Scandium(III)-zeolites as new heterogeneous catalysts for imino-diels-alder reactions

Article abstract of DOI:10.1002/chem.201103624

This study demonstrates the first zeolite-catalyzed synthesis of piperidine derivatives, including peptidomimetics and indoloquinolizidine alkaloids. The approach developed utilizes a highly effective one-pot reaction cascade, through imine formation and

Full text of DOI:10.1002/chem.201103624

DOI: 10.1002/chem.201103624
ScandiumACHTUNGTRENNUNG(III)-Zeolites as New Heterogeneous Catalysts for Imino-Diels–
Alder Reactions
Andrea Olmos, Benoit Louis, and Patrick Pale*[a]
Abstract: This study demonstrates the
first zeolite-catalyzed synthesis of pi-
peridine derivatives, including peptido-
mimetics and indoloquinolizidine alka-
loids. The approach developed utilizes
a highly effective one-pot reaction cas-
cade, through imine formation and
imino-Diels–Alder reactions, promoted
by scandium-loaded zeolites as a heter-
ogeneous catalyst. The methodology
described benefits from very low cata-
lyst loadings (ꢀ5 mol% of ScIII), com-
mercially and readily available starting
materials, and mild reaction conditions.
Furthermore, the ScIII-zeolite catalyst
can be readily reused more than 10
times without any loss in efficiency.
Keywords: heterogeneous catalysis ·
imine · peptidomimetics · piperi-
dine · scandium · zeolites
Introduction
fire-ant venom,[6] or by coniine, from the plant Conium mac-
ulatum, which has a mammalian LD50 of less than 1 mgkgÀ1
and is also the insect-paralyzing component of a pitcher
plant, which is active at the nanogram level.[7]
Furthermore, piperidine probably is the most common
heterocycle occurring in drugs,[8] not only as a scaffold but
also contributing to a wide range of bioactivities, as repre-
sented by, for example, the antidepressant paroxetine, the
anti-parkinsonian agent biperiden, the analgesic fentanyl or
the anaesthetic bupivacaine, or the recently patented anti-
Alzheimer series of compounds (Scheme 2).[9]
Piperidine derivatives: Piperidine heterocycles represent
a large important class of natural products from the alkaloid
family.[1] Piperidine alkaloids are usually found in plants,[2]
and they can exhibit very simple structures (such as con-
iine), up to more complex structures (such as quinolizidine
and indolic alkaloids, for example, lasubine and yohimbine-
type alkaloids), through sugar-like nojirimycine and related
compounds (Scheme 1). They can also be found in other or-
ganisms such as dendrobatid frogs,[3] myrmicine ants, and
other insects,[4] although they can be connected to plant al-
kaloids through ecological relationships (Scheme 1).[5] Piper-
idine alkaloids often have toxic or deterrent activity as illus-
trated by solenopsin A, one of the active components of
Scheme 2. Selected examples of drugs containing the piperidine motif.
Scheme 1. Selected examples of piperidine natural products.
Previous efforts and challenges: With such diversity of struc-
tures and applications, numerous synthetic approaches to-
wards piperidine derivatives have been developed.[10]
Among them, the [4+2] cycloaddition between imines and
some dienes has long been recognized as one of the most
convenient and rapid synthetic methods.[11] Such so-called
imino-Diels–Alder reactions have been reported with vari-
ous soft Lewis acid catalysts, such as zinc,[12] zirconium,[13]
bismuth[14]and lanthanide[15], and more recently niobium
[a] Dr. A. Olmos, Dr. B. Louis, Prof. Dr. P. Pale
Institute of Chemistry, UMR 7177
Universitꢀ de Strasbourg
4 rue B. Pascal, 67000 Strasbourg (France)
E-mail: ppale@unistra.fr
Supporting information for this article (including full experimental
details) is available on the WWW under http://dx.doi.org/10.1002/
chem.201103624.
4894
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 4894 – 4901

FULL PAPER
Scheme 4. Metal-loaded zeolite synthesis.
Scheme 3. Known and possible routes to the piperidine motif.
metallic ions. All these modified zeolites were fully charac-
terized.[22]
salts[16] (Scheme 3, top). Brønsted acids also proved efficient
catalysts for this transformation.[17] Despite the interest in
such reactions, heterogeneous reusable catalysts have so far
never been used for such applications.[18]
With these metal-loaded zeolites as catalysts (5% mol of
metal ion), the behavior of the stable imine 3a, either in
situ or already formed from the corresponding benzalde-
hyde 1a and aniline 2a, was examined in the presence of
the electron-rich Danishefsky diene[23] 4a (Table 1).
Due to our interest in the latter aspect, and based on our
precedents with Cu- and Sc-zeolites,[19,20] we thus wondered
if such metal-loaded zeolites could catalyze imino-Diels–
Alder reactions (Scheme 3, bottom). Zeolites with their
acidic aluminosilicate framework act as Brønsted acids in
various reactions.[21] Depending on the nature of the metal,
metal ions loaded into zeolites could still act as a Lewis
acid. Therefore, metal-loaded zeolites should combine both
types of activation, but as solids, they are thus easy to
handle and recover. However, an aldehyde, an amine, and
a diene must encounter each other within a zeolite pore in
order to react. On the other hand, imine formation and sta-
bility, as well as diene stability, could be a problem in the
presence of zeolites, which are acidic by nature. The fact
that CuI-zeolites could catalyze the formation of propargyl
amines from aldehydes, amines, and terminal alkynes[19d]
strongly suggests the feasibility of metal-zeolite-catalyzed
imino-Diels–Alder reactions.
