ScIII-Doped zeolites as new heterogeneous catalysts: mukaiyama aldol reaction

Article abstract of DOI:10.1002/chem.200901647

Sc^III-doped solids based on zeolite materials have been investigated for the first time as catalysts in organic synthesis. Sc ^III-USY zeolite proved to be a novel and very efficient heterogeneous catalyst for the Mukaiyama aldol reaction. This easy-to-prepare catalyst exhibited wide scope and compatibility with functional groups and is very simple to use, easy to remove (by simple filtration), and is recyclable (up to three times without loss of activity).

Full text of DOI:10.1002/chem.200901647

FULL PAPER
DOI: 10.1002/chem.200901647
ScIII-Doped Zeolites as New Heterogeneous Catalysts:
Mukaiyama Aldol Reaction
Andrea Olmos,[a] Aurꢀlien Alix,[b] Jean Sommer,*[a] and Patrick Pale*[b]
Abstract: ScIII-doped solids based on zeolite materials have been investigated for
the first time as catalysts in organic synthesis. ScIII–USY zeolite proved to be a
Keywords: aldol reaction · hetero-
novel and very efficient heterogeneous catalyst for the Mukaiyama aldol reaction.
This easy-to-prepare catalyst exhibited wide scope and compatibility with function-
al groups and is very simple to use, easy to remove (by simple filtration), and is re-
cyclable (up to three times without loss of activity).
geneous catalysis
zeolites
·
scandium
·
Introduction
copper complexes[6] or scandium salts[7] as the active species.
A
microencapsulation cross-linking technique provided
The aldol reaction is one of the most important reactions
scandium triflate incarcerated in polymer micelles.[8] Alumi-
na has also been reported to be both catalyst and support,
but sonication was required.[9]
[1]
ꢀ
for the formation of C C bonds. Numerous methods have
been and are still being developed to increase the efficiency
and stereoselectivity of this reaction. Among them, the Mu-
kaiyama reaction emerged in the eighties as one of the most
effective routes to b-hydroxyketones, the so-called aldols.[2]
In the last few decades, further work has led to the develop-
ment of catalytic versions of the Mukaiyama aldol reaction,
especially with lanthanides or rare earth salts as the cata-
lyst.[3] Such catalyzed reactions offer a first step toward
greener routes to aldols. Heterogeneous catalysts could be a
good means of going further toward more environmentally
friendly aldolization because they allow recovery and recy-
cling. Surprisingly however, only a few versions based on
supported or heterogeneous catalysts have been reported so
far. Polystyrene-supported catalysts have been described
with phosphoramide[4] and oxazaborolidinone[5] and with
To the best of our knowledge, no zeolite-catalyzed version
of this reaction has been reported so far. Involved in zeo-
lite[10] and other superacid-catalyzed chemistry,[11] we have
recently demonstrated that CuI-modified zeolites are excel-
lent stable heterogeneous CuI catalysts for various reac-
tions,[12] including regioselective [3+2] cycloadditions that
offer an alternative to “click chemistry”.[13] Therefore, we
wondered if such principles could be applied to other metal-
catalyzed reactions, such as the Mukaiyama aldol reaction
(Scheme 1). Herein, we show that ScIII-modified zeolites are
indeed very effective catalysts for the Mukaiyama aldol re-
action and could, therefore, be a green alternative to other
conditions.
[a] A. Olmos, Prof. Dr. J. Sommer
Laboratoire de Physicochimie des Hydrocarbures
Institut de Chimie de Strasbourg
4 rue Blaise Pascal
67000 Strasbourg (France)
E-mail: sommer@chimie.u-strasbg.fr
Scheme 1. Mukaiyama aldol reaction.
[b] A. Alix, Prof. Dr. P. Pale
Laboratoire de Synthꢁse et Rꢂactivitꢂ Organiques
Institut de Chimie de Strasbourg
4 rue Blaise Pascal
67000 Strasbourg (France)
E-mail: ppale@chimie.u-strasbg.fr
Results and Discussion
Catalyst setup: Zeolites are solids that are mainly character-
ized by their topologies (cage or channel type), pore size
(typically 6–8 ꢀ), and acidity (correlated to their Si/Al
ratio). USY-type zeolites have internal frameworks with
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200901647.
Chem. Eur. J. 2009, 15, 11229 – 11234
ꢃ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
11229

large cages and various active sites, which allow large mole-
cules to react within the zeolite. Of the zeolite family, un-
modified and metalated USY zeolites often proved to be
the best catalysts for organic synthesis applications.[10,12–13]
Therefore, we prepared scandiumACHTUNTRGNEUNG(III)-modified USY and
then explored its catalytic properties in Mukaiyama aldol re-
action.
