Hosting of [Cs(crown ether)₂]⁺ type host–guest complexes by a nano-sized molecular cuboid

Abstract

Encapsulation of a guest by host-forming (G-in-H)-type host–guest complexes is the central theme of supramolecular chemistry. Binding of typical hosts (H_in) in the cavity of a larger host (H_out) to form (H_in-in-H_out) type assemblies is also known. However, the incorporation of host–guest complexes in a larger host to form [(G-in-H_in)-in-H_out] type assemblies, where (G-in-H_in) is composed of one guest and one host is relatively unexplored. A recent report describes the binding of host–guest complexes consisting of one guest and two hosts in a larger host molecule. In the present work, we have prepared a nano-sized polycationic molecular cuboid of the type Pd(L′)₈(L)₄ as the H_out, which is capable of encapsulating two units of H_in such as dibenzo[3n]crown-n (where n ranges from 6 to 8). The molecular cuboid is found to selectively recognize the larger-sized dibenzo[24]crown-8 and dibenzo[21]crown-7 over the smaller-sized dibenzo[18]crown-6. The polycationic molecular cuboid can bind a family of [Cs(crown ether)₂]⁺ type cationic host–guest complexes composed of one unit of a guest (Cs⁺) and two units of the H_in (crown ether). All these complexes are prepared via self-assembly in a water medium; the surrogate nuclear waste Cs⁺ is deeply internalized, being encapsulated by two units of crown ether, which are in turn sheltered by the molecular cuboid. All the [G-in-(H_in)₂-in-H_out] type assemblies are characterized using spectroscopic techniques and single-crystal structures. The relative orientation of the pair of crown ethers around the Cs⁺ ion is found to be different in the native form compared to the encapsulated form in the molecular cuboid.

---

1 Introduction

Host–guest chemistry is a well acknowledged central theme of supramolecular chemistry that many researchers have pursued intensely, employing a variety of host and guest entities to achieve intriguing results.^1–7 Host–guest complexes composed of one variety of host and one variety of guest can be formulated as [(G)_x⊂(H)_y] where the values of x and y are variable but small, the simplest varieties being [(G)⊂(H)], [(G)₂⊂(H)] and [(G)⊂(H)₂].^4,8,9 A smaller host molecule may behave like a guest molecule when one or more units of the host are accommodated in the cavity of a larger host molecule to form host-in-host type complexes. The molecular recognition of a host, H_in, by another host, H_out, is relatively less studied; such host-in-host species can be formulated as [(H_in)_y⊂(H_out)] where y ≥ 1. The larger host, i.e., H_out, could be a coordination cage or a covalent cage entity.^10–16 If H_in comes in contact with a complementary guest molecule G either (i) prior to binding with H_out, (ii) during binding with H_out, or (iii) after binding with H_out, then the combined interactions among G, H_in and H_out should result in higher-order structures such as [{(G)_x⊂(H_in)_y}_z⊂(H_out)]. Encapsulation of one unit of [(G)⊂(H_in)] in the cavity of H_out to form a [{(G)⊂(H_in)}⊂(H_out)] type super-assembly (where x = 1, y = 1 and z = 1) is known.^17–29 Additionally, there is an example of the encapsulation of two units of [(G)⊂(H_in)] in the nanospace of H_out to give a [{(G)⊂(H_in)}₂⊂(H_out)] type assembly (where x = 1, y = 1 and z = 2).³⁰ A very recent report by Fujita demonstrated the assembly of G, H_in and H_out in which the guest molecule G is sandwiched by two units of H_in to form a [(G)⊂(H_in)₂] type entity located inside the cavity of H_out, giving rise to a series of elaborate [{(G)⊂(H_in)₂}⊂(H_out)] type molecules (where x = 1, y = 2 and z = 1).³¹ The encapsulation of either a suitable host or host–guest entity via molecular confinement in another larger host is relatively underexplored. Such phenomena are worthy of investigation as they are anticipated to provide novel supramolecular behaviours including the stabilization of otherwise unstable species, the imposition of unusual conformations and novel behaviours on the bound species, the modulation of guest binding strength/selectivity, unusual product formation via catalysis reactions in the confined space, etc.

In the present work, we disclose the preparation of [(H_in)₂⊂(H_out)] and [{(G)⊂(H_in)₂}⊂(H_out)] type super-assemblies in a water medium using the same G and H_out units with different H_in as shown in Fig. 1. The outer-host, i.e., H_out, explored here is a new cuboid-shaped self-assembled nanosized coordination cage with the formula [Pd₈(L′)₈(L)₄]¹⁶⁺, where L′ represents tetramethylethylenediamine (tmeda) and L is a rectangular panel-shaped 4-pyridyl appended tetrakis monodentate ligand. The family of inner-hosts, i.e., H_in, used in this study includes a range of classical crown ethers such as dibenzo[3n]crown-n (n = 6, 7, and 8), and a surrogate nuclear waste, namely, Cs⁺, is used as the cationic guest G. For the [(H_in)₂⊂(H_out)] type assemblies, the selective binding of one type of H_in (e.g., dibenzo[24]crown-8 or dibenzo[21]crown-7) over the binding of another H_in (i.e., dibenzo[18]crown-6) by the outer host H_out is observed. The relative orientations of the two crown ether units in the structure of [(Cs)(crown ether)₂]⁺ in the native state and encapsulated state (in the cavity of H_out) showed remarkable differences.

Fig. 1 Conceptualizing the synthesis of assemblies of (i and ii) [(H_in)₂⊂(H_out)] and (iii–vi) [{(G)⊂(H_in)₂}⊂(H_out)] type molecules. (vii) H_in exchange reaction showing selective inclusion of one type of H_in over another type of H_in by H_out.

