Synthesis of Cage-Type Molecules with π-Cavity
NMR (CDCl3) δ 5.00 (s, 2H), 6.83 (d, J ) 8.7 Hz, 2H), 7.21-
7.39 (m, 7H); 13C NMR (CDCl3) δ 158.5, 137.2, 132.9, 129.3,
128.8, 128.1, 117.3, 113.7, 70.8; MS (EI) m/z calcd for C13H11-
BrO 262.00 [M]+, found 262.03.
have not observed the difference in the guest selectivity
between meta-cage 3c and para-cage 4b. Thus, a different
+
binding process may be proposed in this case. The NH4
ion may slide into the cavity in a “dissociative” fashion:
H+ followed by NH3 sliding through the gate. Once in
the cavity, the NH4+ ion can have good contact with the
surrounding aryl groups. In the case of meta analogue
3c, such a sliding entry is also possible, but, because of
its poor cavity-like shape, the inclusion complex would
be less favored than the case of 4b.11
1,3,5-Tris(3-hydroxybenzyl)-2,4,6-trimethylbenzene (7a).
To a stirred suspension of magnesium powder (183 mg, 7.5
mmol) in THF (5 mL) were added 6 (1.32 g, 5.0 mmol) and
1,2-dibromoethane (one drop) at 25 °C under argon, and the
resulting mixture was stirred for an additional 1 h. This
Grignard solution was transferred dropwise to a mixture of
1,3,5-tris(bromomethyl)mesitylene 5a (399 mg, 1.0 mmol) and
CuI (19 mg, 0.1 mmol) in THF (5 mL) at 60 °C, and the
resulting mixture was further stirred for 12 h at the same
temperature. The mixture was allowed to warm to 25 °C and
quenched with a saturated aqueous NaHCO3 solution (10 mL).
The resulting solution was diluted with EtOAc (10 mL), and
the organic layer was washed with brine, dried, and concen-
trated in vacuo. The residue was filtered through a short pad
of silica gel (hexane/EtOAc, 9:1) to give the coupled compound.
This crude compound was dissolved in ethanol (30 mL) and
was subjected to hydrogenolysis in the presence of 10 wt %
Pd/C (100 mg) under H2 atmosphere (about 1 atm) at 25 °C
for 2 h. The reaction mixture was filtered through Celite, and
the filtrate was concentrated. Purification of the residue by
chromatography (hexane/EtOAc, 7:3) afforded 7a (336 mg,
77%) as a white solid: Rf ) 0.3 (hexane/EtOAc, 7:3); mp 251-
253 °C; 1H NMR (CDCl3/DMSO-d6) δ 2.14 (s, 9H), 4.05 (s, 6H),
6.44 (s, 1H), 6.60 (dd, J ) 2.3, 7.5 Hz, 6H), 7.06 (t, J ) 7.8 Hz,
3H), 8.68 (s, 3H); 13C NMR (CDCl3/DMSO-d6) δ 157.1, 141.5,
134.4, 134.3, 129.1, 119.0, 114.3, 112.6, 35.7, 16.5; MS (FAB)
m/z calcd for C30H30O3 439.22 [M + 1]+, found 439.22.
1,3,5-Tris(3-hydroxybenzyl)-2,4,6-triethylbenzene (7b).
This compound was similarly synthesized as above from 6 (1.32
g, 5.0 mmol) and 1,3,5-tris(bromomethyl)-2,4,6-triethylbenzene
5b (441 mg, 1.0 mmol) in 59% yield (287 mg) as a white solid.
Rf ) 0.3 (hexane:EtOAc, 7:3); mp 200-201 °C; 1H NMR (CDCl3/
DMSO-d6) δ 1.11 (t, J ) 7.5 Hz, 9H), 2.47 (q, J ) 7.5 Hz, 6H),
4.01 (s, 6H), 6.35 (s, 3H), 6.62 (dd, J ) 1.8, 7.9 Hz, 3H), 6.76
(d, J ) 7.5 Hz, 3H), 7.12 (t, J ) 7.8 Hz, 3H), 8.37 (s, 3H); 13C
NMR (CDCl3/DMSO-d6) δ 157.6, 143.2, 141.8, 134.3, 129.8,
120.2, 114.9, 113.4, 40.4, 34.8, 24.1, 15.6; MS (FAB) m/z calcd
for C33H36O3 481.27 [M + 1]+, found 481.28.
1,3,5-Tris(4-hydroxybenzyl)-2,4,6-trimethylbenzene (9).
This compound was similarly synthesized as above from the
bromobenzene derivative 8 (3.5 g, 13.3 mmol) and 5a (1.06 g,
2.66 mmol) in 68% yield (791 mg) as a white solid: Rf ) 0.2
(hexane/EtOAc, 7:3); mp 251-253 °C; 1H NMR (CDCl3/DMSO-
d6) δ 2.12 (s, 9H), 4.01 (s, 6H), 6.72 (d, J ) 8.3 Hz, 6H), 6.83
(d, J ) 8.3 Hz, 6H), 8.16(s 3H); 13C NMR (CDCl3/DMSO-d6) δ
155.2, 135.7, 134.8, 131.7, 129.2, 115.9, 35.7, 17.1; MS (FAB)
m/z calcd for C30H30O3 439.22 [M + 1]+, found 439.23.
meta-Cage 3a. To a stirred solution of Cs2CO3 (1.3 g, 4
mmol) in DMF (25 mL) at 80 °C under argon was added
dropwise the phenolic compound 7a (88 mg, 0.2 mmol) and
5a (80 mg, 0.2 mmol) in DMF (15 mL) by a syringe pump for
6 h, and the resulting mixture was further stirred for 60 h at
the same temperature. The mixture was cooled to room
temperature, and the solvent was removed at low pressure.