The incorporation of scandium into zeolite frameworks
proved critical to get active catalysts. Reports dealing with
the preparation of scandium-containing zeolites were ex-
tremely limited and only ion exchange[14] and isomorphous
substitution sol-gel method[15] have been reported. Slow
vapor diffusion, which proved efficient for doping zeolites
with CuI ions,[12–13] was the best way to produce active cata-
lysts. However, inorganic scandium derivatives exhibit high
melting and boiling points and only a few of them melt
below 5008C. Interestingly, scandium triflate, known to pro-
mote Mukaiyama aldolization, was among those salts. Heat-
ing scandium triflate under a nitrogen flow over H–USY
zeolite at 4508C for two days gave a grey solid that proved
effective as a catalyst. From the amount of scandium triflate
displaced, the loading can be estimated at 10%.[16]
In the absence of catalyst or in the presence of native H–
USY, no reaction took place at room temperature in any of
the solvents used (Table 1, entries 1–2). Rewardingly, the
ScIII-modified USY gave the expected adduct at room tem-
perature as a mixture of stereoisomers. However, the reac-
tion proved to be highly dependent on the nature of the sol-
vent used (Table 1, entries 4–9). Nonpolar and aromatic sol-
vents, such as toluene, did not allow the reaction to proceed
and the starting materials were recovered (Table 1, entry 4).
In sharp contrast, complete conversion was achieved in non-
polar hexane and the adduct was quantitatively obtained as
a mixture of diastereoisomers (Table 1, entry 5). In a slightly
more polar solvent, dichloromethane, the same excellent
conversion and yield were obtained (Table 1, entry 6). The
latter conditions were thus more efficient than the classical
scandium-catalyzed version in homogeneous conditions
(Table 1, entries 5 and 6 vs. 3). More polar solvents proved
deleterious, slowing down the reaction so that almost no
conversion could be detected (Table 1, entries 7–9).
Interestingly, the trimethylsilyl group was always still pres-
ent in the recovered starting materials after experiments
with no or poor conversion, except when performed in
water (Table 1, entry 9). The trimethylsilyl group was also
present in the adduct, revealing a complete silyl transfer
from the starting enol ether to the aldol hydroxyl group.
Because the Mukaiyama aldol reaction is usually per-
formed at very low temperatures (Table 1, entry 3), we in-
vestigated the role that temperature plays in the zeolite-cat-
alyzed version. The reaction with ScIII–USY as catalyst was
still very efficient at low temperatures down to ꢀ208C
(Table 1, entry 11 vs. 6). Below this temperature, the reac-
tion was slower and no conversion could be detected at or
below ꢀ608C (Table 1, entries 12 and 13). On the other
hand, heating proved deleterious, inducing some degrada-
tion and thus a lower yield (Table 1, entry 10 vs. 6).
Condition screening: With the ScIII–USY zeolite in hand, we
then looked at its reactivity. The classical condensation be-
tween the trimethylsilyl enol ether (derived from cyclohexa-
none 1a) and benzaldehyde 2a was used to explore the effi-
ciency of the ScIII-modified USY (Table 1).
Table 1. Screening of conditions for the aldol condensation between 1-
trimethylsilyloxycyclohexene 1a and benzaldehyde 2a.[a]
As expected, the catalyst loading played a key role in the
reaction kinetics. In dichloromethane at room temperature,
the reaction was complete in 4 h in the presence of only 1%
catalyst and in only 30 min with twice this amount of cata-
lyst (Table 1, entry 14 vs. 6). Further increase the catalyst
loading increased the reaction rate; the reaction was almost
instantaneous with 10% of catalyst (Table 1, entry 16 vs. 15
vs. 14).
Entry
Catalyst
(loading [%])
Solvent
T [8C]
t [h]
Yield[b] [%]
ACHTUNGTRENNUNG
[c]
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
none
H–USY
CH2Cl2
CH2Cl2
CH2Cl2
PhMe
20
20
26
22
2
16
0.5
4
16
15
15
2
0[d]
0[d]
[c]
Sc
N
ꢀ78
81[e]
ScIII–USY
ScIII–USY
ScIII–USY
ScIII–USY
ScIII–USY
ScIII–USY
ScIII–USY
ScIII–USY
ScIII–USY
ScIII–USY
ScIII–USY (2)
ScIII–USY (5)
ScIII–USY (10)
20
0[d]
hexane
CH2Cl2
THF
DMF
H2O
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
20
20
20
20
20
50
ꢀ20
ꢀ42
ꢀ60
20
100
100
trace[d]
trace[d]
trace[d,f]
61[f]
Catalyst leaching and recycling: To be sure that the actual
catalyst is the scandium loaded in the zeolite and not scandi-
um ions leaching from the zeolite into solution, we per-
formed a series of experiments in which ScIII–USY was
stirred in dichloromethane for up to 72 h and then filtered.
