Resolving the magnetic asymmetry of the inner space in self-assembled
dimeric capsules based on tetraurea-calix[4]pyrrole components.
Mónica Espelt,1 Gemma Aragay,1 Pablo Ballester*1,2
1

Institute of Chemical Research of Catalonia (ICIQ), The Barcelona Institute of Science
and Technology, Avgda. Països Catalans 16, 43007 Tarragona, Spain. Fax: (34) 977
920228; Tel: (34) 977 920206;
2

Catalan Institution for Research and Advanced Studies (ICREA), Passeig Lluís
Companys, 23, 08018, Barcelona, Spain

*Author to whom correspondence should be addressed; E-Mail: pballester@iciq.es;
Tel.: +34 977920200 (Ext. 316); Fax: +34 977920221.
Abstract.
The encapsulation of N,N,N’,N’-tetramethyl-1,6-hexanediamine-N,N’-dioxide 2 in a
non-chiral capsular assembly formed by dimerization of tetraurea-calix[4]pyrrole 1a
produced the observation of the N-methyl groups of the encapsulated guest as two
separated singlets resonating highly upfield in the 1H NMR spectrum. In order to clarify
the origin of the observed signal splitting we assembled and studied a series of
structurally related dimeric capsules. We used the tetraurea-calix[4]pyrrole 1a, the
enantiomerically pure tetraurea-calix[4]pyrrole R-1b and the tetraurea-bisloop
calix[4]pyrrole 1c as components of the produced assemblies. The 1H NMR spectra of
the assembled encapsulation complexes with bis-N-oxide 2 evidenced diverse splitting
patterns of the N-methyl groups. In addition, 2D EXSY/ROESY NMR experiments
revealed the existence of chemical exchange processes involving the separated methyl
signals of the encapsulated guest. The capsular assemblies were mainly stabilized by a
belt of eight head-to-tail hydrogen bonded urea groups. The interconversion between the
two senses of rotation of the unidirectionally oriented urea groups was slow on the 1H
NMR timescale. These characteristics determined the appearance of a new asymmetry
element (supramolecular conformational chirality) in the assemblies that accounted for
some of the magnetic asymmetries featured by the capsule’s inner space. The
racemization of the supramolecular chirality element was fast on the EXSY timescale
and produced the chemical exchange processes detected for the encapsulation
complexes.

Keywords: calix[4]pyrrole, magnetic asymmetry, molecular encapsulation,
supramolecular chirality

Introduction
Chiral supramolecular structures play a pivotal role in nature i.e. DNA double helix,
ribozymes, ribosomes etc. and in the development of synthetic molecular aggregates
with novel properties and functions.i,ii Supramolecular chirality emerges from the
assembly of a defined number of molecular components, being identical or not, into
chiral architectures that are stabilized by weak intermolecular interactions.iii,iv
Interestingly, the reversible intermolecular interactions used to maintain together the
assemblies’ molecular components, usually hydrogen bonds or coordination bonds, may
generate by themselves additional structural chiral elements. For example, the efficient
establishment of arrays of intermolecular hydrogen bonds might induce the assembly of
the components in conformations that are inherently chiral,v,vi while the use of metalligand interactions could generate the emergence of metal-based stereogenic centers.vii
In general, syntheticviii and biologicalix,x supramolecular systems displaying
supramolecular chirality rely on the presence of stereogenic centers (chirality centers)
installed in at least one of the assembled molecular units. In the same vein, axial
chirality has also demonstrated to be effective in yielding chiral synthetic assembly
processes. In other words, the diastereoselective synthesis of chiral assemblies demands
the use of chiral components that do not easily racemize. In addition, the point or
center-chirality of the molecular unit/s must be efficiently transferred during the
assembly of the supramolecular aggregate. xi
The dimerization of chiral and achiral molecular objects constitutes paradigmatic
examples to highlight the possible outcomes of self-assembly processes from the
viewpoints of self-sorting and supramolecular chirality (Figure 1 and Figure 2Error!
Reference source not found.).
Figure 1.
To start with, the statistical self-sorting dimerization process of a chiral racemic
compound is expected to produce a mixture of homodimers and heterodimer in a 1:1:2
ratio, respectively (Figure 1). On the one hand, the exclusive or preferential formation
of the homodimers will qualify the assembly process as narcissistic (self-recognition)
and diastereoselective with respect to its self-sorting properties.xii On the other hand, the
quantitative or favored dimerization of the chiral racemic component into the
heterodimer (R,S) also corresponds to a self-sorting diastereoselective process, normally
referred as social self-assembly (self-discrimination).xiii,xiv
From the supramolecular chirality point of view, the homodimers are chiral assemblies
and will be produced as a racemic mixture (R,R and S,S, Figure 1) while the heterodimer
is meso and non-chiral (R,S). The exclusive assembly of one of the enantiomeric
homodimers requires the use of the enantiomerically pure monomer.
Based on the above considerations, the social self-assembly dimerization of a chiral
monomer with a non-chiral counterpart will induce the formation of a chiral
heterodimer as a single enantiomer. However, if two elements of chirality emerge from