Table 1. Catalyst and solvent screening.
Entry[a]
Catalyst
Solvent
Time
[h]
Yield of 5a
[%][b]
[c]
1
2
3
4
5
6
7
8
None
H-USY
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
PhMe
CH2Cl2
THF
MeCN
MeOH
MeCN
MeOH
72
24
24
24
20
20
20
16
16
1
0
0–30
traces
0
[c]
[c]
[c]
CuI-USY
CuII-USY
ScIII-USY
ScIII-USY
ScIII-USY
ScIII-USY
ScIII-USY
ScIII-USY
ScIII-USY
0
0
20
48
57
99
57
9
10[d]
11[d]
We report here our results in this area, revealing the first
metal-zeolite-catalyzed cycloaddition of imines with dienes
and its application to the synthesis of peptide mimics and pi-
peridinoindole alkaloid analogues.
2
[a] Reaction performed at room temperature on 0.5 mmol in the pres-
ence of 5 mol% of cat. [b] Yields of isolated and purified product; the
imine accounted for the mass balance. [c] Other solvents have also been
screened, including acetonitrile. [d] Starting from the pre-formed imine
3a.
Results and Discussion
Control experiments showed that no reaction took place
without catalyst (Table 1, entry 1), whereas the native H-
USY exhibited no or a poor catalytic activity depending on
the solvent (Table 1, entry 2). It is interesting to note that
the recovery of the Danishefsky diene was possible for
a short reaction time (<3 h) but not after a longer reaction
time, since it underwent gradual hydrolysis. These results re-
vealed some stability of such dienes despite the acidic zeolit-
ic medium. Both CuI- and CuII-USY were not as effective as
catalysts; almost no transformation was observed at room
temperature resulting in the recovery of the starting materi-
als and the corresponding imine (Table 1, entries 3–4). Nev-
ertheless, with CuI-USY as catalyst, traces of adduct could
be detected after prolonged reaction time (Table 1, entry 3
vs. 4). Rewardingly, ScIII-USY promoted the formation of
the expected and known[12] dehydropiperidinone adduct 5a,
but the transformation efficiency proved very dependent on
the solvent nature (Table 1, entries 5–11). In this reaction,
the more polar the solvent, the better the transformation.
Acetonitrile was the more efficient solvent, especially with
pre-formed imine, quantitatively giving the dehydropiperidi-
Imines, being more or less stable and easily hydrolyzed to
their starting components, that is, aldehyde and amine, are
usually produced and used in situ, despite the release of
water due to their formation, which may be a problem for
the next step. In this work, we compared both routes when
possible, through either a one- or two-pot process, starting
either from aldehyde 1 and amine 2 or from the pre-formed
imine 3.
Catalyst and conditions: The commercially available Ultra-
stable Y (USY) was selected due to its large pores, which
are able to accommodate several molecules together as re-
quired for imine formation and for Diels–Alder type reac-
tions. USY, modified or not, was also the most efficient zeo-
lite for other reactions.[19,20] The commercial ammonium
USY was thus converted to the corresponding copper or
scandium-doped USY in two steps with salt sublimation, as
already reported (Scheme 4a).[19,20] The amount of salts was
calculated to replace 5–10% of the acidic sites of USY, lead-
ing to zeolites loaded with the corresponding amount of
Chem. Eur. J. 2012, 18, 4894 – 4901
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4895

P. Pale et al.
Table 2. Scope of the imine component.
none 5a (Table 1, entry 10). Al-
though it should have affected
the imine formation, methanol
unexpectedly gave reasonable
amounts of adduct, with the use
of either protocol, that is, with
in situ or pre-formed imine
(Table 1, entries 9 and 11).
Scandium ions, despite their
dispersion into zeolites, are thus
clearly available to organic moi-
eties and are able to promote
Entry[a]
Aldehyde
Amine
PhNH2
T
t
Yield
[%][b]
Adduct
[8C]
[h]
1
1a
1a
2a
2a
20
20
18
1
48
99
5a
5a
2[c]
nBuNH2
2b
imino-Diels–Alder
Even at low loading, ScIII-USY
zeolite performs as very
reaction.
3
1a
60
18
traces
–
a
active catalyst for the imino-
Diels–Alder reaction.
4[c]
1a
2b
tBuNH2
2c
20
18
34
5b
5
1a
1a
60
60
18
18
0
0
–
6[c]
2c
Catalyst recycling: The recov-
ery and recyclability of this new
heterogeneous
catalyst
for
imino-Diels–Alder reaction was
evaluated with the same reac-
tion. After the reaction, the
ScIII-USY catalyst was recov-
ered through filtration over
a nylon membrane. After wash-
ing and drying under vacuum at
room temperature for 30 mi-
nutes, the catalyst was re-engag-
ed in the next run. This proce-
dure was applied at each run,
and the catalyst was still highly
active for up to ten times (at
least),[24] without a noticeable
decrease in activity (Figure 1).
