Enhanced thermodynamic and kinetic stability of calix[4]arene dimers locked in the cone conformation

Article abstract of DOI:10.1021/jo049128f

Wide rim tetraurea derivatives (2a,b) have been prepared from a calix[4]arene rigidified in the cone conformation by two diethyleneglycol ether bridges between adjacent oxygens. In comparison to the analogous tetraurea derivatives (3a,b) of a tetrapentoxy

Full text of DOI:10.1021/jo049128f

En h a n ced Th er m od yn a m ic a n d Kin etic Sta bility of Ca lix[4]a r en e
Dim er s Lock ed in th e Con e Con for m a tion
Myroslav O. Vysotsky,† Oliver Mogck,† Yuliya Rudzevich,† Alexander Shivanyuk,†,‡
Volker Bo¨hmer,†* Marcus S. Brody,‡ Young Lag Cho,‡ Dmitry M. Rudkevich,‡ and
J ulius Rebek, J r.*,‡
Abteilung Lehramt Chemie, Fachbereich Chemie und Pharmazie, J ohannes Gutenberg-Universita¨t Mainz,
Duesbergweg 10-14, D-55099, Mainz, Germany, and The Skaggs Institute for Chemical Biology,
The Scripps Research Institute, La J olla, California 92037
vboehmer@mail.uni-mainz.de
Received May 24, 2004
Wide rim tetraurea derivatives (2a ,b) have been prepared from a calix[4]arene rigidified in the
cone conformation by two diethyleneglycol ether bridges between adjacent oxygens. In comparison
to the analogous tetraurea derivatives (3a ,b) of a tetrapentoxy calix[4]arene, 2a ,b show an increased
thermodynamic stability in mixtures of CDCl3 and DMSO-d6. Their kinetic stability as expressed
by the rate of guest exchange (benzene or cyclohexane against the solvent benzene-d6) is also strongly
increased by factors of 30-38. Noticeable differences for the inclusion of selected guests are found.
In tr od u ction
urea dimers.6 Half-life times are shorter in benzene, but
the introduction of bulkier residues (tritylphenyl instead
of tolyl) at the urea functions7 effectively increased the
kinetic stability by several orders of magnitude, e.g., half-
life times up to 60 h for the exchange of the included
benzene.8 In the special case of the homodimer of a
tetratritylurea, the formation of which could be induced
only by the tetramethylammonium cation as guest, the
half-life time could be increased even in DMSO-d6 to
about 90 h.9
O-Alkyl groups, Y g propyl, fix the calix[4]arene in one
of four possible conformations: cone, partial cone, 1,3-
alternate, or 1,2-alternate. The apparent C4v symmetry
expressed by the NMR spectrum of the cone conformer
is due to a rapid interconversion of two identical C2v-
symmetrical pinched cone conformations. Connection of
adjacent oxygens at the narrow rim by two di(ethylene
glycol) tethers10 leads to a more rigid calixarene skeleton
with a nearly C4v symmetrical shape, as shown by several
X-ray structures.11 This improved the complexation
properties of calixarene-based receptors designed for the
inclusion of both neutral organic molecules11b,12 and
cations.13 We therefore concluded that the stability of
tetraurea capsules could be increased as well. Here we
Previously, we1,2 had shown that calix[4]arene tetra-
ethers in the cone conformation substituted by urea
functions on the wide rim reversibly dimerize in apolar
solvents to form capsules.3 These dimers show large
association constants (Ka ≈ 106-108 M-1)4 and appear
1
kinetically stable on the H NMR time scale. Both X-ray
structure2b and H NMR studies2c demonstrate that this
1
association occurs via the formation of a seam of 16 (eight
strong and eight weak) hydrogen bonds between overall
eight urea groups. However, these capsules dissociate
upon the addition of small amounts of hydrogen-bonding
competitors such as DMSO. Initial attempts to improve
their stability by forming an unimolecular capsule5 met
with little success.
The kinetic stability of a capsular assembly, character-
ized for instance by the rate of the guest release or
exchange, strongly depends on the solvent and on the
included guest. Half-life times of more than 1000 h have
been observed in cyclohexane-d12 when such inert guests
as tetrachloromethane, 1,4-difluorobenzene, methyl-cy-
clohexane, or cyclohexane are encapsulated in tetratolyl-
† J ohannes Gutenberg-Universita¨t Mainz.
‡ The Scripps Research Institute.
