Selectivities in the 1,3-dipolar cycloaddition of nitrile oxides to dicyclopentadiene and its derivatives

Article abstract of DOI:10.1016/j.tet.2003.12.026

The 1,3-dipolar cycloaddition of nitrile oxides, generated from aldoximes and nitroalkanes, to dicyclopentadiene proceeds with complete chemo- and stereoselectivity. The approach of the dipole takes place exclusively from the exo-face of the bicycloheptane moiety providing a mixture of regioisomers in approximately 55:45 ratio. On the other hand, nitrile oxide cycloaddition to dimethyldicyclopentadiene dicarboxylate (Thiele's ester), besides exhibiting chemo- and stereoselectivity as in the case of dicyclopentadiene, exhibits complete regioselectivity as well providing a single isomer in good yield. The Influence of remote substituents, including sterically 'sterile' ones, on the regioselectivity has also been investigated using 8-hydroxy and 1-keto derivatives of dicyclopentadiene. These experimental observations have been investigated through gas phase and solvent model MO calculations on the transition state geometries at semiempirical (PM3) and hybrid ab initio-DFT levels of theory. The Computational methods employed in this study were rigorously tested by performing model calculations on well-established experimental observations.

Full text of DOI:10.1016/j.tet.2003.12.026

Tetrahedron 60 (2004) 1453–1462
Selectivities in the 1,3-dipolar cycloaddition of nitrile oxides to
dicyclopentadiene and its derivatives
a
b
c
Irishi N. N. Namboothiri,a Namrata Rastogi, Bishwajit Ganguly, Shaikh M. Mobin
,
,
*
*
and Miriam Cojocarud
aDepartment of Chemistry, Indian Institute of Technology, Bombay, Mumbai 400 076, India
bAnalytical Science Division, Central Salt and Marine Chemicals Research Institute, Bhavnagar Gujarat 364 002, India
cNational Single Crystal X-ray Diffraction Facility, Indian Institute of Technology, Bombay, Mumbai 400 076, India
dDepartment of Chemistry, Bar-Ilan University, Ramat Gan 52900, Israel
Received 6 August 2003; revised 24 November 2003; accepted 11 December 2003
Dedicated to Professor S. N. Balasubrahmanyam on the occasion of his 72nd birthday
Abstract—The 1,3-dipolar cycloaddition of nitrile oxides, generated from aldoximes and nitroalkanes, to dicyclopentadiene proceeds with
complete chemo- and stereoselectivity. The approach of the dipole takes place exclusively from the exo-face of the bicycloheptane moiety
providing a mixture of regioisomers in approximately 55:45 ratio. On the other hand, nitrile oxide cycloaddition to dimethyldicyclopenta-
diene dicarboxylate (Thiele’s ester), besides exhibiting chemo- and stereoselectivity as in the case of dicyclopentadiene, exhibits complete
regioselectivity as well providing a single isomer in good yield. The Influence of remote substituents, including sterically ‘sterile’ ones, on
the regioselectivity has also been investigated using 8-hydroxy and 1-keto derivatives of dicyclopentadiene. These experimental observations
have been investigated through gas phase and solvent model MO calculations on the transition state geometries at semiempirical (PM3) and
hybrid ab initio-DFT levels of theory. The Computational methods employed in this study were rigorously tested by performing model
calculations on well-established experimental observations.
q 2003 Elsevier Ltd. All rights reserved.
1. Introduction
unsymmetrical acyclic6,7 and simple cyclic8 dipolarophiles
2 has been investigated experimentally6,8 and theoreti-
cally.7,8 For instance, addition of nitrile oxide 1 to methyl
acrylate 2 (X¼CO2Me) provides 5-isoxazoline 3a in
overwhelming predominance, compared to its regioisomer,
4-isoxazoline 3b (ratio ,95:5).6,7 As for stereochemistry, p
face selection9,10 in the addition of nitrile oxides to various
dipolarophiles such as norbornene,11 2,3-dioxabicy-
clo[2.2.2]octane,12 cis-3,4-dichlorocyclobutene,13 a-chiral
alkenes14 etc. has been investigated.15 In the case of
norbornene 4 (X¼H), the approach of the dipole preferen-
tially takes place from the exo face.11 Further, in the case of
unsymmetrically substituted norbornenes 4 (X¼hexyl,
SiMe3, CO2Et) formation of single stereo- and regioisomers
has been observed, i.e. the exo isomer 5 in which oxygen of
the dipole is attached to the more substituted center of the
dipolarophile.16 A handful of other reports on the inter-13
and intramolecular14 cycloadditions of nitrile oxides to
bicyclic systems, viz. norbornenes17,18 and norborna-
dienes18,19,20 also reflected this feature. However, in the
intermolecular reactions of norbornadienes18,19 and in
presence of sterically demanding groups on the exo face
of norbornenes,17,18 formation of considerable amount of
endo isomers is observed.
1,3-Dipolar cycloadditions offer convenient one-step
routes for the construction of a variety of five-membered
heterocycles.1,2 In particular, cycloaddition of nitrile
oxides to olefins are of considerable interest as the
resulting isoxazolines are versatile intermediates in the
synthesis of a variety of natural products.3,4 Recently,
isoxazolines fused to bicyclic frameworks have been
subjected to molybdenum mediated N–O bond cleavage
to afford stereoselectively substituted cyclopentane rings.5
Achievement of a high degree of selectivity is, therefore,
of paramount importance for further expanding the scope
and exploiting the potential of this elegant synthetic
strategy.
Regioselectivity in the addition of nitrile oxides 1 to
Keywords: Nitrile oxide; 1,3-Dipolar cycloaddition; Dicyclopentadiene;
Thiele’s ester.
*
Corresponding authors. Tel.: þ91-22-2576-7196; fax: þ91-22-2576-
7152 (I.N.N.N.); Tel.: þ91-278-256-7760 ext. 750; fax: þ91-278-256-
7562 (B.G.);
e-mail addresses: irishi@iitb.ac.in; gang_12@rediffmail.com
0040–4020/$ - see front matter q 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2003.12.026

1454
I. N. N. Namboothiri et al. / Tetrahedron 60 (2004) 1453–1462
of formation of 8 monocycloadducts and 16 biscycloadducts
taking the total number of products expected to 24.
The dipoles, acetonitrile oxide 1a and benzonitrile oxide 1b,
were generated from two different precursors, viz. aldox-
imes28 (Scheme 1, path A) and nitroalkanes (Scheme 1, path
B),29,30 in order to probe whether or not the selectivities are
influenced by the method by which the dipole is generated
1
(Tables 1 and 2). The H and 13C NMR spectra of the
products formed from dicyclopentadiene 6a indicated the
formation of mixtures of two monocycloadducts 7a/8a
and 7b/8b, respectively, in ,55:45 ratio (Table 1).31
In contrast to the behavior of dicyclopentadiene 6a, the
reaction of its dicarboxylate, Thiele’s ester, 6b with
acetonitrile oxide 1a and benzonitrile oxide 1b provided
single monocycloadducts 7c and 7d, respectively
(Table 2).32
Although nitrile oxide cycloadditions to dienes (e.g.,
norbornadienes, vide supra) and polyenes (e.g., fulvene)21
have been reported in the literature, to our knowledge,
selectivities in the cycloaddition of nitrile oxides to systems
possessing multiple p-faces has not been investigated.22,23
Herein, we report the remarkable selectivities observed in
the cycloaddition of nitrile oxides to representative tricyclic
systems.
Further examination of Tables 1 and 2 indicates that the
isomer ratios of the cycloadducts formed in the nitrile oxide
cycloaddition to dicyclopentadiene 6a and its dicarboxylate
6b are independent of the method of generation of nitrile
oxide 133 However, it has been observed that in our hands
path B (nitroalkane, BOC2O, DMAP, THF; Scheme 1, see
Section 5)30 is superior for the generation of acetonitrile
oxide 1a (Table 1, entry 2 and Table 2, entry 2) over path A,
method A (aldoxime, NCS, Et3N, CH2Cl2; see also Table 1,
entry 1 and Table 2, entry 1) and path A, method B
(aldoxime, NaOCl, Et3N, CH2Cl2). On the other hand, path
A, method B (aldoxime, NaOCl, Et3N, CH2Cl2) turned out
to be better for the generation of benzonitrile oxide 1b
(Table 1, entry 3 and Table 2, entry 3) as compared to path B
(Table 1, entry 4 and Table 2, entry 4).