7
1a
1a
2d
2d
60
20
18
18
42
48
5c
5c
8[c]
PhNH2
9
1b
1b
2a
2a
60
20
18
5
47
51
5d
5d
10[c]
11
1c
1c
2a
2a
20
20
18
7
12
85
5e
5e
12[c]
À
EtOOC CHO
13
1d
2a
20
18
82
5 f
[a] Reaction performed on a 0.5 mmol scale in the presence of 5% of cat. [b] Yields of isolated and purified
product; the imine accounted for the mass balance. [c] Starting from the pre-formed imine 3a.
Scope: To look at the potential
and possible limitations of this
new heterogeneous imino-Diels–Alder reaction, various
commercially available aldehydes and amines were submit-
ted to the Danishefsky diene 4a under the conditions set
above (Table 2). As shown above, aryl amines (such as ani-
line), and aryl carbaldehyde (such as benzaldehyde), readily
reacted in the presence of catalytic amounts of Sc-zeolite
and led to the corresponding adduct in modest to good
yields (Table 2, e.g., entry 1), but quantitatively if the corre-
sponding imine was pre-formed (entry 2). In contrast, alkyl-
amines gave variable results, depending on the amine struc-
ture (Table 2, entries 3–10). Simple primary alkylamines
almost did not react, even upon heating (Table 2, entry 3 vs.
1), but pre-forming the imine again helped although the
yield of the isolated corresponding adduct remained modest
(entry 4 vs. 3). The less basic benzylamine proved more re-
active than primary alkylamines, and the one-pot reaction
already gave reasonable yield of the expected adduct,
whereas pre-forming the imine only slightly increased yields
(Table 2, entry 7 vs. 3, entry 8 vs. 4). The bulky tert-butyl-
AHCTUNGTREGaNNNU mine did not react regardless of the protocol used (Table 2,
entries 5,6).
Although enamine formation could have been a problem,
isobutyraldehyde readily reacted with aniline at room tem-
perature; however, a higher reaction rate was achieved
Figure 1. Recycling: Efficiency of recycled ScIII-USY catalyst in the imino
Diels–Alder reaction.
4896
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 4894 – 4901

ScandiumACHTUNGTRENNUNG(III)-Zeolites as New Heterogeneous Catalysts
FULL PAPER
upon heating. The expected adduct was isolated
with reasonable yields, very similar to those ob-
served with benzaldehyde (Table 2, entry 9 vs. 1).
Pre-forming the imine only slightly improved the cy-
cloaddition efficiency (Table 2, entry 10 vs. 9).
Table 3. Scope of the diene component.
Conjugated aldehydes proved to be also reactive.
Cinnamaldehyde gave the expected adduct, albeit
the one-pot process was rather slow, even upon
heating. Under these reaction conditions, some de-
composition occurred and the isolated product yield
was thus lower (Table 2, entry 11). With the imine 2[c]
pre-formed, the reaction was faster, even at room
temperature, giving high yield of the corresponding
Entry[a]
Diene
t
Yield
[%][b]
Adduct
[h]
1
R=H
R=CH3
48
48
0
0
–
–
adduct upon isolation (Table 2, entry 12). Glyoxylate
3
4a
4a
18
2
48
99
5a
5a
4[c]
derivatives proved to be the most reactive among
the screened aldehydes. With aniline at room tem-
perature, it gave the corresponding adduct in a high
yield (Table 2, entry 13).
5
4b
4b
18
2
64
70
6b
7b
In another set of experiments, the nature and the
role of the diene were investigated (Table 3). Simple
dienes such as methylbutadiene (isoprene) or 2,3-di-
methylbutadiene did not react, whereas the more
electron-rich Danishefsky diene 4a readily reacted
at room temperature and the Diels–Alder type
adduct 5a was isolated in good to quantitative yields
depending on the conditions (Table 3, entry 1 and 2
vs. 3 and 4). The electronic density of the diene may
not be the sole factor in this dramatic difference.
Diene S-cis conformation is obviously required for
Diels–Alder cycloaddition, which is less present in
(di)methylbutadiene than in the Danishefsky
diene.[25]
6[c]
7
4c
4c
24
18
0
59
7c
7c
8[c]
9
4d
4d
24
18
15
89
7d
7d
10[c]
[a] Reaction performed at room temperature on a 0.5 mmol scale in the presence of
5% of cat. [b] Yields of isolated and purified product; the imine accounted for the
To check the influence of diene conformation, var-
ious Brassard dienes[26] and the locked methylene- mass balance. [c] Starting from the pre-formed imine 3a.
dioxine diene[27] were prepared and engaged in the
ScIII-USY catalyzed reaction with N-phenyl phenylimine 3a
or its precursors 1a and 2a (Table 3, entries 5–10).
These results thus confirmed that an aldol pathway[28]
always competes with a concerted pathway,[29] even within
zeolites. A common siloxonium pathway cannot neverthe-
less be ruled out.[30] The results also confirmed the role of
diene conformation on the course of the reaction, despite
the constraints imposed by zeolite cavities.