(1) (a) Shimizu, K.; Rebek, J ., J r. Proc. Natl. Acad. Sci. U.S.A. 1995,
92, 12403-12407. (b) Hamann, B. C.; Shimizu, K.; Rebek, J ., J r. Angew.
Chem., Int. Ed. Engl. 1996, 35, 1326-1329. (c) Schalley, C. A.;
Castellano, R. K.; Brody, M. S.; Rudkevich, D. M.; Siuzdak, G.; Rebek,
J ., J r. J . Am. Chem. Soc. 1999, 121, 4568-4579.
(6) Vysotsky, M. O.; Bo¨hmer, V. Org. Lett. 2000, 2, 3571-3574.
(7) There is also an influence of the ether residues, which is not yet
entirely explored.
(8) Vysotsky, M. O.; Thondorf, I.; Bo¨hmer, V. Angew. Chem., Int.
Ed. 2000, 39, 1264-1267.
(9) Vysotsky, M. O.; Thondorf, I.; Bo¨hmer, V. Chem. Commun. 2001,
1890-1891.
(10) Arduini, A.; Fabbi, M.; Mantovani, M.; Mirone, L.; Pochini, A.;
Secchi, A. Ungaro, R. J . Org. Chem. 1995, 60, 1454-1457.
(11) (a) Arduini, A.; McGregor, W. M.; Paganuzzi, D.; Pochini, A.;
Secchi, A.; Ugozzoli, F.; Ungaro, R. J . Chem. Soc., Perkin Trans. 2 1996,
839-846. (b) Arduini, A.; Nachtigall, F. F.; Pochini, A.; Secchi, A.;
Ugozzoli, F. Supramol. Chem. 2000, 12, 273-291.
(12) Arduini, A.; Secchi, A.; Pochini, A. J . Org. Chem. 2000, 65,
9085-9091.
(2) (a) Mogck, O., Bo¨hmer, V., Vogt, W. Tetrahedron 1996, 52, 8489-
8496. (b) Mogck, O.; Paulus, E. F.; Bo¨hmer, V.; Thondorf, I.; Vogt, W.
Chem. Commun. 1996, 2533-2534. (c) Mogck, O.; Pons, M.; Bo¨hmer,
V.; Vogt, W. J . Am. Chem. Soc. 1997, 119, 5706-5712.
(3) For a general review see: Hof, F.; Craig, L. C.; Nuckolls, C.;
Rebek, J ., J r. Angew. Chem., Int. Ed. 2002, 41, 1488-1508.
(4) Castellano, R. K.; Craig, S. L.; Nuckolls, C.; Rebek, J ., J r. J . Am.
Chem. Soc. 2000, 122, 7876-7882.
(5) Brody, M. S., Schalley, C. A., Rudkevich, D. M., Rebek, J ., J r.
Angew. Chem., Int. Ed. 1999, 38, 1640-1644.
10.1021/jo049128f CCC: $27.50 © 2004 American Chemical Society
Published on Web 08/12/2004
J . Org. Chem. 2004, 69, 6115-6120
6115

Vysotsky et al.
SCHEME 1. Ca lix[4]a r en e Tetr a u r ea Der iva tivesa
a
Reagents and conditions: (i) RNCO, CHCl3, 20 °C, 12 h; (ii) 4b, (i-Pr)2EtN, DMF, 70 °C, 15 h.
report thermodynamic and kinetic properties of tetra-
ureas derived from this calix[4]arene with a rigidified
cone conformation.
corresponding to the eight strongest hydrogen bonds, by
four pairs of meta-coupled doublets for calixarene protons
A to b′, and by four pairs of doublets for ArCH2Ar protons
(Figure 1). Tetraurea 2b with bulky 4-(tert-butyl-trityl)-
phenyl units attached to the urea functions showed in
CDCl3 and benzene-d6 also the expected spectrum of a
C2-symmetrical dimer.
Resu lts a n d Discu ssion
Coupling of the tetraamine 1,13 available in three steps
from tert-butyl calix[4]arene (O-alkylation by diethyl-
eneglycol ditosylate,10,14 ipso-nitration with fuming nitric
acid in refluxing acetic acid, reduction with hydrazine
(or hydrogen)), with excess tolyl isocyanate gives 2a in
60% yield (Scheme 1). Yields of tetraurea calixarene 2b
were lower (5-7%) as a result of severe problems during
purification. Better results were obtained in this case
with the active urethane 4b as reagent (an alternative
also for the preparation of 2a ), since 2b precipitates from
the solution.15 The compound 3b can be prepared in two
ways starting either from the tetrapentyl ether tetra-
amino calix[4]arene16 and urethane 4b or from the active
urethane of tetraamino calix[4]arene17 and 4-(tris(4′-tert-
butyl-phenyl)methyl)aniline. The compounds 2a ,b, 3b,
and 4b were fully characterized by all standard methods.
