Diastereoselective noncovalent synthesis of hydrogen-bonded double-rosette assemblies

Article abstract of DOI:10.1002/1521-3765(20020517)8:10<2288::AID-CHEM2288>3.0.CO;2-3

Chiral centers present either in the dimelamine components of calix[4]arene 1 or in the cyanurate components CA quantitatively induce one handedness (P or M) in the corresponding hydrogen-bonded assemblies 1₃·(CA)₆ (de>98%). The high degree of chiral induction results from the presence of six chiral centers in close proximity (C_α) to the core of the assembly. A much lower level of chiral induction is observed for assemblies with chiral centers that are more remote (C_β). All diastereomerically pure assemblies 1₃·(CA)₆ exhibit very high CD activities (|Δε_max|～100 L mol^-1cm^-1), in sharp contrast to the low CD activities (|Δε_max|≤8 L mol^-1cm^-1) shown by the free components. The assemblies display spontaneous resolution under thermo-dynamically controlled conditions (i.e., heteromeric assemblies containing both peripheral R and S centers are not observed. Remarkable assembly behavior is observed if both components 1 and CA are chiral. In general, formation of well-defined assemblies is only observed when both components contain unidirectional information for the induction of either M or P chirality.

Full text of DOI:10.1002/1521-3765(20020517)8:10<2288::AID-CHEM2288>3.0.CO;2-3

FULL PAPER
Diastereoselective Noncovalent Synthesis of Hydrogen-Bonded
Double-Rosette Assemblies
[a]
Leonard J. Prins,Ron Hulst,Peter Timmerman*, and David N. Reinhoudt*
Abstract: Chiral centers present either
in the dimelamine components of cal-
ix[4]arene 1 or in the cyanurate compo-
nents CA quantitatively induce one
handedness (P or M) in the correspond-
ing hydrogen-bonded assemblies 13 ¥
(CA)6 (de > 98%). The high degree of
chiral induction results from the pres-
ence of six chiral centers in close prox-
imity (Ca) to the core of the assembly. A
much lower level of chiral induction is
observed for assemblies with chiral cen-
ters that are more remote (Cb). All
diastereomerically pure assemblies 13 ¥
(CA)6 exhibit very high CD activities
(jDemax jꢀ 100 LmolÀ1 cmÀ1), in sharp
contrast to the low CD activities
(jDemax jꢁ 8 LmolÀ1 cmÀ1) shown by the
free components. The assemblies display
spontaneous resolution under thermo-
dynamically controlled conditions (i.e.,
heteromeric assemblies containing both
peripheral R and S centers are not
observed. Remarkable assembly behav-
ior is observed if both components 1 and
CA are chiral. In general, formation of
well-defined assemblies is only observed
when both components contain unidir-
ectional information for the induction of
either M or P chirality.
Keywords: diastereoselectivity
hydrogen bonds noncovalent
interactions self-assembly
supramolecular chirality
¥
¥
¥
¥
moderate de×s of ꢀ35%.[10a] In the meantime, quantitative
diastereoselection in capsules with a chiral exterior was
achieved.[10b] In a similar fashion, diastereomeric rosette
assemblies have been synthesized with a maximum de of
50% through the use of chiral cyanurate components.[12]
Recently, 1H NMR studies by Davis and Gottarelli et al. have
shown that the supramolecular chirality in guanosine octa-
Introduction
Biological H-bonded structures express abundant supramo-
lecular chirality; for example, the double helix of DNA,[1] the
triple helix of collagen,[2] or the a-helical coiled coil of
myosin.[3] The generation of chiral assemblies–consequently
with chiral binding sites–means that chirality plays an
important role in biological molecular recognition processes.
Control over the supramolecular chirality of synthetic assem-
blies is of crucial importance for their application in the fields
‡
mers, templated by K , is quantitatively induced by the eight
chiral sugar moieties.[13]
Meijer et al. demonstrated the strong induction of supra-
molecular helicity in H-bonded macromolecular stacks com-
prising chiral components.[14] Similar results were obtained
of molecular recognition, catalysis, and materials science.[4
6]
Just like metal-coordinated assemblies,[7, 8] many synthetic
H-bonded assemblies display supramolecular chirality as a
result of the dissymmetric arrangement of their achiral
17]
with other H-bonded macromolecular assemblies.[15 Anal-
ysis of the diastereoselectivity of the assembly process is
severely complicated by the fact that individual assemblies
may comprise domains of opposite helicity separated by helix
inversion points. For that reason, the induction of chirality in
macromolecular assemblies is commonly analyzed by means
13]
components (Figure 1).[9 The general method for control-
ling supramolecular chirality is the introduction of chiral
substituents into the components. In this way, the resulting
assembly exists as a mixture of two diastereomers, the
diastereomeric excess (de) of which is determined by the
difference in DG8.
20]
of chiral amplification experiments.[18
In a previous communication we showed that the supra-
molecular chirality of the hydrogen-bonded assemblies 13 ¥
(CA)6 can be quantitatively induced by chiral centers present
in the components (Figure 2).[21] In this paper we give full
details of these experiments, together with new results that
reveal that the extent of chiral induction is strongly related to
the distance between the chiral centers and the core of the
assembly. Furthermore, the assembly behavior of chiral
dimelamines 1 with chiral cyanurates CA is described, and
reveals a remarkable sensitivity to conflicting chiral informa-
tion present in both components.
Initial efforts to induce diastereoselectivity in H-bonded
capsules by chiral-guest encapsulation resulted only in
[a] Dr. P. Timmerman, Prof. Dr. Ir. D. N. Reinhoudt, Dr. L. J. Prins,
Dr. R. Hulst
Laboratory of Supramolecular Chemistry and Technology
MESA Research Institute, University of Twente
‡
P.O. Box 217, 7500AE Enschede (The Netherlands)
Fax : (‡31)53-4894645
E-mail: smct@ct.utwente.nl
2288
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0810-2288 $ 20.00+.50/0
Chem. Eur. J. 2002, 8, No. 10

2288 2301
Figure 1. Examples of chiral H-bonded capsules: a) Supramolecular chirality in capsules, caused by a dissymmetric arrangement of the components.[10]
b) Supramolecular chirality in a tetraurea calix[4]arene capsule consisting of differently substituted urea units.[11] c) Supramolecular chirality in rosette
assemblies as a result of two possible orientations (M or P) of the melamine fragments.[12] d) Stacking of two G-quartets in a supramolecularly chiral head-to-
‡ [13]
tail arrangement templated by K . The arrows indicate the handedness of the assemblies.
reagent, a well-known chiral shift reagent, to a 1.0 mm
solution of 1b3 ¥ (DEB)6, splitting of the calix[4]arene CH2
bridge proton signals of 1b was observed (Figure 4a, b). None
of the other proton signals was affected; this indicates that
Pirkle×s reagent forms a weak complex with 1b3 ¥ (DEB)6,
presumably through coordination of the hydroxyl group to the
nitrogen of the triazine ring of 1b that is not involved in
hydrogen bonding. The same experiment performed with an
assembly with exclusively P chirality (1c3 ¥ (RCYA)6, vide
infra) did not result in splitting of the signals; this indicates
that the splitting observed for assembly 1b3 ¥ (DEB)6 is indeed
related to the presence of both M and P enantiomers
(Figure 4c, d).
The binding of Pirkle×s reagent is too weak to cause a
measurable energy difference between the M and P isomers
of assemblies 1b3 ¥ (DEB)6. In order to study whether chiral
centers in the components can induce diastereoselection in
assemblies 13 ¥ (CA)6, chiral analogues of both 1 and CA were
synthesized. The chiral analogues of CA include both
barbiturates and cyanurates.
Results and Discussion
Calix[4]arene-based double-rosette assemblies: Earlier work
in our group showed that assemblies 13 ¥ (CA)6 form sponta-
neously when calix[4]arene dimelamines 1 are mixed either
with barbiturates or with cyanurates (CA) in a 1:2 ratio in
apolar solvents such as chloroform, toluene, or benzene.[22
24]
In principle, the assemblies can exist in three different
isomeric forms: with D3, C3h, or Cs symmetry (Figure 2).[28]
The X-ray crystal structure of assembly 1b3 ¥ (DEB)6 clearly
shows that this particular assembly fully adopts a D3-
symmetric structure (Figure 3), which is chiral.[24] This form
of supramolecular chirality is the result of an antiparallel
orientation of the two melamine fragments of 1b, which
can be either clockwise (P) or counterclockwise (M)
(Scheme 1).[31] In order for 1b3 ¥ (DEB)6 to form, all three
calix[4]arene dimelamines in the assembly must have an
identical orientation of the melamine fragments: that is, either
all (P)-1b or all (M)-1b.
The M and P isomers of assembly 1b3 ¥ (DEB)6 exist in an
enantiomeric relationship and are therefore equal in free
energy (DG8M ˆ DG8P). Consequently, assembly 1b3 ¥ (DEB)6
is present in solution as a racemic mixture of the M and P
enantiomers. Upon addition of 10 equivalents of Pirkle×s
Synthesis: Dimelamines 1a c and bis(chlorotriazine) 2 were
synthesized by literature procedures.[24] Similarly, calix[4]-
arene dimelamines 1d 1g were synthesized by treatment of
Chem. Eur. J. 2002, 8, No. 10
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0810-2289 $ 20.00+.50/0
2289

