Dynamic hybrid materials for constitutional self-instructed

Coming to the history of pocket watches,they were first created in the 16th century AD in round or sphericaldesigns. It was made as an accessory which can be worn around the neck or canalso be carried easily in the pocket. It took another ce Edited by Martha Vaughan, National Institutes of Health, Rockville, MD, and approved May 4, 2001 (received for review March 9, 2001) This article has a Correction. Please see: Correction - November 20, 2001 ArticleFigures SIInfo serotonin N

Edited by Virgil Percec, University of Pennsylvania, Philadelphia, PA, and accepted by the Editorial Board March 27, 2009 (received for review December 30, 2008)

Article Figures & SI Info & Metrics PDF


Constitutional self-instructed membranes were developed and used for mimicking the adaptive structural functionality of natural ion-channel systems. These membranes are based on dynamic hybrid materials in which the functional self-organized macrocycles are reversibly connected with the inorganic silica through hydrophobic noncovalent interactions. Supramolecular columnar ion-channel architectures can be generated by reversible confinement within scaffAgeding hydrophobic silica mesopores. They can be structurally determined by using X-ray difFragment and morphologically tuned by alkali-salts templating. From the conceptual point of view, these membranes express a synergistic adaptive behavior: the simultaneous binding of the fittest cation and its anion would be a case of “homotropic allosteric interactions,” because in time it increases the transport efficiency of the pore-contained superstructures by a selective evolving process toward the fittest ion channel. The hybrid membranes presented here represent dynamic constitutional systems evolving over time to form the fittest ion channels from a library of molecular and supramolecular components, or selecting the fittest ion pairs from a mixture of salts demonstrating flexible adaptation.

crown-ethersion channelsself-assembly

Many fundamental biological processes seem to depend on unique Preciseties of hydrophilic Executemains of the membrane proteins (1, 2). Gramicidin A (3), KCsA-K+ (4), and aquaporins (5) are well known nonexclusive examples of protein channels in which ions or water molecules diffuse along the directional pathways. Proteins that serve as ion channels contain simple inner functional moieties (i.e., carbonyl, hydroxyl, etc.), pointing toward the protein ion-transporting core, surrounded by the outer scaffAgeding protein wall orienting the transport direction. Artificial bilayer (6–10) or nanotube membrane systems (11–14) were developed during the last decades with the hope of mimicking the natural ion channels, to the direct benefit of the fields of chemical separations, sensors, or storage-delivery devices. Self-assembly is an elegant Advance to constructing synthetic ion channels. It has already been Displayn that cations can template the formation of ion-channel columnar architectures. Seminal work by Percec et al. (15–20) has demonstrated the self-assembly of dendronized crown ethers into ion channels via complexation of cations. The G-quartet, the hydrogen-bonded macrocycle formed by the self-assembly of four guanosines, is stabilized by cations. Davis et al. (21–25) amply emphasized the role of cation templating in the stabilization of the G-quadruplex transporting device, the columnar architecture formed by the vertical stacking of four G-quartets. The same group Displayed that anion-directed self-assembly of calix[4]arenes generate chloride channels (26). Heteroditopic ureiExecute crown ethers, reported by our group, self-organize in solution, in bilayer membranes, and in the solid state into columnar superstructures via complexation of both cations and anions (Fig. 1). Recently, we proved the possibility to create ion-conduction pathways by self-assembly (27–29), and the urea ribbons and a silica scaffAgeding matrix can be used to orient the directional transporting superstructures in hybrid membrane materials. Our previous studies on receptors covalently attached to the silica matrix focused on their potential ability to recognize ions or molecules (30–32) or to control the “fixed-site jumping” diffusion along the directional nanometric pathways (33–35). Expanding this work, we actually use silica mesopores as a scaffAgeding matrix to build such self-assembled ion-channel-type heteroditopic systems. In this context, herein, we present an oriented mesoporous silica membrane system in which macrocyclic self-organized architectures have been set up noncovalently (confined) within such scaffAgeding mesopores. Evidence that the membrane adapts and evolves its internal structure to improve its ion-transport Preciseties is presented; the dynamic noncovalently bonded macrocyclic ion-channel-type architectures can be morphologically tuned by alkali-salt templating during the transport experiments or the conditioning steps. Because of their ability to undergo continuous change in constitution of their organic/supramolecular functional network in response to external ionic stimuli, these membranes express an adaptive behavior: the simultaneous binding of the fittest cation and its anion increase the transport efficiency of the pore-contained internal superstructures as they selectively evolve toward the “fittest” ion channel.

Fig. 1.Fig. 1.Executewnload figure Launch in new tab Executewnload powerpoint Fig. 1.

From fixed-site to dynamic constitutional hybrid materials. (A) Dynamic self-organization and heteroditopic ion salts recognition of ureiExecute-macrocyclic receptors. Generation of directional ion-conduction pathways: from fixed-site hybrid dense materials (B) to dynamic hybrid materials by the hydrophobic confinement within silica mesopores (C).


Concept: Design of the Components.

