If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Superphanes, compounds in which the two benzene rings clamped parallel on top of each other by six bridges, have been attracting considerable interest because of their structural beauty and symmetry. However, progress in the research of superphane chemistry and beyond has been seriously hampered by their poor availability. Herein, we report the facile and scalable synthesis of a collection of superphanes with structural diversity and their unique photophysical properties, as well as their unusual host–guest behavior. A set of secondary amine–linked superphanes are obtained via dynamic self–assembly of a hexakis–amine and various readily derived aromatic dialdehyde in one pot, followed by in–situ reduction with NaBH4. Further post–functionalization was exemplified by secondary amine to tertiary amine conversion. All superphanes under study were found to exhibit interesting fluorescence either in solution or in the solid state. Finally, the unusual host–guest binding properties was exemplified by the stabilization of a labile 2Cl–·H2O cluster with fully protonated 6a both in solution and in the solid state. With the easy and versatile synthesis, modification, as well as unique photophysical and host–guest properties reported herein, we believe that this study offers new possibilities for superphane chemistry, supramolecular host design, advanced functional materials and beyond.
], cyclophane chemistry has blossomed into an active research area and the cyclophane–based systems have spurred considerable interest in the fields of supramolecular chemistry, natural products, organometallic chemistry, asymmetric synthesis, functional π systems, polymer chemistry, and materials science [
]. However, as stated by Vögtle in 1972 “the ultimate achievement of work in the cyclophane field would be the synthesis of the fully bridged [188.8.131.52.2.2](1,2,3,4,5,6)cyclophane (1, Fig. 1)” that was trivially termed as superphane initially proposed by Boekelheide and Hopf in 1979 [
]. As a tour de force in man–made organic molecules, superphanes, featuring aesthetically pleasing structures with high symmetry (D6h), were initially designed as a highly strained synthetic goal for the discovery of new preparative procedures, and attractive physicochemical properties such as chemical bond, ring strain, and unusual π–electron interactions [
]. In theory, superphanes could be rendered to be unique yet useful supramolecular hosts for specific guest species via well–designed extension and decoration of the toroidal bridges to furnish suitable binding cavities. Nevertheless, most, if not all, of these possibilities have been limited by their availability over the past 40 years or so [
] superphane–based supramolecular hosts are expected to have many advantages and unique characteristics. Firstly, the exceptionally high structural symmetry (D6h or C6v) may simplify its synthesis; Secondly, the highly dense binding sites (6n, where n is the number of binding site(s) on each bridge) may allow for tight guest binding, that is of significance to biomolecular recognition, ion extraction, and cluster stablization [
]; Thirdly, the uniform three–dimensional distribution of the six connecting bridges around the near–enclosed internal cavity may prevent the embedded species from solvation or attack by the external substances. Given the complexity and symmetry of superphanes per se, we envisioned that highly efficient and selective dynamic covalent reactions are beneficial to access such a class of complicated superstructures [
]. For example, in 2020, our group successfully introduced the imine dynamic chemistry to construct the first–ever superphane bearing imine bonds as a supramolecular receptor for hosting either small anions or neutral species, e.g. a water dimer, via self–assembly of two simple substrates in a 2:6 ratio in one pot [
]. Almost simultaneously and independently, Badjić group reported a hexapodal capsule for the recognition of anions, where the capsule was prepared through a dynamic reaction of a hexa–amine and a readily premade hexa–podal aldehyde precursor in a 1:1 fashion using a similar strategy [
]. These two examples provided some interesting results for superphane chemistry. However, both systems 1) consist of multiple unstable and reversible imine bonds; 2) didn't show structural diversity and modification; 3) prepared in milligram scale. Thus, the best is still yet to come.
