NMR studies of the reversible dimerization and guest exchange processes of tetra urea calix[4]arenes using a derivative with lower symmetry

Article abstract of DOI:10.1021/ja970078o

The calix[4]arene derivative 3 substituted at the wider rim by four urea residues was prepared in three steps from the corresponding tert-butyl calix[4]arene 2a. Due to the different ether residues attached to the narrow rim its constitution is C(2v)-symm

Full text of DOI:10.1021/ja970078o

5706
J. Am. Chem. Soc. 1997, 119, 5706-5712
NMR Studies of the Reversible Dimerization and Guest
Exchange Processes of Tetra Urea Calix[4]arenes Using a
Derivative with Lower Symmetry
Oliver Mogck,† Miquel Pons,*,‡ Volker Bo1hmer,*,† and Walter Vogt†
Contribution from the Institut fu¨r Organische Chemie, UniVersita¨t Mainz,
J.-J.-Becher Weg 34 SB1, D-55099 Mainz, Germany, and Departament de Qu´ımica Orga`nica,
UniVersitat de Barcelona, Mart´ı Franque`s, 1-11, E-08028 Barcelona, Spain
ReceiVed January 9, 1997X
Abstract: The calix[4]arene derivative 3 substituted at the wider rim by four urea residues was prepared in three
steps from the corresponding tert-butyl calix[4]arene 2a. Due to the different ether residues attached to the narrow
rim its constitution is C2V-symmetrical, and the hydrogen bonded dimers 3‚3 formed in benzene have C2-symmetry.
Thus, the 1H NMR spectrum not only gives an unambiguous proof for the dimerization but also allows the determination
of the exchange rates for four sets of protons by NOESY experiments. The rate constant for the dissociation/
dimerization kd ) 0.26 ( 0.06 s-1 is in reasonable agreement with the rate constant for the exchange of included
and free benzene ke ) 0.47 ( 0.1 s-1. It was also shown that the formation of dimers is induced by the presence
of suitable guest molecules like benzene.
Introduction
tion of N,N′-dialkyl or -diaryl urea derivatives in the crystal
lattice via hydrogen bonds between the NH- and CdO-
functions8 has inspired the attachment of urea functions to a
calixarene skeleton9 in order to obtain molecules 1 which are
therefore simultanueously equipped with hydrogen bond donor
and acceptor sites. These molecules are self-complementary
and able to associate via hydrogen bonds in apolar organic
solvents.10,11
Self-organization is found as a general principle in nature,
and consequently the self-assembly of suitably functionalized
synthetic molecules to well defined supramolecular structures
in two or three dimensions has become a topic of current
interest.1 Various examples of self-complementary molecules
have been reported which form definite dimers in solution via
hydrogen bonds or via apolar forces.2 When this dimerization
occurs between suitably curved, concave molecules, the forma-
tion of molecular containers can be achieved which are able to
encapsulate smaller guest molecules in a reversible manner. The
first example reported by Rebek and Mendoza3 has been
modified in numerous ways meanwhile.4,5
1
The formation of dimers 1‚1 was first deduced from the H
NMR pattern.10 The splitting of the signal of the aromatic
protons of the calixarene skeleton into two meta coupled
doublets and the large separation of the singlets for the NH-
protons are especially characteristic. Additional evidence came
from high field shifted signals of the guest included in the cavity
of such dimers.12,13 In polar solvents, like DMSO-d6, only one
singlet for the aromatic calixarene protons and two closely
spaced singlets for the aromatic protons are found, indicating
that no dimers are formed.
Calix[4]arene derivatives which are fixed in the cone
conformation by O-alkylation offer a more or less preorganized
cavity. They are easily available and may be functionalized in
various ways by substitution at the wider rim.6,7 The organiza-
† Universita¨t Mainz.
We could prove this dimerization of tetra urea calix[4]arenes
‡ Universitat de Barcelona.
1
1 in apolar solvents independently by H NMR spectroscopy,
X Abstract published in AdVance ACS Abstracts, May 15, 1997.
(1) (a) Lehn, J.-M. Angew. Chem., Int. Ed. Engl. 1990, 29, 1304-1319.
(b) Lawrence, D. S.; Jiang, T.; Levett, M. Chem. ReV. 1995, 95, 2229-
2260. (c) Whitesides, G. M.; Mathias, J. P.; Seto, C. T. Science 1991, 254,
1312-1319. (d) Lindsey, J. S. New J. Chem. 1991, 15, 153-180.
(2) (a) Ghadiri, M. R.; Granja, J. R.; Milligan, R. A.; McRee, D. E.;
Khazanovich, N. Nature 1993, 366, 324-327. (b) Ghadiri, M. R.;
Kobayashi, K.; Granja, J. R.; Chadha, R. K.; McRee, D. E. Angew. Chem.,
Int. Ed. Engl. 1995, 34, 93-95. (c) Kobayashi, K.; Granja, J. R.; Ghadiri,
M. R. Angew. Chem., Int. Ed. Engl. 1995, 34, 95-98. (d) Cram., D. J.;
Choi, H.-J.; Bryant. J. A.; Knobler, C. B. J. Am. Chem. Soc. 1992, 114,
7748-7765. (e) Bonar-Law, R. P.; Sanders, J. K. M. Tetrahedron Lett.
1993, 34, 1677-1680.
(6) The expressions “wide” and “narrow” rim are preferred to the
somewhat ambiguous “upper” and “lower” rim.
