Another aspect of the reaction behavior of cyclopentadienones: 1,5-Sigmatropic rearrangement of the 1,4-addition products of amines

Article abstract of DOI:10.1002/ejoc.200300521

Treatment of 2,5-bis(methoxycarbonyl)-3,4-diphenylcyclopentadienone (1) with aniline (2a) gave the stereoisomeric 2-anilino-2,5-bis(methoxycarbonyl)-3, 4-diphenylcyclopentenones (cis- and trans-4a). ¹H NMR spectroscopic monitoring of the reaction indicates that both compounds are derived from the 1,5-sigmatropic rearrangement of the 1,4-addition products cis- and trans-3a. The structure of 3a was established by the X-ray analysis of the product after methylation (3a-Enol-Me) with diazomethane. The sequential pericyclic reaction behavior of 1 with amines is discussed on the basis of X-ray crystal structures and the optimized structure of the transition state at the B3LYP/6-31G(d) level of theory. Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim, Germany, 2004.

Full text of DOI:10.1002/ejoc.200300521

FULL PAPER
Another Aspect of the Reaction Behavior of Cyclopentadienones:
1,5-Sigmatropic Rearrangement of the 1,4-Addition Products of Amines
Koki Yamaguchi,[a] Chika Kai,[a] Yasuyuki Yoshitake,[a] and Kazunobu Harano*[a]
Keywords: Amines / Cyclopentadienones / Pericyclic reactions / Reaction mechanisms
Treatment of 2,5-bis(methoxycarbonyl)-3,4-diphenylcyclo-
pentadienone (1) with aniline (2a) gave the stereoisomeric
2-anilino-2,5-bis(methoxycarbonyl)-3,4-diphenylcyclo-
pentenones (cis- and trans-4a). 1H NMR spectroscopic
monitoring of the reaction indicates that both compounds are
derived from the 1,5-sigmatropic rearrangement of the 1,4-
addition products cis- and trans-3a. The structure of 3a was
established by the X-ray analysis of the product after methyl-
ation (3a-Enol-Me) with diazomethane. The sequential
pericyclic reaction behavior of 1 with amines is discussed on
the basis of X-ray crystal structures and the optimized struc-
ture of the transition state at the B3LYP/6−31G(d) level of
theory.
( Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2004)
Introduction
the progress of the reaction between 1 and 2a showed the
existence of an intermediary compound ascribable as the
1,4-addition product 3a, which transformed into the 1,5-
sigmatropic rearrangement product 4a.[3a]
Cyclopentadienones are powerful 4π-synthons with very
low LUMO (lowest unoccupied molecular orbital) energy
levels, although there are not many cyclopentadienones ex-
isting as monomers. 2,5-Bis(methoxycarbonyl)-3,4-di-
phenylcyclopentadienone (1) does exists in this state,
though, and the pericyclic reactions of 1 with various olef-
ins, including medium-sized ring polyenes, have been stud-
ied systematically.[1]
During the course of a study of the pericyclic reaction
behavior of cyclopentadienones,[2] we found novel cascade
reactions[3] in which prop-2-ynylamines added to 1 to give
two cyclization products in preference to the DielsϪAlder
(DA) reaction (Scheme 1). We considered compound C to
be the intramolecular ene reaction product of the 1,4-ad-
dition product (A), and compound D to have been formed
via the 1,5-sigmatropic rearrangement product (B) of A, fol-
lowed by an intramolecular DA reaction with the styrene
moiety as diene and dehydrogenation of the DA adduct.
On the basis of the structural features of C and D, we
thought that the 1,4-addition product (A) should be a com-
mon intermediate of both cascade reactions. In connection
with these reactions, in order to ascertain the addition reac-
tion behavior of amines toward the cyclopentadienone and
the migratory aptitude of the amino groups of the 1,4-ad-
dition products, we also studied the reaction between 1 and
The reactions between 1 and amines (2a, o-phenylenedia-
mine, cyclohexylamine, ethylamine, etc.) had been studied
by Eistert et al.[4] In the reaction between 1 and 2a, these
workers unknowingly isolated the 1,4-addition product (3a,
m.p. 109 °C dec.), which isomerizes to the 1,5-sigmatropic
rearrangement product 4a, and reported that 3a isomerized
to the enol form, the 5-anilino-2-hydroxycyclopentadiene
derivative 3a-Enol, the structure of which was determined
1
by H NMR spectroscopy through its methylation product
3a-Enol-Me with diazomethane (Scheme 2).
Our experimental result is not consistent with their ob-
servation in view of the sigmatropic rearrangement behavior
of the 1,4-addition product 3a.
These findings prompted us to reinvestigate the reactions
between 1 and 2a and related amines. The results for the
sequential reactions between cyclopentadienones and am-
ines are discussed here in detail on the basis of X-ray crys-
tallographic data and further additional data that we have
obtained.
Results and Discussion
We considered the overall reaction pathway for the reac-
tion between 1 and 2a in terms of frontier molecular orbital
(FMO) theory,[5] as a guiding principle for pericyclic reac-
tions. A possible reaction scheme is shown in Scheme 3.
There are two modes of the addition reaction: 1,2- (direct)
and 1,4- (conjugate) additions of 2a to 1. Both primary ad-
ducts satisfy the steric requirement of the thermally allowed
1,5-sigmatropic rearrangement of the substituents.
1
aniline (2a) as a model reaction. H NMR monitoring of
[a]
Graduate School of Pharmaceutical Sciences, Kumamoto
University,
5-1 Oe-hon-machi, Kumamoto 862-0973, Japan
Fax: (internat.) ϩ 81-96-371-4639
E-mail: harano@gpo.kumamoto-u.ac.jp
Supporting information for this article is available on the
WWW under http://www.eurjoc.org or from the author.
826
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/ejoc.200300521
Eur. J. Org. Chem. 2004, 826Ϫ834

Reaction Behavior of Cyclopentadienones
FULL PAPER
Scheme 2
Scheme 1
clopent-2-enone derivative, indicating that the 1,5-sigma-
tropic rearrangement of the anilino group had taken place
even at room temperature. The high-field shift of the methyl
1
As shown in Scheme 3, once 2a adds to 1, the anilino protons of the C5-CO2Me group in the H NMR spectrum
group probably migrates to a new position along the cyclo- is attributable to the ring-current effect of the cis-oriented
pentadiene π system to give a more stable isomer.
phenyl group (C4-Ph).
For clarification of the reaction pathway, we reexamined
The single crystal of the stereoisomeric 5-anilinocyclo-
the reaction between 1 and 2a. A mixture of 1 and 2a in pent-2-enone derivative cis-4a could be obtained under dif-
acetonitrile was allowed to stand at room temperature for ferent reaction conditions. The X-ray structure of cis-4a is
19 h to give a mixture of 1:1 adducts (trans-4a/cis-4a, 9:1).
depicted in Figure 2 (see also Table 1).