1-Trimethylsilyloxycyclohexene 1a and benzaldehyde 2a
were added to the resulting solutions. After 18 h, the only
observed transformation was a slow desilylation of 1a.
These experiments revealed that no ScIII leaching occurred
in this ScIII–USY-catalyzed aldolization.
4
90
24
24
0.5
0.1
45[d]
trace[d]
100
20
20
100
100
0.01
[a] Reaction conditions: 1a (1.2 equiv), 2a (1.0 equiv), concentration
0.2m, catalyst (1 mol%, i.e., ꢁ1 mol% of ScIII[16]). [b] Yields of isolated
pure product 3a (as a mixture of stereoisomers) after complete conver-
sion, unless otherwise stated. [c] Other solvents were also examined with-
out success. [d] The starting materials were recovered. [e] The desilylated
aldol product was obtained. [f] The desilylated cyclohexanone was
formed.
We also looked at recycling this catalyst. The ScIII–USY
used in our model reaction under the best conditions was re-
moved by filtration after reaction completion, dried, and
11230
www.chemeurj.org
ꢃ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 11229 – 11234

ScIII-Doped Zeolites
FULL PAPER
Table 2. Scope of silyl enol ethers in the ScIII–USY-catalyzed Mukaiyama
aldol reaction.[a]
used again in a next reaction under the same conditions.
This cycle could be repeated three times without any modifi-
cation in reaction rate and yield (100% within 20 min start-
ing with 5% catalyst loading). However, the fourth cycle led
to a net decrease in yield although it is still reasonable
(67%; see Figure 1 for details).
Entry Silyl enol ether
Product
syn/anti Yield[b] [%]
ratio
1
33:66
33:66
100
100
2[c]
3
4
56:44
46:54
100
100
Figure 1. ScIII–USY recycling and reuse for the aldol condensation be-
tween 1a and 2a.
5
21:79
23:77
56[d]
52[d]
6[e]
7
–
–
46[d]
Scope and limitations: With the catalyst and conditions in
hand, we screened different silyl enol ethers and aldehydes
to briefly explore the scope of this reaction (Tables 2 and 3,
respectively).
8
trace[d]
100
9[c]
As mentioned above, trimethylsilyl enol ether 1a, quanti-
tatively derived from cyclohexanone, reacted with benzalde-
hyde 2a at room temperature in the presence of ScIII-modi-
fied USY as the catalyst. A mixture of diastereoisomers was
produced in which the major isomer was anti. The diastereo-
selectivity remained unchanged with different catalyst load-
ing (Table 2, entry 1 vs. 2).
10
trace[d]
100
11[c]
–
–
12
trace[d]
100
13[c]
The role of the silyl enol ether structure was then ex-
plored (Table 2). Decreasing or increasing the ring size did
not change the reactivity; quantitative reactions were ach-
ieved in both cases under the same conditions. However, the
diastereoselectivity was lower and close to 1:1 in these cases
(Table 2, entries 3, 4 vs. 1). In contrast, increasing the size of
the silyl moiety improved the diastereoselectivity, as with
tert-butyldimethylsilyl enol ether 1d derived from cyclohex-
anone (Table 2, entry 5 vs. 1) but this modification lowered
product yield by slowing down the reaction (Table 2, entry 5
vs. 1). As anticipated, this kinetic effect was more important
at lower temperatures, but only slightly, and unfortunately
without the expected beneficial effect on diastereoselectivity
(Table 2, entry 6 vs. 5). A similar influence of bulkiness was
observed with trimethylsilyl enol ether 1e derived from tert-
butylmethylketone. In this case, the aldolization reaction
proceeded, but only half of the starting materials were con-
verted under the same conditions and the adduct was isolat-
ed in around 50% yield (Table 2, entry 7).
[a] Reaction conditions: 1 (1.2 equiv), 2 (1.0 equiv), concentration 0.2m,
ScIII–USY (1 mol% unless otherwise stated),[16] dichloromethane, 15 h,
RT. [b] Yields of isolated pure product 3. [c] Conditions: ScIII–USY
(10 mol%), 15 min, RT. [d] Recovery of the starting materials. [e] Per-
formed at ꢀ788C.
steric and electronic effects, analogs of 1 f with electron-
withdrawing or -donating groups at the para position, that is,
compounds 1g and 1h, were submitted to the same condi-
tions. At low catalyst loading, almost no reaction occurred
(Table 2, entries 10, 12) whereas at high catalyst loading
rapid quantitative transformations were obtained regardless
of the substitutent (Table 2, entries 11, 13).
