Anion coordinated capsules and pseudocapsules of tripodal amide, urea and thiourea scaffolds

Abstract

This review aims to deliver a detailed and comparative account of the reported examples of anion (halide and oxyanions) coordinated capsules and pseudocapsules of tripodal receptors that employ hydrogen bonds and/or electrostatic hydrogen bonds offered by specific binding sites from amide, urea and thiourea functionalities. The review discusses both the structural aspects of anion binding and solution-state anion binding affinities of N-bridged and aryl-bridged tripodal receptors. Discussions relating to selective anion recognition and separation, carbondioxide uptake, and transmembrane anion transport, as demonstrated by some of these tripodal receptors have also been included in this review.

---

Sandeep Kumar Dey

Dr Sandeep Kumar Dey received his MSc degree in 2007 from North Eastern Hill University (NEHU) Shillong, India and PhD in Supramolecular Chemistry from Indian Institute of Technology Guwahati in 2013 under the supervision of Prof. Gopal Das. He then joined the group of Prof. Shih-Sheng Sun as a postdoctoral research fellow in Institute of Chemistry, Academia Sinica, Taiwan. Presently, he is an Alexander von Humboldt (AvH) research fellow working in the group of Prof. Dr Christoph Janiak at Heinrich-Heine-Universität Düsseldorf, Germany. His research interests are mainly in the field of supramolecular chemistry and functional porous materials for various applications.

Arghya Basu

Dr Arghya Basu received his MSc degree in 2008 from Indian Institute of Engineering Science and Technology, Shibpur, India. He obtained his PhD in Supramolecular Chemistry from Indian Institute of Technology Guwahati in 2014 under the supervision of Prof. Gopal Das. He then joined the group of Prof. Hiroyuki Furuta as a postdoctoral research fellow in Kyushu University, Japan (2014–2015). Presently, he is a DST Young Scientist, working in the group of Prof. Rahul Banerjee at CSIR-NCL Pune, India. His research interests are mainly in the field of supramolecular chemistry and π-conjugated systems for material applications.

Romen Chutia

Dr Romen Chutia received his BSc degree in 2008 from Sibsagar College affiliated to Dibrugarh University and MSc degree in 2010 from Gauhati University, India. He obtained his PhD in Supramolecular Chemistry from Indian Institute of Technology Guwahati in 2016 under the supervision of Prof. Gopal Das. Presently, he is a faculty member in the Department of Chemistry, Sibsagar College, and his research interests are mainly focused on supramolecular chemistry.

Gopal Das

Prof. Gopal Das received his MSc degree in 1993 from Calcutta University, India. He obtained his PhD from Indian Institute of Technology Kanpur in 1999 under the supervision of Prof. Parimal K. Bharadwaj. He then joined the group of Prof. Stefan Matile as a postdoctoral research associate in University of Geneva, Switzerland. In 2004, he joined the Indian Institute of Technology Guwahati as an Assistant Professor and from 2013 he is Full Professor. He has published about 150 research papers till date and his current research interests are in the field of supramolecular chemistry and development of sensors and biosensors.

---

Introduction

The significance of anion coordination chemistry has been increasingly appreciated in recent years because of the important roles that anions play in many biological, environmental, and chemical processes.^1–4 Over the past two decades, supramolecular chemistry has emerged to achieve recognition and sensing of anionic guests mostly in organic solvent media,^5–10 and also in competitive aqueous environments.^11–15 Indeed, cationic macrocyclic hosts had been employed in some early elegant works to achieve anion binding in water by electrostatic interactions.¹⁵ One of the most extended approaches towards the design of anion receptor is based on the use of –NH functions that interact through hydrogen bonds with the anionic guests. Along this line, neutral anion receptors containing multiple hydrogen bond donor groups, such as amide, urea/thiourea and pyrrole has been extensively used for the effective and selective binding of anions,^5–10 although deprotonation of the most acidic hydrogen bond donor occurs in some cases.^16–21 In 2005, Bowman-James et al. has categorized the binding of anions based on their coordination numbers which is helpful in defining the notions of complementarity for a given anion and can aid to the design of optimal anion binding host structures.²² At this juncture, it would not be erroneous to say that the development and systematic studies of numerous acyclic and macrocyclic anion receptors have at least partially solved the problem of selective anion binding and, perhaps more relevantly, the problem of discrimination between different anions. More recently, there has been a growing interest in the development of highly selective anion transporters^23–25 with the potential to become future therapeutics and anion induced formation of supramolecular self-assemblies and materials.^26–29 The concept of anion coordination chemistry has been proficiently employed to generate a range of exciting supramolecular structures such as catenanes, rotaxanes, foldamers, helices, and capsules.^30–35

Since the beginning of anion coordination chemistry, macrocyclic and acyclic podand type hosts functionalized with polarized –NH and –CH donors have widely been employed for anion binding and sensing. However, it has been realized that most of the inorganic anions have very high hydration energies that must be compensated for by the host molecule for effective anion recognition.³⁶ Thus, it is the design of sophisticated three dimensional receptors that is essential to fully encapsulate an anion by creating a highly specific anion binding cavity as observed in various anion binding proteins.^37,38 In comparison to macrocyclic hosts, podands are more readily synthesized, can display rapid complexation/decomplexation kinetics, may undergo significant conformational changes upon anion binding and thus, may form the basis of switchable sensing devices.³⁹ Along this line, numerous hydrogen bonding tripodal scaffolds have been developed to form stable host–guest complexes with anions of different geometries such as, spherical halides, planar carbonate/nitrate and tetrahedral sulfate/phosphate.^40–43 Tripodal receptors constitute a special class of acyclic ionophores with C_3v symmetry, whose side arms upon functionalization with appropriate anion binding elements can selectively recognize one or more anionic guest species via topological complementarity. The geometry and orientation of the receptor side arms in N-bridged and aryl-bridged tripodal scaffolds favour the formation of capsules or pseudocapsules upon anion encapsulation by multiple hydrogen bonding. Based on its selectivity, a tripodal receptor capsule can create a distinct microenvironment to isolate an anionic guest from the bulk of solvent media by encapsulation, and thereby, leads to the phenomenon of selective anion encapsulation when possible formation of different capsules are present in the same solution. Numerous reviews describing the coordination and supramolecular chemistry of anions have been published in recent years and over the last decade. However, reviews that discussed the anion induced capsular assembly formation are only few^{30–32,41,42} and require more attention to be paid. Anion induced capsules and pseudocapsules of tripodal scaffolds have shown a range of exciting properties such as, encapsulation of hydrated anion clusters, fixation of aerial carbon dioxide as carbonate, liquid–liquid extraction of anions from water, selective separation of anions by crystallization and transmembrane anion transportation. This review aims to deliver a comprehensive compilation of the examples reported till date related with the anion induced capsules and pseudocapsules of tripodal receptors that employ hydrogen bonds and/or electrostatic hydrogen bonds offered by specific binding sites from amide, urea, thiourea and hybrid urea–amide functionalities (Schemes 1–5). The review has been divided into four main sections based on the hydrogen bond donating functionality incorporated in the tripodal scaffold for anion coordination. The primary objective of the review is to provide a detailed and comparative account of the solid- and solution-states binding of anions (halides and oxyanions) within the cavity of hydrogen bonding tripodal scaffolds. Some of these tripodal receptors have been found to display selective anion recognition, aerial carbondioxide capture, and selective anion separation, which have also been discussed here. This review covers some of the significant research works from our group and from the groups of Ghosh et al., Custelcean et al., Gale et al., Wu et al., Steed et al., Sun et al., Gunnlaugsson et al. and Hossain et al.

Scheme 1 Structures of tripodal tris-amide receptors 1–12.

Scheme 2 Structures of tripodal tris-urea receptors 13–36.

Scheme 3 Structures of hexaurea receptors 37, 40–42, and pyridinium-based tris-urea receptors 38–39.

Scheme 4 Structures of tripodal tris-thiourea receptors 44–57.

Scheme 5 Structures of tripodal urea–amide hybrid receptors 58–65 and squaramide based tripodal receptors 66–68.

Design principles of tripodal receptors

Tripodal tris-amide receptors (Scheme 1) offer three –NH hydrogen bonds for the recognition of anionic guest. Tris(2-aminoethyl)amine (Tren)-based amide receptors (1–5), due to its small cavity size mostly recognize a spherical F⁻/Cl⁻ within its molecular cavity. In comparison to Tren-based receptors (1–5), aryl-bridged tris-amide receptors (7–10) provide a larger cavity size and thereby, can encapsulate even larger anions such as AcO⁻, NO₃⁻ and SiF₆²⁻ in a 1 : 1/2 : 2 or 2 : 1 host–guest stoichiometry. The cavity size of N-bridged tripodal amide receptors can be increased by certain structural modifications (6 and 11) to encapsulate larger anions such as, NO₃⁻, SiF₆²⁻ and even discrete halide–water clusters.

On the other hand, tripodal tris-urea/thiourea receptors (Schemes 2 and 4) offer six –NH hydrogen bonds and hence, can recognize anions of different geometry including spherical (F⁻/Cl⁻/Br⁻), trigonal planar (AcO⁻/NO₃⁻), trigonal pyramidal (SO₃²⁻) and tetrahedral (SO₄²⁻/PO₄³⁻) anions in a 1 : 1/2 : 2 or 2 : 1 host–guest stoichiometry. In general, the acidity of thiourea –NH proton is higher than that of urea (pK_a of 21.1 and 26.9 in DMSO, respectively),⁴⁴ and thus thiourea receptors are expected to establish stronger hydrogen bonds and form more stable complexes with anions than their urea analogues. However, if the –NH protons are acidic enough due to the incorporation of a highly electron-withdrawing substituent into the receptor framework; deprotonation may occur in the presence of a highly basic anion such as, F⁻ and HCO₃⁻. The use of thiourea groups offers additional advantages, including generally enhanced solubility in lower polarity solvents and a lower propensity to undergo self-association.

In 2005, theoretical investigation by Hay et al. showed that the incorporation of electron withdrawing substituents on the peripheral aryl function of tripodal receptors significantly enhance the stability of the anion complexes.⁴⁵ Notably, a nitro group at the meta-position of the peripheral aryl function increases the acidity of the ortho-CH protons to participate in anion binding. The lipophilicity of the receptors can be significantly enhanced by incorporation of pyridyl or fluorinated aryl functions to the tripodal scaffold, to achieve anion recognition in aqueous or semi-aqueous medium. Attachment of peripheral pyridyl functions offers additional advantage of metal coordination, and hence, anion complexation can be achieved with metal salts in aqueous or semi-aqueous media. The lipophilicity of the receptors can also be increased by functionalization of each receptor side-arm with cationic moieties such as pyridinium functions. Further, due to the charged nature of the pyridinium ring, the –CH protons subsequent to the pyridinium nitrogen can act as potent hydrogen bond donor to the encapsulated anion.

Besides, a few tripodal hexaurea receptors have also been synthesized based on the concepts of complementarity and hydrophobic effect. The ortho-phenylene bridged tripodal hexaurea receptors (40–42) due to its tetrahedral cone conformation and 12 –NH hydrogen bond donors are ideal for tetrahedral oxyanion encapsulation.

Tripodal amide receptors

In the early nineties, Beer et al. has first reported the anion recognition properties of tripodal tris-amide receptors 1a–b (Scheme 1) having redox-active ferrocene and cobaltocene as the electrochemical signaling unit.^46–48 Cyclic voltammetry and ¹H NMR experiments revealed that the combinations of positively charged metallocene units together with amide groups are the essential components for the electrochemical anion recognition effects. Nonetheless, 1a could detect H₂PO₄⁻ by large cathodic shifts of up to 240 mV in the presence of a tenfold excess of Cl⁻ and HSO₄⁻ ions in acetonitrile. During the same period, Reinhoudt et al. reported the anion binding properties of a series of neutral tripodal tris-amide receptors 2a–d (Scheme 1) by ¹H NMR and conductometric experiments.⁴⁹ The association constants for the anion complexation of 2a–d revealed a preferential binding to dihydrogenphosphate (H₂PO₄⁻ > Cl⁻ > HSO₄⁻), and the highest binding affinity was observed for 2a (Table 1). They rationalized that the simple structure of the tripodal scaffold offers the possibility for synthetic manipulation to achieve higher association constants and anion selectivity. Since then, a range of N-bridged and aryl-bridged tripodal amide receptors have been developed to accomplish selective anion recognition, although selectivity has been achieved only in few cases. For example, a systematic study with a set of three positional isomers of nitro-phenyl functionalized tris-amide receptor 3a–c (Scheme 1) showed that the ortho-isomer (3a) is capable of selective fluoride recognition, as revealed from ¹H NMR titration experiments (DMSO-d₆) with different halides.⁵⁰ The ortho-isomer experienced a large downfield shift of amide –NH resonance (Δδ = 1.05 ppm) upon titration with F⁻, and yielded an association constant (log K) of 5.63 for a 1 : 1 stoichiometry. However, the other two isomers (3b and 3c) were found to recognize both fluoride and chloride anions, with greater selectivity for fluoride (Table 1). In recent years, studies employing tripodal receptors generally highlight the structural aspects of anion binding along with detailed solution-state anion binding studies mostly using ¹H NMR spectroscopy. Tris(2-aminoethyl)amine (Tren) based tris-amide receptors demonstrated the halide induced formation of monomeric capsules (semi-capsules) in its neutral form, whereas, aryl-bridged tris-amide receptors have been established to form either monomeric or dimeric capsules depending on the nature of the anion and/or terminal aryl function.