Under the conditions used for the Danishefsky diene,
Brassard dienes 4b–c readily reacted, but two adducts were
produced and none of them proved to be cyclic as expected
from an imino-Diels–Alder type reaction (Table 3, en-
tries 3,4). Spectroscopic evidences revealed the formation of
a- and g-adducts. Starting from 1-ethoxy-1-silyloxybutadiene
4b, the g product was the major adduct and the yields were
good with, as before, a slight increase with the pre-formed
imine (Table 3, entries 5,6). With the more electron-rich 1-
ethoxy-1-silyloxy-3-methoxybutadiene 4c, the reaction only
occurred when the imine was pre-formed. Only the g adduct
was observed but it was produced as a E/Z mixture
(Table 3, entries 7,8). These results were reminiscent of the
Mukaiyama-aldol reaction. As suspected from the results
with Brassard dienes, the diene 4d, locked in S-trans confor-
mation, gave a single adduct, resulting from an imino-Mu-
kaiyama-aldol reaction. The later was obtained in high yield
from the pre-formed imine but in low yield from its starting
precursors (Table 3, entries 9,10).
It is worth noting that these Mukaiyama-type adducts
could nevertheless be converted to the corresponding piperi-
dinones in excellent yields (Scheme 5).[31]
Synthetic application: To highlight the usefulness of this
Scheme 5. Cyclization of the Mukaiyama-type adducts to piperidine de-
rivatives.
Chem. Eur. J. 2012, 18, 4894 – 4901
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4897

P. Pale et al.
zeo-click catalysis, we applied the present ScIII-USY promot-
ed formation of piperidinones to the synthesis of peptidomi-
metics and of indolic piperidine alkaloids.
Peptide analogues and modified peptides are of increasing
interest as mimics of biologically active peptides and thus as
potent drugs.[32] For example, opioid derivatives based on pi-
peridinone as a scaffold have been described and prepared
through solid-phase synthesis (Scheme 6).[18c] We thus won-
dered if the ScIII-zeolite could replace resins, thereby avoid-
ing side reactions during cleavage and low loading.
Scheme 8. Indoloquinolizidine synthesis.
(Table 4). Heating was nevertheless necessary to achieve
high yields within reasonable reaction times and again, pre-
forming the imine clearly improved the efficiency of the re-
action (Table 4, entries 2 vs. 1, 4 vs. 3, 6 vs. 5, 8 vs. 7, 10 vs.
9, and 12 vs. 11).
Scheme 6. Opioid peptide mimics based on piperidinone.
Phenylalanine gave the corresponding adduct in good
yields when the imine was pre-formed, whereas no reaction
was observed from the starting components (Table 4, entry 2
vs. 1). Tyrosine, as its methyl ether, behaved similarly
(Table 4, entry 4 vs. 3) but proved slightly less reactive and
the yields were slightly lower (entry 4 vs. 2). Interestingly,
tryptophane esters readily reacted with the Danishefsky
diene without a protecting group at the N-indole moiety.
However, the corresponding adduct was obtained with
modest yields with an allyl as the carboxylic acid protecting
group but in quantitative yields with a methyl group
(Table 4, entries 7 and 8 vs. 5 and 6). The allyl group in-
duced the formation of unidentified side-products, whereas
the methyl group led to a very clean reaction. Surprisingly,
protection at the N-indole with carbamates induced a de-
crease in reactivity, leading to modest but reasonable yields
(Table 4, entries 9–12 vs. 7 and 8).
On the other hand, indoloquinolizidine alkaloids such as
reserpine, yohimbine hirsutine, and corynantheidol
(Scheme 7) exhibit various and interesting properties,
Scheme 7. Some indoloquinolizidine alkaloids.
In contrast to solution or resin phase reactions,[12–18] the
adducts were always obtained here as the same mixture of
diastereoisomers, even when starting from a single imine.
Whatever the starting amino acid, the ratios were indeed
very similar, with a 3 to 1 mixture, whereas the solution or
resin phase synthesis gave ratios from 1 to 1 to 91 to 9.
NMR studies of the adducts (and especially of the cyclized
products (see below) as well as comparison with the few re-
lated known compounds),[34] allowed the assignment of the
diastereoisomer stereochemistry as shown in Table 4.
These facts led to interesting mechanistic considerations.
Whatever the exact reaction mechanism,[28–30] the fact that
the stereodivergent methoxy group in the primary adduct is
finally eliminated does not allow for distinguishing between
endo or exo pathways, if any. The stereodifference is thus
limited to the induction brought by the amino acid stereo-
center. Interestingly, an outside Houk-type model[36] can ra-
tionalize the stereochemical outcome; interaction of the
ester group either with the zeolite or scandium (or both)
probably favors the outside arrangement, whereas the spher-
ical shape of the zeolite used (USY) favors a more or less
mostly due to their interaction with adrenoceptors in the
nervous system.[33] Some derivatives have also been devel-
oped as apoptosis inducers[34] or as inhibitors for protein ty-
rosine phosphatase B.[35] Such compounds could be obtained
in a three-step sequence, with a Diels–Alder type reaction
on imine derived from tryptophane or tryptamine, followed
by intramolecular cyclization (Scheme 8). The latter is usual-
ly performed in acidic media, and the whole sequence can
be performed on resin.[35] Due to the dual nature of our het-
erogeneous catalyst (containing ScIII ions but with protons
still present), we expected that further cyclization could also
occur, directly leading to indoloquinolizidinones in a one-
pot operation and therefore ScIII-zeolites could replace
resins and their limitations.
We thus applied the ScIII-zeolite conditions detailed above
first to phenylalanine derivatives and then to tryptophane
(protected or unprotected). Regardless of the selected start-
ing amino acid, the reaction proceeded well but stopped at
the first stage, and gave with the Danishefsky diene the ex-
pected piperidinone adducts in good to high yields
4898
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 4894 – 4901

ScandiumACHTUNGTRENNUNG(III)-Zeolites as New Heterogeneous Catalysts
FULL PAPER
Table 4. ScIII-catalyzed imino-Diels–Alder reaction of amino acid derivatives
(X=NH, NPG or -CH=CH-).