the assembly process up to four chiral diastereomers might be expected (Figure 2). Due
to structural constrains, the newly emerged chiral elements may have a fixed
relationship (same or inverse absolute configurations), which in combination with an
efficient transfer of the point-chirality might result in the exclusive formation of one of
the four possible diastereomers. Such assembly process displays exquisite levels of
diastereoselectivity from the viewpoints of self-sorting and supramolecular chirality.
Figure 2.

Chiral assemblies featuring capsular topology and displaying one or multiple sizeable
encapsulated molecular guests are well-known.xv,xvi In particular, the hydrogen-bonding
driven dimerization of chiral and concave-shaped calix[4]arenexvii and
resorcin[4]arenexviii derivatives, as well as their social self-assembly with non-chiral
counterparts produced chiral capsular aggregates that were kinetically and
thermodynamically stable. Unfortunately, the obtained chiral assemblies displayed low
levels of molecular recognition for the encapsulation of chiral substrates.xix This
limitation resulted from the difficulty of transferring the point-chirality of the molecular
components into their spherical or cylindrical chiral inner cavities. Nevertheless, the
magnetic asymmetry present in their chiral inner spaces was easily revealed using NMR
spectroscopy. Owing to the chiral nature of the assemblies, some of the protons of a
non-chiral encapsulated guest became diastereotopic and resonated as separated signals
in the 1H NMR spectra of the corresponding encapsulation complexes. Using chiral
dimeric hydrogen-bonded assemblies derived from resorcin[4]arene scaffolds, Rebek et
al. studied and dissected the steric and magnetic asymmetry effectsxx caused to the
encapsulated guests by the stereogenic carbon atoms located at different distances from
the inner cavity.xxi
The study of the magnetic asymmetry displayed by the inner-spaces of chiral hydrogen
bonded supramolecular assemblies, as well as a deeper understanding of their chiral
elements (structural, stereogenic centers, axial chirality etc.) that are responsible of the
former, constitute topics of current research interest.
Some time ago, we introduced the non-chiral tetraurea- calix[4]pyrrole 1axxii and its
enantiopure chiral version 1b (Figure 4).xxiii We showed that both tetraureas reversibly
dimerized forming a cyclic array of 16 hydrogen-bonds and encapsulating one molecule
of 4,4’-bipyridine N,N’-dioxide 3 (Figure 3). The resulting capsular assemblies
constituted a rare example of molecular container with polar interior able to encapsulate
polar guests in an ordered fashion.
Figure 3.
More recent studies revealed that the encapsulation of a series of alkyl bis-dimethylN,N’-dioxide guests in the meso-capsule 1a1a produced the observation of their methyl
protons as two separated singlets that resonated highly upfield shifted.xxiv We undertook
the present work to clarify the origin of the observed splitting of the N-methyl groups
for the encapsulated ,-alkanediamine N,N,N’N’-dimethyl-N,N’-dioxides. In an effort
to widen the study, we decided to investigate three tetraurea-calix[4]pyrroles 1a-c as

