Peri- and enantioselectivity of thermal, scandium-, and [Pybox/Scandium]-catalyzed diels-alder and hetero-diels-alder reactions of methyl (E)-2-Oxo-4-aryl-butenoates with cyclopentadiene

Article abstract of DOI:10.1002/chem.200700995

The cycloaddition between methyl (E)-2-oxo-4-aryl-3-butenoates (2a-d) and cyclopentadiene, in addition to the expected normal Diels-Alder (DA) adducts endo-3a-d and exo-4a-d, gives the less expected endo-5a-d products of the [4 + 2] heteroDiels-Alder (HDA

Full text of DOI:10.1002/chem.200700995

DOI: 10.1002/chem.200700995
Peri- and Enantioselectivity of Thermal, Scandium-, and [Pybox/Scandium]-
Catalyzed Diels–Alder and Hetero-Diels–Alder Reactions of Methyl (E)-2-
Oxo-4-aryl-butenoates with Cyclopentadiene
[a]
Giovanni Desimoni,*[a] Giuseppe Faita,[a] Marco Toscanini, and Massimo Boiocchi[b]
Abstract: The cycloaddition between
methyl (E)-2-oxo-4-aryl-3-butenoates
(2a–d) and cyclopentadiene, in addi-
tion to the expected normal Diels–
Alder (DA) adducts endo-3a–d and
exo-4a–d, gives the less expected endo-
5a–d products of the [4+2] hetero-
Diels–Alder (HDA) reaction in which
the a-ketoester behaves as a hetero-
diene.If a comparison is made between
of (4’S,5’S)-2,6-bis
A
stereospecific thermal Claisen rear-
rangement of the optically active
5a,b,d into 3a,b,d completes the corre-
lation between their absolute configu-
ration.The [3,3]-sigmatropic rearrange-
ments can be easily carried out under
AHCTREUNG
yl]pyridine (1) and both the DA and
the HDA products are obtained with
excellent enantiomeric excess, up to
>99% ee.The X-ray crystallographic
structure determination of 5c assigns it
the 4R,4aS,7aR absolute configuration.
The thermal retro-Claisen rearrange-
ment of 3c into (4R,4aS,7aR)-5c allows
the correlation of their absolute config-
uration, and 3c has therefore the
2R,3R configuration.By analogy the
same absolute configuration can be as-
signed to 3a,b,d and 5a,b,d, and the
catalytic conditions with scandium(III)
A
triflate, which promotes the equilibra-
tion between 3a–d and 5a–d, with a
different degree of enantioselectivity
characterizing the process starting from
3a–d or 5a–d.The unambiguous attri-
butions of the configuration to the
products allows us to propose a ration-
ale of the stereochemical outcome of
the catalyzed cycloaddition and to in-
vestigate the reaction mechanism of
the competing DA and HDA reactions
and shifts in products distribution by
acid catalysis.
the thermal and the scandium(III) tri-
A
flate-catalyzed conditions, the perise-
lectivity changes and whereas under
thermal conditions the main products
are those from the DA reaction (3a–
d), in the presence of Sc(OTf)3 (OTf=
N
triflate), the HDA products 5a–d
become largely predominant.The reac-
tions are enantioselectively catalyzed
Keywords: asymmetric catalysis
cycloaddition · enantioselectivity ·
periselectivity · pybox
by the scandium(III) triflate complex
G
Introduction
points of view.[4–9] The periselectivity of the competing
Diels–Alder (DA) and hetero-Diels–Alder (HDA) reactions
may arise from 1) two distinct reaction pathways; 2) a single
reaction pathway that bifurcates after the transition state;
3) a partial conversion of the primary cycloadduct of the
pericyclic reaction into the second one.A way to infer these
different options can be the comparison of the change of the
periselectivity observed in the uncatalyzed and Lewis acid
catalyzed reactions.
The [4 +2] cycloaddition between two different dienes,[1–3]
such as that involving a diene and an heterodiene, is an up-
to-date argument from both the theoretical and synthetic
[a] Prof.Dr.G.Desimoni, Prof.Dr.G.Faita, Dr.M.Toscanini
Dipartimento di Chimica Organica
dellꢀUniversità di Pavia
A typical example is given by the reaction of cyclopenta-
diene (Cp) with an a,b-unsaturated carbonyl derivative: Cp
may behave either as the 4p component to give the normal
DA cycloadduct or, less usually, as the 2p component to fur-
nish the HDA cycloadduct.Two specific cycloadditions that
satisfy such features have been studied: the reactions of Cp
with either a-keto-b,g-unsaturated phosphonates or (E)-2-
oxo-4-phenyl-butenoate.
Viale Taramelli 10, 27100 Pavia (Italy)
Fax : (+39)382-987323
E-mail: desimoni@unipv.it
[b] Dr.M.Boiocchi
Centro Grandi Strumenti
dellꢀUniversità di Pavia
Via Bassi 21, 27100 Pavia (Italy)
Supporting information for this article is available on the WWW
under http://www.chemeurj.org/ or from the author. pybox=pyridine-
bis(oxazoline).
9478
ꢁ 2007 Wiley-VCH Verlag GmbH & Co.KGaA, Weinheim
Chem. Eur. J. 2007, 13, 9478 – 9485

FULL PAPER
In the first case, synthetic studies were carried out inde-
pendently by Evans et al.[10,11] and Hanessian and Com-
pain[12] with and without a Lewis acid catalyst, whereas a
computational investigation has been more recently been re-
ported by Houk and co-workers.[13] Evans further contribut-
ed to insight into the reaction by investigating enantioselec-
tive catalysis with a CuII complex of 4-tert-butylbis(oxazo-
line): both the DA and HDA cycloadducts were obtained
with good or excellent diastereo- and enantioselectivities.
The second example concerns the catalyzed reaction be-
tween Cp and methyl (E)-2-oxo-4-phenyl-butenoate (2a).
After initial studies by Kanemasa et al.[14] and Zhou and
Tang,[15] the effect of the lanthanide–triflate complexes of
Cp were investigated in the absence of any catalyst at ambi-
ent temperature.The reaction conditions and the results are
reported in Table 1 (entries 1–4): the yields were excellent
Table 1.Uncatalyzed and Sc III-catalyzed DA and HDA reactions be-
tween 2a–d and Cp in CH2Cl2.