40 mol% of protons theoretically present in the
zeolite. Treatment with the fully acidic H-USY,
even in excess, did not help (Table 5, entries 4–6).
Only conventional methods were able to produce
useful amounts of the desired tetracyclic indoloqui-
nolizidinones (Table 5, entries 7,8), and trifluoro-
acetic acid in large excess (50–100 equiv) proved to
be really efficient, leading to a high yield of the ex-
pected indoloquinolizidinones. A small amount of
dichloromethane was added for solubility reasons
(Table 5, entry 8). It is interesting to note that each
dehydropiperidinone diastereoisomer 5syn or 5anti
led after cyclization under these conditions to
a single indoloquinolizidinone diastereoisomer 9cis
or 9trans (Scheme 10; Table 5). Thus, regardless of
the stereochemistry of the phenyl group adjacent to
the nitrogen atom, the same stereochemistry is pro-
duced during the acidic cyclization.
Entry[a]
Amino acid
T
t
Yield
[%][b]
Adduct
ACHTUNGTRNE(NUNG d.r.)
[8C]
[h]
1
2e
2e
60
60
15
15
0
60
–
2[c]
5g (24:76)
3
2 f
2 f
60
60
15
15
0
50
–
4[c]
5h (25:75)
Conclusion
We have developed a new heterogeneous version of
the imino-Diels–Alder and Mukaiyama reactions
based on ScIII-zeolite as catalyst. The catalyst
proved highly efficient, requiring only low loading
(ꢀ5 mol% of Sc), and was readily reusable without
any loss in reaction efficiency.
5
2g
2g
60
60
15
24
traces
40
–
6[c]
5i (25:75)
Furthermore, this ScIII-USY catalyzed imino-
Diels–Alder has been applied to the synthesis of
some peptidomimetics and highly complex indolo-
quinolizidinones. The latter alkaloid related com-
pounds were conveniently obtained in only two
steps starting from simple components (amino
acids, aldehydes, and electron-rich dienes) with
overall yields over 83%.
7
2h
2h
60
60
15
15
51
98
5j
8[c]
5j (25:75)
9
2i
2i
60
60
15
15
traces
44
–
10[c]
5k (24:76)
Experimental Section
Preparation of ScIII-USY: Commercial NH4USY was loaded in
an oven and heated at 5508C during 4 h to give H-USY. H-
USY (1 g) and ScACHTNUTRGNEUNG(OTf)3 (214 mg, 0.1 equiv) were mixed using
11
2j
2j
60
60
15
15
traces
50
–
12[c]
5l (24:76)
a mortar and charged in a closed reactor. The mixture of pow-
ders was heated at 3508C over three days under a nitrogen
flow, quantitatively yielding ScIII-USY.
General procedure for the ScIII-zeolite catalyzed synthesis of
pyranones: A suspension of ScIII-USY (34 mg, 0.05 equiv) in
acetonitrile (1.5 mL) was added the aldehyde (0.3 mmol,
1 equiv), amine (0.3 mmol, 1 equiv) and the Danishefsky diene
[a] Reaction performed at room temperature on a 0.5 mmol scale in the presence of
5% of cat.. [b] Yields of isolated and purified product; the imine accounted for the
mass balance. [c] Starting from the pre-formed imine 3a.
spherical rather than extended transition-state, leading to
the anti diastereoisomer (Scheme 9).
(0.45 mmol, 1.5 equiv). Depending on the substrates, the mixture was
either maintained at room temperature or heated at 608C. The reaction
was monitored by TLC. After completion of the reaction, the mixture
was taken up in dichloromethane (10 mL) then filtered on nylon mem-
branes (0.20 mm). Solvent evaporation provided the resulting crude prod-
uct, usually at> 95% purity as judged by NMR spectroscopy. When nec-
essary, purifications were performed by silica column chromatography
using a mixture of ethyl acetate/cyclohexane as the eluent to afford pure
products.
The presence of protons left in our heterogeneous ScIII-
zeolite catalyst suggested that direct cyclization of the
Diels–Alder adduct could also occur. However, even upon
heating and long exposure, no further reaction was ob-
served, regardless of the solvent used (Table 5, entries 1–3).
This might be due to the low amount of protons since only
5 mol% of ScIII-USY was used, leading to an approximate
Chem. Eur. J. 2012, 18, 4894 – 4901
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4899