The effect of the aldehyde structure was also explored
(Table 3). Electronic effects were examined with a series of
benzaldehydes substituted by either electron-withdrawing or
-donating groups (Table 3, entries 2–4). At high catalyst
loading, only slight reactivity differences were observed and
the diastereoselectivity remained almost identical. Fluoro
derivative 2b reacted as fast as benzaldehyde 2a in a very
efficient reaction (Table 3, entry 2 vs. 1). The more electron-
withdrawing nitro substituent in 2c induced a slightly slower
reaction and thus a slightly lower yield in the same reaction
time (Table 3, entry 3 vs. 1). Electron-rich para-methoxyben-
zaldehyde 2d gave the aldol product almost as fast and effi-
ciently as 2a (Table 3, entry 4 vs. 1).
The same steric effect could also be operative in the reac-
tion of trimethylsilyl enol ether 1 f, derived from acetophe-
none. Under the same conditions and reaction time, this
silyl enol ether remained almost completely unreacted and
the adduct could barely be detected (Table 2, entry 8). How-
ever, with a higher catalyst loading, the reaction proceeded
within 15 min (Table 2, entry 9). To distinguish between
Chem. Eur. J. 2009, 15, 11229 – 11234
ꢃ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
11231

P. Pale, J. Sommer et al.
Table 3. Scope of aldehydes in the ScIII–USY-catalyzed Mukaiyama aldol
reaction.[a]
Regardless of the conditions, the catalyst loading, and
most of the reagents, syn/anti mixtures of diastereoisomers
were always produced, usually in a 30:70 ratio in favor of
the anti isomer (see Tables 2 and 3). This is in sharp contrast
to conventional Mukaiyama conditions[2] and especially lan-
thanide-catalyzed versions,[18] which usually give the oppo-
site ratio. The results obtained with ScIII–USY as the catalyst
seem closer to those reported for certain metals[19] or bis-
muth[20] catalysts in terms of diastereoselectivity and also re-
gioselectivity. Conventional Mukaiyama conditions and cata-
lysts give 1,4-adducts with conjugated aldehydes, while the
ScIII–USY-catalyst still give aldols; that is, 1,2-addition prod-
ucts.[17]
Entry Silyl enol
ether
Aldehyde
Product
syn/
anti
Yield[b]
[%]
ratio
1[c]
33:66 100
35:65 100
35:65 75
36:64 92
2[c]
1a
1a
These comparisons suggested a concerted mechanism with
a Zimmerman–Traxler-type transition state rather than an
open (Noyori) mechanism (Scheme 2, route b vs. a). Howev-
3[c]
4[c]
1a
1a
5[c]
48:52 72[d]
66:33 88[d]
6
7
2e
1a
–
–
trace[e]
Scheme 2. Proposed mechanisms for the ScIII–USY-catalyzed Mukaiyama
aldol reaction.
[a] Reaction conditions: 1 (1.2 equiv), 2 (1.0 equiv), concentration 0.2m,
ScIII–USY j(1 mol% unless otherwise stated),[16] dichloromethane, 15 h,
RT. [b] Yields of isolated pure product 3. [c] Conditions: ScIII–USY
(10 mol%), 15 min, RT. [d] Only the 1,4-adduct was formed and isolated;
polar products were also formed. [e] Mainly recovery of the starting ma-
terials.
er, as usual in such reactions[1–2,19] it is unclear whether
transmetalation occurred or if a siliconate species was in-
volved. The fact that a complete silyl transfer from the start-
ing enol ether to the aldol product occurred, contrary to
classical Mukaiyama conditions,[1–2] also suggested a concert-
ed cyclic transition state and an siliconate intermediate
(Scheme 2, route b).
Such dramatic differences between the zeolite-catalyzed
reaction and the reaction in solution could be due to the ap-
proximately spherical shape of the USY cavities. Its shape
and diameter (7–8 ꢀ) may only allow compact transition
states. This reaction would thus be one of the first, if not the
first, example of size-selective transition-state discrimination
in such chemistry.
From these results and those for the acetophenone deriva-
tives (Table 2, entries 9, 11, 13), it seemed that electronic ef-
fects did not play important role in this ScIII–USY-catalyzed
aldolization.
