Synthesis and characterization of chiral calixarene analogs locked in the cone conformation by the photocycloaddition

Article abstract of DOI:10.1055/s-2003-36803

A bisphenol derivative was used as the starting material. [2+2] Photocycloaddition of tetraolefin gave the desired calix[4]arene analogs having chiral and achiral structure in 12-77percent yields. The optical resolution was successfully achieved by using the HPLC method. The calixarene analog formed a 1:1 complex with alkali metal ions and extracted large metal picrates better than small ones.

Full text of DOI:10.1055/s-2003-36803

LETTER
199
Synthesis and Characterization of Chiral Calixarene Analogs Locked in the
Cone Conformation by the Photocycloaddition
S
ynthesis of Chira
u
l
C
alixarene
k
A
nalogs ihiro Okada, Masashi Yoshida, Jun Nishimura*
Department of Chemistry, Gunma University, Kiryu 376-8515, Japan
E-mail: okada@chem.gunma-u.ac.jp
Received 17 October 2002
macrocycles. Although many calix[n]arenes were report-
ed as powerful host molecules in the last decade, difficul-
ties in the synthesis or modification toward chiral ones
still remained.5–9 Therefore, we were prompted to develop
Abstract: A bisphenol derivative was used as the starting material.
[2+2] Photocycloaddition of tetraolefin gave the desired
calix[4]arene analogs having chiral and achiral structure in 12–77%
yields. The optical resolution was successfully achieved by using
the HPLC method. The calixarene analog formed a 1:1 complex another simple synthetic route to chiral calixarene analogs
with alkali metal ions and extracted large metal picrates better than
small ones.
with styrene derivatives by the photocycloaddition. We
would like to report here the synthesis and characteriza-
Key words: calixarene, cyclophane, photocycloaddition, alkali
metal
tion of chiral calix[4]arene analogs by the intramolecular
photocycloaddition.
The synthesis of calix[4]arenes is shown in Scheme 1 and
Table 1. Bisphenol derivative 1 was used as the starting
material. Monoetherification of 1 was performed with
Li2CO3 and CH3I in anhyd DMF at room temperature for
12 hours to give monomethyl ether 2 in 93% yield. Tetra-
bromides 3 were obtained in 77–95% yields by the ether-
ification with 2, the corresponding oligo(ethylene glycol)
ditosylate, and K2CO3 in anhyd DMF at 100 °C for 18
Intramolecular [2+2] photocycloaddition of styrene deriv-
atives is a powerful and convenient method for the forma-
tion of a cyclophane skeleton.1,2 [2+2] Photocycloaddition
easily makes the rigid cyclophane skeleton having a C2 or
Cs symmetrical unit by the transformation from the vinyl
group to a cyclobutane ring.3,4 Accordingly, this method
can be adapted for the synthesis of stable enantiomers of
OH
OCH3
OH
OH
TsO O n-1
Br
Br
Br
Br
K2CO3,
OTs
Li2CO3, CH3I
DMF, 100 °C, 18 h
r.t., 12 h
DMF,
2
1
(n-Bu)3SnCH=CH2
n-1
O
O
O
OCH3
OCH3
n-1
O
OCH3
OCH3
O
O
Br
Br
Br
Br
PdCl2(PPh3)2
DMF, LiCl
80 °C, 6 h
3
4
O
n-1
O
OCH3
O
n
OCH3
O
n-1
a
(Pyrex)
hν
1
2
3
+
b
c
benzene
O
O
OCH3
OCH3
d
e
4
5
6
5
Scheme 1
Scheme 1
Synlett 2003, No. 2, Print: 31 01 2003.
Art Id.1437-2096,E;2003,0,02,0199,0202,ftx,en;U10102ST.pdf.
© Georg Thieme Verlag Stuttgart · New York
ISSN 0936-5214