Table 1 Binding constants (log K) of amide receptors with different anions

Receptors	F^−	Cl^−	Br^−	HSO4^−	H2PO4^−
2a	—	3.24	—	2.23	3.78
3b	3.76	3.32	—	—	—
3c	4.06	2.29	—	—	—
5a	3.35	3.26	1.56	—	—
5b	3.94	2.87	2.66	—	—
9	1.59	1.58	1.56	1.96	0.87
10	4.86	3.83	2.97	—	—

---

Monomeric capsules (semi-capsules) of tripodal amide receptors

In our work with tripodal amide receptors, we have shown that the dinitrophenyl-functionalized tris-amide receptor 4 (Scheme 1) serves as an efficient fluoride sensor with characteristic solvent dependent absorptions in the optical spectrum.⁵¹ Single crystals of fluoride complex [TBA(4·F)] were obtained from a library of polar aprotic solvents viz. acetonitrile (MeCN), tetrahydrofuran (THF), dimethylformamide (DMF), dimethylsulfoxide (DMSO) and acetone (Me₂CO) as solvates. Solvatomorphs of [TBA(4·F)] showed the encapsulation of a fluoride anion within the receptor cavity by six strong hydrogen bonds involving the amide –NH protons and three aryl –CH protons, irrespective of the solvent of crystallization (Fig. 1a). In solution state, the π-acidic cavity gives rise to intense colorimetric changes (colourless to red/blue) upon fluoride recognition, due to strong anion–π charge transfer interactions. ¹H NMR analyses of the isolated fluoride complexes showed a significant downfield shift of the amide –NH and ortho-CH resonances (Δδ ∼3.50 and ∼0.70 ppm, respectively), suggesting a strong solution-state binding of fluoride. The high fluoride selectivity of 4 (log K > 7.0) has been employed in the transformation of charged anion complexes [(H4)⁺A⁻] (A⁻ = Cl⁻, Br⁻, ClO₄⁻, HSO₄⁻) into [TBA(4·F)], as evidenced in ¹H NMR titration experiments (DMSO-d₆).⁵² It is to be noted that, attempted crystallization of a 1 : 1 KF complex in solvents such as, DMSO and DMF yielded the respective solvates of 4.⁵³ However, it is possible to crystallize the KF complex from an acetonitrile media with the composition [4·KF(H₂O)₂], where a hydrated KF contact ion-pair is coordinated to the receptor molecule(s) outside the tripodal cavity.⁵³ In the ¹H NMR spectrum of KF complex (CD₃CN/DMSO-d₆), the amide –NH resonance could not be observed and the ortho-CH resonance showed a comparatively minor downfield shift of ∼0.30 ppm in comparison to [TBA(4·F)]. Thus, a binding discrepancy of fluoride in quaternary ammonium (TBAF) and alkali (KF) salts has been demonstrated both in solid and solution-states by tris-amide receptor 4.

Fig. 1 X-ray structures of monomeric capsules, (a) [4·(F⁻)], and (b) [5a·(F⁻·H₂O)]. TBA cations are omitted for clarity. Colour code: C = green, N = blue, O = red, F = chartreuse yellow.

Another interesting example of fluoride binding within a tripodal tris-amide scaffold has been established with a pentafluorophenyl-based receptor, 5a reported by Ghosh et al.^54,55 Solvent sealed monomeric F⁻ capsule [TBA(5a·F)S] (S = H₂O/CHCl₃), can either be obtained from an aqueous or chloroform solution. Whereas, solvent sealed monomeric Cl⁻ capsule [TBA(5a·Cl)CHCl₃], can only be obtained from a chloroform solution and an aqueous solution give non-solvated complex [TBA(5a·Cl)]. In the solvent sealed monomeric capsules, a solvent molecule (H₂O/CHCl₃) is coordinated to the encapsulated halide anion (Fig. 1b). Notably, 5a together with 18-crown-6-ether has efficiently been employed in the liquid–liquid extraction (H₂O/CHCl₃) of KF and KCl from aqueous solution by a dual host approach.

Sun et al. has demonstrated the reversible encapsulation of nitrate anion within the preorganized cavity of a series of N-bridged tripodal amide receptors 6a–6k (Scheme 1), via solvent polarity controlled formation of dimeric receptor capsules.^56,57 The structural scaffold responsible for the capsule formation was prepared by in situ nitration of the N-terminal aromatic ring of the tripodal receptors bearing methyl (6a) and methoxy (6b) substituents at para-position. The studies showed that the nitro group at the ortho-position of the N-terminal aromatic ring is mandatory for hydrogen bonded capsule formation and the mode of hydrogen bonding depends upon the electronic nature of the N-terminal aromatic ring. In the crystal structure of a nitrate-encapsulated complex [(H6a)⁺NO₃⁻], the amide –NH protons form strong hydrogen bonds with the nitrate anion (Fig. 2) and, the etheric oxygen atoms are hydrogen bonded to the proton attached to the bridgehead N-atom. They have also demonstrated that the incorporation of dansyl groups at the periphery of the tripodal amide scaffold resulted in a nitrate selective fluorescent probe (6g) in an aqueous DMSO solution.⁵⁸ They rationalized that the ionization of sulfonic acid in the dansyl moiety assists the transformation of the fluorescent receptor into its zwitterion form by protonation of the bridgehead nitrogen and thereby, attain a cone shape conformation through intramolecular hydrogen bonding.

Fig. 2 X-ray structure of complex [(H6a)⁺(NO₃⁻)]. Colour code: C = green, N = blue, O = red.

Ghosh et al. and coworkers have further explored the anion recognition properties of two positional isomers of an aryl-bridged tris-amide receptor having ortho-nitrophenyl (7) and para-nitrophenyl (8) peripheral functions (Scheme 1).^59,60 The aryl-platform provides a larger preorganized cavity in comparison to the Tren-based receptors and thereby, enables the encapsulation of hydrated anions. The ortho-isomer 7 showed encapsulation of a monohydrated chloride/acetate within the tripodal cavity in complexes [TBA(7·A)H₂O] (A = Cl⁻/AcO⁻), while monotopic recognition of anion has been observed in the nitrate complex [TBA(7·NO₃)] (Fig. 3).⁵⁹

Fig. 3 X-ray structures of monomeric capsules, (a) [7·(AcO⁻·H₂O)], and (b) [7·(NO₃⁻)]. Colour code: C = green, N = blue, O = red. TBA cations are not shown.

The same group has also reported the detailed solution-state isothermal calorimetry (ITC) studies of a pentafluorophenyl-based tris-amide receptor 10 with halides and oxyanions.⁶¹ Fluoride, chloride, bromide, acetate and benzoate showed an exothermic binding profile with 1 : 1 host–guest stoichiometry. The binding of bromide and acetate is equally facilitated by entropy and enthalpy factors, whereas chloride and benzoate binding is strongly enthalpy driven. Structures of halide complexes [TBA(10·X)] (X = F⁻/Cl⁻), has further confirmed the 1 : 1 binding via encapsulation of an anion within the tripodal cavity by 3 –NH hydrogen bonds.

A cationic tripodal receptor 12, based on amide-pyridinium as recognition sites and nitrobenzene as signaling unit showed high selectivity and strong binding affinity for AcO⁻ (log K > 5.0) over other anions.⁶² UV-Vis and ¹H NMR experiments indicated that the selectivity can be attributed to synergistic effects arising from hydrogen bonding, electrostatic interactions and conformational change.

Dimeric capsules of tripodal amide receptors

To gain further insight into the anion binding properties of aryl-bridged tris-amide scaffold, the para-isomer 8 was crystallized in presence of fluoride, chloride, acetate, and nitrate, to obtain the respective complexes from dioxane solutions.⁶⁰ Fluoride complex [TBA(8·F)(H₂O)₃], showed the formation of a dimeric capsular assembly via encapsulation of a fluoride–water cluster [F₂(H₂O)₆]²⁻ (Fig. 4a). Similarly, chloride and acetate complexes showed encapsulation of [A₂(H₂O)₄]²⁻ (A = Cl⁻/AcO⁻) anion–water cluster within the dimeric capsular assembly of 8. Each capsular assembly is stabilized by several receptor–anion, receptor–water and anion–water interactions. Nitrate complex [TBA(8·NO₃)], also showed the formation of a dimeric capsule (Fig. 4b). The capsular size of acetate and nitrate complexes (10.24 Å and 10.02 Å, respectively) is significantly larger than the fluoride and chloride complexes (9.45 Å and 9.56 Å, respectively).

Fig. 4 X-ray structures of dimeric capsules, (a) [(8)₂·(F⁻)₂(H₂O)₆], (b) [(8)₂·(NO₃⁻)₂]. Colour code: C = green/gray, N = blue, O = red, F = chartreuse yellow. TBA cations are not shown.

Although receptor 7 did not showcase the formation of dimeric capsules upon complexation with anions such as, chloride, acetate and nitrate, however, hydrated fluoride ion [F₂(H₂O)₆]²⁻ induced formation of dimeric capsule has been observed in complex [TBA(7·F)(H₂O)₃].⁵⁹ Thus, simple variation in the position of the electron withdrawing nitro-substituent in the aryl terminals of the tripodal amide receptor showed substantial alterations in the anion/hydrated anion induced capsular assembly formation. The ¹H NMR titration experiments (DMSO-d₆) with AcO⁻ showed 1 : 3 host–guest binding for receptor 7 and 8, whereas the single-crystal structure showed 1 : 1 complex.

Another aryl-bridged tris-amide receptor with a pyridyl terminal 9 (Scheme 1) has recently been reported by the same group, which showed encapsulation of hydrated halides [X₂(H₂O)₄]²⁻ (X = F⁻/Cl⁻) within its dimeric capsular assembly when crystallized from 1 : 1 (v/v) aqueous acetone solution.⁶³ On the other hand, attempted complexation with fluoride in 1 : 1 (v/v) dioxane–acetone solvent system resulted in the formation of [SiF₆(H₂O)₂]²⁻ encapsulated dimeric assembly. The binding constants calculated from the ¹H NMR titration experiments with halides and oxyanions revealed that the receptor shows a comparable binding constant with halides (log K ∼ 1.56), whereas NO₃⁻ and HSO₄⁻ show slightly higher values (log K ∼ 2.0) (Table 1).

Recently, we have shown the encapsulation of a cyclic halide–water tetramer [X₂(H₂O)₂]²⁻ (X = Br⁻/Cl⁻) inside the dimeric capsular assembly of a new N-bridged tripodal amide receptor 11 (Scheme 1) with para-nitrophenyl terminals.⁶⁴ The encapsulated halide–water tetramer in complexes [(H11)·X·H₂O] is hydrogen bonded to both the receptor cations of the dimeric capsule of size ∼20.60 Å (Fig. 5a). It is important to mention that the etheric oxygen atoms are hydrogen bonded to the proton attached to the bridgehead-N of receptor cation, which eventually organizes the flexible receptor arms in one direction and drives the formation of capsular assembly. However, the receptor cation prefers to adopt a non-capsular polymeric aggregation in the presence of higher homologous iodide, which confirms the anion selective conformational adaptability of the molecule.

Fig. 5 X-ray structures of dimeric capsules, (a) [(H11)₂⁺·(Br⁻·H₂O)₂], and (b) [(H11)₂⁺·(SiF₆)²⁻]. Colour code: C = green, N = blue, O = red, Br = brown, F = chartreuse yellow.

Further studies with receptor 11 have revealed the solvent dependent encapsulation of an octahedral hexafluorosilicate anion within its dimeric capsular assembly in the solid-state.⁶⁵ Two hexafluorosilicate complexes [(H11)₂(SiF₆)] and [(H11)₂(SiF₆)·2DMF·4H₂O] were obtained upon reaction of 11 with HF in DMSO and DMF respectively, presumably as a result of glass corrosion. In the non-solvated complex crystallized from DMSO, two protonated receptor molecules create a dimeric capsular assembly of 19.33 Å to encapsulate a hexafluorosilicate anion by multiple hydrogen bonds (Fig. 5b). Whereas, in the solvated complex crystallized from DMF, a hexafluorosilicate anion is involved in side-cleft binding with four receptor cations, outside the receptor cavity. Thus, by a mere change of crystallizing solvent, we have been able to isolate two structurally different SiF₆²⁻ complexes of receptor 11. It is important to mention that the capsular size of [(H11)₂(SiF₆)] complex is less by ca. 1.30 Å than the [(H11)·X·H₂O] (X = Cl⁻/Br⁻) complexes.