P. Pale et al.
Data for selected products: (S) Ethyl 2-(4-oxo-2-phenyl-3,4-dihydropyri-
din-1ACTHNUTRGNE(NUG 2H)-yl)-3-phenylpropanoate: 5g colorless oil (60% yield); diaste-
reoisomeric ratio (d.r.) 24:76 syn/anti. Anti isomer: 1H NMR (400 MHz,
CDCl3): d=7.36 (d, J=8.0 Hz, 1H, 7.31–7.28 (m, 4H), 7.18 (t, J=7.5 Hz,
2H), 6.99 (dd, J=6.0, 3.1 Hz, 2H), 6.81 (dd, J=7.5, 1.5 Hz, 2H), 5.24 (d,
J=7.7 Hz, 1H), 4.62 (dd, J=13.1, J=5.4, 1H), 4.11 (q, J=7.3, 2H), 3.74
(dd, J=10.0, 5.8 Hz, 1H), 3.16 (dd, J=14.0, 5.7 Hz, 1H), 2.95 (dd, J=
14.0, 10.0 Hz, 1H), 2.62 (dd, J=16.3, 13.7 Hz, 1H), 2.49 (ddd, J=16.5,
5.0, 0.7 Hz, 1H), 1.21 ppm (t, J=7.2, 3H); 13C NMR (100 MHz, CDCl3):
d=191.7 (C), 170.7 (C), 151.0 (CH), 137.9 (C), 136.2 (C), 129.7 (2CH),
129.2 (2CH), 128.8 (CH), 128.7 (2CH), 127.84 (2CH), 127.31 (CH), 101.7
(CH), 64.5 (CH), 62.7 (CH2), 61.9 (CH), 44.8 (CH2), 36.7 (CH2),
14.4 ppm (CH3); HRMS (ESI+): m/z calcd for C22H24NO3 350.175 [M+
H]+; found: 350.171 . Syn isomer: 1H NMR (300 MHz, CDCl3): d=7.46
(d, J=7.9 Hz, 1H), 7.30–7.23 (m, 5H), 7.11–7.06 (m, 4H), 5.16 (d, J=
7.9 Hz, 1H), 4.06 (q, J=7.1 Hz, 2H), 4.01 (dd, J=7.7, 7.5 Hz, 1H), 3.89
(dd, J=10.0, 5.4 Hz, 1H), 3.16 (dd, J=13.6, 5.1 Hz, 1H), 2.96 (dd, J=
13.7, 10.0 Hz, 1H), 2.52 (d, J=7.7 Hz, 2H), 1.13 ppm (t, J=7.1 Hz, 3H);
13C NMR (75 MHz, CDCl3): d=190.7 (C), 170.5 (C), 150.9 (CH), 138.9
(C), 136.3 (C), 129.4 (2CH), 129.1 (2CH), 129.0 (2CH), 128.6 (CH), 127.5
(CH), 127.3 (2CH), 100.2 (CH), 64.7 (2CH), 61.8 (CH2), 44.1 (CH2), 38.8
(CH2), 14.1 ppm (CH3). HRMS (ESI+): m/z calcd for C22H24NO3:
350.175 [M+H]+, found: 350.171.
Scheme 9. Possible model accounting for the diastereoselectivity ob-
served in the Sc-USY-catalyzed reaction of imines derived from amino
acids.
(4R,6S,12bR) Methyl 2-oxo-4-phenyl-1,2,3,4,6,7,12,12b-octahydroindolo-
Table 5. Cyclization of indolic dehydropiperidinones to indoloquinolizidi-
nones.
AHCTUNGTRENNUNG
[2,3a]quinolizine-6-carboxylate 9cis: Colorless oil, 85% yield. 1H NMR
(400 MHz, CDCl3): d=8.03 (bs, 1H), 7.47–7.42 (m, 3H), (7.36 (t, J=
7.4 Hz, 2H), 7.31 (d, J=7.4 Hz, 2H), 7.15 (dd, J=7.8 Hz, 1H), 7.10 (dd,
J=7.9, 6.8 Hz, 1H), 5.08 (dd, J=5.8, 4.5 Hz, 1H), 4.15 (dd, J=9.3,
4.2 Hz, 1H), 3.96 (dd, J=6.6, 2.0 Hz, 1H), 3.57 (s, 3H), 3.18 (dd, J=14.5,
6.2 Hz, 1H), 3.12 (ddd, J=16.5, 2.1, 1.7 Hz, 1H), 3.03 (ddd, J=16.4, 6.8,
2.2 Hz, 1H), 2.84–2.78 (m, 2H), 2.48 ppm (ddd, J=14.1, 4.3, 2.2 Hz, 1H);
13C NMR (100 MHz, CDCl3): d=207.45 (C), 173.13 (C), 141.2 (C), 136.2
(C), 131.7 (C), 129.1 (2CH), 128.3 (CH), 127.6 (2CH), 127.2 (C), 122.2
(CH), 119.9 (CH), 118.4 (CH), 111.4 (CH), 106.2 (C), 62.25 (CH), 57.92
(CH), 52.3 (CH3), 51.6 (CH), 48.8 (CH2), 43.8 (CH2), 20.45 ppm (CH2);
HRMS (ESI+): m/z calcd for C23H22N2O3Na: 397.1528 [M+Na]+; found:
397.152.
Entry[a] Indolic
dehydropiperidinone
Reagent
Sc-USY
Solvent
T
AHCTUNGTERG[NNUN 8C]
Yield
[%][b]
1
2
3
4
5
6
R=H, CHO, Boc
R=H, CHO, Boc
R=H, CHO, Boc
R=H
CH2Cl2 RT to
reflux
MeCN RT to
reflux
0
0
0
0
0
0
Tol
RT to
reflux
H-USY
H2SO4
CH2Cl2 RT to
reflux
MeCN RT to
reflux
Acknowledgements
R=H
A.O. thanks the French CNRS and the Spanish government for her suc-
cessive post-doctoral fellowships. The authors thank the French Ministry
of Research and the CNRS for their financial support.