2 Results and discussion

2.1 Conceptualization

The key objective of this work is to accomplish [{(G)⊂(H_in)₂}⊂(H_out)] type water soluble molecules in a modular fashion by choosing a suitable set of G, H_in and H_out, as shown in Fig. 1. Self-assembled polycationic coordination cages having a nano-space could be ideal H_out type units, in view of the proven ability of such cages to encapsulate large-sized hydrophobic guest molecules.^32–35 The choice of H_out should be made on the basis of the overall shape and size of H_in or vice versa. Crown ethers such as dibenzo[3n]crown-n are macro-monocyclic hydrophobic host molecules are well known to accommodate alkali metal ions as guests. It should be possible to encapsulate a typical crown ether like dibenzo[18]crown-6 in the cavity of a suitably designed coordination cage H_out. The flexible polyether chain of the crown ether can undergo conformational changes, if necessary, to become encapsulated in the available inner space of H_out. Thus, crown ethers, which themselves are typical hosts, may act as guests in a suitably designed large coordination cage (Fig. 1(i) and (ii)). In the encapsulated form, the crown ether may retain its initial ability to encapsulate a suitable alkali metal ion (Fig. 1(iii) and (iv)). In that case, the crown ether would play dual roles as both host and guest. For convenience, we herein use the term H_in for the crown ether (and H_in-a and H_in-b to distinguish two different crown ethers). The Cs⁺ ion is larger than the cavity of dibenzo[18]crown-6; thus, it perches upon the crown ether cavity. For efficient binding, two units of this crown ether are required to sandwich one unit of Cs⁺, forming [(Cs)⊂(dibenzo[18]crown-6)₂]⁺, for which crystal structures have been reported. The design principle of the current work revolves around the use of a self-assembled coordination cage as H_out, typical crown ethers as H_in, and Cs⁺ as the guest.

We are involved in the synthesis of Pd(II)-based self-assembled coordination cages using pyridine-appended multidentate ligands.³⁶ In a recent study, we combined a di-cationic cis-protected Pd(II), i.e., (PdL′), and a rectangular panel-shaped 3-pyridyl appended tetrakis-monodentate ligand to prepare a self-assembled dodeca-cationic molecular prism of the type Pd₆(L′)₆(L)₃ that is capable of encapsulating small hydrophobic guest molecules (neutral/polar/anionic) in a water medium.³⁷ In the current work, the complexation of cis-protected Pd(II) with a 4-pyridyl appended tetrakis monodentate ligand is undertaken in order to expand the cavity size to prepare a new molecular cuboid of the type Pd₈(L′)₈(L)₄ for the encapsulation of larger-sized hydrophobic molecules in a water medium.

2.2 Nano-sized self-assembled molecular cuboid: H_out-type super-host

The tetrakis monodentate ligand 3,3′,5,5′-tetra(pyridin-4-yl)-1,1′-biphenyl, L (Fig. 2), which has a biphenyl core and four units of terminal 4-pyridyl rings on the periphery, was synthesized via a Suzuki–Miyaura cross-coupling reaction following a literature procedure.³⁸ Both the length and width of the panel-shaped rectangular ligand are greater than one nanometre. Surprisingly, the potential of this simple ligand for complexation with metal ions has not been explored to date. Complexation of cis-protected Pd(II) with the ligand L under optimized conditions produced the new molecular cuboids [Pd₈(tmeda)₈(L)₄](NO₃)₁₆, H1·16NO₃; [Pd₈(tmeda)₈(L)₄](PF₆)₁₆, H1·16PF₆; and [Pd₈(en)₈(L)₄](NO₃)₁₆, H2·16NO₃ (Fig. 2), as described in this section.

Fig. 2 Synthesis of the polycationic molecular cuboids [Pd₈(tmeda)₈(L)₄](X)₁₆ (X = NO₃: H1·16NO₃; X = PF₆: H1·16PF₆), and [Pd₈(en)₈(L)₄](NO₃)₁₆, H2·16NO₃. Crystal structure of H1·16PF₆ showing the [Pd₈(tmeda)₈(L)₄]¹⁶⁺ architecture (counter anions and H atoms are not shown for clarity).

Initially, a mixture of a slight excess of cis-Pd(tmeda)(NO₃)₂ (0.015 mmol) and the ligand L (0.005 mmol) was taken in 1 mL of D₂O and stirred at room temperature, whereupon the metal component dissolved immediately and the ligand remained suspended even after 2 days of stirring. However, upon stirring at 60 °C, the ligand was solubilized within 2 days. The reaction mixture was then cooled and centrifuged, and the supernatant containing the product was separated. The ¹H NMR spectrum of the in situ prepared sample (D₂O, 298 K, i.e., 25 °C) showed a single set of slightly broad peaks (SI Fig. S11) along with signals corresponding to the unutilized metal component, indicating the formation of a single discrete coordination cage as the product. A mixed solvent such as H₂O : CH₃CN could lower the time required for the complexation reaction in this work, as already established by us for a related cage molecule. Thus, the complexation reaction was carried out in D₂O : CD₃CN (1 : 1) with stirring at 60 °C to afford a clear solution within 6 h. The ¹H NMR spectrum of the in situ prepared sample (D₂O : CD₃CN (1 : 1), 298 K, i.e., 25 °C) also showed a single set of slightly broad peaks (SI Fig. S12).

The reaction was then repeated in a similar manner in H₂O : CH₃CN (1 : 1) followed by evaporation and washing with CH₃CN to isolate the targeted self-assembled complex as an off-white solid. Stock solutions were prepared by dissolving the isolated solid in either D₂O or D₂O : CD₃CN (1 : 1); these solutions were used for ¹H NMR spectral study at 298 K, i.e., 25 °C. The obtained spectra of the isolated samples very closely resembled the spectra of the in situ prepared samples in D₂O and D₂O : CD₃CN (1 : 1), respectively (SI Fig. S17 and S18). Comparison of the peak positions of the ensuing coordination cage with the free ligand could not be made due to insolubility of L in D₂O or D₂O : CD₃CN (1 : 1). However, a comparison of the ¹H NMR spectra of L in CDCl₃ and the isolated cage is shown in Fig. 3(i) and (ii).

Fig. 3 400 MHz ¹H NMR spectra of (i) ligand L in CDCl₃ at 298 K, i.e., 25 °C, and (ii and iii) isolated [Pd₈(tmeda)₈(L)₄](NO₃)₁₆, H1·16NO₃ in D₂O at 298 K/333 K, i.e., 25 °C/60 °C.