The NMR spectra of both isomers resemble each other
As can be seen in Figure 1 and 2, the major product is
very closely except for the chemical shifts of the meth- the trans isomer (in the spatial orientation of the 4-phenyl
oxycarbonyl groups. The methyl protons of a methoxycar- and 5-anilino groups), whereas the minor product has a cis
bonyl group of trans-4a showed a high-field shift [δ ϭ configuration. In trans-4a, the NH hydrogen of the anilino
3.38 ppm for CO2Me] in relation to the methyl group [δ ϭ group makes a hydrogen bond with the carbonyl oxygen of
˚
3.61 ppm for CO2Me] of cis-4a, implying that the methyl the CO2Me group (NϪH···OϭC 2.310 A). In the case of
protons of trans-4a are shielded by the electronic ring cur- cis-4a, the hydrogen bond is found between the PhNH and
1
rent of a neighboring phenyl ring. The H NMR and 13C the enone carbonyl group (NϪH···OϭC 2.276 A). The
˚
NMR spectroscopic data of the products are not sufficient cyclopentenone ring of cis-4a is considerably distorted, as
for determination of the substitution position of the anilino judged from the dihedral angle (C2ϪC1ϪC5ϪC4 18.9°).
group and its stereochemistry. To clarify the structure of This may be due to the presence of the CϪH···Oϭ type
the product, the single-crystal X-ray analysis was under- hydrogen-bond[6] interaction (2.372 A) between C4ϪH and
˚
taken. The crystal data and a computer-generated represen- the carbonyl oxygen of the C5-CO2Me group.
tation of trans-4a are shown in Table 1 and Figure 1, respec-
On the basis of these results, we tried to isolate the 1,4-
tively. As shown in Figure 1, the product is the 5-anilinocy- addition product 3a. However, we encountered difficulties
Eur. J. Org. Chem. 2004, 826Ϫ834
www.eurjoc.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
827

K. Yamaguchi, C. Kai, Y. Yoshitake, K. Harano
FULL PAPER
Careful monitoring of the progress of the reaction be-
tween 1 and 2a by 1H NMR spectroscopy revealed the over-
all reaction scheme. A plot of the concentrations of 1, cis-
3a, trans-3a, cis-4a, and trans-4a against time is depicted in
1
Figure 4, and the H NMR assignments of the compounds
are shown in Scheme 3.
The configurations of cis/trans-3a with respect to the spa-
tial relationship between the anilino and methoxycarbonyl
groups were determined by comparison of the chemical
shifts of the methyl protons of the CO2Me groups of both
isomers, the upfield shift being observed in trans-3a bearing
the cis-oriented CO2Me and phenyl groups.
At early stages of the reaction, the concentrations of
trans-3a, cis-3a, and cis-4a increased rapidly with the de-
crease in 1 until maxima were reached and then decreased
gradually over time with an increase in trans-4a. The con-
version of cis-4a into trans-4a is slower than the others. In-
spection of the reaction profile suggests that the 1,4 adduct
also involves the cis and trans isomers.
1
The H NMR spectral changes of the reaction mixture
also clarified the presence of the cis/trans isomerism in both
3a and 4a. The cis/trans isomerism of 3a is due to
ketoϪenol tautomerism and the isomerism of 4a presum-
ably originates from the 1,5-sigmatropic rearrangement of
the C4-hydrogen to the oxygen atom of the enone group,
followed by the reverse reaction. In the reaction, 2a may act
as an organic base to stabilize the enolate form essential for
the cis/trans isomerization of the adducts.
From the experimental results, it becomes clear that the
1,4-addition takes place in preference to the 1,2-addition,
in accordance with the known selectivity of amines toward
enones.[5] In the 1,5-sigmatropic rearrangement reaction,
cis-4a is considered to be the kinetically controlled product
and trans-4a the thermodynamically controlled product.
To confirm the 1,5-sigmatropic rearrangement, a molecu-
lar modeling study was carried out. Density functional the-
ory (DFT)[7] calculations at the B3LYP/6Ϫ31G(d) level
were performed through the use of the structures optimized
by the semiempirical PM3 calculations. The transition state
has one negative eigenvalue corresponding to the vibration
of the shifting of the anilino moiety from C4 to C5. The
transition state (TS1) was further verified by an intrinsic
reaction coordinate (IRC) calculation. The NϪC4 and
Scheme 3
in the isolation of 3a in the crystalline state. After some NϪC5 bond lengths of the three-membered aziridine moi-
˚
reexamination, we were able to isolate the 1,4 adduct trans- ety are 1.677 and 1.624 A, respectively (see Figure 5).
3a as crystals (m.p. 128 °C) under reaction conditions set up
The B3LYP/6Ϫ31G(d)-calculated transition structure for
so that the product was separated from the reaction mixture a simple model compound (1,4 adduct of cyclopen-
through the use of the least possible amount of solvent at a tadienone and 2a) showed the NϪC4 and NϪC5 bond
constant low temperature. Close inspection of the 1H NMR lengths to be 2.260 A, indicating that the NϪC4 and NϪC5
˚
spectrum of the residue of the filtrate indicated the presence bond lengths of TS1 are significantly shortened by the hy-
of the 1,5-sigmatropic rearrangement products cis-4a and drogen bond between the PhϪNH group and the ester car-
trans-4a. The structure of the 1,4-addition product was bonyl group, which can be expected from the X-ray struc-
˚
firmly established by X-ray analysis of the methylated prod- ture (PhNϪH···OϭCϽ 2.420 A) of 3a-Enol-Me (see Fig-
uct[4] 3a-Enol-Me, obtained from treatment with diazo- ure 3). The bond lengths of the transition structure for the
methane (see Figure 3).
2-carboxy derivative (1,4 adduct of 2-carboxycyclopentadi-
˚
A small amount of the methylated enol derivative 4a- enone and 2a) are 1.693 and 1.659 A, similar to those of
Enol-Me of the 1,5-sigmatropic rearrangement product 4a TS1. The calculated reaction barrier is 42.5 kcal/mol at
was also isolated from the methylation reaction mixture.