2. Results and discussion
Dicyclopentadiene 6a and its derivatives, viz. dicarbox-
ylate, Thiele’s ester, 6b,24 alcohol 6c25 and enone 6d25,26
were chosen as the dipolarophiles as three degrees of
differentiation, viz. chemo, stereo and regio, in their
cycloaddition with suitable dipoles such as nitrile oxides
would be possible (Fig. 1). Although norbornene has been
shown to be much more reactive than cyclopentene in the
cycloaddition with nitrile oxides (vide infra),27 the dipole
could, in principle, react with the C2–C3 double bond or
C5–C6 double bond exhibiting chemoselectivity. The
approach of the dipole could take place from the bb-face,
ba-face, ab-face or aa-face. Furthermore, formation of
two regioisomers in which the dipole oxygen is bonded to
C2 or C3/C5 or C6 is also a possibility for every
cycloaddition. Therefore, there is a statistical possibility
In view of the above, only 1b, generated via path A, method
B, has been employed for the subsequent cycloaddition with
alcohol 6c and enone 6d. A mixture of isomers 7e/8e,
similar to that observed in the case of dicyclopentadiene
6a, has been isolated when syn-alcohol 6c was reacted
with benzonitrile oxide 1b (Table 3, entry 1). Finally, the
enone 6d, when treated with benzonitrile oxide 1b, provided
a mixture of isomers 7f and 8f in 34:66 ratio (Table 3,
entry 2).
Figure 1. Chemo-, stereo- and regioselectivity in the cycloaddition of nitrile oxides 1 to dicyclopentadiene moiety 6.

I. N. N. Namboothiri et al. / Tetrahedron 60 (2004) 1453–1462
1455
Scheme 1. 1,3-Dipolar cycloaddition of nitrile oxides 1 to dicyclopentadiene 6a and its derivatives 6b–d.
1
1
The H and 13C NMR spectra of all the products revealed
the preferential reactivity of the bicycloheptenyl (C5–C6)
the major isomer 7a of 7a/8a pair, the H NMR chemical
shift values for the key resonances viz. the two endo
hydrogens and the Me are d 4.36, 2.97 and 1.80,
respectively. The corresponding values for the minor isomer
8a are d 4.28, 2.92 and 1.83, respectively. In the 7b/8b pair,
the endo hydrogens appear at d 4.66 and 3.63 for the major
isomer 7b and at d 4.58 and 3.58 for the minor isomer 8b.
This was confirmed by NOESY experiment in that, in 7b,
besides the positive NOE between Hb and the aromatic
protons, the key positive NOE between Ha and the endo-
methylene protons as well as between Hb and the olefinic
proton H-3 were discernible.
`
double bond in 6a–d vis-a-vis the cyclopentenyl (C2–C3)
double bond. There is no evidence for the formation of the
cycloadduct arising from reaction of the cyclopentenyl
(C2–C3) double bond with the nitrile oxides. This is
broadly consistent with the reactivity of dicyclopentadiene
6a18 and Thiele’s ester 6b34 although evidence to the
contrary also exists in the literature.35,36 In any event, the
lower reactivity amounting to inertness of the endo-oriented
cyclopentenyl (C2–C3) double bond in the dipolar
cycloaddition is attributable to the preferential entry of the
approaching dipole from the exo face of the bicycloheptane
skeleton.
Quite remarkably, a single stereo- and regioisomer 7c or 7d
is formed from Thiele’s ester 6b in its reaction with nitrile
oxide 1a or 1b (Table 2). That Hb in 7c (d 3.43) and in 7d (d
4.00) is endo oriented (Fig. 2) is evident from the fact that it
appears either as a singlet (in 7c no coupling with Hc) or
shows only very weak coupling (,2 Hz) with Hc (dihedral
angle of close to 908 between Hb and Hc). As for the
regiochemistry, the regioisomer 7c or 7d in which the
oxygen of the nitrile oxide bonded to the more substituted
(ester-bearing) olefinic carbon C6 is preferentially formed.
This is consistent with the reactivity of aceto- and
benzonitrile oxides with methyl acrylate6,7 as well as with
unsymmetrically substituted norbornenes.16 The 13C-SEFT
(APT) spectra show that the carbons attached to the oxygen
in the isoxazoline rings of 7c and 7d appearing at d 93.1 and
94.7, respectively, are quaternary carbons. The above
assignment is further confirmed by NOESY experiment.
For instance, Hb in 7d has a positive NOE with Hc, the
olefinic proton H-3 and the aromatic protons (presumably
the ortho protons). This, taken together with the absence of
any NOE between Hb and Hd, confirms structure 7d and,
therefore, by analogy, structure 7c for the cycloadducts.
Having confirmed the chemoselectivity in the dipolar
cycloaddition, the stereo and regio preferences observed
in the cycloaddition had to be ascertained. It is evident from
1H NMR spectra that all the cycloadditions follow the ‘exo
rule’37 of Alder and Stein providing exclusively the exo-
cycloadducts. This is also in accord with the formation of
exo-cycloadducts as the exclusive or predominant products
in the cycloaddition of nitrile oxides to norbornenes11,17,18,38
and norbornadienes.18,19,20
Reaction of dicyclopentadiene 6a with nitrile oxides 1a and
1b proceeds with high stereoselectivity providing exclu-
sively the exo cycloadducts (Table 1). The protons Ha
(Xa¼Ha) and Hb in the cycloadducts appear as doublets
(J¼8.25 Hz) coupled only with each other, but not with Hc
or Hd) indicating their endo orientation (Fig. 2). However,
unlike the case of Thiele’s ester 6b where single
regioisomer 7c or 7d is formed (vide infra), regioselectivity
in the case of 6a is low, as expected, providing a mixture of
regioisomers 7a/8a and 7b/8b in ,55:45 ratio (Table 1). In
Table 1. 1,3-Dipolar cycloaddition of nitrile oxides 1 to dicyclopentadiene
6a
Table 2. 1,3-Dipolar cycloaddition of nitrile oxides 1 to Thiele’s ester 6b
Entry
1
Path
Yielda (%)
7:8b
Entry
1
Path
Yielda (%)
7:8b
1
2
3
4
1a
1a
1b
1b
A (A)c
B
A (B)c
B
57
84
86
52
53:47
53:47
54:46
53:47
1
2
3
4
1a
1a
1b
1b
A (A)c
B
A (B)c
B
52
72
73
50
.99:1
.99:1
.99:1
.99:1
a
a
Isolated yield after column chromatography.
Obtained by 1H NMR (400 MHz) integration of the crude product.
Method in parenthesis.
Isolated yield after column chromatography.
Obtained by 1H NMR (400 MHz) integration of the crude product.
Method in parenthesis.
b
c
b
c

1456
I. N. N. Namboothiri et al. / Tetrahedron 60 (2004) 1453–1462
Table 3. 1,3-Dipolar cycloaddition of nitrile oxides 1 to alcohol 6c and
enone 6d
Entry
1
6
Path
Yielda (%)
7:8b
1
2
1b
1b
6c
6d
A (B)c
A (B)c
70
68
55:45
34:66
a
Isolated yield after column chromatography.
Obtained by 1H NMR (400 MHz) integration of the crude product.
Method in parenthesis.
b
c
Subsequently, the influence of remote substituents on the
reactivity of the bicycloheptenyl double bond has been
investigated using syn-alcohol 6c and enone 6d (Table 3,
entries 1 and 2). Interestingly enough, despite the presence
of a hydroxy group syn to the bicycloheptenyl (C5–C6)
double bond, the behavior of alcohol 6c is analogous to that
of 6a. A mixture of regioisomers 7e and 8e (7e/8e¼55:45) is
formed when 6c reacts with benzonitrile oxide 1b (Table 3,
entry 1). When the endo-methylene group in dicyclopenta-
diene 6a is replaced by a carbonyl group (as in 6d),
substantial alteration in the ratio of the regioisomers (7f/
8f¼34:66) arising from the reaction of the bicycloheptenyl
double bond (in 6d) is observed (Table 3, entry 2).