(7) For a recent review on calixarenes, see: Bo¨hmer, V. Angew. Chem.,
Int. Ed. Engl. 1995, 34, 713-745.
(8) Etter, M.; Urbanczyk-Lipkowska, Z.; Zia-Ebrahimi, M.; Panunto, T.
W. J. Am. Chem. Soc. 1990, 112, 8415-8426.
(9) Urea functions have been also attached to the wider rim in order to
obtain host molecules for anions: (a) Scheerder, J.; van Duynhoven, J. P.
M.; Engbersen, J. F. J.; Reinhoudt, D. N. Angew. Chem., Int. Ed. Engl.
1996, 35, 1090-1093. (b) Scheerder, J.; Engbersen, J. F. J.; Casnati, A.;
Ungaro, R.; Reinhoudt, D. N. J. Org. Chem. 1995, 60, 6448-6454. (c)
Casnati, A.; Fochi, M.; Minari, P.; Pochini, A.; Reggiani, R.; Ungaro, R.;
Reinhoudt, D. N. Gazz. Chim. Ital. 1996, 126, 99-106.
(3) Wyler, R.; de Mendoza, J.; Rebek, Jr., J. Angew. Chem., Int Ed. Engl.
1993, 32, 1699-1701.
(10) Shimuzu, K. D.; Rebek, Jr., J. Proc. Natl. Acad. Sci. U.S.A. 1995,
92, 12403-12407.
(4) (a) Branda, N.; Wyler, R.; Rebek, Jr., J. Science 1994, 263, 1267-
1268. (b) Branda, N.; Grotzfeld, R. M.; Valdes, C.; Rebek, Jr., J. J. Am.
Chem. Soc. 1995, 117, 85-88. (c) Valdes, C.; Spitz, U. P.; Toledo, L. M.;
Kubik, S. W.; Rebek, Jr., J. J. Am. Chem. Soc. 1995, 117, 12733-12745.
(d) Meissner, R.; de Mendoza, J.; Rebek, Jr., J. Science 1995, 270, 1485-
1488. (e) Grotzfeld, R.; Branda, N.; Rebek, Jr., J. Science 1996, 271, 487-
489. (f) Kang, J.; Rebek, Jr., J. Nature 1996, 382, 239-241.
(5) For recent reviews on assembly and encapsulation with self-
complementary molecules, see: (a) Rebek, Jr., J. Acta Chim. Scand. 1996,
50, 707-716. (b) Rebek, Jr., J. Pure Appl. Chem. 1996, 68, 1261-1266.
(c) Rebek, Jr., J. Chem. Soc. ReV. 1996, 255-264.
(11) Cavitands derived from resorcarenes, which are widely used as
building blocks for the synthesis of covalently linked carcerands may be
also assembled via hydrogen bonds to carcerand-like capsules: Chapman,
R. G.; Sherman, J. C. J Am. Chem. Soc. 1995, 117, 9081-9082.
(12) Hamann, B. C.; Shimuzu, K. D.; Rebek, Jr., J. Angew. Chem., Int.
Ed. Engl. 1996, 35, 1326-1329.
(13) For the first example of a definite dimer formed between different
calix[4]arenes substituted at the wider rim by pyridine and carboxylic acid
functions no inclusion of guest molecules was reported. Kok, K.; Araki,
K.; Shinkai, S. Tetrahedron Lett. 1994, 35, 8255-8258.
S0002-7863(97)00078-4 CCC: $14.00 © 1997 American Chemical Society

NMR Studies of Tetra Urea Calix[4]arenes
J. Am. Chem. Soc., Vol. 119, No. 24, 1997 5707
Figure 2. Partial 1H NMR spectrum (400 MHz, 298 K) of 3 in
benzene-d6 containing ∼2% DMSO-d6.
Scheme 1
Figure 1. Schematic representation of the reversible dimerization of
tetra urea calix[4]arenes 1.
since a solution, containing two different urea derivatives, e.g.,
1a and 1b, shows signals of the heterodimer 1a‚1b in addition
to those of the homodimers 1a‚1a and 1b‚1b.14 This demon-
strates also that dimerization is a general phenomenon for urea
derivatives of type 1, which seems to be independent both from
the ether residues Y attached to the phenolic oxygens and the
residues R at the urea unit.15 In addition we recently confirmed
the dimerization for one example by single crystal X-ray
analysis, which allowed for the first time an exact geometrical
description of such a dimer.16 A highly disordered guest
molecule (benzene) was found in the cavity formed by the two
calixarenes supporting the encapsulation properties of these self-
assembled dimers, already suggested by Rebek et al.12
we show that the dimerization of tetra urea calix[4]arenes is a
template driven process
Results and Discussion
Synthesis and Characterization. Compound 3 was syn-
thesized as previously described for compounds 1 by treatment
of the corresponding amine 2c with a slight excess of p-
tolylisocyanate in chloroform at room temperature.14 The
coupling gave exclusively the tetra urea calix[4]arene 3 which
was obtained in 89% yield as a colorless solid after recrystal-
lization from chloroform/methanol. The amino derivative 2c
was obtained from the tert-butyl calixarene derivative 2a in two
steps. Ipso nitration of 2a yielded the p-nitro calixarene 2b.
Reduction with hydrogen and Raney nickel in toluene gave the
desired amino compound 2c in high yield (Scheme 1).