It is worthwhile to note that these dimers are already
chiral by virtue of the two C2v symmetrical calix[4]arenes
(D2-symmetry). The directionality of the hydrogen bonded
belt further reduces the symmetry from D2 to C2 but does
not lead to additional stereoisomers. This contrasts with
chiral C4-symmetrical heterodimers, where the chirality
is due only to the directionality of the hydrogen bonded
belt and the change of this directionality leads to the
opposite enantiomer.18
Disproportionation experiments between dimers of the
rigid cone calixarenes 2a ,b and dimers of the arylurea
of the flexible cone 3a were used to judge the selectivity
of the self-assembly process. Only the substituents at-
tached to the narrow rims differ. Usually, the hetero-
dimerization between two simple arylurea calixarenes is
entropically driven and gives a statistical mixture of
products,2a whereas arylurea 3a and sulfonylurea 3c
heterodimerize exclusively. This has been explained by
the matching of the acid/base properties of their hydrogen
bonding side, although the phenomenon is not yet totally
understood.19
1
The behavior of 2a was studied by H NMR in CDCl3
(Figure 1), in which it dutifully assembles as the dimeric
capsule 2a ‚2a . As a result of two short bridges connecting
the proximal oxygens of the calix[4]arene skeleton, the
monomeric tetraurea 2a has C2v-symmetry. The two
symmetry planes intersect two pairs of opposite meth-
ylene bridges. When two of those tetraureas 2a are
combined into a homodimer, the symmetry observed by
1H NMR is reduced to C2. This is clearly indicated by
the four urea N-H singlets between δ ) 9.1 and 9.4
Equimolar solutions of 2a ‚2a and 3a ‚3a were combined
and allowed to come to equilibrium over 20 h. The 1H
NMR spectrum (Figure 2) showed only the sets of peaks
from each of the homodimers 2a ‚2a and 3a ‚3a and no
peaks for the heterodimer 2a ‚3a . Another spectrum
acquired after an additional 40 h was identical to the
first, indicating that this was indeed the thermodynami-
cally favored state. In fact, even after addition of 8 equiv
of 3a , no new signals for the heterodimer appeared in
(13) Arduini, A.; Bo¨hmer, V.; Delmau, L.; Desreux, J .-F.; Dozol, J .-
F.; Garcia Carrera, M. A.; Lambert, B.; Musigmann, C.; Pochini, A.;
Shivanyuk, A. Chem. Eur. J . 2000, 6, 2135-2144.
(14) Compound 1 is prepared according to the procedure in ref 10.
We found that prewashing the NaH was crucial for successful alky-
lation.
(15) Obviously the rigidified calix[4]arene skeleton causes steric
problems, since a tert-butyl-trityl-substituted tetra urea could not be
obtained reproducibly.
(16) J acobi, R. A.; Bo¨hmer, V.; Gru¨ttner, C.; Kraft, D.; Vogt, W. New
J . Chem. 1996, 20, 493-501.
(17) Vysotsky, M. O.; Bolte, M.; Thondorf, I.; Bo¨hmer, V. Chem. Eur.
J . 2003, 9, 3375-3382.
(18) (a) Castellano, R. K.; Kim, B. H.; Rebek, J ., J r. J . Am. Chem.
Soc. 1997, 119, 12671-12672. (b) Castellano, R. K.; Rebek, J ., J r. J .
Am. Chem. Soc. 1998, 120, 3657-3663.
(19) More detailed studies of the heterodimerization between 2a and
3a and the corresponding tetratosyl-ureas are presented in a forth-
coming article.
6116 J . Org. Chem., Vol. 69, No. 18, 2004

Rigid Cone Calix[4]arene Dimers
F IGURE 1. Section of the 1H NMR spectrum of 2a ‚2a 400 MHz, CDCl3, 25 °C).
DMSO-d6. The homodimers 2b‚2b are more resistant to
DMSO-d6, showing only 4% of the monomer at 11 vol %
of DMSO-d6 and 40% of the monomer at 18% of DMSO-
d6 in the mixture (see Supporting Information, Figure
S1).