We propose the formation of [Pd₈(tmeda)₈(L)₄](NO₃)₁₆, H1·16NO₃ on the basis of geometrical considerations of the chosen metal and ligand components.^32–34 Slow conformational changes at room temperature are likely responsible for the broad peaks of H1·16NO₃. This could be due to restricted rotation of the pyridine units around the pyridine–biphenyl linkages imposed by the methyl groups of tmeda, which are located near the bound pyridine units. Thus, we used cis-Pd(en)(NO₃)₂ to prepare [Pd₈(en)₈(L)₄](NO₃)₁₆ (H2·16NO₃); the ¹H NMR spectrum of this complex in D₂O at room temperature (298 K, i.e., 25 °C) showed sharp peaks (see SI Fig. S19). ¹H NMR spectra of H1·16NO₃ were recorded at an elevated temperature (333 K, i.e., 60 °C) in D₂O or D₂O : CD₃CN (1 : 1), and sharp peaks were obtained (see Fig. 3 and SI Fig. S18), indicating fast conformational changes at elevated temperature. In our prior experience, good crystals are more easily obtained when the cis-protecting group is tmeda rather than en; hence, we used H1·16NO₃ for the planned host–guest interaction study.

Further NMR characterization, including ¹³C, H–H COSY, NOESY, HSQC and DOSY, was carried out for H1·16NO₃ in D₂O : CD₃CN (1 : 1) due to the better solubility in this mixed solvent, which allowed the possibility to obtain samples at higher concentration (see SI Fig. S21–S29). However, the host–guest study was performed in D₂O, keeping in mind the role of water as a universal solvent in biosystems. The ¹³C NMR spectrum of the sample showed a single set of peaks even at room temperature, and DOSY showed a single diffusion band (D = 1.77 × 10⁻¹⁰ m² s⁻¹) for all the protons, thereby supporting the formation of a single discrete product. The ESI-MS study was not successful for H1·16NO₃; however, it was successful for the related complex [Pd₈(tmeda)₈(L)₄](PF₆)₁₆ (H1·16PF₆), which was prepared by adding NH₄PF₆ to a solution of H1·16NO₃ in H₂O. The experimentally obtained isotopic pattern peaks at m/z = 846.44, 1045.11 and 1342.65 corresponding to the polycationic components [M–6PF₆]⁶⁺, [M–5PF₆]⁵⁺ and [M–4PF₆]⁴⁺ were found to be comparable with the simulated patterns, thus supporting the formation of the proposed octa-nuclear complexes (see SI Fig. S30 and S31).

Fortunately, we obtained good-quality crystals suitable for data collection via slow diffusion of 1,4-dioxane into a concentrated solution of H1·16PF₆ in CH₃CN. The complex crystallized in the tetragonal space group I4/m. The asymmetric unit contains one-eighth of a molecule of the polycationic cage, which includes half of a molecule of ligand L coordinated to one Pd(II) center, four hexafluorophosphate anions (all half-occupancy, with one disordered by symmetry about a four-fold rotoinversion axis). The crystal structure (Fig. 2, see SI Section S6 for details) confirmed the proposed formula and revealed the relative arrangement of the ligand units in the molecular cuboid architecture [Pd₈(tmeda)₈(L)₄]¹⁶⁺.

The molecular cuboid H1·16NO₃ is water-soluble and possesses a confined hydrophobic cavity that should be suitable for binding large hydrophobic guest molecules of suitable shapes/sizes. Thus, the hexadecacationic cuboid, i.e., (H1)¹⁶⁺, is considered here as the potential H_out type unit. A cartoon representation depicting the shape of the cationic cavity is shown in Fig. 4(A).

Fig. 4 (A) Cuboid-shaped self-assembled coordination cages as H_out. (B) Spherical monoatomic alkali metal ion Cs⁺ as G. (C and D) Selected podands and crown ethers as H_in.

2.3 Ether podands and crown ethers: H_in type hosts

The ether podands and crown ethers (H3 to H8) evaluated in this work are shown in Fig. 4. The podands H3 and H4 are ortho-substituted benzene derivatives having a total of four and six oxygen donor sites, respectively, located on two symmetrical arms. Dibenzo[18]crown-6 (H6) is the first crown ether ring prepared by Pedersen that has two ortho-substituted benzene rings and six oxygen donor sites arranged in a cyclic array.³⁹ The crown ethers H7 and H8 are higher homologues having seven and eight oxygen donors, respectively. The macrocyclic rings H5 and H6 are of comparable size, but H5 is less hydrophobic as it has only one aromatic ring. The host molecules H3 to H8 were considered as H_in, as we expected that they could enter the cavity of H_out and still act as a host in this encapsulated form.

2.4 Cs⁺ ions as the guest G of the podands/crown ethers: [(G)⊂(H_in)_x]

Podands and crown ethers are known for the binding and extraction of alkali metal ions.^40,41 Typically, crown ethers are more efficient than podands; however, podands may enjoy a kinetic advantage over crown ethers. We were interested in exploiting the binding of the Cs⁺ ion by H_in type hosts inside the cavity of an H_out type host in water. The binding of commercially available Cs⁺ in water is of immense importance, as Cs⁺ is a surrogate of the hazardous radionuclides ¹³⁴Cs and ¹³⁷Cs that are produced as the main nuclear waste of the uranium fission reaction.⁴² Crystal structures^43,44 have been reported for [(Cs)⊂(H6)₂]⁺, in which the ratio of host to guest is 2 : 1. The larger crown ethers H7 and H8 are known to bind Cs⁺ to form [(Cs)⊂(dibenzo[21]crown-7)(X)]⁺ and [(Cs)⊂(dibenzo[24]crown-8)(X)]⁺, where X represents additional coordination sites provided by suitable additives.^45,46 We believed that one or more units of H6 might become encapsulated in the cavity of (H1)¹⁶⁺, and that the encapsulated crown ether would still retain its host-like ability for capturing alkali metal ions. In view of the above-mentioned 2 : 1 binding mode and the availability of space inside the cavity of H_out for hosting [(Cs)⊂(H6)₂]⁺, it should be possible to generate a [{(G)⊂(H_in)₂}⊂(H_out)] type entity in which Cs⁺ is securely lodged as a guest and encapsulated by multiple layers. In this context, we wished to evaluate the hosting behaviour of the podands and the larger crown ethers when encapsulated inside H_out.