The 1H NMR spectrum of compound 3 in benzene-d6,
containing ∼2% DMSO-d6, is in agreement with the monomeric,
C2V-symmetrical molecule.17 Especially characteristic are the
protons in the region of 8.7-6.7 ppm (Figure 2) where four
singlets for the NH-protons and four doublets with ortho
coupling for the protons of the tolyl residues were observed.
The aromatic protons of the calixarene skeleton appeared as
two singlets and in addition in the region between 4.4-3.1 ppm
one pair of doublets was found for the protons of the bridging
methylene groups.
Due to the fact that the methoxy groups can pass the annulus
of a calix[4]arene the single molecule 3 could assume, in
principle, the partial cone or 1,3-alternate conformation with
one or two inverted anisol units. Figure 2 clearly shows that
monomeric 3 exists exclusively in the cone conformation, while
for instance the tetranitro calix[4]arene 2b is present as a mixture
of 91% partial cone and 9% cone conformers in CDCl3.18
General Symmetry Considerations. Calix[4]arenes 1 with
four identical ether and urea groups have a C4V-symmetrical
constitution. Their dimers 1‚1 are composed of pairs of
enantiomers with opposite chirality due to the opposite direction
Due to their easy synthetic access (in comparison to the other
hydrogen bonded capsules described in the literature) and to
their nearly unlimited potential of modification, these dimers
of urea calix[4]arenes may also gain some practical importance.
Therefore a more detailed knowledge of their kinetic and
thermodynamic stability is desirable. Obviously the dimers 1‚1
are stable on the usual NMR time scale, and their monomeric
form has not yet been observed in equilibrium with the dimer.
For 1b in CDCl3 the dimer is the only species detectable by 1H
NMR down to concentrations of about 2 mM, while further
dilution leads to ill defined associates. Rebek et al. were able
to follow the appearance of an additional signal, attributed to
benzene trapped in the cavity, as a function of time after the
addition of benzene to a solution of the dimer 1‚1 in p-xylene-
d10.12 They estimated a half-life time of about 8 min for this
uptake process which must be related somehow to the reversible
dissociation of the dimer. An association constant of K ) 230
M-1 was deduced for the encapsulation. Our own experiments
show, on the other hand, that the formation of heterodimers is
complete in less than 2 min after mixing two solutions
containing the homodimers.
We report here for the first time a detailed and systematic
NMR study of these exchange processes which was possible
using a dimer 3‚3 with reduced symmetry in which the two
calix[4]arene parts are distinguishable by NMR. In addition
(14) Mogck, O.; Bo¨hmer, V.; Vogt, W. Tetrahedron 1996, 52, 8489-
8496.
(17) The addition of small amounts of polar solvents to solutions in apolar
solvents leads to the formation of monomeric species.
(15) Dimerization has also been reported for some calix[4]arenes
substituted by two urea residues at the wider rim. Scheerder, J.; Vreekamp,
R. H.; Engbersen, J. F. J.; Verboom, W.; van Duynhoven, J. P. M.;
Reinhoudt, D. N. J. Org. Chem. 1996, 61, 3476-3481.
(16) Mogck, O.; Paulus, E. F.; Bo¨hmer, V.; Thondorf, I.; Vogt, W. Chem.
Commun. 1996 2533-2534.
(18) Under similar conditions, also the monomeric form of the tetra-
methoxy urea derivative 1c exists only in the cone conformation, and a
recently described “conformational control through self-assembly” could
not be observed as such in our experiments. Compare: Castellano, R. K.;
Rudkevich, D. M.; Rebek, Jr., J. J. Am. Chem. Soc. 1996, 118, 10002-
10003.

5708 J. Am. Chem. Soc., Vol. 119, No. 24, 1997
Mogck et al.
1
Figure 3. Partial H NMR spectrum (500 MHz, 298 K) of 3 in benzene-d6. The schematic formula shows the NOEs/ROEs (r - f) observed in
the dimer 3‚3 and the labeling scheme for the protons.
of the carbonyl groups. This directionality reduces the sym-
metry of each subunit to C4. Interdigitation of the urea
substituents converts the C4 axis of each monomer to an overall
S8 axis for the dimer.
Calix[4]arene 3 has effective C2V symmetry which conse-
quently is reduced to C2 when the urea substituents form an
ordered circular array of hydrogen bonds with a definite
orientation of the CdO groups. However, the lower symmetry
of 3 prevents the formation of a meso dimer with an Sn axis.
Thus the C2 axis is the only symmetry element in the dimer,
and, in particular, the two calixarene units are nonequivalent
within the dimer.
1H NMR Assignmemt of the Resonances of Calixarene 3.
For the assignment the different protons in the dimer 3‚3 are
labeled as follows (Figures 3 and 4).
(a)Different letters are used for aromatic protons belonging
to rings with different substituents: A/a refers to anisole rings
and B/b to rings with a pentyl ether substituent.
(b)Capital and small letters are used to distinguish between
nonequivalent protons in the same aromatic ring.
(c) A dash is used to differentiate between protons belonging
to the two different calixarenes in the dimer.
This labeling convention is based on the fact that a phenol
ether ring of a given type (e.g., an anisole ring) is facing two
rings of different types in the opposite calixarene in the dimer
(an anisole and a pentyloxy substituted ring). Capital letters
are assigned to the protons pointing to a ring of the same type,
and small letters are used for the protons pointing to a ring of
the opposite type. The two calixarenes in the dimer differ by
the orientation of their urea carbonyl groups. A dash is used
for both aromatic protons of a phenol ether ring if the carbonyl
group of the urea unit bound to it, is pointing to the side where
an aromatic ring of the same type is located in the opposite
calixarene.