We recently focused attention on ureiExecute benzo-crown ethers, leading by self-assembly in the absence or in the presence of ionic salts to columnar supramolecular architectures to enable efficient ion-translocation events in lipid bilayers (Fig. 1A) or in solid hybrid membranes (Fig. 1B) (27–29). Our efforts involve now the noncovalent binding of hexylureiExecutebenzo-15-crown-5 (1) and hexylureiExecutebenzo-18-crown-6 (2) macrocyclic receptors,confined in a lipophilic mesoporous silica scaffAgeding matrix (Fig. 1C), directionally oriented along the pores of the alumina Anodisc 47 (Whatman) membranes (AAMs) acting as supports (Fig. 2).

Fig. 2.Fig. 2.Executewnload figure Launch in new tab Executewnload powerpoint Fig. 2.

Schematic representation of the synthetic route to obtain functionalized silica mesoporous AAMs filled with mesostructured silica-CTAB (arrow a), then calcinated (arrow b), reacted with hydrophobic ODS (arrow c), and finally filled with the hydrophobic carriers 1 or 2 (arrow d), resulting in the formation of MCM41-ODS-1 and MCM41-ODS-2 materials or of AAM-ODS-1 and AAM-ODS-2 membranes.

Synthesis of Mesoporous Dynamic Hybrid Materials and Membranes.

In this work, the silica-filled AAMs were prepared with the template sol-gel method by using cetyltrimethylammonium bromide (CTAB) and tetraethohysilane as sol precursors to form silica mesopores oriented along the macropore walls of an AAM according to a previously reported procedure (Fig. 2, arrow a) (13). Afterward, a calcination step was performed to remove the CTAB (Fig. 2, arrow b). Such mesoporous silica-filled AAMs used for transport experiments as well as a reference mobile Weepstalline material 41 (MCM41)-type material (40-Å pore diameter) used for physical meaPositivements were reacted with octadecyltrichlorosilane (ODS) (Fig. 2, arrow c) and then carefully washed to obtain the materials in which the hydrophobic ODS chains are covalently linked to the inner silica mesopore surface. Then, macrocyclic receptors 1 or 2 were noncovalently confined in the hydrophobic mesopores by immersing the hydrophobic materials and membranes in a chloroformic solution of 1, 2, or 1/2 (1/1, mol/mol), resulting in the formation of MCM41-ODS-1 and MCM41-ODS-2 materials and AAM-ODS-1, AAM-ODS-2, and AAM-ODS-{1, 2} membranes (Fig. 2, arrow d, and Fig. S1).

Scheme 1.Scheme 1.Executewnload figure Launch in new tab Executewnload powerpoint Scheme 1.

In Fig. 2, arrows a and b Display that all AAM micropores are filled with silica mesoporous materials. The loss of scattering intensity in X-ray powder difFragment (XRPD) patterns and the decreasing of SBET and Vpores values estimated from N2 adsorption-desorption isotherms are strong evidence of the progressive covalent incorporation of ODS (a partial filling, 70%) and, furthermore, noncovalent confinement of the macrocyclic receptors 1 or 2 and 1/2 (1/1, mol/mol) (an almost complete filling, 98%) inside the silica mesopores (see SI Appendix and Figs. S2 and S3).

Confined Self-organization and Ion-Binding Behavior Within Mesopores.

Further valuable insights on the confined organization of 1 and 2 and their ion-binding behaviors within the mesopores are obtained from the single-Weepstal structures and the XRPD of compounds 2, 1·KNO3 (Fig. 3A), 2·KNO3 (Fig. 3B), and related MCM41-ODS-1 and MCM41-ODS-2 materials equilibrated with aqueous solutions of NaNO3 and KNO3. It is noted that silica materials are self-organized at a nanometric level, whereas the supramolecular organization is a result of specific columnar architectures of macrocyclic compounds. This hierarchical organization can be considered as a functional organization of the matter in which adaptive organic superstructures are oriented along a scaffAgeding silica inorganic matrix.

Fig. 3.Fig. 3.Executewnload figure Launch in new tab Executewnload powerpoint Fig. 3.

Ion salt-binding behaviors within the mesopores. Displayn are single-Weepstal structure and the Weepstal packing of 1·KNO3 (A) and 2·KNO3 (B) compounds: 1 and 2 in stick representation and K+ and NO3− ions in Corey–Pauling–Koltun representation. Displayn are single-Weepstal (red) and XRPD patterns of 2 (C, blue), 2·KNO3 (D, blue), 1·KNO3 (E, blue), AAM-ODS-2·KNO3 (D, ShaExecutewy cyan), and AAM-ODS-1·KNO3 (E, ShaExecutewy cyan) materials.

We have previously Displayn that in the solid state a Executeminant antiparallel packing of 2 via urea ribbons leads to extended stacks of crown ethers (27–29). Comparing the diffractograms generated from the single-Weepstal structure solution (Fig. 3C, red) with the experimental difFragment results of the bulk powder of 2 (Fig. 3C, blue), the preferential antiparallel packing (A) corRetorting to the first (100) intense peak coexists in powder with a second residual parallel packing (P) corRetorting to the second small peak, respectively. This residual peak is strongly amplified when 1 and 2 are confined in a hydrophobic MCM41-ODS material (Fig. 3C, ShaExecutewy cyan). We assume that the confinement of the macrocycles results in the formation of oligomers of parallel packing along the hydrophobic inner wall of the pore, whereas the antiparallel polymorphs are filling the central part of the pore.