2.5 A typical host H_in as the guest of a super-host H_out: [(H_in)_x⊂(H_out)]

Before proceeding to the synthesis of the targeted [{(G)⊂(H_in)₂}⊂(H_out)] type assembly, we decided to study the hosting abilities of (H1)¹⁶⁺, i.e., H_out, towards the podands/crown ethers (H3 to H8). The podands H3 and H4 are water soluble, as is the crown ether H5. However, the crown ethers H6, H7 and H8 are not soluble in water. The encapsulation study was performed by portion-wise addition of a variety of water-soluble H_in (i.e., H3 to H5) to a solution of H_out (i.e., H1·16NO₃) in D₂O, and the progress of the reactions was followed by recording ¹H NMR spectra of the samples to evaluate the value of “x” in the probable [(H_in)_x⊂(H_out)] type entities. The water-soluble podands H3 and H4 could be encapsulated by (H1)¹⁶⁺, as can be seen from the shift of the signals observed in the ¹H NMR spectra recorded during the monitoring (see SI Fig. S35 and S36). However, changes in the peak positions could be seen even after the addition of several equivalents of the podands. The formation of [(H3)_x⊂(H1)]·16NO₃ (1) and [(H4)_x⊂(H1)]·16NO₃ (2) is proposed, although the value of “x” could not be deduced. This situation was also observed for the binding of water-soluble crown ether H5 (see SI Fig. S37), for which the value of x in [(H5)_x⊂(H1)]·16NO₃ (3) could not be calculated.

An excess amount of each water-insoluble H_in (i.e., H6, H7 or H8) was added to a solution of H_out (i.e., H1·16NO₃) in D₂O, and the reaction mixture was stirred at room temperature for 12 h. The mixture was then centrifuged, and the supernatant was evaporated to isolate the ensuing [(H_in)_x⊂(H_out)] type assemblies. We propose that the cavity of (H1)¹⁶⁺ accommodated two units of crown ethers in D₂O, resulting in [(H6)₂⊂(H1)]·16NO₃ (4), [(H7)₂⊂(H1)]·16NO₃ (5) and [(H8)₂⊂(H1)]·16NO₃ (6) (Fig. 5, step (i) and step (ii); SI Fig. S117, step (ii)). This binding stoichiometry, i.e., x = 2, is proposed based on the calculated integration ratios of the signals related to H_out and encapsulated H_in in the ¹H NMR spectra of isolated samples recorded in D₂O (see Fig. S38–S40).

Fig. 5 Designer coordination cage (H1)¹⁶⁺ as super-host H_out for encapsulation of (i and ii) two units of typical crown ethers (H6/H7) as H_in to form [(H_in)₂⊂(H_out)]. (iii and iv) Addition of Cs⁺ as G after accommodation of the two units of crown ether to form [{(G)⊂(H_in)₂}⊂(H_out)]. (v and vi) Combination of one equivalent of Cs⁺ ions and two equivalents of typical crown ethers in a single pot to form [{(G)⊂(H_in)₂}⊂(H_out)]. (vii) Encapsulation of one crown ether over another in a selective manner, showing the preference for H7 over H6 (the behaviour of H8 is found to be comparable with that of H7).

Proton d of (H1)¹⁶⁺ shows a remarkable upfield shift upon uptake of the crown ethers (compare Fig. 6(i) with Fig. 6(ii), (iv) and (vi) for uptake of H6, H7 and H8, respectively) indicating its location above the plane of the aromatic rings of the encapsulated crown ethers. This effect is especially pronounced in the case of H6 (compare Fig. 6(i) with Fig. 6(ii)), indicating restricted movement of H6 in the cavity. In addition to ¹H NMR, compounds 4, 5 and 6 were also characterised using various NMR spectroscopic techniques including ¹³C, COSY, HSQC, and NOESY (see SI Fig. S41–S72). The DOSY NMR data for compounds 4, 5 and 6 (see Fig. S73–S75) supports the encapsulation of the dibenzo-crown ethers (H_in) in the cavity of the molecular cuboid (H_out), as all signals for all of these compounds present as a single diffusion band.

Fig. 6 ¹H NMR spectra (400 MHz, D₂O, 298 K, i.e., 25 °C) for isolated samples of (i) H1·16NO₃, (ii) [(H6)₂⊂H1]·16NO₃ (4), (iii) [{G⊂(H6)₂}⊂H1]·17NO₃ (8), (iv) [(H7)₂⊂H1]·16NO₃ (5), (v) [{G⊂(H7)₂}⊂H1]·17NO₃ (9), (vi) [(H8)₂⊂H1]·16NO₃ (6), and (vii) [{G⊂(H8)₂}⊂H1]·17NO₃ (10).

To determine whether the molecular cuboid had any preference for the binding of a particular crown ether over another, we conducted competition experiments and recorded the corresponding ¹H NMR spectra. Addition of H7 to a solution of [(H6)₂⊂(H1)]·16NO₃ (4) in D₂O resulted in the formation of [(H7)₂⊂(H1)]·16NO₃ (5) (Fig. 5, step (vii) and SI Fig. S76), whereas addition of H6 to a solution of 5 in D₂O resulted in no changes. This data suggested the preferential binding of H7 over H6 by (H1)¹⁶⁺. Similar experiments revealed that the binding of H8 was preferred over that of H6 (see SI Fig. S117, step (vii) and Fig. S77). However, addition of H8 to a D₂O solution of 5 or addition of H7 to a D₂O solution of 6 resulted in the formation of a mixture of products, as observed in their ¹H NMR spectra. We believe that the host (H1)¹⁶⁺ does not favour the binding of H7 over H8 or vice versa (see SI Fig. S118, step (vii) and (viii), and Fig. S78). This lack of competition possibly generates multiple compounds including [(H7)(H8)⊂(H1)]·16NO₃ (7). Thus, the affinity is proposed to follow the order H8 ≈ H7 > H6.