On the basis of our previously reported imine dynamic strategy, herein, we describe a general synthetic route accessible to a series of highly stable superphanes (2, Fig. 1) with irreversible covalent bonds via in–situ reduction of the corresponding imino–based superphanes using NaBH4 as the reductive reagent. The easy accessibility and high efficiency were exemplified with successful gram–scale preparation of the secondary–amine–based superphane 6c, which was successfully post–functionalized with ethyl iodide, allyl bromide, propargyl bromide or 1–bromobut–2–yne bromide, giving the corresponding peraza–substituted superphanes. Notably, the synthesis of all superphanes under current study was chromatography free. More importantly, unprecedented solid–state and solution–state fluorescent properties were also discussed. Finally, protonated superphane 6a proves capable of binding a labile trimeric chloride hydrate cluster (2Cl–·H2O). To the best of our knowledge, this is the first report on the facile and gram–scale synthesis of superphanes that feature liquid and solid fluorescence, as well as effective encapsulation of a 2Cl–·H2O cluster.
2. Results and discussion
2.1 One–pot, gram–scale synthesis and pre–modification
According to our previously reported work, the imino–containing superphane 5a could be accessible via dynamic self–assembly of hexakis–amine (3) and m–phthalaldehyde (4a) in a 2:6 ratio in one pot [
]. Inspired by the fact that macrobicyclic azacryptands and sandwich–like azacyclophanes could be easily prepared through reversible and thermodynamically driven imine condensation, followed with a reduction step [
] we next sought to transform imine superphane 5a into its stable amine form. To our delight, treatment of 5a with NaBH4 in a mixture of CH2Cl2 and CH3OH (1:1, v/v) at room temperature over 5 h led to the formation of amine superphane 6a in 81% yield (Scheme 1). Similarly, self–assembly of 3 with 5–bromoisophthaldehyde 4b gave imine–based superphane 5b in 39% yield. After reduction with NaBH4, 5b was transformed into 6b in a yield of 74% (Schemes S1 and S2). Satisfyingly, we managed to further simplify the synthesis of 6a and 6b via in–situ reduction of 5a or 5b after the completion of self–assembly process in one pot with decent yields (69% for 6a and 59% for 6b, respectively) (see Scheme 1).
To confirm the versatility of this approach to synthesis of amine–based superphanes of interest, different variants of m–phthalaldehyde, viz. 2,6–pyridinedicarboxaldehyde (4c), 2,5–furandicarboxaldehyde (4d) and larger 4,4′–oxydibenzaldehyde (4e), were utilized to replace 4a. Expectedly, using a similar imine condensation strategy, compounds 6c–6e were successfully obtained by the simple one–pot self–assembly of hexakis–amine 3 with 4c, 4d and 4e, respectively, followed by in–situ reduction reactions in the presence of NaBH4 (Scheme S3). Of particular note is that, during the course of synthesis, no chromatography was needed for the purification of either 5a–5e or 6a–6e. All imine–based superphanes 5b–5e and amine–bearing superphanes 6a–6e have been well characterized by 1H NMR, 13C NMR spectroscopy, and electrospray ionization high–resolution mass spectrometry (ESI–HRMS) (see supporting information).
Subsequently, the availability of this facile and concise synthetic approach to a wide range of amine–based superphanes was further examined by the demonstrative gram–scale synthesis of superphane 6c in one pot. Concretely, the synthesis started from 2,6–pyridinedicarboxaldehyde 4c (1.27 g), which was subject to imine condensation with hexakis–amine 3 (2.36 g) in methanol at 65 °C for 12 h. Then, the resulting mixtures with a substantial number of precipitates were in–situ treated with excess NaBH4 for a further 12 h at RT, where the precipitates disappeared gradually thereafter. After solvent removal, the resulting residuals were dispersed in water and repeatedly extracted with chloroform. The organic phases were combined, dried, and evaporated to yield crude product, which was washed with excessive acetonitrile to offer 1.32 g of desired superphane 6c as a light yellow powder in 47% overall yield. Notably, most of the yields obtained are far from perfect (100%) presumably due in large part to the occurrence of polymers or oligomers as the by-products. But they are acceptable in consideration of the structural complexity of superphanes per se.
