Substituent effects on regioselectivity in the photorearrangement of a few naphthobarrelenes

Article abstract of DOI:10.1016/S0040-4020(00)00450-6

The photorearrangement of a few 9,10-disubstituted-naphthobarrelenes to the corresponding naphthosemibullvalenes has been examined. It was observed that the regioselectivity in these photoisomerizations depends on the relative stabilities of the diradical intermediates. AM1 semi-empirical calculations are in support of this view. Thermal transformation of the naphthosemibullvalenes gave the corresponding water added naphthopentalenofurans, in nearly quantitative yields. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4020(00)00450-6

TETRAHEDRON
Pergamon
Tetrahedron 56 (2000) 5421±5428
Substituent Effects on Regioselectivity in the Photorearrangement
of a Few Naphthobarrelenes1
Meledathu C. Sajimon,a Danaboyina Ramaiah,a S. Ajaya Kumar,a Nigam P. Rathb and
Manapurathu V. Georgea,c,
*
aPhotochemistry Research Unit, Regional Research Laboratory (CSIR), Trivandrum 695 019, India
bDepartment of Chemistry, University of MissouriÐSt. Louis, St. Louis, MO 63121, USA
cJawaharlal Nehru Centre for Advanced Scienti®c Research, Bangalore 560 064, India
Dedicated to Professor Rolf Huisgen on the occasion of his 80th birthday
Received 14 March 2000; revised 9 May 2000; accepted 16 May 2000
AbstractÐThe photorearrangement of a few 9,10-disubstituted-naphthobarrelenes to the corresponding naphthosemibullvalenes has been
examined. It was observed that the regioselectivity in these photoisomerizations depends on the relative stabilities of the diradical inter-
mediates. AM1 semi-empirical calculations are in support of this view. Thermal transformation of the naphthosemibullvalenes gave the
corresponding water added naphthopentalenofurans, in nearly quantitative yields. q 2000 Elsevier Science Ltd. All rights reserved.
Introduction
naphthobarrelenes, the major pathway involves an initial
naphtho±vinyl bridging.5a,b Although attempts have been
made to rationalize these observations by deuterium-
labeling studies of 1,2-naphthobarrelenes, the effects of
substituents on the initial mode of bonding as well as the
observed regioselectivity in the photorearrangement of 2,3-
naphthobarrelenes have not so far been examined.
The di-p-methane rearrangement of barrelenes,2 benzo-
barrelenes,3 dibenzobarrelenes4 and naphthobarrelenes5
has been extensively investigated. All these derivatives
undergo photorearrangement to the corresponding semi-
bullvalenes and/or cyclooctatetraenes depending upon the
reaction conditions. The semibullvalenes are known to arise
through a triplet state mediated pathway, whereas direct
irradiation leads to the singlet state mediated cyclo-
octatetraene derivatives.6 There has been considerable
interest in recent years to understand the factors that are
responsible for the observed regioselectivity in the photo-
rearrangement of barrelenes, leading to semibullvalenes.
As in the case of substituted dibenzobarrelene derivatives,
which yield two regioisomeric dibenzosemibullvalenes, on
irradiation,7 the corresponding naphthobarrelenes can, in
principle, yield four regioisomers (Scheme 1). For example,
a substituted naphthobarrelene such as 3 can undergo an
initial naphtho±vinyl bridging leading to the formation of
regioisomers 1 and 4, whereas the initial vinyl±vinyl
bridging can lead to the regioisomers 2 and 5. It has been
observed that, as in the case of benzobarrelenes3 and 2,3-
anthracenobarrelenes,8 an initial vinyl±vinyl bonding is the
preferred mode of bridging in the photorearrangement of
2,3-naphthobarrelenes.5a,b However, in the case of 1,2-
Keywords: di-p-methane rearrangement; naphthobarrelenes; naphthosemi-
bullvalenes; regioselectivity.
* Corresponding author. Fax: 191-471-490186/1712;
e-mail: mvg@csrrltrd.ren.nic.in
Scheme 1.
0040±4020/00/$ - see front matter q 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020(00)00450-6

5422
M. C. Sajimon et al. / Tetrahedron 56 (2000) 5421±5428
In the present investigation, we have examined the photo-
rearrangement of a few disubstituted 2,3-naphthobarrelenes
containing 1,2-dibenzoylalkene moieties (6a±e). These
results indicate that in the rearrangement of naphtho-
barrelenes to the corresponding semibullvalenes, steric
effects exerted by the substituents have a bearing on the
observed regioselectivity. Another aspect of the present
study deals with the thermal transformations of the naphtho-
semibullvalenes formed in these reactions. Interestingly, the
thermal transformations of representative naphthosemi-
bullvalenes 10a, 10b and 12e, yielded the corresponding
dihydropentalenofurans as the primary products, which
underwent water addition during work-up, to yield the
corresponding tetrahydropentalenofuran derivatives 14a,b,e.
acetylene (DBA) by neat heating or in the presence of a
Lewis acid catalyst such as aluminium chloride. The corre-
sponding dibenzobarrelenes were also isolated under these
conditions. The structures of the naphthobarrelenes 6a±e
have been arrived at on the basis of analytical results and
1
spectral data. The H NMR spectrum of 6a, for example,
showed the bridgehead protons at d 5.18 and 5.87 as
doublets. In the case of 6d, the 13C NMR spectrum showed
®ve carbons corresponding to the cyclopentyl group over
the range d 27.76±40.00, indicating a restricted orientation
for the cyclopentyl group.
Photolysis of 6a in benzene gave the regiospeci®c naphtho-
semibullvalene 10a in a 90% yield (Scheme 2). Yields
reported herein are the isolated yields of products, after
puri®cation through recrystallization, unless otherwise
1
stated. The H NMR spectrum of 10a showed the allylic
Results
bridgehead proton as a doublet at d 4.42, while the proton
attached to the cyclopropane ring appeared as a singlet at d
4.82. The 13C NMR spectrum of 10a showed four bridge-
head carbons over the range d 48.08±71.69. The structure of
10a was further con®rmed through X-ray crystallographic
analysis.10
Steady-state photolysis and product identi®cation
The starting naphthobarrelenes 6a±e, containing 1,2-
dibenzoylalkene moieties were prepared through the
reaction of the appropriate anthracenes with dibenzoyl-
Scheme 2.