2.6 Host–guest complex [(G)⊂(H_in)₂] as the guest of the super-host H_out: [{(G)(H_in)₂}⊂(H_out)]

Since H6 is known to form a [Cs⊂(H6)₂]⁺ type complex, we anticipated the formation of [{(Cs)⊂(H6)₂}⊂(H1)]·17NO₃ (8) via the direct addition of excess CsNO₃ to a solution of [(H6)₂⊂(H1)]·16NO₃ (Fig. 5, step (iii)) or one-pot mixing of an excess of CsNO₃, an excess of H6 and (H1)·16NO₃ in D₂O (Fig. 5, step (v)). The ¹H NMR spectra of the resulting assembly (proposed as 8) prepared by either method were found to be comparable. The spectrum of resulting assembly 8 is shown in Fig. 6(iii); it revealed the retention of 2 units of encapsulated H6 and an appreciable downfield/upfield shift of the aromatic/aliphatic proton signals of the H6 moiety compared to those in the spectrum of 4 (compare Fig. 6(iii) with Fig. 6(ii)). Interestingly, the geminal methylene protons of H6 (geminal protons k₆, as well as l₆) were found to be non-equivalent. Additionally, the para-substituted pyridine rings of coordinated L in 8 showed four signals instead of the expected two signals. These observations suggest unsymmetrical environments for these protons due to the restricted rotation of the pyridyl groups with respect to the biphenyl core and the restricted rotation of the ethylene spacers of the H6 units in the structure of 8, as shown in Fig. 7(A). This proposal is also supported by the COSY and NOESY data (Fig. 7(B) and (C)). The NOESY data for 8 reveals cross-peak correlations between the aromatic proton g₆ and the aliphatic protons k₆, , l₆ and ; however, no such correlations are observed between the aromatic proton h6 and the aliphatic protons (Fig. 7(B)). The ¹³³Cs NMR spectrum of a sample containing a tenfold excess of Cs⁺ showed an upfield-shifted signal for the bound Cs⁺ present in compound 8 (Δδ of ∼35 ppm) compared to the signal of free Cs⁺ as shown in Fig. 7(D), confirming the presence of a well-shielded Cs⁺. The single diffusion band in the DOSY spectrum of 8 supported the existence of only one product (Fig. 7(E)). On the basis of the ¹H NMR spectrum, we propose that the conformational changes in the backbone of H6 are highly restricted, as it wraps around Cs⁺ and is located in the cavity of (H1)¹⁶⁺. It is likely that all the oxygen atoms of H6 are coordinated with Cs⁺, resulting in restricted rotation.

Fig. 7 (A) Partial representation of [{G⊂(H6)₂}⊂H1]·17NO₃, 8, showing one of the walls of the super-host and part of the encapsulated host–guest complex. (B and C) Partial ¹H–¹H COSY and ¹H–¹H NOE spectra (400 MHz, D₂O, 298 K) emphasising correlations between certain protons (for COSY: a with b and a′ with b′; for NOE: g₆ with k₆, , , and l₆) for characterization of 8. (D) ¹³³Cs NMR spectrum (66 MHz, D₂O, 298 K) of 8 in the presence of excess Cs⁺, supporting the encapsulation of G, i.e., Cs⁺. (E) ¹H-DOSY NMR spectrum (500 MHz, D₂O, 298 K) of 8 showing a single band for which the calculated diffusion coefficient (D) value is 1.67 × 10⁻¹⁰ m² s⁻¹.

The compounds [{(Cs)(H7)₂}⊂(H1)]·17NO₃ (9) and [{(Cs)(H8)₂}⊂(H1)]·17NO₃ (10) were prepared in a manner very similar to 8. The aromatic protons of the H7 moiety of 9 were reasonably downfield-shifted, whereas the aliphatic protons showed a marginal downfield shift and broadening as compared to those of 5 (compare Fig. 6(v) with Fig. 6(iv)). The aromatic/aliphatic protons of the H8 moiety of 10 are marginally downfield-shifted and broadened (compare Fig. 6(vii) with Fig. 6(vi)). The para-substituted pyridine rings of coordinated L in 9 and 10 showed only two signals in contrast to the four signals observed in the case of 8. Thus, the conformational movement of the L moiety and the H7/H8 moiety in complexes 9/10 is not restricted. It is likely that not all the oxygen atoms of H7/H8 are coordinated with Cs⁺ simultaneously, thus allowing conformational changes. The compounds 8, 9 and 10 were characterised using various NMR spectroscopic techniques including ¹H, ¹³C, COSY, NOESY and HSQC (see SI Fig. S80–S110). The DOSY NMR data for the compounds (see SI Fig. S111–S113) supports the formation of a single discrete product, as all the signals for all of these compounds belong to a single diffusion band. Stacked ¹H NMR spectra depicting the portion-wise addition of an aqueous solution of CsNO₃ to aqueous solutions of 4, 5 and 6 to generate 8, 9 and 10 are shown in SI Fig. S115–S117. For the binding of Cs⁺ by the host-in-host system 4, a slow guest-exchange was observed and a high binding constant (K_a > 10⁵ M⁻¹) is proposed,⁴⁷ as one equivalent of Cs⁺ was sufficient to populate 4. However, for the binding of Cs⁺ by the host-in-host systems 5 and 6, fast guest exchange was observed with a low binding constant in both cases; for the former case, K_a is approximately 10² M⁻¹, while for the latter case the value was too small to be estimated.