Comparison of the reactivity of 6a and 6d indicates that
the reversal in the regioselectivity is attributable to
electronic effects (vide infra) as the exo face of the
bicycloheptenyl double bond in both 6a and 6d experiences
similar steric environment.
Figure 3. X-ray crystal structure of 8f.39
Though aqueous model has been used to examine the
selectivity in all the cases at both the levels of theory, THF
has been employed only with B3LYP/6-31Gp//PM3 basis
set. The reliability of these calculations for predicting the
selectivities observed in the 1,3-dipolar cycloaddition were
first examined by performing model calculations as
described below.
The relative reactivities of cyclopentene and norbornene
with acetonitrile oxide 1a were first examined. As
mentioned earlier, norbornene has been shown to be much
more reactive than cyclopentene in their cycloaddition with
nitrile oxides by Huisgen and co-workers (vide infra).27 In
addition, we have examined the regioselectivity in the
cycloaddition of nitrile oxide 1a to methylacrylate,
b-methyl methylacrylate and b,b-dimethyl methylacrylate
that has earlier been investigated by Huisgen and Houk.6,7
The structure and stereochemistry of the pairs of cycload-
ducts 7e/8e and 7f/8f were confirmed as described in the
case of the cycloadducts arising from dicyclopentadiene 6a
and Thiele’s ester 6b. The structure of 8f has been further
established by single crystal X-ray crystallography
(Fig. 3).39
Our calculated results on the relative reactivities of
ethylene, cyclopentadiene and norbornene with acetonitrile
oxide 1a showed good qualitative agreement with Huisgen’s
results. Transition states were located for the cycloaddition
of acetonitrile oxide 1a with ethylene, cyclopentene and
norbornene, at PM3 level and the relative activation barriers
calculated at B3LYP/6-31Gp//RHF/PM3 level were 12.2,
17.1 and 11.1 kcal mol21, respectively. The corresponding
rate constants k (sec21) calculated using Arrhenius equation
are 0.00224, 0.000193 and 0.00388, respectively. The
relative rate constants for the cycloaddition of cyclopentene
and norbornene with respect to ethylene (0.086 and 1.73,
respectively) clearly indicate the preferential reactivity of
3. Theoretical calculations
In order to probe the selectivities observed during the 1,3-
dipolar cycloaddition of nitrile oxides 1 to dicyclopenta-
diene 6a and Thiele’s ester 6b and other derivatives 6c–d,
TS energy calculations were carried out at semiempirical
(PM3) and hybrid ab initio-DFT levels of theory. All the TS
geometries were optimized at PM3 level and characterized
with one imaginary frequency.40 Single point calculations
were performed using B3LYP/6-31Gp level of theory at
PM3 TS geometries.41 Solvent corrections42 were modeled
with aqueous model43 and organic solvent (THF).44 It may
be noted that THF was used in the generation of nitrile
oxides 1 from nitroalkanes and its subsequent cycloaddition
with 6a and 6b. Solvent continuum model, IPCM was
employed to calculate the energies in presence of THF.42,43
`
norbornene towards nitrile oxide 1a visa-vis cyclopentene.
The fact that the cyclopentene moiety in 6 is endo fused
to the norbornene moiety further diminishes the reactivity
of the former and, therefore, points to the observed
chemoselectivity.
As far as our model calculations on the regioselectivity are
concerned, results obtained at semi-empirical (PM3) and
hybrid ab initio-DFT levels of theory, for the cycloaddition
of nitrile oxide 1a to methylacrylate, b-methyl methylacryl-
ate and b,b-dimethyl methylacrylate, by locating the
transition states (TS’s) of cycloadducts, viz. 5-isoxazoline
9 and 4-isoxazoline 10, concurred well with the observed
experimental results6,7 (see Table 4).
Figure 2.

I. N. N. Namboothiri et al. / Tetrahedron 60 (2004) 1453–1462
1457
Table 4. Semi-empirical (PM3) and hybrid ab initio/DFT (B3LYP/6-31Gp) level calculated relative TS energy differences in kcal/mol for the cycloaddition of
acetonitrile oxide 1a with substituted acrylates
Row/col
1
2
3
4
5
6
7
8
9
Method
RHF/PM3 (A)
9a
10a
0.3
1.6
2.0
3.4
2.2
9b
10b
0.0
0.1
0.0
0.0
0.0
9c
10c
0.0
0.0
0.0
0.0
0.0
11
12
1
2
3
4
5
0.0
0.0
0.0
0.0
0.0
0.9
0.0
2.7
1.8
2.7
1.4
0.1
8.6
6.9
8.2
0.2
0.0
0.0
0.0
0.0
0.0
1.8
2.5
4.2
2.6
PM3 SM5.4 (B)a
B3LYP/6-31Gp//PM3 (C)
B3LYP/6-31Gp//PM3 SM5.4 (D)a
B3LYP/6-31Gp//PM3 (E)b
a
Aqueous model.
THF model.
b
The gas phase and solvent model calculations at semi-
empirical (PM3) and hybrid ab initio/DFT levels of theory
predicted the predominant formation of 5-isoxazoline 9a
over 4-isoxazoline 10a (Table 4, cols 2 and 3) in the
cycloaddition of methylacrylate with acetonitrile oxide 1a
(experimental ratio6 9a:10a¼95:5). The selectivity is
reversed in the cycloaddition of acetonitrile oxide 1a with
b-methyl methylacrylate in that the 4-isoxazoline 10b is the
major isomer (cols 4 and 5, experimental ratio6
9b:10b¼36:64). Further methyl substitution at the b-pos-
ition of acrylate, i.e. b,b-dimethyl methylacrylate, leads to
exclusive formation of 4-isoxazoline 10c (cols 6 and 7,
experimental ratio6 9c:10c¼0:100). The regio and stereo-
selectivity observed in the cycloaddition of acetonitrile
oxide 1a to norbornene carboxylate 4 (X¼CO2Me)16 is also
qualitatively supported by our calculations at various levels
of theory (cols 8 and 9, experimental ratio, X¼CO2Et,16
11:12¼100:0).
dipole 1a towards both 6a and 6b and the corresponding
approaches of the dipole 1b towards dipolarophile 6d were
found to be prohibitively high and, therefore, were omitted
from Table 5 for simplicity. The analysis of PM3 TS
geometries suggested that the computed TS’s for addition of
dipole 1a to dipolarophiles 6a and 6b and those for the
addition of dipole 1b to dipolarophile 6d are concerted and
slightly asynchronous in nature.45
The gas phase and aqueous model calculations at semi-
empirical level for the cycloaddition of 1a to dicyclopenta-
diene 6a (Table 5, entry 1), though favor the bb-approach of
the dipole 1a over the ab and other approaches, make no
distinction between the TS’s leading to two regioisomers 7
and 8. However, calculations at higher level, i.e. B3LYP/6-
31Gp (including gas phase, aqueous and THF model, Table 5,
entry 2), predicted the marginal preference for the formation
of isomer 7 over isomer 8, as observed experimentally
(7a:8a¼53:47).
Having established the efficacy of our approach to
satisfactorily predict the selectivities observed in the
previous experimental studies,6,16 we turned to the chemo-,
stereo- and regioselectivities observed in our laboratories in
the cycloaddition of nitrile oxides 1 to dicyclopentadiene
6a, its dicarboxylate, Thiele’s ester, 6b, and other
derivatives 6c–d and the results are summarized in
Table 5. Since the experimental results provided no
evidence for the formation of any bis-cycloadducts, they
were excluded from calculations. Further, the relative TS
energies corresponding to ba- and aa-approach of the
As for the cycloaddition of nitrile oxide 1a to Thiele’s ester
6b, calculations at semi-empirical level (PM3, gas phase
and aqueous model, Table 5, entry 3) showed the marginal
preference for the approach of nitrile oxide 1a towards the
C2–C3 double bond (ab-approach) of Thiele’s ester 6b.
However, the bb-approach of the dipole 1a in which the
dipole oxygen is bonded to the ester-bearing carbon C6 of
6b is preferred over the bb-approach in which the dipole C
is bonded to C6 by 0.6 and 1.4 kcal mol21, respectively.