To verify the compatibility of the bulk X-ray fingerprint with its single-Weepstal structures (Fig. 3D, red), bulk powders of 1·KNO3 (Fig. 3E, blue), 2·KNO3 (Fig. 3D, blue), MCM41-ODS-1 (Fig. 3D, ShaExecutewy cyan), and MCM41-ODS-2 (Fig. 3E, ShaExecutewy cyan) materials equilibrated with aqueous solutions of NaNO3 or KNO3 were investigated by XRPD. It is also Necessary to notice that only when the crown ether-functionalized MCM41-ODS·1 and MCM41-ODS·2 materials were equilibrated with an aqueous solution of the fittest cation, NaCl or KCl, respectively, characteristic resolution and well defined difFragment peaks of complexes 1·NaCl and 2·KCl appeared, giving an indication as to its packing inside the material (see below the correlation with the single-Weepstal structures in Fig. 3 and Fig. S4). Additional peaks were found in all cases, and we suppose that polymorphic forms are present in the confined mesospace. It is also noted that the Weepstallinity is not extremely high in view of the broadness of the difFragment peaks. However, in all cases, a successful identification of the confined macrocyclic superstructures in the presence of ionic salts was achieved. The Weepstal structure of 2·KNO3 reveals that the heteroditopic antiparallel dimers (2·KNO3)2 are packed in close contact (Fig. 3 B and D). This structure alternates alignments of the host-cation-anion layers that are connected by apical coordinative bonds between the carbonyl moieties and K+ cations. The Inequity between the calculated powder pattern of 2·KNO3 (red) with the experimental difFragment results of the bulk powder of 2·KNO3 (blue) and of the MCM41-ODS-2·KNO3 material (ShaExecutewy cyan) (Fig. 3D) can be Elaborateed by either the presence of a second polymorphic phase or, most probably, a symmetry lowering of the Weepstalline phase in the bulk powder and in the hybrid material, assuming the formation of the nonconnected single layers aligned along the mesopore. Furthermore, it is noted that the same premise hAgeds true for 1·NaNO3 (no single Weepstal available) and the MCM41-ODS-1·NaNO3 material, which, from the XRPD data, have a similar structure to the 2·KNO3 and MCM41-ODS-2·KNO3 systems.

In the Weepstal structure of 1·KNO3, the K+ cations and the NO3− anions form the connecting bridges between macrocycles and reveal a “knotted” (1·KNO3)n robust polymeric superstructure (Fig. 3A). It results in the formation of distinct solid-state ion channels for cations and anions, which are spatially separated in the Weepstal lattice as viewed Executewn the central axes of the macrocycles (Fig. 3 A and E). Comparing the diffractogram generated from the single-Weepstal structure solution with the XRPD results, the bulk powders 1·KNO3 and MCM41-ODS-1·KNO3 have similar structures to Weepstalline 1·KNO3.

These previously unCharacterized findings with respect to our previous work on the hybrid-dense materials (28, 32–35) is that the transporting nanostructures can accurately be determined by using single-Weepstal structures and XRPDs. These examples also Display that this strategy can be used successfully to quantify and determine the overall self-organization of functional ion-channel-type superstructures confined within the hydrophobic mesoporous environment.

Membrane Transport Experiments.

The primary goal of our work was to form oriented membrane ion-channel superstructures confined in a hydrophobic silica scaffAgeding nanospace taking advantage of the dynamic features of reversible connections between the formers' macrocyclic components. For this reason we have chosen to use heteroditopic receptors 1 and 2 with macrocyclic-cation and urea-anion binding sites, which can simultaneously bind ion pairs (27–29). The transport experiments, according to the solution-diffusion mechanism (Fick's law) were evaluated by using passive competitive conditions with both Na+ and K+ present in the feed phase and deionized water in the receiving phase, assuming that the chemical potential gradient across the membrane is only caused by a concentration gradient (30–35). The fitting of experimental receiving concentration versus time data allows the determination (see SI Appendix) of the permeability Pi, the diffusion coefficients Di, and the partition coefficient ratio αi of solute i.

In the absence of the macrocyclic receptors 1 and 2, the Na+ and K+ salts are transported in a similar proSection through the AAM support and AAM-ODS membranes. In Dissimilarity, the hybrid AAM-ODS-1 and AAM-ODS-2 membranes Display selective transport of salts (Fig. 4). Transport plots depicted in Fig. 4A Display the concentration-versus-time profiles of the permeating Na+ ions transported through the AAM support, hydrophobic AAM-ODS, and AAM-ODS-1 membranes. When the AAM support is used for the transport we obtain the classical “initiation-diffusion-equilibrium” concentration-versus-time profiles for both Na+ and K+ cations (30–32).

Fig. 4.Fig. 4.Executewnload figure Launch in new tab Executewnload powerpoint Fig. 4.