The residue was diluted with EtOAc (5 mL) and poured into
brine (10 mL). The organic phase was dried, filtered, and
concentrated in vacuo. Purification of the residue by chroma-
tography (hexane/EtOAc, 4:1) afforded 3a (25 mg, 4%) as a
white solid: Rf ) 0.7 (hexane/EtOAc, 7:3); mp 320-344 °C dec;
1H NMR (CDCl3) δ 2.06 (s, 9H), 2.21 (s, 9H), 3.95 (s, 6H), 5.11
(s, 6H), 5.89 (s, 3H), 7.05-7.09 (m, 6H), 7.28 (t, J ) 7.9 Hz,
3H); 13C NMR (CDCl3) δ 160.1, 141.9, 137.7, 135.7, 134.0,
132.6, 129.5, 124.6, 120.7, 118.6, 70.8, 36.4, 17.5, 16.2; MS
(FAB) m/z calcd for C42H42O3 595.31 [M + 1]+, found 595.42.
Conclusion
We have synthesized cage-type molecules 3 and 4 as
potential hosts for the selective recognition of alkali metal
and ammonium cations mainly through cation-π inter-
actions. The synthesis involved a key cyclization step, of
which yield was markedly dependent on the capping
component. The cage-type hosts have a π-cavity that is
expected to accommodate a K+ ion by a molecular
modeling study. However, competitive binding studies by
electrospray ionization mass spectrometry showed that
3c binds selectively Li+ while 4b binds both Li+ and NH4
+
in the presence of other cations examined. A gate-
selective binding process may explain these results.
Under such a process, a dissociative binding process
explains the complexation of host 4b with NH4+. A
structural modification of cage-type host 4b to address
the gate-selective binding process is under investigation.
Experimental Section
General Methods. 1H NMR spectra are reported as fol-
lows: chemical shift in ppm from internal tetramethylsilane
on the delta scale, multiplicity, coupling constant (in hertz),
and integration. Melting points are uncorrected ones. Chro-
matography means flash column chromatography using silica
gel of 230-400 mesh. Solvents were dried and distilled under
standard conditions before use.
1-Bromo-3-benzyloxybenzene (6). To a stirred solution
of sodium hydride (720 mg, 18 mmol, 60% dispersion in
mineral oil) in DMF (20 mL) was added 3-bromophenol (2.6 g,
15 mmol) at a water-ice bath temperature under argon
atmosphere, and the resulting mixture was stirred for 20 min
at the same temperature. To the mixture was added benzyl
bromide (1.8 mL, 15 mmol), the ice bath was removed, and
the mixture was stirred at ambient temperature for 2 h. The
reaction mixture was treated with a saturated aqueous am-
monium chloride solution (5 mL) and diluted with diethyl ether
(30 mL). The organic layer was dried and concentrated, and
the residue was purified by chromatography (hexane/EtOAc,
9:1) to afford 6 as a white solid (4.0 g, 100% yield): Rf ) 0.6
1
(hexane/EtOAc, 9:1); mp 61-62 °C; H NMR (CDCl3) δ 5.06
(s, 2H), 6.88-6.91 (m, 1H), 7.07-7.16 (m, 3H), 7.31-7.41 (m,
5H); 13C NMR (CDCl3) δ 160.2, 137.0, 131.2, 129.3, 128.8,
128.2, 124.7, 123.5, 118.8, 114.5, 70.8; MS (EI) m/z calcd for
C13H11BrO 262.00 [M]+, found 262.05.
1-Bromo-4-benzyloxybenzene (8). This compound was
similarly synthesized as above from 4-bromophenol (3.46 g,
20 mmol) and benzyl bromide (2.4 mL, 20 mmol) in 92% yield
(4.86 g) as a colorless liquid: Rf ) 0.7 (hexane/EtOAc, 9:1); 1H
(11) Such a dissociative process, occurring under the dynamic
+
process, may not achieve an equilibrium state; otherwise, the NH4
-
inclusion complex with better size complementarity would be preferred
over the Li+ complex (From the results of Shinkai and co-workers,4 it
is expected that the NH4+-inclusion complex surrounded by four
phenyl rings in an ideal case would be thermodynamically more favored
over the Li+ complex stabilized by one phenyl ring).
meta-Cage 3b. This compound was synthesized similarly
as above by coupling of the meta-phenol derivative 7a (46 mg,
0.104 mmol) with 1,3,5-tris(bromomethyl)-2,4,6-(trimethoxy)-
benzene 5c (47 mg, 0.104 mmol) in 31% yield (21 mg) as a
J. Org. Chem, Vol. 70, No. 18, 2005 7091