Because the signals of dimers and monomers in the
1H NMR spectra can be separately integrated (Figure 3)
it is possible to estimate and to compare the equilibrium
constants for the dimerization at a given composition of
the mixed solvent. For 11.7% DMSO-d6 in CDCl3 a value
of Ka ) 13-23 M-1 was found for 2a at c ) 1.40 × 10-2
- 3.3 × 10-3 M, and Ka ) 8000-10000 M-1 was derived
for 2b where a lower concentration of c ) 1.2 × 10-4 to
2.0 × 10-4 M had to be chosen to observe meaningful
amounts of monomer and dimer in equilibrium. Thus,
the stability of 2b is about 500 times higher than of that
of 2a . However, this ratio can be taken only as a rough
estimate, since in both cases the Ka values increased with
decreasing concentration of the tetraurea. This may be
due to the fact that the description as a simple equilib-
rium 2 h 2‚2 is oversimplified and that further species
(leading eventually to broad signals in the NMR-
spectrum) are involved. The sensitivity of these measure-
ments to traces of water may be an additional factor
influencing the accuracy.
F IGURE 2. Downfield window of the 1H NMR spectra ob-
served for the attempted heterodimerization between 2a ‚2a
(b) (0.7 mM) and 3a ‚3a (Y ) C10H21) ([) (600 MHz, CDCl3, 25
°C): (a) 1 equiv of 3a ‚3a ; (b) 2 equiv of 3a ‚3a ; (c) 4 equiv of
3a ‚3a ; (d) 8 equiv of 3a ‚3a .
1
the H NMR. The behavior of mixtures of 2b and 3a was
similar, i.e., no heterodimerization was observed.
Denaturation of 2a ‚2a with DMSO-d6 (Figure 3) was
done to measure its tolerance toward hydrogen bond
competitors.20 DMSO-d6 was titrated into a solution of
2a or 2b (at c ) 1 mmol) in CDCl3 until a 1:1 ratio of
assembly to dissociated monomer (as observed in 1H
NMR) was reached. The capsule 2a ‚2a did not reach this
equilibrium point until the solution contained 8 vol %
DMSO. This is roughly four times the amount of DMSO-
d6 tolerated by the conformationally flexible calixarene
dimers, which exist solely as the monomer in 10 vol %
Kinetic measurements were carried out in benzene-d6
for the exchange of encapsulated benzene (C6H6) or
cyclohexane (C6H12) against the solvent to characterize
the kinetic stabilities of the dimers. When the benzene
complex of a dimer of 2a or 2b is dissolved in benzene-
d6, the intensity of the signal of encapsulated benzene
(Figure 4a) decreases according to a first-order rate law
(Figure 4b). The half-life time of this process is 82 s for
2a ‚C6H6‚2a (at 25 °C) and 2460 s for the 2b‚C6H6‚2b. For
the flexible tolyl urea 3a the rate of exchange (benzene
for benzene-d6) was too fast to be measured in this way,21
whereas t1/2 ) 550 s was found for 3b‚C6H6‚3b. Because
tetraurea dimers with included cyclohexane show a much
(20) For the studies on double-rosettes motif thermodynamically
stable in mixtures with DMSO-d6 (up to 70%), see: Felix, O.; Crego-
Calama, M.; Luyten, I.; Timmerman, P.; Reinhoudt, D. N. Eur. J . Org.
Chem. 2003, 1463-1474.
J . Org. Chem, Vol. 69, No. 18, 2004 6117

Vysotsky et al.
F IGURE 3. Denaturation experiments of the dimer 2a ‚2a in CDCl3/DMSO-d6 mixtures (600 MHz, 25 °C, [2a ] ) 1 mmol): (a)
pure CDCl3; (b) 0.4% DMSO; (c) 1.4% DMSO; (d) 2.3% DMSO; (e) 4.2% DMSO; (f) 6.0% DMSO; (g) 7.7% DMSO; (h) 11% DMSO.
F IGURE 4. (a) Time-dependent 1H NMR spectra (400 MHz, benzene-d6) of the 2b‚C6H6‚2b. (b) First-order kinetic plot of the
guest exchange reaction C6H6 vs C6D6.
higher kinetic stability, we planned to carry out similar
kinetic experiments with cyclohexane as guest. However,
C6H12 could not be encapsulated by refluxing 2b‚C6H6‚
2b in cyclohexane for 48 h; under the same conditions
of a tetraurea 3a (with four pentyl groups at the narrow
rim) and of 2a are 6.3 and 245 h, respectively.