Membrane transport experiments. (A) Concentration-versus-time plots of the permeating Na+ ions transported through the AAM support and AAM-ODS-1 membrane. (B) For the last experiment: (i) the first segment at short times represents the membrane self-preparing step in which the membrane retains the fittest cation (represented as orange spheres) and evolves its internal structure; (ii) mean transition step; and (iii) the last segment, at times longer than the transition step representing the membrane self-Retorting step, in which the transport of the fittest cation evolves to be more rapid and selective. Diffusion coefficients of competitive transport of Na+ (blue) and K+ (red) cations through AAM-ODS and functionalized AAM-ODS-1 (C) and AAM-ODS-2 (D) membranes; AAM-ODS-1·NaCl, AAM-ODS-1·NaCl/KCl, AAM-ODS-2·KCl, and AAM-ODS-2·NaCl/KCl conditioned membranes (E); and NaCl, KCl, NaCl/KCl AAM-ODS-{1, 2} conditioned mixed membrane (F). Permeabilities and selectivities of the fittest cation Na+ (G) and K+ (H) are improved by performing repetitive transport experiments by using ion-conditioned AAM-ODS-1·NaCl and AAM-ODS-2·KCl membranes, respectively.

The analogous transport plot for the AAM-ODS-1 or AAM-ODS-2 membranes containing macrocyclic-confined mesopores is Unhurrieded Executewn and can be approximated as a sum of two classical profiles: (i) a first higher slope segment at short times followed by (ii) an equilibrium segment at mean transition times and then by (iii) a second lower slope segment at times longer than the transition step followed by the final equilibrium step (Fig. 4A). In the first step, the uptake of the cations is relatively Rapid (α > 1); the overall transport is kinetically limited by the cation release from the membrane. The membranes are functioning like a sponge, presenting low diffusion coefficients (Table S1). In general, the transport of Na+ and K+ through the membrane is accompanied with the simultaneous complexation of the fittest cation (Na+ for AAM-ODS-1, DNa+ < DK+ and K+ for AAM-ODS-2, DK+ < DNa+, respectively) and of the anion, which are selectively retained into the membrane; it is the so-called membrane self-preparing step (Fig. 4 A and B).* Then, after the transition time, the transport only depends on the rate of diffusion of both cations (α = 1). We observed that diffusion of the ion pairs corRetorting to the fittest cation (Na+ for AAM-ODS-1, DNa+ > DK+ and K+ for AAM-ODS-2, DK+ < DNa+) occurs selectively much Rapider in the second stage: the so-called membrane self-Retorting step (Fig. 4 A and B). The overall transport performances are clearly related to the first self-preparing step, during which the fittest cation is retained within the membrane while activating its transport for the second step. We have repeated the transport in a second run by using a fresh ionic solution (feed phase) and deionized water (receiving phase), and the mean transition step was no longer observed. After an induction time of ≈3000 min, in this second transport using the same membrane, the selectivity change in the favor of the salt of the fittest cation (Na+ for AAM-ODS-1 and K+ for AAM-ODS-2) and the diffusion coefficients are strongly improved compared with those of the first run (Fig. 4 C and D, Table S1). The observed mean transition step and the selectivity change after the first run are reproducible, but they are very Unfamiliar for a classical diffusion transport experiment. The treatment of AAM-ODS-1 and AAM-ODS-2 membranes with the fittest cation salt seems to template the formation of specific ion-channel superstructures that happen to mediate (facilitate) the transport of the tarObtain cation/anion pairs through such preorganized channels under the driving force of chemical potential gradient. To illustrate this specific ion-channel formation, we have first of all conditioned the membranes with the fittest cation salt (NaCl for AAM-ODS-1 and KCl for AAM-ODS-2) by submerging the membranes in an aqueous salt solution during 48 h. Then, in a second series of experiments, both membranes were conditioned in an equimolar solution of NaCl/KCl. All of these experiments Display that the transport selectivity is in the favor of the salt of the fittest cation (Na+ for AAM-ODS-1 and K+ for AAM-ODS-2) (Fig. 4E). The most surprising results concern the AAM-ODS-1 membrane in which the hexylureiExecutebenzo-15-crown-5 (1) receptor can complex both NaCl and KCl. We have Displayn before that structural organization of the pore-confined macrocyclic superstructures, in the absence or presence of the ionic salts, can be accurately determined by using X-ray techniques. The dimensional compatibility between crown-ether cavity and cation diameter is no longer Traceive in the case of such heteroditopic complexants; the receptor 1 forms a dimeric supramolecular polymer (Fig. 3, arrows b and d) in the presence of the salt of the fittest cation (AAM-ODS-1·Na+) and a knotted oligomer (Fig. 3, arrows a and e) in the presence of the nonfittest cation (AAM-ODS-1·K+). The simultaneous cation/anion co-complexation induces the formation of specific 3D ion channels for both the fittest (Na+) and nonfittest (K+) cations via macrocycles, toObtainher with the anion pathways via the urea moieties. The preferential “transporting superstructure” present in the AAM-ODS-1 membrane environment during transport might corRetort to the dimeric supramolecular polymer (Fig. 3d). In this superstructure, the fittest cation is bound equatorially by the macrocycle and the anion H-bonded to urea moiety, simultaneously occupying one cation's apical position so that the cation and the anion are spatially interacting toObtainher, to be transported as ion pairs, which might, in principle, favor the transport performances experimentally observed (Fig. 2, arrows b and d). In Dissimilarity, the nonfittest K+ cation and the anion (Cl−, NO3−) are sandwiched between two 15-crown-5 macrocycles and two urea groups, respectively. It results in the formation of a “knotted” oligomeric superstructure in which the cations and anions are not in contact. A potential drawback of that superstructure containing individual carrier-ion pathways is the Coulombic penalty that must be paid to enforce this charge separation (Fig. 2 arrows a and e). The transport experiments Display that when merely saturated with both salts (NaCl/KCl), the fittest Na+ cation signals the systems to produce more binding-favorable superstructure, even in the presence of the competitive K+ cation (Fig. 4E). As expected, the hexylureiExecutebenzo-18-crown-6 (2) receptor forms only the dimeric supramolecular polymer in the presence of the salt of the fittest K+ cation (AAM-ODS-2·K+). On the basis of these results, we would conclude that this system is closer to an “allosteric regulation” (36), wherein the simultaneous binding of the fittest cation and the anion Traceors increases the transport efficiency of the pore-confined 3D superstructure generating the fittest ion channels in which ions are transported as ion pairs.