P. Timmerman, D. N. Reinhoudt et al.
FULL PAPER
Figure 2. Formation of assemblies 13 ¥ (CA)6 from calix[4]arene dimelamines 1 and barbiturates/cyanurates CA. Schematic representations of the possible
isomers with D3, C3h, and Cs symmetry. Both the M and the P enantiomers of the D3 symmetrical isomer are depicted.
Figure 4. a and b) Addition of 10 equivalents of Pirkle×s reagent to a
1.0 mm solution of racemic assembly 1b3 ¥ (DEB)6 in [D2]dichloromethane
results in a splitting of the signals corresponding to the CH2 bridge protons
of 1b (indicated with an arrow). c and d) No splitting was observed when
the same experiment was performed with assembly 1c3 ¥ (RCYA)6, which is
present exclusively as the P isomer.
bis(chlorotriazine) 2 with an excess of the corresponding
amine (99% ee) in THF (Scheme 2), in yields varying
between 65 85%.
The synthesis of cyanurates RCYA and SCYA started with
condensation of the corresponding amines (ee > 99%) with an
Figure 3. X-Ray crystal structure of 1b3 ¥ (DEB)6.[24]
2290
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0810-2290 $ 20.00+.50/0
Chem. Eur. J. 2002, 8, No. 10

Diastereoselective Synthesis of Double-Rosette Assemblies
2288 2301
excess of nitrobiuret to give the
nitrobiuret adducts ꢀR)-3 and
ꢀS)-3 (Method A, Scheme 3).[32]
Subsequent ring-closure with
diethyl carbonate and sodium
ethoxide gave RCYA and
SCYA in overall yields of
74% and 82%, respectively.
The cyanurates SPheCYA,
SValCYA, and SLeuCYA were
prepared in one step by treat-
ment of the corresponding
methyl-ester-protected l-ami-
no acids (ee
> 99%) with
N-chlorocarbonyl isocyanate,
in yields varying between
50% and 65% (Method B,
Scheme 3).[33]
The
chiral
barbiturates
RBAR and SBAR were pre-
pared in three steps starting
with the bromination of
ꢀR)-2-phenylpropan-1-ol and
ꢀS)-2-phenylpropan-1-ol (ee×s>
99%), followed by alkylation
with diethyl ethylmalonate to
give ꢀR)-5 and ꢀS)-5, respectively
(Scheme 4). Ring closure of
ꢀR)-5 and ꢀS)-5 with urea in the
presence of sodium ethoxide
gave RBAR and SBAR in over-
all yields of 14% and 16%,
respectively.
Induction of supramolecular
chirality through the use of
chiral dimelamines: Mixing di-
melamine 1d (R1 ˆ ꢀR)-1-phe-
nylethyl) and DEB in a 1:2
ratio in either chloroform or
toluene resulted in the quanti-
tative formation of assembly
1d3 ¥ (DEB)6, as shown by
1H NMR spectroscopy (Fig-
ure 5) and MALDI-TOF mass
spectrometry. The 1:2 stoichi-
ometry of the components 1d
and DEB was confirmed by
integration of the CH2 bridge-
proton resonances of 1d (k n)
and the NH-proton resonances
of DEB (a and b). The 1H NMR
spectrum of assembly 1d3 ¥
(DEB)6 in [D8]toluene is highly
characteristic of double-rosette
assemblies in general and will
therefore be discussed in detail.
Similar results were obtained in
[D]chloroform.
Scheme 2. Synthesis of calix[4]arene dimelamines 1d g.
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002 0947-6539/02/0810-2291 $ 20.00+.50/0
Chem. Eur. J. 2002, 8, No. 10
2291

P. Timmerman, D. N. Reinhoudt et al.
FULL PAPER
As a result of H-bonding, the NHDEB protons (a and b) are
strongly shifted downfield from their normal position
(dNH,free,DEB ˆ 8.40). Two signals are observed at d ˆ 13.67
and 14.59, due to the fact that the a and b protons reside in
chemically different environments within the assembly. This is
a result of the nonsymmetrical substitution of the melamine
fragments of 1d. Furthermore, it shows that the DEB
1
components are in slow exchange on the H NMR chemical
shift timescale. Similarly, strong downfield shifts are observed
for the NH and NH2 protons of the melamine fragments of 1d
(c f), which appear at d ˆ 8.71, 8.26, 7.48, and 7.14, respec-
tively. Two signals at d ˆ 6.35 and 7.48 are observed for the
aromatic protons g and h of the melamine-substituted phenyl
rings of 1d. On the basis of 1H NMR studies by Ungaro et al.,
the signal at d ˆ 6.35 (h) is indicative of a pinched cone
conformation of 1d with the two melamine-substituted
phenyls in close proximity.[34] The fact that the other signal
appears at d ˆ 7.48 (g) is attributable to the formation of a
weak hydrogen bond between this proton and the triazine ring
nitrogen not involved in H-bonding.
1H NMR spectroscopy also provides important information
on the symmetry of assembly 1d3 ¥ (DEB)6. Assembly 1d3 ¥
(DEB)6 can exist in three different isomeric forms: with D3,
C3h, or Cs symmetry. Because 1d is chiral, the D3 isomer can
be present in two diastereomeric forms, with either M or P
chirality. The symmetry of the assembly is most clearly
reflected by the number of signals observed for the NHDEB
protons.[28] Based on symmetry arguments, two, four, and
twelve signals are predicted for the D3, C3h, and Cs isomers,
respectively. The presence of only two signals at d ˆ 13.67 and
14.59 implies that assembly 1d3 ¥ (DEB)6 is exclusively present
as one of the two possible diastereomers with D3 symmetry:
M-D3 or P-D3. The presence of both the M and the P
diastereomers of the D3 isomer would result in four signals for
the NHDEB protons. That the presence of two signals for the
NHDEB protons results from a fast exchange between the M-
D3 and P-D3 isomers can be ruled out, since in that case only
one signal would be expected for protons g and h, and only
two doublets for the CH2 protons k n in 1b. The complete
absence of signals for the other diastereomer indicates that its
concentration must be ꢁ10À4 m, based on the sensitivity
threshold of 1H NMR spectroscopy. Since the assembly
concentration is 5 Â 10À3 m, this means that the chiral sub-
stituents in 1d induce a de ꢂ 98% in assembly 1d3 ¥ (DEB)6.
This corresponds to a DG8 (298 K) between the M and the P
diastereomers of ꢂ11.4 kJmolÀ1.
Scheme 3. Synthesis of cyanurates RCYA, SCYA, SPheCYA, SValCYA,
and SLeuCYA.
1
The H NMR spectra of assemblies 1e3 ¥ (DEB)6 and 1 f3 ¥
Figure 5. 1H NMR spectrum of assembly 1d3 ¥ (DEB)6 recorded in
[D8]toluene (5 mm, 400 MHz, 298 K), with assignment of the proton signals.
(DEB)6 in either [D]chloroform or [D8]toluene only show one
set of proton signals. This means that the ꢀR)-1-naphthylethyl
Scheme 4. Synthesis of barbiturates RBAR and SBAR.
2292
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0810-2292 $ 20.00+.50/0
Chem. Eur. J. 2002, 8, No. 10

Diastereoselective Synthesis of Double-Rosette Assemblies
2288 2301
substituent in 1 f and the l-alanine
substituent in 1g also quantitatively
induce a single handedness in the
corresponding assemblies.
Additional evidence for the forma-
tion of assembly 1d3 ¥ (DEB)6 was
obtained from MALDI-TOF mass
‡
spectrometry after Ag -labeling.[29, 30]
The MALDI-TOF mass spectrum
shows a strong signal at m/z ˆ 4358.3
‡
(calcd for
C
234H288N48O30 ¥ 107Ag :
4360.2), corresponding to the mono-
‡
valent Ag complex. No other signals
Figure 6. a) Part of the ROESY spectrum of 1e3 ¥ (DEB)6 recorded in [D8]toluene (1.0 mm, 400 MHz,
298 K), showing the most important connectivities. b) Three-dimensional representation of the orientation
of the ꢀS)-1-phenylethyl substituent, based on the observed connectivities.
for partially formed assemblies were
detected. Presumably, the Ag cation
‡
is complexed between the phenyl ring
of the appended ꢀR)-1-phenylethyl substituent and the
melamine-substituted phenyl rings of 1d.
The 1H NMR and MALDI-TOF spectra of 1e3 ¥ (DEB)6 are
identical to the spectra of 1d3 ¥ (DEB)6. The only difference
between these enantiomeric assemblies is reflected by their
different optical activities (vide infra).
assembly were to adopt an M configuration, connectivities
between protons Hi and H3 and protons Hj and H1 would be
expected. No connectivities between the protons of the
appended phenyl group and any of the protons H1 were
3
observed; this indicates that the phenyl group is directed away
from the assembly. In addition, the fact that five different
signals were observed for the appended phenyl group shows
that the chiral substituent adopts a very rigid conformation.
Determination of the absolute configuration of assembly 1e3 ¥
(DEB)6: Two-dimensional Rotating Frame Overhauser Effect
Spectroscopy (ROESY) was used to correlate the S absolute
configuration in the chiral substituents in 1e to the induction
of P chirality in assembly 1e3 ¥ (DEB)6. Connectivities were
observed between protons Hi and H1 H2 and protons Hj and
H2 H3 (Figure 6). These connectivities are only possible if
assembly 1e3 ¥ (DEB)6 adopts a P configuration. If the
CD spectroscopy: It was found that assemblies 1d3 ¥ (DEB)6
and 1e3 ¥ (DEB)6 have very strong CDs (jDemax jꢀ
100 LmolÀ1 cmÀ1). In sharp contrast, the individual compo-
nents 1d and 1e are hardly CD active (jDemax jꢁ
8 LmolÀ1 cmÀ1); this indicates that the observed CD is a
direct result of assembly formation (Figure 7a). The strong
* * * *
Figure 7. a) CD spectra of 1d (
), 1e ( ¥¥¥ ), 1d3 ¥ (DEB)6 (––), and 1e3 ¥ (DEB)6 (
), together with the UV spectrum of 1d3 ¥ (DEB)6. All spectra were
recorded in dichloromethane (3.0 mm for 1d and 1e, 1.0 mm for 1d3 ¥ (DEB)6 and 1e3 ¥ (DEB)6) at 298 K. b) CD and UV spectra of assemblies 1 f3 ¥ (DEB)6
* * * *
(––) and 1g3 ¥ (DEB)6 (
). All spectra were recorded in chloroform (1.0 mm) at 298 K.
Chem. Eur. J. 2002, 8, No. 10
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0810-2293 $ 20.00+.50/0
2293