In further experiments we were able to verify the
consequences of the rigidification of the cone conforma-
tion on the inclusion of selected guests (G, methyl-
substituted cyclohexanones and cyclopentanones) in com-
parison to methylcyclopentane (M). Ratios of binding
constants KG/KM were determined for the dimers of 2a
and 3a in CDCl3 in the presence of an excess of a 1:1
mixture of both guests by integration of the methyl
signals of the included guests in the 1H NMR spectra
(examples in Figure 5, see also Figures S2 and S3). The
2b‚CHCl3‚2b was converted in only ∼30% to 2b‚C6H12
‚
2b.22 Half-life times found for the cyclohexane complexes
(21) For a similar dimer derived from a mixed tetraether bearing
two pentyl and two methyl groups on distal oxygens the rate of guest
exchange (benzene, k ) 0.47 s-1) is comparable to the rate of the
dissociation/recombination; see ref 2c.
(22) The observation that no heterodimers are found when 2a and
2b are mixed in a 1:1 ratio in benzene-d6 (even after 15 h at 60 °C)
may be explained in a similar way by kinetic problems.
6118 J . Org. Chem., Vol. 69, No. 18, 2004

Rigid Cone Calix[4]arene Dimers
Exp er im en ta l Section
The tetraamine 113,23 and the tetratolylurea calix[4]arenes
2a
3a (Y ) C5H11
,
C10H211c) were synthesized in accordance with
the published procedures.
5,11,17,23-Tetr a tolylu r ea -25,27,26,28-biscr ow n -3-ca lix-
[4]a r en e (2a ). Tolyl isocyanate (0.193 g, 1.45 mmol) was added
to a solution of amine 1 (0.280 g, 0.447 mmol) in chloroform
(20 mL), and the reaction mixture was stirred for 12 h at 20
°C. Methanol (50 mL) was added, and a white precipitate
formed was filtered off, washed with methanol (3 × 10 mL),
and crystallized additionally from
a chloroform/methanol
mixture, yielding a white crystalline compound (0.310 g,
60%): mp >300 °C dec; 1H NMR (600 MHz, DMSO-d6) δ 8.41
(s, 4H, NH), 8.17 (s, 4H, NH), 7.27 (d, 8H, 3J H,H ) 8.3 Hz, CH),
3
7.11 (br d, 4H, CHcal), 7.08 (br d, 4H, CHcal), 7.03 (d, 8H, J H,H
2
) 8.3 Hz, CH), 4.92 and 3.19 (two d, 4H, J H,H ) 11.7 Hz,
2
ArCH2Ar), 4.40 and 2.92 (two d, 2H, J H,H ) 11.8 Hz, ArCH2-
Ar), 4.20-4.13 (m, 12H, OCH2CH2O), 3.64-3.59 (m, 4H, OCH2-
1
CH2O), 2.21 (s, 12H); H NMR (400 MHz, CDCl3, integration
is given for a dimeric species): δ 9.35 (s, 2H, NH), 9.26 (s, 2H,
F IGURE 5. High-field window of 1H NMR spectra (600 MHz,
chloroform-d, 25 °C) recorded for ureas 2a and 3a in the
presence of mixtures of the indicated guest and methylcyclo-
pentane. The signals of the methyl group of the included
methylcyclopentane are indicated (b). [Host] ) 1 mM, [guest]
) 20 mM, [methylcyclopentane] ) 20 mM.
4
NH), 9.22 (s, 2H, NH), 9.13 (s, 2H, NH), 7.80 (d, 2H, J H,H
)
2.0 Hz, ArH), 7.78 (d, 2H, 4J H,H ) 2.0 Hz, ArH), 7.72-7.67 (m,
16H, ArH), 7.64 (d, 2H, 4J H,H ) 2.0 Hz, ArH), 7.61 (d, 2H, 4J H,H
) 2.0 Hz, ArH), 7.38 (s, 2H, NH), 7.23-7.06 (m, 22H, ArH
4
4
and NH), 5.98 (d, 2H, J H,H ) 2.0 Hz, ArH), 5.94 (d, 2H, J H,H
) 2.0 Hz, ArH), 5.92 (d, 2H, 4J H,H ) 2.0 Hz, ArH), 5.86 (d, 2H,
4J H,H ) 2.0 Hz, ArH), 4.69 (d, 2H, J H,H ) 11.8 Hz, ArCH2Ar),
2
TABLE 1. Rela tive Bin d in g Con sta n ts KG/KM of Dim er s
of Rigid a n d F lexible Tetr a u r ea Ca lix[4]a r en es 2a a n d 3a
(Y ) C10H21) for Differ en t Gu ests vs Meth ylcyclop en ta n ea
2
4.67 (d, 2H, J H,H ) 11.6 Hz, ArCH2Ar), 4.23-3.94 (m, 28H,
2J H,H ) 11.8 Hz, ArCH2Ar and OCH2), 3.65-3.56 (m, 8H,
OCH2), 2.88-2.77 (m, 8H, ArCH2Ar), 2.26 (s, 24H, ArCH3);
13C NMR (75 MHz, DMSO-d6) δ 153.4; 151.2, 149.6, 138.1,
136.0, 135.4, 134.7, 131.1, 129.9, 129.4, 120.3, 119.3, 118.9,
77.3, 74.9, 21.1; ESI/MS m/z (relative intensity, %) 1179.5 (M
+ Na+, 100).