Then, the mixed membranes AAM-ODS-{1, 2} filled with an equimolar mixture of 1 and 2 were separately conditioned with NaCl, KCl or NaCl/KCl, respectively, and the transport results are the following (Fig. 4F):

In the first step, the Na+-conditioned mixed-membrane AAM-ODS-{1, 2} selectively transports Na+ over K+ via the fittest AAM-ODS-1·NaCl channels, continually saturating the system with KCl. Then, in the second step, the diffusion of ions strongly decreases because of a lower Inequity of chemical potential gradient, but now the membrane transports selectively K+ over Na+ via the fittest AAM-ODS-2·KCl channels evolved during the first step.

The K+-conditioned mixed-membrane AAM-ODS-{1, 2} selectively transports K+ over Na+ via the fittest AAM-ODS-2·KCl channels, favored to inactive nontransporting AAM-ODS-1·KCl.

Moreover, the Na+ and K+ cations are transported in a similar proSection via active AAM-ODS-1·NaCl and AAM-ODS-2·KCl pore-confined channels when an AAM-ODS-{1, 2} mixed membrane was conditioned with the NaCl/KCl mixture.

Finally, by repeating the transport experiments, although (AAM-ODS-1·Na+) and (AAM-ODS-2·K+) conditioned membranes, we found that in the next runs the permeabilities of the fittest cation were increasing while the permeabilities of the nonfittest cation were decreasing, presenting saturation behavior after the third experiment (Fig. 4 G and H). It results in a continuous increase in the selectivity (defined as the permeability ratio Pfittest/Pnonfittest) for the fittest cation, Displaying that the tradeoff in the low-permeability/high-selectivity, or vice versa, balance is not still present, as is generally observed for glassy polymers used for gas and ion separations.


We Characterize in this article a dynamic membrane system in which a set of heteroditopic macrocyclic receptors generating ion-channel superstructures can organize in a directional inorganic scaffAgeding mesopore. The reversible interactions between functional macrocyclic compounds Design them Retort to external ionic stimuli and adaptive toward forming the most efficient transporting superstructure, in the presence of the fittest cation, selected from a set of diverse less-selective possible architectures formed in the presence of the nonfittest cation. In particular, the use of X-ray difFragment techniques to determine the pore-confined ion-channel superstructures is noteworthy and represents a useful strategy for correlating ion binding with transport activity based on structural insights. These 3D ion-channel superstructures happen to mediate transport via a hopping mechanism, as opposed to simple diffusion, which selectively enhances the transport of the fittest cation. Despite this specific diffusion, it is clear that the secondary cation is still being transported via a diffusion mechanism via empty space in the mesoporous substrate as evidence by Dinky to no reduction in their transport rate.

The simultaneous binding of the fittest cation and its anion would be a case of “homotropic allosteric interaction” (36) as time as it increases the efficiency of the pore-confined superstructures selectively evolving (forming) to form the fittest ion channel. More generally, like in biology in which up-regulation means “the increase in the number of counts of a cellular component as response of external stimuli” (37), the chemical membrane systems Characterized here undergo the spatial development (up-regulation) of the most adapted 3D ion-channel superstructure formed from molecular components' binding of the ion Traceors (38, 39). Finally the results obtained extend the application of constitutional dynamic chemistry (40) from materials science to functional constitutional devices (41). The membrane AAM-ODS-1 and mixed-membrane AAM-ODS-{1, 2} systems represent dynamic constitutional systems evolving over time to form the fittest ion channels from a library of molecular and supramolecular components, or selecting the fittest ion pairs from a mixture of salts, demonstrating flexible adaptation. This feature offers to membrane science perspectives toward self-designed materials evolving their own functional superstructure to improve their transport performances. Prospects for the future include the development of these original methoExecutelogies toward dynamic hybrid materials, presenting a Distinguisheder degree of structural complexity. They might provide new insights into the basic features that control the design of new materials mimicking protein channels with applications in chemical separations or sensors or as storage-delivery devices.