Generally, post–synthetic functionalization of the core macrocycles or molecular cages has been ascertained to be a most effect synthetic route to access new analogues without a need to rework the synthesis of the core structures [
]. However, the binding properties of the core systems in question could be modulated with greater fidelity. Additionally, direct (post–)functionalization of macrocycles or cages is also key to their widespread uses in the areas of material science, biology, and environmental science et al. With this in mind, we next sought to explore the post–functionalization of amine–based superphanes. As a representative example, 6c was then utilized for modification with alkyl halides. In the first test, ethyl iodide (7a) was subject to reaction with 6c pretreated with NaH in DMF. As expected, the per–substituted product 8a was obtained in yield of 8% (Schemes 2 and S4). Given the fact that allyl and propargyl groups are two of the most commonly used functionalizable moieties for fabrication of functional molecules and materials via olefin metathesis, alkyne metathesis or click reactions [
], then commercially available allyl bromide (7b), propargyl bromide (7c) and 1–bromobut–2–yne bromide (7d) were employed for direct modification of 6c. Specifically, treatment of 6c with excess allyl bromide in the presence of NaOH (in DMF) at 65 °C for 20 h led to the expected per–azasubstituted product 8b in 21% yield. Similarly, triple bond–containing groups proved capable of being incorporated into 6c via the substitution of 6c with excess propargyl bromide or 1–bromobut–2–yne bromide in the assistance of NaH (in DMF) at room temperature, offering 8c and 8d in yields of 11% and 5%, respectively. Notably, the post-functionalization yields are apparently poor (ranging from 5% to 21%) presumably due to the occurrence of partially alkylated by-products. However, it is reasonable when one considers that we are dealing with “one-pot” reaction forming 12 new chemical bonds. Superphanes 8a–8d bearing 12 allyl substituents have been well characterized by 1H NMR, 13C NMR spectroscopy, and electrospray ionization high–resolution mass spectrometry (ESI–HRMS) (see supporting information). Interestingly, based on the 1H NMR spectrum of 8a–8d in CDCl3 or DMSO–d6 at 298 K, all three sets of methylene proton signals were found split, respectively, indicating that these methylene protons were distinguishable even at room temperature probably due to the restriction of intramolecular rotation of methylene units in question (Figs. S1–S4).
2.3 Solid–state structures
To obtain detailed insight into the structure, conformation, and potential binding properties of superphanes 6a–6e, substantial trials have been dedicated to growing their single crystals in either neutral form or protonated fashion. In the case of superphane 6c, many efforts have been made to grow single crystals of either 6c or its protonated form in the presence of strong acids, e.g. HCl, HBr, TsOH (4–methylbenzenesulfonic acid) and CF3COOH, but to no avail. Much to our surprise, upon slowly diffusing acetone vapor into a solution of 6c in a mixture of chloroform and methanol in the presence of excess CF3COOLi, colorless block–like single crystals suitable for X–ray single crystal structure analyses were obtained. Unexpectedly, the yielding crystal structure revealed a 2CH3[email protected]6c·2H+ complex, instead of an expected CF3COOLi inclusion complex of 6c (Fig. S5). In the solid state, the framework of 6c·2H+ displayed D2h symmetry and a tunnel right through the center of the cage (Fig. 2a). Unambiguously, one methanol molecule was observed to be entrapped within the center of 6c and the other one was clamped by two out of six pyridyl–containing bridges (Fig. S6).
To elucidate the structures in greater details of the post modified superphanes (8a–8d) with 12 alkyl groups, great efforts were devoted to growing single crystals with 8a–8d. Luckily, suitable single crystals were obtained by slow diffusion of ethyl ether vapor into a solution of 8b in a mixture of CH3OH, CHCl3 and Et2O at low temperature (0–5 °C). The resulting crystal structure gave an expected super cage with exceptionally high symmetry (near D6h) (Fig. 2b and c). Due to steric congestion, all allyl groups are oriented opposite to the core of superphane 8b to form two identical bowl–shaped outer cavities with the depth of 3.83 Å (Fig. S7). Meanwhile, the two face–to–face benzene rings are somewhat extruded to generate a close benzene···benzene contact of 4.57 Å, indicating the occurrence of weak π···π interactions and the disappearance of internal void of 8b. In this case, no solvent molecules were found within the internal cavity of superphane 8b. Instead, an ethyl ether molecule was observed in each outer cavity formed by six allyl moieties as the fences (Fig. S8). Each pyridinyl group at the periphery of 8b was aligned nearly perpendicular to the corresponding bridge plane and almost parallel to the other one at its opposite site. Interestingly, each 8b cage was found surrounded tightly by other six 8b molecules at the same plane via multiple cooperative Csp2–H···π, offset π···π, and Csp3–H···π interactions, giving rise to a regular hexagon with the side length of 14.23 Å (Figs. S9 and S10). In the case of propargyl–derived superphane 8c, the molecular geometry and molecular packings were observed to be analogous to what was seen in the case of 8b, as inferred from the single crystal structure of 8c (Figs. 2d and S11–S14).