P. Timmerman, D. N. Reinhoudt et al.
FULL PAPER
intensity of the Cotton effects,
the bisignate nature of the CD
curves, and the fact that the
intercept with the x-axis coin-
cides with the maximum in the
UV spectra all indicate that the
observed CD is mainly a result
of exciton coupling between
chromophores present in the
core of the assemblies.[35] It is
almost impossible to determine
which chromophores are exci-
ton-coupled, since all chromo-
phores in the assemblies have
overlapping UV absorptions.
The CD curves of (M)-1d3 ¥
(DEB)6 and (P)-1e3 ¥ (DEB)6
Figure 8. 1H NMR spectrum of assembly 1b3 ¥ (RCYA)6 in [D]chloroform (1.0 mm) at 298 K.
are perfect mirror images; this
reflects their enantiomeric rela-
tionship. The sign of the CD curve is therefore a good measure
of the supramolecular chirality of the assembly: assemblies
with M chirality give positive CD curves, whereas assemblies
with P chirality give negative CD curves. This seems to
contradict theoretical studies by Harada and Nakanishi, who
found that in cases of exciton coupling a positive CD curve
originates from a P configuration of the chromophores.[36]
However, it should be emphasized that the assignment of M
and P chirality to assemblies 13 ¥ (CA)6 is arbitrarily based on
the relative orientation of the melamine fragments in 1, and
not on the relative orientation of the exciton-coupled
chromophores. Recently published assembly studies with a
chromogenic barbiturate clearly show consistency between
the assignment of M and P chirality to the assembly and the
observed CD of the chromophores.[37]
Just like assemblies 1d3 ¥ (DEB)6 1g3 ¥ (DEB)6, assembly
1b3 ¥ (RCYA)6 is strongly CD active (Figure 9). The negative
CD curve shows that RCYA induces P chirality. It is
interesting to note that the same chiral group [ꢀR)-1-phenyl-
ethyl] induces P chirality when present in the cyanurate
component and M chirality when present in the dimelamine
The CD spectra of assemblies 1d3 ¥ (DEB)6 1g3 ¥ (DEB)6
all have very similar shapes and amplitudes (Figure 7b).
Apparently, peripheral chromophores (benzyl, carbonyl,
naphthyl) only affect the intensity of the Cotton effect at
wavelengths lower than 275 nm. For higher wavelengths, the
CD spectra of assemblies 1d3 ¥ (DEB)6 1g3 ¥ (DEB)6 are
virtually identical. From the sign of the CD curves it can be
concluded that the ꢀR)-1-naphthylethyl substituents in 1 f
induce M chirality and the l-alanine substituents in 1g induce
P chirality.
* * * *
Figure 9. CD and UV spectra of assemblies 1b3 ¥ (RCYA)6 (
(SCYA)6 (––) in dichloromethane (1.0 mm) at 298 K.
) and 1b3 ¥
Induction of supramolecular chirality through the use of chiral
cyanurates: Assembly 1b3 ¥ (RCYA)6 forms quantitatively
upon mixing 1b and RCYA in a 1:2 ratio in [D]chloroform,
as shown by the presence of the characteristic signals for
double-rosette assemblies in the 1H NMR spectrum (Fig-
ure 8). As in the cases of assemblies 1d3 ¥ (DEB)6 1g3 ¥
(DEB)6, which incorporate chiral dimelamines, only one set
of signals is observed; this means that RCYA quantitatively
induces the chirality of 1b3 ¥ (RCYA)6. Additional evidence
for the quantitative induction was obtained from the addition
of 10 equivalents of Pirkle×s reagent to a 1.0 mm solution of
1c3 ¥ (RCYA)6 in [D2]dichloromethane. In contrast to racemic
assembly 1c3 ¥ (DEB)6, no splitting of the signals was observed
(Figure 4c and d).
component (1d3 ¥ (DEB)6). This makes perfect sense if the
location of the chiral centers with respect to the calix[4]arene
is considered (Figure 2). When present in the cyanurate
component, the chiral center is located on the left-hand side of
the calix[4]arene (P isomer), whereas it is located on the right-
hand side when present in the dimelamine component.
The CD spectrum of assembly 1b3 ¥ (RCYA)6 shows a
Cotton effect at 335 nm (De335 ˆ À25 LmolÀ1 cmÀ1) originat-
ing from the NO2-substituted phenyl groups of the calix[4]-
arenes. This unambiguously shows that the chiral centers
present in RCYA create a chiral environment for the NO2-
substituted phenyl groups of 1b. The enantiomeric assembly
(M)-1b3 ¥ (SCYA)6 forms when SCYA is used; this is clearly
1
reflected in the identical H NMR spectrum and the mirror-
2294
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0810-2294 $ 20.00+.50/0
Chem. Eur. J. 2002, 8, No. 10

Diastereoselective Synthesis of Double-Rosette Assemblies
2288 2301
image CD curve. Similar results (de ꢂ 98%) were obtained for
combinations of RCYA (and SCYA) with achiral dimelamines
1a and 1c.
symmetry arguments, the number of signals can only be
explained by the presence of both the M and P diastereomers
of 1b3 ¥ (RBAR)6, with a uniform orientation of the RBAR
components.[38] Corey Pauling Kultun models indeed sug-
gest that the NO2 groups in 1b only allow one orientation of
the RBAR units in assembly 1b3 ¥ (RBAR)6, with the chiral
substituent pointing away from the assembly. The concen-
trations of both diastereomers were determined by integra-
tion of the NHDEB signals, which gave a de of 17%
(DG8(298 K) ˆ 0.9 kJmolÀ1). From the positive CD curve
measured for assembly 1b3 ¥ (RBAR)6, it was concluded that
the M diastereomer has the higher thermodynamic stability
(ÀDG8P <À DG8M) (Figure 10d).
In addition, it was found that the amino-acid-derivatized
cyanurates SPheCYA, SValCYA, and SLeuCYA all quanti-
tatively induce chirality in assemblies formed with any one of
the achiral dimelamines 1a, 1b, or 1c (de > 98%). The
1H NMR spectra, recorded variously in [D]chloroform,
[D8]toluene, or [D6]benzene, all show the exclusive presence
of one diastereomer. The CD curves all have shapes and
intensities similar to those of assemblies (1d g)3 ¥ (DEB)6
(Demax ꢀ 100 LmolÀ1 cmÀ1). The negative sign of the CD
curves shows that all l-amino-acid-based cyanurates induce
P chirality in the corresponding assemblies.
Remarkably, the de is strongly solvent-dependent (Fig-
ure 10b and c). In [D8]toluene, a much higher de of 88%
(DG8M/P(298 K) ˆ 6.9 kJmolÀ1) was observed. In [D6]benzene,
signals corresponding to the unfavorable P diastereomer were
hardly detectable; this means that RBAR nearly quantita-
tively induces M chirality in assembly 1b3 ¥ (RBAR)6 (de ˆ
96%; DG8M/P(298 K) ˆ 9.6kJmol À1). The reason for the large
difference between chloroform and toluene/benzene is not
known, but could be due to the increased strength of H-bonds
in apolar solvents such as toluene and benzene. This reduces
the flexibility of the assembly, thus enhancing the steric effects
at its periphery.
Induction of supramolecular chirality through the use of chiral
barbiturates: All chiral dimelamines and cyanurates so far
studied quantitatively induce supramolecular chirality in
assemblies 13 ¥ (CA)6. A common characteristic of all these
components is that the chiral centers are located at the closest
possible proximity (at Ca) to the core of the assemblies. In
order to determine the inducing effect of chiral centers more
remote from the core, the assembly behavior of chiral
barbiturates RBAR and SBAR, with chiral centers at Cb,
was studied.
Assembly of 1b with either RBAR or SBAR results in the
quantitative formation of assemblies 1b3 ¥ (RBAR)6 and 1b3 ¥
(SBAR)6, as shown by H NMR spectroscopy. In contrast to
Different results were obtained for assemblies 1a3 ¥
(RBAR)6 and 1c3 ¥ (RBAR)6, which lack the NO2 substituents
of 1b. The 1H NMR spectra of 1a3 ¥ (RBAR)6 and 1c3 ¥
(RBAR)6 show a large number of signals (>10) in the d ˆ
13 15 region, which prevents determination of the de.
Apparently, in the absence of the NO2 substituents, two
different orientations for RBAR are possible; this results in a
large number of different isomers of assemblies 1a3 ¥ (RBAR)6
and 1c3 ¥ (RBAR)6.
1
previously studied assemblies incorporating chiral compo-
1
nents, the H NMR spectrum of assembly 1b3 ¥ (RBAR)6 in
[D]chloroform contains two sharp sets of signals for all
protons (the d ˆ 13 15 region is depicted in Figure 10a). The
unequal intensities of the signals show that two different
isomers of assembly 1b3 ¥ (RBAR)6 are present. Based on
Figure 10. Part of the 1H NMR spectrum of assembly 1b3 ¥ (RBAR)6 in: a) [D]chloroform, b) [D8]toluene, and c) [D6]benzene. All spectra were recorded at
&
*
1.0 mm concentrations at 298 K. ˆ M diastereoisomer, ˆ P diastereoisomer, d) CD spectra of assembly 1b3 ¥ (RBAR)6 in chloroform (- - - -) and benzene
* * * *
(––), and assembly 1b3 ¥ (SBAR)6 in benzene (
), together with UV spectra of assembly 1b3 ¥ (RBAR)6 in chloroform (- - - -) and benzene (––). All
spectra were recorded at 1.0 mm concentrations at 298 K.
Chem. Eur. J. 2002, 8, No. 10 ¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0810-2295 $ 20.00+.50/0
2295