guest
2a ‚2a
3a ‚3a
4-methylcyclohexanone
3-methylcyclohexanone
2-methylcyclohexanone
3-methylcyclopentanone
2,2-dimethylcyclopentanone
2-methylcyclopentanone
0.28
1.10
0.60
0.59
2.00
1.50
1.10
0.71
0.76
1.10
1.90
2.60
4-[Tr is(4′-ter t-bu tylp h en yl)m eth yl]p h en yl Ca r ba m ic
Acid 4-Nitr op h en yl Ester (4b). A solution of 4-(tris(4′-tert-
butyl-phenyl)methyl)aniline24 (1.525 g, 3.03 mmol) and 4-ni-
trophenylchloroformiate (0.671 g, 3.33 mmol) in chloroform (20
mL) was stirred for 3 h at 20 °C. Then hexane (50 mL) was
added, and the precipitate formed was filtered off, washed with
hexane (3 × 10 mL), and dried, yielding a white crystalline
compound (1.59 g, 78%): mp 218-223 °C; 1H NMR (400 MHz,
a
[host] ) 1 mM, [guest] ) 20 mM, [methylcyclopentane] ) 20
mM; KG and KM ) binding constants for the guest and for
methylcyclopentane, respectively.
values collected in Table 1 show the differences in the
relative binding constants for calixarenes 2a and 3a ,
which, however, are difficult to rationalize. Most probably
they are not only caused by differences in size and shape
of the cavity due to the different flexibility of the host.
From the chemical shift of the methyl signals also
changes in the preferred orientations of the guest can be
deduced that cause differences in the magnetic shielding
experienced by the guest. These factors cannot be teased
apart.
3
DMSO-d6) δ 8.26 and 7.36 (two d, 4H, J H,H ) 8.8 Hz), 7. 29
3
and 7.20 (two d, 4H, J H,H ) 8.9 Hz), 7.25 and 7.08 (two d,
3
12H, J H,H ) 8.5 Hz), 6. 95 (s, 1H), 1.30 (s, 27H); 13C NMR
(100 MHz, DMSO-d6): δ ) 155.40, 150.07, 148.51, 145.03,
143.84, 143.70, 134.15, 132.07, 130.66, 125.21, 124.16, 122.08,
117.70, 63.30, 34.29, 31.36. ESI/MS m/z (relative intensity, %)
691.5 (M + Na+, 25).
5,11,17,23-Tet r a {4-[t r is(4′-ter t-b u t ylp h en yl)m et h yl]-
p h en yl}u r ea -25,27,26,28-biscr ow n -3-ca lix[4]a r en e (2b).
Diisopropylethylamine (76 µL, 0.411 mmol) was added in one
portion to a solution of tetraamine 1 (42.8 mg, 0.069 mmol)
and active urethane 4b (229 mg, 0.343 mmol) in DMF (25 mL)
and chloroform (1 mL). The solution was stirred at 70 °C for
15 h. After cooling to room temperature, a precipitate was
filtered off, washed with DMF (3 × 5 mL), and dried. The
resulting solid was dissolved in chloroform (2 mL) and
precipitated by adding methanol (20 mL). A white crystalline
compound was filtered off, washed with methanol (4 × 5 mL),
and dried, yielding 66.5 mg (35%) of the desired compound:
mp >330 °C dec; 1H NMR (400 MHz, benzene-d6, integration
is given for a dimeric species) δ 9.88 (s, 2H, NH), 9.86 (s, 2H,
NH), 9.62 (s, 2H, NH), 9.60 (s, 2H, NH), 8.18 and 5.94 (two d,
Con clu sion s
Rigidification of the calix[4]arene skeleton by two di-
(ethylene glycol) bridges between adjacent oxygen atoms
leads to remarkable changes in the properties of the
respective tetraurea derivatives. Dimeric capsules formed
by 2a and 2b show increased thermodynamic and kinetic
stabilities in comparison with the more flexible analogues
3a and 3b. In both pairs the stability is higher for bulky
urea residues (2b/3b vs 2a /3a ). Differences in the selec-
tivity of guest inclusion are also noticed between both
types of tetraurea calix[4]arenes. Most remarkable is the
exclusive formation of both homodimers and the complete
absence of heterodimers in a solution containing a 1:1
mixture of a flexible and a rigidified tetraurea. This
observation will be further discussed in the sequel.