2.7 Crystal structures of [{(G)(H_in)₂}⊂(H_out)] type complexes

We could not obtain suitable HRMS data for the [{(guest)⊂(host_in)₂}⊂(host_out)] type molecules, but single-crystal X-ray crystallographic analysis unequivocally supported the formation of 8, 9 and 10 (Fig. 8(A), see SI Section S6 for details). Single crystals were obtained by slow evaporation of their aqueous solutions at room temperature over a period of 3 weeks. Compound 8 crystallized in the tetragonal crystal system with the I4₁/a space group. The asymmetric unit of 8 comprises one unit of ligand L coordinated to two units of cis-protected Pd(II) ions and half a molecule of the crown ether coordinated to a Cs⁺ ion, which lies on the fourfold rotoinversion axis. Two nitrate ions were located and refined, one of which is disordered over three positions. In contrast, compounds 9 and 10 crystallized in the triclinic space group P1̄. In case of 9, the asymmetric unit contains half the complex, which includes two units of ligand L coordinated to four units of cis-protected Pd(II), five nitrate anions (four with full occupancy and one with half occupancy, which is disordered by symmetry), and one crown ether molecule coordinated to a Cs atom disordered over three positions (with occupancies of 0.15, 0.15 and 0.21). In the case of 10, the asymmetric unit contains half the complex, which includes two units of the ligand L coordinated to four units of cis-protected Pd(II), six nitrate anions (five with full occupancy and one with half occupancy, which is disordered by symmetry), and one crown ether molecule coordinated to a Cs atom disordered over two positions (with occupancies of 0.15 and 0.08). The crystal structures of 8, 9 and 10 confirm the inclusion of the caesium cation in the cavity of the cationic molecular cuboid; the electrostatic repulsion between the cationic entities is minimized by the three-dimensional screening of the Cs⁺ ion by the two crown ether units (see Fig. 8). It is relevant here to look again at the architecture of the free cage (H1)¹⁶⁺ through its crystal structure (Fig. 2) and cartoon depiction (Fig. 4). In the architecture of (H1)¹⁶⁺, there are two square-shaped open windows, one at the top and the other at the bottom of the cuboid-cage, each of which is demarcated by the four Pd(II) units. Additionally, there are four narrow slits located along the edges of the cuboid, vertically located between the square windows, with each being demarcated by two units of Pd(II). There are four phenylene units from the two units of encapsulated dibenzo crown ether molecules per cuboid cage, as shown in the crystal structures of 8, 9 and 10. In all cases, these four phenylene units are located in the four narrow slits of the cuboid, with one in each slit. The ethylene-oxide units of the crown ether are located at the core of the cuboid, and the oxygen atoms are pointed approximately towards the centre of the cuboid, interacting with the deeply confined Cs⁺ ion. The overall structures of the encapsulated [(Cs)⊂(crown ether)₂]⁺ complexes are shown in a simplified manner in Fig. 8(B) for clarity, although there is disorder in the crystal structures.

Fig. 8 (A) Crystal structures of 8, 9 and 10 showing the super-host framework as a wireframe and the [Cs⊂(crown ether)₂]⁺ as a space filling model. (B) Chemical drawings showing the relative orientation of the pair of encapsulated crown ether rings wrapped around Cs⁺, i.e., the [Cs⊂(crown ether)₂]⁺ moieties as observed in the crystal structures of 8, 9 and 10.

2.8 Adaptive deformation of the super-host H_out

The stimulus-induced adaptive deformation of coordination architectures, in line with enzymatic behaviour, is a subject of contemporary interest.^48–50 A selective stimulus-induced adaptive deformation of the molecular cuboid super-host is observed in this work, where the unique stimulus is a specific host–guest complex, i.e., [(Cs)⊂(H6)₂]⁺. This finding can be described by comparing the initial structure of (H1)¹⁶⁺ and the deformed structure [{(Cs)⊂(H6)₂}⊂(H1)]¹⁷⁺ as depicted in Fig. 9 in cartoon format.

Fig. 9 Cartoon representations for comparison of the architectures of (A and B) cuboid-shaped [H1]¹⁶⁺ and (C and D) the deformed cuboid in [{Cs⊂(H6)₂}⊂H1]¹⁷⁺ as viewed from the side and top.

The dimensions of both the square-shaped open windows of the Pd₈L₄ type cuboid (H1)¹⁶⁺ described by the four co-planar metal centres are 13.40 × 13.40 Å, with bisecting diagonals of 18.95 Å. The remaining four ligand-panelled walls of (H1)¹⁶⁺ are rectangular-shaped and are described by the four metal centres, having dimensions of 13.40 × 11.78 Å with bisecting diagonals of 17.84 Å in each case. The cuboid architecture of (H1)¹⁶⁺ became deformed upon the encapsulation of [(Cs)⊂(H6)₂]⁺ to form the cationic part of the compound 8; the four metal centres, which are not co-planar, of both the open windows described a puckered rhombus with arms of 13.33 × 13.34 Å and non-bisecting diagonals of unequal lengths, namely, 21.89 and 14.04 Å. The adjacent bond angles of the rhombus are 110° and 63°, respectively. Further, the pair of puckered open windows are located in a staggered configuration with respect to each other. The four panelled walls of compound 8 are also mildly puckered, having rectangular dimensions of 13.33/13.34 × 12.25 Å and non-intersecting diagonals of 17.63 and 17.73 Å, respectively. However, no deformation was observed due to uptake of either [(Cs)⊂(H7)₂]⁺ or [(Cs)⊂(H8)₂]⁺ by (H1)¹⁶⁺ to form the cationic parts of 9 and 10, respectively.

2.9 Structure of [(Cs)⊂(H6)₂]⁺ in native and encapsulated forms

A comparison of the solid-state structure of [(Cs)⊂(H6)₂]⁺ in its native form⁴³ with that in its encapsulated form revealed interesting conformational details, as shown in Fig. 10. The caesium atom is sandwiched between two H6 molecules in each case, and the crown ether is not flat but bowl-shaped. In the native state, one unit of H6 provides its convex face and the other provides its concave face for coordination with caesium (Fig. 10(A)). However, in the encapsulated state, both crown ether units provide their convex face for binding caesium (Fig. 10(B)). The structural change in [(Cs)⊂(H6)₂]⁺ after being encapsulated as a guest by H_out is due to π⋯π and C–H⋯π interactions between the protons of H6 and the aromatic rings of the ligand in H1. The planes of the two phenylene rings of a crown ether unit in the native and encapsulated states are disposed in a divergent manner, subtending dihedral angles of ∼115° and 90°, respectively. This observation suggests that the molecular cuboid induces bending of the guest (H6) by ∼25° to fit into the cavity.