2.4 Photophysical properties
Fluorescence is a property of matter that is critical for many research fields, such as chemistry, biology, material science and the like [
]. Interestingly, although no specific fluorofores were integrated, genuine blue–green fluorescent emissions were observed when chloroform solutions of 6c, 6d and 6e were exposed to 365–nm UV light (Fig. 3a, inset). Upon excitation of 6c, 6d, and 6e in chloroform at 340, 373, and 367 nm, respectively, they showed similar emission bands of 400–600 nm with an emission maxima at 452, 449, and 439 nm, respectively (Figs. S15–S17). Due to the poor solubility of 6a and 6b in commonly used solvents, they were subject to protonation with HCl and the resulting salts were found soluble in DMSO. Upon excitation of the protonated 6a, 6b in DMSO at 365 and 376 nm, respectively, emission bands of 400–600 nm with corresponding emission maxima at 443 and 442 nm were seen (Figs. S18 and S19). Notably, all of these superphanes in solution displayed poor fluorescence quantum yields (Φf < 0.01). In contrast, both imino–based superphanes 5a–5c, 5e and the alkyl–functionalized superphanes 8a–8d in solutions proved nearly fluorescence–silent. These findings indicated that the observed weak emission of 6a–6e in solution might be attributed to an intramolecular proton transfer mechanism.
Interestingly, as opposed to the weak/non–fluorescent features in solution, relatively strong solid–state fluorescent emission of the imino–based compounds 5a–5c, and 5e was observed (Fig. 3b). Specifically, when exited with light at 433, 446, 497 and 446 nm for 5a, 5b, 5c, and 5e, respectively, their solid samples emitted strong visible fluorescent light with the corresponding emission maxima at 505, 517, 551 and 527 nm (Figs. S20–S23). Similarly, the secondary amine–based superphanes 6a–6e in solid state also exhibited relatively strong fluorescence with the emission maxima ranging from 439 nm to 549 nm (Figs. S24–S28). Structurally, compounds 5a–5e, inter alia the former four, are quite similar. Nevertheless, their solid–state emitting color differs from one another significantly ranging from chartreuse to near white with fluorescence quantum yields (Φf) of 7.1, 5.8, 3.5 and 14.0 for 5a, 5b, 5c and 5e, respectively (Fig. 3c). Likewise, cages 6a–6e as powders emit near white, blueing, yellowish, near–white, and white fluorescence, respectively, with the corresponding fluorescent quantum yields of 5.2, 9.0, 10.0, 17.1 and 4.0. As a representative example of the alkyl–functionalized systems, 8c was also observed to exhibit relatively strong light–yellow fluorescence in the solid state (Fig. S29). Taken together, based on the framework of superphane, a series of emissive fluorescent super cages were easily obtained via incorporating various substituents into the bridges between the top and bottom benzene rings. The acquired fluorescence is likely to be finely tuned by modifying the bridges with specific units or by reducing the imine bonds of the superphanes. Given the weak solution–state emission and strong solid–state fluorescence, these findings might be somewhat rationalized by aggregation induced emission (AIE) effect [
]. The powder X-ray diffraction (PXRD) pattern of the bulk material (e.g. 5a) does not correspond with the simulated PXRD pattern using the data from the single-crystal structure of the respective cage, presumably indicating a more complicated emission mechanism (Fig. S30). Thus, more experimental and theoretical studies are needed to shed light on the detailed mechanism, inter alia, involving the unusual (near–) white emission.