4
4
4H, J H,H ) 1.8 Hz, CHcal), 8.14 and 5.91 (two d, 4H, J H,H
)
3
1.8 Hz, CHcal), 8.07 and 7.55 (two d, 8H, J H,H ) 8.8 Hz, CH),
8.07 and 7.51 (two d, 8H, J H,H ) 8.8 Hz, CH), 7.97 and 5.81
3
(23) Arduini, A.; Mirone, L.; Paganuzzi, D.; Pinalli, A.; Pochini, A.;
Secchi, A.; Ungaro, R. Tetrahedron 1996, 52, 6011-6018.
(24) Gibson, H. W.; Lee, S.-H.; Engen, P. T.; Lecavalier, P.; Sze, J .;
Shen, Y. X.; Bheda, M. J . Org. Chem. 1993, 58, 3748-3756.
J . Org. Chem, Vol. 69, No. 18, 2004 6119

Vysotsky et al.
(two br d, 4H, CHcal), 7.96 and 5.81 (two br d, 4H, CHcal), 7.93
and 7.44 (two d, 8H, J H,H ) 8.8 Hz, CH), 7.93 and 7.41 (two
evaporated under reduced pressure, and the residue was
dissolved in 250 mL of chloroform. which was washed with 1
M solution of potassium carbonate (3 × 50 mL) and water (3
× 50 mL), dried over MgSO4, and evaporated. The solid
obtained after treatment with ethyl acetate was additionally
crystallized from a chloroform/methanol mixture, giving a
white crystalline compound (0.75 g, 64%), which is the
chloroform complex of the desired tetraurea dimer.
Den a tu r a tion Exp er im en ts w ith DMSO. Tetraurea de-
rivatives were dried in high vacuum (0.5 mbar) at 100 °C over
3 h and stored in a desiccator over diphosphorus pentoxide
and sodium hydroxide. CDCl3 was treated with aluminum
oxide (basic) for at least 12 h prior to all measurements.
DMSO-d6 was used without any additional purification. 1H
NMR denaturation experiments were carried out directly in
a NMR tube by adding a solvent mixture to a solution of the
tetraurea calix[4]arene in chloroform-d.
3
3
3
d, 8H, J H,H ) 8.8 Hz, CH), 7.37 and 7.13 (two d, 24H, J H,H
)
3
8.8 Hz, CH), 7.36 and 7.12 (two d, 24H, J H,H ) 8.8 Hz, CH),
3
7.35 and 7.23 (two d, 24H, J H,H ) 8.8 Hz, CH), 7.34 and 7.21
3
(two d, 24H, J H,H ) 8.8 Hz, CH), 7.01 (s, 2H, NH), 6.88 (s,
2H, NH), 6.82 (s, 2H, NH), 6.76 (s, 2H, NH), 5.18 and 3.31
2
(two d, 4H, J H,H ) 10.4 Hz, ArCH2Ar), 5.14 and 3.28 (two d,
4H, 2J H,H ) 10.4 Hz, ArCH2Ar), 4.48 and 3.30 (two d, 8H, 2J H,H
) 10.0 Hz, ArCH2Ar), 4.3-3.7 (m, 32H, OCH2CH2O), 1.28 (s,
54H, C(CH3)3), 1.27 (s, 54H, C(CH3)3), 1.22 (s, 108H, C(CH3)3);
MALDI-TOF/MS m/z (relative intensity, %) 2461.5 (M + Na+,
10).
5,11,17,23-Tet r a {4-[t r is(4′-ter t-b u t ylp h en yl)m et h yl]-
p h en yl}u r ea -25,26,27,28-tetr a p en toxy ca lix[4]a r en e (3b).