B3LYP/6Ϫ31G(d) level with ZPE correction, considerably
828
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2004, 826Ϫ834

Reaction Behavior of Cyclopentadienones
FULL PAPER
Table 1. Crystal data and intensity measurement for trans-4a, cis-4a, cis-4b, and 3a-Enol-Me
Compound
trans-4a
cis-4a
3a-Enol-Me
cis-4b
Formula
M.p. (C)
C27H23NO5
175
C27H23NO5
129
C28H25NO5
148
C29H27NO5
175
Molecular weight
Crystal System
Lattice Type
Lattice Parameters
441.48
triclinic
primitive
441.48
monoclinic
primitive
455.51
triclinic
primitive
469.54
monoclinic
primitive
˚
a (A)
12.624(3)
20.235(3)
10.668(1)
98.84(1)
94.29(1)
76.96(1)
17.737(2)
10.409(1)
12.867(2)
10.252(1)
24.778(4)
9.828(4)
99.34(1)
104.11(1)
93.43(1)
10.204(4)
15.145(3)
16.339(4)
˚
b (A)
˚
c (A)
α (°)
β (°)
106.13(1)
106.86(2)
γ (°)
¯
¯
Space Group
Z
P1 (#2)
P21/a (#14)
4
1.285
P1 (#2)
P21/n (#14)
4
1.291
4
4
Dcalcd. (g/cm3)
Dobsd.
1.110
1.150
1.273
1.280
1.287
1.301
Solvent
EtOH/EtOAc
EtOH/EtOAc
2θ Ͻ 55.0°
12581
12130
8889
EtOH/EtOAc
EtOH/EtOAc
Scan range
Reflections collected
Unique data collected
Unique data used
12684
12133
6363
11763
10920
4858
6088
5551
4108
IϾ3.0σ(I)
0.085
0.137
I Ͼ 2.0σ(I)
0.131
0.136
I Ͼ 3.0σ(I)
0.088
0.096
I Ͼ 3.0σ(I)
0.071
0.134
R
Rw
Figure 1. Computer-generated representation of the X-ray struc-
ture of trans-4a; the thermal ellipsoids are omitted for clarity
Figure 3. Computer-generated representation of the X-ray struc-
ture of 4a-Enol-Me; the thermal ellipsoids are omitted for clarity
higher than expected from the reaction conditions. To ob-
tain accurate energies, much higher levels of theory seem to
be required.
Interestingly, the reaction between 1 and 4-ethylaniline
(2b) gave the cis form (cis-4b) of the 1,5-sigmatropic re-
arrangement product of the 1,4 adduct. The crystal struc-
ture is depicted in Figure 6. X-ray analysis suggests that the
stability of the cis isomer may arise from the effective
NϪH···OMe hydrogen-bond interaction due to the elec-
tronic effect of the 4-ethyl group on the aniline ring.
Figure 2. Computer-generated representation of the X-ray struc-
ture of cis-4a; the thermal ellipsoids are omitted for clarity
Eur. J. Org. Chem. 2004, 826Ϫ834
www.eurjoc.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
829

K. Yamaguchi, C. Kai, Y. Yoshitake, K. Harano
FULL PAPER
Figure 6. Computer-generated representation of X-ray structure of
cis-4b; the thermal ellipsoids are omitted for clarity
Figure 4. 1H NMR spectroscopic monitoring of the reaction be-
tween 1 and 2a
2-Methylaniline (2c) gave both cis and trans adducts 4c.
The introduction of the methyl group at the 2-position of
aniline accelerates the decomposition reaction of the ad-
duct.
Aliphatic amines (propyl-, isopropyl-, butyl-, isobutyl-
and allylamines) and alcohols (methanol, ethanol) also
showed similar reaction behavior, to give the corresponding
1,4 adducts. However, the NMR spectral information could
not be obtained because of the facile dissociation to the
starting materials in the NMR solvents.
The instability of the 1,5-sigmatropic rearrangement
product is primarily the result of a structural feature of the
β-keto-α-amino acid (see Figure 7).
Figure 7. Enol forms of the 1,5-sigmatropic rearrangement prod-
ucts calculated by the DFT method at the B3LYP/6Ϫ31G(d) level
The mechanism of the elimination of amines or alcohols
is interpreted in terms of the three-system FMO interaction
based on the reactant dissection method.[8]
Figure 5. TS geometry of the 1,5-sigmatropic rearrangement of 3a
(enol form) calculated at the B3LYP/6Ϫ31G(d) level
Table 2. Energies of the GS and TS structures for the reaction between 1 and 2a
Compound
cis-3a
3a-Enol
trans-3a
TS1
PM3[a]
Ϫ85.9
Ϫ84.8[b]
Ϫ86.5
Ϫ1473.5785
Ϫ36.6[b]
B3LYP /6Ϫ31G(d)[c]
Ϫ1473.5781[f]
Ϫ1473.4880[f]
Ϫ1473.5781
Ϫ1473.5746 (Ϫ1473.1087)[d][e]
Ϫ1473.5030 (Ϫ1473.0409)[d,e]
Ϫ1473.4971[f]
Compound
trans-4a
4a-Enol
cis-4a
PM3
Ϫ87.7
Ϫ1473.5909
Ϫ84.9
Ϫ1473.5631
Ϫ87.7
Ϫ1473.5920
B3LYP /6Ϫ31G(d)
Ϫ1473.5925[f]
[b]
[c]
[d]
[a] ϩ Kcal/mol.
Reaction barrier 48.2 kcal/mol. Hartree.
E°, the thermal corrections to energy were scaled by a factor of 0.9804
(298.15°, 1.0 atm). [e] Reaction barrier 42.5 kcal/mol. [f] Values for conformational isomers due to rotation around the CϪCO2Me bonds.
830  2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.eurjoc.org
Eur. J. Org. Chem. 2004, 826Ϫ834

Reaction Behavior of Cyclopentadienones
FULL PAPER
An alternative reaction pathway for the elimination of showed broadening of the methoxy and phenyl signals due
aniline is possible: compound 4 might undergo a 1,5-sigma- to complex formation. Use of tertiary amines (e.g., triethyl-
tropic rearrangement to give the 1,2-addition product (i.e., amine) did not produce similar transient color changes with
the hemiaminal, 5).[9] However, we can rule out this reac- 1. The hydrogen bonding to the carbonyl groups of 1 lowers
tion pathway since we were unable to observe any 1H NHR the LUMO energy level (ΕLUMO Ϫ1.88 eV Ǟ Ϫ2.11 eV),
signals attributable to the symmetric structure of 5a.
resulting in an enhancement of the electron-accepting
The reaction between phencyclone (6) and allylamine ability favorable for donorϪacceptor interaction between
gave the 2-allylamino-3-cyclopentenone derivative 7, which addends. On the other hand, 2,5-diethyl-3,4-diphenylcyclo-
may be formed by the proposed reaction mechanism and pentadienone (8; ΕLUMO ϭ Ϫ1.11 eV) did not show color-
stabilized by the recovery of aromaticity in the phen- ation to give the DA adduct. These indicate that 1,4-ad-
anthrene ring (see Scheme 4).
dition of amines to 1 might be affected by charge-transfer
complex formation.[10] In the cascade reactions between 1
and unsaturated amines, the n-HOMO (lone pair electron)
of amines lying above the π-HOMO (NHOMO) of the un-
saturated bond acting as a 2π-source in the DA reaction
plays a crucial role interacting with the π-LUMO of 1, lead-
ing to the 1,4-addition reaction followed by the sequential
pericyclic reactions.
Conclusion
In conclusion, the reaction between 1 and 2a gave the
1,4-addition product, which then rearranges to give the 1,5-
sigmatropic rearrangement product. The structures of the
1,4 adduct and the 1,5-sigmatropic rearrangement products
were verified by single-crystal X-ray analysis. The detailed
1
analysis of H NMR monitoring revealed the existence of
the 1,5-sigmatropic rearrangement of the 1,4-addition prod-
uct. This investigation strongly supports the assumption
that the 1,4-addition product plays a key role as a common
intermediate in the cascade reactions between prop-2-ynyla-
mines and the cyclopentadienone. Electron-deficient cyclo-
pentadienones, in conjunction with the use of the 4π-syn-
thon, are potential enone components for the skeletal con-
struction of heterocycles.
On the basis of this study, we are currently investigating
the cascade reactions between cyclopentadienones and
prop-2-yn-1-ols in the presence of tertiary amines.
Experimental Section
General: Melting points were uncorrected. The IR spectra were
taken with a Hitachi 270Ϫ30 spectrophotometer. 1H NMR and 13
C
NMR spectra were taken with JEOL JNM-EX 270 (270 MHz),
JNM-AL 300 (300 MHz), and JNM-A 500 (500 MHz) spec-
trometers on ca. 10% solutions with TMS as an internal standard;
chemical shifts are expressed as δ values and the coupling constants
(J) are expressed in Hz. Mass spectra were obtained with JEOL
JMS-DX 303 and JEOL JMS-BU20 (GCmate) instruments. UV
spectra were recorded with a Shimadzu UV-2500PC spectrophoto-
meter.
Scheme 4
The reaction between 2,5-diethyl-3,4-diphenylcyclopen-
tadienone (8) and diallylamine, which had reacted with 1 to
afford the cascade reaction product,[3] did not give the 1,4-
addition product but instead gave the corresponding DA
adducts 9.
In the reactions between 1 and primary and secondary
amines, perceptible transient color changes (orange Ǟ
green) were observed in the reaction mixture at an early
Materials:
2,5-Bis(methoxycarbonyl)-3,4-diphenylcyclopenta-
dienone (1),[11a] phencyclone (6),[11b] and 2,5-diethyl-3,4-diphenyl-
cyclopentadienone (8)[11a] were prepared by the previously re-
ported method.
1
stage of the reaction. The H NMR spectrum of the reac-
Reaction between 1 and Aniline (2a): A mixture of 1 (500 mg,
tion mixture (e.g., 1 ϩ dipropylamine) at an early stage 1.4 mmol) and 2a (300 mg, 3.2 mmol) in acetonitrile (8 mL) was
Eur. J. Org. Chem. 2004, 826Ϫ834
www.eurjoc.org  2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 831

K. Yamaguchi, C. Kai, Y. Yoshitake, K. Harano
FULL PAPER
stirred at room temperature for 3 days. After evaporation of the
solvent, the residue was chromatographed over silica gel with a
CϪPh), 162.5, (COOCH3), 164.7, 166.8 (COOCH3) ppm. MS (EI,
70 eV): m/z (%) ϭ 445 (24) [Mϩ], 423 (100), 396 (34), 363 (97).
benzene/EtOAc mixture (20:1) as eluent to give the 1,5-sigmatropic C28H25NO5 (455.50), calcd. C 73.83, H 5.53, N 3.08; found C
1
rearrangement product trans-4a (581 mg, 90%) and cis-4a (72 mg,
10%).
73.61, H 5.47, N, 3.09. The H NMR spectroscopic data are essen-
tially identical to the previously reported data.[4] The structure was
established by the X-ray analysis.
Isomer trans-4a: Yellow prisms, m.p. 129 °C. IR (KBr): ν˜ ϭ 3368
(ϾNH), 1758, 1734, 1710 (CϭO) cmϪ1 1H NMR (CDCl3): δ ϭ
.
Compound 4a-Enol-Me: Orange oil. IR (KBr): ν˜ ϭ 3396 (ϾNH),
1734, 1666 (CϭO) cmϪ1 1H NMR (CDCl3): δ ϭ 3.47 [s, 3 H,
.
3.38 (s, 3 H, COOCH3), 3.90 (s, 3 H, COOCH3), 5.29 (s, 1 H,
methine), 5.40 (s, 1 H, NH), 6.59Ϫ7.40 (m, 15 H, ArH) ppm. 13C
NMR (CDCl3): δ ϭ 52.0, 52.5 (COOCH3 ϫ 2), 52.8 (ϾCHϪ), 75.0
(ϾC(NHϪPh)COOCH3), 142.6 (ϾCϭCϪCOOCH3), 164.4 (Cϭ
CϪPh), 167.8, 170.3, 196.0 (ϾCϭO) ppm. MS (EI, 70 eV): m/z
(%) ϭ 441 (62) [Mϩ], 382 (100), 350 (23), 247 (74). The stereochem-
istry was established by X-ray analysis.
C(NHϪPh)COOCH3), 3.67 (s, 3 H, CϪOCH3), 3.94 (s, 3 H, CO-
OCH3), 4.48 (s, 1 H, NH), 6.26Ϫ7.35 (m, 15 H, ArH) ppm. 13C
NMR
(CDCl3):
δ
ϭ
52.0 ppm
(ϪOCH3),
54.4
[C(NHϪPh)COOCH3), 63.3 (COOCH3), 71.5, 98.6, 164.3
[C(NHϪPh)COOCH3), 164.7 (COOCH3) ppm. MS (FAB): m/z
(%) ϭ 456 (46) [Mϩ ϩ 1], 455 (10) [Mϩ], 424 (53), 363 (100). The
previously reported spectroscopic data for cis-3a should be reas-
signed as for cis-4a. The formation of trans-3a was only recognized
in the 1H NMR spectrum of the reaction mixture by analysis of
data behavior in comparison with those of the related compounds.
The peak assignments are shown in Scheme 3.