Initially TOCSY experiments were used to correlate protons
belonging to the same spin system such as protons belonging
to the same aromatic ring (e.g., A, a; B, b; etc.) or pairs of
methylene protons in the calixarene skeleton. HMBC experi-
ments were used as previously described to assign protons
belonging to aromatic rings with different substituents.19 In this
case the carbon atoms bearing the ether residues show a three
bond coupling to both the aromatic protons of the same ring
and the R-protons in the aliphatic ether group. This experiment
shows that the signals at 8.41, 8.23, 6.41, and 6.35 ppm belong
This symmetry is entirely reflected by the 1H NMR spectrum
(Figure 3). The dimer 3‚3 contains four different phenolic units.
Each has two different aromatic protons. This leads to eight
doublets with meta coupling for the protons a, a′, A, A′, b, b′,
B, and B′ which can be clearly discerned in Figure 3. (For the
exact assignment, see below.) Correspondingly for the protons
of the NH groups directly connected to the calixarene skeleton
1
1
1
1
the four singlets R, R′, â, and â′ are found in the region
between 7.2 and 7.7 ppm, while the remaining NH resonances
2R, 2R′, 2â and 2â′ appear as three signals around 9.7-9.9 ppm
in the ratio 2:1:1. In the region of 3.1-4.5 ppm there are eight
doublets found for the protons of the methylene bridges c, d, e,
and f next to two singlets for the methoxy protons. The
p-methyl groups of the four different urea residues give singlets
in the ratio 1:2:1. The aromatic protons of these tolyl residues
3R, 3R′, 3â, and 3â′ appear as four pairs of ortho coupled
doublets, showing the identity of both sides due to fast rotation
around the aryl-nitrogen bond. The multiplet in the region ot
3.54-3.4 ppm is attributed to the O-CH2 protons of the pentyl
ether residues.
This NMR pattern unambiguously proves that 3 exists
exclusively in form of dimers 3‚3 which are stable on the NMR
time scale.
(19) Molins, M. A.; Nieto, P. M.; Sa´nchez, C.; Prados, P.; de Mendoza,
J.; Pons, M. J. Org. Chem. 1992, 57, 6924-6931.

NMR Studies of Tetra Urea Calix[4]arenes
J. Am. Chem. Soc., Vol. 119, No. 24, 1997 5709
(I) The change of the direction of the CdO‚‚‚HN hydrogen
bonds transfers a f a′, A f A′, b f b′, etc. This process can
occur either within a dimer by a more or less concerted rotation
(flipping) of the urea moieties around the aryl-NH bonds or
via dissociation of the dimer followed by recombination.
(II) The dissociation and recombination of the dimer occurs
without a change in the direction of the carbonyl groups, but
the two subunits are turned by 90° with respect to each other.
This process transfers a f A′, A f a′, b f B′, etc.
(III) The dissociation and recombination occurs with a rotation
of the subunits by 90° accompanied by a change in the direction
of the carbonyl groups. This is formally a combination of the
processes I and II and it transfers a f A, A f a, b f B, etc.
A fourth possibility for the recombination, namely the
formation of the starting dimer, cannot be observed by NMR.
Figure 4 summarizes all these possibilities.
Exchange information is normally obtained from EXSY
experiments, which are identical to NOESY experiments.
Therefore discrimination between exchange and cross-relaxation
(NOE) cross-peaks which have the same sign, except for very
small molecules, has to be exercised carefully. Rotating frame
experiments provide a useful way of discriminating between
exchange and direct cross-relaxation as the corresponding cross-
peaks have opposite signs. Complications arise, however, from
three spin effects and TOCSY transfer since both give rise to
peaks of the same sign as those originating from exchange.
Furthermore, differential off-resonance effects complicate the
quantitative interpretation of cross-peak integrals. In the case
of the dimer 3‚3 cross-peaks from the process III correlate sites
between which scalar coupling exists, and therefore TOCSY
contamination of the exchange cross peaks cannot be neglected.
Off-resonance ROESY20 strongly attenuates TOCSY transfer
and offset-dependent intensity variations. In this experiment
spin-locking is carried out at an effective field which forms an
angle Θ with the external magnetic field B0. The cross-
relaxation rate (σ) for this experiment is a linear combination
of those effective in the laboratory frame σN (i.e., in NOE
experiments, Θ ) 0) and in the on-resonance rotating frame
σR (i.e., in ordinary ROE experiments, Θ ) 90°). For molecules
outside the extreme narrowing condition σN and σR have
different sign. Therefore, there is an angle for which the two
contributions mutually cancel, and cross-relaxation is eliminated
leaving only exchange cross-peaks. Figure 5 shows a com-
parison of NOESY, on-resonance and off-resonance ROESY
(Θ ) 35.6°) spectra of dimer 3‚3 in benzene-d6 which
demonstrates that the indicated cross peaks are in fact exchange
cross-peaks.
Figure 4. Schematic representation of the homomerization pathways
for the dimer 3‚3 (upper part). Observed exchange cross-peaks in
NOESY experiments are indicated as arrows in the schematic formula
of the dimer 3‚3 (lower part).
to protons located in O-methylated phenolic units (A, a, A′,
a′), while those at 8.01, 7.71, 6.26, and 6.24 ppm are resonances
of protons in pentyl ether rings (B, b, B′, b′).
