Guest-induced capsule formation based on concerted interactions in water at neutral pH

Article abstract of DOI:10.1039/c0cc02394a

The interaction of a tetracationic calixarene (TAC4) with dianionic guest (BS) in water was studied with NMR, nano-ITC and ESI-MS experiments; the convergent evidence provided by these different techniques shows that the addition of a suitable guest molecule triggers the formation of a homodimeric capsule in aqueous neutral solution.

Full text of DOI:10.1039/c0cc02394a

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Guest-induced capsule formation based on concerted interactions
in water at neutral pHw
Carmela Bonaccorso, Antonio Ciadamidaro, Carmelo Sgarlata,z Domenico Sciotto* and
Giuseppe Arena*
Received 6th July 2010, Accepted 12th August 2010
DOI: 10.1039/c0cc02394a
The interaction of a tetracationic calixarene (TAC4) with
dianionic guest (BS) in water was studied with NMR,
nano-ITC and ESI-MS experiments; the convergent evidence
provided by these different techniques shows that the addition of
a suitable guest molecule triggers the formation of a homodimeric
capsule in aqueous neutral solution.
We have previously investigated the inclusion properties of
the tetracationic calix[4]arene (TAC4) depicted in Scheme 1
and have shown its power to promote the inclusion of organic
anions in water in neutral solution.6 If anionic guests can be
included in the cavity of TAC4, can a gemini-like guest trigger
the formation of a capsule? To answer this question we
designed and synthesized a dianionic guest (BS) (Scheme 1)7
and studied its interaction with TAC4 by 1H NMR titrations,
DOSY, ROESY, nano-ITC and ESI-MS measurements, to
show that they may form a capsule in neutral aqueous
solution, that would rely on concerted hydrophobic and
electrostatic interactions.
In 1H NMR titrations8 the signals for the guest shift upfield
from where they occur in free solution as the host concentration
increases: the complexation induced shifts (CIS) follow the
order Me > H2 > H1 > H3 > Ha > Hb > Hg (Fig. S3,
ESIw). This shows that the guest is included via its aromatic
moiety, most likely to maximize the burial of lipophilic
surfaces from water, with CH3 groups pointing deeply into
the cavity. The large CIS (up to 2 ppm for the methyl group)
reveals that the aromatic moieties of BS are included
and experience the magnetic shielding provided by the four
aromatic rings that shape the cavity. Guest signals are detected
as single resonance due to the fast exchange between the free
and the complexed species on the NMR timescale. Complexation
of BS could not be described using a 1 : 1 or a 2 : 1 host–guest
binding model; a good fit of the data was instead obtained
with the 1 : 1 and 2 : 1 association model (Fig. 1 and Fig. S4,
ESIw), and this yields log K values of 4.34 ꢀ 0.01 and
3.06 ꢀ 0.01, respectively (Table 1). The overall binding
constant for the formation of the capsule (log b21 = 7.4)
One of the vital current tasks in supramolecular chemistry is to
develop simple yet comprehensive and reliable methods to
effectively hold together separate parts of complex architectures,
exclusively through cooperative non-covalent interactions.
Only in the past couple of decades have researchers learned
how to program well-designed building blocks which self-
assemble into defined aggregates.1 In this context, assembling
molecular structures that are able to encapsulate guest
molecules in solution has turned out to be a very interesting
yet challenging area that has attracted the interest of many
research groups.2 Gibb and Simin have elegantly shown that
these structures can be used to perform reactions that are
normally difficult to carry out in aqueous solution.2
Hence, water solubility is a key issue for these structures.
However, working in water implies disadvantages not involved
when dealing with organic solvents; these include solubility,
the use of a buffer and the drop of the binding constant values.