Entry Sub-
strate
Catalyst[a] T [8C]
Time
Yield
[%]
ACHTREUNG
1
2
3
4
5
6
7
8
9
2a
2b
2c
2d
2a
2b
2c
2d
2a
2b
2c
2d
–
–
–
–
Sc
Sc
Sc
Sc
Sc
Sc
Sc
Sc
ambient 24 h
ambient 5 days quant.89:11
ambient 10 h quant.92:8
quant.82:18
83:17
87:13
81:19
ambient 4 days quant 92:8
15 min quant.35:65
77:23
(OTf)3 ꢀ50
89:11
95:5
92:8
91:9
92:8
>95:<5
91:9
89:11
G
15 min quant.86:14
(4’S,5’S)-2,6-bis[4’-(triisopropylsilyl)oxymethyl-5’-phenyl-
G
E
15 min quant.90:10
E
5 h
55[b]
38:62
1’,3’-oxazolin-2’-yl]pyridine (1) as chiral catalysts was inves-
tigated (Scheme 1).[16] It was found that the best enantiose-
lective catalyst was based on the ScIII ion since both the DA
product, endo-3a, and the HDA product, endo-5a, were ob-
tained with almost complete enantioselectivity.
N
3 days quant.41:59
10
11
12
G
3 days 97
73:27
U
3 days quant.85:15
C
3 days 80[b]
47:53
[a] Loading=10 mol%.[b] Consistent amount of 2d was isolated.
Whereas the absolute configuration of the DA product
between Cp and dimethyl (but-2-enoyl)phosphonate was de-
termined through chemical correlation,[11] the absolute con-
figuration of product 3a, with a negative optical rotation,
was assumed to be the 2R,3R isomer[14–16] on the basis of
confident considerations about the structure of the reactive
intermediates.Furthermore, the stereospecific [3,3]-Claisen
rearrangement of 5a into 3a, carried out under thermal con-
ditions, allowed the correlation of the absolute configuration
of these products.[16]
With these results in hand, the cycloaddition involving
methyl (E)-2-oxo-4-aryl-butenoates (2) was chosen as a
model to further infer the periselectivity between DA and
HDA reactions under either thermal or catalyzed condi-
tions, to determine the absolute configuration of the prod-
ucts, and to verify the possible interconversion between 3
and 5.
and in each case the main product was derived from the
endo-DA pathway (3a–d), with the DA selectivity ranging
from 82 to 92% and the endo selectivity from 77 to 87%.
When the same reactions were run in the presence of Sc-
A
change in the periselectivity was evident in the case of 2b
and 2c (Table 1, entries 6 and 7), while cycloadditions start-
ing from 2a and 2d clearly showed an inversion of the peri-
selectivity in favor of the HDA adducts 5a and 5d (Table 1,
entries 5 and 8), which became the main reaction products
as previously observed in the reaction involving a-keto-b,g-
unsaturated phosphonates.[12]
The next step was the investigation of the same reactions
under enantioselective catalytic conditions with the complex
of Sc
N
N
methyl-5’-phenyl-1’,3’-oxazolin-2’-yl]pyridine) (1).The reac-
tions were carried out with 10% of catalyst and 3-Š molecu-
lar sieves (MS; Table 2).
Results and Discussion
The DA products 3 and 4 were separated from 5 by
column chromatography, the DA adduct ratios were deter-
mined by H NMR spectroscopic analysis and the enantio-
To test the selectivity of several a,b-unsaturated carbonyl
derivatives, four different methyl (E)-2-oxo-4-aryl-but-3-
enoates (2a–d) were synthesized, and their reactions with
1
meric excess of 3–5 by chiral HPLC (Table 2), except in the
Scheme 1.The lanthanide triflate complexes of 1 as chiral catalysts in the DA and HDA reactions between cp and 2.
Chem. Eur. J. 2007, 13, 9478 – 9485
ꢁ 2007 Wiley-VCH Verlag GmbH & Co.KGaA, Weinheim
www.chemeurj.org
9479

G.Desimoni et al.
Table 2.DA and HDA reactions between 2a–d and Cp in CH2Cl2 in the presence of 10 mol% of the 1/Sc-
(OTf)3 complex and 3-Š MS.
mer (4R,4aS,7aR)-5c was iso-
lated with 99% ee, which
allows us to attribute the
2R,3R configuration to the DA
adduct 3c as well its enantio-
meric purity (Table 2, entry 3).
From the univocal attribu-
tion of the absolute configura-
tion of 3c and 5c, by analogy,
the same absolute configura-
tions 2R,3R and 4R,4aS,7aR
can be proposed for the DA
adducts 3a,b,d and the HDA
ACHTREUNG
Entry Substrate T [8C] Time
Yield [%][a]
A
3
5
%ee[b] conf.%
ee[c] conf.
1[d]
2
3
2a
2b
ꢀ50
ꢀ70
ꢀ70
ꢀ70
15 min 99
15 min 93
15 min 99
34:66
33:67
35:65
39:61
95:5
95:5
95:5
93:7
99.5 (2R,3R)-3a 99.5 (4R,4aS,7aR)-5a
99.5 (2R,3R)-3b >99.5 (4R,4aS,7aR)-5b
2c
99 (2R,3R)-3c
99 (2R,3R)-3d
99.5 (4R,4aS,7aR)-5c
99 (4R,4aS,7aR)-5d
4
2d[e]
4 h
84
[a] Yields of the isolated products.[b] Determined by HPLC with a Chiralpak AD column (for adduct 3c, see
the Experimental Section for details).[c] Determined by HPLC with a Chiralcel OJ column (for adducts 5b
and 5d, see the Experimental Section for details).[d] Data confirming those of ref.[16].[e] Compound 2d is
poorly soluble in CH2Cl2.
cases of 3c and 5b whose enantiomers could not be separat-
ed.In the case of 5c, an accurate crystallization allowed us
to obtain crystals suitable for X-ray crystal structure deter-
mination: the absolute configuration of 5c was therefore un-
ambiguously determined to be 4R,4aS,7aR (Figure 1).
adducts 5a,b,d, respectively.This attribution confirms the
previous proposal based on considerations about the struc-
ture of the octahedral reactive intermediate 6 involved in
the catalytic process (Figure 2).[16]
Scheme 2.The interconversion between 3 and 5 under thermal condi-
tions.
Figure 1.X-ray structure of 5c, thus confirming the absolute 4R,4aS,7aR
configuration.
Having determined the absolute configuration of 5c, the
relationship between 5c and 3c was a key point in assigning
an absolute configuration to the DA adduct.
Figure 2.The assumed reactive intermediate 6 in the DA and HDA reac-
tions between 2 and Cp catalyzed by the Sc(OTf)3 complex of pybox 1,
A
which gives (2R,3R)-3 and (4R,4aS,7aR)-5.pybox =pyridinebis(oxazo-
line).
Previously, we easily obtained the correlation between
HDA and DA cycloadducts by a thermal conversion of 5a
into 3a following a stereospecific Claisen [3,3]-sigmatropic
rearrangement, whereas, under the same conditions, 3a gave
only the retro-DA reaction.[16] Hence, the stereospecific sig-
matropic rearrangement 3!5 could be a useful tool both to
correlate absolute configurations and to determine the enan-
tiomeric purity when chiral HPLC analysis is unable to sep-
arate the enantiomers of 3c or 5b or when the separation is
somewhat difficult (e.g., 5d).