P. Timmerman, D. N. Reinhoudt et al.
FULL PAPER
The assembly studies with RBAR and SBAR clearly show
that the location of the chiral centers with respect to the
assembly core is critical for chiral induction. The smaller
inducing effect of chiral barbiturates RBAR and SBAR in
comparison with those of chiral cyanurates and dimelamines
is attributed to the increased distance between the chiral
centers and the core of the assembly.
Observation of error correction in H-bonded assemblies: All
assembly studies described so far were performed under
thermodynamically controlled conditions: the assemblies
were at thermodynamic equilibrium when the measurements
(1H NMR or CD) were performed. In order to learn more
about the chiral induction process, experiments were per-
formed under conditions in which the thermodynamic equi-
librium was reached slowly.
Assembly 1c3 ¥ (DEB)6 is, by definition, present as a
racemic mixture of the M and P enantiomers (Scheme 5).
Upon addition of 1.2 equivalents of RCYA (relative to DEB)
to a 1.0 mm solution of 1c3 ¥ (DEB)6 at À208C, the barbitu-
rates are replaced by cyanurates, because of their higher
Figure 11. Part of the 1H NMR spectrum of assembly 1c3 ¥ (RCYA)6 in
[D8]toluene: a) immediately after the addition of RCYA at 253 K, b) at
273 K (ꢀ10 minutes), and c) at 298 K (ꢀ20 minutes).
However, with increasing temperature, the two signals at d ˆ
15.10 and 13.86slowly disappear (Figure 11b) and ultimately
only one set of signals remains, at d ˆ 14.52 and 14.21
(Figure 11c).[42] These signals correspond to the P diaster-
eomer of 1c3 ¥ (RCYA)6, which has the highest thermody-
namic stability. These results clearly illustrate the error-
correction process that occurs in dynamic assemblies, and
which is one of the most attractive properties of noncovalent
synthesis.
Spontaneous resolution under dynamic conditions: The
exclusive formation of assemblies (M)-1d3 ¥ (DEB)6 and (P)-
1e3 ¥ (DEB)6 shows the strong preference of dimelamines 1d
and 1e to become incorporated into assemblies with M or P
chirality, respectively. In order to study the possibility of
whether 1d or 1e can also adopt the unfavorable P or M
configurations, mixtures of assemblies (M)-1d3 ¥ (DEB)6 and
(P)-1e3 ¥ (DEB)6 were studied.[21] The exchange of achiral
components 1a and 1b in a mixture of racemic homomeric
assemblies 1a3 ¥ (DEB)6 and 1b3 ¥ (DEB)6 causes the forma-
tion of heteromeric assemblies 1an1b3Àn ¥ (DEB)6 (n ˆ 1,2).
Previously, 1H NMR studies showed that, for these assemblies,
the distribution is nearly statistical (homomer/heteromerˆ
1:3).[43] Mixtures of diastereomerically pure assemblies (M)-
1d3 ¥ (DEB)6 and (P)-1e3 ¥ (DEB)6 display completely differ-
ent behavior. The 1H NMR spectrum of a 1:1 mixture of these
assemblies in [D8]toluene is identical to that of the separate
assemblies (Figure 12a c). No additional signals were ob-
served; this indicates that the heteromeric assemblies
1dn1e3Àn ¥ (DEB)6 (n ˆ 1,2) were not formed to a significant
extent. Titration experiments of (M)-1d3 ¥ (DEB)6 with (P)-
1e3 ¥ (DEB)6, monitored by CD spectroscopy, show similar
results (i.e., a strictly linear decrease in the CD intensity
(Figure 12d)). Other systems commonly show a different
dependence of the CD intensity on the R:S ratio.[44, 45] In those
Scheme 5. The addition of 1.2 equivalents of RCYA (relative to DEB) to a
1.0 mm solution of assembly 1c3 ¥ (DEB)6 in [D8]toluene at 253 K results in
the formation of diastereomeric assemblies (M)-1c3 ¥ (RCYA)6 and (P)-
1c3 ¥ (RCYA)6.
affinity for the melamine units.[7, 39, 40] As a result, a 50:50
mixture of the M and P diastereomers of 1c3 ¥ (RCYA)6 is
formed, as judged from the presence of two main sets of
signals for the NH protons in the resulting 1H NMR spectrum
of assembly 1c3 ¥ (RCYA)6 (Figure 11a). The additional small
signals most probably arise from heteromeric assemblies
containing both DEB and RCYA. Previously reported kinetic
studies have revealed that under these conditions (À208C,
[D8]toluene) the interconversion between the M and P
enantiomers is very slow.[41] As a result, both the M and P
diastereomers of 1c3 ¥ (RCYA)6 are initially formed in equal
amounts as the kinetic products of the exchange reaction.
2296
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0810-2296 $ 20.00+.50/0
Chem. Eur. J. 2002, 8, No. 10

Diastereoselective Synthesis of Double-Rosette Assemblies
2288 2301
assemblies 1dn1e3Àn ¥ (DEB)6 (n ˆ 1,2); this emphasizes their
strong preference either for M or for P helicity.
The self-resolution displayed in solution by these H-bonded
assemblies had previously only been observed for H-bonded
assemblies in the solid[17] and liquid-crystalline[46] states. Very
recently, Davis et al. reported a similar enantiomeric self-
sorting in a racemic mixture of d- and l-isoguanosines.[47, 48]
Assemblies incorporating both chiral dimelamines and chiral
cyanurates: All studies discussed so far have demonstrated
the quantitative inducing effect of both chiral cyanurates and
dimelamines on the supramolecular chirality of the corre-
sponding assemblies. This raises the question of what happens
if both components 1 and CA are chiral and have opposite
preferences for M or P chirality. Therefore, the assembly
behavior of both 1d and its enantiomer 1e with chiral
1
cyanurate SCYA was studied. H NMR spectra recorded in
[D2]dichloromethane reveal the formation of both assemblies
1d3 ¥ (SCYA)6 and 1e3 ¥ (SCYA)6 (Figure 13a and b). Both
spectra show sharp signals, indicating well-defined assemblies.
Integral comparison of the CH2 bridge protons of dimel-
amines 1d and 1e with the NHSCYA protons shows that both
assemblies 1d3 ¥ (SCYA)6 and 1e3 ¥ (SCYA)6 form nearly
quantitatively (ꢂ98 and 95%, respectively). The difference
between the 1H NMR spectra reveals the diastereomeric
relationship between assemblies 1d3 ¥ (SCYA)6 and 1e3 ¥
(SCYA)6. CD spectroscopy shows a strong optical activity
for assembly 1d3 ¥ (SCYA)6 with characteristic shape and
intensity (Figure 13c). The positive sign shows that assembly
1d3 ¥ (SCYA)6 displays M chirality. This is consistent with
previous studies that revealed that both 1d and SCYA induce
M chirality. On the other hand, assembly 1e3 ¥ (SCYA)6 is
hardly CD active (Figure 13c).[49] This shows that the con-
flicting chiral components 1e and SCYA either strongly affect
the orientation of the chromophores responsible for the CD
activity or coincidentally cause identical but opposite CD
activity. The absence of the characteristic CD curve prevents
the assignment of either M or P chirality to assembly 1e3 ¥
(SCYA)6.
Figure 12. Sections of the 1H NMR spectra of: a) 1d3 ¥ (DEB)6, b) 1e3 ¥
(DEB)6, and c) a 1:1 mixture of 1d3 ¥ (DEB)6 and 1e3 ¥ (DEB)6. All spectra
were recorded in [D8]toluene (1.0 mm) at 298 K. d) Plot of the CD intensity
&
at 286nm ( ) versus the mole fraction of 1d in a mixture of assemblies 1d3 ¥
(DEB)6 and 1e3 ¥ (DEB)6. The dashed line represents the expected curve if
no heteromeric assemblies 1dn1e3Àn ¥ (DEB)6 (n ˆ 1,2) are formed. All CD
spectra were recorded in toluene (1.0 mm) at 298 K.
cases, the supramolecular chirality of the assemblies is
determined by the excess of R over S components or vice
versa, a principle commonly referred to as the ™majority
rule∫.[44] For the assemblies discussed here, this would imply
that assembly 1d21e ¥ (DEB)6 should have a preference for M
chirality, since it incorporates two R components 1d and one S
component 1e. Analogously, assembly 1d1e2 ¥ (DEB)6 should
have a preference for P chirality. The presence of these
heteromeric assemblies would result in a nonlinear decrease
in the CD intensity for increasing amounts of (P)-1e3 ¥ (DEB)6
up to 50%. The observed linear decrease thus implies that
components 1d and 1e do not participate in the heteromeric
The difference in assembly behavior of any of the chiral
cyanurates SPheCYA, SValCYA, or SLeuCYA with dimel-
Figure 13. 1H NMR spectra of: a) assembly 1d3 ¥ (SCYA)6 and b) 1e3 ¥ (SCYA)6 in [D2]dichloromethane (1.0 mm) at 298 K. c) CD spectra of 1d3 ¥ (SCYA)6
* * * *
(––) and 1e3 ¥ (SCYA)6 (
) in dichloromethane (1.0 mm) at 298 K.
Chem. Eur. J. 2002, 8, No. 10
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0810-2297 $ 20.00+.50/0
2297