Water soluble non-covalent assemblies have been obtained
mainly by metal–ligand interactions.3 For example, in order to
explain the rapid capture and crystallisation of (di)cationic
cyclic (aza)crown ethers in molecular capsules based on two
p-sulfonatocalix[4]arenes, Raston et al. carried out a diffusion
study in water indicating that more than one species seems to
be present in solution, at least one of which displays diffusion
behaviour that would be consistent with a 2 : 1 host-guest
complex.3 Only few papers, though, deal with structures that
are not based on metal–ligand interactions; however the
systems reported are not entirely soluble in water and need
mixtures of water/organic solvent.2,4 To the best of our
knowledge there are only two examples of self-assembling
capsules that are not based on metal–ligand interactions and
are soluble in water;2,5 both these assemblies are not soluble in
neutral region and reach full solubility in water at pH 9.
Dipartimento di Scienze Chimiche, Universita` degli Studi di Catania,
Viale Andrea Doria 6, 95125 Catania, Italy. E-mail: garena@unict.it,
dsciotto@unict.it; Fax: +39 (0)95 337678
w Electronic supplementary information (ESI) available: Experimental
details. See DOI: 10.1039/c0cc02394a
z Current address: Lawrence Berkeley National Laboratory,
1 Cyclotron Road Berkeley, CA 94720, USA.
Scheme 1 Host and guest investigated.
c
This journal is The Royal Society of Chemistry 2010
Chem. Commun., 2010, 46, 7139–7141 7139

View Article Online
Fig. 1 Schematic representation of the assemblies formed.
Table 1 Log K values and thermodynamic parameters for the host–guest complexes
ITC
NMR
log Ka/dm3 molꢂ1
log Ka/dm3 molꢂ1
DGa/kJ molꢂ1
DHa/kJ molꢂ1
DSa/J Kꢂ1 molꢂ1
Equilibrium
H + G # HG
HG + H # H2G
4.34 (1)
3.06 (1)
3.6 (2)
2.9 (3)
ꢂ20.5 (1)
ꢂ16.5 (2)
ꢂ11.9 (1)
ꢂ9.1 (2)
30 (4)
25 (5)
a
s in parentheses.
nicely compares with the value reported for a deep cavitand-
based capsule.2
The ESI-MS analysis of an aqueous solution of TAC4/BS
(4/1) showed signals at 1861.8 and 1111.2 corresponding to
[2(TAC4)ꢁBSꢁH2PO4ꢁ2H]+ and [TAC4ꢁBSꢁH]+, respectively,
confirming the co-existence of a 2 : 1 (capsule) and a 1 : 1
host–guest complex (Fig. S5, ESIw).
This model was checked by nano-ITC. A satisfactory fit of
the nano-ITC data to the 1 : 1 and a 2 : 1 host–guest association
model was again obtained, yielding log K values equal to
3.6 ꢀ 0.3 and 2.9 ꢀ 0.3 for the two species, respectively, which
1
is consistent with the H NMR experiments.6,9
Fig. 2 Thermograms obtained in the absence (red) and in the
The enthalpic and entropic values show that both steps are
favored by enthalpy and entropy. This indicates that the
entropic penalties associated with organizing the guest and
the host for complex formation have been paid for by the
desolvation of the guest and the host. The burying of the
aromatic moiety from water, which is also facilitated by
the ionic interactions upon self-assembly of the complexes,
contributes to the free energy of formation; the enthalpically
favorable contribution overrides the cost in energy needed
for the desolvation of all the interacting particles prior to the
self-assembly process.
presence (blue) of NaNO3.
To prove that the ionic interaction between the host(s) and
the guest is crucial for the formation of the supramolecular
assemblies we run the same experiments in 0.1 mol dmꢂ3
NaNO3. The introduction of high salt concentrations determined
a dramatic drop in the heat developed and the binding
constant (Fig. 2); in any case the model obtained in the
absence of sodium nitrate both via NMR and ITC could no
longer be reproduced. This clearly shows that the driving
forces for the complex formation include both electrostatic
as well as CH–p and p–p interactions.10
The formation of a ‘‘capsule’’ in solution was further
confirmed using diffusion NMR spectroscopy (DOSY). This
method has proved to be useful for the detection of aggregates
which were difficult to characterize due to their high symmetry
and non-covalent character.11
Fig. 3 BS self-diffusion coefficients. Red full circles, experimental
data points; broken line, HG model; solid line, HG + H2G model.