To exclude the possibility that the periselectivity of the
competing DA and HDA reactions arises from a partial con-
version between primary cycloadducts, the stability of 3a–d
and 5a–d toward Sc(OTf)3 and the possible loss of stereo-
G
chemistry during their eventual conversion must be checked
(Scheme 3).
First, the reactions of 2a–d with Cp at low temperature in
Following the protocol used for 5a, the enantiomeric
purity of 5b,d was easily obtained by determining the enan-
tiomeric excess of 3b,d, derived from their thermal conver-
sion (Table 2, entries 2 and 4).In the case of 3c, despite the
possibility of the retro-DA reaction, the importance of de-
termining its absolute configuration forced us to attempt the
3c!5c conversion under thermal conditions (Scheme 2).
Besides 2c deriving from the retro-DA process, the enantio-
the presence of Sc(OTf)3 were carried out for a longer reac-
A
tion time (3 days) to show possible equilibration of the
products (Table 1, entries 5–8 versus 9–12).The yields were
again good, but the isomer distribution was somewhat
changed.These results suggest that the periselectivity ob-
tained from the Sc(OTf)3-catalyzed reaction may be altered
G
(even if not dramatically) by product isomerization occur-
ring under the reaction conditions.
9480
www.chemeurj.org
ꢁ 2007 Wiley-VCH Verlag GmbH & Co.KGaA, Weinheim
Chem. Eur. J. 2007, 13, 9478 – 9485

Diels–Alder and Hetero-Diels–Alder Reactions
FULL PAPER
time (Table 1, entries 5, 6, 9, and 10).Since their formation
occurs in the presence of 2a and 2b, respectively, the test of
the stability of the HDA products at ꢀ508C was therefore
repeated in the presence of 0.5 equivalents of either 2a or
2b, thus simulating, as far as possible, the reaction condi-
tions (see the Supporting Information).Under these condi-
tions, the equilibration of the HDA products to the corre-
sponding DA adducts occurred without any decomposition
(Table 3, entries 4, 5, 9, and 10), and the influence of the a-
dicarbonyl fragment of 2a,b, able to coordinate the scandi-
um cation, cannot be ignored.
Scheme 3.Investigation into the stability of 3a–d and 5a–d to Sc(OTf)3.
G
In the study of DA/HDA cycloadditions, the retro-Claisen
rearrangement of products catalyzed with a Lewis acid has
been already reported in the case of the DA product of di-
ethyl a-crotonylphosphonate,[12] therefore it should be inter-
esting to study the conversions of 3 and 5 in the presence of
The conversion of 3a–d into 5a–d (or vice versa) cata-
lyzed by Sc(OTf)3 could either occur through a process that
G
does not modify the chiral centers of the DA and HDA ad-
ducts (in analogy to the thermal processes) or through a
non-stereoselective rearrangement.To cast aside any doubt,
a pair of enantiopure isomers must be submitted to the cata-
lyzed conversion, thus checking the variation of the optical
purity of each stereoisomer.Only 3d and 5d ensure easy
HPLC analysis and product stability, therefore they were
Sc(OTf)3 and to extend the study to the optically active
G
products to check the eventual loss of stereochemistry, a
subject that to our knowledge has not yet been explored.
The conditions first employed were those under which the
catalyzed reaction was carried out (Table 1, entries 5–8); if
the adducts were stable, then they were warmed up at ambi-
ent temperature (Table 3).
treated with Sc(OTf)3 at ambient temperature (Table 4).
H
Table 4.
[3,3]-Sigmatropic
rearrangement
(OTf)3.[a]
of
(2R,3R)-3d
and
(4R,4aS,7aR)-5d catalyzed by Sc
AHCTREUNG
Table 3.
ACHTREUNG
5 catalyzed by Sc
ACHTREUNG
Entry
Starting
isomer
Time
[min]
3d
Yield
[%]
4d
Yield
[%]
5d
ee
[%]
Product distribution (3/5)[b]
G
ee
[%]
Yield
[%][b]
Entry
Reagent
T [8C]
Time
24 h[c]
24 h
15 min
24 h
1
2
3a
3a
ꢀ50
89:11 (3a/5a)
85:15 (3a/5a)
decomposition of 5a (no 3a)
38:62 (3a/5a)
85:15 (3a/5a)
89:11 (3b/5b)
86:14 (3b/5b)
decomposition of 5b (no 3b)
8:92 (3b/5b)
81:19 (3b/5b)
3c
89:11 (3c/5c)
5c
87:13 (3c/5c)
3d
86:14 (3d/5d)
5d
83:17 (3d/5d)
1
2
3
4
5
6
7
3d
3d
3d
5d
5d
5d
5d
0
10
60
0
10
60
96
85
71
–
70
81
52
99
99
49
–
99
99
rac
4
5
15
–
3
4
–
10
14
100
27
15
–
ambient
ꢀ50
99
61
99
99
99
37
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
5a
5a[d]
5a[d]
3b
3b
5b
5b[f]
5b[f]
3c
ꢀ50
ambient
ꢀ50
3 h
24 h[e]
6 h
ambient
ꢀ70
300
30
18
15 min
24 h[g]
3 h
10 h
45 min
10 h
8 h
5 days
3 h
5 days
3 h
ꢀ70
[a] At ambient temperature.[b] exo-4d has always a significant ee value.
ambient
ꢀ70
3c
5c
5c
3d
3d
5d
5d
ambient
ꢀ70
Both 3d (containing 4% 4d) and 5d, which are stable at
ꢀ708C, begin to rearrange within five minutes when treated
ambient
ꢀ70
with Sc(OTf)3 at ambient temperature.The composition of
A
ambient
ꢀ70
the mixtures obtained from both 3d and 5d after 10 and 60
minutes was determined to check the enantiomeric excess of
both the residual and developing products.After 10 minutes
the mixtures were not yet fully equilibrated, and the enan-
tiomeric excess was that of the starting product (Table 4, en-
tries 2 and 5).After one hour, either starting from 3d or 5d,
ambient
[a] About 0.1 mmol of reagent (when 3 is used as the reagent, it is conta-
minated with 4); 10% catalyst in CH2Cl2.[b] Variable amounts 4 are
present.[c] Compound 5a begins to form after 35 min.[d] Addition of
0.05 mmol of 2a.[e] Compound 5b begins to form after 35 min.[f] Ad-
dition of 0.05 mmol of 2b.[g] Compound 3b begins to form after 1h.
the ratio of 3d(+4d)/5d was the same as that found in the
A
previous equilibrations (Table 3, entries 16 and 18), but two
different results were observed.When 3d is the starting
product, the enantiomeric excess of both 3d and 5d drops
and the amount of 4d strongly increases (Table 4, entry 3).