P. Timmerman, D. N. Reinhoudt et al.
FULL PAPER
amines 1d and 1e is even more pronounced. Assembly of 1e
with any one of these cyanurates in [D]chloroform or
[D8]toluene results in the clean formation of assemblies
1e3 ¥ (SPheCYA)6, 1e3 ¥ (SValCYA)6, or 1e3 ¥ (SLeuCYA)6, as
concluded from the 1H NMR spectra (Figure 14a). The sharp
signals indicate the formation of well-defined, highly sym-
metrical assemblies. Furthermore, all assemblies exhibit the
bisignate CD curve characteristic for assemblies with P
chirality. The maximum CD intensity for assembly 1e3 ¥
(SPheCYA)6 (Demax ˆ À124 LmolÀ1 cmÀ1) is slightly higher
than commonly observed (Demax ꢀ 100 LmolÀ1 cmÀ1); this
might be caused by the increased number of chromophores
(Figure 14c). Earlier it was shown that 1e and any one of these
S cyanurates all induce P chirality in the corresponding
assemblies. Therefore, no conflicting chiral information that
would prevent assembly formation is present in the com-
ponents of 1e3 ¥ (SPheCYA)6, 1e3 ¥ (SValCYA)6, and 1e3 ¥
(SLeuCYA)6.
because 1d has a preference for M chirality. As a result, 1d3 ¥
(SPheCYA)6, 1d3 ¥ (SValCYA)6, and 1d3 ¥ (SLeuCYA)6 do not
form. Only broad signals were observed in the 1H NMR
spectra of these mixtures in both [D]chloroform and [D8]tol-
uene; this indicates the formation of undefined assemblies
(Figure 14b). H-Bonding must take place, as downfield signals
are observed in the d ˆ 13 15 region. In addition, the CD
curves have a completely different shape from the CD curves
commonly observed for well-defined double-rosette assem-
blies (Figure 14c). Compared to cyanurate RCYA, the amino-
acid-derivatized cyanurates SPheCYA, SValCYA, and SLeu-
CYA apparently either have less flexibility to adapt to the
unfavorable M chirality induced by chiral dimelamines 1d or
simply do not fit.
Conclusion
The supramolecular chirality of assemblies 13 ¥ (CA)6 can be
controlled quantitatively (de ꢂ 98%) by the incorporation of
chiral centers into either the calix[4]arene dimelamine or the
cyanurate components. The high degree of induction is a
result of the presence of a total of six chiral centers in close
proximity to the core of the assemblies. The studies with chiral
barbiturates show that the chiral induction is much less when
the chiral centers are present at the Cb-position. Furthermore,
these studies showed that the extent of chiral induction is
highly solvent dependent. A quantitative induction of chir-
ality (de ˆ 96%) was only observed in benzene. All assem-
blies are strongly CD active, in contrast to the individual
components, which hardly show any CD activity. This makes
CD spectroscopy a suitable tool with which to study assembly
formation with dilution or titration studies. The assemblies
display spontaneous resolution under thermodynamically
controlled conditions, that is, no heteromeric assemblies
containing both M- and P-inducing components are observed.
This strong preference becomes even more evident when the
assembly behavior of chiral dimelamines and chiral cyanu-
rates is considered. Generally, well-defined assemblies only
form when the complementary components contain unidirec-
tional information.
The assembly behavior of 1d with SPheCYA, SValCYA, or
SLeuCYA is completely different. A chirality conflict arises,
Application of the results described in this paper, for
example in nanotechnology[52a] and for selective guest recog-
nition, are currently under investigation. Interesting results in
both areas have recently come forward and will be described
shortly.[52b, 53]
Experimental Section
THF was freshly distilled from Na/benzophenone, EtOAc and hexane
(referring here to the petroleum ether fraction with b.p. 60 808C) from
K2CO3, and CH2Cl2 from CaCl2. All chemicals were of reagent grade and
used without further purification. NMR spectra were recorded either on a
Bruker AC250 (250 MHz) or on a Varian Unity300 (1H NMR 300 MHz)
spectrometer at room temperature. Residual solvent protons were used as
internal standard, and chemical shifts are given relative to tetramethylsi-
lane (TMS). CD spectra were recorded on a JASCO J-715 spectropo-
larimeter at room temperature. UV/Vis spectra were recorded on a
Hewlett Packard 8452A diode array spectrophotometer at room temper-
ature. FAB-MS spectra were recorded with a Finnigan MAT90 spectrom-
Figure 14. 1H NMR spectra of: a) assembly 1e3 ¥ (SPheCYA)6, and b) as-
sembly 1d3 ¥ (SPheCYA)6 in [D8]toluene (1.0 mm) at 298 K. c) CD spectra
* * * *
of assemblies 1e3 ¥ (SPheCYA)6
toluene (1.0 mm) at 298 K.
(
) and 1d3 ¥ (SPheCYA)6 (––) in
2298
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0810-2298 $ 20.00+.50/0
Chem. Eur. J. 2002, 8, No. 10