Fig. 10 Structural comparison of [Cs⊂(H6)₂]⁺ when (A) exists in its native state (reported work) and (B) encapsulated in the molecular cuboid (this work).

3 Conclusions

In summary, we have successfully synthesized a new water-soluble template-free polycationic molecular cuboid. The molecular cuboid (H_out) encapsulates selected crown ether molecules (H_in) in its nano-confinement, providing a family of host-in-host type complexes of the type [(H_in)₂⊂(H_out)]. Preferential binding of certain crown ethers over others is demonstrated in these complexes. Encapsulation of the caesium ion (G) at the centre of the host-in-host complex resulted in [{(G)⊂(H_in)₂}⊂(H_out)] type molecules in which a cationic guest is accommodated in a cationic host due to the screening effect provided by the crown ether units. From another view-point, host–guest complexes of suitable shapes/sizes are recognized in larger host molecules. This finding encourages the binding study of small-sized hosts, and host–guest complexes in the cavity of large-sized host molecules to prepare ensembles that are expected to exhibit a variety of novel behaviours.

Author contributions

D. K. C. and M. R. conceptualized the idea, carried out the research, analysed the data and wrote the manuscript. S. K. carried out X-ray diffraction experiments and refined the crystal structures and contributed to the manuscript preparation. D. K. C. is the principal investigator and managed the project.

Conflicts of interest

There are no conflicts to declare.

Data availability

CCDC 2423199–2423201 and 2423283 contain the supplementary crystallographic data for this paper.^51a–d

All relevant data has been included in the manuscript and SI. See DOI: https://doi.org/10.1039/d5sc03202d.

Acknowledgements

D. K. C. thanks the Anusandhan National Research Foundation (ANRF) (formerly Science and Engineering Research Board (SERB)), Government of India (Project no. CRG/2022/004413), for financial support. D. K. C. thanks IIT Madras for financial support through a Mid-Career Institute Research and Development Award (IRDA-2019) and through a center under Institute of Eminence program (IoE Center of Molecular Architecture). We thank SAIF, IIT Madras for the single crystal X-ray diffraction facility. We also thank the Department of Chemistry, IIT Madras for the NMR facility and DST-FIST funded ESI-MS facility. We thank Dr R. Baskar, Department of Chemistry, IIT Madras for DOSY NMR measurements.