Tripodal urea receptors

After the seminal papers by Wilcox⁶⁶ and Hamilton⁶⁷ on urea–anion interactions, a variety of anion receptors have been reported in which one or more urea/thiourea fragments are incorporated in an acyclic or macrocyclic framework. In 1995, Morán et al. has demonstrated that the Tren-based tris-phenylurea receptor 13 (Scheme 2) and its thiourea analogue 44 (Scheme 4) can efficiently bind tetrahedral oxyanions due to topological complementarity and directionality of hydrogen bonding.⁶⁸ Based on the ¹H NMR experiments, he suggested a 1 : 1 complex formation upon binding of a dihydrogenphosphate anion within the tripodal cavity via six hydrogen bonds. Based on similar ideas, Wu et al. has reported the recognition of tetrahedral oxyanions (H₂PO₄⁻ and HSO₄⁻) by complexation-prompted fluorescence enhancement of the Tren-based tris-naphthylurea receptor 14 (Scheme 2) in DMF solution.⁶⁹ He has also reported the thiourea analogue 45 (Scheme 4) of the naphthylurea receptor, which showed substantial absorption spectral changes upon complexation with H₂PO₄⁻ and HSO₄⁻ in DMF solutions.⁷⁰ The association constants obtained by fluorescence/UV-Vis titrations showed a higher selectivity for the H₂PO₄⁻ anion with a 1 : 1 complex stoichiometry for both the receptors.^69,70 However, no structural data were available to support the host–guest complementarity and the shape recognition for these tetrahedral oxyanions. Nonetheless, following these solution-state anion binding studies and theoretical calculations by Hay et al.,⁴⁵ a number of tripodal urea and thiourea receptors mostly with electron withdrawing aryl terminals have been developed for the recognition and separation of anionic guests, particularly sulfate and phosphate. Most of these studies highlight the structural aspects of anion binding with the tripodal urea/thiourea receptors, which were then related to the observed selectivity in solution. Oxyanions such as, CO₃²⁻, SO₄²⁻ and HPO₄²⁻ induced formation of dimeric capsular assembly and halide induced formation of monomeric capsules have commonly been observed in most of the structural reports on anion complexes of tripodal urea/thiourea receptors. However, oxyanion binding within the unimolecular receptor cavity has also been perceived in certain cases.

Dimeric capsules of tripodal urea receptors

Tren-based tris-urea receptor 15 with peripheral 3-cyanophenyl group (Scheme 2) has been recognized as a versatile anion receptor accounting for the selective separation of SO₄²⁻ over other anions via selective crystallization of Ag⁺ coordinated framework,⁷¹ encapsulation of staggered conformer of divalent oxalate anion,⁷² and uptake of aerial CO₂ as CO₃²⁻ complex that efficiently extracts sulfate, thiosulfate and chromate from water by anion exchange metathesis.⁷³

Complexation of 15 with 0.5 equivalents of Ag₂SO₄ in water/acetone binary solvent mixture yielded a coordination polymer of capsular aggregate having the composition [Ag₂SO₄(15)₂](Me₂CO)_1.5(H₂O)_3.7, reported by Custelcean et al.⁷¹ Two receptor molecules that encapsulate a SO₄²⁻ anion are held together by CN–Ag⁺ and C=O–Ag⁺ coordinative bonds to form a rigidified capsule of 9.81 Å size. Attempted complexation of 15 with other soluble silver salts having anions of different geometry and basicity such as, AgBF₄, AgNO₃, AgMeSO₃ and AgMeCO₂ failed to produce any coordination complexes and yielded crystals of 15 in each case.

The same cyanophenyl-based receptor 15 has also been utilized by Ghosh et al. to capture the staggered conformer of divalent oxalate anion within its dimeric capsular assembly of 9.81 Å size (Fig. 6a).⁷² They have further showed that the planar conformer of the divalent oxalate anion can be captured with the fluorophenyl-based receptor 19. Encapsulation of planar form of the anion resulted in an increase of capsular size by ∼1.0 Å due to the more C–C single bond character of planar C₂O₄²⁻ than its staggered conformer.

Fig. 6 X-ray structures of oxyanion encapsulated dimeric assemblies, (a) [(15)₂·(C₂O₄)²⁻], (b) [(15)₂·(CO₃)²⁻], (c) [(15)₂·(SO₄)²⁻], and (d) [(15)₂·(CrO₄)²⁻]. Colour code: C = green/gray, N = blue, O = red, S = yellow, Cr = purple. TBA cations are not shown.

Thus, recognition of different conformers of C₂O₄²⁻ inside the dimeric capsule of two different tripodal receptors has been accomplished by simple tuning of the peripheral aromatic substituents. ¹H NMR titration data showed that 15 has slightly higher binding affinity towards C₂O₄²⁻ (log K = 4.82) compared to 19 (log K = 4.29).

However, the most interesting property of 15 is its ability to capture atmospheric CO₂ as CO₃²⁻ within its dimeric capsular assembly. The aerial CO₂ uptake is facilitated in a DMSO solution of the receptor containing equivalent amount of (TBA)OH, which eventually form diamonded crystals of CO₃²⁻ encapsulated complex [(TBA)₂CO₃(15)₂] (Fig. 6b) in about 85% yield.⁷³ Because of its excellent solubility in chloroform and dichloromethane, [(TBA)₂CO₃(15)₂] has been efficiently employed for the extraction of SO₄²⁻, S₂O₃²⁻ and CrO₄²⁻ from water by anion-exchange metathesis.⁷³

Quantification by weighing the bulk extract shows that the CO₃²⁻ capsule can separately extract ca. 90% of the above three anions from aqueous solutions. In each case, the bulk extracted solid was crystallized from DMSO to obtain 2 : 1 host–guest capsular complexes [(TBA)₂SO₄(15)₂] (Fig. 6c), [(TBA)₂S₂O₃(15)₂] and [(TBA)₂CrO₄(15)₂] (Fig. 6d), of identical capsular size (9.61–9.71 Å). The extraction of SO₄²⁻ ions from water has also been demonstrated under alkaline conditions (pH 12.5) and in the presence of excess nitrate ions.

Tris(3-pyridylurea) receptor 17 upon complexation with Mg²⁺ salts of divalent oxyanions in aqueous methanol solutions, self-assemble into crystalline capsules with composition [Mg(H₂O)₆·(17)₂A] (A = SO₄²⁻, SeO₄²⁻, CO₃²⁻, SO₃²⁻), as reported by Custelcean et al.⁷⁴ Competitive crystallization experiments in the presence of these anions established the anion selectivity order, SO₄²⁻ > SeO₄²⁻ > CO₃²⁻ > SO₃²⁻ which is different from both the Hofmeister bias and anion basicity scale, whereas the trend was in agreement with the lattice energy calculations on the crystal structures. The lower selectivity of the [Mg(H₂O)₆]²⁺ bridged capsule for planar CO₃²⁻ and pyramidal SO₃²⁻ was attributed to the mismatch of the anion binding cavity with these anions. Oxyanion encapsulation within the urea-lined dimeric cavity in a 2 : 1 host–guest stoichiometry is characteristic to the solid-state, as only 1 : 1 complexation could be detected in solution. The selective separation of SO₄²⁻ from an aqueous solution containing excess of NO₃⁻ has also been achieved with this capsule. However, the capsule has limited utility for sulfate separation from alkaline nuclear wastes, as they do not form under basic conditions (pH > 10), due to Mg(OH)₂ precipitation. Structurally similar [M(H₂O)₆]²⁺ (M = Mn/Zn) bridged dimeric SO₄²⁻ capsules have also been reported by Wu and Janiak et al.⁷⁵

It has been realized that, the oxyanion binding cavity in [Mg(H₂O)₆]²⁺ bridged capsule is large enough to accommodate either SO₄²⁻ (Fig. 7a) or SeO₄²⁻, which resulted in poor selectivity for sulfate against selenate. Thus, to improve the selectivity for SO₄²⁻ against SeO₄²⁻, the anion binding cavity has been fine-tuned by substituting cationic [Mg(H₂O)₆]²⁺ subunits with [Li(H₂O)]⁺, which yielded a coordination capsule [Li₂SO₄(17)₂(H₂O)₂] (Fig. 7b) with substantially reduced cavity size (capsule size drops by ∼0.50 Å).⁷⁶ Such allosteric regulation of the cavity size by the judicious choice of external bridging cation has allowed for the effective separation of SO₄²⁻ over SeO₄²⁻ by competitive crystallization, with an observed selectivity exceeding that of sulfate-binding protein. Thus, Custelcean et al. has been able to establish both shape recognition of SO₄²⁻ against SO₃²⁻ and CO₃²⁻, and size recognition against SeO₄²⁻ using tripodal urea receptor.

Fig. 7 X-ray structures of sulfate encapsulated metal complexes, (a) [Mg(SO₄)(H₂O)₆·(17)₂], (b) [Li₂(SO₄)(H₂O)₂(17)₂]. Colour code: C = green, N = blue, O = red, S = yellow, Mg = pink, Li = purple.

Custelcean et al. has also demonstrated sulfate separation from a highly competitive aqueous alkaline solution (∼6 M Na⁺, pH = 14) by selective crystallization of sodium-based coordination capsule [Na₂SO₄(17)₂(H₂O)₄] and thereby, provided for the first time a viable approach to sulfate separation from nuclear wastes.⁷⁷ The alkali metal coordination capsules are held together by coordination and hydrogen bonding water bridges, with a SO₄²⁻ anion encapsulated inside urea-lined dimeric cavity.

Very recently, Wu et al. has reported the tris(4-pyridylurea) receptor 18 and its self-assembly with various metal chloride salts.⁷⁸ The receptor has shown the formation of 2 : 2 dimeric capsules with chloride salts of alkali (Na/K) and alkaline earth metals (Mg/Ca), where two Cl⁻ encapsulated semi-capsule formed a staggered dimeric capsule by hydrogen bonding interactions between the peripheral pyridyl nitrogen and water molecules of the cationic [M₂(H₂O)₆]²⁺/[M(H₂O)₆]²⁺ subunit. However, formation of 2 : 1 host–guest dimeric capsules were observed with carbonate salts of alkali metals (Na/K). The sizes of the (Cl⁻)₂ encapsulated coordination capsules of 18 (11.18–11.33 Å) are significantly larger than that of oxyanion (CO₃²⁻/SO₄²⁻) encapsulated coordination capsules of 17 and 18 (9.20–9.96 Å).

A pentafluorophenyl-substituted tris-urea receptor 21 has been extensively studied by Ghosh et al. and established as a versatile receptor for anions of different geometry.^79–83 Combined ¹H NMR, isothermal titration calorimetry (ITC) and single crystal X-ray diffraction studies revealed that monovalent anions such as, F⁻, OH⁻ and H₂PO₄⁻ formed 1 : 1 complexes, and divalent oxyanions such as, CO₃²⁻, SO₄²⁻ and HPO₄²⁻ formed 2 : 1 host–guest capsular complexes. It is interesting to note that, a DMSO solution of 21 upon treatment with (TBA)CN formed crystals of hydroxide-encapsulated complex [TBA(21·OH)].⁷⁹ Two hydroxide-encapsulated receptor units are held together by intermolecular π–π interactions to form a pseudo-dimeric capsule of 14.95 Å. It is to be noted that, no crystals were formed from a DMSO solution of 21 and (TBA)CN in anaerobic condition, suggesting that atmospheric moisture must have a role in the formation of OH⁻ complex. Pseudo-dimeric capsule formation has also been observed in fluoride complex [TBA(21·F)] via halogen bonding interactions between the pentafluorophenyl rings.⁸⁰ In dihydrogenphosphate complex [TBA(21·H₂PO₄)], a (H₂PO₄⁻)₂ dimer is encapsulated within a dimeric assembly of 13.79 Å, demonstrating the formation of 2 : 2 host–guest capsular complex in the solid-state (Fig. 8a).⁸¹ However, hydrogenphosphate complex [(TBA)₂HPO₄(21)₂], obtained from a DMSO solution of the receptor containing TBA salts of H₂PO₄⁻ and OH⁻ ions, showed encapsulation of a HPO₄²⁻ anion within a dimeric assembly of 9.92 Å (Fig. 8b).⁷⁹ Thus, the receptor has been shown to encapsulate both monovalent and divalent phosphate anion(s) within dimeric capsular assemblies of different sizes. Arsenate-encapsulated complex [(TBA)₂HAsO₄(21)₂] having capsular size 10.22 Å, was obtained from a receptor solution containing 2 equiv. of TBAI and excess sodium hydrogen arsenate in a 5% H₂O–DMSO solvent system.⁸² Thus, the authors were able to furnish the first structural evidence of arsenate encapsulation by a tripodal urea receptor.

Fig. 8 X-ray structures of phosphate encapsulated dimeric assemblies, (a) [(21)₂·(H₂PO₄⁻)₂], and (b) [(21)₂·(HPO₄)²⁻]. Colour code: C = green/gray, N = blue, O = red, P = orange, F = chartreuse yellow. TBA cations are not shown.