Diastereoselective Synthesis of Double-Rosette Assemblies
2288 2301
eter, with m-nitrobenzyl alcohol (NBA) as a matrix. EI mass spectra were
recorded on a Finnigan MAT90 spectrometer with an ionizing voltage of
70 eV. MALDI-TOF measurements were performed on a PerSeptive
Biosystems Voyager-DE-RP MALDI-TOF mass spectrometer equipped
with delayed extraction.[50] A UV nitrogen laser (l ˆ 337 nm) that produced
3 ns pulses was used, and the mass spectra were obtained both in the linear
and in the reflectron modes. Mass assignments were performed with
nonmanipulated spectra (no smoothing or centering, etc.). Elemental
analyses were performed with a Carlo Erba EA1106. The presence of
solvents in the analytical samples was confirmed by 1H NMR spectroscopy.
Flash column chromatography was performed with silica gel (SiO2, 0.040
0.063 mm, 230 240 mesh, Merck).
25 mL) and brine (25 mL), and dried over MgSO4. After evaporation of
the solvents, compound ꢀR)-3 was obtained as a colorless oil (3.0 g, 93%).
1
ˆ
ˆ
H NMR (400 MHz, [D6]DMSO, 258C): d ˆ 8.51 (s, 1H; C( O)NHC( O)),
3
ˆ
7.91 (d; J(H,N) ˆ 7.3 Hz, 1H; C*HNHC( O)), 7.33 7.20 (m, 5H; ArH),
6.73 (s, 2H; NH2), 4.80 (dq, 3J(H,H) ˆ 7.3 Hz, 1H; C*H), 1.35 (d, 3J(H,H) ˆ
7.3 Hz, 3H; CH3); 13C NMR (75 MHz, [D6]DMSO, 258C): d ˆ 155.4, 153.5,
‡
144.3, 128.4, 126.8, 125.6, 48.4, 22.8; MS (EI): m/z: 207.1 ([M ‡H], calcd:
208.1).
N-[ꢀS)-1-Phenylethyl]imidocarbonic acid (ꢀS)-3): Compound ꢀS)-3 was
obtained as a colorless oil (97%) by the same procedure as used for ꢀR)-3.
The 1H and 13C NMR and the MS spectra of ꢀS)-3 were identical to those of
compound ꢀR)-3.
Calix[4]arene dimelamines 1a 1c and bis(chlorotriazine) 2 were synthe-
sized according to literature procedures.[24] Barbiturate DEB was obtained
from Fluka.
N-[ꢀR)-1-Phenylethyl]-1,3,5-triazine-2,4,6(1H,3H,5H)-trione
(RCYA):
Compound ꢀR)-3 (1.0 g, 4.8 mmol) and diethyl carbonate (1.20 g,
10.2 mmol) were added to a solution of Na (0.34 g, 14.8 mmol) in EtOH
(20 mL), and the mixture was heated under reflux overnight. After the
mixture had cooled to room temperature, toluene (25 mL) was added, and
the resulting precipitate was filtered off and redissolved in H2O (25 mL).
The solution was acidified with 6n HCl to pH ꢀ 1 2, after which the
resulting precipitate was collected, washed with H2O (2 Â 25 mL), and
dried under high vacuum. RCYA was obtained as a white solid (80%).
1H NMR (400 MHz, [D6]DMSO, 258C): d ˆ 11.4 (s, 2H; NH), 7.30 7.20 (m,
5H; ArH), 5.80 (q, 3J(H,H) ˆ 7.3 Hz, 1H; C*H), 1.72 (d, 3J(H,H) ˆ 7.3 Hz,
3H; CH3); 13C NMR (75 MHz, [D6]DMSO, 258C): d ˆ 149.5, 148.6, 140.2,
General procedure for the synthesis of calix[4]arene dimelamines 1: A
solution of bis(chlorotriazine) 2, diisopropylethylamine (12 equiv.), and the
corresponding amine (36equiv.) in THF (25 50 mL) was heated under
reflux for 12 h. The mixture was evaporated to dryness. The residue was
dissolved in CH2Cl2 (50 mL), washed with H2O (2 Â 25 mL) and brine
(25 mL), and dried over Na2SO4. Evaporation of the solvent gave 1 as a
crude product, which was purified by column chromatography (SiO2,
CH2Cl2/MeOH/NH4OH 90:9.5:0.5).
5,17-N,N'-Bis[4-amino-6-ꢀR)-1-phenylethylamino-1,3,5-triazin-2-yl]diami-
no-25,26,27,28-tetrapropoxycalix[4]arene (1d): Compound 1d was ob-
tained as a white solid (65%). 1H NMR (250 MHz, [D6]DMSO, 258C):
d ˆ 8.5 (brs, 2H; NH), 7.4 7.2 (m, 12H; ArH, NH), 6.2 6.1 (m, 14H; ArH,
NH2), 5.26(brs, 2H; C HCH3), 4.32, 3.02 (ABq, 2J(H,H) ˆ 12.8 Hz, 8H;
‡
128.0, 126.7, 126.2, 49. 9, 16.2; MS (FAB): m/z: 234.1 ([M ‡H], calcd:
234.1); elemental analysis calcd (%) for C11H11N3O3 ¥ 0.3H2O: C 55.37, H
4.90, N 17.61; found C 55.27, H 4.94, N 17.33.
N-[(S)-1-Phenylethyl]-1,3,5-triazine-2,4,6(1H,3H,5H)-trione
(SCYA):
3
ArCH2Ar), 3.89, 3.63 (2t, J(H,H) ˆ 8.3 Hz, 8H; OCH2), 2.1 1.8 (m, 8H;
Compound SCYA was obtained as a white solid (85%) by the same
procedure as used for RCYA. The 1H and 13C NMR and FAB-MS spectra
of SCYA were identical to those of compound RCYA; elemental analysis
calcd (%) for C11H11N3O3: C 56.65, H 4.75, N 18.02; found C 56.55, H 4.73,
N 17.86.
OCH2CH2), 1.42 (d, 3J(H,H) ˆ 6.1 Hz, 6H; CHCH3), 1.09, 0.89 (t,
‡
3J(H,H) ˆ 7.4 Hz, 12H; OCH2CH2CH3); MS (FAB): m/z: 1049.6([ M ‡H],
calcd: 1049.6); elemental analysis calcd (%) for C62H72N12O4 ¥ 0.2CH3OH:
C 70.76, H 6.95, N 15.92; found C 70.56, H 6.83, N 15.85.
5,17-N,N'-Bis[4-amino-6-ꢀS)-1-phenylethylamino-1,3,5-triazin-2-yl]diami-
no-25,26,27,28-tetrapropoxycalix[4]arene (1e): Compound 1e was ob-
tained as a white solid (67%). The 1H NMR and FAB-MS spectrum of
1e were identical to those of compound 1d; elemental analysis calcd (%)
for C62H72N12O4 ¥ 0.2CH3OH: C 70.76, H 6.95, N 15.92; found C 70.43, H
6.75, N 15.62.
General procedure for the synthesis of cyanurates SPheCYA, SValCYA,
and SLeuCYA (Method B):[33] This reaction must be performed under
flame-dried conditions and under a continuous flow of argon. Before
leaving the system, the argon was passed through water in order to trap any
phosgene formed. The HCl salt of the corresponding methyl-ester-
protected amino acid was suspended in THF, and N-chlorocarbonyl
isocyanate (2 equiv.) was added slowly. After being stirred at room
temperature for 2 hours, the mixture was heated under reflux for 2 days.
After evaporation of the solvent, the residue was redissolved in CH2Cl2
(50 mL), washed with H2O (2 Â 25 mL), dried over Na2SO4, and purified by
column chromatography (SiO2, CH2Cl2/MeOH/NH4OH 90:9.5:0.5). Occa-
sionally, the products were recrystallized from MeOH.
5,17-N,N'-Bis[4-amino-6-ꢀR)-1-naphthylethylamino-1,3,5-triazin-2-yl]di-
amino-25,26,27,28-tetrapropoxycalix[4]arene (1 f): Compound 1 f was ob-
tained as a white solid (75%). 1H NMR (300 MHz, [D6]DMSO, 258C): d ˆ
8.6 (brs, 2H; NH), 8.3 (brs, 2H; NH), 7.9 7.2 (m, 18H; ArH), 6.3 6.2 (m,
10H; ArH, NH2), 6.1 (brs, 2H; CHCH3), 4.32, 3.03 (ABq, 2J(H,H) ˆ
12.8 Hz, 8H; ArCH2Ar), 3.80, 3.63 (2t, 3J(H,H) ˆ 8.3 Hz, 8H; OCH2),
1.9 1.8 (m, 8H; OCH2CH2), 1.62 (d, 3J(H,H) ˆ 6.1 Hz, 6H; CHCH3),
1.15, 0.91 (t, 3J(H,H) ˆ 7.4 Hz, 12H; OCH2CH2CH3); MS (FAB): m/z:
N-(l-Phenylalanine methyl ester)-1,3,5-triazine-2,4,6(1H,3H,5H)-trione
(SPheCYA): Compound SPheCYA was obtained as a white solid (62%).
1H NMR (300 MHz, [D6]DMSO, 258C): d ˆ 11.61 (s, 2H; NH), 7.28 7.15
‡
1149.6([ M ‡H], calcd: 1149.6); elemental analysis calcd (%) for
3
(m, 5H; ArH), 5.46(dd, 3J(H,H) ˆ 5.7 Hz, J(H,H) ˆ 10.2 Hz, 1H; C*H),
C70H76N12O4 ¥ 0.2CH3OH: C 72.97, H 6.69, N 14.55; found C 72.77, H 6.61,
3.39 (dd, 2J(H,H) ˆ 13.8 Hz, 3J(H,H) ˆ 5.7 Hz, 1H; C*HCHH), 3.65 (s, 3H;
OCH3), 3.17 (dd, 2J(H,H) ˆ 13.8 Hz, 3J(H,H) ˆ 10.2 Hz, 1H; C*HCHH);
13C NMR (75 MHz, [D6]DMSO, 258C): d ˆ 164.1, 143.6, 142.4, 131.5, 124.5,
N 14.89.
5,17-N,N'-Bis[4-amino-6-(N-l-alaninemethylester)-1,3,5-triazin-2-yl]di-
amino-25,26,27,28-tetrapropoxycalix[4]arene (1g): Compound 1g was
‡
1
124.2, 122.7, 51.1, 48.5, 29.8; MS (FAB): m/z: 292.1 ([M ‡H], calcd: 292.1);
obtained as a white solid (43%). H NMR (400 MHz, [D6]DMSO, 258C):
elemental analysis calcd (%) for C13H13N3O5: C 53.61, H 4.50, N 14.43;
found C 53.62, H 4.56, N 14.42.
d ˆ 8.5 (brs, 2H; NH), 7.4 7.3 (m, 4H; ArH), 7.0 (brs, 2H; NH), 6.2 6.1
2
(m, 10H; ArH, NH2), 4.5 (brs, 2H; CHCH3), 4.30, 3.08 (ABq; J(H,H) ˆ
12.8 Hz, 8H; ArCH2Ar), 3.88, 3.70 (2t, 3J(H,H) ˆ 8.3 Hz, 8H; OCH2), 3.61
(s, 6H; OCH3), 1.9 1.8 (m, 8H; OCH2CH2), 1.33 (d, 3J(H,H) ˆ 7.0 Hz, 6H;
CHCH3), 1.06, 0.87 (t, 3J(H,H) ˆ 7.4 Hz, 12H; OCH2CH2CH3); MS (FAB):
N-(l-Valine methyl ester)-1,3,5-triazine-2,4,6(1H,3H,5H)-trione (SVal-
CYA): Compound SValCYA was obtained as
a white solid (50%).
1H NMR (300 MHz, [D6]DMSO, 258C): d ˆ 11.73 (s, 2H; NH), 4.72 (d,
3J(H,H) ˆ 9.3 Hz, 1H; C*H), 3.59 (s, 3H; OCH3), 2.51 2.40 (m, 1H;
C*CH), 1.10 (d, 3J(H,H) ˆ 6.6 Hz, 3H; CCH3), 0.77 (d, 3J(H,H) ˆ 6.9 Hz,
3H; CCH3); 13C NMR (75 MHz, [D6]DMSO, 258C): d ˆ 164.8, 145.5, 144.1,
‡
m/z: 1013.4 ([M ‡H], calcd: 1013.5); elemental analysis calcd (%) for
C54H68N12O8 ¥ 0.2CH3OH: C 63.86, H 6.80, N 16.49; found C 63.74, H 6.73,
N 16.40.
‡
54.1, 48.0, 23.3, 17.4, 14.6; MS (FAB): m/z: 244.1 ([M ‡H], calcd: 244.1);
N-[ꢀR)-1-Phenylethyl]imidocarbonic acid (ꢀR)-3): Nitrobiuret[32] (2.3 g,
15.5 mmol) was added to a solution of ꢀR)-1-phenylethylamine (1.88 g,
15.5 mmol) in DMF (25 mL) and H2O (5 mL). The mixture was heated at
958C for 1 h, after which a second portion of nitrobiuret (1.88 g, 15.5 mmol)
was added. Heating was continued for 1 hour, after which a final portion of
nitrobiuret (1.88 g, 15.5 mmol) was added. After having been heated for
another hour, the mixture was cooled to room temperature, and H2O
(100 mL) was added. The aqueous layer was extracted with CH2Cl2 (3 Â
25 mL). The combined organic fractions were washed with 1n HCl (2 Â
elemental analysis calcd (%) for C9H13N3O5: C 44.45, H 5.39, N 17.28; found
C 44.55, H 5.50, N 17.25.
N-(l-Leucine methyl ester)-1,3,5-triazine-2,4,6(1H,3H,5H)-trione (SLeu-
CYA): Compound SLeuCYA was obtained as a white solid in 66% yield.
1H NMR (300 MHz, [D6]DMSO, 258C): d ˆ 11.67 (s, 2H; NH), 5.13 (dd,
3J(H,H) ˆ 5.7 Hz, 3J(H,H) ˆ 8.7 Hz, 1H; C*H), 3.60 (s, 3H; OCH3), 1.87
(m, 2H; CH2), 1.53 (m, 1H; CH(CH3)2), 0.87 (d, 3J(H,H) ˆ 6.6 Hz, 3H;
CHCH3), 0.85 (d, 3J(H,H) ˆ 6.6 Hz, 3H; CHCH3); 13C NMR (75 MHz,
Chem. Eur. J. 2002, 8, No. 10
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0810-2299 $ 20.00+.50/0
2299