When 5d is the starting product (Table 4, entry 6), after one
hour the enantiomeric purities of both the developing 3d
and the residual 5d do not change and the rearrangement is
stereospecific.When 5d is in the presence of scandium tri-
flate for 5 hours, the amount of 4d strongly increases, both
3d and 5d loose their enantiomeric purities, and the behav-
ior of 5d tends to that observed for 3d (Table 4, entry 7).
The data in Table 3 show a significant constancy in the
product distribution at ambient temperature (with a prefer-
ence for the DA products), starting either from 3a–d or
5c,d, the ratio of which can be the result of an equilibrium.
Two results (Table 3, entries 3 and 8) seem at a first glance
to be illogical.In the presence of Sc (OTf)3, 5a and 5b de-
A
compose at ꢀ50 (ꢀ70)8C within a short time.On the other
hand, adducts 5a,b have been obtained under similar condi-
tions from the reaction of 2a,b and Cp catalyzed by Sc-
(OTf)3 and were found to be stable for a longer reaction
Chem. Eur. J. 2007, 13, 9478 – 9485
ꢁ 2007 Wiley-VCH Verlag GmbH & Co.KGaA, Weinheim
www.chemeurj.org
9481

G.Desimoni et al.
Therefore, the beginning of the scandium-catalyzed rear-
rangement can be classified as a stereospecific [3,3]-sigma-
tropic rearrangement as in the case of the thermal reaction.
Then, starting either from 3d or 5d, the prolonged interac-
tion between the substrate and Lewis acid promotes other
processes whose mechanisms deserve further investiga-
tion.[17]
and to determine the enantiomeric purity of products not
separable by chiral HPLC analysis.
The stability of either racemic 3a–d and 5a–d or enantio-
pure 3d and 5d in the presence of Sc(OTf)3 was checked at
A
low and ambient temperature, with and without 1.From the
results the following features can be pointed out:
Finally, about the isomer distribution and the competing
DA and HDA cycloadditions in the enantioselective reac-
1.The equilibrations set up at ambient temperature starting
from racemic 3a–d and 5a–d are complete within a few
hours, thus giving rise to a DA/HDA product ratio of
about 85:15 independent of the starting materials.
2.If the previous reactions are carried out at low tempera-
ture, equilibration is slower and incomplete in the case of
3a,b and 5a,b or completely absent for 3c,d and 5c,d.
tions catalyzed by the complex between 1 and Sc(OTf)3, the
A
same enantiomeric excess obtained for both the DA and
HDA products (Table 2) certainly does not suggest two dis-
tinct reaction pathways.To understand whether their forma-
tion occurs from a single reaction pathway that bifurcates
after the transition state, thus giving the different products,
or from the partial stereospecific conversion of one single
product into the second one, (2R,3R)-3a–d and
3.The complex 1/Sc(OTf)3 does not promote any isomeri-
G
zation under the reaction conditions, and both the DA
and HDA products are stable.
(4R,4aS,7aR)-5a–d were treated with the Sc(OTf)3/pybox 1
G
complex under the reaction conditions (Table 2) and no
product equilibration was observed.This control experiment
has been already been carried out by Evans et al.[11] on the
enantiopure DA and HDA adducts obtained from the reac-
tion between a-keto-b,g-unsaturated phosphonates and Cp
From this evidence, great care is required in the compari-
son between the periselectivity observed in the uncatalyzed
and the Sc(OTf)3-catalyzed reactions, and it cannot be ex-
A
cluded that the DA/HDA ratio could sometimes be altered
from a partial equilibration of the primary reaction products
induced by the ScIII cation.These problems are fully avoided
catalyzed by a tert-butyl-box/Cu(SbF6)2 (box=bis(oxazo-
A
line)) complex, and also in this case no product conversion
occurred.Hence, also the DA and HDA cycloadducts ob-
tained in the reactions between 2a–d and Cp in the presence
when the reactions are catalyzed by the complex 1/Sc(OTf)3
G
since cycloadducts 3a–d and 5a–d have been found to be
the kinetically controlled products.Since the enantiomeric
purities of both the DA and HDA isomers are the same, we
can conclude that the discovery by Houk and co-workers[13]
that the competition between the DA and HDA processes
can arise from a single reaction pathway that bifurcates
after the transition state rationalizes the results of the above
study concerning the enantioselective reaction between 2a–
of the Sc(OTf)3/pybox 1 catalyst are primary products, thus
A
confirming the existence of a single transition state that
leads to adducts 3 and 5.
Conclusions
d and Cp catalyzed by the complex 1/Sc(OTf)3.
G
Pybox units are excellent ligands that give rise to catalysts
for several enantioselective reactions.[18] Especially, the lan-
thanide complexes of pybox 1 have already been used as ef-
ficient catalysts in the enantioselective DA and Mukaiya-
ma–Michael reactions with acryloyl- and crotonoyloxazolidi-
nones[19] and in the Mukaiyama–aldol reaction of pyru-
Experimental Section
General: Melting points were determined by the capillary method and
are uncorrected. 1H and 13C NMR spectra were recorded at 300 and
75 MHz, respectively (CDCl3, 258C, trimethylsilane (TMS)).IR spectra
were registered on a Perkin-Elmer RX I spectrophotometer.Optical ro-
tations were measured on a Perkin-Elmer 241 polarimeter.Separation
and purification of the products was carried out by column chromatogra-
phy on Merck silica gel 60 (230–400 mesh).The enantiomeric excess ( ee)
of the products was determined by HPLC using Daicel columns: Chiral-
pak AD for 3a,b,d and Chiralcel OJ for 5a,c,d.
vates.[20] The complex 1/Sc
(OTf)3 is a very efficient stereose-
A
lective catalyst of the DA and HDA reactions between
cyclopentadiene and methyl (E)-2-oxo-4-aryl-3-butenoates
(2a–d), thus allowing excellent control of both the diaster-
eo- and enantioselectivities, both in the DA (3a–d) and
HDA (5a–d) reactions, with enantiomeric excess not easily
obtainable with other optically active catalysts.[14,15]
The X-ray structure of 5c allowed us to determine its ab-
solute configuration as 4R,4aS,7aR, and the conversion of
(4R,4aS,7aR)-5c into 3c, through a stereospecific retro-
Claisen rearrangement, allowed us to determine the abso-
lute configuration of the latter as 2R,3R, thus confirming the
previously proposed configuration on the basis of mechanis-
tic considerations.[16] By analogy, the same absolute configu-
ration can be attributed to 3b,d and 5b,d.In this context,
the stereospecific rearrangement between 3 and 5 has been
found to be a useful tool to correlate absolute configurations
Materials: Dichloromethane, hydrocarbon-stabilized ACS grade from Al-
drich, was distilled from calcium hydride and used immediately.Scandi-
um triflate was obtained from Aldrich as the ACS reagent.Powdered
molecular sieves (3 Š) were obtained from Aldrich and were heated
under vacuum at 3008C for 5 h and kept in sealed vials in a dryer.