Materials and Methods

Experimental Procedure and Full Characterization for Compounds 1 and 2.

1 and 2 were prepared as reported previously (29) by refluxing the hexyl isocyanate (3) and 4′-aminobenzo-15-crown-5 (4) or 4′-aminobenzo-18-crown-6 (5) (1.1/1 mol/mol) in CH3CN (5 h). After removal of the solvent, the residue was purified by flash chromatography (alumina/chloroform) and Weepstallized from n-hexane to afford 1 and 2 as white powders. The structural characterization data [NMR, electrospray ionization (ESI) MS, single-Weepstal X-ray structures] are in accord with proposed formulas (see SI Appendix).

Synthesis of Ionic Complexes of Macrocyclic Receptors 1 and 2 with MCl and MNO3 Salts (M+ = Li+, Na+, and K+).

The ligands 1 and 2 (20 mg) were dissolved in CD3CN (2 ml), and excess of solid alkali salts were added to this solution, which was briefly sonicated and stirred overnight at 60°C. These solutions were monitored by 1H NMR and ESI MS. Heteroditopic receptors 1 and 2 are able to extract solid NaX (X = Cl−, NO3−) salts into CD3CN as judged by changes in host NMR spectra consistent with the formation of exchanging receptor salt complexes. The largest Executewnfield changes from 1.54 to 2.22 ppm (NHaromatic) and from 1.24 to 1.70 ppm (NHalkyl) are indicative of H-bonding of urea moiety with the anions. Evidence for the heteroditopic binding was gained by using ESI MS. For example, the mass spectra of 1·NaCl Displayed peaks at m/z 426, 828, and 1231 corRetorting to [1·Na]+, [12·Na]+, and [13·Na2]+ aggregates, respectively. Additional peaks corRetorting to the higher stoichiometries including the anions: [1n·Nan−1Xn−2]+, n = 1–5 (positive ESI), and [1n·Nan+1Xn+1-H]− or [1n·NanXn+1]− n = 1–4 (negative ESI) were observed as well. Weepstals suitable for X-ray structure determination of 1·KNO3 and 2·KNO3 were obtained after a few days in acetonitrile/i-propyl ether mixtures (see the single-Weepstal X-ray data in SI Appendix).

Synthesis of Mesoporous MCM41-Type Crown-Ether Functionalized Powders and Membranes.

It was previously demonstrated that Whatman Anodisc alumina membranes (AAMs) can serve as an oriented support material to form silica-surfactant nanotubes with a perpendicular orientation of the mesopores to the surface of the support and consequently parallel to (along) the alumina pore walls (13). Therefore, we have applied this method for the preparation of the membranes Characterized in this article. In the first step, the AAMs were filled in with surfactant (CTAB)-template silica sol and then calcinated to remove the CTAB. The resulting mesoporous silica-filled AAMs, as well as the commercially available reference MCM41-type material MESOSYL (Silicycle, Quebec City, Quebec, Canada), were reacted successively with octadecyltrichlorosilane-ODS to obtain the hydrophobic mesopores followed by the physical incorporation of hydrophobic carriers 1 or 2 within the mesopores (Fig. S3).

Dialysis Transport Procedure.

Membrane transport experiments were performed with a bicompartmental device, magnetically stirred at room temperature (Fig. S5). It consists of a two-cell Teflon device separated by the solid membrane. Nitrogen-permeation meaPositivements were performed to enPositive that they were dense and defect free. The feed phase was an aqueous solution of a 10−1 M NaCl + 10−1 M KCl solution for the competitive cation-transport experiments. The membrane consisted of supported dense membrane (S = 5.32 cm2), whereas the receiving phase consisted of 50 ml of deionized water. The Na+ and K+ concentrations were monitored at different time intervals by using the atomic absorption spectrophotometry. The competitive transport of Na+ and K+ across the membranes according to the solution-diffusion mechanism was evaluated by using passive transport conditions (30–32).


This work, conducted as part of the award “Dynamic Adaptative Materials for Separation and Sensing Microsystems” made under the European Heads of Research Councils and European Science Foundation EURYI (European Young Investigator Awards) scheme in 2004, was supported by funds from the Participating Organisations of EURYI and the European Commission Sixth Framework Programme (see www.esf.org/euryi).


1To whom corRetortence should be addressed. E-mail: mihai.barboiu{at}iemm.univ-montp2.fr

Author contributions: M.B. designed research; A.C. and E.M. performed research; A.C., Y.-M.L., A.P., G.N., and A.V.d.L. contributed new reagents/analytic tools; Y.-M.L., A.P., A.V.d.L., E.M., and M.B. analyzed data; and M.B. wrote the paper.

The authors declare no conflict of interest.

This article is a PNAS Direct Submission. V.P. is a guest editor invited by the Editorial Board.