NOESY spectra show close contacts between aromatic
protons in adjacent rings of the same calixarene (i.e., B-a, A-b,
B′-a′, and A′-b′) and to those of the bridging methylene groups
(i.e., A-ceq-b, B-deq-a, B′-eeq-a′, and A′-feq-b′). These NOEs
and the information from TOCSY spectra provide the complete
connectivity around each calixarene in the dimer. This could
be extended to the urea substituents by consecutive correlations
between aromatic-NH, NH-NH, and NH-tolyl protons. In order
to complete the assignment it is necessary to establish the
relative position of the two calixarene rings. This could be
accomplished by the observation of a full set of ROE contacts
between aromatic protons in the skeleton of one calixarene to
both urea and tolyl protons from the opposite calixarene. All
these connectivities are summarized in the schematic representa-
tion of the dimer 3‚3 in Figure 3.
Rate Constants. The rate constants were derived from
NOESY experiments according to the method described by Abel
et al.21 in which the matrix containing measured peak volumes
(A) is related to the matrix containing all pairwise exchange
and cross-relaxation rates (L) by the equation
Homomerization of Dimers. The fact that the two calix-
arene units forming the dimer can be distinguished by NMR
gives the unique chance to study in detail the dynamic exchange
or regrouping processes at equilibrium in these dimers and to
elucidate their mechanisms. For instance the aromatic protons
of the anisole units, which are equivalent in the monomeric
calix[4]arene 3, are found in the four positions a, a′, A, and A′
in the dimer 3‚3. This allows, in principle, the study of three
site exchange processes, namely a f a′, a f A, and a f A′,
while the two different aromatic protons in dimers of calix[4]-
arenes 1 allow the study of only one exchange process.
These exchange processes can be realized by the following
pathways (Figure 4).
A ) Pe(Lt )
m
where tm is the mixing time and P is an array containing the
relative populations of each site. Therefore L is defined to be
1
tm
L ) ln(AP-1)
(20) Desvaux, H.; Berthault, P.; Birlirakis, N.; Goldman, M. J. Magn.
Reson. 1994, 108, 219-229.
(21) Abel, E. W.; Coston, T. P. J.; Orrell, K. G.; Sik, V.; Stephenson,
D. J. Magn. Reson. 1986, 70, 34-53.

5710 J. Am. Chem. Soc., Vol. 119, No. 24, 1997
Mogck et al.
Table 1. Rate Constants kI-kIII for the Three Observable
Homomerization Pathways of Dimers 3‚3 in Benzene-d6 Derived
from NOESY Experiments with Different Mixing Times tm (500
MHz, 298 K)
entry tm [ms]
kI [s-1
]
kII [s-1
]
kIII [s-1
]
1a
2a
3b
4b
5c
200
1000
500
1000
300
500
1000
500
0.054 ( 0.004 0.041 ( 0.004 0.053 ( 0.003
0.059 ( 0.004 0.069 ( 0.004 0.054 ( 0.003
0.151 ( 0.013 0.078 ( 0.010 0.038 ( 0.005
0.149 ( 0.040 0.073 ( 0.008 0.051 ( 0.008
0.074 ( 0.009 0.072 ( 0.006 0.056 ( 0.005
0.069 ( 0.025 0.065 ( 0.015 0.059 ( 0.028
0.075 ( 0.007 0.073 ( 0.007 0.063 ( 0.007
0.064 ( 0.012 0.080 ( 0.026 0.066 ( 0.022
6c
7c
8c,d
av
0.066e
0.069
0.055
a c(3) ) 7.3 mg mL-1, freshly prepared. b Solution of a stored for 3
months. c c(3) ) 6.3 mg mL-1, freshly prepared and sealed. d XD-
NOESY experiment. e Calculated without entries 3 and 4.
intensity of the cross-peaks used to quantify the rates of
exchange. A control XD-NOESY22 experiment in which the
urea NH protons were removed from the relaxation pathways
showed that indeed the rates obtained by the full relaxation plus
exchange matrix method are pure exchange rates, not affected
by cross-relaxation. In a XD-NOESY experiment selected
resonances are inverted in the middle of the relaxation time,
and, therefore, cross-relaxation to these protons is effectively
suppressed. The intensities of the exchange cross-peaks are
different in NOESY and XD-NOESY experiments with the same
mixing time, but the rates extracted by the method of Abel et
al. were effectively the same (Table 1).
Rate constants for the three exchange processes calculated
from different experiments are collected in Table 1. Comparing
these values it must be considered that the experimental error
for kII should be slightly higher, since the cross peaks from
which this rate was derived fall closer to the diagonal.
Therefore, with the exception of entries 3 and 4 of kI (Vide infra)
it can be stated that all three exchange processes occur with
the same rate of kI-III ) 0.065 ( 0.015 s-1. This can be
explained only as the result of a single mechanism, the
dissociation of the dimer, and its statistical recombination.
Keeping in mind that a fourth combination possibility exists,
the overall rate constant for this dissociation/dimerization
process must be 4 x kI-III ) kd ) 0.26 ( 0.06 s-1
.
Experiments 3 and 4 led to a higher value for kI while kII
and kIII remain unchanged. It seems reasonable to assume that
in these cases the proton exchange additionally occurs via the
reorientation of the CdO‚‚‚HN hydrogen bonds within the
dimer. This process might be catalyzed by traces of water
accumulated in the sample during storage, which is indicated
also by the broadening of the NH-peaks.