Higher level (B3LYP/6-31Gp) calculated results very
Table 5. Semi-empirical (PM3) and B3LYP/6-31Gp calculated relative transition state energy differences for the cycloaddition of 1a with 6a and 6b and 1b
with 6d in kcal/mola
Entry
Computational level
6
bb(XC,O)b 7
bb(XC,C)c 8
ab(XC,O)
ab(XC,C)
1
2
3
4
5
6
RHF/PM3
B3LYP/6-31Gp//PM3
RHF/PM3
6a
6a
6b
6b
6d
6d
0.0 (0.0)
0.0 (0.0, 0.0)
0.4 (0.2)
0.0 (0.0, 0.0)
0.4 (0.0)
0.5 (0.0, 0.3)
0.0 (0.0)
0.1 (0.05, 0.1)
1.0 (1.6)
3.0 (4.0, 3.4)
0.0 (0.9)
0.0 (0.5, 0.0)
0.7 (0.7)
6.1 (6.0, 5.9)
0.9 (0.0)
3.9 (3.2, 3.6)
1.1 (0.5)
6.9 (6.0, 6.4)
0.7 (0.8)
6.0 (5.9, 6.0)
0.0 (1.7)
6.7 (8.8, 7.1)
0.4 (2.3)
2.3 (3.6, 2.8)
B3LYP/6-31Gp//PM3
RHF/PM3
B3LYP/6-31Gp//PM3
a
Solvent calculated (aqueous model, THF model) energy differences in parentheses.
(XC,O): dipole oxygen forms bond with C–X.
(XC,C): dipole carbon forms bond with C–X.
b
c

1458
I. N. N. Namboothiri et al. / Tetrahedron 60 (2004) 1453–1462
clearly predicted the formation of the experimentally
observed product 7c over 8c and the products arising from
ab-approach of the dipole 1a (Table 5, entry 4).
Furthermore, upon incorporating the solvent model (aqu-
eous and THF) at B3LYP/6-31Gp level, the calculated
results show even larger preference for the formation of
experimentally observed product 7c (7c:8c¼.99:1).
tative systems possessing multiple p-faces, viz. dicyclo-
pentadiene and its derivatives, has been investigated. The
`
greater reactivity of bicycloheptenyl double bond vis-a-vis
cyclopentenyl double bond in the dicyclopentadienyl
moiety towards the nitrile oxide dipole is amply evident
from the exclusive formation of the exo-cycloadduct arising
from approach of the dipole from the exo face of the
bicycloheptenyl double bond. Furthermore, in the case of
substituted dicyclopentadienes, viz. the dicarboxylate
(Thiele’s ester), the exclusive regioisomer is the one in
which the dipole oxygen is attached to the carbon bearing
the substituent. Influence of remote substituents, including
sterically ‘sterile’ ones, on the regioselectivity has also been
demonstrated. The combination of semi-empirical (PM3)
and hybrid ab initio/DFT (B3LYP) methods was used to
predict the chemo-, stereo- and regioselectivity of dicyclo-
pentadienes and its derivatives. These more economical
methods have successfully predicted the selectivity in the
1,3-dipolar cycloadditions of nitrile oxide to model systems
as well as complex real systems. Regioselectivities observed
in these cases were rationalized on the basis of Mulliken
population analysis. The commonly used frontier molecular
orbital analysis failed to predict the regio-selectivity in these
cases at the levels of theory employed.
In the case of 6d, PM3 and B3LYP/6-31Gp calculated
results predicted the preferential approach of nitrile oxide
1b towards the bb-face of C5–C6 double bond over the ab-
face of C2–C3 double bond. The regioselectivity obtained
from our calculations is in agreement with our experimental
results. Aqueous model, however, altered the preference in
favor of bb(XC,O) approach, and can be understood on the
basis of the dipole moment of the transition states. The
calculated dipole moment of the TS corresponding to
bb(XC,O) approach is relatively higher and gets more
stabilized in the polar solvent. However, the calculations
performed in THF are in accordance with gas phase results.
The overall calculated results suggest that the hybrid ab
initio/DFT (B3LYP) levels predict more accurately the
experimental observations. It has been found that for 1,3-
dipolar cycloadditions, B3LYP calculations often give
comparable or slightly better results than other high level
MP2, CASSCF or CCSD(T) calculations.46 The pertur-
bation MO treatment of cycloaddition reactivity pioneered
by Sustmann47 has generally been applied to explain the
regioselectivity qualitatively of 1,3-dipolar cycloaddi-
tions.48 Such frontier molecular orbital analysis performed
for 6a, 6b and 6d does not provide any clear picture towards
the origin of regioselectivity in these cases. However,
turning to the Mulliken population analysis, the charges
calculated for 6b and 6d at B3LYP/6-31Gp shows that the
C6 carbon of 6b bears a positive charge (0.100), while C5
has a negative charge of (20.118) and in the case of 6d, C6
carries slightly more negative charge (20.111) in compari-
son to C5 (20.106). This result explains the regioselectivity
observed for 6b qualitatively, where the negative end of the
dipole 1a (oxygen) is attached to the C6 carbon of 6b and
the positive end of the dipole 1a (carbon) is attached to C5
of 6b. The difference in the charge between C5 and C6 is
smaller in the case of 6d, and leads to a mixture of
regioisomers. Mayo et al. have performed similar charge
analysis to rationalize the regioselectivity observed for the
1,3-dipolar cycloaddition of nitrile oxides with unsymme-
trically substituted norbornenes.16
5. Experimental
5.1. General
The melting points are uncorrected. IR spectra were
recorded on an Impact 400/Nicolet FT spectrometer.
1
1
NMR spectra (1H, 13C, H–1H COSY, H–1H NOESY)
were recorded on a JEOL-400, AMX-400 or VXR-300S
spectrometer with TMS as the internal standard. High
resolution mass spectra (CI in methane or i-butane) were
recorded at 60–70 eV on a VG-Fisons ‘Autospec’ spec-
trometer. X-ray data were collected on a MACH3S-
CAAD4/NONIUS diffractometer.
5.2. Procedure for the generation of acetonitrile oxide 1a
from acetaldoxime and its reaction with 6 (path A,
method A)
To a stirred solution of 6 (0.5 mmol) and N-chlorosuccini-
mide (134 mg, 1 mmol, 2 equiv.) in CH2Cl2 (10 ml) at 0 8C
under N2 was added acetaldoxime (59 mg, 1 mmol,
2 equiv.) followed by Et3N (22 mg, 0.22 mmol,
0.44 equiv.). The reaction mixture was stirred at rt overnight
(12 h). The reaction mixture was then diluted with CH2Cl2
(20 ml), washed with 5% HCl (20 ml), brine (20 ml), dried
over anh. Na2SO4 and concentrated in vacuo. The crude
The above calculated results for the cycloaddition of nitrile
oxide 1 to dicyclopentadiene 6a and its derivatives 6b–d have
shown that the norbornene units of 6a–d are comparatively
more reactive than cyclopentene units as predicted and
observed in the case of isolated norbornenes and cyclopen-
tenes.27 Regioselectivities predicted and observed for the
addition of nitrile oxide 1a to Thiele’s ester 6b are in
accordance with the norbornene esters.16 Overall, it appears
that the reactivity of norbornene units in 6a–d are not
perturbed in presence of cyclopentene units.
1
residue was subjected to H NMR (400 MHz) analysis in
order to determine the isomeric purity/composition and then
purified by silica gel column chromatography by eluting
with ethylacetate/pet. ether.
5.3. Procedure for the generation of benzonitrile oxide
1b from benzaldoxime and its reaction with 6 (path A,
method B)
4. Summary and conclusions
To a stirred solution of 6 (0.5 mmol), oxime (1 mmol,
2 equiv.) and few drops of Et3N in CH2Cl2 (10 ml) at 0 8C
The 1,3-dipolar cycloaddition of nitrile oxides to represen-

I. N. N. Namboothiri et al. / Tetrahedron 60 (2004) 1453–1462
1459
was added dropwise 4% NaOCl solution (10 ml, excess).