200
Y. Okada et al.
LETTER
hours. The vinylation of 3 was performed with tributyl(vi-
nyl)tin, PdCl2(PPh3)2, and LiCl in anhyd DMF at 80 °C for
6 hours to give tetraolefins 4 in 47–88% yields. [2+2]
Photocycloaddition of 4 (0.31–1.2 mM) was carried out
by irradiation with a 400W high-pressure Hg lamp (Pyrex
filter) in benzene for 6–12 hours. After evaporation,
calix[4]arenes 5 and 6 were isolated with column chroma-
tography (SiO2, benzene–ethyl acetate = 2:1). The results
are summarized in Table 1. The olefin 4a did not give any
desired product at all due to the difficulty of parallel ar-
rangement between two olefin moieties under the photo-
cycloaddition. On the other hand, 4b gave calixarenes in
excellent 77% yield (entry 2). These results show that the
chain length of 4a is too short to make a calixarene skele-
ton, but that of 4b is suitable for the formation of calixare-
ne. In fact, MM2 calculations show that the steric energy
of 5a and 6a is higher than that of 5b and 6b as shown in
Figure 1. When the oligoethylene chain became longer
from 4c to 4e, the yields of calixarenes were gradually de-
creased from 27% to 12% (entries 4, 7, and 10). This re-
sult shows that the molecular stability of calixarenes is
decreased from n = 2 to 5, judging from the MM2 calcu-
lations as shown in Figure 1. Thus, new calix[4]arenes,
which show the potential of the introduction of functional
group on upper rim, were simply obtained with styrene
derivatives by the [2+2] photocycloaddition.
5
6
100
90
80
5
1
2
3
4
Chain length (n)
Figure 1 MM2 calculation of calixarene isomers 5 and 6.
energy of 5 is ca. 2–3 kcal/mol lower than that of 6 in the
case of n = 2 and 3 and nearly the same in the case of n =
4 as shown in Figure 1. The highest regioselectivity for 5
and 6 is recorded as 3.6:1 in the system of n = 3.
[2+2] Photocycloaddition of cinnamoyl derivatives was
reported to change the product distribution from a trans to
a cis adduct by the addition of metal perchlorate in ben-
zene due to the affinity of olefins on metal salt.10 There-
fore, we examined the conditions of photocycloaddition to
be high regioselectivity by adding metal perchlorate. In
fact, the product ratio (5:6) and total yield (5 + 6) of 4d are
dramatically changed from 1.1:1 to 2.1:1 and from 23% to
46% when KClO4 (1 equiv) was added in the benzene so-
lution (entries 7 and 8 in Table 1). Similarly, those of 4c
The product ratio of chiral or achiral isomers as regioiso-
mers is important to prepare the new chiral calixarenes. In
fact, the chiral calixarene 5 was produced more than the
achiral 6 in the system of n = 2–4 in benzene solution (see
Table 1). It is clear that 5 is a more stable isomer than 6.
The MM2 calculations also demonstrated that the steric
Table 1 The Product Distribution of Photocycloaddition in Benzenea
Product
Entry
Comp.
n
Salt
Salt equiv
Yield (%)
Ratio
5
6
Total
–
5:6
1
2
4a
4b
4b
4c
4c
4c
4d
4d
4d
4e
4e
4e
1
2
2
3
3
3
4
4
4
5
5
5
–
–
–
1
–
1
1
–
1
1
–
1
1
–
–
–
–
43
11
21
25
27
12
31
20
34
77
19
27
30
32
23
46
37
12
45
27
1.3:1
1.4:1
3.6:1
4.9:1
5.2:1
1.1:1
2.1:1
1.2:1
0.73:1
0.61:1
0.56:1
3
NaClO4
–
8.0
4
5.8
5.1
5.2
5
NaClO4
KClO4
–
6
7
11
8
KClO4
RbClO4
–
15
17
9
10
11
12
5.1
17
9.5
7.0
KClO4
RbClO4
28
17
a Conditions: 0.31–1.2 mmol/L, 6–12 h.
Synlett 2003, No. 2, 199–202 ISSN 0936-5214 © Thieme Stuttgart · New York