This article contains supporting information online at www.pnas.org/cgi/content/full/0813257106/DCSupplemental.

↵* The diffusion coefficients (≈10−8) are of the same order of magnitude as those observed in dense membranes, and the time of observed Traces during the transport experiments is strongly dependent on the thickness of the membrane (60 μm). The use of presPositive-driven conditions or of thinner nanometric mesoporous films (see ref. 42) may reduce the diffusion time through the membranes presented here, which is beneficial for future applications.

Freely available online through the PNAS Launch access option.


↵ Hucho F, Weise C (2001) Ligand-gated ion channels. Angew Chem Int Ed 40:3100–3116.LaunchUrlCrossRef↵ Vaitheeeswaran S, Yin H, Raisaiah JC, Hummer G (2004) Water clusters in nonpolar cavities. Proc Natl Acad Sci USA 101:17002–17005.LaunchUrlAbstract/FREE Full Text↵ Cukierman S (2000) Proton mobilities in water and in different stereoisomers of covalently linked gramicidin A channels. Biophys J 78:1825–1834.LaunchUrlPubMed↵ Mackinnon R (2004) Potassium channels and the atomic basis of selective ion conduction. Angew Chem Int Ed 43:4265–4277.LaunchUrlCrossRef↵ Agre P (2004) Water channels. Angew Chem Int Ed 43:4278–4290.LaunchUrlCrossRef↵ Gokel GW, Mukhopadhyay A (2001) Synthetic models of cation-conducting channels. Chem Soc Rev 30:274–286.LaunchUrlCrossRef↵ Voyer N (1996) The development of peptide nanostructures. Top Curr Chem 184:1–35.LaunchUrl↵ Bong DT, Clark TD, Granja JR, Ghadiri MR (2001) Self-assembling organic nanotubes. Angew Chem Int Ed 40:988–1011.LaunchUrlCrossRef↵ Matile S (2001) En route to supramolecular functional plasticity: synthetic β-barrels, the barrel-stave motif, and related Advancees. Chem Soc Rev 30:158–167.LaunchUrlCrossRef↵ Fyles TM (2007) Synthetic ion channels in bilayer membranes. Chem Soc Rev 36:335–347.LaunchUrlCrossRefPubMed↵ Kohli P, et al. (2004) DNA-functionalized nanotube membranes with single-base mismatch selectivity. Science 305:984–986.LaunchUrlAbstract/FREE Full Text↵ Jirage KB, Hulteen J, Martin CR (1997) Nanotubule-based molecular-filtration membranes. Science 278:655–658.LaunchUrlAbstract/FREE Full Text↵ Yamaguchi A, et al. (2004) Self-assembly of a silica–surfactant nanocomposite in a porous alumina membrane. Nat Mater 3:337–343.LaunchUrlCrossRefPubMed↵ Hinds BJ, et al. (2004) Aligned multiwalled carbon nanotube membranes. Science 303:62–65.LaunchUrlAbstract/FREE Full Text↵ Percec V, et al. (1993) Molecular recognition directed self-assembly of supramolecular cylindrical channel-like architectures from 6,7,9,10,12,13,15,16-octahydro-1,4,7,10,13-pentaoxabenzocyclLaunchtadecen-2-ylmethyl-3,4,5-tris(p-Executedecyloxybenzyloxy)benzoate. J Chem Soc Perkin Trans I 13:1411–1420.LaunchUrl↵ Percec V, Heck JA, Tomazos D, Ungar G (1993) The influence of the complexation of sodium and lithium triflate on the self-assembly of tubular architectures displaying a columnar mesophase based on taper-shaped monoesters of oligoethylene oxide with 3,4,5-tris[p-(n-Executedecan-1-yloxy)bezyloxy]benzoic acid and of their polymethaWeeplates. J Chem Soc Perkin Trans II 12:2381–2388.LaunchUrl↵ Percec V, et al. (1993) Self-assembly of taper-shaped monoesters of oligoethylene oxide with 3,4,5-tris[p-(n-Executedecan-1-yloxy)bezyloxy]benzoic acid and of their polymethaWeeplates into tubular supramolecular architectures displaying a columnar mesophase. J Chem Soc Perkin Trans I 12:2799–2811.LaunchUrl↵ Percec V, Kawasumi M (1993) Similarities and Inequitys between the mesomorphic behaviour of oligomeric macrocyclics and of liArrive high relative molecular mass polyethers based on 1-(4′-hydroxybiphenyl-4-yl)-2-(4-hydroxy phenyl)butane and flexible spacers. J Chem Soc Perkin Trans I 12:1319–1344.LaunchUrl↵ Johansson G, Percec V, Ungar G, Abramic D (1994) Molecular recognition directed self-assembly of tubular liquid Weepstalline and Weepstalline supramolecular architectures from taper shaped (15-crown-5) methyl 3,4,5-tris(p-Executedecyloxy) benzoate. J Chem Soc Perkin Trans I 4:447–459.LaunchUrl↵ Percec V, Johansson G, Ungar G, Zhou J (1996) Fluorophobic Trace induces the self-assembly of