capsular precursors (Figure 4). Tetraurea 1a lacks of stereogenic carbon atoms and its
dimerization produced a meso capsular assembly 1a1a (vide infra).
Figure 4.
Tetraurea R-1b is chiral and possesses four stereogenic benzylic carbon atoms with the
same absolute configuration. Its dimerization yielded the enantiopure chiral capsule R1bR-1b. The dimerization process of equimolar amounts of non-chiral 1a and
enantiopure R-1b took place with high levels of diastereoselective self-sorting and
supramolecular chirality affording exclusively the heterodimeric capsular aggregate
1aR-1b. Finally, equimolar amounts of non-chiral tetraurea 1a and non-chiral bisloop
tetraurea 1c assembled almost exclusively into the heterodimeric capsule 1a1c. All the
studied capsules, displayed an element of supramolecular chirality that derived from the
unidirectional orientation of their urea groups. This unidirectional sense of rotation of
the ureas forced the two halves of the capsule to adopt opposite cyclochiral
conformations (designated as P for the clockwise and M for the counterclockwise sense
of rotation, Figure 3.Error! Reference source not found.). The sense of rotation of the urea
groups was kinetically stable on the 1H NMR chemical shift timescale. Thus, depending
on the chiral relationship that existed between the two halves (enantiomeric or
diastereomeric), their intrinsic chiral nature and the symmetry properties of the
assemblies, different forms and levels of supramolecular chirality emerged and could be
revealed from the careful analysis of their 1H NMR spectra. We selected the N,N,N’,N’tetramethyl-1,6-hexanediamine-N,N’-dioxide 2 as template in all the dimerization
studies (Figure 4). The encapsulated bis-N-oxide 2 folds into a conformation that nicely
complemented both the volume and the functionality of the inner space of all the
studied dimeric capsular assemblies.xxiv The magnetic asymmetry featured by the innerspace of both chiral and non-chiral capsular assemblies caused diverse and
characteristic splitting patterns of the methyl groups of the encapsulated N-oxide 2. The
observed splitting patterns were rationalized on the basis of the symmetry and the
supramolecular chirality properties of the assembled capsules. Furthermore,
EXSY/ROESY experiments performed in the series of encapsulation complexes
evidenced distinctive chemical exchange processes involving the methyl proton signals
of encapsulated 2. These processes derived from the interconversion between the two
senses of rotation of urea groups that was slow on the 1H NMR timescale but fast on the
EXSY/ROESY timescale.
Results and discussion
Magnetic asymmetries in the inner spaces of homodimers: non-chiral homodimeric
assembly 1a1a and enantiopure homodimeric assembly R-1bR-1b.
As commented above, we described the dimerization process of tetraureacalix[4]pyrrole 1a in CDCl3 solution induced by encapsulation of one molecule of N,N’-