The most exciting property of 21 is its ability to capture atmospheric CO₂ as CO₃²⁻ encapsulated complex [(TBA)₂CO₃(21)₂] in almost quantitative yield (∼98%), from a DMSO solution of the receptor containing equivalent amount of (TBA)OH.⁸³ The disassembly of the CO₃²⁻ capsule and recovery of the free receptor (∼85%) has been accomplished by simply treating the complex with aqueous methanol. As expected, no crystals were formed from a DMSO solution of 21 and (TBA)OH under anaerobic condition in a glove box.

The carbonate complex [(TBA)₂CO₃(21)₂] has been efficiently employed in the liquid–liquid (chloroform–water) extraction of SO₄²⁻ from water via an anion exchange process. Almost quantitative and clean extraction of SO₄²⁻ from water (99% from extracted pure mass and >95% shown by gravimetrically) has been achieved by this process.⁷⁹ Selective SO₄²⁻ extraction from water, using CO₃²⁻ capsule has also been demonstrated in the presence of other competitive oxyanions such as, H₂PO₄⁻ and NO₃⁻. The composition of the extracted mass was confirmed by ¹H NMR and single crystal X-ray diffraction, which showed sulfate-encapsulated dimeric assembly [(TBA)₂SO₄(21)₂] of 9.18 Å. It is to be noted that, the size of the capsular assembly did not change upon exchange of tetrahedral SO₄²⁻ for trigonal-planar CO₃²⁻, which perhaps plays a key role in the facile extraction of SO₄²⁻ due to topological complementarity with the urea-functionalized anion-binding cavity. However, an impure SO₄²⁻ extraction has been observed when an organic layer containing an equivalent mixture of 21 and (TBA)Cl (phase transfer agent) was used as an extractant, further validating the significance of persistent dimeric capsule in SO₄²⁻ extraction and separation.

Detailed ¹H NMR titration experiments of 21 with halides and oxyanions in DMSO-d₆ solutions yielded the anion selectivity order as: H₂PO₄⁻ > SO₄²⁻ > CH₃COO⁻ > F⁻ > Cl⁻ (Table 2). Similarly, isothermal titration calorimetric (ITC) studies in dry DMSO solutions showed the highest binding affinity towards H₂PO₄⁻ and the anion binding affinity follows the order: H₂PO₄⁻ > SO₄²⁻ ∼ F⁻ > CH₃COO⁻ > Cl⁻, which is nearly in agreement with the ¹H NMR titration results.^80,81 The anion binding affinity of the receptor has also been studied in aqueous DMSO (DMSO-d₆/D₂O, 9 : 1 v/v) by performing ¹H NMR titrations with 1 : 1 D₂O/DMSO-d₆ solutions of various anions as their sodium salts.⁸² A 1 : 1 binding stoichiometry has been observed with all investigated oxyanions e.g. HAsO₄²⁻, HPO₄²⁻, SO₄²⁻, CO₃²⁻, and showed the highest affinity for HAsO₄²⁻ (log K = 4.42) followed by H₂PO₄²⁻ (log K = 3.62) > SO₄²⁻ (log K = 3.48) > CO₃²⁻ (log K = 2.68).

Table 2 Binding constants (log K) of urea receptors with different anions

Receptors	F^−	Cl^−	H2PO4^−	SO4^2−	AcO^−
a K determined by ^1H NMR titration experiments.b K determined by UV-Vis/fluorescence titration experiments.
13^a	—	2.94	4.04	>4.0	—
14^b	—	—	4.60	3.43	—
16^a	4.51	3.09	4.20	4.70	—
20^a	—	2.75	2.65	>4.0	—
21^a	4.06	3.42	5.52	4.73	4.45
22^a	—	2.60	2.38	>4.0	—
25^a	3.87	—	4.38	4.78	3.27
26^a	—	—	4.41	4.97	3.21
27^b	5.31	4.28	4.62	6.70	4.95
33^a	—	3.38	4.39	5.74	3.41

---

Gale et al. has reported the transmembrane anion transport properties of a series of fluorinated tripodal tris-urea receptors 20–23 (Scheme 2) and their thiourea analogues 46–49 (Scheme 4), along with the detailed solid- and solution-state anion binding studies.⁸⁴ Based on the experimental results obtained, he has suggested that increasing the degree of fluorination increases the lipophilicity of the tripodal receptors and this has shown to be the major contributing factor in the superior transport activity of the fluorinated compounds, with a maximum transport rate achieved with compound 23. Vesicle anion transport assays using ion-selective electrodes showed that, this class of fluorinated urea/thiourea compounds are capable of transporting chloride through a lipid bilayer via a variety of mechanisms, including Cl⁻/H⁺ cotransport and Cl⁻/NO₃⁻, Cl⁻/HCO₃⁻ antiport processes. The ¹H NMR studies revealed that the fluorinated receptors can bind anions in DMSO-d₆ solutions according to the trend SO₄²⁻ > H₂PO₄⁻ > Cl⁻ > HCO₃⁻ ≫ NO₃⁻ (Table 2), with various deprotonation events occurring upon titration with H₂PO₄⁻ and HCO₃⁻. Receptor 23 showed the formation of a 2 : 1 host–guest capsular complex [(TEA)₂CO₃(23)₂] of size 9.16 Å, when crystallized in presence of (TEA)HCO₃.

Recently, a set of three positional isomers of nitro-phenyl functionalized tris-urea receptor (24–26) has been extensively studied by our group, towards recognition of various oxyanions.^85–87 The para-isomer 26 has previously been shown to form a large capsular complex with SO₄²⁻ via encapsulation of a rugby ball shaped (SO₄²⁻·3H₂O·SO₄²⁻) adduct within a dimeric receptor capsule assembled by electrostatic nitro–nitro interactions (Fig. 9a).⁸⁸ In our studies, we have shown that the para-isomer can encapsulate a carbonate or a divalent terephthalate anion in a 2 : 1 host–guest stoichiometry.⁸⁵ The terephthalate complex [(TBA)₂C₈H₄O₄(26)₂] was obtained from a DMSO solution of 26 containing excess of 1 : 2 mixture of terephthalic acid and (TBA)OH, while carbonate complex [(TEA)₂CO₃(26)₂] was obtained from a DMSO solution of 26 containing excess of (TEA)HCO₃. In both the complexes, a total of 14 –NH hydrogen bonds stabilize an anion within the respective capsular assemblies (Fig. 9b). Interestingly, the para-isomer in the presence of excess TBA(H₂PO₄) self-assembled into a tetrahedral molecular cage by encapsulation of a tetrameric tetrahedral mixed phosphate cluster (H₂PO₄⁻·HPO₄²⁻)₂ via 24 –NH hydrogen bonds (Fig. 10a).⁸⁷ In the tetrahedral cage complex [(26)₄(H₂PO₄⁻·HPO₄²⁻)₂], each of the four receptor units coordinate to a phosphate anion (H₂PO₄⁻/HPO₄²⁻) axially by six –NH hydrogen bonds donated from the three urea functions (Fig. 10b). Further, the fluoride complex [TEA(26·F)] of the para-isomer showed the formation of a self-assembled (2 + 2) pseudodimeric cage with 1 : 1 host : guest stoichiometry.

Fig. 9 X-ray structures of oxyanion encapsulated dimeric assemblies, (a) [(26)₂·[(SO₄²⁻)₂·(H₂O)₃], (b) [(26)₂·(C₈H₄O₄)²⁻]. C = green/gray, N = blue, O = red, S = yellow. TBA cations are not shown.

Fig. 10 X-ray structures of tetrahedral cage complex [(26)₄(H₂PO₄⁻·HPO₄²⁻)₂], (a) tetrameric (H₂PO₄⁻·HPO₄²⁻)₂ encapsulated tetrahedral cage of 26, and (b) hydrogen bonding interactions between (H₂PO₄⁻·HPO₄²⁻)₂ cluster and a receptor unit. C = green/gray/yellow/pink, N = blue, O = red, P = orange. TBA cations are not shown.

Unlike 1 : 1 complexation of SO₄²⁻ by the para-isomer 26, the meta-isomer 25 can encapsulate SO₄²⁻ in a 2 : 1 host–guest stoichiometry to form a dimeric capsule [(TBA)₂SO₄(25)₂].⁸⁶ However, the meta-isomer is incapable to self-assemble into a dimeric capsule in presence of terephthalate dianion, and adopts a flat trigonal planar like geometry where each receptor side arm is in interaction with a carboxylate group of the dianion in complex [(TBA)₃(C₈H₄O₄)_1.5(25)]. The most important property of the meta-isomer is its ability to capture atmospheric CO₂ as CO₃²⁻ to form capsular complex [(TBA)₂CO₃(25)₂], triggered by the presence of (TBA)OH/(TBA)F in a DMSO solution.⁸⁶ It is remarkable to note that above 90% yield of CO₃²⁻ encapsulated complex can be obtained from the basic (OH⁻) solution mixture after 3 days of exposure to an unmodified atmosphere. Structural comparison revealed an increase of 1.0 Å in the capsular size of carbonate complex of the meta- isomer (9.0 Å) as compared to its para-analogue (8.0 Å), although both the capsular assemblies has the same symmetry and encapsulate a CO₃²⁻ with an array of 14 –NH hydrogen bonds, which is indeed a consequence of positional isomeric effect. The effect of positional isomerism has also been observed for sulfate binding by the isomeric receptors, where the para-isomer in presence of SO₄²⁻ formed a large dimeric capsule of 15.77 Å engulfing a (SO₄²⁻·3H₂O·SO₄²⁻) adduct, whereas the meta-isomer formed a SO₄²⁻ encapsulated dimeric assembly of 9.60 Å. Above all, the terephthalate complexes of the isomeric receptors revealed the greatest evidence of positional isomerism, where the para-isomer formed a dimeric capsular assembly of 12.25 Å and the meta-isomer formed a terephthalate coordinated polymeric assembly.

The meta-isomer has also been shown to encapsulate HPO₄²⁻ within a dimeric assembly of 9.86 Å, assembled by π–π interactions of the nitrophenyl aromatics.⁸⁵ The complex [(TBA)₂HPO₄(25)₂] was obtained from a DMSO solution of the receptor containing equal amounts of excess (TBA)H₂PO₄ and (TBA)OH, since crystallization of 25 in the presence of excess (TBA)H₂PO₄ was not fruitful. However, successful crystallization of the ortho-isomer 24, in the presence of TBA salts of different oxyanions was not fruitful presumably due to the steric effect provided by the nitro groups at the ortho-position, which hinders the facile inclusion and coordination of an anion due to electrostatic factor, as confirmed by 2D NOESY NMR analysis of the free receptor.

Recently, we have reported a tripodal heteroditopic tris-urea receptor 29 which is capable of both anion and cation recognition.⁸⁹ The receptor in the presence of (TEA)HCO₃ self-assemble into a molecular capsule having small cavity size of 3.97 Å via intermolecular CH–π interactions and carbonate coordination (Fig. 11a).

Fig. 11 X-ray structures of complexes (a) [29·(TEA)₂CO₃] and (b) [29·K₂HPO₄]. Colour code: C = green/gray, N = blue, O = red, P = orange, K = purple.

Interestingly, the receptor when treated with K₂HPO₄ in a 2 : 1 DMSO-H₂O solution, afforded crystals of K⁺ coordinated centrosymmetric capsule assembled by weak π–π interactions of the phenyl rings, and resulted in three C₂-symmetric urea-lined clefts for the recognition of HPO₄²⁻ anion (Fig. 11b). The binding constants calculated from ¹H NMR titration experiments with different oxyanions showed the anion selectivity order as HCO₃⁻ (log K = 2.88) > H₂PO₄⁻ (log K = 2.77) > HSO₄⁻ (log K = 2.07) for 1 : 1 complexation.

The tripodal urea receptors 30–33 on a cyanuric acid-platform has been shown to display different modes of sulfate coordination by Ghosh et al.^90–92 SO₄²⁻ encapsulation within dimeric assembly of 30 and 31 were observed in the presence of MgSO₄ and (TBA)₂SO₄ respectively. Interestingly, 30 in the presence of ZnSO₄·7H₂O formed a 3D coordination polymer [(30)₃Zn₃(SO₄)₄][Zn(H₂O)₆][(H₂O)₁₈], the secondary building unit (SBU) of which constitutes four Zn²⁺ coordinated receptor units in a bowl shaped conformation encapsulating a sulfate–water [(SO₄)₄(H₂O)₁₂]⁸⁻ cluster in its cavity (Fig. 12a).⁹⁰ Binding constant calculation yielded log K value of 5.47 for 30 with ZnSO₄ in DMSO-d₆. In contrast, receptors 32 and 33 showed a 1 : 1 binding with SO₄²⁻, which will be discussed in the next section.^91,92

Fig. 12 X-ray structures of (a) the secondary building unit (SBU) of [(30)₃Zn₃(SO₄)₄][Zn(H₂O)₆][(H₂O)₁₈], (b) tetrahedral cage complex [(37)₄(PO₄³⁻)₄]. Colour code: C = green/gray/yellow/pink, N = blue, O = red, P = orange.