Conjugated aldehydes, such as cinnamaldehyde, reacted
at a slightly lower rate and some polymeric side-products
were observed. Therefore, yields were slightly lower under
the same conditions (Table 3, entry 5 vs. 1). Interestingly,
only the 1,2-addition product was observed, regardless of
the silyl enol ether used (Table 3, entries 5–6). No trace of
the 1,4-adduct could be detected, as under other Mukaiyama
aldol conditions.[17] With aliphatic aldehydes, only traces of
product seemed to be formed (Table 3, entry 7 vs. 1).
Conclusions
Mechanism: Although not investigated so far, the mecha-
nism of the ScIII–USY-catalyzed Mukaiyama aldol reaction
is an interesting point because of the confinement effect and
other effects due to the peculiar zeolite environment could
induce differences compared with the mechanism commonly
envisaged for this reaction. The present results already give
some insights into this mechanism and indeed revealed
some differences.
In conclusion, we have reported for the first time the use of
ScIII-modified zeolites as catalysts in organic synthesis. With
these catalysts, we have developed a simple and efficient
method for the Mukaiyama aldol reaction. Moreover, this
heterogeneous catalyst can be reused three times without
significant loss of activity. Furthermore, the product re-
mained fully silylated and not partially deprotected as in the
classical methods.
11232
www.chemeurj.org
ꢃ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 11229 – 11234

ScIII-Doped Zeolites
FULL PAPER
31.8, 28.2, 24.8, 24.7, 23.2 ppm; IR (neat): n˜ =3419, 2925, 2855, 1688,
Further work to expand the use of modified zeolites as
catalysts in organic synthesis, especially ScIII-modified zeo-
lites, and to better understand the mechanism of this ScIII–
USY-catalyzed reaction is now underway in our group.
1603, 1493, 1450, 1409, 1364, 1332, 1307, 1252, 1222, 1199, 1157, 1061,
1036, 1024, 1010 cmꢀ1
.
1
Anti isomer: H NMR (300 MHz, CDCl3): d=7.33–7.23 (m, 5H), 4.93 (d,
J=4.9, 1H), 3.34 (brs, 1H), 3.02 (ddd, J=11.3, J=4.9, J=4.2, 1H), 2.26–
2.22 (m, 2H), 2.10–2.01 (m, 1H), 1.94–1.86 (m, 1H), 1.81–1.29 ppm (m,
8H); 13C NMR (75 MHz, CDCl3): d=221.5, 141.9, 128.2, 127.4, 126.2,
74.2, 55.5, 44.3, 28.6, 28.2, 25.0, 25.0, 23.2 ppm; IR (neat): n˜ =3419, 2925,
2855, 1688, 1603, 1493, 1450, 1409, 1364, 1332, 1307, 1252, 1222, 1199,
1157, 1061, 1036, 1024, 1010 cmꢀ1; HRMS (ESI, positive mode): m/z
calcd for C15H20O2Na: 255.1361; found: 255.1396 [M+Na]+.
Experimental Section
General: All starting materials were commercially available (used as re-
ceived) or prepared according to known literature procedures (see the
Supporting Information for details). The reactions were monitored by
thin-layer chromatography carried out on silica plates (silica gel 60 F254,
Merck) by using UV light and phosphomolybdic acid with cerium(IV)
for visualization. Column chromatography was performed on silica gel 60
(0.040–0.063 mm, Merck) by using mixtures of pentane and dichloro-
1-(4-Nitrophenyl)-3-phenyl-3-(trimethylsilyloxy)propan-1-one (3g): The
general procedure was performed with ScIII–USY (80 mg, 0.1 equiv), tri-
methyl(1-(4-nitrophenyl)vinyloxy)silane 1g, and benzaldehyde 2a to give
the expected silylated aldol as a white solid (yield 100%). M.p. 858C;
1H NMR (300 MHz, CDCl3): d=8.29 (d, J=8.9, 1H), 8.11 (d, J=8.9,
1H), 7.45–7.24 (m, 5H), 5.34 (dd, J=9.0, J=3.7, 1H), 3.60 (dd, J=15.4,
J=9.0, 1H), 3.02 (dd, J=15.4, J=3.7, 1H), ꢀ0.05 ppm (s, 9H); 13C NMR
(75 MHz, CDCl3): d=197.3, 150.2, 144.0, 142.0, 129.4, 128.4, 127.6, 125.6,
123.6, 71.8, 50.0, ꢀ0.2 ppm; IR (neat): n˜ =2961, 2905, 1691, 1602, 1521,
1493, 1405, 1369, 1341, 1316, 1298, 1264, 1248, 1192, 1091, 1045,
1026 cmꢀ1; HRMS (ESI, positive mode): m/z calcd for C18H24NO4SiNa:
366.1138; found: 366.1940 [M+H]+.