2.5 Stabilization of chloride hydrate clusters (2Cl–·H2O)
Cryptands and aza–cryptands have been well–established to work as versatile anionic receptors, inter alia, in their protonated form, for negatively charged guest species [
]. Inspired by those findings in the literature, a selected example, 6a containing 12 NH units, was subject to protonation with strong acids, e.g. hydrochloric acid and H2SO4. Specifically, superphane 6a proves not soluble in commonly used solvents, but it becomes soluble in either DMSO or a mixture of DMSO and water (5:1, v/v) and only one set of resonances were observed upon treatment of 6a with excess HCl or H2SO4 (Figs. S31 and S32). These findings allowed us to suggest that 6a was capable of being fully protonated with HCl or H2SO4 in relatively flexible conformation at 298 K.
We next sought to explore whether NH–bearing superphane 6a in its protonated manners could serve as a receptor for binding guest species of interest. Initial evidence for anion encapsulation of protonated 6a came from the single crystal X–ray structure of its hydrochloric complex. Specifically, suitable crystals were obtained by allowing a solution of HCl–protonated 6a in water to evaporate slowly at room temperature. The resulting crystal structure revealed a protonated complex in the solid state (Fig. 4a). Due to the limitations of crystallography and the disorder of the solvent molecules and the counter chloride outside the cavity, the degree of protonation of 6a was not able to be exactly determined. In terms of the structure of protonated 6a, the two face–to–face benzene rings are clamped parallel on top of each other by six bridges bearing m–xylylenediamine moieties and the six toroidal bridges are almost uniformly distributed around the internal cavity, generating a lantern–like superstructure (Figs. S33 and S34). Notably, each bridge of superphane 6a has at least two NHs and one Ar–H as the binding sites, pointing to the inside cavity. These enable protonated 6a to feature a unique supramolecular receptor for binding specific guests with up to 18 hydrogen bonding donors. Interestingly, a closer inspection at this crystal structure revealed that a trimeric cluster consisting of two Cl− anions and one H2O molecule were found embedded right within the cavity of 6a, wherein the Cl−···Cl− and averaged Cl−···O(water) distances were measured to be 4.15 and 3.19 Å, respectively (Figs. 4b–d and S35). Each hydrogen atom of the water molecule points to one chloride anion, leading to a Cl−···H–O–H⋯Cl− complex, which as a whole was stabilized by multiple hydrogen bonds.
The ability of protonated 6a to bind chloride in solution was further probed via 1H NMR spectroscopy using a mixture of DMSO–d6 and D2O (5:1, v/v) as the solvent. As discussed above, spectroscopic analysis of the H2SO4–protonated 6a revealed only one set of resonance signals in the complex, indicating that the dominant species is 6a·H1212+, with all secondary amino groups protonated. Of particular note is that the aromatic Csp2–He in protonated 6a was observed to be located at 6.55 ppm, a relatively upfield region, presumably suggesting that no counter sulfate anion was captured within the central cavity of 6a. However, upon addition of excess Cl− (as its TBA salt) to a 1.0 mM solution of H2SO4–protonated 6a, the broad singlet associated with the Csp2–H protons seen at 6.55 ppm (e) underwent a pronounced downshift to 7.01 ppm. Meanwhile, the aromatic Csp2–H protons at 7.02 (c) and 6.93 (d) suffered a shift to 6.91 and 6.76 ppm, respectively, and the methylene protons showed chemical shift changes from 4.43 (a) and 4.04 ppm (b) to 4.51 and 4.23 ppm, respectively (Fig. S36). In contrast, fluoride and bromide complicated the host-guest interactions while iodide and nitrate were found ineffective to be bound by protonated 6a (Fig. S37). These findings led us to suggest that protonated superphane 6a binds effectively the Cl− anion well in solution.