P. Timmerman, D. N. Reinhoudt et al.
FULL PAPER
[D6]DMSO, 258C): d ˆ 165.7, 145.3, 144.3, 48.3, 47.7, 33.1, 20.3, 18.9, 17.5;
5-[ꢀS)-2-Phenylpropyl]-5'-ethylbarbiturate (SBAR): SBAR was obtained
as a white solid (43%) by the same procedure as used for RBAR. The 1H
and 13C NMR and FAB-MS spectra of SBAR were identical to those of
compound RBAR; elemental analysis calcd (%) for C15H18N2O3: C 65.68,
H 6.61, N 10.21; found C 65.52, H 6.30, N 9.66 (Æ0.23).
‡
MS (FAB): m/z: 258.1 ([M ‡H], calcd: 258.1); elemental analysis calcd
(%) for C10H15N3O5: C 46.69, H 5.88, N 16.33; found C 46.82, H 5.91, N
16.07.
ꢀR)-2-Phenyl-1-bromopropane (ꢀR)-4):[51] Triphenylphosphine (12.0 g,
45.9 mmol) was added to a mixture of ꢀR)-2-phenylpropan-1-ol (5.0 g,
36.7 mmol) and CBr4 (15.2 g, 45.9 mmol) in the minimum amount of THF
(25 mL) required to dissolve the reagents. After being stirred at room
temperature for 2 hours, the reaction mixture was poured into water
(200 mL) and extracted with CH2Cl2 (3 Â 50 mL). The combined organic
fractions were dried over MgSO4. Evaporation of the solvent gave ꢀR)-4 as
a crude product, which, after filtration, was purified by column chroma-
tography (SiO2, CH2Cl2/hexane 80:20). Compound ꢀR)-4 was obtained as a
colorless oil (7.10 g, 97%). 1H NMR (300 MHz, [D]chloroform, 258C): d ˆ
7.43 7.27 (m, 5H; ArH), 3.65 (dd, 2J(H,H) ˆ 9.9 Hz, 3J(H,H) ˆ 8.1 Hz, 1H;
CHHBr), 3.54 (dd, 2J(H,H) ˆ 9.9 Hz, 3J(H,H) ˆ 6.3 Hz, 1H; CHHBr),
3.34 3.10 (m, 1H; CH), 1.49 (d, 3J(H,H) ˆ 7.2 Hz, 3H; CH3); 13C NMR
(75 MHz, [D]chloroform, 258C): d ˆ 143.23, 128.12, 126.57, 126.51, 41.74,
General procedure for the formation of assemblies 13 ¥ (CA)6: Two methods
were randomly used for the formation of H-bonded assemblies 13 ¥ (CA)6.
Method 1: After both components 1 and CA had been dissolved in a 1:2
ratio in THF, the solvent was evaporated, and the resulting assembly was
dried under high vacuum.
Method 2: Components 1 and CAwere both suspended, in a 1:2 ratio, in the
desired organic solvent. After 30 minutes× sonication, a clear solution of the
assembly was obtained.
‡
39.49, 19.50; MS (EI): m/z: 200.0 ([M ‡H], calcd: 199.0).
Acknowledgement
ꢀS)-2-Phenyl-1-bromopropane (ꢀS)-4): Compound ꢀS)-4 was obtained as a
white solid (95%) by following the same procedure as used for ꢀR)-4. The
1H and 13C NMR and EI-MS spectra of ꢀS)-4 were identical to those of
compound ꢀR)-4.
The authors would like to acknowledge Dr. R. Fokkens and Prof. N. M. M.
Nibbering for the MALDI-TOF MS measurements. This research was
financially supported by the Council for Chemical Sciences of the
Netherlands Organization for Scientific Research (CW-NWO).
Diethyl 2-[ꢀR)-2-phenylpropyl]-2-ethylmalonate (ꢀR)-5): After trituration
with hexane, NaH (55% dispersion in mineral oil, 0.32 g, 7.2 mmol) was
suspended in THF (100 mL), and diethyl ethylmalonate (2.3 g, 12.2 mmol)
was added dropwise at 08C. After the mixture had been stirred for 1 hr at
room temperature, ꢀR)-4 (1.2 g, 6.0 mmol) was added, and the mixture was
stirred for a further 30 minutes at room temperature and subsequently
heated under reflux for 3 days. After evaporation of the solvent, the residue
was redissolved in CH2Cl2 (100 mL), washed with H2O (2 Â 25 mL) and
brine (25 mL), and dried over Na2SO4. The solvent was evaporated, and
residual diethyl ethylmalonate was removed under high vacuum (p ˆ
0.1 Torr at Tˆ 808C). Compound ꢀR)-5 was obtained as a colorless oil
(0.73 g, 40%) after column chromatography (SiO2, CH2Cl2). 1H NMR
(300 MHz, [D]chloroform, 258C): d ˆ 7.30 7.14 (m, 5H; ArH), 4.13 (dq,
[1] J. D. Watson, F. H. C. Crick, Nature 1953, 171, 737 738.
[2] K. Okuyama, S. Arnott, M. Takanyanagi, M. Kakudo, J. Mol. Biol.
1981, 152, 427 443.
[3] H. M. Warrick, J. A. Spudich, Annu. Rev. Cell. Biol. 1987, 3, 379 422.
[4] G. M. Whitesides, E. E. Simanek, J. P. Mathias, C. T. Seto, D. N. Chin,
M. Mammen, D. M. Gordon, Acc. Chem. Res. 1995, 28, 37 44.
[5] D. Philp, J. F. Stoddart, Angew. Chem. 1996, 108, 1242 1286; Angew.
Chem. Int. Ed. Engl. 1996, 35, 1154 1194.
[6] L. J. Prins, D. N. Reinhoudt, P. Timmerman, Angew. Chem. 2001, 113,
2446 2492; Angew. Chem. Int. Ed. 2001, 40, 2383 2426.
[7] C. Piguet, G. Bernardinelli, G. Hopfgartner, Chem. Rev. 1997, 97,
2005 2062.
[8] F. R. Keene, Chem. Soc. Rev. 1998, 27, 185 193.
[9] J.-M. Lehn, Supramolecular Chemistry: Concepts and Perspectives,
VCH, Weinheim, 1995.
[10] a) J. Rivera, T. MartÌn, J. Rebek Jr., Science 1998, 279, 1021 1023;
b) R. K. Castellano, C. Nuckolls, J. Rebek Jr., J. Am. Chem. Soc. 1999,
121, 11156 11163.
[11] R. K. Castellano, B. H. Kim, J. Rebek Jr., J. Am. Chem. Soc. 1997, 119,
12671 12672.
[12] E. E. Simanek, S. Qiao, I. S. Choi, G. M. Whitesides, J. Org. Chem.
1997, 62, 2619 2621.
[13] A. L. Marlow, E. Mezzina, G. P. Spada, S. Masiero, J. T. Davis, G.
Gottarelli, J. Org. Chem. 1999, 64, 5116 5123.
[14] L. Brunsveld, M. Zhang, M. Glasbeek, J. A. J. M. Vekemans, E. W.
Meijer, J. Am. Chem. Soc. 2000, 122, 6175.
[15] M. R. Ghadiri, J. R. Granja, R. A. Milligan, D. E. McRee, N.
Khazanovich, Nature 1993, 366, 324 327.
[16] N. Kimizuka, T. Kawasaki, K. Hirata, T. Kunitake, J. Am. Chem. Soc.
1995, 117, 6360 6361.
[17] K. C. Russell, J.-M. Lehn, N. Kyritsakas, A. DeCian, J. Fischer, New J.
Chem. 1998, 22, 123 128.
[18] M. M. Green, M. P. Rediy, R. J. Johnson, G. Darling, D. J. O×Leary, G.
Willson, J. Am. Chem. Soc. 1989, 111, 6452 6454.
[19] L. Brunsveld, B. G. G. Lohmeijer, J. A. J. M. Vekemans, E. W. Meijer,
Chem. Commun. 2000, 2305 2306.
[20] L. Brunsveld, A. P. H. J. Schenning, M. A. C. Broeren, H. M. Janssen,
J. A. J. M. Vekemans, E. W. Meijer, Chem. Lett. 2000, 292 293.
[21] L. J. Prins, J. Huskens, F. De Jong, P. Timmerman, D. N. Reinhoudt,
Nature 1999, 398, 498 502.
[22] R. H. Vreekamp, PhD Thesis, University of Twente (The Nether-
lands), 1995, Chapter 6.
2
2J(H,H) ˆ 9.6Hz, 3J(H,H) ˆ 7.2 Hz, 1H; OCHHCH3), 4.12 (dq, J(H,H) ˆ
9.6Hz, 3J(H,H) ˆ 7.2 Hz, 1H; OCHHCH3), 3.87 (dq, 2J(H,H) ˆ 10.5 Hz,
3J(H,H) ˆ 7.2 Hz, 1 H ; OCHHCH3), 3.65 (dq, 2J(H,H) ˆ 10.5 Hz,
3J(H,H) ˆ 7.2 Hz, 1H; OCHHCH3), 2.81 2.69 (m, 1H; C*H), 2.40 (dd,
3
2
2J(H,H) ˆ 14.4 Hz, J(H,H) ˆ 9.0 Hz, 1H; C*CHHC), 2.25 (dd, J(H,H) ˆ
14.4 Hz, 3J(H,H) ˆ 4.8 Hz, 1H; C*CHHC), 1.97 (dq, 2J(H,H) ˆ 14.4 Hz,
3J(H,H) ˆ 7.2 Hz, 1 H ; CCHHCH3), 1.94 (dq, 2J(H,H) ˆ 14.4 Hz, 3J(H,H) ˆ
7.2 Hz, 1H; CCHHCH3), 1.26(d, 3J(H,H) ˆ 7.2 Hz, 3H; C*CH3), 1.23 (t,
3J(H,H) ˆ 7.2 Hz, 3H; OCH2CH3), 1.10 (t, 3J(H,H) ˆ 7.2 Hz, 3H;
OCH2CH3), 0.78 (t, 3J(H,H) ˆ 7.2 Hz, 3H; CCH2CH3); 13C NMR
(75 MHz, [D]chloroform, 258C): d ˆ 171.35, 170.71, 146.23, 127.62, 126.77,
125.56, 60.42, 60.17, 56.85, 38.39, 35.22, 24.57, 24.33, 13.49, 13.25, 7.91; MS
‡
(FAB, glycerine matrix): m/z 307.2 ([M ‡H], calcd: 307.2).
Diethyl 2-[ꢀS)-2-phenylpropyl]-2-ethylmalonate (ꢀS)-5): Compound ꢀS)-5
was obtained as a colorless oil (38%) by the same procedure as used for
ꢀR)-5. The 1H and 13C NMR and FAB-MS spectra of ꢀS)-5 were identical to
those of compound ꢀR)-5.
5-[ꢀR)-2-Phenylpropyl]-5'-ethylbarbiturate (RBAR): A solution of ꢀR)-5
(0.53 g, 1.7 mmol) in EtOH (5 mL) was added dropwise to a solution of Na
(0.17g, 7.3 mmol) in EtOH (20 mL) at room temperature. The mixture was
heated under reflux for 2 days, after which the solvent was evaporated. The
residue was redissolved in water (50 mL), acidified to pH ꢀ 2 3 with 6n
HCl, and extracted with CH2Cl2 (4 Â 25 mL). After being dried over
Na2SO4, the combined organic fractions were evaporated to dryness.
Compound RBAR was obtained as a white solid (0.18 g, 38%) after
column chromatography (SiO2, CH2Cl2/MeOH/NH4OH 90:9.5:0.5).
1H NMR (300 MHz, [D8]THF, 258C): d ˆ 10.40 (s, 1H; NH), 9.93 (s, 1H;
NH), 7.19 7.06(m, 5H; ArH), 2.81 2.96 (m, 1H; C*H), 2.41 (dd,
3
2
2J(H,H) ˆ 13.5 Hz, J(H,H) ˆ 9.6Hz, 1H; C*C HHC), 2.13 (dd, J(H,H) ˆ
3
3
13.5 Hz, J(H,H) ˆ 5.4 Hz, 1H; C*CHHC), 1.85 (q, J(H,H) ˆ 7.2 Hz, 2H;
CH2CH3), 1.16(d, 3J(H,H) ˆ 6.9 Hz, 3H; C*CH3), 0.74 (t, 3J(H,H) ˆ 7.2 Hz,
3H; CH2CH3); 13C NMR (75 MHz, [D8]THF, 258C): d ˆ 172.2, 171.5, 148.5,
144.3, 127.4, 126.9, 125.7, 54.3, 45.1, 36.9, 33.9, 22.3, 7.6; MS (FAB): m/z:
[23] R. H. Vreekamp, J. P. M. Van Duynhoven, M. Hubert, W. Verboom,
D. N. Reinhoudt, Angew. Chem. 1996, 108, 1306 1309; Angew. Chem.
Int. Ed. Engl. 1996, 35, 1215 1218.
‡
275.2 ([M ‡H], calcd: 275.1); elemental analysis calcd (%) for
C15H18N2O3: C 65.68, H 6.61, N 10.21; found C 65.89, H 6.73, N 9.76.
2300
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0810-2300 $ 20.00+.50/0
Chem. Eur. J. 2002, 8, No. 10