ACHTUNGTRENNUNGmethane as the eluent. Evaporation of solvents was conducted under re-
duced pressure at temperatures of less than 308C. Melting points (m.p.)
were measured in open capillary tubes and are uncorrected. IR spectra
were recorded by using a Bruker ALFA FTIR spectrometer (neat) and
values are reported in cmꢀ1 1H and 13C NMR spectra were recorded by
.
using a Bruker Avance 300 spectrometer at 300 and 75 MHz, respectively.
Chemical shifts (d) and coupling constants J are given in ppm and Hz, re-
spectively. Chemical shifts (d) are reported relative to residual solvent as
an internal standard ([D1]chloroform: d=7.26 and 77.0 ppm for 1H and
13C, respectively). Carbon multiplicities were determined by DEPT135
experiments. Electrospray low/high-resolution mass spectra (ESIMS)
were obtained from the mass spectrometry department of the Service
Commun d’Analyses, Institut de Chimie, Strasbourg.
3-Hydroxy-1-(4-nitrophenyl)-3-phenylpropan-1-one (3g aldol desilylated
adduct): After acidic treatment of 3g, the expected aldol was isolated as
a yellowish solid (yield 100%). M.p. 908C; 1H NMR (300 MHz, CDCl3):
d=8.28 (d, J=8.9, 1H), 8.08 (d, J=8.9, 1H), 7.47–7.26 (m, 5H), 5.35
(dd, J=9.0, J=3.1, 1H), 3.48 (dd, J=17.6, J=9.0, 1H), 3.32 (dd, J=17.6,
J=3.1, 1H), 3.05 ppm (brs, 1H); 13C NMR (75 MHz, CDCl3): d=198.1,
150.4, 142.6, 140.9, 129.1, 128.6, 127.9, 125.6, 123.8, 69.8, 48.0 ppm; IR
(neat): n˜ =3466, 1679, 1600, 1586, 1573, 1512, 1460, 1418, 1398, 1370,
1341, 1318, 1296, 1270, 1202, 1184, 1159, 1112, 1086, 1059, 1015,
1003 cmꢀ1; HRMS (ESI, positive mode): m/z calcd for C15H13NO4Li:
278.1005; found: 278.0929 [M+H]+.
General procedure for the ScIII–zeolite catalyzed Mukaiyama aldol reac-
tion: Aldehyde
2 (0.4 mmol, 1 equiv) and then silyl enol ether 1
(0.48 mmol, 1.2 equiv) were added to a suspension of ScIII–USY (8 mg,
0.01 equiv) in dichloromethane (2.5 mL). After stirring at RT (unless oth-
erwise stated) for 15 h, the mixture was taken up in dichloromethane
(10 mL) then filtered on nylon membranes (0.20 mm). Solvent evapora-
tion provided the resulting crude product usually at >95% purity as
judged by NMR spectroscopy. The syn/anti ratio was determined from
1H NMR spectra of the crude product. Column chromatography was per-
formed to separate the syn/anti isomers by using pentane/dichloro-
Detailed experimental procedures and NMR spectra for all new com-
pounds are given in the Supporting Information.
ACHTUNGTRENNUNGmethane (1:1) as the eluent. All desilylated aldol products, except 3c and
3g, are known compounds and exhibited spectroscopic data consistent
Acknowledgements
with the previously reported ones in the literature.
2-(Phenyl(trimethylsilyloxy)methyl)cyclooctanone (3c): The general pro-
cedure was performed by using cyclooctenyloxytrimethylsilane 1c and
benzaldehyde 2a to give the expected silylated aldol as a colorless oil
(yield 100%).
The authors thank the CNRS, the French Ministry of Research, and the
Loker Hydrocarbon Institute, Los Angeles, for financial support. A.O.
thanks the University of Valencia, Spain, for an exchange program and
A.A. thanks the ANRS for a PhD fellowship.
Syn isomer: 1H NMR (300 MHz, CDCl3): d=7.33–7.23 (m, 5H), 4.78 (d,
J=9.5, 1H), 3.09 (ddd, J=12.3, J=9.5, J=3.3, 1H), 2.48–2.32 (m, 2H),
2.22–2.07 (m, 1H), 1.84–1.29 (m, 9H), ꢀ0.11 ppm (s, 9H); 13C NMR
(75 MHz, CDCl3): d=218.9, 142.9, 128.1, 127.6, 127.1, 77.6, 57.5, 45.4,
31.0, 28.5, 24.8, 22.8, ꢀ0.2 ppm; IR (neat): n˜ =2972, 2926, 1698, 1453,
[1] Modern Aldol Reaction (Ed.: R. Mahrwald), Wiley-VCH, Wein-
heim, 2004.