than in its free form, which is consistent with an increase of its
effective size due to complex formation.12
Two models have been used to fit the experimental data
points, i.e. a model that takes into account the existence of HG
only and a second one that considers the formation of
HG + H2G (see ESIw, DOSY experiments). Noteworthy,
the mole fraction values needed to generate the theoretical
curves were computed by using the log K values obtained via
ITC (Fig. S7, ESIw).13 The use of data from NMR and ITC,
that probe different physical observables, provides a way to
cross-check the results. Fig. 3 indisputably shows that whilst
the first model (HG only) fails to reproduce the experimental
The increase of TAC4 concentration markedly affects the
motion of BS, while it has a smaller effect on the TAC4
diffusion coefficients (Fig. 3); the guest diffuses more slowly
c
7140 Chem. Commun., 2010, 46, 7139–7141
This journal is The Royal Society of Chemistry 2010

View Article Online
their spatial proximity; the signals of the Hb groups overlap
with those of the a methylene groups and cannot be used as
reporters for the interaction. Noteworthy, all these spatial
correlations are no longer observed when HG is the main
species (H/G o 4) (Fig. S10, ESIw). Taken together, these
observations prove unambiguously the existence of a self-
assembled homodimeric capsule.
This work shows that a suitable choice of the anchoring
groups of the upper rim of the hydrophobic cavity can be a
powerful approach to high definition self-assembly in water,
provided that the upper rim substituents match the electro-
static and steric requirements of the guest. NMR, MS and
ITC, that probe different physical observables, cross-validate
one another. We have demonstrated that (i) a simple host like
TAC4 may form a homodimeric capsule; (ii) the capsule forms
in neutral aqueous solution; (iii) significant amounts of the
capsule (50%) form even with relatively small (H/G) ratios.
We are grateful to MIUR and University of Catania for
funding. Dr A. Giuffrida, G. Grasso, V. Zito (CNR-IBB) and
D. Garozzo (CNR-ICTP) are acknowledged for performing
and discussing the ESI-MS experiments; we are grateful to Dr
Peter Gans (www.hyperquad.co.uk) for helping with and
discussing the diffusion data treatment.
Notes and references
1 D. Philp and J. F. Stoddart, Angew. Chem., Int. Ed. Engl., 1996, 35,
1154.
2 C. Sgarlata, J. S. Mugridge, M. D. Pluth, B. E. F. Tiedemann,
V. Zito, G. Arena and K. N. Raymond, J. Am. Chem. Soc., 2010,
132, 1005; L. Simin and B. C. Gibb, Chem. Commun., 2008, 3709;
S. M. Biros and J. Rebek, Jr., Chem. Soc. Rev., 2007, 36, 93 and
references therein; M. D. Pluth, R. G. Bergman and
K. N. Raymond, Science, 2007, 316, 85; S. J. Dalgarno,
J. L. Atwood and C. L. Raston, Chem. Commun., 2006, 4567.
3 P. Jin, S. J. Dalgarno and J. L. Atwood, Coord. Chem. Rev., 2010,
254, 1760; M. D. Pluth, R. G. Bergman and K. N. Raymond, Acc.