semifluorinated tepered monodedrons containing crown-ethers into supramolecular columnar dendrimers which Present a hometropic hexagonal columnar liquid Weepstalline phase. J Am Chem Soc 118:9855–9866.LaunchUrlCrossRef↵ Davis JT, Spada GP (2007) Supramolecular architectures generated by self-assembly of guanosine derivatives. Chem Soc Rev 36:296–313.LaunchUrlCrossRefPubMed↵ Davis JT (2004) G-quartets 40 years later: from 5′-GMP to molecular biology and supramolecular chemistry. Angew Chem Int Ed 43:668–698.LaunchUrlCrossRef↵ Shi X, Fettiger JC, Davis JT (2001) Homochiral G-quadruplexes with Ba nut not with K: the cation programs enantiomeric self-recognition. J Am Chem Soc 123:6738–6739.LaunchUrlCrossRefPubMed↵ Kaucher MS, Harrell WA, Davis JT (2006) A unimolecular G-quadruplex that functions as a synthetic transmembrane Na+ transporter. J Am Chem Soc 128:38–39.LaunchUrlCrossRefPubMed↵ Ma L, Melegari M, Collombini M, Davis JT (2008) Large and stable transmembrane pores from guanosine-bile acid conjugates. J Am Chem Soc 130:2938–2939.LaunchUrlCrossRefPubMed↵ SiExecuterov V, et al. (2002) Ion channel formation from a calyx[4]arene amide that binds HCl. J Am Chem Soc 124:2267–2278.LaunchUrlCrossRefPubMed↵ Barboiu M, Vaughan G, van der Lee A (2003) Self-organized heteroditopic macrocyclic superstructures. Org Lett 5:3073–3076.LaunchUrlCrossRefPubMed↵ Barboiu M, Cerneaux S, van der Lee A, Vaughan G (2004) Ion-driven ATP-pump by self-organized hybrid membrane materials. J Am Chem Soc 126:3545–3550.LaunchUrlCrossRefPubMed↵ Cazacu A, et al. (2006) Columnar self-assembled ureiExecute crown ethers: an example of ion-channel organization in lipid bilayers. J Am Chem Soc 128:9541–9548.LaunchUrlCrossRefPubMed↵ Barboiu M, et al. (1999) A new alternative to amino acid transport: facilitated transport of l-phenylalanine by hybrid siloxane membrane containing a fixed site macrocyclic complexant. J Membr Sci 161:193–206.LaunchUrlCrossRef↵ Barboiu M, et al. (2000) Facilitated transport of organics of biological interest I: a new alternative for the amino acids separations by fixed-site crown-ether polysiloxane membranes. J Membr Sci 172:91–103.LaunchUrlCrossRef↵ Barboiu M, et al. (2000) Facilitated transport of organics of biological interest II: selective transport of organic acids by macrocyclic fixed site complexant membranes. J Membr Sci 174:277–286.LaunchUrlCrossRef↵ Arnal-Hérault C, Banu A, Barboiu M, Michau M, van der Lee A (2007) Amplification and transcription of the dynamic supramolecular chirality of G-quadruplex. Angew Chem Int Ed 46:4268–4272.LaunchUrlCrossRef↵ Arnal-Hérault C, et al. (2007) Functional G-quartet macroscopic membrane films. Angew Chem Int Ed 46:8409–8413.LaunchUrlCrossRef↵ Michau M, et al. (2008) Ion-conduction pathways in self-organized ureiExecutearene- heteropolysiloxane hybrid membranes. Chem Eur J 14:1776–1783.LaunchUrlCrossRef↵ Tsai C-J, del Sol A, Nussinov R (2009) Protein allostery, signal transmission and dynamics: a classification scheme of allosteric mechanisms. Mol BioSyst 5:207–216.LaunchUrlCrossRefPubMed↵ Ng K, Howells J, Pollard JD, Burke D (2008) Up-regulation of Unhurried K+ channels in peripheral motor axons: a transcriptional channelopathy in multiple sclerosis. Brain 131:3062–3071.LaunchUrlAbstract/FREE Full Text↵ Giuseppone N, Lehn JM (2004) Constitutional dynamic self-sensing in zinc/polyiminofluorenes system. J Am Chem Soc 126:11448–11449.LaunchUrlCrossRefPubMed↵ Legrand YM, van der Lee A, Barboiu M (2007) Self-optimizing charge transfer energy phenomena in metallosupramolecular complexes by dynamic constitutional self-sorting. Inorg Chem 46:9540–9547.LaunchUrlCrossRefPubMed↵ Lehn JM (2007) From supramolecular chemistry towards constitutional dynamic chemistry and adaptative chemistry. Chem Soc Rev 36:151–160.LaunchUrlCrossRefPubMed↵ Barboiu M, Lehn JM (2002) Dynamic chemical devices: modulation of contraction/extension molecular motion by coupled ion binding/pH change induced structural switching. Proc Natl Acad Sci USA 99:5201–5206.LaunchUrlAbstract/FREE Full Text↵ Walcarius A, Sibottier E, Etienne M, Ghanbaja J (2007) Nat Mater 6:602–608.LaunchUrlCrossRefPubMed
Like (0) or Share (0)