Guest Exchange. When normal benzene is added to a
solution of 3‚3 in benzene-d6 a signal appears at 4.22 ppm,
which is attributed to benzene included in the cavity of the
dimer. We were able to prove the origin of this signal by the
observation of an exchange cross-peak with “free” benzene (at
7.15 ppm) in a NOESY experiment. This allows also the
estimation of the exchange rate (ke) of benzene in and out of
the cavity. The value of ke ) 0.47 ( 0.1 s-1 is comparable
with the rate constant for the dissociation/recombination process
of the dimer 3‚3 and is much higher than the rate for the uptake
of the guest under the conditions reported by Rebek et al.12 This
probably reflects the different concentration of the guest in a
mixed solvent system as compared to the pure solvent.
A simple experiment shows that the presence of a suitable
guest can even be decisive for the dimerization. In solvents of
Figure 5. Partial 2D NMR spectra (500 MHz, 298 K) of 3 in benzene-
d6, tm ) 300 ms: NOESY (a), ROESY (b), and off-resonance ROESY
(Θ ) 35.6°) (c). Negative cross-peaks in (b) arising from cross-
relaxation are plotted with a reduced number of contours.
The matrix logarithm requires diagonalization of the original
matrix and taking the logarithm of the eigenvalues. This method
avoids the acquisition of a series of experiments with different
mixing times, and it corrects for the effects of spin-difussion
or the effect of cross-relaxation on the observed exchange rates.
Separation of exchange and NO effects is very important for
this molecule as the protons that allow the monitorization of
the exchange processes show strong NOE contacts and com-
posite NOE-exchange pathways are expected to affect the
(22) Fejzo, J.; Westler, W. M.; Macura, S.; Markley, J. J. Magn. Reson.
1991, 92, 195-202.

NMR Studies of Tetra Urea Calix[4]arenes
J. Am. Chem. Soc., Vol. 119, No. 24, 1997 5711
relevant protons. Spectra were recorded in phase sensitive mode using
the method of States et al.24 with 1024 points in F2 and from 256 to
512 points in F1. Spectra were zero filled to 1024 × 2048 and
processed with a Gaussian apodization in both dimensions.
Integration of the signals was carried out using the built in routines
present in the VNMR software package. Before calculating the rates,
cross-peak volumes were symmetrized by averaging volumes measured
at both sides of the diagonal. The integral of the diagonal peak for the
aromatic proton b could not be determined accurately because of signal
overlapping, and, therefore, an average of the remaining diagonal peaks
was used instead. Overlapping cross-peaks between protons which have
contributions from processes I and III were divided in a way such that
the ratio of the volumes assigned to the two components were the same
as the average ratios determined for the same processes in the rest of
the well resolved peaks. All cross-peaks corresponding to type II
processes were assigned the same volume which was the average of
the best resolved peaks of this type in the same spectrum. All rates
are an average of four to six calculations in which the experimental
peak-volumes have been randomly modified within a 10% limit which
is an estimate of the integration errors, after averaging both sides of
the diagonal. Signal-to-noise issues have been addressed by adding to
the experimental value a small random number with a normal
distribution with a variance equal to the one obtained from the
integration of 50 empty regions around the area where cross peaks have
been measured.
Figure 6. Partial 1H NMR spectra (200 MHz, 298 K) of 1b in CD2Cl2
(a) and after the addition of ∼5% benzene-d6 (b).
the wrong size or shape, such as CD2Cl2, just ill defined
associates of 1b are present, as indicated by broad signals in
1
the H NMR spectrum. The addition of small amounts of a
suitable guest molecule like benzene-d6, however, induces the
formation of dimers 1b‚1b, as shown in Figure 6.23
Conclusion
The C2V-symmetrical tetra urea calix[4]arene 3 was synthe-
sized and shown to self-assemble in benzene to C2-symmetrical
hydrogen bonded dimers 3‚3 in which the two calix[4]arenes
are nonequivalent. Therefore it is possible to follow by NOESY
in this particular dimer three exchange processes between four
sets of protons, which have been unabiguously assigned by a
combination of NOESY, TOCSY, HMBC, and ROESY experi-
ments. Consequently a distinction can be made between
homomerization within the undissociated dimer and via dis-
sociation/dimerization. In contrast, a C4V-symmetrical urea
derivative 1 forming a S8-symmetrical dimer 1‚1 allows only a
single exchange process to be followed, and a distinction
between different homomerization pathways is impossible.
It could be shown that the homomerization occurs in dry
solution exclusively via dissociation/dimerization. The rate for
this process is in reasonable agreement with the rate observed
for the exchange between free and included benzene molecules.
In addition it was demonstrated that the presence of a suitable
guest can induce the dimerization under certain conditions.
From a methodological point of view, the use of a molecule
with reduced symmetry combines the advantages achieved by
a heterodimer formed from two different urea calixarenes
avoiding, however, the drawback of the simultaneous presence
of different dimers. Similarly, the direct measurement of the
exchange of a given solvent has an obvious advantage over
encapsulation studies in solvent mixtures. The use of experi-
ments which allow for distinguishing unequivocally among
cross-relaxation (NOESY), coherence transfer (TOCSY), and
exchange (EXSY) peaks demonstrates the possibility to study
exchange processes in complex systems. Finally, our results
demonstrate the advantages of using a full relaxation plus
exchange matrix to derive exchange rates uncontaminated by
cross-relaxation.