M. C. Sajimon et al. / Tetrahedron 56 (2000) 5421±5428
5423
Table 1. Regioisomeric product distribution in the phototransformations of
naphthobarrelenes 6a±e (irradiated in benzene with RPR 300 nm light
source and the yields are calculated on the basis of 100% product balance)
ing effect of 0.62 ppm. Interestingly, irradiation of the
9-benzoyl-10-phenylnapthobarrelene 6e gave the regio-
speci®c isomer 12e in a 58% yield, along with the
unchanged starting material 6e (30%). The structure of
12e was con®rmed through X-ray crystallographic
analysis.10
Entry Disubstituted naphthobarrelene 10 (%) path A 12 (%) path B
1
2
3
4
5
6a (methyl, benzoyl)
6b (methoxy, benzoyl)
6c (acetyl, methyl)
6d (cyclopentyl, benzoyl)
6e (phenyl, benzoyl)
100
100
47
35
0
0
0
53
65
100
Thermal transformations of naphthosemibullvalenes
10a, 10b and 12e
It has been observed earlier that dibenzosemibullvalenes
containing dibenzoyl groups attached to the cyclopropane
ring undergo interesting isomerization to the corresponding
dibenzodihydropentalenofuran derivatives.7,9 In order to
examine the type of products formed in the case of naphtho-
semibullvalenes, a few representative examples were
subjected to thermolysis. The thermal isomerization of
10a in o-dichlorobenzene for 20 min gave a 82% of the
corresponding water added tetrahydropentalenofuran
derivative 14a, after puri®cation by column chromato-
graphy (Scheme 3). Similarly, the thermolysis of 10b and
12e, under analogous conditions, gave the water added tetra-
hydropentalenofuran derivatives 14b and 14e in 72% and
60% yields, respectively. The 1H NMR spectrum of 14a, for
example showed the tertiary hydroxyl proton at d 3.09, as a
singlet. The two allylic bridgehead protons were observed at
d 4.63 and 4.94, whereas the vinylic protons appeared at d
4.94 and 5.88, respectively as multiplets. The 13C NMR
spectrum of 14a showed the hemiacetal carbon at d
108.55, whereas the benzylic carbon, substituted with the
ether linkage was observed at d 85.10. The structure of 14a
was also con®rmed through DEPT-135 NMR technique.
Compounds 14b and 14e also displayed similar spectral
features. As a representative example, the structure of 14b
was further con®rmed through X-ray crystallographic
analysis.10
Irradiation of 6b, similarly gave the regiospeci®c isomer
10b in a 88% yield. The structure of 10b was also con®rmed
through X-ray crystallographic analysis.10 Interestingly,
irradiation of 9-acetyl-10-methylnaphthobarrelene 6c,
under analogous conditions, yielded both the regioisomers
10c (31%) and 12c (35%), along with 20% of the unchanged
starting material (Scheme 2). Yields of various regioisomers
calculated on the basis of 100% product balance are
1
summarized in Table 1 and were con®rmed by H NMR
spectroscopy monitoring of these photoreactions. The struc-
tures of the regioisomers 10c and 12c were arrived at on the
basis of spectral data and analytical evidence. The 1H NMR
spectrum of 10c showed the allylic bridgehead proton as a
doublet at d 4.74, whereas it appeared as a doublet at d 4.55
in its regioisomer 12c, showing a net shielding effect of
0.19 ppm. Similarly, the cyclopropane ring proton of 10c
appeared as a singlet at d 4.62, whereas it was observed at d
4.75 in the case 12c, indicating a net deshielding effect of
0.13 ppm.
Irradiation of 9-benzoyl-10-cyclopentylnaphthobarrelene
6d, similarly, gave a mixture of regioisomers 10d (23%)
and 12d (43%) in the ratio of 1:1.9. The spectral patterns
of 10d and 12d closely resemble those of the regioisomers
10c and 12c. The allylic bridgehead proton of 10d appeared
as a doublet at d 4.70, while it appeared as a doublet at d
4.38 in the case of 12d with a net shielding of 0.32 ppm.
Similarly, the cyclopropane ring proton of 10d appeared as a
singlet at d 4.67, whereas it was observed along with one of
the vinylic protons at d 5.28 in 12d, showing a net deshield-
In order to con®rm the formation of naphthopentalenofurans
13a, 13b and 13e as primary products in these reactions,
control experiments were carried out in NMR tubes under
dry conditions. For example, compound 10a was subjected
to neat thermolysis at 1808C for 2 min in a NMR tube in an
oil bath. The solvent CDCl3 was added afterwards and
1
recorded the H and 13C NMR spectra. Spectral analysis
showed the formation of the corresponding dihydro-
pentalenofuran derivative 13a as the exclusive product.
The 1H NMR spectrum showed the allylic bridgehead
proton at d 4.63 whereas the adjacent vinylic proton
appeared as a multiplet at d 5.80. The other vinylic proton
appeared as a doublet at d 6.46, while the benzylic proton of
the pentalenofuran ring was observed at d 6.43 as a singlet.
The 13C NMR spectrum showed the carbonyl carbons at d
198.68 and 199.77. In a separate experiment, when 10b was
thermolyzed and analyzed as in the case of 10a, it showed
the formation of the corresponding dihydropentalenofuran
derivative 13b as the exclusive primary product.
Discussion
The mechanism for the formation of regioisomeric products
10 and 12 in the phototransformations of naphthobarrelenes
6a±e can be rationalized in terms of the pathways shown in
Scheme 3.

5424
M. C. Sajimon et al. / Tetrahedron 56 (2000) 5421±5428
orbital calculations13 were carried out for the model
diradical intermediates 15a±f (Table 2, Chart 1). These
calculations were performed, subject to constraints, for the
maximum steric effects by the substituents on the hydrogen
atom of the radical center. A quantitative account of the
extent of distortion in planarity of the radical center can
be obtained from the torsional angle abcd. It was found
that, as the bulkiness of the substituent increases the
torsional angle also increases, thereby indicating a net
distortion in planarity. For example, the model diradical
intermediate 15a (corresponding 8a), where the methyl
group is syn to the radical center exhibited a dihedral
angle of 0.498, whereas the model diradical intermediate
15d (corresponding 11a) showed a dihedral angle of
0.948, indicating a higher extent of destabilization in the
latter case.
Table 2. The torsional angles associated with the diradical intermediates
15a±f, generated by AM1 semi-empirical molecular orbital calculations
(subject to constraints, for the maximum steric effects by the substituents
on the radical center)
Entry
Torsional angle, abcd (8)
15a
15b
15c
15d
15e
15f
0.49
0.40
0.59
0.94
9.68
12.17
Interestingly, in the case of 6c, where both the regioisomers
(10c and 12c) are formed in nearly equal amounts, both the
pathways are equally favored. In contrast, in the case of 6d,
the product (12d) arising through path B was found to be
nearly double that of the product (10d) arising through path
A. This again supports the view that with increase in the
bulkiness of the substituent (cyclopentyl) the diradical inter-
mediate 8 is destabilized, thereby favoring a pathway
mediated through the diradical intermediate 11. The AM1
calculations for the respective model diradical 15e, showed
a torsional angle of 9.688, in support of this view. Electronic
effects such as electron withdrawing ability of the C-9
substituent in controlling the regioselectivity can be ruled
out in these cases since the photoproduct distribution in
6a±d are signi®cantly different. Additional support for the
role of steric effects in controlling regioselectivity was
derived through the photorearrangement of the naphtho-
barrelene 6e, having benzoyl and phenyl groups at 9 and
10 positions, respectively. Depending on the excited state
orientation, the phenyl group can exert effective steric inter-
ference (torsional angle 12.178) through the perihydrogens14
(Chart 1, 15f). It is therefore not surprising to note that the
photoisomerization of 6e gave exclusively the regioisomer
12e, channeled through path B.
Chart 1.
Scheme 2. As mentioned earlier, 2,3-naphthobarrelenes
6a±e, containing 1,2-dibenzoylalkene moieties, undergo
photoisomerization, channeled exclusively through an
initial vinyl±vinyl bonding. This bonding can proceed
either through a [3,11] bridging (path A) or [2,12] bridging
(path B) leading to the primary diradical intermediates 7 and
9, respectively. These diradicals can subsequently undergo
transformation to the corresponding secondary diradical
intermediates 8 and 11 and leading ultimately to the regio-
isomeric naphthosemibullvalenes 10 and 12, respectively.
The observed regioselectivity in these cases may be
dependent on the stabilities of the primary (7 and 9) and
secondary (8 and 11) diradical intermediates, respectively.
In the case of the primary diradicals 7 and 9, the effects of
the substituents are not expected to be signi®cant since the
substituents are far removed from the radical site. Therefore,
the observed regioselectivity in these reactions may be
directly related to the stabilities of the secondary 1,3-
diradical intermediates 8 and 11, wherein one can anticipate
signi®cant steric effects of the substituents.
The thermal isomerization of naphthosemibullvalenes 10a,
10b and 12e is analogues to that of the dibenzosemibullva-
lenes which are reported to give dihydropentalenofuran
derivatives.7,9 However, the naphthodihydropentaleno-
furans formed in the thermolysis of the corresponding semi-
bullvalenes were found to readily add a molecule of water,
under the conditions of work-up, yielding the water-added
products. It is interesting to note that a 1,2 addition of water
is the preferred mode and that the ease of addition may
re¯ect the release of strain in going from 13a,b,e to the
corresponding water added products 14a,b,e.
If we consider the diradical intermediate 11a derived from
6a, for example, one can expect some amount of steric
interference between the bulky benzoyl substituent at the
C-8 position and the hydrogen atom at the radical center
at 2b position. The consequence of this will be a distortion
in the planarity of the radical center, thereby resulting in a
net destabilization of the diradical intermediate 11a11,12 (i.e.
the extent of delocalization of the free electron in the p
orbital of the sp2 hybridized carbon atom with the naphtha-
lene moiety is decreased due to the distortion in planarity).
In the case of diradical intermediate 8a, however, such steric
effects are minimal as the methyl group at the C-8 position is
well separated from the radical center. Therefore, the
reaction of 6a is channeled through path `a' involving the
stable diradical intermediate 8a. A similar rationalization
holds good for the formation of 10b from 6b.
Conclusions
The observed regioselectivity in the photoisomerization of
naphthobarrelenes 6a±e can be rationalized in terms of the
relative stabilities of the diradical intermediates (8a±e and
11a±e) derived from the initially formed primary diradical
intermediates (7a±e and 9a±e). It was observed that steric
effects of the nearby substituents to the radical centre play a
In order to have a qualitative picture of the steric effects
exerted by the substituents, AM1 semi-empirical molecular

M. C. Sajimon et al. / Tetrahedron 56 (2000) 5421±5428
5425
dominant role with respect to the stabilities of 8a±e and
11a±e. AM1 semi-empirical calculations of the respective
model diradicals support this view. Thermal isomerization
of representative naphthosemibullvalenes 10a, 10b and 12e
led to the corresponding naphthodihydropentalenofurans
13a,b,e in nearly quantitative yields.
9-Benzoyl-10-methoxynaphthobarrelene (6b). Reaction
of 9-benzoyl-10-methoxyanthracene (1.56 g, 5 mmol) with
DBA (1.17 g, 5 mmol) in the presence of AlCl3 (0.67 g,
5 mmol) in dichloromethane (40 mL) and work-up as in
the earlier case gave 624 mg (40%) of the unchanged start-
ing anthracene derivative, mp 174±1758C (mixture mp),
followed by 625 mg (20%) of the dibenzobarrelene deriva-
tive, mp 246±2478C (mixture mp) and 950 mg (35%) of 6b,
mp 187±1888C, after recrystallization from cyclohexane: IR
nmax (KBr) 1666 cm21; UV lmax (CH3CN) 240 nm (e
Experimental
1
The equipment and procedure for melting point determi-
nation and spectral recordings are described in earlier publi-
cations.15 All steady-state irradiation experiments were
carried out in a Srinivasan±Grif®n±Rayonet Photochemical
Reactor (RPR 300 nm) or by using a Pyrex ®ltered light
from a Hanovia 450 W medium pressure mercury lamp,
under inert conditions. Solvents for photolysis experiments
were puri®ed and distilled before use. Petroleum ether used
was the fraction with a bp 60±808C.
76,700); H NMR (CDCl3) d 4.18 (3H, s), 5.21 (1H, d,
Jˆ5.7 Hz), 5.95 (1H, d, Jˆ5.8 Hz), 7.02±8.19 (21H, m);
13C NMR (CDCl3) d 45.57, 48.89, 63.13, 122.36, 128.14,
128.27, 128.42, 129.75, 133.80, 139.08, 150.84, 152.09,
152.88, 193.08, 194.59, 197.76; exact mol wt calcd for
C38H26O4: 546.1831. Found: 546.1820 (high-resolution
mass spectrometry). Anal. Calcd for C38H26O4: C, 83.52;
H, 4.76. Found: C, 83.28; H, 4.84.
9-Acetyl-10-methylnaphthobarrelene (6c). A mixture of
9-acetyl-10-methylanthracene (1 g, 5 mmol) and DBA (1 g,
5 mmol) was well powdered and heated in an oil bath at
180±2008C for 20 min. The crude reaction mixture was
recrystallized from a mixture (1:4) of methanol and
dichloromethane to give 800 mg (40%) of the correspond-
ing dibenzobarralene derivative, mp 226±2278C (mixture
mp). The mother liquor was chromatographed over silica
gel. Elution with a mixture (1:4) of ethyl acetate and hexane
gave 400 mg (40%) of the unchanged anthracene derivative
mp 136±1378C (mixture mp). Further elution gave 150 mg
(8%) of 6c, mp 209±2108C, after recrystallization from a
mixture (4:1) of dichloromethane and methanol: IR nmax
(KBr) 1730, 1667 cm21; UV lmax (CH3CN) 251 nm (e
Starting materials
Dibenzoylacetylene (DBA),16 mp 110±1118C; 9-benzoyl-
10-methylanthracene,17 mp 175±1768C, 9-benzoyl-10-
methyoxyanthracene,18 mp 174±1758C, 9-acetyl-10-methyl-
anthracene,19 mp 135±1368C; 9-benzoyl-10-cyclopentyl-
anthracene,19 mp 135±1368C and 9-benzoyl-10-phenyl-
anthracene,20 mp 174±1758C were prepared by reported
procedures.
Preparation of the naphthobarrelene derivatives 6a±e
The naphthobarrelenes were prepared through the Diels±
Alder reaction of the appropriate anthracene derivatives
with DBA under neat heating or in the presence of
anhydrous aluminium chloride. Small amounts of the corre-
sponding dibenzobarrelenes were also isolated, in each case,
under these conditions.
1
72,300); H NMR (CDCl3) d 2.73 (3H, s), 2.76 (3H, s),
5.46 (1H, d, Jˆ4.2 Hz), 5.85 (1H, d, Jˆ4.2 Hz), 7.09±
8.04 (16H, m); 13C NMR (CDCl3) d 14.50, 33.63, 48.00,
48.73, 124.86, 125.05, 126.37, 126.51, 126.74, 128.32,
128.44, 128.56, 130.58, 132.92, 132.96, 133.27, 136.23,
137.66, 137.73, 138.23, 138.31, 139.12, 152.36, 152.56,
194.31, 194.42, 206.57; exact mol wt calcd for C33H24O3:
469.1804. Found: 469.1820 (high-resolution mass
spectrometry). Anal. Calcd for C33H24O3: C, 84.61; H,
5.13. Found: C, 84.92; H, 5.00.
9-Benzoyl-10-methylnaphthobarrelene (6a). To a mixture
of 9-benzoyl-10-methylanthracene (0.74 g, 2.5 mmol) and
AlCl3 (310 mg, 2.5 mmol) in dichloromethane (20 mL) was
added DBA (0.58 g, 2.5 mmol) in dichloromethane (20 mL)
and stirred for 1 h at room temperature. The reaction
mixture was acidi®ed with hydrochloric acid (6.5 N,
200 mL) and extracted with dichloromethane. Removal of
the solvent gave a solid residue, which was chromato-
graphed over silica gel. Elution with a mixture (1:9) of
ethyl acetate and hexane gave 148 mg (20%) of the
unchanged anthracene derivative, mp 176±1778C (mixture
mp). Further elution with a mixture (1:4) of ethyl acetate
and hexane gave 157 mg (12%) of the corresponding
dibenzobarrelene derivative, mp 226±2278C (mixture mp),
followed by 750 mg (57%) of 6a, mp 203±2048C: IR nmax
(KBr) 1668 cm21; UV lmax (CH3CN) 238 nm (e 73,800);
1H NMR (CDCl3) d 2.90 (3H, s), 5.18 (1H, d, Jˆ5.4 Hz),
5.87 (H, d, Jˆ5.4 Hz), 7.00±8.10 (21H, m); 13C NMR
(CDCl3) d 14.52, 48.22, 49.02, 128.21, 128.42, 128.47,
129.74, 138.74, 138.82, 152.23, 152.89, 193.27, 194.60,
198.35; exact mol wt calcd for C38H26O3: 530.1882.
Found: 530.1876 (high-resolution mass spectrometry).
Anal. Calcd for C38H26O3: C, 86.04; H, 4.90. Found: C,
85.93; H, 5.00.
9-Benzoyl-10-cyclopentylnaphthobarrelene (6d). A well
powdered mixture of 9-benzoyl-10-cyclopentyl-anthracene
(700 mg, 2 mmol) and DBA (600 mg, 2.5 mmol) was
heated in an oil bath at 180±2008C for 90 min. The solid
residue was chromatographed over silica gel. Elution with a
mixture (1:4) of ethyl acetate and hexane gave 200 mg
(28%) of the unchanged anthracene derivative, mp 173±
1748C (mixture mp) followed by 150 mg (15%) of the
adduct 6d, mp 231±2328C, after recrystallization from a
mixture (1:1) of dichloromethane and methanol: IR nmax
(KBr) 1681 cm21; UV lmax (CH3CN) 258 nm (e 73,100);
1H NMR (CDCl3) d 1.90±2.13 (8H, m), 4.23 (1H, m),
5.19 (1H, d, Jˆ5.6 Hz), 5.98 (1H, d, Jˆ5.2 Hz), 7.05±
8.22 (21H, m); 13C NMR (C6D6) d 27.76, 27.87, 33.65,
33.98, 40.00, 48.99, 49.68, 125.13, 126.07, 127.10,
127.70, 128.34, 128.91, 129.77, 130.06, 132.65, 133.53,
138.82, 152.29, 152.66, 192.99, 194.60, 198.03. Anal.
Calcd for C42H32O3: C, 86.30; H, 5.48. Found: C, 86.60;
H, 5.72.

5426
M. C. Sajimon et al. / Tetrahedron 56 (2000) 5421±5428
9-Benzoyl-10-phenylnaphthobarrelene (6e). To a mixture
of 9-benzoyl-10-phenylanthracene (1.79 g, 5 mmol) and
anhydrous AlCl3 (0.7 g, 5 mmol) in dichloromethane
(30 mL) was added a solution of DBA (1.17 g, 5 mmol) in
dichloromethane (20 mL). The mixture was stirred for 1.5 h
at room temperature, acidi®ed with hydrochloric acid
(6.5 N, 500 mL) and extracted with dichloromethane.
Removal of the solvent gave a solid residue, which was
chromatographed over silica gel. Elution with a mixture
(1:9) of ethyl acetate and hexane gave 800 mg (44%) of
the unchanged anthracene derivative, mp 218±2198C
(mixture mp). Further elution with a mixture (1:4) of ethyl
acetate and hexane gave 600 mg (40%) of 6e, mp 213±
2148C: IR nmax (KBr) 1664, 1652 cm21; UV lmax
(CH3CN) 241 nm (e 73,300); 1H NMR (CDCl3) d 5.3
(2H, m), 7.01±7.98 (26H, m); 13C NMR (CDCl3) d 48.63,
49.05, 125.33, 126.20, 126.44, 126.85, 127.66, 127.99,
128.08, 128.23, 128.35, 128.56, 128.64, 129.66, 130.11,
130.43, 131.09, 132.52, 133.77, 135.15, 136.97, 137.21,
138.31, 138.94, 151.08, 153.56, 192.79, 194.46, 198.01;
mass spectrum m/e (relative intensity) 592 (M1, 21), 487
(16), 105(100). Anal. Calcd for C43H28O3: C, 87.16; H, 4.73.
Found: C, 87.38; H, 4.77.
acetate and hexane gave 55 mg (35%) of 12c, mp 198±
1998C, after recrystallization from a mixture (1:1) of
dichloromethane and methanol: IR nmax (KBr) 1731,
1657 cm21; UV lmax (CH3CN) 239 nm (e 71,700); 1H
NMR (CDCl3) d 2.65 (3H, s), 2.85 (3H, s), 4.55 (1H, d,
Jˆ2.1 Hz), 4.75 (1H, s), 5.58 (1H, d, Jˆ5.1 Hz), 6.06
(1H, m), 7.08±8.11 (14H, m); 13C NMR (CDCl3) d 16.17,
32.77, 47.95, 58.21, 65.37, 71.17, 123.05, 124.56, 125.06,
125.87, 126.32, 126.52, 128.28, 128.45, 128.61, 128.81,
129.31, 132.46, 132.99, 133.25, 133.63, 134.60, 136.01,
136.68, 143.92, 195.00, 195.27, 204.75; exact mol wt
calcd for C33H24O3: 468.1725. Found: 468.1728 (high-
resolution mass spectrometry). Anal. Calcd for C33H24O3:
C, 84.61; H, 5.13. Found: C, 84.49; H, 5.01.
Further elution gave 30 mg (20%) of the unchanged starting
naphthobarrelene 6c, mp 209±2108C (mixture mp),
followed by 45 mg (31%) of 10c, mp 189±1908C, after
recrystallization from a mixture (1:1) of dichloromethane
and methanol: IR nmax (KBr) 1732, 1654 cm21; UV lmax
1
(CH3CN) 243 nm (e 71,200); H NMR (CDCl3) d 2.65
(3H, s), 2.77 (3H, s), 4.62 (1H, s), 4.74 (1H, d, Jˆ2.2 Hz),
5.59 (1H, d, Jˆ5.2 Hz), 5.99 (1H, m), 7.39±8.02 (16H, m);
13C NMR (CDCl3) d 14.99, 32.77, 48.42, 57.09, 65.37,
71.73, 123.16, 124.19, 125.28, 126.24, 126.32, 128.19,
128.42, 128.62, 128.74, 129.11, 129.56, 132.01, 132.04,
133.07, 133.18, 134.78, 136.53, 145.67, 194.81, 195.18,
204.62; exact mol wt calcd for C33H24O3: 468.1725.
Found: 468.1740 (high-resolution mass spectrometry).
Anal. Calcd for C33H24O3: C, 84.61; H, 5.13. Found: C,
84.38; H, 5.11.
Photolysis of 6a. A solution of 6a (500 mg, 0.9 mmol) in
benzene (300 mL) was irradiated (Hanovia 450 W) for
60 min and removal of the solvent under vaccum gave a
residual solid, which was chromatographed over silica gel.
Elution with a mixture (1:4) of ethyl acetate and hexane
gave 450 mg (90%) of 10a, mp 180±1818C, after recrystal-
lization from acetonitrile: IR nmax (KBr) 1670 cm21; UV
1
lmax (CH3CN) 239 nm (e 74,900); H NMR (CDCl3) d
2.90 (3H, s), 4.42 (1H, d, Jˆ2.4 Hz), 4.82 (1H, s), 5.57
(1H, d, Jˆ5.3 Hz), 6.01 (1H, m), 7.25±7.96 (19H, m); 13C
NMR (CDCl3) d 16.13, 48.08, 57.88, 65.09, 71.69, 123.02,
128.36, 128.48, 128.72, 129.93, 136.59, 145.33, 195.10,
197.84; exact mol wt calcd for C38H26O3: 530.1882.
Found: 530.1871 (high-resolution mass spectrometry).
Anal. Calcd for C38H26O3: C, 86.04; H, 4.90. Found: C,
85.76; H, 4.87.
Photolysis of 6d. A solution of 6d (150 mg, 0.20 mmol) in
benzene (50 mL) was irradiated using RPR 300 nm for
60 min. Removal of the solvent under vacuum gave a
residual solid, which was chromatographed over silica gel
using a chromatotron. Elution with a mixture (1:9) of ethyl
acetate and petroleum ether gave (65 mg, 43%) of 12d, mp
211±2128C, after recrystallization from a mixture (1:1) of
dichloromethane and methanol: IR nmax (KBr) 1664 cm21
;
1
UV lmax (CH3CN) 265 nm (e 72,900); H NMR (C6D6) d
1.92±2.48 (8H, m), 3.93 (1H, m), 4.38 (1H, d, Jˆ1.7 Hz),
5.28 (2H, s), 5.88 (1H, m), 6.81±8.13 (19H, m); 13C NMR
(C6D6) d 27.17, 27.64, 32.05, 34.64, 41.30, 50.00, 57.43,
64.27, 75.04, 123.82, 124.72, 125.68, 126.19, 127.70,
128.34, 128.93, 129.43, 130.06, 131.42, 132.51, 133.60,
136.47, 138.17, 143.15, 146.76, 194.51, 194.90, 197.61.
Anal. Calcd for C42H32O3: C, 86.30; H, 5.48. Found: C,
86.60; H, 5.49.
Photolysis of 6b. A benzene solution (100 mL) of 6b
(200 mg, 0.4 mmol) was irradiated using RPR 300 nm for
60 min and work-up of the reaction mixture as in the earlier
case gave 176 mg (88%) of 10b, mp 183±1848C, after
recrystallization from acetonitrile: IR nmax (KBr)
1668 cm21; UV lmax (CH3CN) 245 nm (e 45,600); 1H
NMR (C6D6) d 3.91 (3H, s), 4.41 (1H, d, Jˆ2.3 Hz), 5.21
(1H, s), 5.23 (1H, d, Jˆ5.3 Hz), 5.94 (1H, m), 6.87±8.12
(19H, m); 13C NMR (C6D6) d 48.11, 57.93, 60.93, 64.32,
74.98, 120.28, 123.10, 123.89, 125.64, 126.01, 127.22,
127.70, 128.34, 128.61, 128.90, 130.06, 130.18, 132.58,
132.79, 133.46, 136.41, 137.11, 138.72, 149.44, 156.48,
194.50, 196.96; exact mol wt calcd for C38H26O4:
546.1831. Found: 546.1830 (high-resolution mass
spectrometry). Anal. Calcd for C38H26O4: C, 83.52; H,
4.76. Found: C, 83.27; H, 4.95.
Further elution gave a mixture of 6d and 10d, which were
separated by fractional recrystallization from a mixture
(2:1) of dichloromethane and acetonitrile to give 30 mg
(20%) of 6d, mp 231±2328C (mixture mp) and 35 mg
(23%) of 10d, mp 203±2048C: IR nmax (KBr) 1667 cm21
;
1
UV lmax (CH3CN) 261 nm (e 70,700); H NMR (C6D6) d
2.21±2.91 (8H, m), 3.40 (1H, m), 4.67 (1H, s), 4.70 (1H, d,
Jˆ1.5 Hz), 5.32 (1H, d, Jˆ5.1 Hz), 5.62 (1H, m), 6.64±7.91
(19H, m); 13C NMR (C6D6) d 27.74, 27.85, 33.64, 33.98,
42.53, 48.77, 58.25, 65.13, 71.48, 123.17, 127.69, 128.33,
128.90, 133.54, 138.81, 139.55, 146.74, 152.31, 152.66,
194.21, 194.59, 198.04. Anal. Calcd for C42H32O3: C,
86.30; H, 5.48. Found: C, 86.25; H, 5.38.
Photolysis of 6c. A solution of 6c (150 mg, 0.23 mmol) in
benzene (50 mL) was irradiated using RPR 300 nm for
90 min. Removal of the solvent under vacuum gave a
residual solid, which was chromatographed over silica gel
using a chromatotron. Elution with a mixture (1:9) of ethyl

M. C. Sajimon et al. / Tetrahedron 56 (2000) 5421±5428
5427
Photolysis of 6e. A benzene solution (300 mL) of 6e
(590 mg, 1 mmol) was irradiated (Hanovia 450 W) for
90 min. Removal of the solvent under vaccum gave a
residual solid, which was chromatographed over silica gel.
Elution with a mixture (1:9) of ethyl acetate and petroleum
ether gave (342 mg, 58%) of 12e, mp 209±2108C, after
recrystallization from a mixture (1:1) of chloroform and
methanol: IR nmax (KBr) 3056, 1686, 1680 cm21; UV
135.27, 137.86, 144.25, 155.37, 198.20, 199.80; exact mol
wt calcd for C38H28O5: 564.1937. Found 564.1941. (high-
resolution mass spectrometry). Anal. Calcd for C38H28O5: C,
80.45; H, 4.96. Found: C, 80.51; H, 4.86.
14e (60%), mp 218±2198C : IR nmax (KBr) 1667 cm21; UV
l
max (CH3OH) 257 nm (e 74,300); 1H NMR (CDCl3) d 3.96
(1H, s, D2O exchangeable), 4.16 (1H, m), 4.59 (3H, s), 4.90
(1H, m), 5.84 (1H, m), 6.16 (1H, s), 7.19±7.84 (19H, m);
13C NMR (CDCl3) d 62.32, 68.14, 73.85, 85.11, 108.98,
125.63, 125.86, 126.26, 126.66, 127.34, 127.74, 128.00,
128.35, 128.55, 128.83, 129.98, 130.80, 131.54, 132.08,
133.03, 133.90, 135.33, 140.32, 198.61, 199.64; 13C NMR
(DEPT-135, CDCl3) d 62.36, 68.00, 85.81, 125.67, 126.30,
126.71, 128.05, 128.41, 128.60, 128.89, 130.02, 131.58,
133.14135.40. Anal. Calcd for C43H30O4: C, 84.59; H,
4.92. Found: C, 84.96; H, 5.25.
1
lmax (CH3CN) 241 nm (e 66,600); H NMR (CDCl3) d
4.38 (2H, m), 5.52 (1H, d, Jˆ5.2 Hz), 5.78 (1H, m),
7.25±7.74 (24H, m, aromatic); 13C NMR (CDCl3) d
48.55, 57.53, 64.95, 71.84, 122.79, 126.37, 128.57,
128.82, 128.87, 135.81, 137.75, 145.65, 194.75, 194.96,
197.79. mass spectrum m/e (relative intensity) 592 (M1,
20), 487 (14), 105 (100). Anal. Calcd for C43H28O3: C,
87.16; H, 4.73. Found: C, 86.93; H, 5.00.
Further elution gave 177 mg (30%) of the unchanged
naphthobarrelene derivative 6e, mp 213±2148C (mixture
mp).
NMR monitoring of the thermal transformations of 10a
and 10b
1H NMR monitoring of the photoreactions of 6a±e
Thermal transformations of representative naphthosemi-
bullvalenes 10a and 10b (20 mg) to the corresponding
naphthopentalenofurans 13a and 13b were achieved by
carrying out the neat thermolysis in NMR tubes in an oil
bath maintained at 1808C for 2 min. CDCl3 (0.5 mL) was
immediately added to the thermolyzed mixtures and
Solutions of 6a±e (10±20 mg) in CDCl3 or C6D6 (0.5 mL)
in NMR tubes were irradiated with RPR 300 nm light source
for ,30 min and analyzed the product mixture by 1H NMR
(300 MHz). The 1H NMR spectrum, in each case, con®rmed
the ratios of photoproducts as reported earlier under the
section dealing with steady-state photolysis, and thereby
ruling out the possible formation of any other products in
these phototransformations.
1
1
recorded both the H and 13C NMR spectra. The H and
13C NMR spectra revealed the formation of only the corre-
sponding naphthopentalenofurans in these reactions.
1
13a. H NMR (CDCl3) d 2.99 (3H, s), 4.63 (1H, s), 5.80
Thermal transformations of 10a, 10b and 12e
(1H, m), 6.43 (1H, s), 6.46 (1H, d, Jˆ5.4 Hz), 7.28±7.89
(19H, m); 13C NMR (CDCl3) d 15.00, 55.60, 78.01, 87.85,
122.85, 126.62, 128.00, 128.70, 128.87, 133.01, 137.21,
148.38, 198.68, 199.77.
Thermal transformations of representative naphthosemi-
bullvalenes 10a, 10b and 12e (0.2 mmol) to the correspond-
ing water added naphthopentaleneofurans 14a, 14b and 14e
were achieved by re¯uxing in o-dichlorobenzene (20 mL)
for 20 min. The solvent was removed under vacuum and the
residual solid thus obtained was chromatographed over
silica gel using a chromatotron. Elution with a mixture
(1:5) of ethyl acetate and hexane and recrystallization
from a mixture (1:1) of acetonitrile and dichloromethane
gave the desired product.
1
13b. H NMR (CDCl3) d 4.38 (3H, s), 4.62 (1H, s), 5.82
(1H, m), 6.48 (1H, d, Jˆ5.2 Hz), 6.60 (1H, s), 7.25±8.31
(19H, m); 13C NMR (CDCl3) d 52.39, 61.30, 77.47, 85.51,
121.96, 125.40, 127.54, 127.66, 127.86, 132.89, 135.91,
146.73, 155.33, 197.30, 198.51.
X-Ray structure determination of 10a, 10b, 12e and 14b
14a (82%), mp 239±2408C: IR nmax (KBr) 1665 cm21; UV
l
Good quality crystals of 10a, 10b, 12e and 14b were
mounted on glass ®bers in random orientations and
subjected to X-ray crystallographic analysis, employing a
Siemens R3 automated four circle diffractometer. Data
reduction and structure solution were achieved by
shelxtl-plus (VMS) structure solution package.10a Details
of the X-ray structures may be found at the Cambridge
Crystallographic Data Center.10b
max (CH3OH) 249 nm (e 64,500); 1H NMR (CDCl3) d 2.93
(3H, s), 3.09 (1H, s, D2O exchangeable), 4.06 (1H, s), 4.63
(1H, s), 4.94 (1H, m), 5.88 (1H, m), 6.69 (1H, s), 7.25±8.15
(19H, m); 13C NMR (CDCl3) d 14.88, 61.29, 67.15, 72.37,
85.10, 108.55, 128.28, 127.43, 127.58, 127.73, 132.72,
136.79, 140.40, 197.56, 199.23; 13C NMR (DEPT-135,
CDCl3) d 14.00, 61.39, 66.29, 84.17, 124.37, 126.11,
126.54, 126.69, 126.84, 127.65, 128.01, 131.23, 131.84,
133.84. Anal. Calcd for C38H28O4: C, 83.21; H, 5.11.
Found: C, 83.48; H, 5.45.
Acknowledgements
14b (72%), mp 246±2478C: IR nmax (KBr) 1661 cm21; UV
l
(1H, s, D2O exchangeable), 3.98 (1H, s), 4.31 (3H, s), 4.54
(1H, s), 4.86 (1H, m), 5.82 (1H, m), 6.76 (1H, s), 7.16±8.26
(19H, m); 13C NMR (CDCl3) d 60.97, 62.89, 73.84, 83.60,
85.73, 109.35, 122.07, 127.30, 127.56, 129.59, 129.70,
We (M. C. S., D. R., S. A. K. and M. V. G.) thank the
Council of Scienti®c and Industrial Research, Department
of Science and Technology, Government of India, the
Regional Research Laboratory, Trivandrum, the
Volkswagen Foundation, Germany (D. R.), the University
of MissouriÐSt. Louis (N. P. R.) and the Jawarharlal
max (CH3OH) 251 nm (e 69,500); 1H NMR (CDCl3) d 3.05

5428
M. C. Sajimon et al. / Tetrahedron 56 (2000) 5421±5428
Kutateladze, A. G.; Mackawa, Y.; Mangette, J. E. J. Am. Chem.
Soc. 1994, 116, 9795±9796.