[2] For pioneering reports, see: a) T. Mukaiyama, K. Banno, K. Narasa-
ka, J. Am. Chem. Soc. 1974, 96, 7503–7509; b) For recent reviews,
see: c) T. Mukaiyama, Angew. Chem. 2004, 116, 5708–5733; Angew.
Chem. Int. Ed. 2004, 43, 5590–5614; d) C. Palomo, M. Oiarbide,
J. M. Garcꢄa, Chem. Soc. Rev. 2004, 33, 65–75; e) J. L. Vicario, D.
Badia, L. Carrillo, E. Reyes, J. Etxebarria, Curr. Org. Chem. 2005, 9,
219–235; f) B. Schetter, R. Mahrwald, Angew. Chem. 2006, 118,
7668–7687; Angew. Chem. Int. Ed. 2006, 45, 7506–7525.
[3] E. M. Carreira in Comprehensive Asymmetric Catalysis, Vol 3 (Eds.:
E. N. Jacobsen, A. Pfaltz, Y. Yamamoto), Springer, Berlin, 1999,
pp. 997–1066.
[4] R. A. Flowers, X. Xu, C. Timmons, G. Li, Eur. J. Org. Chem. 2004,
2988–2990.
[5] S. Itsuno, S. Arima, N. Haraguchi, Tetrahedron 2005, 61, 12074–
12080.
[6] M. Benaglia, M. Cinquini, F. Cozzi, G. Celentano, Org. Biomol.
Chem. 2004, 2, 3401–3407.
1375, 1329, 1251, 1201, 1158, 1085, 1046 cmꢀ1
.
1
Anti isomer: H NMR (300 MHz, CDCl3): d=7.26–7.16 (m, 5H), 4.76 (d,
J=8.4 Hz), 3.00 (ddd, J=11.3, J=8.4, J=3.1, 1H), 2.18–2.08 (m, 2H),
2.01–1.83 (m, 1H), 1.74–1.18 (m, 9H), ꢀ0.03 ppm (s, 9H); 13C NMR
(75 MHz, CDCl3): d=219.3, 143.4, 128.0, 127.2, 126.7, 76.6, 58.9, 44.6,
30.6, 28.2, 25.5, 24.9, 23.6, 0.1 ppm; IR (neat): n˜ =2926, 2856, 1704, 1493,
1453, 1412, 1364, 1328, 1313, 1249, 1199, 1157, 1078, 1055, 1027 cmꢀ1
;
HRMS (ESI, positive mode): m/z calcd for C18H28O2SiLi: 311.2019;
found: 311.1999 [M+Li]+.
2-(HydroxyACHTUNGTRENNUNG(phenyl)methyl)cyclooctanone (3c desilylated aldol adduct):
After acidic treatment of 3c, the expected aldol was isolated as a color-
less oil (yield 100%).
Syn isomer: 1H NMR (300 MHz, CDCl3): d=7.37–7.24 (m, 5H), 4.82 (d,
J=7.5, 1H), 3.21 (bs, 1H), 3.12 (ddd, J=11.9, J=7.5, J=4.4, 1H), 2.33–
2.29 (m, 2H), 2.16–2.01 (m, 1H), 1.83–1.22 ppm (m, 9H); 13C NMR
(75 MHz, CDCl3): d=221.2, 142.5, 128.4, 127.7, 126.4, 76.4, 55.7, 45.0,
Chem. Eur. J. 2009, 15, 11229 – 11234
ꢃ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
11233

P. Pale, J. Sommer et al.
[7] a) S. Kobayashi, I. Hachiya, M. Yasuda, Tetrahedron Lett. 1996, 37,
5569–5572; b) S. Iimura, K. Manabe, S. Kobayashi, Tetrahedron
2004, 60, 7673–7678.
[8] a) S. Kobayashi, S. Nagayama, J. Am. Chem. Soc. 1998, 120, 2985–
2986; b) M. Takeuchi, R. Akiyama, S. Kobayashi, J. Am. Chem. Soc.
2005, 127, 13096–13097.
Tetrahedron 2008, 64, 8922–8929; d) M. Keller, A. Sani-Souna-Sido,
P. Pale, J. Sommer, Chem. Eur. J. 2009, 15, 2810–2817.
[14] a) G. V. Tsitsishvili, A. Y. Krupennikova, M. N. Burdzhanadze,
Dokl. Akad. Nauk SSSR 1978, 240, 648–650; b) F. H. Khalil, A. K.
Aboul-Gheit, M. M. Selim, Revis. del Inst. Mexicano del Petroleo
1974, 6, 56–66 ( CA 82:113844).