Rates of exchange of benzene in and out of the cavity were
determined at 298 K from NOESY experiments with mixing times of
300 and 500 ms, respectively, in a sample of 5.4 mg mL-1 of 3 in a
mixture of 77% benzene-H6 and 23% benzene-d6. Solvent suppression
was carried out using a Sklenar-Bax read pulse to avoid saturation of
the solvent signal.25
Syntheses. Compound 2a was prepared following the procedure
described in the literature for similar compounds.26
5,11,17,23-Tetra-tert-butyl-25,27-bis(methyloxy)-26,28-bis-
(pentyloxy)calix[4]arene 2a. Methyl iodide (5 mL, 84 mmol) was
reacted with 5,11,17,23-tetra-tert-butyl-25,27-dihydroxy-26,28-bis-
(pentyloxy)calix[4]arene (4g, 5.06 mmol) in DMF with NaH as a base.
The white solid obtained from the reaction was recrystallized from
CHCl3/MeOH to give pure 2a (3.48 g, 84%): mp 141-142 °C; the 1H
NMR spectra in CDCl3 and DMSO-d6 show just very broad, nonde-
scriptive signals of a mixture of different conformers; MS (FD) 817.4
(M+, calcd 817.2).
5,11,17,23-Tetranitro-25,27-bis(methyloxy)-26,28-bis(pentyloxy)-
calix[4]arene 2b. Compound 2a (2 g, 2.45 mmol) was dissolved in
dry CH2Cl2 (250 mL). A mixture of fuming nitric acid (15 mL, 360
mmol) and glacial acetic acid (15 mL, 260 mmol) was added with
vigorous stirring at room temperature. After 1-2 h, when the color
of the solution had changed from black-violet to orange-yellow, water
(150 mL) was added, and stirring continued for 15 min. The organic
layer was then washed with water (3 × 150 mL), dried over Na2SO4,
and evaporated to dryness. The resulting residue was dissolved in
CHCl3 (50 mL), and MeOH (150-200 mL) was added to precipitate
a crude product. The material was recrystallized from CHCl3/MeOH
to give pure pale yellow solid 2b (1.68 g, 89%) as a mixture of two
1
conformers (91% partial cone, 9% cone): mp 255-257 °C dec; H
NMR 200 MHz (CDCl3) δpc 8.24 (s, 2H, ArH), 8.19 (s, 2H, ArH),
7.83 (d, 2H, ArH, J ) 2.6 Hz), 7.19 (d, 2H, ArH, J ) 2.5 Hz), 4.07 (d,
2H, ArCH2Ar, ax, J ) 14.2 Hz), 3.93 (s, 3H, OCH3), 3.91-3.65 (m,
8H, ArCH2Ar and OCH2), 3.33 (d, 2H, ArCH2Ar, eq, J ) 14.3 Hz),
3.01 (s, 3H, OCH3), 1.99-1.87 (mbr, 4H, CH2), 1.52-1.34 (mbr, 8H,
CH2), 0.95 (t, 6H, CH3, J ) 6.8 Hz) ppm; δc 8.19 (s, 4H, ArH), 7.12
(s, 4H, ArH), 4.41 (d, 4H, ArCH2Ar, ax, J ) 13.8 Hz), 3.97 (s, 6H,
OCH3) ppm, the signals for the remaining protons are superimposed
by signals originating from the partial cone conformation; MS (FD)
772.8 (M+, calcd 772.8). Anal. Calcd for C40H44N4O12: C, 62.17; H,
5.74; N, 7.25. Found: C, 62.05; H, 5.92; N, 7.11.
Experimental Section
NMR Procedures. NMR spectra were recorded on Varian VXR-
500 or on Bruker AM400 and AC200 spectrometers at 298 K with the
solvents as internal lock and internal references (1H NMR: benzene-
d6, 7.15 ppm; CD2Cl2, 5.32 ppm). Recycle times for all 2D experiments
were chosen to be at least three times the longest T1 value for the
(24) States, D. J.; Haberkorn, R. A.; Ruben, D. J. J. Magn. Reson. 1982,
48, 286-292.
(25) Sklenar, V.; Bax, A. J. Magn. Reson. 1987, 75, 378-383.
(26) Jakobi, R. A.; Bo¨hmer, V.; Gru¨ttner, C.; Kraft, D.; Vogt, W. New.
J. Chem. 1996, 20, 493-501.
(23) Recently a similar observation was reported; see Meissner, R.;
Garcias, X.; Mecozzi, S.; Rebek, Jr., J. J. Am. Chem. Soc. 1997, 119, 77-
85.

5712 J. Am. Chem. Soc., Vol. 119, No. 24, 1997
Mogck et al.
5,11,17,23-Tetraamino-25,27-bis(methyloxy)-26,28-bis(pentyloxy)-
calix[4]arene 2c. The tetranitro compound 2b (1 g, 1.29 mmol) was
dissolved in toluene (150 mL) and hydrogenated in the presence of
Raney nickel at room temperature. After the hydrogen uptake was
complete, the catalyst was filtered off and washed with warm toluene.
