Yoshikawa, M. Shibasaki, Tetrahedron 2001, 57, 2569–2579.
12] a) N. Yoshikawa, N. Kumagai, S. Matsunaga, G. Moll, T. Ohshima,
T. Suzuki, M. Shibasaki, J. Am. Chem. Soc. 2001, 123, 2466–2467;
b) B. M. Trost, H. Ito, E. R. Silcoff, J. Am. Chem. Soc. 2001, 123,
3
386–3389.
[
[
[
[
28] G. Bartoli, M. C. Bellucci, M. Bosco, R. Dalpozzo, E. Marcantoni,
L. Sambri, Chem. Eur. J. 2000, 6, 2590–2598.
29] N. Ichinose, K. Mizuno, T. Tamai, Y. Otsui, J. Org. Chem. 1990, 55,
3
367–3368.
4
079–4083.
30] K. Miura, T. Nakagawa, S. Suda, A. Hosomi, Chem. Lett. 2000, 150–
51.
[
13] For the examples of direct aldol reactions of ethyl ketones see: a) R.
Mahrwald, Org. Lett. 2000, 2, 4011–4012; b) R. Mahrwald, B.
Ziemer, Tetrahedron Lett. 2002, 43, 4459–4461; c) R. Mahrwald, B.
Schetter, Org. Lett. 2006, 8, 281–284.
1
31] K. Soai, M. Okudo, M. Okamoto, Tetrahedron Lett. 1991, 32, 95–96.
[
14] a) M. S. Kulpinski, F. F. Nord, J. Org. Chem. 1943, 8, 256–270;
b) F. J. Villani, F. F. Nord, J. Am. Chem. Soc. 1947, 69, 2608–2608;
c) A. Baramee, N. Chaichit, P. Intawee, C. Thebtaranonth, Y. Thebt-
aranonth, J. Chem. Soc. Chem. Commun. 1991, 1016–1017; d) P. M.
Received: March 14, 2006
Published online: July 26, 2006
Chem. Eur. J. 2006, 12, 8158 – 8167
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8167