dioxide 2 (Figure 4).xxiv The resulting homodimeric capsular assembly 21a•1a (Figure
5) is not chiral, however the methyl protons of the encapsulated N-oxide 2 split into two
singlets that resonate highly upfield shifted. What can be the origin of the magnetic
asymmetry present in the inner cavity of a non-chiral homodimer 1a•1a? The dimeric
assembly 1a•1a is stabilized by the head-to-tail unidirectional orientation of their eight
urea groups creating a seam of 16 hydrogen bonds. This peculiar hydrogen bonding
arrangement forces the four urea groups in each one of the two tetraureacalix[4]pyrrole monomers, 1a, to display a complementary sense of rotation when
observed from a position above their cavities (Figure 3.Error! Reference source not found.).
This is a general property exhibited by all the assembled dimeric capsules studied in this
work. To comply with the above requirement the teraurea- calix[4]pyrrole monomers 1a
assemble in the dimer as conformationally cyclochiral enantiomeric compounds, which
we designate as P and M (Figure 5a). Each hemisphere in the 1a•1a capsule is chiral
and features C4 symmetry, however the overall dimeric assembly 1a•1a, consisting of
two cyclochiral enantiomeric halves, is meso and possesses S8 symmetry. The C4
symmetry of the halves indicated that 2 was experiencing a fast spinning process on the
chemical shift timescale.
The inclusion of the alkyl-dimethyl-N-oxide knobs of 2 in the deep aromatic cavity of
1a was driven by the formation of four hydrogen bonds between the oxygen atom of the
N-oxide and the pyrrole NHs. The inclusion in the aromatic cavity is evidenced by the
large upfield shift experienced by the signals of the encapsulated methyl protons (∆δ 
2.5 ppm). The inclusion process of the alkyl-dimethyl-N-oxide residue in a chiral half of
the 1a•1a dimer induced that the two methyl groups of the same N-oxide knob became
diastereotopic and resonated as separated signals. The other chiral half of the 1a•1a
capsule had an enantiomeric relationship with the former and consequently all their
proton signals must have identical chemical shift values. The observation of separated
signals for the diastereotopic methyl groups of the encapsulated 2 also indicated that the
unidirectional orientation of the eight urea groups in the 1a•1a capsular assembly was
kinetically stable on 1H NMR chemical shift timescale. Otherwise the 1a•1a capsular
assembly and its two hemispheres will display C4v symmetry and they would not exist
in solution as conformationally cycloenantiomers.
Figure 5.
The 1H NMR spectrum of 21a•1a in CDCl3 solution also reflected the chiral nature of
its two hemispheres and the kinetic stability of the unidirectional orientation of the urea
belt through the observation of two sets of separated signals for the ortho and meta
aromatic protons with respect to the urea groups in the meso-phenyl substituent (Ha,
Ha’, Hb and Hb’, Figure 6a). The observed desymmetrization of the meso-aromatic
protons also required that the rotation of the Cmeso-phenyl bond was slow on the 1H
NMR chemical shift timescale. Not surprisingly, the benzylic protons (Hh/ Hh’, Figure
4) of 1a•1a also appeared as diastereotopic signals (Figure 6a).

An EXSY experiment performed on a CDCl3 solution of 21a•1a evidenced the
existence of a chemical exchange process between the two singlets of the diastereotopic
methyl groups for the encapsulated 2 (Figure 6c). Integration of the diagonal and
chemical exchange cross-peaks for the methyl proton signals provided an energy barrier
of 16.0 kcal/mol for the exchange process. The calculated energy barrier coincided with
the one determined for the change in the unidirectional sense of rotation of the urea
groups (16.6 kcal/mol) using the integrals of the diagonal and the cross-peaks of the
diastereotopic aromatic protons ortho to the urea groups in the meso-phenyl
susbtituents.xxv In homodimeric capsular assemblies of non-chiral building blocks (the
same tetraurea- calix[4]pyrrole forms the two hemispheres), the change in the sense of
rotation of the urea groups defines exactly the same molecule and simply interconverts
one hemisphere into its enantiomer (the P hemisphere is converted into M and vice
versa). Most likely, the cycloenantiomerization process of the halves was responsible of
the chemical exchange process observed for the diastereotopic methyl groups. The
diastereotopic methyl groups of bound 2 could also be involved in a chemical exchange
process by the tumbling motion of the guest inside the capsule. However, we considered
that the reduced space available inside the 1a•1a capsule together with the coincidence
in the values of the energy barriers for the cyloenatiomerization process of the two
halves and the exchange of the diastereotopic methyl groups constitute strong evidence
to rule out this latter possibility.
The enatiopure chiral tetraurea- calix[4]pyrrole R-1b differs from the non-chiral
tetraurea 1a in the substitution of one of its hydrogen atoms at its four benzylic
positions by a methyl group. The structure contains four stereogenic carbon centers with
the same absolute configuration (R). The enantiomerically pure chiral tetraurea R-1b
and guest 2 were dissolved in CDCl3 solution in a 2:1 molar ratio. The analysis of the
resulting solution using 1H NMR spectroscopy revealed the diagnostic protons signals
expected for their quantitative assembly into the dimeric capsule 2R-1bR-1b. We
have mentioned above that the senses of rotation of the four urea groups in the two
halves of the capsule must be complementary. Thus, the combination of this element of
asymmetry with the existence of one stereogenic carbon in each urea arm of R-1b
determined a diastereomeric relationship (PR and MR) between the two hemispheres of
the dimer. Each hemisphere was chiral and featured C4 symmetry. The entire
enantiomerically pure dimeric assembly 2PR-1bMR-1b also showed C4 symmetry.
Because the two hemispheres have a diastereomeric relationship they are distinct:
chemically non-equivalent. Consequently, their proton signals, as well as those of the
included N-methyl groups of the guest might have different chemicals shifts. The
reduction in symmetry described for the capsule 2PR-1bMR-1b compared to the
previous homodimeric 1a•1a capsule, was reflected in a larger magnetic asymmetry of
its inner space and the concomitant increase in the number of proton signals of the
encapsulated guest. We observed two sets of two diastereotopic singlets for the methyl
groups of the encapsulated 2 in the 1H NMR spectrum of 2PR-1bMR-1b. Each Noxide knob is included in a distinct chiral hemisphere and yielded one set of two
diastereotopic singlets. The existence of chemically non-equivalent hemispheres in the