Then the toluene solution was evaporated in vacuum to give the pure
6.24 (d, 2H, ArH, J ) 2.5 Hz), 4.48 (d, 2H, ArCH2Ar, ax, J ) 11.6
Hz), 4.39 (d, 2H, ArCH2Ar, ax, J ) 11.6 Hz), 4.37 (d, 2H, ArCH2Ar,
ax, J ) 11.7 Hz), 4.32 (d, 2H, ArCH2Ar, ax, J ) 11.8 Hz), 3.84 (s,
6H, OCH3), 3.83 (s, 6H, OCH3), 3.53-3.41 (m, 8H, OCH2), 3.37 (d,
2H, ArCH2Ar, eq, J ) 11.8 Hz), 3.23 (d, 2H, ArCH2Ar, eq, J ) 11.8
Hz), 3.21 (d, 2H, ArCH2Ar, eq, J ) 11.7 Hz), 3.11 (d, 2H, ArCH2Ar,
eq, J ) 11.8 Hz), 1.97 (s, 12H, ArCH3), 1.94 (s, 6H, ArCH3), 1.90 (s,
6H, ArCH3), 1.92-1.81 (m, 8H CH2), 1,33-1.28 (m, 16H CH2), 0.94-
1
tetraamino calixarene 2c (735 mg, 87%): mp 147-148 °C dec; H
NMR 200 MHz (CDCl3) δ 6.47 (sbr, 4H, ArH), 5.79 (sbr, 4H, ArH),
4.19 (dbr, 4H, ArCH2Ar, ax, J ) 11.5 Hz), 3.85-3.15 (mbr, 18H, NH2,
OCH3, OCH2), 2.93 (dbr, 4H, ArCH2Ar, eq, J ) 11.4 Hz), 1.92-1.71
(mbr, 4H, CH2), 1.59-1.31 (mbr, 8H, CH2), 0.98-0.84 (mbr, 6H, CH3)
ppm; MS (FD) 652.75 (M+, calcd 652.9).
1
089 (m, 12H, CH3) ppm; H NMR 200 MHz (98% benzene-d6, 2%
DMSO-d6) δ 8.65 (s, 2H, NH), 8.62 (s, 2H, NH), 8.32 (s, 2H, NH),
8.28 (s, 2H, NH), 7.7 (d, 4H, ArH, J ) 7.7 Hz), 7.61 (s, 4H, ArH),
7.41 (d, 4H, ArH, J ) 7.5 Hz), 7.02 (d, 4H, ArH, J ) 7.8 Hz), 6.93 (d,
4H, ArH, J ) 8.0 Hz), 6.87 (s, 4H, ArH), 4.35 (d, 4H, ArCH2Ar, ax,
J ) 12.5 Hz), 3.94 (s, 6H, OCH3), 3.57-3.44 (m, 4H, OCH2), 3.17 (d,
4H, ArCH2Ar, eq, J ) 12.3 Hz), 2.10 (s, 6H, ArCH3), 2.07 (s, 6H,
ArCH3), 1.81-1.62 (m, 4H, CH2), 1.47-1.21 (m, 8H, CH2), 0.84 (t,
6H, CH3, J ) 6.8 Hz) ppm. Anal. Calcd for C72H80N8O8: C, 72.95;
H, 6.8; N, 9.45. Found: C, 73.13; H, 7.16; N, 9.34.
5,11,17,23-Tetrakis(N′-p-tolyl-N-ureido)-25,27-bis(methyloxy)-
26,28-bis(pentyloxy)calix[4]arene 3. To a solution of 2c (340 mg,
0.52 mmol) in dry CHCl3 (30 mL) p-tolylisocyanate (295 mg, 2.2 mmol)
was added, and the mixture was kept under argon at 40 °C for 2 h.
The solution was then concentrated under reduced pressure. MeOH
was added to precipitate essentially pure 3 (561 mg, 91%). The material
was recrystallized from CHCl3/MeOH: mp 223-224 °C; 1H NMR 500
MHz (benzene-d6) δ 9.87 (s, 2H, NH), 9.86 (s, 2H, NH), 9.77 (s, 4H,
NH), 8.41 (d, 2H, ArH, J ) 2.4 Hz), 8.23 (d, 2H, ArH, J ) 2.4 Hz),
8.15 (d, 4H, ArH, J ) 8.5 Hz), 8.09 (d, 4H, ArH, J ) 8.7 Hz), 8.06 (d,
4H, ArH, J ) 8.6 Hz), 8.01 (d, 2H, ArH, J ) 2.5 Hz), 7.96 (d, 4H,
ArH, J ) 8.4 Hz), 7.71 (d, 2H, ArH, J ) 2.5 Hz), 7.61 (s, 2H, NH),
7.56 (s, 2H, NH), 7.29 (s, 2H, NH), 7.24 (s, 2H NH), 7.03 (d, 4H,
ArH, J ) 8.5 Hz), 6.94 (d, 4H, ArH, J ) 8.6 Hz), 6.92 (d, 4H, ArH,
J ) 8.7 Hz), 6.83 (d, 4H, ArH, J ) 8.4 Hz), 6.41 (d, 2H, ArH, J ) 2.5
Hz), 6.35 (d, 2H, ArH, J ) 2.5 Hz), 6.26 (d, 2H, ArH, J ) 2.4 Hz),
Acknowledgment. O.M., V.B., and W.V. acknowledge
financial support from the Deutsche Forschungsgemeinschaft
and M.P. acknowledges financial support from CICYT PB94-
0924. This work was carried out using the NMR facilities of
the Serveis Cient´ıfico Te`cnics de la Universitat de Barcelona.
The authors thank M. A. Molins for recording the HMBC
spectra.
JA970078O