Meth od a . A solution of the respective tetraamino calix[4]-
arene16 (0.101 g, 0.132 mmol), the active urethane 4b (0.396
g, 0.592 mmol), and diisopropylethylamine (145 µL, 0.789
mmol) in THF (25 mL) was refluxed for 6 h. After cooling, the
mixture was evaporated under reduced pressure, and the
residue was passed through a silica column (ethyl acetate/
hexane mixture 1:5, Rf ) 0.93). The crude product was treated
first with ethyl acetate and crystallized from a chloroform/
methanol mixture, giving a white powder, which is the
chloroform complex of the desired tetraurea dimer (0.207 g,
54%): mp 274-276 °C; 1H NMR (400 MHz, chloroform-d,
integration is given for a dimeric species) δ 9.56 (s, 8H, NH),
Kin etic Stu d ies. Benzene complexes of dimeric capsules
were prepared as follows. The tetraurea calix[4]arene was
dissolved in benzene (p.a.), stirred with aluminum oxide (basic,
50 mg per 10 mg of the substance) for 15 min, filtered, and
evaporated to dryness. Complexes with cyclohexane were
prepared from benzene (or chloroform) complexes by heating
them up in cyclohexane till a clear solution was formed. The
solvent was evaporated under reduced pressure, and the
residue was dried (0.5 mbar). Benzene or cyclohexane com-
plexes of dimeric capsules were dissolved in benzene-d6 and
3
1
7.83 and 7.14 (two d, 32H, J H,H ) 8.8 Hz, CH), 7.78 and 5.59
transferred to a NMR tube, and a series of H NMR spectra
(two d, 16H, 4J H,H ) 2.1 Hz, CHcal), 7.08 and 6.93 (two d, 96H,
3J H,H ) 6.9 Hz, CH), 6.45 (s, 4H, NH), 4.22 and 2.89 (two d,
(25 °C) was immediately recorded in appropriate time intervals
using an automated procedure for the spectra acquisitions.
Kinetic runs for each complex were repeated three times,
leading to deviations of less than (10% from the averaged
value.
2
16H, J H,H ) 11.4 Hz, ArCH2Ar), 3.61 and 3.53 (two q, 16H,
3
2J H,H ) 3J H,H ) 8.7 Hz, OCH2), 1.93 (quintet, 16H, J H,H ) 7.8
Hz, OCH2CH2), 1.24 (s, 216H, C(CH3)3), 1.30-1.15 (m overlap-
ping with the previous singlet, 32H, CH2CH2CH3), 0.86 (t, 24H,
3J H,H ) 7.8 Hz, CH2CH3); 13C NMR (100 MHz, chloroform-d) δ
153.78, 151.15, 148.11, 143.76, 142.41, 137.72, 134.90, 134.74,
133.48, 132.67, 130.72, 123.89, 117.66, 116.22, 116.11, 75.64,
63.03, 34.20, 31.38, 30.01, 29.81, 28.23, 22.80, 14.26; ESI/MS
m/z (relative intensity, %) 1465.1 (M + 2Na+, 100), 1945.9 (2M
+ 3Na+, 19.7), 2907.3 (2M + Na+, 82.01).
Ack n ow led gm en t. We are grateful to the Deutsche
Forschungsgemeinschaft (project BO 523/14-2), The
Skaggs Institute, and the National Institute of General
Medical Sciences for support of this research.
Su p p or tin g In for m a tion Ava ila ble: Denaturation ex-
periments with urea 2b in chloroform-d/DMSO-d6 mixtures
(1H NMR) and competitive inclusion of methylcyclopentane vs
selected guests into the rigid-calix dimer 2a ‚2a and into the
flexible-calix dimer 3a ‚3a (1H NMR). This material is available
free of charge via the Internet at http://pubs.acs.org.
Meth od b. A solution of the active urethane of the tetra-
amino calix[4]arene17 (0.543 g, 0.404 mmol) and 4-(tris-4′-tert-
butyl-phenyl)methyl)aniline (1.041 g, 2.02 mmol) in a mixture
of DMF (30 mL) and chloroform (2 mL) was stirred at room
temperature for 0.5 h. Then diisopropylethylamine (0.744 mL,
4.04 mmol) was added in one portion, and the solution was
stirred at 70 °C for 12 h. After cooling, the solvent was
J O049128F
6120 J . Org. Chem., Vol. 69, No. 18, 2004