To obtain greater insights into the binding of chloride by protonated 6a, 1H NMR spectroscopic titrations were carried out in a mixture of DMSO–d6 and D2O (5:1, v/v) using TBACl as the chloride anion source (Fig. S38). Specifically, upon incremental addition of 0–4 equivalents of Cl−, all proton signals ascribable to protonated 6a were observed to gradually decrease while a new set of resonances progressively increase, reflecting the first chloride binding event in slow exchange on the NMR time scale. Interestingly, apart from the increase, the new set of signals (e.g. e) corresponding to the chloride complex simultaneously underwent either downfield or upfield shifts regardless of the first saturation point achieved, presumably indicating the second chloride binding event in fast exchange on the NMR time scale. Consequently, the binding constants of K11 = (1.8 ± 0.3) × 103 and K12 = (1.1 ± 0.1) × 103 M−1 can be estimated by fitting the 1H NMR data corresponding to the Csp2–Hc, Csp2–Hd, and Csp2–He proton resonances to a 1:2 binding model (Fig. S39). Notably, although protonated 6a was weakly fluorescent in a mixture of DMSO and water (5:1, v/v), it proved silent to anions (e.g. Cl−) in the fluorescent spectroscopic studies (Fig. S40). These findings allowed us to conclude that the fully protonated 6a was probably capable of encapsulating two chloride anions, as reflected in the single crystal structure mentioned above.
A careful inspection of the 1H NMR spectrum of HCl–protonated 6a in DMSO–d6 at room temperature revealed the occurrence of a broad singlet around 8.47 ppm, which was able to be deuterated with D2O within 30 min, reflecting the active role of protons in question (Fig. S41). In contrast, no such active protons were seen in the case of H2SO4–protonated 6a in a mixture of DMSO–d6 and H2O (5:1, v/v). However, upon addition of excess TBACl to the beforementioned solution, a broad singlet at 8.50 ppm occurred (Fig. S42). These observations led us to suggest that water molecules might get involved in the binding of the chloride dimer, resulting a possible complexation of a Cl−···H–O–H⋯Cl− cluster in solution as observed in the single crystal structure. Anionic solvate clusters are omnipresent and their recognition and stabilization have been appreciated as a challenging topic and perpetual task. In this regard, superphanes might be appealing and promising.
A series of dodecaazasuperphanes, featuring exceptionally high (D6h) symmetry and multiple (up to 18) binding sites, have been established via a facile and efficient in–situ reductive amination strategy. The structural diversity could be achieved by the pre–modification of the dialdehydes. Furthermore, the feasible post–modification of secondary amine–based superphanes was exemplified by furnishing superphane 6e with various functional groups, viz. ethyl, allyl, propargyl and but–2–yn–1–yl. Importantly, compounds 5a–5e and 6a–6e under study were found to exhibit solid–state fluorescence with various emission colors ranging from bluish to (near–) white. Meanwhile, 6a–6e proved emissive in solution as well. Last but not least, with suitable internal apertures and appropriate binding sites as demonstrated by crystallography, and 1H NMR spectroscopy, superphane 6a in its protonated manners proved able to capture a 2Cl–·H2O cluster both in the solid state and in solution. This work establishes the first facile and efficient synthetic approach to emissive superphanes that could be broadly utilized for development of advanced emissive materials or as a new generation of supramolecular hosts for neutral species, anions, cations, ion pairs, as well as labile ionic solvate clusters or solvent clusters.
Conceptualization and supervision: QH; Synthesis, characterization, NMR, and photophysical experiments: AL, YL and YJ; Binding studies: WZ; Single crystal growing, data collection and analysis: AL and QH; Writing – original draft: ZL and QH; Writing – review & editing, QH. All authors proofread, commented on, and approved the final version of this manuscript.
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
This research was funded by the National Natural Science Foundation of China ( 22071050 and 21901069 to Q. H.), the Science and Technology Plan Project of Hunan Province, China (Grant No. 2019RS1018 to Q. H.), and Fundamental Research Funds for the Central Universities (Startup Funds to Q. H.). We thank Dr Zhenyi Zhang from Bruker (Beijing) Scientific Technology Co., Ltd for helpful discussions on X–ray crystallography.
Appendix A. Supplementary data
The following is the Supplementary data to this article: