Synthesis of a ditopic cyclophane based on the cyclobutane ring by chalcone photocycloaddition

Article abstract of DOI:10.1016/S0040-4020(03)00475-7

The intramolecular photocycloaddition of chalcones to give cyclobutanes has proven to be a fast and simple method to shrink a cyclophane ring to a tricyclic system, in order to prepare potential ditopic receptors. In particular, the chalcone 1, having dioxyethylene chains as spacers, is converted in high yield to the cyclobutane 2. NOESY spectroscopy indicates that the formation of 2 occurs by a head-to-head syn ring closure.

Full text of DOI:10.1016/S0040-4020(03)00475-7

Tetrahedron 59 (2003) 3455–3459
Synthesis of a ditopic cyclophane based on the cyclobutane ring by
chalcone photocycloaddition
*
Francesca R. Cibin, Giancarlo Doddi and Paolo Mencarelli
*
`
Dipartimento di Chimica ed ICCOM-Sez di Roma, Universita di Roma ‘La Sapienza’, P.le Aldo Moro 5, 00185 Roma, Italy
Received 28 January 2003; revised 3 March 2003; accepted 21 March 2003
Abstract—The intramolecular photocycloaddition of chalcones to give cyclobutanes has proven to be a fast and simple method to shrink a
cyclophane ring to a tricyclic system, in order to prepare potential ditopic receptors. In particular, the chalcone 1, having dioxyethylene
chains as spacers, is converted in high yield to the cyclobutane 2. NOESY spectroscopy indicates that the formation of 2 occurs by a head-to-
head syn ring closure. q 2003 Elsevier Science Ltd. All rights reserved.
1. Introduction
2. Results and discussion
The cyclobutane ring may constitute a rigid nucleus for the
insertion of a variety of cyclic ligand moieties, in order to
build complex ditopic receptors.
The dioxyethylene substituted trans-benzalacetophenone
(chalcone) 1 has been synthesized according to the route
indicated in Scheme 2.
A fast method to obtain cyclobutane rings is the photo-
chemical dimerization of a,b-unsaturated carbonyl or
carboxyl compounds and in particular of benzalaceto-
phenones (chalcones). These reactions can be carried out
in solution, solid state and molten state by sunlight or UV–
vis irradiation, with variable results in terms of yield and
product composition.1–11
A 4£1023 mol dm23 chloroform solution of this compound,
that showed an intense UV absorption band centred at
Due to our interest in host–guest chemistry,12–16 we have
tested the feasibility to produce ditopic cyclophanes that
could act as receptors, exploiting the intramolecular
photocycloaddition of the cyclic chalcone 1 to cyclobutane
2 (Scheme 1).
Scheme 1.
Keywords: cyclophanes; enones; carbocycles; cycloaddition;
photochemistry.
*
Corresponding authors. Tel.: þ39-649913077; fax: þ39-6490631;
giancarlo.doddi@uniromal.it; gdoddi@uniroma1.it;
paolo.mencarelli@uniroma1.it
Scheme 2.
0040–4020/03/$ - see front matter q 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0040-4020(03)00475-7

3456
F. R. Cibin et al. / Tetrahedron 59 (2003) 3455–3459
334 nm, after ,1 h exposure to sunlight on a sunny day in
July, yielded 2 in 74% yield. The irradiation for ,20 min of
a 6.5£1023 mol dm23 chloroform solution in a Rayonet
apparatus, with fourteen 350 nm centred lamps, gave the
cyclobutane 2 in 92% yield. Thermal treatment in the
absence of light was ineffective.
by (H,H)-COSY, (H,C)-COSY and NOESY-Phase sensitive
spectra.
1H and 13C NMR spectra showed that the ortho and meta
protons (and carbons), of the phenylene groups directly
bonded to the cyclobutane ring of 2, are not chemically
equivalent (Figs. 3 and 4), because of the restricted rotations
of the aromatic rings due to the short spacer.†
The ready occurrence of this cyclization with respect to the
examples reported in Refs. 1–11, is related to its
intramolecular nature, that favors the approach of the
reacting groups. This feature was already observed in
the photocycloaddition of cinnamoyl derivatives, where the
photoactive groups were linked together by a short
spacer,17–19 and0 could account for the different behaviour
shown from 4,4 -dimethoxybenzalacetophenone, which is
electronically very similar to 2, but that under irradiation
yields only brown resin.3
Figure 3.
The cycloaddition of trans-chalcones may give four
possible stereoisomers, namely syn/anti head-to-head and
head-to-tail (Fig. 1). Since we had difficulties obtaining
1
suitable crystals of 2 for X-ray studies, we resorted to H
NMR spectrometry to gain information about its
stereochemistry.
Figure 4. 1H NMR of 2 in CDCl3 (detail).
The upfield shift of H-2d00 and to a lesser extent of H-3c00,
suggests that these protons are in some degree directed
inwards to the opposite phenylene group, experiencing its
ring current shielding effect.
Figure 1.
Two symmetrical multiplets (AA0BB0 system) were
observed for the cyclobutyl protons. Simulation of these
NMR patterns has allowed the estimation of the coupling
The non-chemical equivalence of the H-200 protons allowed
us to distinguish the through-space interactions of proton 2a00
from those of 2d00. In particular, the NOESY spectrum of 2
shows, inter alia, that the protons that we have named H-2a00,
have a through-space coupling exclusively with the protons
H-2, whereas the protons that we have named H-2d00, couple
through the space only with the protons H-1 (Fig. 5, filled
contours).
0
constants of the cyclobutyl protons (Fig. 2): JAA ¼11.3 Hz,
20
0
0
JAB¼6.3 Hz, JAB ¼-0.8 Hz, JBB ¼10.5 Hz.
These results are in accordance with a trans relationship
between protons H-1 and H-2, and with a cis relationship
between the two H-1 protons (and obviously between the
two H-2 protons), caused by a head-to-head syn junction
(Fig. 6, 2 h–h syn). Results obtained from 15 ns of
molecular dynamics simulations indicate (see Table 1)
that, for the 2 h–h syn stereoisomer, the atomic distances
between H-2a–H-2 and H-2d–H-1 are lower than 3.5 A,
whereas the distances H-2a00–H-1 and H-2d00 –H-2 are greater.
In this situation we should observe exactly the correlation
pattern reported in Figure 5.
00
00
Figure 2. Simulated (above) and experimental (below) 1H NMR spectra of
cyclobutyl protons of 2.
˚
The values of these coupling constants suggest that 2 was
formed by the head-to-head coupling, but they do not allow
a certain assignment with respect to the syn/anti stereo-
chemistry.21,22
†
The atom numbers indicated in Figures 3–7, do not follow the IUPAC
rules used in Section 3 for naming the compounds, and were utilized only
for the sake of simplicity.
A more accurate structural determination has been attained

F. R. Cibin et al. / Tetrahedron 59 (2003) 3455–3459
3457
Table 1. Relative average energies (kEl) and average distances obtained
from 15 ns of molecular dynamics simulations for syn and anti
˚
stereoisomers of 2, at 298 K in CHCl3. Distances greater than 3.5 A are
not reported (see Section 3 for details)
2 h–h syn
2 h–h anti
kEl (kJ/mol)
0
36.28
Atom pair type kdl (A) (1/kd 26l)1/6 (A) kdl (A) (1/kd 26l)1/6 (A)
˚
˚
˚
˚
H2a00 –H2
H2a00 –H1
H2d00 –H1
H2d00 –H2
2.409
–
3.324
2.434
2.338
2.714
2.593
2.359
2.827
–
2.870
–
2.745
–
2.783
–
–
–
–
–
–
–
–
–
–
–
2.438
–
2.441
2.375
–
2.350
2.645
–
–
2.599
2.574
3.427
–
2.416
–
2.420
–
2.373
–
2.379
–
2.537
Figure 5. NOESY spectrum of 2 (detail).
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
Figure 7. Head-to-tail syn and anti structures (the dioxyethylene chains
have been omitted).
Figure 6. Head-to-head syn and anti structures (the dioxyethylene chains
have been omitted).
could have implications on the potential ligand properties of
the corresponding cyclophane moiety, and could be
exploited to construct more complex structures.
On the contrary, for the 2 h–0h0 anti stereoisomer, (Fig. 6, 2
h–h anti), both the H-2a –H-1 and the H-2a00 –H-2
˚
atomic distances are lower than 3.5 A and therefore the
protons H-2a00 would show cross peaks with both H-1 and
H-2.
3. Experimental
The NOESY pattern confirms that the head-to-tail junction
does not occur, because in this case the H-2 protons should
correlate with both the H-200 protons, either in the syn
structure or the highly strained anti structure (Fig. 7). In
part0i0cular, in 020 h–t0s0 yn, symmetry considerations show that
H-2a and H-2d (H-3b and H-3c00) protons would be chemically
equivalent, in contrast with the NMR results.
3.1. General and instrumentation
Melting points were determined with a Bu¨chi capillary
instrument and are uncorrected. Thin-layer chromatography
(TLC) was carried out on Merck precoated 60 Kiesegel F254
silica gel plates and Merck precoated RP-18 F254S reverse
phase plates. Flash chromatography was carried out on
columns with flash silica gel 60 Merck (40–63 mm). All
reactions requiring anhydrous conditions were conducted in
flame-dried apparatus. 1H NMR and 13C NMR spectra were
measured on a Bruker AC 200 or a Bruker AC 300
spectrometer, at 258C if not specified. Chemical shifts are
expressed in ppm with reference to the solvent signals. 2D
In the photocycloadditions of cinnamoyl derivatives linked
with short spacers, the cyclobutane products were formed by
head-to-head syn coupling.17–19 It is noteworthy that also
for the chalcone 1 the only kind of coupling process
observed again allows a syn relationship for the two
carbonyl groups. This feature is of importance because it

3458
F. R. Cibin et al. / Tetrahedron 59 (2003) 3455–3459
NMR spectra (H,H-COSY, H,C-COSY, NOESY-Phase
sensitive) were measured using standard Bruker software.
NOESY-Phase sensitive spectra in TPPI mode, were
recorded with mixing-times of 50–800 ms. IS-MS spectra
were obtained on a Perkin–Elmer SCIEX API 365 triple
quadrupole spectrometer. UV–vis spectra were carried out
on a Cary 300, at 258C. IR spectra were recorded on an
FT-IR Nicolet 510. Microanalyses were performed on a
Carlo Erba EA 1110 CHNS-O analyser. Photoreactions
were carried out in a Rayonet apparatus.
water. Recrystallization from MeOH yielded the bis-(4-
hydroxychalcone) 5 as a yellow solid (62%, mp 188–
1908C).
1H NMR (DMSO, 408C),‡ d 3.3 [bs, OH], 3.81–3.86 [m,
4H, –O–CH2], 4.16–4.21 [m, 4H, –O–CH2], 6.88
[AA0XX0, JH – H¼8.8 Hz, 4H, H-30], 7.01 [AA0XX0,
3
00
3
3JH–H¼8.8 Hz, 4H, H-3 ], 7.63 [d, JH–H¼15.6 Hz, 2H,
H-1], 7.72 [d, JH–H¼15.6 Hz, 2H, H-2], 7.78 [AA0XX0,
3
0
0
0
3JH–H¼8.3 Hz, 4H, H-2 ], 8.03 [AA XX , JH–H¼8.8 Hz,
3
4H, H-200].
3.2. Molecular modeling
13C NMR (DMSO, 508C),‡ d 67.26 [–O–CH2], 68.77 [–O–
CH2], 114.75 [C-300], 115.00 [C-30], 119.82 [C-2], 127.46
[C-quat.], 129.24 [C-200], 129.96 [C-20], 130.50 [C-quat.],
142.09 [C-1], 160.05 [C-quat.], 161.57 [C-quat.], 186.96
[CvO].
Molecular dynamics calculations were carried out with the
MacroModel 6.0 molecular modeling system,23 running on
a Silicon Graphics O2 R10000 workstation. The MMFF
force field24 was used in all the calculations. The
electrostatic interactions were calculated by using the
partial atomic charges of the MMFF force field. The effect
of the solvent was taken into account by using the GB/SA
model for chloroform as implemented in the MacroModel
package.25 A time step of 0.5 fs was chosen without the use
of SHAKE.
IR (CHCl3), nmax (CvO) 1652 cm21
MS (IS) m/z¼551 (MþH)þ.
.
Elemental analysis was unsatisfactory owing to the
presence, in traces, of non removable impurities.
3.3. Materials
3.3.3. 9,12,15,27,30,33-Hexaoxapentacyclo [32.2.2.
25,8.216,19.223,26] tetratetraconta-1(36), 3,5,7,16,18,20,
23,25,34,37,39,41,43-tetradecaene-2,22-dione (1). A sol-
ution containing 3 (0.414 g, 1 mmol) in dry DMF (200 mL)
was added dropwise (ca. 5 h), to a stirred suspension of 5
(1 mmol) and dry K2CO3 (690 mg, 1 mmol) in dry DMF
(800 mL) and kept under argon at 808C. The stirring was
continued at this temperature for ,3 days until complete
consumption of 5, as established by reverse phase TLC
(CH3CN/H2O 6/4). The resulting mixture was cooled and
filtered to remove K2CO3. The recovered K2CO3 was
washed with DMF, the organic phases were collected and
solvent removed in vacuo to yield a solid residue that was
dissolved in CHCl3, washed with water and dried on
Na2SO4. The solvent was removed in vacuo to yield a solid
that was purified by flash chromatography (CHCl3/acetone
9/1). Recrystallization from CHCl3/MeOH yielded the
cyclic bis-chalcone 1 as yellow solid (68%, mp 191–
1928C).
4-Hydroxybenzaldehyde (98%), 40-hydroxyacetophenone
(99%), p-toluenesulphonyl chloride (98%) and diethylene-
glycol (99%) were purchased from Aldrich and have been
used without further purification. Dry K2CO3 was obtained
by overnight heating analytical grade K2CO3 (Carlo Erba) at
1108C. Dry DMF was purchased from Fluka. Spectroscopy
grade chloroform (Carlo Erba) was used in the synthetic and
kinetic experiments without further purification.
3.3.1. 1,7-Bis-(4-formylphenyl)-1,4,7-trioxaheptane (4).
A dry DMF solution (50 mL) containing the diethylene
glycol bis-(p-toluenesulfonate) 3 (14.63 g, 35.3 mmol),26
was added dropwise (ca. 1 h), to a stirred suspension in dry
DMF (50 mL) of 4-hydroxybenzaldehyde (9.40 g,
77 mmol) and dry K2CO3 (26 g, 188 mmol), kept under
argon at 808C. The stirring was continued at this
temperature for 3.5 h. The resulting mixture was cooled
and poured on ice water. The precipitate was filtered and
washed with cold water to remove K2CO3. The solid
residue was crystallized from AcOEt affording 4 (9.32 g,
29.6 mmol, 84%, mp 141–1438C); lit. (60%, 140–
1418C).27
1H NMR (CDCl3),‡ d¼3.86–3.96 [m, 8H, –O–CH2],
4.14–4.24 [m, 8H, 00 –O–CH2], 6.77 [AA0XX0,
0
0
3
3JH–H¼8.7 Hz, 4H, H-3 ], 6.83 [AA XX , JH–H¼8.8 Hz,
4H, H-30], 7.30 [d, JH – H¼15.6 Hz, 2H, H-2], 7.44
3
3.3.2. 1,7-Bis-(4-[trans 3-(4-hydroxyphenyl)-3-oxo-pro-
penyl]-phenyl)-1,4,7-trioxaheptane (5). 40-Hydroxyaceto-
phenone (1.51 g, 11.1 mmol) and the aldehyde 4 (1.16 g,
3.69 mmol), were added to a stirred solution of NaOH
(1.34 g, 33.5 mmol) in EtOH/H2O (8/2, 25 mL), kept at
,58C under argon atmosphere. After the addition, the
suspension was stirred at 758C for ,3 h until complete
dissolution of 4. The solution was subsequently stirred at
508C for 70 h, until complete consumption of the bis-
aldehyde 4, as established by reverse phase TLC (eluent:
CH3CN/H2O 6/4þ0.1% TFA). The mixture was cooled,
carefully acidified to pH 5.5 by HCl 6 M and then ice water
was added until complete separation of a yellow powdered
solid, that was filtered and repeatedly washed with ice
[AA0XX0, 3JH – H¼8.5 Hz, 4H, H-200], 7.73 [d,
0
0
3JH–H¼15.6 Hz, 2H, H-1], 7.89 [AA XX , JH–H¼8.8 Hz,
3
4H, H-20].
13C NMR (CDCl3),‡ d 67.81 [–O–CH2], 69.95 [–O–CH2],
114.64 [C-300], 115.34 [C-30], 118.99 [C-2], 127.80
[C-quat.], 129.86 [C-200], 130.41 [C-20], 131.30 [C-quat.],
143.80 [C-1], 160.67 [C-quat.], 162.43 [C-quat.], 188.26
[CvO].
‡
For the sake of simplicity, the atom numbers indicated in NMR spectra
are referred to those indicated in Scheme 2 (compounds 1 and 5), and
Figure 3 (compound 2).

F. R. Cibin et al. / Tetrahedron 59 (2003) 3455–3459
3459
IR (CHCl3), nmax(CvO) 1655 cm21
.
di Roma La Sapienza’ is gratefully acknowledged for the
microanalyses.
UV–vis (CHCl3), lmax 334 nm, 1 5.8£104 cm21
-
dm3 mol21
.
MS (IS) m/z¼621 (MþH)þ.
References
Calcd. C 73.53, H 5.85; Found C 73.80, H 5.95.
1. Stobbe, H. Ber. 1919, 52, 666–672.
2. Stobbe, H.; Hensel, A. Ber. Dtsch. Chem. Ges. 1926, 59,
2254–2265.
3.3.4. 9,12,15,27,30,33-Hexaoxaheptacyclo-[32.2.2.
25,8.216,19.223,26.02,22.03,21]-tetratetraconta-1(36),5,
7,16,18,23,25, 34,37,39,41,43-dodecaene-4,20-dione (2).
Photoisomerization in sunlight. A solution of 1 (98 mg,
0.163 mmol) in 50 mL of chloroform, kept in a Pyrex flask,
was exposed to sunlight on a sunny day in July. The
progress of the reaction was followed by silica gel TLC
(CHCl3/acetone 9/1) and reverse phase TLC (CH3CN/H2O
8/2). The reaction was stopped after ,1 h. The solution was
evaporated and the residue purified by flash chromatography
(CHCl3/acetone 94/6, 72 mg, 74%). Recrystallization from
chloroform–methanol gave microscopic white needles of
an analytically pure sample (mp 258–2598C).
3. Stobbe, H.; Bremer, K. J. Prakt. Chem. 1929, 123, 1–60.
4. Mustafa, A. Chem. Rev. 1952, 51, 1–23.
5. Schmidt, G. M. J. Pure Appl. Chem. 1971, 27, 647–678.
6. Montaudo, G.; Caccamese, S. J. Org. Chem. 1973, 38,
710–716.
7. Zahir Sheik, A. J. Appl. Polym. Sci. 1979, 23, 1355–1372.
8. Abdel-Megeid, F. M. E.; Bose, A. K.; Elkaschef, M. A.-F.;
Elsayed, A. S.; Mokhtar, K.-E. M.; Sharma, S. D. Indian
J. Chem. 1975, 13, 482–484.
9. Hasegawa, M.; Saigo, K.; Mori, T.; Uno, H.; Nohara, M.;
Nakanishi, H. J. Am. Chem. Soc. 1985, 107, 2788–2793.
10. Toda, F.; Tanaka, K.; Kato, M. J. Chem. Soc., Perkin Trans. 1
1998, 1315–1318.
Photoisomerization at 350 nm. A 6.44£1023 mol dm23
solution of 1 (0.100 g in 25 mL) in chloroform, contained
in a Pyrex flask, was irradiated in a Rayonet reactor by 14
lamps whose emission was centred at 350 nm. After ca.
30 min the reaction appeared to be complete. The solvent
was removed and the residue purified by flash chromato-
graphy (yield: 92%).
11. D’Auria, M. Heterocycles 2001, 54, 475–496.
12. Doddi, G.; Ercolani, G.; Mencarelli, P.; Scalamandre, C.
`
J. Org. Chem. 1991, 56, 6331–6336.
13. Doddi, G.; Ercolani, G.; La Pegna, P.; Mencarelli, P. J. Chem.
Soc., Chem. Commun. 1994, 1239–1240.