Davis et al. has recently reported three new families of tripodal urea receptors, 34, 35 and 36 (Scheme 2) and their thiourea analogues (55, 56 and 57, Scheme 4) for transmembrane anion transport studies using vesicle anion transport assays.^93,94 For biologically relevant anion transport the major target is chloride, so the binding properties of these receptors were assessed for chloride (Bu₄N⁺Cl⁻) by ¹H NMR titrations in DMSO-d₆. It was found that, the use of more electron-withdrawing aryl groups increases the chloride affinities of all three series, while the thioureas are nearly twice as powerful as the ureas. The Cl⁻ binding affinity of the urea receptors are typically in the range of log K 1.40–2.60. All of these results will be discussed in detail under the “thiourea receptor” section.

Wu et al. has reported an interesting example of self-assembly where a hexaurea receptor 37 (Scheme 3) in presence of PO₄³⁻ self-assemble into a tetrahedral anion cage [(37)₄(PO₄)₄]¹²⁻ held together by 48 hydrogen bonds (Fig. 12b).⁹⁵ In the complex, the phosphate ions occupy the vertices of a tetrahedron and the ligands lie on the faces with an estimated internal volume of 181 Å and PO₄³⁻–PO₄³⁻ distance of 15 Å. The complex is stable up to 80 °C and can persist in different aprotic solvents such as, acetone, acetonitrile and DMSO.

Monomeric capsules (semi-capsules) of tripodal urea receptors

The para-isomer of the cyanophenyl-based tris-urea receptor 16 has been extensively studied by Hossain et al. for halide and oxyanion binding using ¹H NMR spectroscopy, X-ray structure analyses of the protonated receptor complexes and density functional theory (DFT) calculations.⁹⁶ The ¹H NMR titration studies revealed that the receptor forms 1 : 1 complexes with all studied anions, showing a binding trend in the order of F⁻ > H₂PO₄⁻ > Cl⁻ ∼ HSO₄⁻ > Br⁻ (Table 2). DFT calculations showed formation of monomeric halide capsules, revealing the highest binding energy for the fluoride complex. Likewise, the para-isomer of the fluorophenyl-based tris-urea receptor 20 has revealed the formation of monomeric chloride capsule [TBA(20·Cl)].⁸⁴

In the oxyanion-encapsulated metal (M = Mg/Mn/Zn) complexes of tris(3-pyridylurea) receptor 17, it has been observed that the pyridyl groups of the tripodal receptor do not coordinate with the metal ion, but interact with the hydrated metal cation [M(H₂O)₆]²⁺ through hydrogen bonding.^74,75 These results implied that the oxyanion capsules may also be generated via second-sphere coordination, i.e. by employing metal complexes of hydrogen bond donating ligands having free counter anion(s). Based on this concept, Wu, Janiak and Yang et al. has employed [Fe(DABP)₃]SO₄ (DABP = 5,5′-diamino-2,2′-bipyridine) and [Fe(BP)₃]SO₄ (BP = 2,2′-bipyridine) complexes to self-assemble with 17.⁹⁷ However, crystal structures of both the isolated complexes [Fe(DABP)₃][17·SO₄]·10H₂O and [Fe(BP)₃][17·SO₄]·9H₂O, showed formation of monomeric sulfate capsules (Fig. 13a).

Fig. 13 X-ray structures of monomeric capsules, (a) [17·(SO₄)²⁻], and (b) [27·(TBA)(SO₄)]⁻. C = green/gray, N = blue, O = red, S = yellow, Fe = orange.

The encapsulation of sulfate by a single receptor molecule rather than two is rationalized by the better crystal packing of the cationic units [Fe(DABP)₃]²⁺ or [Fe(BP)₃]²⁺ without sufficient secondary hydrogen bond donors and anion-encapsulated receptor units.

Integration of the well-defined tripodal urea scaffold and the redox-active ferrocene group has resulted in an electrochemical sensor 27 for SO₄²⁻ and H₂PO₄⁻ anions, as demonstrated by Wu and Yang et al.⁹⁸ Large cathodic shifts of both the reduction and oxidation peaks were observed after the addition of 1.0 equivalents of SO₄²⁻ (ΔE_pc = −190 and ΔE_pa = −150 mV) to a 1.0 mM chloroform solution containing 0.10 M (TBA)PF₆ salt. Sulfate complex [(TBA)₂SO₄(27)H₂O], showed the formation of a cation (TBA) sealed semi-capsule, where the TBA cations provide –CH hydrogen bonds to the encapsulated SO₄²⁻ anion (Fig. 13b). Whereas, fluoride complex [TBA(27·F)] showed the formation of semi-capsule. Detailed ¹H NMR experiments showed that the binding affinity of different anions follows the order SO₄²⁻ > F⁻ > AcO⁻ ∼ H₂PO₄⁻ > HSO₄⁻ ∼ Cl⁻ > Br⁻, in accordance with the UV/Vis titration profiles (Table 2). The receptor has been found to selectively bind a SO₄²⁻ anion (log K = 6.70) even in the presence of highly competitive fluoride ions (log K = 5.31), as confirmed by competitive ¹H NMR experiments. Furthermore, a quinoline-based tripodal tris-urea receptor 28 in conjunction with the ferrocene-based receptor 27 has been employed to control intermolecular SO₄²⁻ transfer between the two receptors driven by an electrochemical stimulus, which is detected by the change of the fluorescence intensity before and after electrochemical oxidation of the ferrocene groups.⁹⁹

Ghosh et al. has shown the solid- and solution-state evidences of temperature dependent formation of a cation (TBA) sealed semi-capsule [(TBA)₂SO₄(33)], employing a pentafluorophenyl-based urea receptor 33 on a cyanuric acid platform (Fig. 14).⁹² The solution-state recognition of (TBA)₂SO₄ was made evident from the diffusion coefficient values observed for the receptor and (TBA)₂SO₄ from the ¹H DOSY NMR analyses at variable temperatures. The authors suggested that the higher binding affinity of 33 with SO₄²⁻ (log K = 5.74) compared to H₂PO₄⁻ (log K = 4.39) could be due to the existence of ion–pair interaction without the loss of electrostatic energy arising from charge separation.

Fig. 14 X-ray structure of cation sealed sulfate capsule, [(TBA)₂SO₄(33)]. C = green/gray, N = blue, O = red, S = yellow, F = chartreuse yellow.

Steed et al. has reported two sets of cationic tripodal tris-urea receptors on an aryl platform, each arm of which is functionalized with both urea and pyridinium functions.^100,101 The first set of receptors 38a and 38b (Scheme 3) containing p-tolyl and n-octyl substituents, displayed significant binding affinity towards halides (Cl⁻ and Br⁻) and oxyanions (NO₃⁻, AcO⁻ and H₂PO₄⁻) with the formation of pyridinium C–H⋯A⁻ interactions despite the presence of six urea –NH donors, as evidenced in ¹H NMR titration experiments.¹⁰⁰ Density functional theory (DFT) calculations suggested that, in the lowest energy conformation of [(38a)Cl]²⁺, the pyridinium–urea receptor formed a monomeric capsule that encapsulate a chloride ion in its cavity by three C–H⋯Cl⁻ and three N–H⋯Cl⁻ hydrogen bonds, and is sealed at the periphery by C–H⋯π interactions between the p-tolyl rings (Fig. 15a). Halide binding with a –NH and a –CH proton from each receptor side arm rather than the urea –NH protons has also been supported by the magnitude of chemical shift changes in the ¹H NMR experiments in DMSO-d₆. Whereas, in the lowest energy conformation of [38b·Cl]²⁺ complex, a less symmetrical “2-up, 1-down” conformation is favoured where a chloride ion is bound to urea functions of the two receptor side arms that are oriented upwards, and this has further been supported by ¹H and variable temperature (VT) NMR experiments in CD₃CN.

Fig. 15 Chloride-coordinated monomeric capsules of pyridinium-based tris-urea receptors 38a and 39a, as obtained from DFT optimization studies.

The second set of receptors 39a and 39b (Scheme 3) has been designed with two well-separated binding compartments on the same aryl platform, the bottom one comprising a 3-aminopyridinium pocket and the upper one involves three remote urea groups.¹⁰¹ DFT optimization and ¹H NMR experiments (CD₃CN) suggested that the 3-aminopyridinium pocket can bind strongly to a chloride ion via charge assisted N–H⋯Cl⁻ and C–H⋯Cl⁻ hydrogen bonding, and the upper urea groups zip-up the anion bound monomeric capsule by N–H⋯O=C tape motif (Fig. 15b). The sharpening of the downfield shifted pyridinium –CH resonance upon addition of one equivalent of chloride suggested that the anion has a role in locking-in the 3-up conformation, working synergically with the intramolecular urea hydrogen bonding. The authors rationalized that in less polar solvents, the tendency of the urea groups to self-associate favoured a more preorganized receptor molecule for anion binding, whereas in more polar media, less preorganization may result due to poor self-association between the urea groups.

In their pioneering work on tripodal anion receptors, Wu and Li et al. has demonstrated the highly efficient extraction of sulfate from an aqueous to an organic phase (chloroform) by employing a nitrophenyl-substituted tripodal hexaurea receptor 40 (Scheme 3) to obtain complex [(TBA)₂(40·SO₄)DMSO].¹⁰² The receptor is capable of encapsulating a sulfate anion within a complementary tetrahedral cage that is protected by aromatic rings, and represents a successful strategy for overcoming the “Hofmeister bias” by taking advantage of a combination of complementarity, the chelate effect, and the hydrophobic effect. This is the first example of a receptor that satisfies the optimal coordination of 12 hydrogen bonds for a sulfate anion by encapsulation within its complementary molecular cavity with each urea group chelating an edge of the tetrahedral sulfate (Fig. 16a). The association constant (log K) estimated from the ¹H NMR titration data (DMSO-d₆) was found to be larger than 4.0, even in the presence of 25% D₂O. The selective extraction of sulfate from a nitrate rich aqueous solution was achieved by thoroughly mixing with a equimolar solution of 40 and (TBA)Cl in CDCl₃ and immediate analysis of the organic phase to ascertain the quantitative formation of the sulfate-encapsulated complex.

Fig. 16 X-ray structures of monomeric capsules, (a) [40·(SO₄)²⁻], and (b) [41·(CO₃)²⁻]. C = green/gray, N = blue, O = red, S = yellow, F = chartreuse yellow. Countercations are not shown.

A pentafluorophenyl-substituted tripodal hexaurea receptor 41 has recently been reported by Hossain et al. for the absorption of atmospheric CO₂ as CO₃²⁻, induced by the presence of excess TBAF in DMSO solution to obtain complex [(TBA)₂(41·CO₃)] (Fig. 16b).¹⁰³ The ¹H NMR titration data with (TEA)HCO₃ yielded a binding constant (log K) of 3.25 for a 1 : 1 (host/guest) binding model. Besides, the receptor showed strong affinity for HSO₄⁻ with an association constant (log K) of 3.35, presumably due to receptor-anion complementarity as established by Wu and Li et al.

Wu et al. has recently reported a set of three ferrocenyl-functionalized tripodal hexaurea receptors with ortho-, meta- and para-phenylene bridges for the bis-urea arms, designed for anion binding and electrochemical signalling purposes.¹⁰⁴ In sulfate complex [(TBA)₂(42·SO₄)2H₂O], the ortho-phenylene bridged hexaurea receptor 42 adopts a folded cone conformation for optimal coordination of sulfate by 12 –NH hydrogen bonds. Interestingly, the meta-phenylene bridged receptor 43 can encapsulate two SO₄²⁻ anions in its “inner” and “outer” tripodal clefts, as evidenced by distinct stepwise 1 : 1 and 1 : 2 host/guest binding in the ¹H NMR titration experiments (DMSO-d₆), and was also supported by molecular modelling studies (Fig. 17). Such a two-step binding process is facilitated by the specific trigonal bipyramidal conformation of the meta-phenylene bridged receptor, which features two clefts rather than one tetrahedral cage as observed in ortho-phenylene bridged analogue. The association constants for the first and second step sulfate coordination were evaluated to be log K >4.0 and >2.0, respectively. Molecular modelling studies revealed that the para-phenylene bridged receptor forms a cavity that is too large for the second sulfate coordination, which may rationalize its weaker binding than the meta-analogue.

Fig. 17 1 : 2 host–guest capsule of hexaurea receptor 43 with sulfate anion.

Tripodal thiourea receptors

Several Tren-based tris-thiourea receptors (Scheme 4) mostly analogous to the above discussed urea receptors have been synthesized, and their anion binding affinity and stoichiometry have been assessed both in solution and solid-states. It is interesting to note that, a few of these thiourea receptors mainly 46, 53 and 54 behaved very differently towards their anion binding affinity in solution-state and also solid-state host–guest complex stoichiometry, as compared to their urea analogues. Thus, the increased acidity of the thiourea –NH protons has a significant role to play in influencing the anion binding affinity and complexation, which is different from their urea analogues.