Isomer cis-4a: Yellow crystal; m.p. 175 °C. IR (KBr): ν˜ ϭ 3368
1
(ϾNH), 1716 (br. CϭO) cmϪ1. H NMR (CDCl3): δ ϭ 3.62 ppm
(s, 3 H, COOCH3), 3.82 (s, 3 H, COOCH3), 4.93 (s, 1 H, NH),
5.45 (s, 1 H, methine), 6.42Ϫ7.41 (m, 15 H, ArH) ppm. 13C NMR
(CDCl3): δ ϭ 52.5, 53.9 ppm (COOCH3 ϫ 2), 59.0 (ϾCHϪ), 77.5
(ϾC(NHϪPh)COOCH3), 134.4 (ϾCϭCϪCOOCH3), 144.5 (Cϭ
CϪPh), 169.6, 178.3, 195.1 (ϾCϭO) ppm. MS (EI, 70 eV): m/z
(%) ϭ 441 (56) [Mϩ], 382 (100), 350 (18), 247 (80). The stereochem-
istry was established by X-ray analysis.
Adducts of 1- and 4-Ethylaniline (2b) and 2-Methylaniline (2c): The
following compounds were obtained by essentially the same pro-
cedure as above; cis-4b (410 mg, 61%) from 2b, and cis-4c (76 mg,
12%) and trans-4c (15 mg, 2.3%) from 2c.
Compound cis-4b: Orange crystals, m.p. 175 °C. IR (KBr): ν˜ ϭ
Isolation and Identification of 3a (Reinvestigation of the Previous
1712, 1730 (CϭO), 3392 (ϾNH) cmϪ1 1H NMR (CDCl3): δ ϭ
.
Reports):
A mixture of 1 (2.5 g, 7.0 mmol) and 2a (1.5 g,
1.12 ppm (t, J ϭ 7.5 and 7.7 Hz, 3 H, CH3), 2.47 (q, J ϭ 7.5 Hz,
2 H, CH2), 3.57 (s, 3 H, COOCH3), 3.78 (s, 3 H, COOCH3), 4.85
(s, 1 H, NH), 5.44 (s, 1 H, methine), 6.35Ϫ7.43 (m, 14 H, ArH)
ppm. 13C NMR (CDCl3): δ ϭ 15.6 ppm (ϪCH3), 27.7 (ϪCH2),
16.0 mmol) in acetonitrile (40 mL) was stirred at 5 °C for 24 h. The
precipitates were collected and washed with cold diethyl ether. The
product was dried in vacuo to give a yellow powder (trans-3a, 1.9 g,
60%). After the filtrate had been evaporated to dryness, the residue
was chromatographed over silica gel with a benzene/EtOAc mixture
(30:1) as eluent to give the 1,5-sigmatropic rearrangement products
cis-4a (70 mg, 2.2%) and trans-4a (202 mg, 6.4%).
52.3,
53.7 (COOCH3
ϫ
2),
59.0 (ϾCHϪ),
77.0
(ϾC(NHϪPh)COOCH3), 142.2 (ϾCϭCϪCOOCH3), 163.5 (Cϭ
CϪPh), 169.7, 178.1, 195.2 (ϾCϭO) ppm. MS (EI, 70 eV): m/z
(%) ϭ 469 (55) [Mϩ], 410 (100), 247 (68). C29H27NO5 (469.53):
calcd. C 74.18, H 5.80, N, 2.98; found C 73.89, H 5.75, N 2.92.
The stereochemistry was established by the X-ray analysis.
Compound trans-3a: Yellow powder; m.p. 128 °C. IR (KBr): ν˜ ϭ
3436 (ϾNH), 1750, 1732, 1704 (CϭO) cmϪ1 1H NMR (CDCl3):
.
δ ϭ 3.26 ppm (s, 3 H, COOCH3), 3.81 (s, 3 H, COOCH3), 4.46 (s,
1 H, methine), 4.67 (s, 1 H, NH), 6.70Ϫ7.54 (m, 15 H, ArH) ppm.
MS (FAB): m/z (%) ϭ 442 (24) [Mϩ ϩ 1], 441 (11) [Mϩ], 307 (37),
154 (100), 136 (57). C27H23NO5 (441.46): calcd. C 73.46, H 5.25,
N 3.17; found C 73.75, H 5.14, N 3.27.
Compound trans-4c: Yellow crystals, m.p. 145 °C. IR (KBr)): ν˜ ϭ
1752, 1728, 1700 (CϭO), 3424 (ϾNH) cmϪ1 1H NMR(CDCl3):
.
δ ϭ 2.00 (s, 3 H, CH3), 3.18 (s, 3 H, COOCH3), 3.65 (s, 3 H,
COOCH3), 4.32 (s, 1 H, NH), 4.62 (s, 1 H, methine), 6.92Ϫ7.35
(m, 14 H, ArH) ppm. 13C NMR (CDCl3): δ ϭ 15.6 ppm (ϪCH3),
1
When the H NMR sample solution was allowed to stand at room
50.5,
50.6 (COOCH3
ϫ
2),
65.2 (ϾCHϪ),
68.2
temperature, trans-3a gradually transformed into a mixture of 3a,
4a, and 1a.
(ϾC(NHϪPh)COOCH3), 140.1 (ϾCϭCϪCOOCH3), 160.9 (Cϭ
CϪPh), 164.9, 179.4, 193.6 (ϾCϭO) ppm. MS (EI, 70 eV): m/z
(%) ϭ 456 (36) [Mϩ ϩ 1], 391 (22), 349 (67), 257 (39), 107 (97), 44
(100). C28H25NO5 (455.50): calcd. C 73.83, H 5.53, N, 3.08; found
C 73.77, H 5.68, N 2.87.
Reaction between trans-3a and Diazomethane: A mixture of trans-
3a (500 mg, 1.1 mmol) in MeOH (3 mL) and diazomethane in di-
ethyl ether (40 mL; a solution of diazomethane in diethyl ether was
prepared from N-methyl-N-nitroso-p-toluenesulfonamide (4.3 g)
(by the established method) was stirred at 0 °C for 3 h. The precipi-
tated crystals were collected and washed with cold diethyl ether.
The product was dried under vacuum to give a yellow powder (3a-
Enol-Me, 350 mg, 68%). After the filtrate had been evaporated to
dryness, the residue was chromatographed over silica gel with a
benzene/EtOAc mixture (3:1) as eluent to give the 1,5-sigmatropic
rearrangement product 4a-Enol-Me (14 mg, 2.7%).
Compound cis-4c: Orange crystals, m.p. 165 °C. IR (KBr)): ν˜ ϭ
1724 (br. due to three carbonyls), 3440 (ϾNH) cmϪ1
.