LETTER
Synthesis of Chiral Calixarene Analogs
201
are changed from 3.6:1 to 5.2:1 and from 27% to 32% complexation with 6 and metal ions. Ionophores showed
when KClO4 was added (entries 4 and 6). These results excellent extractability for larger alkali metal ions like K+,
showed that 5 was more sensitive than 6 in the presence or Rb+, and Cs+ than for smaller ones like Li+ and Na+. The
absence of metal ions.
ion selectivity of 5 dramatically changes by the length of
the ether linkage; i.e., 5b having a diethylene glycol unit
(n = 2) could not extract the alkali ions due to the small
size of crown ether. Ionophore 5c having a triethylene gly-
col unit (n = 3) apparently shows sharp K+ ion selectivity.
Ionophore 5d having a tetraethylene glycol unit (n = 4)
clearly shows Rb+ ion selectivity. Moreover, 5e having a
pentaethylene glycol unit (n = 5) shows Cs+ ion selectivi-
ty. These results clearly suggest that the ring size of crown
ether was forced to form a suitable cavity size by the rigid
calixarene skeleton. In fact, CPK model examination and
MM2 calculations indicate that the cavity size of crown
ether on calixarenes 5 is approximately 2.6–2.9 Å for 5c,
2.9–3.3 Å for 5d, and 3.3–3.7 Å for 5e. These are really
complementary to the complexed ions, because the ionic
diameters of K+, Rb+, and Cs+ are 2.76 Å, 2.98 Å, and 3.40
Å, respectively.14
1
The structure of 5 and 6 was mainly elucidated by H
NMR spectroscopy.11 All the protons could be assigned in
the usual way by using several experiments like COSY
and NOESY and considering the symmetry of C2 for 5 and
Cs for 6. We concluded that they took completely the
cone-type conformation, as depicted in Scheme 1, from
the following findings: their methylene bridges show AB
type coupling (d = 3.13–3.17 and 4.16–4.22 with J = 15
Hz for 5 and d = 3.17 and 4.21–4.26 with J = 15 Hz for
6),2 which is the same as those ascribed to the calixarene
cone-form. Their cyclobutane methine protons resonate at
d = 4.13–4.60 to demonstrate the typical cis configura-
tion.1,3 Their methoxy groups show a singlet at d = 3.63–
3.75 for 5 and 3.20–3.26 for 6 to keep the same environ-
ment, respectively.1 The chirality was easily demonstrated
by the optical resolution of 5d using the HPLC method
(Chiralpak AD, hexane–2-propanol = 23/1 as an eluent, The best extractability for alkali metal ions among all ion-
αD = –40.2 and +41.1).9
ophores is exhibited by 5d possessing a tetraethylene gly-
col unit. In both cases of tri- and pentaethylene glycol
units, their extractability considerably decreased, al-
though their ion selectivity increased. Thus, the regioiso-
mers 5 and 6 showed quite different behavior for metal
binding. Accordingly, this is a sophisticated modification
of the calix[4]arene geometry, which dramatically
changed the ion selectivity and extractability.
First, the stoichiometry for the complexation was exactly
examined by Job’s plots (3 mmol/L) between ionophore
5d and alkali metal thiocyanates in CDCl3–CD3OD (7:3).
When the mole fraction of 5d was 0.5, the complex con-
centration reached maximum for K+, Rb+, and Cs+. This
result clearly demonstrates that 5 forms a 1:1 complex
with alkali metal ions.
In conclusion, we have synthesized calix[4]arene analogs
having chiral and achiral structure in excellent yield firm-
ly locked in the cone conformation. The optical resolution
of chiral calixarene was successfully achieved by using
the HPLC method. The calix[4]arene analog formed a 1:1
complex with alkali metal ions and extracted larger metal
picrates than small ones. Further investigation is now in
progress and will be reported elsewhere.
Based on these observations, we determined the extract-
ability of ionophores 5 and 6 with alkali metal ions from
the aqueous phase to an organic phase.12–16 The extraction
experiments were carried out with 2.5 × 10–4 M of iono-
phores in CH2Cl2 and 2.5 × 10–5 M of picric acid in 0.01
M of metal hydroxide at 22 °C. These results are summa-
rized in Table 2. Generally speaking, 5 acted as excellent
ionophores, but 6 did not. It is clear that the steric hin-
drance of cyclobutane methine protons prohibited the
a
Table 2 Extraction (%) of Alkali Metal Picrates in CH2Cl2
+
Compd
5b
Li+
<1
Na+
<1
K+
Rb+
<1
Cs+
<1
NH4
1.2
1.3
5.9
2.6
<1
<1
5c
2.8
7.7
1.7
3.4
<1
20.9
22.8
5.6
<1
4.5
43.0
9.4
<1
2.0
5d
1.7
1.9
36.5
12.8
<1
5e
6b
<1
6c
<1
<1
<1
<1
3.3
6.8
6.2
<1
<1
<1
6d
1.2
1.5
9.9
12.1
8.7
1.9
2.5
6e
11.6
a Extraction conditions: 2.5 × 10–4 M of ionophore in CH2Cl2; 2.5 × 10–5 M of picric acid in M of MOH at 22 °C. Ionophore solution (5.0 mL)
was shaken (10 min) with picrate solution (5.0 mL) and % extraction was measured by the absorbance of picrate in CH2Cl2. Experimental error
was ± 2%.
Synlett 2003, No. 2, 199–202 ISSN 0936-5214 © Thieme Stuttgart · New York