The cooling bath was removed and the reaction mixture,
while stirring continued, was allowed to warm to rt
overnight (12 h). The layers were separated, the organic
layer was washed with 5% HCl (20 ml), brine (20 ml), dried
over anh. Na2SO4 and concentrated in vacuo. The crude
residue was subjected to H NMR (400 MHz) analysis as
before and then purified by silica gel column chromatog-
raphy by eluting with ethylacetate/pet. ether.
7.65 (m, 2H); 13C NMR (CDCl3) d 31.4, 35.2, 39.3, 42.4,
47.5, 51.8, 52.9, 83.3, 126.6, 128.5, 129.1, 129.5, 131.1,
132.1, 157.6; MS (CI, CH4) m/e (rel intensity) 251 (Mþ,
100), 183 (22), 157 (19), 156 (17), 155 (19), 117 (15), 105
(18), 91 (18); HRMS calcd for C17H17NO (Mþ): 251.1310,
found: 251.1314.
1
3-Phenyl-4,4a,5,7a,8,8a-hexahydro-3aH-4,8-methanoin-
deno[5,6-d]isoxazole 8b (minor isomer). Colorless crystal-
line solid; mp 95–96 8C; IR (KBr) cm21 2960 (s), 1604 (m),
1459 (s), 1348 (s); 1H NMR (CDCl3) d 1.40 (d, J¼10.3 Hz,
1H), 1.60 (d, J¼10.3 Hz, 1H), 2.40 (m, 2H), 2.60 (m, 2H),
2.75 (d, J¼5.6 Hz, 1H), 3.20 (m, 1H), 3.58 (d, J¼8.4 Hz,
1H), 4.58 (d, J¼8.4 Hz, 1H), 5.70 (m, 2H), 7.36 (m, 3H),
7.65 (m, 2H); 13C NMR (CDCl3) d 32.6, 34.6, 40.8, 44.0,
45.7, 49.5, 50.2, 85.7, 126.6, 128.5, 129.2, 129.4, 130.6, 131.2,
157.3; MS (CI, CH4) m/e (rel intensity) 251 (Mþ, 100), 183
(26), 155 (33), 141 (17), 115 (15), 105 (16); HRMS calcd for
C17H17NO (Mþ): 251.1310, found: 251.1290.
5.4. Procedure for the generation of nitrile oxide 1 from
nitroalkane (nitroethane or nitrophenylmethane) and its
reaction with 6 (path B)
To a stirred solution of 6 (0.5 mmol) and nitroalkane
(1 mmol, 2 equiv.) in THF (10 ml) under N2 was added
DMAP (12 mg, 0.1 mmol, 20 mol%) followed by BOC2O
(327 mg, 1.5 mmol, 3 equiv.). The reaction mixture was
stirred at rt overnight (12 h). The reaction mixture was then
diluted with water (20 ml) and extracted with ether
(3£15 ml). The combined organic layer was washed with
5% HCl (20 ml), brine (20 ml), dried over anh. Na2SO4 and
concentrated in vacuo. The crude residue was subjected to
1H NMR (400 MHz) analysis as before and then purified by
silica gel column chromatography by eluting with ethyl-
acetate/pet. ether.
5.4.3. Cycloaddition of acetonitrile oxide 1a with
dimethyldicyclopentadiene dicarboxylate 6b. Yield of
cycloadduct dimethyl 3-methyl-3a,4,4a,7,7a,8-hexahydro-
8aH-4,8-methanoindeno-[5,6-d]isoxazole-6,8a-dicarboxyl-
ate 7c: path A, method A: 52%, path B: 72% (see also
Table 2, entries 1 and 2); colorless crystalline solid; mp
175–177 8C; IR (KBr) cm21 2960 (s), 1737 (s), 1716 (s),
1314 (m), 1262 (s); 1H NMR (CDCl3) d 1.48 (d, J¼11.0 Hz,
1H), 1.79 (d, J¼11.0 Hz, 1H), 1.94 (s, 3H), 2.34 (m, 1H),
2.48 (m, 1H), 2.62 (d, J¼5.1 Hz, 1H), 2.82 (m, 1H), 2.88 (d,
J¼4.4 Hz, 1H), 3.37 (m, 1H), 3.43 (s, 1H), 3.73 (s, 3H), 3.75
(s, 3H), 6.60 (d, J¼1.5 Hz, 1H); 13C NMR (CDCl3) d 11.7,
30.3, 37.4, 40.1, 42.9, 49.1, 51.4 (£2), 52.4, 56.7, 93.1,
137.8, 141.5, 156.6, 164.6, 168.5; MS (CI, CH4) m/e (rel
intensity) 306 (MHþ, 5), 305 (Mþ, 1), 274 (8), 246 (7), 232
(8), 214 (11), 204 (15), 183 (100), 173 (14), 172 (26), 151
(75), 117 (23); HRMS calcd for C16H20NO5 (MHþ):
306.1342, found: 306.1360.
5.4.1. Cycloaddition of acetonitrile oxide 1a with
dicyclopentadiene 6a. Total yield of cycloadducts 3-
methyl-4,4a,7,7a,8,8a-hexahydro-3aH-4,8-methanoindeno
[5,6-d]isoxazole 7a þ3-methyl-4,4a,5,7a,8,8a-hexahydro-
3aH-4,8-methanoindeno[5,6-d]isoxazole 8a (inseparable
mixture; 7a:8a¼53:47): path A, method A: 57%, path B:
84% (see also Table 1, entries 1 and 2); colorless crystalline
solid; mp 60–62 8C; IR (KBr) cm21 2960 (s), 1453 (s),
1262 (s); 1H NMR (CDCl3) d 1.30 (d, J¼9.2 Hz, 1H), 1.45
(d, J¼9.2 Hz, 1H), 1.80, 1.83 (d, J¼0.8 Hz, 3H), 2.05–2.32
(m, 3H), 2.38 (m, 1H), 2.45–2.60 (m, 1H), 2.92, 2.97 (d,
J¼8.3 Hz, 1H), 3.08 (m, 1H), 4.28, 4.36 (d, J¼8.3 Hz, 1H),
5.45–5.65 (m, 2H); 13C NMR (CDCl3) d 10.6, 11.8, 31.5,
32.4, 34.6, 35.0, 39.3, 40.7, 41.3, 42.9, 45.5, 47.3, 50.2,
51.6, 53.4, 56.6, 81.6, 84.0, 130.5, 131.2 (£2), 131.8, 156.2,
156.4; MS (CI, CH4) m/e (rel intensity) 190 (MHþ, 12) 189
(Mþ, 42), 169 (80), 168 (100), 141 (96), 131 (62), 115 (78);
HRMS calcd for C12H15NO (Mþ) 189.1154,
found:189.1164.
5.4.4. Cycloaddition of benzonitrile oxide 1b with
dimethyldicyclopentadiene dicarboxylate 6b. Yield of
cycloadduct 3-phenyl-3a,4,4a,7,7a,8-hexahydro-8aH-4,8-
methanoindeno-[5,6-d]isoxazole-6,8a-dicarboxylate
7d:
path A, method B: 73%, path B: 50% (see also Table 2,
entries 3 and 4); colorless crystalline solid; mp 160–161 8C;
IR (KBr) cm21 2939 (s), 1738 (s), 1716 (s); 1H NMR
(CDCl3) d 1.50 (d, J¼10.6 Hz, 1H), 1.89 (d, J¼10.6 Hz,
1H), 2.40 (m, 1H), 2.52 (m, 1H), 2.78 (dd, J¼5.1 and
2.2 Hz, 1H), 2.88 (m, 1H), 2.98 (d, J¼4.4 Hz, 1H), 3.39 (m,
1H), 3.75 (s, 3H), 3.77 (s, 3H), 4.0 (d, J¼2.2 Hz, 1H), 6.74
(d, J¼1.8 Hz, 1H), 7.40 (m, 3H), 7.68 (m, 2H); 13C NMR
(CDCl3) d 30.3, 37.5, 40.1, 44.1, 49.2, 51.5 (£2), 52.5, 53.2,
94.7, 127.0, 128.3, 128.7, 130.2, 137.9, 141.5, 158.1, 164.7,
168.3; MS (CI, CH4) m/e (rel intensity) 367 (Mþ, 23), 331
(24), 308 (10), 280 (24), 232 (12), 183 (100), 151 (49);
HRMS calcd for C21H21NO5 (Mþ) 367.1420, found:
367.1420.