(4’S,5’S)-2,6-Bis[4’-(triisopropylsilyl)oxymethyl-5’-phenyl-1’,3’-oxazolin-2’-
A
yl]pyridine (1) was prepared as previously described.[19] (E)-2-Oxo-4-phe-
nylbut-3-enoic acid methyl ester (2a) was prepared, following a previous-
ly reported method,[21,22] from esterification with methanol of the potassi-
um salt, obtained from benzaldehyde and pyruvic acid in the presence of
KOH.Yellow needles were obtained from diisopropyl ether (mp.6. 9–
708C, ref.[22] mp. .70–71 8C).Following the same procedure, and starting
from the suitable aldehyde, the following products were prepared: (E)-2-
9482
www.chemeurj.org
ꢁ 2007 Wiley-VCH Verlag GmbH & Co.KGaA, Weinheim
Chem. Eur. J. 2007, 13, 9478 – 9485

Diels–Alder and Hetero-Diels–Alder Reactions
FULL PAPER
oxo-4-(4-methylphenyl)but-3-enoic acid methyl ester (2b) in a yield of
45% (a sample suitable for this study was purified by crystallization from
diisopropyl ether to give yellow needles (m.p. 80–818C, ref.[23]
m.p. 818C)); (E)-2-oxo-4-(4-bromophenyl)but-3-enoic acid methyl ester
(2c) in a yield of 35% as yellow needles from methanol (m.p. 1208C,
ref.[24] mp. .122 8C); (E)-2-oxo-4-(4-nitrophenyl)but-3-enoic acid methyl
ester (2d) in a yield of 32% as bright-orange crystals from ethyl acetate
(m.p. 182–1838C, ref. [25] m.p. 182.5–183.58C).
5c: 1H NMR (CDCl3): d=7.47 (d, 3J
tons), 7.13 (d, 3J(H,H)=8.3 Hz, 2H, aromatic protons), 6.25 (dd, 3J-
(H,H)=3.0 Hz, 4J
(H,H)=1.3 Hz, 1H, H-3), 6.06 (s, 2H, H-6 and H-7),
5.11 (dd, 3J(H,H)=5.8 Hz, 4J(H,H)=1.3 Hz, 1H, CH), 4.05 (dd, 3J-
(H,H)=6.8 and 2.8 Hz, 1H, CH), 3.82 (s, 3H, OCH3), 2.86 (m, 1H, CH),
2.07 (dd, 2J(H,H)=16.3 Hz, 3J
(H,H)=8.4 Hz, 1H, CHH), 1.73 ppm (dd,
2J(H,H)=16.3 Hz, 3J(H,H)=7.4 Hz, 1H, CHH); 13C NMR (CDCl3): d=
162.7, 145.9, 140.4, 137.8, 131.1, 130.2, 128.9, 120.0, 110.6, 81.6, 51.8, 42.8,
37.2, 33.2 ppm; IR (Nujol): n¯ =1721 (C=O), 1653 cmꢀ1 (C=C); elemental
analysis calcd (%) for C16H15BrO3 (335.2): C 57.33, H 4.51; found: C
57.45, H 4.36.
(H,H)=8.3 Hz, 2H, aromatic pro-
AHCTREUNG
U
ACHTREUNG
U
ACHTREUNG
AHCTREUNG
G
ACHTREUNG
E
ACHTREUNG
General procedure for the reaction between cyclopentadiene and (E)-2-
oxo-4-arylbut-3-enoic acid methyl ester (2a–d):
Uncatalyzed reactions (Table 1, entries 1–4): Cyclopentadiene (300 mL;
approximately 4.5 mmol) was added by microsyringe to a solution of 2a–
d (0.30 mmol) in anhydrous CH2Cl2 (0.3 mL) at ambient temperature in a
rubber-septum-sealed vial.Stirring of the reaction mixture was continued
for the time reported in Table 1 until the disappearance of 2 was shown
by TLC.The reaction mixture was separated by column chromatography
(silica gel, 30 cm length, 1.5 cm diameter).
Adducts from 2d: The products were isolated following the procedure
described for 2a; the ratio 3d/4d was determined by 1H NMR spectro-
scopic analysis from the aromatic protons at d=7.41 (3d) and 7.30 ppm
(4d).
3d: 1H NMR (CDCl3): d=8.17 (d, 3J
tons), 7.41 (d, 3J(H,H)=8.7 Hz, 2H, aromatic protons), 6.47 (dd, 3J-
(H,H)=5.6 and 3.0 Hz, 1H, vinylic proton), 6.00 (dd, 3J
(H,H)=5.6 and
3.0 Hz, 1H, vinylic proton), 3.89 (s, 3H, OCH3), 3.71 (dd, 3J
(H,H)=5.1
and 3.5 Hz, 1H, CH), 3.57 (bs, 1H, CH), 3.37 (d, 3J
(H,H)=4.8 Hz, 1H,
CH), 3.12 (bs, 1H, CH), 1.89 (d, 2J
(H,H)=8.9 Hz, 1H, CHH), 1.70 ppm
A
AHCTREUNG
A
ACHTREUNG
Adducts from 2a: Cyclohexane/ethyl acetate (90:10) was the eluant, and
the inseparable mixture of the Diels–Alder products 3a and 4a was
eluted first, then the hetero-Diels–Alder product 5a was separated.The
ratio 3a/4a was determined by 1H NMR spectroscopic analysis from the
vinylic protons at d=5.97 (3a) and 6.10 ppm (4a).
AHCTREUNG
AHCTREUNG
AHCTREUNG
(m, 1H, CHH); 13C NMR (CDCl3): d=192.8, 161.4, 150.8, 145.9, 139.1,
132.6, 127.7, 123.3, 56.4, 52.6, 48.2, 47.2, 46.8, 45.0 ppm; IR (film): n¯ =
1726 cmꢀ1 (C=O); elemental analysis calcd (%) for C16H15NO5 (301.3): C
63.78, H 5.02, N 4.65; found: C 63.92, H 5.48, N 4.67.
3a: Colorless crystals; m.p. 458C (hexane); IR (Nujol): n˜ =1726 cmꢀ1 (C=
1
O); H and 13C NMR spectra are in accordance with those previously re-
ported.[16]
5d: Colorless needles; m.p. 145–1468C (diisopropyl ether); 1H NMR
5a: Colorless needles; m.p. 50–558C (diisopropyl ether/pentane); IR
(Nujol): n¯ =1734 (C=O), 1646 cmꢀ1 (C=C); 1H and 13C NMR spectra are
in accordance with those previously reported.[16]
3
ACHTREUNG
3
(CDCl3): d=8.23 (d, J
(H,H)=8.7 Hz, 2H, aromatic protons), 6.26 (dd, 3J
(H,H)=1.1 Hz, 1H, H-3), 6.08 (s, 2H, H-6 and H-7), 5.15 (d, 3J
5.8 Hz, 1H, CH), 4.22 (dd, 3J
(H,H)=6.7 and 2.8 Hz, 1H, CH), 3.84 (s,
3H, OCH3), 2.91 (m, 1H, CH), 2.07 (dd, 2J(H,H)=16.9 Hz, 3J
(H,H)=
7.7 Hz, 1H, CHH), 1.70 ppm (ddd, 2J(H,H)=16.9 Hz, 3J
(H,H)=7.5 and
A
ACHTREUNG
A
ACHTREUNG
Adducts from 2b: The products were isolated following the procedure
described for 2a; the ratio 3b/4b was determined by 1H NMR spectro-
scopic analysis from the vinylic protons at d=5.97 (3b) and 6.11 ppm
(4b).