202
Y. Okada et al.
LETTER
3.77–3.83 (4 H, m), 4.13 (2 H, m), 4.22 (2 H, d, 15), 4.43 (2
H, m), 5.94 (2 H, dd, 1.8, 8.0), 6.17 (2 H, dd, 1.8, 8.0), 6.32
(2 H, t, 8.0), 7.07 (2 H, dd, 1.8, 8.0), 7.16 (2 H, t, 8.0), 7.38
(2 H, dd, 1.8, 8.0). 6c: Calcd for C42H46O6: 646: 2.24–2.58 (8
H, m), 3.17 (2 H, d, 15), 3.26 (6 H, s), 3.50–3.95 (12 H, m),
4.21 (2 H, d, 15), 4.16 (2 H, m), 4.60 (2 H, m), 6.44 (2 H, dd,
2.0, 7.8), 6.67 (2 H, dd, 2.0, 7.8), 6.69 (2 H, t, 7.8), 6.76 (2
H, dd, 2.0, 7.8), 6.81 (2 H, t, 7.8), 6.86 (2 H, dd, 2.0, 7.8). 5d:
Calcd for C44H50O7: 690: 2.24–2.52 (8 H, m), 3.13 (2 H, d,
15), 3.63 (6 H, s), 3.64–4.06 (16 H, m), 4.16 (2 H, d, 15),
4.26 (2 H, m), 4.36 (2 H, m), 6.40 (2 H, dd, 1.8, 7.5), 6.56 (2
H, dd, 1.8, 7.5), 6.63 (2 H, t, 7.5), 6.67 (2 H, dd, 1.8 & 7.5),
6.86 (2 H, t, 7.5), 6.98 (2 H, dd, 1.8, 7.5). 6d: Calcd for
C44H50O7: 690: 2.26–2.50 (8 H, m), 3.17 (2 H, d, 15), 3.20 (6
H, s), 3.58–3.94 (16 H, m), 4.26 (2 H, d, 15), 4.28 (2 H, m),
4.44 (2 H, m), 6.52 (2 H, dd, 1.8, 7.5), 6.66 (2 H, dd, 1.8,
7.5), 6.72 (2 H, dd, 1.8, 7.8), 6.76 (2 H, t, 7.5), 6.78 (2 H, dd,
1.8, 7.5), 6.83 (2 H, dd, 1.8, 7.5).
Acknowledgment
This work was supported by grants from the Ministry of Education,
Culture, Sports, Science, and Technology, Japan.
References
(1) Okada, Y.; Sugiyama, K.; Wada, Y.; Nishimura, J.
Tetrahedron Lett. 1990, 31, 107.
(2) Okada, Y.; Ishii, F.; Kasai, Y.; Nishimura, J. Chem. Lett.
1992, 755.
(3) Okada, Y.; Mabuchi, S.; Kurahayashi, M.; Nishimura, J.
Chem. Lett. 1991, 1345.
(4) Okada, Y.; Hagihara, M.; Mineo, M.; Nishimura, J. Synlett
1998, 269.
(5) Calixarenes 50th Anniversary; Vicens, J.; Asfari, Z.;
Harrowfield, J. M., Eds.; Kluwer: Dordrecht, 1994.
(6) Böhmer, V. Angew. Chem. Int. Ed. Engl. 1995, 34, 713.
(7) Okada, Y.; Ishii, F.; Kasai, Y.; Nishimura, J. Tetrahedron
Lett. 1993, 34, 1971.
(8) Okada, Y.; Kasai, Y.; Nishimura, J. Tetrahedron Lett. 1995,
36, 555.
(9) Okada, Y.; Mizutani, M.; Ishii, F.; Nishimura, J.
Tetrahedron Lett. 1997, 38, 9013.
(12) Okada, Y.; Mizutani, M.; Ishii, F.; Kasai, Y.; Nishimura, J.
Tetrahedron 2001, 57, 1219.
(13) Okada, Y.; Mizutani, M.; Ishii, F.; Nishimura, J. Synlett
2000, 41.
(14) Okada, Y.; Mizutani, M.; Ishii, F.; Nishimura, J.
Tetrahedron Lett. 1999, 40, 1353.
(15) Okada, Y.; Mizutani, M.; Nishimura, J. Tetrahedron Lett.
1998, 39, 8467.
(10) Kimura, M.; Shimoyama, M.; Morosawa, S. J. Chem. Soc.
Chem. Commun. 1991, 375.
(11) Compd.: MS (M+): 1H NMR d (intensity, multiplicity, J in
Hz). 5c: Calcd for C42H46O6: 646: 2.16–2.62 (8 H, m), 3.17
(2 H, d, 15), 3.52 (2 H, m), 3.58–3.70 (6 H, m), 3.75 (6 H, s),
(16) Okada, Y.; Mizutani, M.; Ishii, F.; Nishimura, J.
Tetrahedron Lett. 1998, 39, 8461.
Synlett 2003, No. 2, 199–202 ISSN 0936-5214 © Thieme Stuttgart · New York