5.4.2. Cycloaddition of benzonitrile oxide 1b with
dicyclopentadiene 6a. Total yield of cycloadducts 3-
phenyl-4,4a,7,7a,8,8a-hexahydro-3aH-4,8-methanoindeno
[5,6-d]isoxazole 7b þ3-phenyl-4,4a,5,7a,8,8a-hexahydro-
3aH-4,8-methanoindeno[5,6-d]isoxazole 8b (7b:8b¼54:46):
path A, method B: 86%, path B: 52% (see also Table 1,
entries 3 and 4).
3-Phenyl-4,4a,7,7a,8,8a-hexahydro-3aH-4,8-methanoin-
deno[5,6-d]isoxazole 7b (major isomer). Separated from the
7b/8b mixture by fractional crystallization from ethanol;
colorless crystalline solid; mp 85–87 8C; IR (KBr) cm21
5.4.5. Cycloaddition of benzonitrile oxide 1b with
dicyclopentadien-8-ol 6c. 3-Phenyl-4,4a,7,7a,8,8a-hexahy-
dro-3aH-4,8-methanoindeno[5,6-d]isoxazol-9-ol 7e (major
isomer) þ3-phenyl-4,4a,5,7a,8,8ahexahydro-3aH-4,8-meth-
anoindeno[5,6-d]isoxazol-9-ol 8e (minor isomer): Total
1
2962 (s), 1600 (m), 1461 (s); H NMR (CDCl3) d 1.40 (d,
J¼10.3 Hz, 1H), 1.60 (d, J¼10.3 Hz, 1H), 2.30 (m, 2H),
2.60 (m, 3H), 3.20 (m, 1H), 3.63 (d, J¼8.4 Hz, 1H), 4.66 (d,
J¼8.4 Hz, 1H), 5.68 (m, 1H), 5.78 (m, 1H), 7.36 (m, 3H),

1460
I. N. N. Namboothiri et al. / Tetrahedron 60 (2004) 1453–1462
chemistry; Padwa, A., Ed.; Wiley: New York, 1984.
(b) Namboothiri, I. N. N.; Hassner, A. Topics in current
chemistry: stereoselective intramolecular 1,3-dipolar cyclo-
additions; Metz, P., Ed.; Springer: Germany, 2001; Vol. 216,
pp 1–49. (c) Karlsson, S.; Hogberg, H.-E. Org. Prep. Proc. Int.
2001, 33, 103. (d) Fisera, L.; Ondrus, V.; Kuban, J.; Micuch, P.;
yield (inseparable mixture, 7e:8e¼55:45) 70% (see also
Table 3, entry 1); colorless crystalline solid; mp 140–
142 8C; IR (KBr) cm21 3399 (b), 2976 (s), 1658 (s), 1351
1
(s), 1038 (s); H NMR (CDCl3) d 1.55 (m, 2H), 2.45 (m,
1H), 2.65 (m, 1H), 2.81 (m, 1H), 3.36 (m, 1H), 3.39, 3.48 (d,
J¼8.4 Hz, 1H), 4.43, 4.49 (d, J¼8.4 Hz, 1H), 4.83 (m, 1H),
5.87–5.95 (m, 1H), 6.02–6.05 (m, 1H), 7.40–7.48 (m, 3H),
7.60–7.69 (m, 2H); 13C NMR (CDCl3) d 35.2, 35.7, 42.4,
43.0, 45.5, 46.6, 49.5, 50.6, 51.0, 51.1, 52.2, 52.6, 75.6,
76.8, 83.5, 85.0, 126.7, 126.8, 128.72, 128.74, 129.0 (£2),
129.9 (£2), 133.6, 134.9, 137.1, 137.4, 157.3, 157.6; MS
(CI, i-butane) m/e (rel intensity) 268 (MHþ, 61), 267 (Mþ,
100), 250 (13); HRMS calcd for C17H17NO2 (Mþ):
267.1259, found 267.1236.
¨
Blanarikova, I.; Jager, V. J. Heterocycl. Chem. 2000, 37, 551.
(e) Huisgen, R. Chem. Pharm. Bull. 2000, 48, 757. (f) Gothelf,
K. V.; Jørgensen, K. A. Chem. Rev. 1998, 98, 863.
3. For reviews: (a) Kozikowski, A. P. Acc. Chem. Res. 1984, 17,
410. (b) Torssell, K. B. G. Nitrile oxides, nitrones and
nitronates in organic synthesis; VCH: New York, 1988.
(c) Grundmann, C.; Grunanger, P. The nitrile oxides. Springer:
Berlin, 1971. (d) Hassner, A.; Murthy, K. S. K.; Maurya, R.;
Dehaen, W.; Friedman, O. J. Heterocycl. Chem. 1994, 31, 687.
(e) Litvinovskaya, R. P.; Khripach, V. A. Russ. Chem. Rev.
2001, 70, 405. (f) Caramella, P.; Gru¨nanger, P. In 1,3-Dipolar
cycloaddition chemistry; Padwa, A. Ed.; Wiley: New York;
Vol. 1; Chapter 3. (g) Padwa, A. In 1,3-Dipolar Cycloaddition
Chemistry; Padwa, A. Ed.; Wiley: New York; Vol. 2; Chapter
2. (h) Padwa, A.; Schoffstall, A. M. Intramolecular 1,3-dipolar
cycloaddition chemistry. In Advances in cycloaddition.
Curran, D. P., Ed.; JAI: Greenwich, 1990; pp 1–89.
¨
5.4.6. Cycloaddition of benzonitrile oxide 1b with
dicyclopentadien-1-one 6d. Total yield of cycloadducts
(7f:8f¼34:66) 68% (see also Table 3, entry 2).
3-Phenyl-3a,4,4a,7a,8,8a-hexahydro-5H-4,8-methanoin-
deno[5,6-d]isoxazol-5-one (major isomer) 8f. Separated
from the mixture by fractional crystallization from ethanol,
colorless crystalline solid; mp 210–211 8C; IR (KBr) cm21
2976 (m), 1697 (s), 1351 (w), 1274 (m), 1184 (w), 1076 (w),
4. For applications to natural product synthesis: (a). Tetrahedron
symposia-inprint; Oppolzer, W., Ed.;, 1985; Vol. 41. p 3447.
¨
¨
(b) Jager, V.; Muller, R.; Leibold, T.; Hein, M.; Schwarz, M.;
1
890 (m); H NMR (CDCl3) d 1.70 (ABq, J¼11.0 Hz, 2H),
2.77 (t, J¼6.1 Hz, 1H), 2.95 (d, J¼4.9 Hz, 2H), 3.36 (m,
1H), 3.44 (d, J¼8.5 Hz, 1H), 4.44 (d, J¼8.5 Hz, 1H), 6.21
(m, 1H), 7.39–7.40 (m, 3H), 7.68 (m, 2H), 7.73 (dd, J¼6.1,
2.4 Hz, 1H); 13C NMR (CDCl3) d 36.5, 43.6, 45.2, 46.5,
50.7, 52.3, 84.2, 126.7, 128.2, 128.69, 129.9, 135.9, 156.9,
164.4, 209.6; MS (CI, i-butane) m/e (rel intensity) 266
(MHþ, 265 (Mþ, 100), 172 (79), 144 (12), 143 (15), 128 (8),
117 (9), 93 (54), 76 (14); HRMS calcd for C17H15NO2
(Mþ): 265.1103, found 265.1106.
¨
Fengler, M.; Jaroskova, L.; Parzel, M.; Le Roy, P.-Y. Bull.
Soc. Chim. Belg. 1994, 103, 491. (c) Brieger, G.; Bennet, J. N.
Chem. Rev. 1980, 80, 63. (d) Broggini, G.; Zecchi, G.
Synthesis 1999, 905. (e) Kotyatkina, A. I.; Zhabinsky, V. N.;
Khripach, V. A. Russ. Chem. Rev. 2001, 70, 641. (f) Kanemasa,
S.; Tsuge, O. Heterocycles 1990, 30, 719.
5. Tam, W.; Tranmer, G. K. Org. Lett. 2002, 4, 4101.
6. (a) Christl, M.; Huisgen, R. Tetrahedron Lett. 1968, 5209.