Chem. Res., 2009, 42, 1650 and references therein; M. Yoshizawa,
M. Tamura and M. Fujita, Science, 2006, 312, 251 and references
therein; S. J. Dalgarno, J. Fisher and C. L. Raston, Chem.–Eur. J.,
2006, 12, 2772 and references therein; T. Inomata and K. Konishi,
Chem. Commun., 2003, 1282; H. J. Kim, D. Moon, M. S. Lah and
J. I. Hong, Angew. Chem., Int. Ed., 2002, 41, 3174; S. Russel-Seidel
and P. J. Stang, Acc. Chem. Res., 2002, 35, 972; O. D. Fox,
N. K. Dalley and R. G. Harrison, J. Am. Chem. Soc., 1998, 120, 7111.
4 A. Shivanyuk and J. Rebek, Jr., Chem. Commun., 2001, 2424;
J. L. Atwood, L. J. Barbour and A. Jerga, Chem. Commun., 2001,
Fig. 4 Top: 2D ROESY (phosphate buffer, D2O, pD 6.8, TAC4/BS = 6);
off-diagonal peaks of interest are encircled. Bottom: optimized model
(see ESIw); arrows indicate the spatially correlated groups.
data, the second model (HG + H2G) nicely fits the experi-
mental self-diffusion coefficients, thus pointing out that the
two species coexist in neutral solution. The self-diffusion
coefficient of BS, ‘‘extrapolated’’ to larger H/G ratio
(2.4 ꢃ 10ꢂ10 m2 sꢂ1), is significantly smaller than that of free
TAC4 (3.1 ꢃ 10ꢂ10 m2 sꢂ1) and free BS (3.8 ꢃ 10ꢂ10 m2 sꢂ1),
and confirms the ‘‘weighed average’’ larger size of the species
in solution; the ‘‘extrapolated’’ value nicely matches the
2376; M. O. Vysotsky, I. Thondorf and V. Bohmer, Chem.
Commun., 2001, 1890.
¨
5 F. Corbellini, L. D. Costanzo, M. Crego-Calama, S. Geremia and
D. N. Reinhoudt, J. Am. Chem. Soc., 2003, 125, 9946.
6 C. Sgarlata, C. Bonaccorso, F. G. Gulino, V. Zito, G. Arena and
D. Sciotto, New J. Chem., 2009, 33, 991.
theoretically predicted value (2.4 ꢃ 10ꢂ10 m2
s
ꢂ1) for the
guest-containing homodimeric capsule.12
2D ROESY experiments were performed to map out the
capsule geometry in details. The cross peaks corresponding to
intermolecular NOE contacts confirmed the spatial proximity
of the different components of the complex. The results are
consistent with the observed changes in the chemical shift and
in the diffusion coefficients, confirming that the TAC4 units
are wrapped around the guest.
7 Details of the synthesis can be found in ESIw.
8 L. Fielding, Tetrahedron, 2000, 56, 6151.
9 M. V. Rekharsky and Y. Inoue, J. Am. Chem. Soc., 2002, 124, 12361.
10 G. Arena, A. Casnati, A. Contino, G. G. Lombardo, D. Sciotto
and R. Ungaro, Chem.–Eur. J., 1999, 5, 738; E. A. Meyer,
R. K. Castellano and F. Diederich, Angew. Chem., Int. Ed.,
2003, 42, 1210; J. C. Ma and D. A. Dougherty, Chem. Rev.,
1997, 97, 1303; M. Nishio, Tetrahedron, 2005, 61, 6923.
11 T. Evan-Salen and Y. Cohen, Chem.–Eur. J., 2007, 13, 7659;
Y. Cohen, L. Avram and L. Frish, Angew. Chem., Int. Ed., 2005,
44, 520.
12 Y. Cohen, L. Avram, T. Evan-Salem and L. Frish, in Analytical
Methods in Supramolecular Chemistry, ed. C. Schalley, Wiley-VCH
Verlag, Weinheim, 2007, pp. 163–216.
13 The use of the values obtained via NMR does not change the
picture.
The H1 and H2 protons correlate with all the calix[4]arene
signals, while the H3 proton interacts with the upper rim
groups (Fig. 4). The Me protons show a NOE effect both with
the calix[4]arene cavity (Hc) and the ammonium methyl
groups (Ha). The off-diagonal peaks between the b and g
methylene protons and the ammonium methyl groups confirm
c
This journal is The Royal Society of Chemistry 2010
Chem. Commun., 2010, 46, 7139–7141 7141