AHCTREUNG
A
ACHTREUNG
A
ACHTREUNG
1.6 Hz, 1H, CHH); 13C NMR (CDCl3): d=162.5, 149.0, 146.4, 146.3,
137.7, 130.2, 128.1, 123.4, 109.0, 81.5, 51.9, 42.5, 37.7, 33.0 ppm; IR
(Nujol): n¯ =1726 (C=O), 1652 cmꢀ1 (C=C); elemental analysis calcd (%)
for C16H15NO5 (301.3): C 63.78, H 5.02, N 4.65; found: C 64.02, H 4.98, N
4.77.
1
3b: H NMR (CDCl3): d=7.20–7.05 (m, 4H, aromatic protons), 6.47 (dd,
3
3J
A
G
2.7 Hz, 1H, vinylic proton), 3.87 (s, 3H, OCH3), 3.76 (dd, 3J
and 3.4 Hz, 1H, CH), 3.50 (m, 1H, CH), 3.23 (dd, 3J
(H,H)=5.1 and
1.2 Hz, 1H, CH), 3.05 (m, 1H, CH), 2.35 (s, 3H, CH3), 1.96 (d, J
8.7 Hz, 1H, CHH), 1.65 ppm (dd, 3J
Reaction catalyzed by scandium triflate: A mixture of 2a–d (0.30 mmol)
and scandium triflate (14 mg, 0.03 mmol) in anhydrous CH2Cl2 (0.3 mL)
was stirred for 15 min at ambient temperature in a rubber-septum-sealed
vial and then cooled at the temperature reported in Table 1.Cyclopenta-
diene (100 mL; approximately 1.5 mmol) was added by microsyringe and
stirring of the reaction mixture was continued at the temperature and for
the time reported in Table 1.The reaction mixture was decomposed in
water, extracted with CH2Cl2, dried, and the mixture of adducts was sepa-
rated by column chromatography as previously described.
13C NMR (CDCl3): d=193.7, 161.8, 139.8, 139.4, 135.3, 132.1, 128.7,
126.8, 56.1, 52.3, 48.8, 47.2, 46.7, 44.7, 20.4 ppm; IR (film): n¯ =1726 cmꢀ1
(C=O); elemental analysis calcd (%) for C17H18O3 (270.3): C 75.53, H
6.71; found: C 75.17, H 6.55.
5b: 1H NMR (CDCl3): d=7.15 (m, 4H, aromatic protons), 6.33 (dd, 3J-
A
N
3
3
ACHTREUNG
5.13 (dd, J
A
and 3.0 Hz, 1H, CH), 3.82 (s, 3H, OCH3), 2.89 (m, 1H, CH), 2.36 (s, 3H,
Reaction catalyzed by the scandium triflate/pybox 1 complex:Compounds
2a–d (0.30 mmol), pybox 1 (22 mg, 0.03 mmol), scandium triflate (14 mg,
0.03 mmol), and MS (about 0.040 g) were added to anhydrous CH2Cl2
(0.3 mL) at ambient temperature in a rubber-septum-sealed vial. The re-
action mixture was stirred for 15 min and then cooled at the temperature
reported in Table 2.Cyclopentadiene (100 mL; approximately 1.5 mmol)
was added by a microsyringe, and stirring of the reaction mixture was
continued for the time reported in Table 2, when all 2 disappeared.The
reaction was decomposed in water, extracted with CH2Cl2, dried, and the
mixture of adducts was separated by column chromatography (silica gel,
30 cmLꢀ1, 1.5 cm diameter).
2
3
ACHTREUNG
CH3) 2.12 (dd, J
C
(H,H)=8.4 Hz, 1H, CHH), 1.75 ppm
A
ACHTREUNG
(CDCl3): d=162.9, 145.6, 138.3, 137.9, 135.7 130.2, 128.7, 127.1, 112.1,
81.8, 51.6, 43.2, 37.3, 33.3, 20.5 ppm; IR (Nujol): n¯ =1733 (C=O),
1646 cmꢀ1 (C=C); elemental analysis calcd (%) for C17H18O3 (270.3): C
75.53, H 6.71; found: C 75.38, H 6.83.
Adducts from 2c: The products were isolated following the procedure
described for 2a; the ratio 3c/4c was determined by 1H NMR spectro-
scopic analysis from the vinylic protons at d=5.96 (3c) and 6.05 ppm
(4c).
3c: 1H NMR (CDCl3): d=7.43 (d, 3J
U
The reaction mixture from 2a was separated by eluting with cyclohexane/
ethyl acetate (90:10).The mixture of DA products 3a and 4a (colorless
oil) was analyzed on a Chiralpak AD column with hexane/2-propanol
(96:4) as the eluant (1.0 mLminꢀ1), and the average retention times were
13 and 15.5 min for methyl (2S,3S)- and (2R,3R)-3-phenylbicyclo-
ACHTREUNG
A
2.7 Hz, 1H, vinylic proton), 3.87 (s, 3H, OCH3), 3.68 (dd, 3J
and 3.5 Hz, 1H, CH), 3.51 (broad s, 1H, CH), 3.21 (dd, J
1.5 Hz, 1H, CH), 3.04 (broad s, 1H, CH), 1.89 (d, 2J
CHH), 1.65 ppm (dd, 2J(H,H)=8.8 Hz, 3J
A
the exo enantiomers 4a.[16]
A
13C NMR (CDCl3): d=193.3, 161.6, 141.9, 139.2, 132.3, 131.1, 128.6,
119.5, 56.3, 52.4, 48.4, 47.2, 46.7, 44.5 ppm; IR (film): n¯ =1726 cmꢀ1 (C=
O); elemental analysis calcd (%) for C16H15BrO3 (335.2): C 57.33, H 4.51;
found: C 57.12, H 4.55.