2PR-1bMR-1b assembly was also substantiated by the observation of two separated
proton signals for the pyrrole NHs (each integrating by four protons). We commented
above that the combination of the same tetraurea monomer in 21a1a defines a
cycloenantiomerization process of the halves (M became P and viceversa) through the
change in the unidirectional orientation of the urea groups and the concomitant
existence of a chemical exchange process between diastereotopic methyl groups of
bound 2. In striking contrast, the analogous change of orientation of the urea groups in
the homocapsule 2PR-1bMR-1b (having chemically non-equivalent hemispheres)
interconverted the two diastereomeric halves (MR into PR and vice versa) but without
inducing their enantiomerization. Accordingly, the EXSY chemical exchange pattern
for the methyl signals of 2 encapsulated in PR-1bMR-1b showed exchange cross-peaks
only between pairs of methyl groups included in distinct hemispheres (Figure 6).

Figure 6.
Magnetic asymmetries in the inner space of chiral heterodimeric capsular assemblies:
1a R-1b and 1a1c
We previously reported that the assembly of 1a with R-1b induced by encapsulating one
molecule of 4,4’-dipyridyl N,N’-dioxide 3 yielded exclusively the heterocapsule
aggregate 3P-1a•MR-1b.xxiii We demonstrated that upon heterodimerization, the
counterclockwise (M) sense of rotation of the urea groups in tetraurea- calix[4]pyrrole
R-1b was energetically favored. This result derived from an efficient transfer of
chirality of the stereogenic center to the cyclochiral conformer. Consequently, the
assembly of tetraurea 1a in a dimeric capsule with R-1b forced the former to adopt the P
conformational cyclochirality.xxiii Theoretical calculations indicated that the
diastereomeric heterocapsule 3M-1aPR-1b was significantly higher in energy
compared to 3P-1aMR-1b. We concluded that the heterodimeric chiral assembly
3P-1aMR-1b did not experience a change in the unidirectional orientation of their
urea groups because the resulting dimer 3 M-1aPR-1b was thermodynamically not
stable.
An equimolar mixture of the non-chiral tetraurea-calix[4]pyrrole 1a and the
entantiomerically pure tetraurea-calix[4]pyrrole R-1b could produce a maximum of four
different capsular assemblies encapsulating one molecule of 2: two homodimeric
caspules (21a•1a and 2R-1b•R-1b) and two heterodimeric assemblies (2(M-1a
PR-1b) and 2(P-1aMR-1b) that are chiral diastereoisomers. Experimentally, the 1H
NMR spectrum of the mixture showed the diagnostic proton signals for the exclusive
assembly of only one of the two heterodimers (Figure 8a). By analogy to the results
commented above, we assigned the absolute configuration 2P-1aMR-1b to the
exclusively produced heterodimer in the present work (Figure 7).
Figure 7.