14. Doddi, G.; Fazio, M. C.; Luvidi, L.; Mencarelli, P.; Ercolani,
G. Tetrahedron 1996, 52, 3509–3520.
15. D’Acerno, C.; Doddi, G.; Ercolani, G.; Mencarelli, P. Chem.
Eur. J. 2000, 6, 3540–3546.
1H NMR (CDCl3),‡ d¼3.55–3.65 [m, 4H, –O–CH2],
3.70–3.85 [m, 4H, –O–CH2], 4.15–4.30 [m, 8H, –O–
CH2], 4.60–4.65 [m, 2H, H-1],§ 4.87–4.91 [m, 2H, H-2],§
16. Doddi, G.; Ercolani, G.; Franconeri, S.; Mencarelli, P. J. Org.
Chem. 2001, 66, 4950–4953.
6.55 [dd, JH–H¼8.4 Hz, JH–H¼2.5 Hz, 2H, H-300c], 6.62
17. Akabori, S.; Kumagai, T.; Habata, Y.; Sato, S. Bull. Chem.
Soc. Jpn 1988, 61, 2459–2466.
3
4
and 6.63 [AA0XX0, JH–H¼8.9 Hz, 6H, H-30 and super-
3
00
3
4
imposed00H-3 b], 6.67 [dd, JH–H¼8.5 Hz, JH–H¼2.0 Hz,
18. Akabori, S.; Kumagai, T.; Habata, Y.; Sato, S. J. Chem. Soc.,
Perkin Trans. 1 1989, 1497–1505.
3
4
2H,00H-2 d], 6.910 [dd, JH–H¼8.5 Hz, JH–H¼2.0 Hz, 2H,
H-2 a], 7.52 [AA XX0, 3JH–H¼8.9 Hz, 4H, H-20].
19. Zitt, H.; Dix, I.; Hopf, H.; Jones, P. G. Eur. J. Org. Chem.
2002, 2298–2307.
13C NMR (CDCl3),‡ d¼42.61 [C-1], 45.60 [C-2)], 68.01
[–O–CH2], 68.81 [–O–CH2], 69.92 [–O–CH2], 70.32
[–O–CH2], 114.27 [C-30], 115.49 [C-300b], 115.95 [C-300c],
126.77 [C-200a], 130.17 [C-quat.], 130.35 [C-20], 131.29
[C-200d], 132.25 [C-quat.], 156.75 [C-quat.], 162.31
[C-quat.], 197.09 [CvO].
20. MestRe-C 2.3 program, Departamento de Quımica Organica,
Universidade de Santiago de Compostela, 1706 Santiago de
Compostela, Spain—http://qobrue.usc.es/jsgroup/MestRe-
C/MestRe-C.html.
´
´
21. Montaudo, G.; Caccamese, S.; Librando, V. Org. Magn. Res.
1974, 6, 534–536.
22. Caccamese, S.; McMillan, J. A.; Montaudo, G. J. Org. Chem.
1978, 43, 2703–2704.
IR (CHCl3), nmax(CvO) 1679 cm21
.
23. MacroModel V6.0 Mohamadi, F.; Richards, N. G. J.; Guida,
W. C.; Liskamp, R.; Lipton, M.; Caulfield, C.; Chang, G.;
Hendrickson, T.; Still, W. C. J. Comput. Chem. 1990, 11,
440–467.
UV–vis (CHCl3), lmax 272 nm,
dm3 mol21
1
3.1£104 cm21
.
MS (IS) m/z¼621 (MþH)þ.
24. Halgren, T. A. J. Comput. Chem. 1996, 17, 490–519, and
references therein.
Calcd. C 73.53, H 5.85. Found C 73.39, H 5.91.
25. Still, W. C.; Tempczyk, A.; Hawley, R. C.; Hendrickson, T.
J. Am. Chem. Soc. 1990, 112, 6127–6129.
26. Ouchi, M.; Kanzaki, T.; Inoue, Y.; Hakushi, T. J. Org. Chem.
1984, 49, 1408–1412.
Acknowledgements
27. Katritzky, A. R.; Belyakov, S. A.; Denisko, O. V.; Maran, U.;
Dalal, N. S. J. Chem. Soc., Perkin Trans. 2 1998, 611–615.
`
Dr Paola Galli of the ‘Dipartimento di Chimica, Universita
§
The estimated J values for this AA0BB0 scheme are reported in the text.