1H
NMR(CDCl3): δ ϭ 1.91 ppm (s, 3 H, CH3), 3.60 (s, 3 H, CO-
OCH3), 3.79 (s, 3 H, COOCH3), 5.47 (s, 1 H, methine), 4.80 (s, 1
H, NH), 7.30Ϫ7.42 (m, 14 H, ArH) ppm. 13C NMR (CDCl3): δ ϭ
17.0 ppm (ϪCH3), 52.3, 53.8 (COOCH3 ϫ 2), 58.7 (ϾCHϪ), 77.3
(ϾC(NHϪPh)COOCH3), 142.3 (ϾCϭCϪCOOCH3), 163.4 (Cϭ
CϪPh), 169.8, 170.9, 195.4 (ϾCϭO) ppm. MS (EI, 70 eV): m/z
(%) ϭ 455 (56) [Mϩ], 396 (100), 247 (88), 91 (47). C28H25NO5
(455.50): calcd. C 73.83, H 5.53, N 3.08; found C 74.08, H 5.75,
N 2.99.
Compound 3a-Enol-Me: Yellow prisms, m.p. 148 °C. IR (KBr): ν˜ ϭ
3392, 3356 (ϾNH), 1724, 1676 (CϭO) cmϪ1 1H NMR (CDCl3):
.
δ ϭ 3.53 ppm (s, 3 H, COOCH3), 3.68 (s, 3 H, OCH3), 4.09 (s, 3
H, COOCH3), 4.74 (s, 1 H, NH), 6.79Ϫ7.36 (m, 15 H, ArH) ppm. Reaction between Phencyclone (6) and Allylamine: The adduct was
13C NMR (CDCl3): δ ϭ 51.1 ppm (COOCH3), 52.1 (C2-OCH3),
obtained by the method described above, from phencyclone (3.8 g,
62.4 (COOCH3), 75.8 [ϾC(NHϪPh)Ph], 113.2, 130.3, 160.0 (Cϭ 10 mmol) and allylamine (1.1 g, 20 mmol) in 1,4-dioxane (10 mL).
832
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2004, 826Ϫ834

Reaction Behavior of Cyclopentadienones
FULL PAPER
The precipitates were washed with diethyl ether to give the adduct
reaction products, the chemical shifts of which are listed in
Scheme 3. A graphical representation of the changes of the concen-
7 (2.4 g, yield 55%). 7: Colorless crystals, m.p. 210Ϫ212 °C. IR
1
(KBr): ν˜ ϭ 1752 (CϭO) cmϪ1. H NMR (500 MHz, CDCl3): δ ϭ trations is depicted in Figure 4.
2.86 ppm (dd, J ϭ 12.8 and 6.7 Hz, 1 H, ϾCH-H), 2.99 (dd, J ϭ
X-ray Crystallographic Study: Single crystals of the compound cis-
4.9 and 12.8 Hz, 1 H, ϾCHϪH), 4.98 (dd, J ϭ 10.4, 1 H, ϭ
CHHϪ), 5.09 (dd, J ϭ 17.1, 1 H, ϭCHH), 5.84 (dd, J ϭ 10.4 and
17.1 Hz, 1 H, -CHϭCH2), 5.25 [s, 1 H, CH(Ph)], 7.13Ϫ7.67 (m, 16
H, aromatic H), 8.74 (d, J ϭ 8.6 Hz, 1 H, Ha), 8.76 (d, J ϭ 8.6 Hz,
1 H, Ha) ppm. 13C NMR (125 MHz CDCl3): δ ϭ 47.8 ppm (ϪCH2
CϭCH2), 56.5 [ϪCH(Ph)], 76.1 [ϪC(NH)Ph], 116.0 (CHϭCH2),
122.9, 123.4, 125.9, 126.2, 126.8, 127.2, 127.4, 127.6, 128.1, 128.6,
128.8 (Ar CH), 128.2, 128.4, 131.1, 131.7, 135.0, 136.5, 136.6, 140.2
(Ar-C), 215.2 (CϭO) ppm. MS (EI, 70 eV): m/z (%) ϭ 411 (18)
[Mϩ Ϫ CO], 382 (92), 370 (100), 354 (80). C32H25NO (439.19),
calcd. C 87.44, H 5.73, N 3.19; found C 87.54, H 5.67, N 3.24.
4a were prepared by slow evaporation of an ethanol/ethyl acetate
solution at room temperature. The cell constants were found by a
least-squares procedure with use of the values of the Bragg angles
of 20 reflections. Systematic absences of reflections indicate the
space group to be P21/a (No. 14). All measurements were made
on a Rigaku AFC-7 four-circle autodiffractometer with graphite
monochromated Mo-Kα radiation. The reflection data were col-
lected by the ω-2θ scan technique to a maximum 2θ value of 55.0°.
Of the unique reflections measured, reflections with Io Ͼ 2.00σ(I)
were used. The structures were solved by direct methods.[12] The
non-hydrogen atoms were refined anisotropically and the hydrogen
atoms were refined isotropically. Some methoxycarbonyl group hy-
drogens were located on calculated positions and refined. Full-ma-
trix, least square refinement was used and the unweighted (R) and
weighted agreement factors (Rw) are given.
Reaction between 2,5-Diethyl-3,4-diphenylcyclopentadienone (8) and
Diallylamine: A solution of 8 (500 mg, 1.7 mmol) and diallylamine
(500 mg, 5.2 mmol) in benzene (2 mL) was heated at reflux for 8 h.
After evaporation of benzene, the residue was treated with Ac2O
(0.6 mL) in pyridine (2 mL). The reaction mixture was poured onto
ice water. The oil was extracted with diethyl ether. The diethyl ether
layer was washed with dil. HCl and then with water. After evapor-
ation of diethyl ether, the residue was purified by chromatography
on silica gel to give the DA adducts as the acetyl derivatives [endo-
9 (193 mg, 26%) and exo-9 (82 mg, 11%)].
Neutral atom scattering factors were taken from the International
Tables for X-ray Crystallography.[13] All calculations were per-
formed on a Silicon Graphics O2 workstation with teXsan Crystal
Structure Analysis Package.[14]
The crystal structures of trans-4a, 3a-Enol-Me, and cis-4b were
solved similarly. The numbering schemes used in this paper and
ORTEP representations are listed in the Supporting Information
(see also the footnote on the first page of this article). CCDC-
212296 to -212299 for trans-4a, cis-4a, 3a-Enol-Me, and cis-4b con-
tain the supplementary crystallographic data for this paper. These
data can be obtained free of charge at www.ccdc.cam.ac.uk/conts/
retrieving.html [or from the Cambridge Crystallographic Data
Centre, 12 Union Road, Cambridge CB2 1EZ, UK; Fax: (internat.)
ϩ 44-1223-336-033; E-mail: deposit@ccdc.cam.ac.uk].