The 1H NMR spectrum of a CDCl3 solution containing the 2P-1aMR-1b heterodimer
showed that the methyl protons of the encapsulated guest 2 appeared as two sets of two
separated singlets (Figure 8a). This splitting pattern was in agreement with the existence
of two distinct chiral hemispheres. Each hemisphere included one dimethyl-N-oxide
terminal group of the encapsulated guest 2. The methyl groups of the two chemically
non-equivalent terminal N-oxide groups of bound 2 also became diastereotopic due to
the chiral nature of the hemispheres and resonated as two separated signals. A
transverse T-ROESYxxvi experiment performed on the above solution showed the total
absence of cross-peaks between the methyl proton signals of the encapsulated N-oxide
2. This fact evidenced the lack of chemical exchange processes involving the methyl
groups and supported the existence of a unique and preferred sense of orientation of the
urea groups in the heterodimeric chiral assembly 2P-1aMR-1b (Figure 8c).
Figure 8.

The non-chiral tetraurea- calix[4]pyrrole 1a (1 equiv.) and the non-chiral bisloop
calix[4]pyrrole derivative 1c (1.2 equiv) assembled almost quantitatively in CDCl3
solution in the presence of 2 (1equiv). The mixture yielded the heterodimeric capsule
21a1c as the major assembly. A careful analysis of the resulting solution using 1H
NMR spectroscopy revealed the presence of traces of the homodimer 21a1a (Figure
8b). None of the components of the 21a1c assembly are chiral, however the required
complementary between the senses of rotation of the urea groups of the two
hemispheres forced the two halves to adopt cyclochiral conformations. Again, the
unidirectional orientation of the urea groups in the two capsules was kinetically stable
on the 1H NMR timescale. This produced the existence of the 21a1c capsule in
solution as a mixture of two enantiomers, 2M-1aP-1c and 2P-1aM-1c, which
interconverted by changing the unidirectional sense of rotation of their urea groups
(Figure 9). In each enantiomeric capsule the two halves are chemically non-equivalent
and conformationally chiral. In short, the methyl groups of encapsulated 2 are expected
to resonate as four separated singlets.
In complete agreement with these symmetry considerations, the 1H NMR spectrum of
the racemic heterocapsule 21a1c in CDCl3 solution showed two separate proton
signals for the pyrrole NHs indicative of two distinct hemispheres (Figure 8b).
However, only three different signals were detected for the methyl protons of
encapsulated 2. Two singlets integrated for 3 protons and one for 6 protons. Most likely,
two of the singlets of the four methyl groups in bound 2 resonate at the same chemical
shift. A 2D EXSY experiment revealed the existence of chemical exchange cross-peaks
between signals that were assigned to the diastereotopic methyl groups of encapsulated
2 (Figure 8d). The energy barrier calculated for the exchange process was in agreement
with the energy required for the inversion of the unidirectional orientation of the urea
groups (17.3 kcal/mol). Taken together, these results suggested that the chemical
exchange between diastereotopic methyl protons was caused by the inversion of the

sense of rotation of the capsule’s urea belt. Possibly and as already commented above,
the tumbling motion of encapsulated 2 is not allowed owing to geometric/size
constrains of the container. This hypothesis was substantiated experimentally by the
lack of exchange cross-peaks between the methyl groups encapsulated in distinct
hemispheres.
Figure 9.
Conclusions
We have assembled and studied a series of dimeric molecular capsules based on
tetraurea-calix[4]pyrrole derivatives. All capsules encapsulated one molecule of the bisN-oxide 2. We observed capsule’s dependent splitting pattern for the N-methyl proton
signals of encapsulated 2 (from 2 to 4 separate singlets) that reflected the differences in
magnetic asymmetry of their interiors. We rationalized the observed splitting patterns
based on the symmetry and the supramolecular chirality properties of the capsular
assemblies, as well as of their components. The explanations of the observed number of
signals for the methyl groups of bound 2 are also supported by the results obtained in
the corresponding 2D EXSY/ROESY NMR experiments. The methyl groups of
encapsulated 2 were involved in chemical exchange processes caused by the fast
inversion, on the EXSY/ROESY timescale, of the sense of rotation of the belt of
hydrogen bonded urea groups that stabilized the capsules. In contrast, the unidirectional
orientation of the urea groups was kinetically stable on the chemical shift timescale and
was responsible of the existence in solution of the capsule’s hemispheres as
complementary cyclochiral conformers. The inclusion of the dimethyl-N-oxide knobs of
2 in the chiral components of the capsules rendered the methyl groups of the former
diastereotopic. The 1H NMR spectra of capsules featuring enantiomeric hemispheres
(1a•1a) displayed one set of two singlets for the methyl protons of encapsulated 2. On
the other hand, the 1H NMR spectra of capsules with diastereomeric (MR-1b•PR-1b) or
chemically non-equivalent hemispheres (1aR-1b and 1a1c) showed two sets of two
singlets for the methyl groups of encapsulated 2. The obtained results ruled out the
existence of a tumbling motion of encapsulated 2. Conversely, the C4 symmetry
displayed by the hemispheres of the capsules suggested a fast spinning process on the
chemical shift timescale of encapsulated 2.
Supplementary Materials: Experimental details, synthesis and characterization of
calyx[4]pyrrole 1c and details on the calculations of the energy barriers of the chemical
exchange processes detected by 2D EXSY experiments.