Nehru Centre for Advanced Scienti®c Research, Bangalore
(M. V. G.) for ®nancial support of this work.
6. (a) Ciganek, E. J. Am. Chem. Soc. 1966, 88, 2882±2883.
(b) Rabideau, P. W.; Hamilton, J. B.; Friedmann, L. J. Am.
Chem. Soc. 1968, 90, 4465±4465. (c) Zimmerman, H. E.;
Grunewald, G. L. J. Am. Chem. Soc. 1966, 88, 183±184.
7. Sajimon, M. C.; Ramaiah, D.; Muneer, M.; Ajithkumar, E. S.;
Rath, N. P.; George, M. V. J. Org. Chem. 1999, 64, 6347±6352.
8. Zimmerman, H. E.; Amick, D. R.; Hemetsberger J. Am. Chem.
Soc. 1973, 95, 4606±4610.
References
1. Contribution No. RRLT-PRU-120 from the Regional Research
Laboratory (CSIR), Trivandrum, India.
2. For some recent reviews on the di-p-methane rearrangement of
barrelenes see: (a) Zimmerman, H. E. In Rearrangements in
Ground and Excited States; de Mayo, P., Ed.; Academic: New
York, 1980; Vol. 3, Chapter 16, pp 131±166. (b) Zimmerman,
H. E. In Organic Photochemistry; Padwa, A., Ed.; Marcell Dekker:
New York, 1991; Vol. 11, pp 1±36. (c) Scheffer, J. R.; Pokkuluri,
P. R. In Photochemistry in Organized and Constrained Media;
Ramamurthy, V., Ed.; VCH: New York, 1991; pp 185±246.
(d) de Lucchi, O.; Adam, W. In Comprehensive Organic Synthesis;
Trost, B. M., Paquette, L. A., Eds.; Pergamon: Oxford, 1991;
Vol. 5, pp 193±214.
9. Pratapan, S.; Ashok, K.; Gopidas, K. R.; Rath, N. P.; Das, P. K.;
George, M. V. J. Org. Chem. 1990, 55, 1304±1308.
10. (a) Sheldrick, G. M. Siemens Analytical X-ray Division,
Madison, WI, 1991. (b) CCDC Nos. 10a (Compound No. in the
text): 143874; 10b: 143875; 12e: 143872; 14b: 143873.
11. RuÈchardt, C. Top. Curr. Chem. 1980, 88, 1±32.
12. Paquette, L. A.; Bay, E. J. Am. Chem. Soc. 1984, 106, 6693±
6701.
13. Evaluated using AM1 semi-empirical method of PC Spartan
software obtained from Wavefunction, Inc., 18401 Von Karman,
Suite 370, Irvine, CA 92612, USA.
3. (a) Zimmerman, H. E.; Givens, R. S.; Pagni, R. M. J. Am. Chem.
Soc. 1968, 90, 4191±4193. (b) Zimmerman, H. E.; Givens, R. S.;
Pagni, R. M. J. Am. Chem. Soc. 1968, 90, 6096±6108. (c) Brewers,
J. P. N.; Heaney, H. J. J. Chem. Soc. Chem. Commun. 1967, 811±
812. (d) Bender, C. O.; Brooks, D. W.; Cheng, W.; Dolman, D.;
O'Shea, S. F.; Sugaeman, S. Can. J. Chem. 1979, 56, 3027±3037.
(e) Bender, C. O.; Bengton, D. L.; Dolman, D.; Herle, L. E. L.;
O'Shea, S. F. Can. J. Chem. 1982, 60, 1942±1952.
14. Sabacky, M. J.; Johnson Jr., C. S.; Smith, R. G.; Gutowsky,
H. S.; Martin, J. C. J. Am. Chem. Soc. 1967, 89, 2054.
15. (a) Junjappa, H.; Saxene, M. K.; Ramaiah, D.; Loharay, B. B.;
Rath, N. P.; George, M. V. J. Org. Chem. 1998, 63, 9801±9805.
(b) Ramaiah, D.; Muneer, M.; Gopidas, K. R.; Das, P. K.; Rath,
N. P.; George, M. V. J. Org. Chem. 1996, 61, 4240±4246.
16. (a) Lutz, R. E. In Org. Synth.; Horning, E. C., Ed.; Wiley: New
York, 1995; Collect. Vol. 3, pp 248±250. (b) Lutz, R. E.; Smithy
Jr., W. R. J. Org. Chem. 1951, 16, 648±654.
4. (a) Chen, J.; Scheffer, J. R.; Trotter, J. Tetrahedron 1992, 48,
3251±3274. (b) Zimmerman, H. E; Sulzbach, H. M.; Tollefson,
M. B. J. Am. Chem. Soc. 1993, 115, 6548±6556. (c) Scheffer, J. R.;
Yang, J. In CRC Handbook of Organic Photochemistry and Photo-
biology; Horspool, W. M., Song, P. S., Eds.; CRC: Boca Raton, FL,
1995; Chapter 16, pp 204±221.
17. Bokava, A. I.; Buchina, I. K. Zh. Org. Khim. 1984, 20, 1318±
1321.
18. Krollpfeiffer, F.; Branschield, F.; Thorn, F.; Schneider, K.
Ann. 1928, 462, 46±72.
5. (a) Zimmerman, H. E.; Bender, C. O. J. Am. Chem. Soc. 1969,
91, 7516±7518. (b) Zimmerman, H. E.; Bender, C. O. J. Am.
Chem. Soc. 1970, 92, 4366±4376. (c) Bender, C. O.; Elder,
J. L. M.; Herbst, A. J.; Miller, L. E. Can. J. Chem. 1972, 50,
395±401. (d) Zimmerman, H. E.; Viriot-Villaume, M. L. J. Am.
Chem. Soc. 1973, 95, 1274±1280. (e) Zimmerman, H. E.;
19. Merritt Jr., C.; Braun, C. E. In Organic Synthesis, Rabjohn, N.,
Ed.; Wiley: New York, 1967; Vol. 4, pp 8±10.
20. Gibson, S.; Mosnaim, A. D.; Nonhebel, D. C.; Russel, J. A.
Tetrahedron 1969, 25, 5047.