Isomer endo-9: Colorless crystals, m.p. 110Ϫ113 °C. IR (KBr): ν˜ ϭ
1770 (bridge CϭO), 1646 (ϾCϭO) cmϪ1 1H NMR (500 MHz,
.
CDCl3): δ ϭ 0.74 ppm (t, J ϭ 7.3 Hz, 3 H, ϪCH2ϪCH3), 0.95 (t,
J ϭ 7.3 Hz, 3 H, ϪCH2ϪCH3), 1.64 Ϫ1.72 (m, 3 H, 2-Hβ and
ϪCH2ϪCH3), 1.81Ϫ1.89 (m, 1 H, ϪCHHϪCH3), 2.09 (s, 3 H,
COϪCH3), 2.06Ϫ2.14 (m, 1 H, 2-Hα), 2.73Ϫ2.77 (m, 1 H, 3-H),
2.99 (dd, J ϭ 4.3 and 12.8 Hz, 1 H, ϾCHϪCHϪN), 3.77 (dd, J ϭ
17.7 and 4.9 Hz, 1 H, NϪCHHϪCHϭ), 3.90 (dd, J ϭ 17.7 and
4.9 Hz, 1 H, N CHH,ϪCHϭ), 4.02 (dd, J ϭ 11.0 and 12.8 Hz, 1
H, ϾCHϪCHHϪN), 5. 12 (d, J ϭ 17.7 Hz, 1 H, ϪCHϭCHH),
5.20 (d, J ϭ 10.4 Hz, 1 H, ϪCHϭCHH), 5.68Ϫ5.75 (m, 1 H,
ϪCHϭCH2), 7.32Ϫ6.98 (m, 10 H, Ar-H) ppm. 13C NMR
(100 MHz, CDCl3): δ ϭ 9.0, 9.2, 18.9, 19.4, 21.5, 32.6, 36.5, 47.7,
51.2, 56.5, 58.9, 116.8, 132.7, 140.5, 145.1, 171.5, 206.0 ppm. MS
(EI, 70 eV): m/z (%) ϭ 427 [Mϩ]. C29H33NO2 (427.58): calcd. C
81.46, H 7.78, N 3.28; found C 81.34, H 7.71, N 3.34.
Computational Details: Semiempirical MO calculations were run
through the CS Chem3D Pro interface with the aid of MOPAC97
on a Macintosh G4 personal computer (dual CPU). The ground-
state (GS) structures were optimized by the EF routine im-
plemented in the MOPAC program packages by use of the MNDO-
PM3 (PM3) approximation. The transition states (TS) were located
by the transition state location routine. The fully optimized geo-
metries calculated by the PM3 method were used as starting geo-
metries for the ab initio and DFT calculations.
Isomer exo-9: Colorless crystals, m.p. 133Ϫ136 °C. IR (KBr): ν˜ ϭ
1752 (bridge CϭO), 1652 (ϾCϭO) cmϪ1 1H NMR (500 MHz,
.
DFT calculations were carried out at the B3LYP/6Ϫ31G(d) level
of theory by use of the Linda Gaussian 98 package[7] on a PC
(Pentium 4, 3.0 MHz) cluster of four processors. Geometry optimi-
zations were undertaken with default convergence limits. Transition
structures were located by use of the TS keyword. Vibrational fre-
quencies were used to characterize the nature of the stationary
points and to obtain thermodynamic parameters. Zero-point en-
ergy (ZPE) corrections were scaled by 0.9804.[15] Graphical analysis
of the MO calculation data was performed on a Macintosh G4
personal computer.
CDCl3): δ ϭ 0.58 ppm (t, J ϭ 7.3 Hz, 3 H, ϪCH2ϪCH3), 0.77 (t,
J ϭ 7.3 Hz, 3 H, ϪCH2ϪCH3), 1.63Ϫ1.69 (m, 2 H, 2-Hβ and
ϪCHHϪCH3), 1.76Ϫ1.88 (m, 4 H, 2-Hα and ϪCH2CH3), 2.11 (s,
3 H, ϪCOCH3), 2.67Ϫ2.71 (m, l H, 3-H), 3.30 (t, J ϭ 12.8 Hz, 1
H, ϾCHϪCHHϪN), 3.52 (dd, J ϭ 4.3 and 12.8 Hz, 1 H,
ϾCHϪCHHϪN), 3.93 (dd,
NϪCHHϪCHϭ), 4.02 (dd,
J
J
ϭ
17.7 and 4.9 Hz,
17.7 and 4.9 Hz,
1
1
H,
H,
ϭ
NϪCHHϪCHϭ), 5.18 (d, J ϭ 17.1 Hz, 1 H, ϪCHϭCHH), 5.25
(d, J ϭ 10.4 Hz, 1 H, CHϭCHH), 5.78Ϫ5.83 (m, l H, ϪCHϭ
CH2), 7.01Ϫ7.28 (m, 10 H, Ar-H) ppm. 13C NMR (100 MHz,
CDCl3): δ ϭ 9.1, 16.6, 19.6, 21.9, 33.3, 38.7, 47.9, 52.3, 55.7, 58.9,
116.9, 135.5, 142.8, 144.7, 170.3, 207.9 ppm. MS (EI, 70 eV): m/z
(%) ϭ 427 [Mϩ]. C29H33NO2 (427.58): calcd. C 81.46, H 7.78, N
3.28; found C 81.19, H 7.82, N 3.45.
The DFT calculation data (Cartesian coordinates) are also avail-
able from our laboratory web site (http://yakko.pharm.kumamoto-
u.ac.jp/).
1H NMR Monitoring of the Reaction between 1 and 2a: A solution
of 1 (500 mg) and 2a (300 mg) in MeCN (8.0 mL) was prepared.
At a given temperature, the ensuing reaction was monitored by
Acknowledgments
We thanks Miss T. Akiyama and Miss M. Yamanami for exper-
imental assistance.
1
analysis of the H NMR signals of the OCH3 groups of 1 and the
Eur. J. Org. Chem. 2004, 826Ϫ834
www.eurjoc.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
833

K. Yamaguchi, C. Kai, Y. Yoshitake, K. Harano
FULL PAPER
Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, V. G.
Zakrzewski, J. A. Montgomery, Jr., R. E. Stratmann, J. C. Bu-
rant, S. Dapprich, J. M. Millam, A. D. Daniels, K. N. Kudin,
M. C. Strain, O. Farkas, J. Tomasi, V. Barone, M. Cossi, R.
Cammi, B. Mennucci, C. Pomelli, C. Adamo, S. Clifford, J.