Acknowledgements
The authors thank Gobierno de España MINECO (projects CTQ2014-56295-R,
CTQ2014-52974-REDC and Severo Ochoa Excellence Accreditation 2014-2018 SEV2013-0319), FEDER funds (project CTQ2014-56295-R) and the ICIQ Foundation for
funding.

Figure 1. Equilibria established in the reversible dimerization process of a chiral racemic monomer into
dimeric aggregates. The putative appearance of additional elements of asymmetry in the dimers has not
been considered for simplicity.

Figure 2. Exclusive assembly of four chiral heterodimers (social-self sorting) in the dimerization of a

chiral monomer with a non-chiral counterpart. To account for the formation of four chiral diastereomers
we considered the appearance of two elements of asymmetry (M/P) during the dimer assembly. The
plausible existence of a structurally favored relationship between the two chiral elements (identical or
inverse) reduces the number of possible diastereomers to just two (black rectangle) or even to only one if
the chirality point of the chiral monomer is effectively transferred to the supramolecular chirality of the
dimer (red ellipsoid). The black rectangle includes the two diastereomers with complementary asymmetry
elements. The detection of these two diastereomers using 1H NMR spectroscopy requires that their
interconversion caused by the inversion of the chiral elements resulting from their assembly
(racemization) is slow on the chemical shift timescale.

Figure 3. Left) Energy minimized MM3 structure of the 31a1a encapsulation complex. Right) The
two calix[4]pyrrole units of the assembly (hemispheres) are shown separately to highlight the opposite
and complementary sense of rotation of their urea groups. The two hemispheres are conformationally and
inherently chiral cycloenantiomers.

Figure 4. Structures of the tetraurea-calix[4]pyrroles and bis-N-oxide guests used in this work and in
previous studies described in literature.

Figure 5 Energy minimized MM3 structures of the homodimer 21a1a (a) and the homodimer 2R1bR-1b (b). The corresponding schematic representations of the complexes are also depicted with the
assignments of the sense of rotation of the urea groups in each hemisphere. Non-polar hydrogen atoms of
the host are omitted for clarity.

Figure 6. Selected regions of the 1H NMR spectra in CDCl3 of: a) 21a•1a homocapsule and b) 2R1b•1b homocapsule. Selected upfield region of the EXSY experiments (mixing time 300 ms) of: c)
21a•1a homocapsule and d) 2R-1b•1b homocapsule. Primed letters indicate diastereotopic protons, 1
and 2 superscripts indicate methyl groups in different hemispheres.

Figure 7. a) Energy minimized MM3 structure of heterodimer 2P-1aMR-1b and its corresponding
schematic representation with indication of the sense of rotation of the urea groups in each hemisphere.
Non-polar hydrogen atoms of the host are omitted for clarity

Figure 8. Selected regions of the 1H NMR spectra in CDCl3 of: a) 21a1b heterocapsule and b)
21a1c heterocapsule. Selected upfield region of: c) the transverse T-ROESY experiments of 21a1b
heterocapsule and d) the EXSY experiment of 21a1c heterocapsule, the small doublet centered at 0.5
ppm corresponds to the small quantity of 21a1a. The spectrum in panel c was acquired using the
transverse T-ROESY pulse program in order to remove artifact cross-peaks (HOHAHA). Primed letters
indicate diastereotopic protons, 1 and 2 superscripts indicate different hemispheres.