References

1. G. W. Gokel, W. M. Leevy and M. E. Weber, Chem. Rev., 2004, 104, 2723–2750.
2. G. Montà-González, F. Sancenón, R. Martínez-Máñez and V. Martí-Centelles, Chem. Rev., 2022, 122, 13636–13708.
3. E. G. Percástegui, T. K. Ronson and J. R. Nitschke, Chem. Rev., 2020, 120, 13480–13544.
4. H. Takezawa and M. Fujita, Bull. Chem. Soc. Jpn., 2021, 94, 2351–2369.
5. R. Banerjee, D. Chakraborty and P. S. Mukherjee, J. Am. Chem. Soc., 2023, 145, 7692–7711.
6. Y. Sun, C. Chen, J. Liu and P. J. Stang, Chem. Rev., 2020, 49, 3889–3919.
7. M. Yoshizawa and L. Catti, Acc. Chem. Res., 2019, 52, 2392–2404.
8. C. J. T. Cox, J. Hale, P. Molinska and J. E. M. Lewis, Chem. Soc. Rev., 2024, 53, 10380–10408.
9. L. Escobar and P. Ballester, Chem. Rev., 2021, 121, 2445–2514.
10. K. Lizuka, H. Takezawa and M. Fujita, J. Am. Chem. Soc., 2023, 145, 25971–25975.
11. K. Lizuka, H. Takezawa and M. Fujita, J. Am. Chem. Soc., 2024, 146, 32311–32316.
12. L. Neira, C. Alvariño, O. Domarco, B. Blanco, C. Peinador, M. D. García and J. M. Quintela, Chem.–Eur. J., 2019, 25, 14834–14842.
13. Y.-H. Huang, Y.-L. Lu, Z.-M. Cao, X.-D. Zhang, C.-H. Liu, H.-S. Xu and C.-Y. Su, J. Am. Chem. Soc., 2024, 146, 21677–21688.
14. A. I. Day, R. J. Blanch, A. P. Arnold, S. Lorenzo, G. R. Lewis and I. A. Dance, Angew. Chem., Int. Ed., 2002, 41, 275–277.
15. T. Kawase, Y. Nishiyama, K. Nakamura, T. Ebi, K. Matsumato, H. Kurata and M. Oda, Angew. Chem., Int. Ed., 2007, 46, 1086–1088.
16. Y. Xia, Y. Jiang, X.-L. Ni, Q. Wang and D. A. Wang, Chin. Chem. Lett., 2024, 35, 109782.
17. T. Kawase, K. Tanaka, N. Shiono, Y. Seirai and M. Oda, Angew. Chem., Int. Ed., 2004, 43, 1722–1724.
18. H. Ueno, T. Nishihara, Y. Segawa and K. Itami, Angew. Chem., Int. Ed., 2015, 54, 3707–3711.
19. S.-Y. Kim, I.-S. Jung, E. Lee, J. Kim, S. Sakamoto, K. Yamaguchi and K. Kim, Angew. Chem., Int. Ed., 2001, 40, 2119–2121.
20. W. Gong, X. Yang, P. Y. Zavalij, L. Isaacs, Z. Zhao and S. Liu, Chem.–Eur. J., 2016, 22, 17612–17618.
21. K. Cai, M. C. Lipke, Z. Liu, J. Nelson, T. Cheng, Y. Shi, C. Cheng, D. Shen, J.-M. Han, S. Vemuri, Y. Feng, C. L. Stern, W. A. Goddard III, M. R. Wasielewski and J. F. Stoddart, Nat. Commun., 2018, 9, 5275.
22. S. Guo, L. Liu, F. Su, H. Yang, G. Liu, Y. Fan, J. He, Z. Lian, X. Li, W. Guo, X. Chen and H. Jiang, JACS Au, 2024, 4, 402–410.
23. X.-Y. Pang, H. Zhou, X. Xie, W. Jiang, Y. Yang, J. L. Sessler and H.-Y. Gong, Angew. Chem., Int. Ed., 2024, 63, e202407805.
24. H. Danjo, Y. Hashimoto, Y. Kidena, A. Nogamine, K. Katagiri, M. Kawahata, T. Miyazawa and K. Yamaguchi, Org. Lett., 2015, 17, 2154–2157.
25. J. Jian, Chem. Res. Chin. Univ., 2022, 38, 407–409.
26. T. N. Parac, M. Scherer and K. N. Raymond, Angew. Chem., Int. Ed., 2000, 39, 1239–1242.
27. D. Zhang, T. K. Ranson, J. L. Greenfield, T. Brotin, P. Berthault, E. Léonce, J.-L. Zhu, L. Xu and J. R. Nitschke, J. Am. Chem. Soc., 2019, 141, 8339–8345.
28. T. Sawada, H. Hisada and M. Fujita, J. Am. Chem. Soc., 2014, 136, 4449–4451.
29. E. Ubasart, O. Borodin, C. Fuertes-Espinosa, Y. Xu, C. García-Simón, L. Gómez, J. Juanhuix, F. Gándaram, I. Imaz, D. Maspoch, M. von Delius and X. Ribas, Nat. Chem., 2021, 13, 420–427.
30. C. Falaise, S. Khlifi, P. Bauduin, P. Schmid, W. Shepard, A. A. Ivanov, M. N. Sokolov, M. A. Shestopalov, P. A. Abramov, S. Cordier, J. Marrot, M. Haousa and E. Cadot, Angew. Chem., Int. Ed., 2021, 60, 14146–14153.
31. K. Lizuka, H. Takezawa and M. Fujita, Angew. Chem., Int. Ed., 2025, 64, e202422143.
32. R. Saha, A. Devaraj, S. Bhattacharyya, S. Das, E. Zangrando and P. S. Mukherjee, J. Am. Chem. Soc., 2019, 141, 8638–8645.
33. I. A. Bhat, R. Jain, M. M. Siddique, D. K. Saini and P. S. Mukherjee, Inorg. Chem., 2017, 56, 5352–5360.
34. P. Howlader, B. Mondal, P. C. Purba, E. Zangrando and P. S. Mukherjee, J. Am. Chem. Soc., 2018, 140, 7952–7960.
35. Y. Yamanoi, Y. Sakamoto, T. Kusukawa, M. Fujita, S. Sakamoto and K. Yamaguchi, J. Am. Chem. Soc., 2001, 123, 980–981.
36. S. Sharma, M. Sarkar and D. K. Chand, Chem. Commun., 2023, 59, 535–554.
37. M. Ray, S. Krishnaswamy, A. K. Pradhan and D. K. Chand, Chem. Mater., 2023, 35, 6702–6712.
38. W. Liu, C. Bobbala, C. L. Stern, J. E. Hornick, Y. Liu, A. E. Enciso, E. A. Scott and J. F. Stoddart, J. Am. Chem. Soc., 2020, 142, 3165–3173.
39. C. J. Pedersen, J. Am. Chem. Soc., 1967, 89, 7017–7036.
40. M. Valente, S. F. Sousa, A. L. Magalhães, C. Freire and J. Solut, Chem, 2010, 39, 1230–1242.
41. G. G. Talanova, N. S. A. Elkarim, R. E. Hanes, H.-S. Hwang, R. D. Rogers and R. A. Bartsch, Anal. Chem., 1999, 71, 672–677.
42. J. Wang and S. Zhuang, Nucl. Eng. Technol., 2020, 52, 328–336.
43. T. Akutagawa, K. Shitagami, M. Aonuma, S.-I. Noro and T. Nakamura, Inorg. Chem., 2009, 48, 4454–4461.
44. A. N. Chekhlov, Russ. J. Coord. Chem., 2009, 35, 222–225.
45. C. Yan, Z. Li, X.-Q. Xiao, N. Wei, Q. Lu and M. Kira, Angew. Chem., Int. Ed., 2016, 55, 1484–1487.
46. M. Hu, X. Wang, Y. Jiang and S. Li, Inorg. Chem. Commun., 2007, 11, 85–88.
47. P. Thordarson, Chem. Soc. Rev., 2011, 40, 1305–1323.
48. H. Xu, T. K. Ronson, A. W. H. Heard, P. C. P. Teeuwen, L. Schneider, P. Pracht, J. D. Thoburn, D. J. Wales and J. R. Nitschke, Nat. Chem., 2025, 17, 289–296.
49. H. Hou, Z. Jiang, Q. Chen, Q.-F. Sun and M. Hong, CCS Chem., 2025, 7 DOI:10.31635/ccschem.025.202505454.
50. S. S. Mishra, S. Krishnaswamy and D. K. Chand, J. Am. Chem. Soc., 2024, 146, 4473–4488.
51. (a) M. Ray, S. Krishnaswamy and D. K. Chand, CCDC 2423199: Experimental Crystal Structure Determination, 2025,  DOI:10.5517/ccdc.csd.cc2mbjq7; (b) M. Ray, S. Krishnaswamy and D. K. Chand, CCDC 2423200: Experimental Crystal Structure Determination, 2025,  DOI:10.5517/ccdc.csd.cc2mbjr8; (c) M. Ray, S. Krishnaswamy and D. K. Chand, CCDC 2423201: Experimental Crystal Structure Determination, 2025,  DOI:10.5517/ccdc.csd.cc2mbjs9; (d) M. Ray, S. Krishnaswamy and D. K. Chand, CCDC 2423283: Experimental Crystal Structure Determination, 2025,  DOI:10.5517/ccdc.csd.cc2mbmf1.

---