[15] a) C. T. Brigden, D. Thompsett, C. D. Williams, Dalton Trans. 2004,
2829–2830; b) C. T. Briden, C. D. Williams, D. Apperley, Inorg.
Mater. 2007, 43, 758–769.
[9] B. C. Ranu, R. Chakraborthy, Tetrahedron 1993, 49, 5333–5338.
[10] For recent examples, see: a) A. Sani-Souna-Sido, S. Chassaing, M.
Kumarraja, P. Pale, J. Sommer, Tetrahedron Lett. 2007, 48, 5911–
5914; b) S. Chassaing, M. Kumarraja, P. Pale, J. Sommer, Org. Lett.
2007, 9, 3889–3892; c) A. Sani-Souna-Sido, S. Chassaing, P. Pale, J.
Sommer, Appl. Catal. A 2008, 336, 101–108.
[11] For recent examples, see: a) K. Y. Koltunov, S. Walspurger, J.
Sommer, Chem. Commun. 2004, 1754–1755; b) M. Haouas, S. Wal-
spurger, F. Taulelle, J. Sommer, J. Am. Chem. Soc. 2004, 126, 599–
606; c) K. Y. Koltunov, S. Walspurger, J. Sommer, Tetrahedron Lett.
2005, 46, 8391–8394; d) A. Vasilyev, S. Walspurger, P. Pale, J.
Sommer, Tetrahedron 2005, 61, 3559–3564; e) S. Walspurger, Y. Sun,
A. Sani-Souna-Sido, J. Sommer, J. Phys. Chem. B 2006, 110, 18368–
18373; f) A. Vasilyev, S. Walspurger, S. Chassaing, P. Pale, J.
Sommer, Eur. J. Org. Chem. 2007, 5740–5748.
[12] a) M. K. Patil, M. Keller, M. B. Reddy, P. Pale, J. Sommer, Eur. J.
Org. Chem. 2008, 4440–4445; b) P. Kuhn, A. Alix, M. Kumarraja, B.
Louis, P. Pale, J. Sommer, Eur. J. Org. Chem. 2009, 423–429; c) P.
Kuhn, P. Pale, J. Sommer, B. Louis, J. Phys. Chem. C 2009, 113,
2903–2910.
[13] a) S. Chassaing, M. Kumarraja, A. Sani-Souna-Sido, P. Pale, J.
Sommer, Org. Lett. 2007, 9, 883–886; b) S. Chassaing, A. Sani-
Souna-Sido, A. Alix, M. Kumarraja, P. Pale, J. Sommer, Chem. Eur.
J. 2008, 14, 6713–6721; c) A. Alix, S. Chassaing, P. Pale, J. Sommer,
[16] 1 mol% of catalyst corresponds to around 1 mol% of ScIII species
based on the theoretical number of native acidic sites of the corre-
sponding H–zeolite. For
a recent method of determination of
Brçnsted acid sites on zeolites, see: B. Louis, S. Walspurger, J.
Sommer, Catal. Lett. 2004, 93, 81–84.
[17] a) N. Iwasawa, T. Mukaiyama, Chem. Lett. 1987, 463–466; b) M.
Reetz, D. N. A. Fox, Tetrahedron Lett. 1993, 34, 1119–1122; c) M.
Terada, T. Nakai, K. Mikami, Inorg. Chim. Acta 1994, 222, 377–380;
d) J. Otera, Y. Fujita, S. Fukuzumi, K. Hirai, J.-H. Gu, T. Nakai, Tet-
rahedron Lett. 1995, 36, 95–98.
[18] a) S. Kobayashi, I. Hachiya, Tetrahedron Lett. 1992, 33, 1625–1628;
b) S. Kobayashi, I. Hachiya, H. Ishitani, M. Araki, Synlett 1993,
472–474.
[19] R. Mahrwald, Chem. Rev. 1999, 99, 1095–1120.
[20] a) C. Le Roux, L. Ciliberti, H. Laurent-Robert, A. Laporterie, J.
Dubac, Synlett 1998, 1249–1251; b) C. Le Roux, H. Gaspardllough-
mane, J. Dubac, J. Jaud, P. Vignaux, J. Org. Chem. 1993, 58, 1835–
1839; c) M. Wada, M. Takeichi, T. Matsumoto, Bull. Chem. Soc. Jpn.
1991, 64, 990–994; d) H. Ohki, M. Wada, K. Akiba, Tetrahedron
Lett. 1988, 29, 4719–4722.
Received: June 16, 2009
Published online: September 11, 2009
11234
www.chemeurj.org
ꢃ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 11229 – 11234