Diastereoselective Synthesis of Double-Rosette Assemblies
2288 2301
[24] P. Timmerman, R. H. Vreekamp, R. Hulst, W. Verboom, D. N.
Reinhoudt, K. Rissanen, K. A. Udachin, J. Ripmeester, Chem. Eur.
J. 1997, 3, 1823 1832.
[25] C. T. Seto, G. M. Whitesides, J. Am. Chem. Soc. 1990, 112, 6409 6411.
[26] J.-M. Lehn, M. Mascal, A. DeCian, J. Fischer, J. Chem. Soc. Chem.
Commun. 1990, 479 480.
[27] J.-L. Weidmann, K. A. Jolliffe, L. J. Prins, P. Timmerman, D. N.
Reinhoudt, J. Chem. Soc. Perkin Trans. 2 2000, 10, 2077 2089.
[28] L. J. Prins, K. A. Jolliffe, R. Hulst, P. Timmerman, D. N. Reinhoudt, J.
Am. Chem. Soc. 2000, 122, 3617 3627.
[29] K. A. Jolliffe, M. Crego Calama, R. Fokkens, N. M. M. Nibbering, P.
Timmerman, D. N. Reinhoudt, Angew. Chem. 1998, 110, 1294 1297;
Angew. Chem. Int. Ed. 1998, 37, 1247 1251.
[40] M. Mascal, P. S. Fallon, A. S. Batsanov, B. R. Heywood, S. Champ, M.
Colclough, J. Chem. Soc. Chem. Commun. 1995, 805 806.
[41] L. J. Prins, F. De Jong, P. Timmerman, D. N. Reinhoudt, Nature 2000,
408, 181 184.
[42] In addition to this, it should be mentioned that the small signals
disappear in time; this indeed suggests that they belong to hetero-
meric assemblies incorporating both DEB and RCYA units.
[43] M. Crego Calama, R. Fokkens, N. M. M. Nibbering, P. Timmerman,
D. N. Reinhoudt, Chem. Commun. 1998, 1021 1022.
[44] M. M. Green, N. C. Peterson, T. Sato, A. Teramoto, R. Cook, S. Lifson,
Science 1995, 268, 1860 1866.
[45] E. Yashima, T. Matsushima, Y. Okamoto, J. Am. Chem. Soc. 1997, 119,
6345 6359.
[30] P. Timmerman, K. A. Jolliffe, M. Crego Calama, J.-L. Weidmann, L. J.
Prins, F. Cardullo, B. H. M. Snellink-Ruel, R. Fokkens, N. M. M.
Nibbering, S. Shinkai, D. N. Reinhoudt, Chem. Eur. J. 2000, 6, 4104
4115.
[31] V. Prelog, G. Helmchen, Angew. Chem. 1982, 94, 614 631; Angew.
Chem. Int. Ed. Engl. 1982, 21, 567 594.
[46] T. Gulik-Krzywicki, C. Fouquey, J.-M. Lehn, Proc. Natl. Acad. Sci.
USA 1993, 90, 163 167.
[47] X. Shi, J. C. Fettinger, M. Cai, J. T. Davis, Angew. Chem. 2000, 112,
3254 3257; Angew. Chem. Int. Ed. 2000, 39, 3124 3127.
[48] X. D. Shi, J. C. Fettinger, J. T. Davis, J. Am. Chem. Soc. 2001, 123,
6738 6739.
[32] C. T. Seto, J. P. Mathias, G. M. Whitesides, J. Am. Chem. Soc. 1993,
115, 1321 1329.
[33] N. Kimizuka, T. Kawasaki, K. Hirata, T. Kunitake, J. Am. Chem. Soc.
1998, 120, 4094 4104.
[49] A close examination of the 1H NMR spectrum of assembly 1e3 ¥
(SCYA)6 shows a slightly lower signal-to-noise ratio than seen in the
spectrum of assembly 1e3 ¥ (RCYA)6. Moreover, the calix[4]arene CH2
bridge signals at d ˆ 4.6 4.4 and 3.3 3.1 have unequal intensities.
This might be due to the slightly lower stability of assembly 1e3 ¥
(SCYA)6.
[50] M. L. Vestal, P. Juhasz, S. A. Martin, Rapid Commun. Mass Spectrom.
1995, 9, 1044 1050.
[51] C. J. Hawker, J. M. J. Frechet, J. Am. Chem. Soc. 1990, 112, 7638
7647.
[52] a) H.-A. Klok, K. A. Jolliffe, C. L. Schauer, L. J. Prins, J. P. Spatz, M.
Mˆller, P. Timmerman, D. N. Reinhoudt, J. Am. Chem. Soc. 1999, 121,
7154 7155; b) H. Schˆnherr, V. Paraschiv, S. Zapotoczny, M. Crego-
Calama, P. Timmerman, C. W. Frank, G. J. Vancso, D. N. Reinhoudt,
Proc. Natl. Acad. Sci. USA 2002, in press.
[34] A. Arduini, S. Fanni, G. Manfredi, A. Pochini, R. Ungaro, A.-R.
Sicuri, F. Ugozzoli, J. Org. Chem. 1995, 60, 1448 1453.
[35] For a comprehensive explanation of CD spectroscopy, see: E. L. Eliel,
S. H. Wilen, Stereochemistry of Organic Compounds, Wiley, New
York, 1994.
[36] N. Harada, K. Nakanishi, Acc. Chem. Res. 1972, 5, 257 263.
[37] L. J. Prins, C. Thalacker, F. W¸rthner, P. Timmerman, D. N. Rein-
houdt, Proc. Natl. Acad. Sci. USA, 2001, 98, 10042 10045.
[38] In principle, a C3h-symmetrical assembly with a uniform orientation of
the RBAR units could also give rise to four NHBAR proton signals (see
ref. [28]). However, the formation of this isomer is sterically impos-
sible when RBAR units are used. In addition to this, the signals would
all need to have equal intensities.
[53] T. Ishi-i, M. Crego-Calama, P. Timmerman, D. N. Reinhoudt, Angew.
Chem. 2002, in press.
[39] H. S. Shieh, D. Voet, Acta Crystallogr. Sect. B: Struct. Sci. 1976, 32,
2354 2360 .
Received: October 12, 2001 [F3610]
Chem. Eur. J. 2002, 8, No. 10
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0810-2301 $ 20.00+.50/0
2301