R=H
Tol
RT to
reflux
7
8
R=H
R=H
MeOH reflux
50
85
CF3COOH CH2Cl2
0
[a] Reaction performed at room temperature on 0.5 mmol. [b] Yields of
isolated and purified product; the imine accounted for the mass balance.
[1] a) D. OꢂHagan, Nat. Prod. Rep. 2000, 17, 435–446; b) G. B. Fodor,
B. Colasanti, Pyridine and Piperidine Alkaloids, in Alkaloids: Chem-
ical and Biological Perspectives Vol. 3, Pergamon, New York, 1996,
pp. 1–90 and M. J. Schneider (update in Vol. 10), 1996, pp. 155–299.
[2] D. S. Seigler, Plant secondary metabolism, Kluwer Academic Pub-
lishers, New York, 2002.
[3] a) J. W. Daly, H. M. Garraffo, T. F. Spande in Alkaloids from Am-
phibian Skins, in Alkaloids: Chemical and Biological Perspectives,
Vol. 13 (Ed.: S. W. Pelletier), Pergamon, New York, 1999, pp. 1–
161; b) J. W. Daly, T. F. Spande, H. M. Garraffo, J. Nat. Prod. 2005,
68, 1556–1575.
[4] a) A. Numata, T. Ibuka, Alkaloids from Ants and other Insects, in
Alkaloids: Chemical and Physiology, Vol. 31 (Ed.: A. Brossi), Aca-
demic Press, San Diego, 1987, pp. 193–316; b) P. Laurent, J-C.
Braeckman, D. Daloze, Top. Curr. Chem. 2005, 240, 167–279.
[5] a) R. A. Saporito, H. M. Garraffo, M. A. Donnelly, A. L. Edwards,
J. W. Daly, Proc. Natl. Acad. Sci. USA 2004, 101, 8045–8050;
b) J. W. Daly, H. M. Garraffo, T. F. Spande, L-A. Giddings, R. A.
Saporito, D. R. Vieites, M. Vences, J. Chem. Ecol. 2008, 34, 252–
279.
Scheme 10. Cyclization of indolyl dehydropiperidinones to indoloquinoli-
zidinones.
4900
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 4894 – 4901

ScandiumACHTUNGTRENNUNG(III)-Zeolites as New Heterogeneous Catalysts
FULL PAPER
[6] J. C. MacConnell, M. S. Blum, H. M. Fales, Tetrahedron 1971, 27,
1129–1139.
[7] a) N. V. Mody, R. Henson, P. A Hedin, D. H. Kokpol, Experientia
1976, 32, 829–830; b) for a review, see: J. Vetter, Food Chem. Toxi-
col. 2004, 42, 1373–1382.
[8] R. E. Dolle, B. Le Bourdonnec, A. J. Goodman, G. A. Morales, C. J.
Thomas, W. Zhang, J. Comb. Chem. 2008, 10, 753–802.
[9] J. C. Hannam, J. J. Kulagowski, A. Madin, M. P. Ridgill, E. M.
Seward PCT Int. Appl. 2006, 043064; CAN 144:432693.
[10] a) F-X. Felpin, J. Lebreton, Eur. J. Org. Chem. 2003, 3693–3712;
b) P. M. Weintraub, J. S. Sabol, J. M. Kane, D. R. Borcherding, Tetra-
hedron 2003, 59, 2953–2989; c) R. W. Bates, K. Sa-Ei, Tetrahedron
2002, 58, 5957–5978; d) S. Laschat, T. Dickner, Synthesis 2000,
1781–1813.
[11] a) G. R. Heintzelman, I. R. Meigh, R. Mahajan, S. M. Weinreb, Org.
React. 2005, 65, 141–599; b) S. M. Weinreb, Heterodienophile addi-
tions to dienes in Comprehensive Organic Synthesis Vol. 5, (Eds.:
B. M. Trost, I. Fleming), Pergamon, Oxford, 1991.
[20] a) A. Olmos, A. Alix, J. Sommer, P. Pale, Chem. Eur. J. 2009, 15,
11229–11234; b) A. Olmos, J. Sommer, P. Pale, Chem. Eur. J. 2011,
17, 1907–1914.
[21] 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.
[22] These results will be reported in a more specialized journal.
[23] S. Danishefsky, T. Kitahara, J. Am. Chem. Soc. 1974, 96, 7807–7809.
[24] We arbitrarily stopped after 10 runs; the catalyst could probably still
be reused for more runs.
[25] a) J. C. Tai, N. L. Allinger, J. Am. Chem. Soc. 1976, 98, 7928–7932;
b) D. R. Lide Jr., M. Jen, J. Chem. Phys. 1974, 40, 252–253; c) S.
Danishefsky, R. K. Singh, R. B. Gammill, J. Org. Chem. 1978, 43,
379–380.
[26] a) J. Savard, P. Brassard, Tetrahedron 1984, 40, 3455–3464; b) L.
Lombardo, Tetrahedron Lett. 1985, 26, 381–384.
[12] J. F. Kerwin, S. Danishefsky, Tetrahedron Lett. 1982, 23, 3739–3742.
[13] a) L. Le Coz, C. Veyrat-Martin, L. Wartski, J. Seyden-Penne, C.
Bois, M. Philoche-Levisalle, J. Org. Chem. 1990, 55, 4870–4879;
b) S. Kobayashi, S. Komiyama, H. Ishitani, Angew. Chem. 1998, 110,
1026–1028; Angew. Chem. Int. Ed. 1998, 37, 979–981.