Ochterski, G. A. Petersson, P. Y. Ayala, Q. Cui, K. Morokuma,
D. K. Malick, A. D. Rabuck, K. Raghavachari, J. B. Foresman,
J. Cioslowski, J. V. Ortiz, A. G. Baboul, B. B. Stefanov, G. Liu,
A. Liashenko, P. Piskorz, I. Komaromi, R. Gomperts, R. L.
Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A.
Nanayakkara, C. Gonzalez, M. Challacombe, P. M. W. Gill, B.
Johnson, W. Chen, M. W. Wong, J. L. Andres, C. Gonzalez,
M. Head-Gordon, E. S. Replogle, J. A. Pople, Gaussian, Inc.,
Pittsburgh PA, 1998.
[1]
[1a] K. Harano, M. Yasuda, K. Kanematsu, J. Org. Chem. 1982,
[1b]
47, 3736Ϫ3743.
M. Yasuda, K. Harano, K. Kanematsu, J.
Am. Chem. Soc. 1981, 103, 3120Ϫ3126. [1c] M. Yasuda, K. Har-
[1d]
ano, K. Kanematsu, J. Org. Chem. 1981, 46, 3836Ϫ3841.
M. Mori, A. Hayamizu, K. Kanematsu, J. Chem. Soc., Perkin
[1e]
Trans. 1 1981, 1259Ϫ1272.
M. Yasuda, K. Harano, K.
[1f]
Kanematsu, J. Org. Chem. 1980, 45, 2368Ϫ2372.
K. Har-
ano, M. Yasuda, T. Ban, K. Kanematsu, J. Org. Chem. 1980,
[1g]
45, 4455Ϫ4462.
T. Ban, Y. Wakita, K. Kanematsu, J. Am.
[1h]
Chem. Soc. 1980, 102, 5415Ϫ5416.
K. N. Houk, R. B.
Woodward, J. Am. Chem. Soc. 1970, 92, 4143Ϫ4145; K. N.
Houk, R. B. Woodward, J. Am. Chem. Soc. 1970, 92,
4145Ϫ4147.
[2] [2a]
Y. Yoshitake, H. Nakagawa, M. Eto, K. Harano, Tetra-
[8]
[9]
[8a] S. Inagaki, H. Fujimoto, K. Fukui, J. Am. Chem. Soc. 1976,
[2b]
hedron Lett. 2000, 41, 4395Ϫ4400.
Harano, Tetrahedron 1999, 55, 6051Ϫ6066.
Eto, K. Harano, J. Chem. Soc., Perkin Trans.
T. Jikyo, M. Eto, K.
[8b]
98, 4693Ϫ4701.
S. Inagaki, Y. Hirabayashi, J. Am. Chem.
[2c]
T. Jikyo, M.
1998,
T. Jikyo, M. Eto, K. Harano, Chem. Pharm.
Soc. 1977, 99, 7418Ϫ7423.
1
The LUMO coefficients of 1 indicate that the largest coefficient
is located at the C3 position, favorably for the 1,4-addition re-
action. If the 1,2-addition product is formed, the initial N-phe-
nyl-substituted hemiaminal may lose water to give the stable
Schiff base.
[2d]
3463Ϫ3470.
[2e]
Bull. 1997, 45, 1961Ϫ1969.
M. Eto, T. Aoki, K. Harano,
Tetrahedron 1994, 50, 13395Ϫ13408. [2f] M. Eto, K. Harano, T.
[2g]
Oya, T. Hisano, Chem. Pharm. Bull. 1993, 41, 97Ϫ107.
Harano, K. Uchida, M. Izuma, T. Aoki, M. Eto, T. Hisano,
K.
[10]
The AM1-calculated FMO energy levels (eV) are as follows: 1,
LUMO Ϫ1.88, HOMO Ϫ9.65; 6, LUMO Ϫ1.60, HOMO
Ϫ8.42; 8, LUMO Ϫ1.11, HOMO Ϫ8.90.
Chem. Pharm. Bull. 1988, 36, 2312Ϫ2322.
[3] [3a]
Y. Yoshitake, K. Yamaguchi, C. Kai, T. Akiyama, C.
Handa, T. Jikyo, K. Harano, J. Org. Chem. 2001, 66,
[11] [11a]
[11b]
White, D. M, J. Org. Chem. 1974, 39, 1951Ϫ1952.
W.
[3b]
8902Ϫ8911.
K. Harano, M. Eto, T. Oya, S. Nakamoto, T.
R. Forsyth, G. A. Weisenburger, K. W. Field, Transactions of
the Illinois State Academy of Science, vol. 89, 1 and 2, pp.
37Ϫ40 (1996).
Hisano, Tetrahedron Lett. 1992, 43, 6473Ϫ6476.
B. Eistert, A. J. Thommen, Chem. Ber. 1971, 104, 3048.
T. J. Arackal, B. Eistert, Chem. Ber. 1973, 106, 3788.
[4] [4a]
[4b]
[12] [12a]
SIR92: A. Altomare, M. C. Burla, M. Camalli, M. Cas-
[5]
I. Fleming, Frontier Orbitals and Organic Chemical Reactions,
Wiley, London, 1976, p. 86Ϫ181.
carano, C. Giacovazzo, A. Guagliardi, G. Polidori, J. Appl.
[12b]
Cryst. 1994, 27, 435.
SAPI91: F. Hai-Fu, Structure Analy-
[6] [6a]
R. Taylor, O. Kennard, J. Am. Chem. Soc. 1982, 104,
sis Programs with Intelligent Control, Rigaku Corporation, To-
kyo, Japan, 1991.
D. T. Cromer, J. T. Waber, in International Tables for X-ray
Crystallography, vol. IV, The Kynoch Press, Birmingham, Eng-
land, Table 2.2A, 1974.
Crystal Structure Analysis Package, Molecular Structure Cor-
poration, 1985 and 1992.
A. P. Scott, L. Radom, J. Phys. Chem. 1996, 100,
[6b]
5063Ϫ5070.
G. R. Desiraju, Acc. Chem. Res. 1991, 24,
290Ϫ296. [6c] T. Steiner, J. A. Kanters Kroon, Chem. Commun.
1996, 1277Ϫ1278.
[13]
[6d]
[6e]
Y. Gu, T. Kar, S. Scheiner, J. Am. Chem. Soc. 1999,
9411Ϫ9422.
G. R. Desiraju, T. Steiner, The Weak Hydrogen Bond in Struc-
tural Chemistry and Biology, Oxford University Press, Ox-
ford, 1999.
[14]
[15]
10502Ϫ10513.
[7]
Gaussian 98, Revision A.7, M. J. Frisch, G. W. Trucks, H. B.
Received August 15, 2003
834
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2004, 826Ϫ834