Dimeric capsules of tripodal thiourea receptors

As stated in the previous section, Gale et al. and coworkers have extensively studied the solid- and solution-state anion binding properties of a series of fluorinated tripodal tris-urea receptors 20–23 (Scheme 2) and their thiourea analogues 46–49 (Scheme 4).⁸⁴ Among the fluorinated thiourea receptors, the anion complexes of the pentafluorophenyl-based receptor 46 have been more elaborately studied in the solid-state than the others. The authors were able to crystallize the anion complexes of 46 with Cl⁻, NO₃⁻, SO₄²⁻ and H₂PO₄⁻ as their quaternary ammonium salts.⁸⁴ The receptor has shown the formation of 1 : 1 host–guest complexes with spherical chloride, trigonal-planar nitrate and tetrahedral sulfate anions. Attempted crystallization of a bicarbonate complex has failed due to the mono-deprotonation of one of the thiourea –NH groups, presumably because HCO₃⁻ functioned as a Brønsted base. However, such deprotonation event has not been observed in the pentafluorophenyl-based urea analogue 21 even in the presence of (TBA)OH, in which case a CO₃²⁻ complex was obtained by aerial CO₂ fixation.⁸³ Complexation with H₂PO₄⁻, revealed the formation of a 2 : 2 host–guest stoichiometric capsule, where a (H₂PO₄⁻)₂ dimer is encapsulated within a dimeric assembly of 13.44 Å, similar to the urea analogue 21.

Sulfate and hydrogenphosphate complexes of the fluorophenyl-substituted thiourea receptor 47 showed the formation of 2 : 1 host–guest capsular complexes,⁸⁴ identical to the urea receptors 21 and 25. Hydrogenphosphate complexes [(TBA)₂HPO₄(47)₂] and [(TBA)₂HPO₄(49)₂] (Fig. 18a) with 2 : 1 host–guest stoichiometry were obtained in the presence of excess (TBA)H₂PO₄ indicating the deprotonation of H₂PO₄⁻ anion in the solution-state while crystallization, as evidenced from ¹H NMR titration experiments (DMSO-d₆). In the ¹H NMR titrations of 48 and 49 with (TBA)H₂PO₄, a combination of fast and slow exchange processes has been observed in wet DMSO-d₆ (0.5/10% D₂O), similar to urea compound 23. This behaviour was explained due to the deprotonation of bound H₂PO₄⁻, and subsequent formation of a HPO₄²⁻ complex.

Fig. 18 X-ray structures of oxyanion encapsulated dimeric assemblies, (a) [(49)₂·(HPO₄)²⁻], and (b) [(46)₂·(NO₃⁻)₂]. C = green/gray, N = blue, O = red, P = orange, F = chartreuse yellow. Countercations are not shown.

A nitrate complex [(TBA)(46·NO₃)] of thiourea receptor has also been obtained in the solid state, although no solution-state nitrate binding has been observed with 46–49.⁸⁴ The complex revealed the formation of a (2 + 2) pseudo-dimeric capsule, where two nitrate encapsulated units are held together via anion–π interactions (Fig. 18b). Formation of (2 + 2) pseudo-dimeric capsules has also been observed in the F⁻ complexes of urea receptor 21 and 26.

The encapsulation of divalent tetrahedral oxyanion via deprotonation of its monovalent state has also been observed in the solid-state by employing the para-tolyl and para-methoxyphenyl-substituted thiourea receptors 50 and 51.¹⁰⁵ Solution-state evidences of the deprotonation events have largely been provided by the tolyl-substituted receptor 50 with HCO₃⁻, HSO₄⁻ and H₂PO₄⁻ anions in DMSO-d₆. Both these receptors formed 2 : 1 host–guest SO₄²⁻ capsules of identical size (∼9.90 Å), when crystallized in the presence of excess (TBA)HSO₄ in DMSO. Similarly, crystallization of 51 in the presence of excess (TBA)H₂PO₄ in DMSO yielded a 2 : 1 host–guest capsular complex with HPO₄²⁻. The same complex has also been obtained from a DMSO solution of the receptor mixed with K₂HPO₄ and TBAI. Capsule size of HPO₄²⁻ complexes of 47, 49 and 51 are almost same, 10.02–10.12 Å (distance between the bridgehead-N atoms of dimeric assembly).

Sulfate and thiosulfate complexes of a chlorophenyl-substituted tris-thiourea receptor 52 also showed the formation of 2 : 1 host–guest capsular complexes,¹⁰⁶ identical to the urea receptor 15. Thiosulfate complex [(TEA)₂S₂O₃(52)₂] (Fig. 19b) was obtained from an DMSO–H₂O (∼2 : 1 v/v) solution of 52 containing an excess (10 equiv.) of equimolar quantities of (TEA)Cl and Na₂S₂O₃, and sulfate complex [(TBA)₂SO₄(52)₂] was obtained from a DMSO solution containing 52 and excess (TBA)₂SO₄ or (TBA)HSO₄. The capsular size of the thiosulfate complex (10.12 Å) is slightly larger (∼0.8 Å) than the sulfate complex (9.33 Å), which may be attributed to the larger size and lower charge density of the S₂O₃²⁻ as compared to SO₄²⁻. Considering a 1 : 1 binding model, the association constants (log K) were determined to be 4.54 and 3.35 for SO₄²⁻ and S₂O₃²⁻ respectively, using ¹H NMR titration data.

Fig. 19 X-ray structures of oxyanion encapsulated dimeric assemblies, (a) [(53)₂·(PO₄)³⁻], and (b) [(52)₂·(S₂O₃)²⁻]; C = green/gray, N = blue, O = red, S = yellow, P = orange. Countercations are not shown.

We have further established that the para-nitrophenyl-substituted tris-thiourea receptor 53 functions as a competent hydrogen bonding scaffold that can selectively encapsulate a trivalent phosphate anion (PO₄³⁻) within the persistent dimeric capsules assembled by aromatic π-stacking interactions between the receptor side-arms.^107,108 An acetonitrile (MeCN) solution of 53 in the presence of excess (TBA)H₂PO₄ crystallized into 2 : 1 host–guest capsular complex [(TBA)₃PO₄(53)₂·2MeCN], where a PO₄³⁻ anion is encapsulated by 12 –NH hydrogen bonds (Fig. 19a).¹⁰⁷ The deprotonation of bound H₂PO₄⁻ was also evident from the origin of a new set of signals in the ¹H NMR titration experiments (DMSO-d₆). Similar solution-state behaviour has not been observed with the urea analogue 26. Further, an acetonitrile solution of 53 in the presence of phosphoric acid (H₃PO₄) yielded a non-capsular complex [(H53)₂HPO₄·3H₂O], with two receptor cations and an externally bound HPO₄²⁻ anion.¹⁰⁸ Phosphate complex [(TBA)₃PO₄(53)₂·2MeCN], can reproducibly be crystallized quantitatively in much higher yield simply by adding an excess of (TBA)F/(TBA)AcO into an acetonitrile solution of complex [(H53)2HPO₄·3H₂O]. Similarly, the TEA salt of PO₄³⁻ encapsulated complex [(TEA)₃PO₄(53)₂], can reproducibly be crystallized from an acetonitrile solution of complex [(H53)₂HPO₄·3H₂O] charged with excess (TEA)AcO/(TEA)HCO₃.¹⁰⁸ The solution-state deprotonation of receptor cations can be attributed to the high basicity of crystallization medium induced by the presence of excess fluoride/acetate/bicarbonate ions whereas, deprotonation of HPO₄²⁻ anion can be explained due to the formation of multiple hydrogen bonding interactions with the in situ generated neutral receptor that lowers the pK_a of the bound HPO₄²⁻ to the extent that it is deprotonated by excess fluoride/acetate/bicarbonate in solution. Competitive crystallization experiments performed in the presence of an excess of anions such as HCO₃⁻, HSO₄⁻, AcO⁻, NO₃⁻, F⁻ and Cl⁻ further establish the phenomenon of selective PO₄³⁻ encapsulation as confirmed by ¹H NMR, ³¹P NMR, and powder X-ray diffraction patterns of the isolated crystals. The size of the PO₄³⁻ capsules were measured to be 9.60 Å and 9.53 Å having TBA and TEA countercations, respectively. However, a much smaller capsular size of 7.92 Å has been observed in the carbonate complex [(TEA)₂CO₃(53)₂], comparable to the urea analogue [(TEA)₂CO₃(26)₂].

The combined solution-phase, solid-phase and phase-interface anion binding properties of a para-cyanophenyl-substituted thiourea 54 has been thoroughly explored by Ghosh et al.¹⁰⁹ Solution-state anion binding studies using ¹H NMR spectroscopy (DMSO-d₆) and ITC measurements in dry acetonitrile showed relatively higher association constants with halides over oxyanions, and follows the binding trend in the order F⁻ > Cl⁻ > HSO₄⁻ > H₂PO₄⁻ (Table 3), which is different from its urea analogue 16. The receptor has further been exploited in the liquid–liquid extraction of sulfate and fluoride from aqueous media. Using an anion-exchange based liquid–liquid extraction strategy the receptor has shown ca. 70% extraction of fluoride and ca. 40% extraction of sulfate from aqueous solutions by employing (TBA)I as the phase transfer agent. Identical to the other thiourea receptors (47, 50, 51, and 52), 54 formed a 2 : 1 host-guest capsular complex with SO₄²⁻ anion with a dimension of 9.51 Å.

Table 3 Binding constants (log K) of thiourea receptors with different anions

Receptors	F^−	Cl^−	H2PO4^−	SO4^2−
a K determined by ^1H NMR titration experiments.b K determined by UV-Vis/fluorescence titration experiments.
44^a	—	2.28	2.40	>4.0
45^b	—	—	5.03	3.93
46^a	—	2.10	2.11	>4.0
47^a	—	2.25	2.35	>4.0
53^a	4.47	3.54	5.87	3.75
54^a	3.57	3.06	2.20	2.83

---

It has been mentioned earlier that three new families of tripodal urea (34–36) and thiourea (55–57) receptors have been reported by Davis et al. showing the chloride binding affinities are strongly dependent on the nature of the binding groups (thiourea > urea) and the aryl substituent (electron-withdrawing groups being more effective).^93,94 In DMSO-d₆, benzene-based 57F2 (log K = 2.65) proved stronger than the desmethyl cyclohexane system 55F2 (log K = 2.0) but slightly weaker than the hexamethyl cyclohexane analogues 56F2 (log K = 2.82). However, in chloroform-d the chloride binding affinity of 57F2 (log K = 8.83) was found to be higher than that of 56F2 (log K = 7.47) and 55F2 (log K = 6.38) as well. Ab initio calculations suggested that the tris-(axial/NH-in) conformation of the cyclohexane-based tris-urea/thiourea systems can form six hydrogen bonds to Cl⁻ with lengths in the range 2.6–2.8 Å. Crystal structure analysis of the chloride complex of 57F2 revealed the formation of a dimeric receptor capsule surrounding a cyclic (Cl⁻–H₂O)₂ dianionic cluster (Fig. 20). Receptor 56F2 has been found to be the most effective anionophore, while 57F2 possesses roughly 70% the activity of 56F2, as measured by approximate half-lives t_1/2 (s) for chloride–nitrate exchange across the large unilamellar vesicles (LUVs, 200 nm), and specific initial rates [I] (s⁻¹) obtained from the fluorescence decay data. The hexamethyl cyclohexane system 56 promotes anion transport by improving binding without substantially affecting structural flexibility.

Fig. 20 X-ray structure of dimeric capsule [(57F2)₂(Cl⁻·H₂O)₂], C = gray, N = blue, O = red, F = chartreuse yellow. Countercations are not shown.

Monomeric capsules (semi-capsules) of thiourea receptors

Thiourea receptors 46 and 47 showed the formation of monomeric capsules with chloride anion [(TBA)(L·Cl)] (L = 46/47), as observed in urea analogue 20.⁸⁴ Formation of monomeric halide capsules have also been observed in tolyl-based receptor 50 in complexes [TBA(50·X)DMSO] (X = Cl⁻/Br⁻).¹⁰⁵ In all these complexes, an anion is coordinated by six –NH hydrogen bonds inside the receptor cavity.

Interestingly, the structural aspects of fluoride binding by a nitrophenyl-based tris-thiourea 53 revealed the formation of solvent (DMSO/MeCN) sealed monomeric capsules.^107,108 Both the complexes [TBA(53·F)S] (S = DMSO/MeCN) showed F⁻ encapsulation within the receptor cavity and hydrogen bonding interactions between a lattice solvent and peripheral nitro-aromatic functions, demonstrating the ditopic binding of a negatively charged species (F⁻) and a neutral molecule DMSO/MeCN (Fig. 21a). Whereas, fluoride complex [TBA(54·F)MeCN] of cyanophenyl-based receptor 54 showed the monotopic recognition of fluoride.¹⁰⁹

Fig. 21 X-ray structures of (a) solvent sealed monomeric capsule [53·(F⁻)DMSO], and (b) cation sealed monomeric capsule [53·(TBA)(SO₄)]⁻. C = green/gray, N = blue, O = red, S = yellow, F = chartreuse yellow. All TBA cations are not shown.