Figure 9. Equilibrium between the two enantiomeric capsules 2M-1aP-1c and 2P-1aM-1c. The red
arrows indicate the unidirectional sense of rotation of the belt of hydrogen bonded ureas assigned by

looking from the top of the bisloop component. The inversion of the unidirectional sense of rotation of the
urea groups interconverts the two enantiomers.

References
i

L. Zhang, L. Qin, X. Wang, H. Cao, M. Liu Adv. Mater. (Weinheim, Ger.) 2014, 26,
6959.
ii
J. Crassous Chem. Soc. Rev. 2009, 38, 830.
iii
A. Scarso, J. Rebek In Supramolecular Chirality; Crego-Calama, M., Reinhoudt, D.
N., Eds.; Springer-Verlag: Berlin Heidelberg, 2006; Vol. 265, p 312.
iv
M. A. Mateos-Timoneda, M. Crego-Calama, D. N. Reinhoudt Chem. Soc. Rev. 2004,
33, 363.
v
A. Szumna Chem. Soc. Rev. 2010, 39, 4274.
vi
M. Grajda, M. Wierzbicki, P. Cmoch, A. Szumna J. Org. Chem. 2013, 78, 11597.
vii
A. V. Davis, D. Fiedler, M. Ziegler, A. Terpin, K. N. Raymond J. Am. Chem. Soc.
2007, 129, 15354.
viii
L. J. Prins, J. Huskens, F. de Jong, P. Timmerman, D. N. Reinhoudt Nature 1999,
398, 498.
ix
C. X. Lin, Y. G. Ke, Z. Li, J. H. Wang, Y. Liu, H. Yan Nano Lett. 2009, 9, 433.
x
T. Imura, N. Ohta, K. Inoue, N. Yagi, H. Negishi, H. Yanagishita, D. Kitamoto ChemEur J 2006, 12, 2434.
xi
Y. Tsunoda, K. Fukuta, T. Imamura, R. Sekiya, T. Furuyama, N. Kobayashi, T. Haino
Angew. Chem., Int. Ed. Engl. 2014, 53, 7243.
xii
P. L. Arnold, J. C. Buffet, R. Blaudeck, S. Sujecki, C. Wilson Chem-Eur J 2009, 15,
8241.
xiii
C. G. Claessens, T. Torres J. Am. Chem. Soc. 2002, 124, 14522.
xiv
M. Yoshizawa, M. Tamura, M. Fujita Angew Chem Int Edit 2007, 46, 3874.
xv
M. Yoshizawa, M. Fujita Pure Appl. Chem. 2005, 77, 1107.
xvi
M. Yoshizawa, J. K. Klosterman, M. Fujita Angew Chem Int Edit 2009, 48, 3418.
xvii
R. K. Castellano, B. H. Kim, J. Rebek J. Am. Chem. Soc. 1997, 119, 12671.
xviii
B. Kuberski, A. Szumna Chem. Commun. (Cambridge, U. K.) 2009, 1959.
xix
A. Szumna Chem-Eur J 2009, 15, 12381.
xx
T. Amaya, J. Rebek J. Am. Chem. Soc. 2004, 126, 6216.
xxi
M. P. Schramm, P. Restorp, F. Zelder, J. Rebek J. Am. Chem. Soc. 2008, 130, 2450.
xxii
P. Ballester, G. Gil-Ramirez Proc. Natl. Acad. Sci. U. S. A. 2009, 106, 10455.
xxiii
M. Chas, G. Gil-Ramirez, E. C. Escudero-Adan, J. Benet-Buchholz, P. Ballester
Org. Lett. 2010, 12, 1740.
xxiv
M. Espelt, P. Ballester Org. Lett. 2012, 14, 5708.
xxv
G. Gil-Ramirez, M. Chas, P. Ballester J. Am. Chem. Soc. 2010, 132, 2520.
xxvi
A. Bax, D. G. Davis J. Magn. Reson. 1985, 63, 207.