(b) Christl, M.; Huisgen, R.; Sustmann, R. Chem. Ber. 1973,
106, 3275. (c) Christl, M.; Huisgen, R. Chem. Ber. 1973, 106,
3345.
3-Phenyl-3a,4,4a,7a,8,8a-hexahydro-7H-4,8-methanoin-
deno[5,6-d]isoxazol-7-one (minor isomer) 7f. Colorless
crystalline solid; mp 158–160 8C; IR (KBr) cm21 2970 (m),
1697 (s), 1345 (w), 1095 (w); 1H NMR (CDCl3) d 1.70 (ABq,
J¼11.0 Hz, 2H), 2.78 (m, 2H), 3.18 (d, J¼4.9 Hz, 1H), 3.35
(m, 2H), 4.52 (d, J¼8.5 Hz, 1H), 6.25 (m, 1H), 7.41 (m, 3H),
7.67 (m, 2H), 7.23 (m, 1H); 13C NMR (CDCl3) d 36.3, 42.0,
47.1, 48.3, 48.9, 50.7, 52.0, 52.8, 83.3, 126.6, 128.5, 128.72,
129.9, 136.5, 157.1, 163.5, 208.5; MS (CI, i-butane) m/e (rel
intensity) 266 (MHþ, 59), 265 (Mþ, 100), 86 (23); HRMS
calcd for C17H15NO2 (Mþ): 265.1103, found 265.1117.
7. (a) Huisgen, R. J. Org. Chem. 1976, 41, 403. (b) Houk, K. N.;
Sims, J.; Duke, R. E., Jr.; Strozier, R. W.; George, J. K. J. Am.
Chem. Soc. 1973, 95, 7287. (c) Houk, K. N.; Sims, J.; Watts,
C. R.; Luskus, L. J. J. Am. Chem. Soc. 1973, 95, 7301.
8. (a) Beltrame, P.; Beltrame, P. L.; Caramella, P.; Cellerino, G.;
Fantechi, R. Tetrahedron Lett. 1975, 3543. (b) Bianchi, G.; De
Micheli, C.; Gandolfi, R. J. Chem. Soc., Perkin Trans. 1 1976,
1518.
9. (a) Cram, D. J.; Abd Elhafez, F. A. J. Am. Chem. Soc. 1952,
74, 5828. (b) Cieplak, A. S. J. Am. Chem. Soc. 1981, 103,
4540. (c) Cheung, C. K.; Tseng, L. T.; Lin, M.-H.; Srivastava,
S.; le Noble, W. J. J. Am. Chem. Soc. 1986, 108, 1598. (d) Wu,
Y.-D.; Houk, K. N. J. Am. Chem. Soc. 1987, 109, 908.
10. For recent reviews on p-facial selectivity: (a) Chem. Rev.
1999, 99(5): Special issue on p-face selection (b) Mehta, G.;
Uma, R. Acc. Chem. Res. 2000, 33, 278. (c) Marchand, A. P.;
Coxon, J. M. Acc. Chem. Res. 2002, 35, 271.
Acknowledgements
I.N.N.N. thanks CSIR (India) for financial support, Dr. J. N.
Moorthy (IIT Kanpur), RSIC (IIT Bombay) and SIF (IISc.,
Bangalore) for NMR spectra. N.R. thanks IIT Bombay for
a Teaching Assistantship. B.G. thanks Dr. P. K. Ghosh,
Director, CSMCRI for his support.
11. Fliege, W.; Huisgen, R. Justus Liebigs Ann. Chem. 1973, 2038.
12. Gandolfi, R.; Tonoletti, G.; Rastelli, A.; Bagatti, M. J. Org.
Chem. 1993, 58, 6038.
13. Bianchi, G.; De Micheli, C.; Gamba, A.; Gandolfi, R. J. Chem.
Soc., Perkin Trans. 1 1974, 137.
14. (a) Houk, K. N.; Moses, S. R.; Wu, Y. D.; Rondan, N. G.;
References and notes
1. (a) Huisgen, R. Angew. Chem., Int. Ed. Engl. 1963, 2, 565.
(b) Huisgen, R. Angew. Chem., Int. Ed. Engl. 1963, 2, 633.
2. For selected reviews: (a). In 1,3-Dipolar cycloaddition
¨
Jager, V.; Schohe, R.; Fronczek, F. R. J. Am. Chem. Soc. 1984,
106, 3880. (b) Kozikowski, A. P.; Ghosh, A. K. J. Org. Chem.

I. N. N. Namboothiri et al. / Tetrahedron 60 (2004) 1453–1462
1461
¨
1984, 49, 2762. (c) Jager, V.; Schohe, R. Tetrahedron Lett.
1983, 24, 5501.
allowed pericyclic reaction products was called periselectivity:
see Ref. 7c.
15. For recent examples of diastereofacial selective 1,3-dipolar
cycloadditions (a) Garcia Ruano, J. L.; Bercial, F.; Gonzalez,
G.; Martin Castro, A. M.; Martin, M. R. Tetrahedron:
Asymmetry 2002, 13, 1993. (b) Melsa, P.; Mazal, C. Collect.
Czech. Chem. Commun. 2002, 67, 353. (c) Garcia Ruano, J. L.;
Alonso de Diego, S. A.; Blanco, D.; Martin Castro, A. M.;
Martin, M. R.; Rodriguez Ramos, J. H. Org. Lett. 2001, 3,
3173. (d) Garcia Ruano, J. L.; Fraile, A.; Martin, M. R.
Tetrahedron 1999, 55, 14491. (e) Padwa, A.; Prein, M.
Tetrahedron 1998, 5, 6957. (f) Fray, A. H.; Meyers, A. I.
J. Org. Chem. 1996, 61, 3362.
33. Presence of excess nitrile oxide precursors (up to 4 equiv. of
oxime or nitroalkane) and/or excess reagents did not alter the
isomer ratios. In presence of excess dipolarophile, on the other
hand, the reaction remained incomplete, but the isomer ratio
remained unchanged. However, best yields were obtained when
nitrile oxides 1 were generated in presence of the dipolarophile.
34. For e.g. the bicycloheptenyl double bond in Thiele’s ester 6b
undergoes hydrogenation much faster compared to the
cyclopentenyl double bond: see Alder, K.; Stein, G. Annales
1931, 485, 223.
35. For e.g. both the double bonds in dicyclopentadiene 6a are
epoxidised at the same rate: see Turner, R. B.; Meador, W. R.;
Winkler, R. E. J. Am. Chem. Soc. 1957, 79, 4116.
16. Mayo, P.; Hecnar, T.; Tam, W. Tetrahedron 2001, 57, 5931.
17. (a) Arjona, O.; de Dios, A.; de la Pradilla, R. F.; Mallo, A.;
Plumet, A. Tetrahedron 1990, 46, 8179. (b) Arjona, O.;
Dominguez, C.; de la Pradilla, R. F.; Mallo, A.; Manzano, C.;
Plumet, J. J. Org. Chem. 1989, 54, 5883.
36. (a) Dols, P. P. M. A.; Klunder, A. J. H.; Zwanenburg, B.
Tetrahedron 1993, 49, 11373. (b) Marchand, A. P.; Sorokin,
V. D.; Rajagopal, D.; Bott, S. G. Tetrahedron 1994, 50, 9933.
37. (a) Alder, K.; Stein, G. Angew. Chem. 1937, 50, 510.
(b) Martin, J. G.; Hill, R. K. Chem. Rev. 1961, 61, 537.
38. The torsional effects and steric hindrance exerted by the
5-endo H’s were blamed for the exo preference: see Ref. 11.
18. Tanaka, K.; Masuda, H.; Mitsuhashi, K. Bull. Chem. Soc. Jpn
1986, 59, 3901.
19. (a) Micheli, C. D.; Gandolfi, R.; Oberti, R. J. Org. Chem. 1980,
45, 1209. (b) Taniguchi, H.; Ikeda, T.; Yoshida, Y.; Imoto, E.
Bull. Chem. Soc. Jpn. 1977, 50, 2694.