Receptors 46 and 53 showed the formation of cation (TBA) sealed monomeric capsules with SO₄²⁻ anion [(TBA)₂(L·SO₄)] (L = 46/53).^84,108 In both the complexes, a sulfate anion is bound within the receptor cavity by six –NH hydrogen bonds, and one of the TBA cations is held in close contact with both the receptor and anion by –CH hydrogen bonds (Fig. 21b).

Tripodal urea–amide hybrid receptors

Recently, a set of three tripodal urea-amide hybrid receptors 58–60 (Scheme 5) based upon the N-methyl-1,3,5-benzenetricarboxamide platform has been reported by Gunnlaugsson et al.¹¹⁰ This ligand platform having N-methyl tertiary amides gives rise to highly preorganized structures and this preorganization has resulted in the formation of self-assembled dimeric capsules with SO₄²⁻ and H₂PO₄⁻ anions. Structure analyses revealed the formation of 2 : 1 host–guest stoichiometric capsule with SO₄²⁻ anion, and 2 : 2 capsules with H₂PO₄⁻ anion, where the hydrogen bonded H₂PO₄⁻ dimer is bridged by a water molecule within the dimeric capsular assembly (Fig. 22). Sulfate capsules of the receptors were obtained from MeCN–DMF solutions, and showed identical mode of anion encapsulation involving all of the 12 urea –NH protons. ¹H NMR titrations of 58–60 with SO₄²⁻ showed the existence of two binding stoichiometry in solution, an initial formation of the desired 2 : 1 host–guest complex followed by the formation of a 1 : 1 stoichiometry.

Fig. 22 X-ray structures of dimeric capsules of urea–amide hybrid receptors, (a) [(58)₂·(SO₄)²⁻] and (b) [(58)₂·(H₂PO₄⁻)₂·H₂O]. C = green/gray, N = blue, O = red, S = yellow, P = orange, F = chartreuse yellow. TBA cations are not shown.

The same group has previously reported another set of tripodal urea–amide hybrid receptors 61–65 (Scheme 5) based on the 1,3,5-benzenetricarboxamide platform¹¹¹.¹H NMR titration studies showed that these receptors can bind anions such as, Cl⁻, AcO⁻, SO₄²⁻ and H₂PO₄⁻ in a 1 : 1 stoichiometry, but failed to form any self-assembled anion capsules in the solid state. Molecular modelling studies indicated that the secondary amides prevented the formation of the desired preorganized cavity, instead giving rise to the formation of more open or bowl-shaped structures. All receptors showed high binding affinities for AcO⁻, and H₂PO₄⁻.

Finally, an anion binding tripodal scaffold that needs to be addressed is the squaramide-based receptors 66–68 (Scheme 5).¹¹² Structure analysis of complex [(TBA)₂(66·SO₄)] revealed an unusual formation of 2 : 2 host–guest stoichiometric capsule with SO₄²⁻ anion, unlike those have been discussed for tris-urea/thiourea receptors. Each SO₄²⁻ anion is coordinated to both the receptors engaging two squaramide arms of one molecule and another arm from the second molecule of the dimeric capsule. DOSY NMR experiments indicated that dimeric structure does not exist as a stable complex in solution, and suggested a 1 : 1 binding stoichiometry, supported by ¹H-NMR titration studies. Detailed ¹H NMR titration experiments (DMSO-d₆) yielded the anion selectivity order (log K) SO₄²⁻ (4.75) > H₂PO₄⁻ (4.15) > HSO₄⁻ (3.65) > CH₃COO⁻ (2.82) > Cl⁻ (2.58) for receptor 66.

Systematic developments and relative anion binding affinities of tripodal receptors

Following their systematic developments over the years, tripodal receptors can be broadly categorized into first and second generation receptors. A vast majority of the tripodal receptors are based on the tris(2-aminoethyl)amine (Tren) scaffold and are regarded as the first generation tripodal receptors (FGTRs) because of their first use in anion receptor chemistry. The growth of anion coordination chemistry has witnessed the development of several other tripodal receptors based on 1,3,5-trialkyl (methyl/ethyl) benzene, cyclohexane and cyanuric acid platform. Because of their structural similarity with the Tren-based system (the bridgehead tertiary-N is replaced by a six member ring structure), these can also be regarded as the FGTRs. Solution and solid state studies of the FGTRs with different anions portrayed a clear picture of the concepts of preorganization, complementarity and hydrophobic effect essential for effective anion recognition by these families of flexible receptors. Based on these acquired information, several groups have designed and synthesized some second generation tripodal receptors (SGTRs) either by incorporation of additional functional groups and/or aromatic rings into the tripodal side arms or increasing the number of hydrogen bond donor sites on each side arm of the receptor molecule. The development of SGTRs have brought into light a range of exciting structures such as, tetrahedral phosphate cage formed by 37 and sulfate encapsulation within the complementary tetrahedral cavity of 40 and 42. However, solution state anion binding affinity of the SGTRs did not vary significantly in comparison to most FGTRs. This could be the reason why FGTRs (mostly with fluorinated aryl functions) have mostly been employed for transmembrane anion transportation applications, with advantageous simple and easy synthetic accessibility. A year wise development of different tripodal scaffolds for anion recognition has been portrayed in Scheme 6, and Table 4 shows a comparative account of the solution state anion binding affinities (log K) of the tripodal receptors for the commonly studied halides (F⁻ and Cl⁻) and oxyanions (AcO⁻, SO₄²⁻/HSO₄⁻ and H₂PO₄⁻).

Scheme 6 Selected scaffolds of first generation and second generation tripodal receptors (FGTRs and SGTRs), showing their systematic developments over the years (1992–2015).

Table 4 A comparative account of the tripodal receptors based on their increasing binding constant values for an anion

Anions	Tripodal receptors and their relative binding constants (log K)
	log K = 1–3	log K = 3–4	log K = 4–5	log K = 5–6	log K > 6
F^−	9	3b, 5a, 5b, 25, 54	3c, 16, 21, 53	3a, 10, 27	4
Cl^−	3c, 5b, 9, 13, 20, 22, 44, 46, 47, 61, 62, 64	2a, 3b, 5a, 10, 16, 21, 33, 53, 54	27	—	55–57F2
AcO^−	61, 62, 66	25, 26, 33, 63, 64, 65	21, 27	12	—
H2PO4^2−	29, 44, 46, 47, 54	2a, 61, 63, 64, 65	13, 14, 16, 25, 26, 27, 33, 62, 66	45, 53	—
SO4^2−/HSO4^−	2a, 9, 29, 54, 61, 63, 64, 65	14, 41, 45, 53, 62	13, 16, 20, 21, 22, 25, 26, 40, 44, 46, 47, 52, 66	30, 33	27

---

Conclusions

We have been involved in the research targeting the anion induced supramolecular self-assemblies of hydrogen bonding tripodal scaffolds since almost a decade, and obtained some interesting results employing a few N-bridged tripodal receptors having either amide or urea/thiourea functions for anion coordination. The purpose of this review is to discuss and generalize the anion coordination induced formation of dimeric and monomeric capsular assemblies in hydrogen bonding tripodal scaffolds. Tren-based tris-amide receptors (1–5) have always been shown to form monomeric capsules with halides with high selectivity towards fluoride and thus, selective fluoride sensing and efficient fluoride extraction from water has been achieved with receptors 4 and 5a, respectively. Whereas, tris-amide receptors (7, 8 and 9) on a mesitylene platform has been established as an ideal scaffold for encapsulation of halide-water clusters via formation of dimeric capsular assembly. Tren-based tris-urea receptors (13–28) have frequently been shown to form dimeric capsules with oxyanions showing high selectivity towards sulfate and thereby, selective sulfate separation by competitive crystallization and liquid–liquid extraction has been achieved with receptors 15, 17 and 21 respectively. Tren-based hexaurea receptors with an ortho-phenylene bridge (40–42) provides an ideal cavity for sulfate encapsulation and as such, receptors 40 and 42 have been employed in the selective separation of sulfate from aqueous solution containing high concentration of nitrate by liquid–liquid extraction. Another interesting property that has been observed with few tripodal urea receptors (tris/hexa-urea) is the capture of atmospheric carbondioxide as carbonate inside the molecular cavity of the tripodal scaffold, as demonstrated by tris-urea receptors 15, 21, 25 and hexaurea 41 in basic DMSO solution mixtures (Table 5).

Table 5 Tripodal amide and urea/thiourea receptors for selective anion recognition, anion separation and transmembrane anion transport

Receptors	Applications	Method(s) involved/remarks
3a	Selective fluoride recognition	^1H-NMR spectroscopy
4	Selective fluoride recognition	^1H-NMR and UV-vis spectroscopy
5a	Extraction of KF and KCl from water	Extraction by 18-crown-6-ether as a cation host
6g	Selective nitrate sensing	Fluorescence spectroscopy
15	Carbondioxide uptake as carbonate complex	DMSO solution containing (TBA)OH
15	Extraction of sulfate, thiosulfate and chromate	Extraction by carbonate complex (in organic phase)
17	Selective sulfate separation	Competitive crystallization
21	Carbondioxide uptake as carbonate complex	DMSO solution containing (TBA)OH
21	Extraction of sulfate from aqueous to organic phase	Extraction by carbonate complex (in organic phase)
23	Transmembrane chloride transport	Vesicle anion transport assay
25	Carbondioxide uptake as carbonate complex	DMSO solution containing (TBA)OH/F
27	Electrochemical sensing of sulfate and phosphate	Cyclic voltammetry
40	Extraction of sulfate from water	(TBA)Cl as phase transfer agent
41	Carbondioxide uptake as carbonate complex	DMSO solution containing (TBA)F
53	Selective phosphate separation	Competitive crystallization
54	Extraction of fluoride and sulfate from water	(TBA)I as phase transfer agent
56–57F2	Transmembrane chloride transport	Vesicle anion transport assay

---

The structural aspects of anion binding by Tren-based tris-thiourea receptors (44–54) are analogues to the tris-urea receptors as observed in fluorinated urea and thiourea receptors. However, receptor 53 (thiourea analogue of tris-urea receptor 26) showed unusual selectivity for trivalent phosphate encapsulation even in the presence of other competitive anions, which has not been observed with any other tripodal urea/thiourea receptors. Most importantly, some significant results have also been achieved in the area of transmembrane anion transport using fluorinated tris-urea/thiourea receptors as anion carriers. Overall, a significant amount of research has been pursued towards the progress of anion induced capsular self-assemblies with hydrogen bonding tripodal scaffolds. Based on the information collected from X-ray structures of numerous anion complexes, chemists have always tried to optimize the receptor-anion complementarity by suitable functioning of the tripodal scaffold with highly ordered convergent binding groups. Although a few examples of anion separation has been discussed in this review, but the utility of molecular capsules for the binding and removal of anionic species is undoubtedly in its infancy. Further research in toxic anion separation using the basic concept of anion receptor chemistry is much waited, which could solve some of the practical aspects of drinking water purification. The expansion of biological applications for synthetic anion transporters is also in its infancy, and there will certainly be more opportunities in the future for using the tripodal receptors to address biochemical and biomedical problems. In consideration of the interesting results achieved, tripodal receptors will continue to play an active role in anion coordination chemistry and will continue to be one of the first options to be considered in the design of receptors for halides and oxyanions.

Acknowledgements

This work was supported by CSIR and SERB through grant 01/2727/13/EMR-II and SR/S1/OC-62/2011, New Delhi, India. We also thank Central Instrument Facility (CIF) at Indian Institute of Technology (IIT) Guwahati for providing instrument facilities, and DST-FIST for sponsoring a single crystal X-ray diffractometer at Department of Chemistry, IIT Guwahati.