The HDA product 2-methoxycarbonyl-4-phenyl-4,4a,5,7a-tetrahydrocylo-
penta[b]pyran (5a) was analyzed on a Chiralcel OJ column with hexane/
2-propanol (90:10) as the eluant (1.0 mLminꢀ1), and the average reten-
tion times were 20 and 36 min for (4R,4aS,7aR)-5a and (4S,4aR,7aS)-5a,
Chem. Eur. J. 2007, 13, 9478 – 9485
ꢁ 2007 Wiley-VCH Verlag GmbH & Co.KGaA, Weinheim
www.chemeurj.org
9483

G.Desimoni et al.
respectively.The sample from entry 1 of Table 2, for which [ a]2D0 =ꢀ226.4
(c=1.2 in CHCl3),[16] could be crystallized from diisopropyl ether/hexane;
m.p. 70–718C.
(5 mL) and heated at 908C for 12 h.The solvent was evaporated, and the
residue was purified by column chromatography on silica gel (eluant: cy-
clohexane/ethyl acetate (95:5)).The first fraction was starting material
3c, then a very small amount of 2c was isolated, followed by 5c (5 mg,
33% yield).This latter compound was submitted to HPLC analysis on a
Chiralcel OJ column with hexane/2-propanol (90:10) as the eluant
(1.0 mLminꢀ1), and the major enantiomer was (4R,4aS,7aR)-5c with
99% ee, thus allowing the determination of both the absolute configura-
tion and enantiomeric excess of starting material 3c.
The reaction mixture from 2b was separated by eluting with cyclohex-
ane/ethyl acetate (90:10).The mixture of DA products 3b and 4b was an-
alyzed on a Chiralpak AD column with hexane/2-propanol (98:2) as the
eluant (1.0 mLminꢀ1), and the average retention times were 8.8 and
10.1 min for methyl (2S,3S)- and (2R,3R)-3-(4-methylphenyl)bicyclo-
A
Determination of the enantiomeric excess of 5d: A sample of 5d (10 mg,
0.033 mmol) from entry 4 in Table 2 was dissolved in anhydrous toluene
(5 mL) and heated at 908C for 12 h.The solvent was evaporated, and the
residue was purified by column chromatography on silica gel (eluant: cy-
clohexane/ethyl acetate (85:15)).The first fraction was (2 R,3R)-3d (5 mg,
50% yield), and HPLC analysis performed on a Chiralpak AD column
with hexane/2-propanol (96:4) as the eluant (1.0 mLminꢀ1) gave a mix-
ture of (2S,3S)- and (2R,3R)-3d in the ratio 0.5:99.5 with [a]2D0 =ꢀ107.3
(c=0.3 in CHCl3).The second fraction was (4 R,4aS,7aR)-5d (5 mg),
which was identical in every respect to the starting product, and whose
enantiomeric excess was therefore 99% ee.
the exo enantiomers 4b.
The HDA product 2-methoxycarbonyl-4-(4-methylphenyl)phenyl-
4,4a,5,7a-tetrahydrocylopenta[b]pyran (5b) was analyzed on a Chiralce-
l OJ column with hexane/2-propanol (90:10) as the eluant (1.0 mLminꢀ1),
and the enantiomers of 5b overlap (retention time=24 min).The sample
from entry 2 of Table 2, for which [a]2D0 =ꢀ198.1 (c=2.2 in CHCl3), could
be crystallized from cyclohexane/hexane; m.p. 96–978C.
The reaction mixture from 2c was separated by eluting with cyclohexane/
ethyl acetate (95:5).The mixture of DA products 3c and 4c was analyzed
on a Chiralpak AD column with hexane/2-propanol (96:4) as the eluant
(1.0 mLminꢀ1), and the average retention times were 11 and 13.5 min for
exo enantiomers 4c, whereas the enantiomers of 3c overlap at 14 min.
General procedure for the [3,3]-sigmatropic rearrangements of (2R,3R)-
3d and (4R,4aS,7aR)-5d catalyzed by Sc
(2R,3R)-3d or (4R,4aS,7aR)-5d (about 0.1 mmol) were dissolved in anhy-
drous CH2Cl2 (1.0 mL), and Sc(OTf)3 (4.8 mg, 0.01 mmol) was added to
A
G
The HDA product 2-methoxycarbonyl-4-(4-bromophenyl)-4,4a,5,7a-tetra-
hydrocylopenta[b]pyran (5c) was analyzed on a Chiralcel OJ column
with hexane/2-propanol (90:10) as the eluant (1.0 mLminꢀ1), and the
average retention times were 22 and 26 min for (4R,4aS,7aR)-5c and
(4S,4aR,7aS)-5c, respectively.The sample from entry 3 of Table 2, which
is nearly enantiomerically pure (4R,4aS,7aR)-5c with [a]2D0 =ꢀ156.1 (c=
2.4 in CHCl3), crystallizes from diisopropyl ether/pentane as colorless
needles; m.p. 105–1068C.
AHCTREUNG
the stirred solution kept at the temperature and for the time reported in
Table 4.The reaction mixture was quenched with water and extracted
with CH2Cl2.The organic layer was dried, the solvent was evaporated,
and the composition of the residue (ratio of 3d/4d/5d]) was determined
by 1H NMR spectroscopic analysis.The overall residue was purified by
column chromatography on silica gel (eluant: cyclohexane/ethyl acetate
The reaction mixture from 2d was separated by eluting with cyclohex-
ane/ethyl acetate (90:10).The mixture of DA products 3d and 4d was an-
alyzed on a Chiralpak AD column with hexane/2-propanol (96:4) as the
eluant (1.0 mLminꢀ1), and the average retention times were 26.5 and
30.5 min for methyl (2R,3R)- and (2S,3S)-3-(4-nitrophenyl)bicyclo-
(85:15)).The first fraction was 3d(+4d), and HPLC analysis performed
N
on a Chiralpak AD column with hexane/2-propanol (96:4) as the eluant
(1.0 mLminꢀ1) allowed both a precise determination of the ratio of 3d/
4d and the enantiomeric composition of 3d.The second fraction was 5d,
and its enantiomeric purity was determined either by HPLC (Chiralcel
OJ column) and the optical rotation of the product.
[2.2.1]hept-5-en-2-ylglyoxylate (3d), respectively, and 19 and 24.5 min for
the exo enantiomers 4d.
CCDC-649656 (5c) contains the supplementary crystallographic data for
this paper.These data can be obtained free of charge from the Cam-
bridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_
request/cif.[26]
The HDA product 2-methoxycarbonyl-4-(4-nitrophenyl)-4,4a,5,7a-tetra-
hydrocyclopenta[b]pyran (5d) was analyzed on a Chiralcel OJ column
with hexane/2-propanol (90:10) as the eluant (1.0 mLminꢀ1), and the
average retention times were 79 and 84 min for (4S,4aR,7aS)- and
(4R,4aS,7aR)-5d, respectively.The sample from entry 4 of Table 2 has
[a]2D0 =ꢀ147.2 (c=1.3 in CHCl3).