[14] H. Laurent-Robert, B. Garrigues, J. Dubac, Synlett 2000, 1160–1162.
[15] a) S. Kobayashi, H. Ishitani, S. Nagayama, Synthesis 1995, 1195–
202; b) S. Kobayashi, S. Nagayama, J. Am. Chem. Soc. 1997, 119,
10049–10053; c) J. Collin, N. Jaber, M. I. Lannou, Tetrahedron Lett.
2001, 42, 7405–7407.
[16] V. Jurcik, K. Arai, M. M. Salter, Y. Yamashita, S. Kobayashi, Adv.
Synth. Catal. 2008, 350, 647–651.
[17] a) T. Akiyama, Y. Tamura, J. Itoh, H. Morita, K. Fuchibe, Synlett
2006, 141–143; b) T. Akiyama, J. Takaya, H. Kagoshima, Tetrahe-
dron Lett. 1999, 40, 7831–7834.
[27] a) J. R. Grunwell, A. Karipides, C. T. Wigal, S. W. Heinzman, J.
Parlow, J. A. Surso, L. Clayton, F. J. Fleitz, M. Daffner, J. E. Stevens,
J. Org. Chem. 1991, 56, 91–95; b) M. Sato, Y. Sugita, Y. Abiko, C.
Kaneko, Tetrahedron: Asymmetry 1992, 3, 1157–1160; c) M. Sato, S.
Sunami, Y. Sugita, C. Kaneko, Chem. Pharm. Bull. 1994, 42, 839–
845.
[28] a) S. Hermitage, J. A. K. Howard, D. Jay, R. G. Pritchard, M. R. Pro-
bert, A. Whiting, Org. Biomol. Chem. 2004, 2, 2451–2460; b) A.
Whiting, C. M. Windsor, Tetrahedron 1998, 54, 6035–6050.
[29] For more theoretical insights into the mechanism of such reactions,
see : a) M. A. McCarrick, Y.-D. Wu, K. N. Houk, J. Org. Chem.
1993, 58, 3330–3343; b) S. A. Kozmin, M. T. Green, V. H. Rawal, J.
Org. Chem. 1999, 64, 8045–8047.
[30] S. J. Danishefsky, E. Larson, D. Askin, N. Kato, J. Am. Chem. Soc.
1985, 107, 1246–1255.
[18] Polystyrene-supported reactions have been described; see: a) Y.
Wang, S. R. Wilson, Tetrahedron Lett. 1997, 38, 4021–4024; b) W.
Zhang, W. Xie, J. Fang, P. G. Wang, Tetrahedron Lett. 1999, 40,
7929–7933; c) C. J. Creighton, C. W. Zapf, J. H. Bu, M. Goodman,
Org. Lett. 1999, 1, 1407–1409.
[31] J. Itoh, K. Fuchibe, T. Akiyama, Angew. Chem. 2006, 118, 4914–
4916; Angew. Chem. Int. Ed. 2006, 45, 4796–4798.
[32] A. Grauer, B. Kçnig, Eur. J. Org. Chem. 2009, 5099–5111.
[33] a) M. M. Murburg, E. C. Villacres, G. N. Ko, R. C. Veith, J. Clin. En-
docrinol. Metab. 1991, 73, 861–865; b) S. W. Tam, M. Worcel, M.
Wyllie, Pharmacol. Ther. 2001, 91, 215–243.
[34] F. Wehner, A. Nçren-Mꢃller, O. Mꢃller, I. R. Correa, A. Giannis, H.
Waldmann, ChemBioChem 2008, 9, 401–405.
[35] A. Nçren-Mꢃller, I. R. Correa, H. Prinz, C. Rosenbaum, K. Saxena,
H. Schwalbe, D. Vestweber, G. Cagna, S. Schunk, O. Schwartz, H.
Schiewe, H. Waldmann, Proc. Natl. Acad. Sci. USA 2006, 103,
10606–10611.
[19] 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,
Tetrahedron 2008, 64, 8922–8929; d) M. K. Patil, M. Keller, M. B.
Reddy, P. Pale, J. Sommer, Eur. J. Org. Chem. 2008, 4440–4445;
e) P. Kuhn, A. Alix, M. Kumarraja, B. Louis, P. Pale, J. Sommer,
Eur. J. Org. Chem. 2009, 423–429; f) P. Kuhn, P. Pale, J. Sommer, B.
Louis, J. Phys. Chem. C 2009, 113, 2903–2910; g) M. Keller, A. Sani
Souna Sido, P. Pale, J. Sommer, Chem. Eur. J. 2009, 15, 2810–2817;
h) V. Bꢀneteau, A. Olmos, T. Boningari, J. Sommer, P. Pale, Tetrahe-
dron Lett. 2010, 51, 3673–3677; i) T. Boningari, A. Olmos, M. B.
Reddy, J. Sommer, P. Pale, Eur. J. Org. Chem. 2010, 6338–6347.
[36] K. N. Houk, M. N. Paddon-Row, N. G. Rondan, Y. D. Wu, F. K.
Brown, D. C. Spellmeyer, J. T. Metz, Y. Li, R. J. Loncharich, Science
1986, 231, 1108–1117.
Received: November 17, 2011
Published online: March 13, 2012
Chem. Eur. J. 2012, 18, 4894 – 4901
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4901