Notes and references

1. J. L. Sessler, P. A. Gale and W.-S. Cho, Anion Receptor Chemistry, RSC Publishing, Cambridge, 2006.
2. C. J. Avers, Molecular Cell Biology, Addison–Wesley, Reading, 1986.
3. J. L. Atwood and A. Szumna, J. Am. Chem. Soc., 2002, 124, 10646.
4. P. A. Gale, Chem. Commun., 2011, 47, 82.
5. P. A. Gale, N. Busschaert, C. J. E. Haynes, L. E. Karagiannidis and I. L. Kirby, Chem. Soc. Rev., 2014, 43, 205.
6. M. Wenzel, J. R. Hiscock and P. A. Gale, Chem. Soc. Rev., 2012, 41, 480.
7. P. A. Gale, Chem. Soc. Rev., 2010, 39, 3746.
8. C. Caltagirone and P. A. Gale, Chem. Soc. Rev., 2009, 38, 520.
9. P. A. Gale, S. E. García-Garrido and J. Garric, Chem. Soc. Rev., 2008, 37, 151.
10. T. Gunnlaugsson, M. Glynn, G. M. Tocci, P. E. Kruger and F. M. Pfeffer, Coord. Chem. Rev., 2006, 250, 3094.
11. S. Kubik, Chem. Soc. Rev., 2010, 39, 3648.
12. T. H. Rehm and C. Schmuck, Chem. Soc. Rev., 2010, 39, 3597.
13. C. Bazzicalupi, A. Bencini and V. Lippolis, Chem. Soc. Rev., 2010, 39, 3709.
14. M. Cametti and K. Rissanen, Chem. Commun., 2009, 2809.
15. E. García-Espana, P. Diaz, J. M. Llinares and A. Bianchi, Coord. Chem. Rev., 2006, 250, 2952.
16. M. Boiocchi, L. D. Boca, D. Esteban-Gómez, L. Fabbrizzi, M. Licchelli and E. Monzani, J. Am. Chem. Soc., 2004, 126, 16507.
17. D. Esteban-Gómez, L. Fabbrizzi and M. Licchelli, J. Org. Chem., 2005, 70, 5717.
18. L. S. Evans, P. A. Gale, M. E. Light and R. Quesada, Chem. Commun., 2006, 965.
19. P. Bose and P. Ghosh, Chem. Commun., 2010, 46, 2962.
20. A. Basu and G. Das, Inorg. Chem., 2012, 51, 882.
21. A. Basu, S. K. Dey and G. Das, RSC Adv., 2013, 3, 6596.
22. K. Bowman-James, Acc. Chem. Res., 2005, 38, 671.
23. J. T. Davis, O. Okunola and R. Quesada, Chem. Soc. Rev., 2010, 39, 3843.
24. P. A. Gale, Acc. Chem. Res., 2011, 44, 216.
25. P. A. Gale, R. Pérez-Tomás and R. Quesada, Acc. Chem. Res., 2013, 46, 2801.
26. H. Maeda, T. Shirai and S. Uemura, Chem. Commun., 2013, 49, 5310.
27. H. Maeda, K. Naritani, Y. Honsho and S. Seki, J. Am. Chem. Soc., 2011, 133, 8896.
28. B. Dong, T. Sakurai, Y. Honsho, S. Seki and H. Maeda, J. Am. Chem. Soc., 2013, 135, 1284.
29. B. Dong, T. Sakurai, Y. Bando, S. Seki, K. Takaishi, M. Uchiyama, A. Muranaka and H. Maeda, J. Am. Chem. Soc., 2013, 135, 14797.
30. K. M. Mullen and P. D. Beer, Chem. Soc. Rev., 2009, 38, 1701.
31. P. Ballester, Chem. Soc. Rev., 2010, 39, 3810.
32. R. Vilar, Struct. Bonding, 2008, 129, 175.
33. H. Juwarker and K.-S. Jeong, Chem. Soc. Rev., 2010, 39, 3664.
34. S. J. Coles, J. G. Frey, P. A. Gale, M. B. Hursthouse, M. E. Light, K. Navakhun and G. L. Thomas, Chem. Commun., 2003, 568.
35. T. Basu and R. Mondal, CrystEngComm, 2010, 12, 366.
36. J. W. Steed and J. L. Atwood, Supramolecular Chemistry: An Introduction, Wiley, Chichester, U.K, 2000.
37. Z. Wang, H. Luecke, N. Yao and F. A. Quiocho, Nat. Struct. Biol., 1997, 4, 519.
38. J. W. Pflugrath and F. A. Quiocho, Nature, 1985, 314, 257.
39. J. W. Steed, Chem. Commun., 2006, 2637.
40. B. Kuswandi, Nuriman, W. Verboom and D. N. Reinhoudt, Sensors, 2006, 6, 978.
41. M. Arunachalam and P. Ghosh, Chem. Commun., 2011, 47, 8477.
42. R. Dutta and P. Ghosh, Chem. Commun., 2014, 50, 10538.
43. A. E. Hargrove, S. Nieto, T. Zhang, J. L. Sessler and E. V. Anslyn, Chem. Rev., 2011, 111, 6603.
44. F. G. Bordwell, Acc. Chem. Res., 1988, 21, 456.
45. V. S. Bryantsev and B. P. Hay, Org. Lett., 2005, 7, 5031.
46. P. D. Beer, D. Hesek, J. Hodacova and S. E. Stokes, J. Chem. Soc., Chem. Commun., 1992, 270.
47. P. D. Beer, C. Hazlewood, D. Hesek, J. Hodacova and S. E. Stokes, J. Chem. Soc., Dalton Trans., 1992, 1327.
48. P. D. Beer, Z. Chen, A. J. Goulden, A. Graydon, S. E. Stokes and T. Wearb, J. Chem. Soc., Chem. Commun., 1993, 1834.
49. S. Valiyaveettil, J. F. J. Engbersen, W. Verboom and D. N. Reinhoudt, Angew. Chem., Int. Ed., 1993, 32, 900.
50. I. Ravikumar, P. S. Lakshminarayanan and P. Ghosh, Inorg. Chim. Acta, 2010, 363, 2886.
51. S. K. Dey and G. Das, Chem. Commun., 2011, 47, 4983.
52. S. K. Dey and G. Das, Cryst. Growth Des., 2011, 11, 4463.
53. S. K. Dey, B. K. Datta and G. Das, CrystEngComm, 2012, 14, 5305.
54. I. Ravikumar, S. Saha and P. Ghosh, Chem. Commun., 2011, 47, 4721.
55. S. Saha, B. Akhuli, I. Ravikumar, P. S. Lakshminarayanan and P. Ghosh, CrystEngComm, 2014, 16, 4796.
56. A. S. Singh and S.-S. Sun, Chem. Commun., 2011, 47, 8563.
57. A. S. Singh and S.-S. Sun, RSC Adv., 2012, 2, 9502.
58. A. S. Singh and S.-S. Sun, J. Org. Chem., 2012, 77, 1880.
59. M. Arunachalam and P. Ghosh, Chem. Commun., 2009, 5389.
60. M. Arunachalam and P. Ghosh, Inorg. Chem., 2010, 49, 943.
61. R. Dutta and P. Ghosh, Eur. J. Inorg. Chem., 2012, 3456.
62. S. Bao, W.-T. Gong, W.-D. Chen, J.-W. Ye, Y. Lin and G.-L. Ning, J. Inclusion Phenom. Macrocyclic Chem., 2011, 70, 115.
63. S. Chakraborty, R. Dutta, M. Arunachalam and P. Ghosh, Dalton Trans., 2014, 43, 2061.
64. A. Basu and G. Das, Chem. Commun., 2013, 49, 3997.
65. A. Basu, R. Chutia and G. Das, CrystEngComm, 2014, 16, 4886.
66. P. J. Smith, M. V. Reddington and C. S. Wilcox, Tetrahedron Lett., 1992, 33, 6085.
67. E. Fan, S. A. V. Arman, S. Kincaid and A. D. Hamilton, J. Am. Chem. Soc., 1993, 115, 369.
68. C. Raposo, M. Almaraz, M. Martin, V. Weinrich, M. L. Mussons, V. Alcazar, M. C. Caballero and J. R. Morán, Chem. Lett., 1995, 37, 2795.
69. H. Xie, S. Yi, X. Yang and S. Wu, New J. Chem., 1999, 23, 1105.
70. H. Xie, S. Yi and S. Wu, J. Chem. Soc., Perkin Trans. 2, 1999, 2751.
71. R. Custelcean, B. A. Moyer and B. P. Hay, Chem. Commun., 2005, 5971.
72. P. Bose, R. Dutta and P. Ghosh, Org. Biomol. Chem., 2013, 11, 4581.
73. R. Dutta, S. Chakraborty, P. Bose and P. Ghosh, Eur. J. Inorg. Chem., 2014, 25, 4134.
74. R. Custelcean, P. Remy, P. V. Bonnesen, D. Jiang and B. A. Moyer, Angew. Chem., Int. Ed., 2008, 47, 1866.
75. B. Wu, J. Liang, J. Yang, C. Jia, X.-J. Yang, H. Zhang, N. Tang and C. Janiak, Chem. Commun., 2008, 1762.
76. R. Custelcean and P. Remy, Cryst. Growth Des., 2009, 9, 1985.
77. A. Rajbanshi, B. A. Moyer and R. Custelcean, Cryst. Growth Des., 2011, 11, 2702.
78. R. Zhang, Y. Zhao, J. Wang, L. Ji, X.-J. Yang and B. Wu, Cryst. Growth Des., 2014, 14, 544.
79. B. Akhuli, I. Ravikumar and P. Ghosh, Chem. Sci., 2012, 3, 1522.
80. I. Ravikumar, P. S. Lakshminarayanan, M. Arunachalam, E. Suresh and P. Ghosh, Dalton Trans., 2009, 4160.
81. P. S. Lakshminarayanan, I. Ravikumar, E. Suresh and P. Ghosh, Chem. Commun., 2007, 5214.
82. R. Dutta, P. Bose and P. Ghosh, Dalton Trans., 2013, 42, 11371.
83. I. Ravikumar and P. Ghosh, Chem. Commun., 2010, 46, 1082.
84. N. Busschaert, M. Wenzel, M. E. Light, P. Iglesias-Hernandez, R. Perez-Tomas and P. A. Gale, J. Am. Chem. Soc., 2011, 133, 14136.
85. R. Chutia, S. K. Dey and G. Das, Cryst. Growth Des., 2013, 13, 883.
86. S. K. Dey, R. Chutia and G. Das, Inorg. Chem., 2012, 51, 1727.
87. R. Chutia, S. K. Dey and G. Das, Cryst. Growth Des., 2015, 15, 4993.
88. D. A. Jose, D. K. Kumar, B. Ganguly and A. Das, Inorg. Chem., 2007, 46, 5817.
89. A. Basu and G. Das, J. Org. Chem., 2014, 79, 2647.
90. R. Dutta, B. Akhuli and P. Ghosh, Dalton Trans., 2015, 44, 15075.
91. R. Dutta and P. Ghosh, Eur. J. Inorg. Chem., 2013, 2673.
92. I. Ravikumar and P. Ghosh, Chem. Commun., 2010, 46, 6741.
93. J. A. Cooper, S. T. G. Street and A. P. Davis, Angew. Chem., Int. Ed., 2014, 53, 5609.
94. H. Valkenier, C. M. Dias, K. L. P. Goff, O. Jurcek, R. Puttreddy, K. Rissanen and A. P. Davis, Chem. Commun., 2015, 51, 14235.
95. B. Wu, F. Cui, Y. Lei, S. Li, N. de, S. Amadeu, C. Janiak, Y.-J. Lin, L.-H. Weng, Y.-Y. Wang and X.-J. Yang, Angew. Chem., Int. Ed., 2013, 52, 5096.
96. A. Pramanik, D. R. Powell, B. M. Wong and M. A. Hossain, Inorg. Chem., 2012, 51, 4274.
97. F. Zhuge, B. Wu, J. Liang, J. Yang, Y. Liu, C. Jia, C. Janiak, N. Tang and X.-J. Yang, Inorg. Chem., 2009, 48, 10249.
98. M. Li, B. Wu, C. Jia, X. Huang, Q. Zhao, S. Shao and X.-J. Yang, Chem.–Eur. J., 2011, 17, 2272.
99. M. Li, Y. Hao, B. Wu, C. Jia, X. Huang and X.-J. Yang, Org. Biomol. Chem., 2011, 9, 5637.
100. D. R. Turner, M. J. Paterson and J. W. Steed, J. Org. Chem., 2006, 71, 1598.
101. D. R. Turner, M. J. Paterson and J. W. Steed, Chem. Commun., 2008, 1395.
102. C. Jia, B. Wu, S. Li, X. Huang, Q. Zhao, Q.-S. Li and X.-J. Yang, Angew. Chem., Int. Ed., 2011, 50, 486.
103. A. Pramanik, M. E. Khansari, D. R. Powell, F. R. Fronczek and M. A. Hossain, Org. Lett., 2014, 16, 366.
104. X. Huang, B. Wu, C. Jia, B. P. Hay, M. Li and X.-J. Yang, Chem.–Eur. J., 2013, 19, 9034.
105. A. Gogoi and G. Das, Supramol. Chem., 2013, 25, 819.
106. A. Basu and G. Das, Dalton Trans., 2012, 41, 10792.
107. S. K. Dey and G. Das, Dalton Trans., 2011, 40, 12048.
108. S. K. Dey and G. Das, Dalton Trans., 2012, 41, 8960.
109. P. Bose, R. Dutta, S. Santra, B. Chowdhury and P. Ghosh, Eur. J. Inorg. Chem., 2012, 35, 5791.
110. K. Pandurangan, J. A. Kitchen, S. Blasco, E. M. Boyle, B. Fitzpatrick, M. Feeney, P. E. Kruger and T. Gunnlaugsson, Angew. Chem., Int. Ed., 2015, 54, 4566.
111. C. Maria Gomes dos Santos, E. M. Boyle, S. D. Solis, P. E. Kruger and T. Gunnlaugsson, Chem. Commun., 2011, 47, 12176.
112. C. Jin, M. Zhang, L. Wu, Y. Guan, Y. Pan, J. Jiang, C. Lin and L. Wang, Chem. Commun., 2013, 49, 2025.

---

Footnotes

† Present address: Institut für Anorganische Chemie und Strukturchemie, Heinrich-Heine-Universität Düsseldorf, 40204 Düsseldorf, Germany.

‡ Present address: Physical/Materials Chemistry Division, CSIR-National Chemical Laboratory, Dr. Homi Bhabha Road, Pune 411008, India.

---