General procedure for the [3,3]-sigmatropic rearrangements of 3a–d and
Acknowledgement
5a–d catalyzed by Sc
A
ACHTREUNG
0.1 mmol) were dissolved in anhydrous CH2Cl2 (1.0 mL), and Sc
AHCTREUNG
This work was supported by the Ministero dellꢀUniversità e della Ricerca
(MUR) and by the University of Pavia.
(4.8mg, 0.01 mmol) was added to the stirred solution kept at the temper-
ature and for the time reported in Table 3.The reaction mixture was
quenched with water and extracted with CH2Cl2.The organic layer was
dried, the solvent was evaporated, and the composition of the residue
(3a–d/5a–d ratio) was determined by 1H NMR spectroscopic analysis.In
[1] L.Toma, S.Romano, P.Quadrelli, P.Caramella,
Tetrahedron Lett.
2001, 42, 5077–5080.
the case of entry 5, 2a (0.05 mmol) and Sc(OTf)3 (0.01 mmol) in anhy-
G
[2] P.Caramella, P.Quadrelli, L.Toma,
1130–1131.
J. Am. Chem. Soc. 2002, 124,
drous CH2Cl2 (0.5 mL) were stirred at ambient temperature for 15 min,
the solution was cooled at ꢀ508C, a cooled solution of 5a (0.1 mmol)
was added, and the reaction mixture was warmed to ambient tempera-
ture for 3 h and worked up as described above.Compound 2a did not in-
terfere with the determination of the ratio of 3a/5a.
[3] P.Quadrelli, S.Romano, L.Toma, P.Caramella,
J. Org. Chem. 2003,
68, 6035–6038.
[4] A.G. Leach, E. Goldstein, K.N. Houk,
J. Am. Chem. Soc. 2003,
125, 8330–8339.
Determination of the enantiomeric excess of 5b: A sample of 5b (12 mg,
0.044 mmol) from entry 2 in Table 2 was dissolved in anhydrous toluene
(5 mL) and heated at 908C for 12 h.The solvent was evaporated and the
residue purified by column chromatography on silica gel (eluant: cyclo-
hexane/ethyl acetate (90:10)).The first fraction was (2 R,3R)-3b (6 mg,
50% yield), and HPLC analysis gave (2R,3R)-3b with >99.5% ee and
[a]2D0 =ꢀ95.3 (c=0.2 in CHCl3).The second fraction was (4 R,4aS,7aR)-5b
(5 mg), which was identical in every respect to the starting product,
whose enantiomeric excess was therefore >99.5% ee.
[5] J.Limanto, K.S.Khuong, K.N.Houk, M.L.Snapper,
J. Am. Chem.
Soc. 2003, 125, 16310–16321.
[6] D.A. Singleton, C. Hang, M.J. Szymanski, M.P. Meyer, A.G.
Leach, K.T.Kuwata, J.S.Chen, A.Greer, C.S.Foote, K.N.Houk,
J. Am. Chem. Soc. 2003, 125, 1319–1328.
[7] T.C. Dinadayalane, G.N. Sastry,
5533.
Organometallics 2003, 22, 5526–
[8] T.C.Dinadayalane, G.Gayatri, G.N.Sastry, J.Leszczynski,
J. Phys.
Chem. A 2005, 109, 9310–9323.
Determination of the enantiomeric excess of 3c: A sample of 3c (15mg,
0.045mmol) from entry 3 in Table 2 was dissolved in anhydrous toluene
[9] B.R.Ussing, C.Hang, D.A.Singleton,
7594–7607.
J. Am. Chem. Soc. 2006, 128,
9484
www.chemeurj.org
ꢁ 2007 Wiley-VCH Verlag GmbH & Co.KGaA, Weinheim
Chem. Eur. J. 2007, 13, 9478 – 9485

Diels–Alder and Hetero-Diels–Alder Reactions
FULL PAPER
[10] D.A.Evans, J.S.Johnson, J. Am. Chem. Soc. 1998, 120, 4895–4896.
be (4S,4aS,7aR)-2-methoxycarbonyl-4-(4-nitrophenyl)-4,4a,5,7a-tet-
[11] D.A. Evans, J.S. Johnson, E.J. Olhava,
J. Am. Chem. Soc. 2000,
rahydrocyclopenta[b]pyran.
122, 1635–1649.
[18] G.Desimoni, G.Faita, P.Quadrelli,
3154.
Chem. Rev. 2003, 103, 3119–
[12] S.Hanessian, P.Compain, Tetrahedron 2002, 58, 6521–6529.
[13] N.C¸ elebi-ÖlÅüm, D.H.Ess, V.Aviyente, K.N.Houk,
Soc. 2007, 129, 4528–4529.
J. Am. Chem.
[19] G.Desimoni, G.Faita, M.Guala, A.Laurenti, M.Mella,
Chem. Eur.
J. 2005, 11, 3816–3824.
[14] S.Kanemasa, Y.Oderaotoshi, S.Sakaguchi, H.Yamamoto, J.
Tanaka, E.Wada, D.P.Curran, J. Am. Chem. Soc. 1998, 120, 3074–
3088.
[20] G.Desimoni, G.Faita, F.Piccinini, M.Toscanini,
Eur. J. Org. Chem.
2006, 5228–5230.
[21] E.D.Secher, H.F.Ryder, J. Am. Chem. Soc. 1952, 74, 4392.
[22] H.Audrain, J.Thorhauge, RG. .Hazell, KA. .Jørgensen,
Chem. 2000, 65, 4487.
[15] J.Zhou, Y.Tang, Org. Biomol. Chem. 2004, 2, 429–433.
J. Org.
[16] G.Desimoni, G.Faita, M.Mella, F.Piccinici, M.Toscanini,
Eur. J.
Org. Chem. 2007, 1529–1534.
[23] M.Reimer, E.Chase, J. Am. Chem. Soc. 1938, 60, 2470.
[24] M.Reimer, E.Tobin, J. Am. Chem. Soc. 1940, 62, 2518.
[17] From the experiments described in entries 3 and 7 of Table 4, a very
small amount of a new product was isolated. 1H NMR (CDCl3,
[25] E.D.Stecher, F.Dunn, E.Gelblum,
J. Am. Chem. Soc. 1957, 79,
TMS, 300 MHz): d=8.23 (d, 3J
G
4748.
tons), 7.44 (d, 3J
A
[26] The H and 13C NMR spectra of 3–5b–d and some significant HPLC
chromatograms of the DA and HDA products obtained from the
thermal and ScIII-catalyzed reactions and the Claisen and retro-
Claisen rearrangements are given in the Supporting Information.
1
ACHTREUNG
H
ACHTREUNG
G
ACHTREUNG
AHCTREUNG
A
Received: June 29, 2007
Published online: September 14, 2007
Chem. Eur. J. 2007, 13, 9478 – 9485
ꢁ 2007 Wiley-VCH Verlag GmbH & Co.KGaA, Weinheim
www.chemeurj.org
9485