39. Selected X-ray crystallographic data for 8f (C17H15NO2):
˚
20. (a) Yip, C.; Handerson, S.; Jordan, R.; Tam, W. Org. Lett.
1999, 1, 791. (b) Yip, C.; Handerson, S.; Trammer, G. K.;
Tam, W. J. Org. Chem. 2001, 66, 276.
Space group: monoclinic
P
˚
21/n; a¼9.45(12) A;
˚
˚
b¼13.629(10) A; c¼10.865(8) A; V¼1304.3(2) A3; Z¼4,
Dcalc¼1.351 g cm23; m¼0.089 mm21; R1¼0.0521; Rw¼
0.0890. Complete crystallographic data (excluding structure
factors) for this structure have been deposited with the
Cambridge Crystallographic Data Centre as supplementary
publication number CCDC 211238. Copies of the data can be
obtained, free of charge, on application to CCDC, 12 Union
Road, Cambridge CB2 1EZ, UK (fax:þ44-1223-336033 or
e-mail: deposit@ccdc.cam.ac.uk).
21. Houk, K. N.; Luskus, L. J.; Bhacca, N. S. J. Am. Chem. Soc.
1970, 92, 6392.
22. Addition of trifluoroacetonitrile oxide to dicyclopentadiene
has been found to give a mixture of isomers, but no attempts
have been made to characterize the isomers; the endo
cyclopentenyl double bond was hydrogenated making the
two regioisomers indistinguishable: see Ref. 18.
23. For addition of azido and azomethine oxide groups to
dicyclopentadiene, see (a) Nagibina, N. N.; Sidorova, L. P.;
Klyuev, N. A.; Charushin, V. N.; Chupakhin, O. N. Russ.
J. Org. Chem. 1997, 33, 1468. (b) Plenkiewicz, J. Bull. Acad.
Polym. Sci., Ser. Sci. Chim. 1977, CA87, 68248.
40. (a) Stewart, J. J. P. J. Comput. Chem. 1989, 10, 209. (b) Becke,
A. D. J. Chem. Phys. 1993, 98, 1372.
41. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Gill, P. M. W.;
Johnson, B. G.; Robb, M. A.; Cheeseman, J. R.; Keith, T.;
Petersson, G. A.; Montgomery, J. A.; Raghavachari, K.;
Al-Laham, M. A.; Zakrzewski, V. G.; Ortiz, J. V.; Boresman,
J. B.; Peng, C. Y.; Ayala, P. Y.; Chen, W.; Wong, M. W.;
Andres, J. L.; Replogle, E. S.; Gomperts, R.; Martin, R. L.;
Fox, D. J.; Binkley, J. S.; Defrees, D. J.; Baker, J.; Stewart,
J. P.; Head-Gordon, M.; Gonzalez, C.; Pople, J. A.
GAUSSIAN 94, Rev. B.2, Gaussian Inc., Pittsburgh, PA,
1995.
24. (a) Thiele, J. Chem. Ber. 1900, 33, 666. (b) Thiele, J. Chem.
Ber. 1901, 34, 68. (c) Peters, D. J. Chem. Soc. 1959, 1761.
(d) Dunn, G. L.; Donohue, J. K. Tetrahedron Lett. 1968, 3485.
(e) Minter, D. E.; Marchand, A. P.; Lu, S.-P. Magn. Reson.
Chem. 1990, 28, 623. (f) Marchand, A. P.; Namboothiri, I. N.
N.; Lewis, S. B. Tetrahedron 1998, 54, 12691.
25. (a) Woodward, R. B.; Katz, T. J. Tetrahedron 1959, 5, 70.
(b) Cookson, R. C.; Hudec, J.; Williams, R. O. J.C.S. Chem.
Commun. 1967, 1382. (c) Dilling, W. L. Chem. Rev. 1966, 373.
26. Ogino, T.; Awano, K. Bull. Chem. Soc. Jpn 1986, 59, 2811.
27. For instance, the relative addition constants k for norbornene
and cyclopentene in their cycloadditions with benzonitrile
oxide are 15.3 and 0.21, respectively. Ethylene (k¼1) was
taken as the standard dipolarophile, see: Bast, K.; Christl, M.;
Huisgen, R.; Mack, W. Chem. Ber. 1973, 106, 3312.
28. (a) Grundmann, C.; Dean, J. M. J. Org. Chem. 1965, 30, 2809.
(b) See also Ref. 2b for recent references.
42. For a recent report on the solvent effect in 1,3-dipolar
cycloadditions, see: Hu, Y.; Houk, K. N. Tetrahedron 2000,
56, 8239, and the references cited therein.
43. Chambers, C. C.; Hawkins, C. J.; Cramer, C. J.; Truhlar, D. G.
J. Chem. Phys. 1996, 100, 16385.
44. Foresman, J. B.; Keith, T. A.; Wiberg, K. B.; Snoonian, J.;
Frisch, M. J. J. Phys. Chem. 1996, 100, 16098.
45. For discussion on the mechanism of 1,3-dipolar cyclo-
additions: (a) Huisgen, R. J. Org. Chem. 1968, 33, 2291.
(b) Ref. 7a. (c) Poppinger, D. Aust. J. Chem. 1976, 29, 465.
(d) Komornicki, A.; Goddard, J.; Schaefer, H. F., III. J. Am.
Chem. Soc. 1980, 102, 1763. (e) Hiberty, P. C.; Ohanessian,
G.; Schlegel, H. B. J. Am. Chem. Soc. 1983, 105, 719.
(f) Dewar, M. J. S.; Olivella, S.; Rzepa, H. S. J. Am. Chem.
Soc. 1978, 100, 5650. (g) Liuch, J. M.; Bertran, J. Tetrahedron
1982, 38, 1847. (h) Liuch, J. M.; Bertran, J. J. Chem. Soc.,
Perkin Trans. 2 1982, 1419. (i) Harcourt, R. D.; Little, R. D.
J. Am. Chem. Soc. 1984, 106, 41. (j) Weingarten, M. D.; Prein,
29. (a) Mukaiyama, T.; Hoshino, T. J. Am. Chem. Soc. 1960, 82,
5339. (b) Shimizu, T.; Hayashi, T.; Shibafuchi, H.; Teramura,
K. Bull. Chem. Soc. Jpn 1986, 59, 2827.
30. Basel, Y.; Hassner, A. Synthesis 1997, 309.
31. While the isomer 7b was isolated in pure form from 7b/8b
mixture by fractional crystallization from ethanol, attempts to
separate isomers 7a and 8a were not successful (see Section 5).
32. Selective formation of one of the several possible thermally

1462
I. N. N. Namboothiri et al. / Tetrahedron 60 (2004) 1453–1462
M.; Price, T.; Snyder, J. P.; Padwa, A. J. Org. Chem. 2001,
1997, 62.
(b) Sustmann, R.; Trill, H. Angew. Chem., Int. Ed. Engl.
1972, 11, 838.
46. (a) Binkley, J. S.; Pople, J. A. Int. J. Quantum Chem. 1975, 9,
229. (b) Pople, J. A.; Binkley, J. S.; Seeger, R. Int. J. Quantum
Chem. Symp. 1976, 10, 1. (c) Roos, B. O. Adv. Chem. Phys.
1987, 69, 399. (d) Raghavachari, K.; Trucks, G. W.; Pople,
J. A.; Head-Gordon, M. Chem. Phys. Lett. 1989, 157, 479.
(e) Nguyen, M. T.; Chandra, A. K.; Sakai, S.; Morokuma, K.
J. Org. Chem. 1999, 64, 65.
48. (a) Houk, K. N. Acc. Chem. Res. 1975, 8, 361. (b) Caramella,
P.; Houk, K. N. J. Am. Chem. Soc. 1976, 98, 6397. (c) Houk,
K. N.; Caramella, P.; Munchausen, L. L.; Chang, Y.-M.;
Battaglia, A.; Sims, J.; Kaufman, D. C. J. Electron Spectrosc.
Related Phenom. 1977, 10, 441. (d) Houk, K. N.; Yamaguchi,
K. 1,3-Dipolar cycloaddition chemistry; Padwa, A., Ed.;
Wiley: New York, 1984; Vol. 2, p 407.
47. (a) Sustmann, R. Tetrahedron Lett. 1971, 12, 2717.