Abstract
It is considered a more formidable task to precisely control the self-assembled products containing purely covalent components, due to a lack of intrinsic templates such as transition metals to suppress entropy loss during self-assembly. Here, we attempt to tackle this challenge by using directing groups. That is, the self-assembly products of condensing a 1:2 mixture of a tetraformyl and a biamine can be precisely controlled by slightly changing the substituent groups in the aldehyde precursor. This is because different directing groups provide hydrogen bonds with different modes to the adjacent imine units, so that the building blocks are endowed with totally different conformations. Each conformation favors the formation of a specific product that is thus produced selectively, including chiral and achiral cages. These results of using a specific directing group to favor a target product pave the way for accomplishing atom economy in synthesizing purely covalent molecules without relying on toxic transition metal templates.
Similar content being viewed by others
Introduction
Mother Nature avoids byproducts in synthesis by taking advantage of reversible bonding forces1,2,3,4. The reversible nature of these supramolecular interactions allows synthetic errors to be checked and corrected. This biological capability inspires chemists to employ either noncovalent forces5,6,7 or dynamic bonds8,9,10,11,12,13,14 as the reaction motifs in synthesizing artificial systems. Henceforth, high-yielding syntheses could be accomplished without relying on tedious stepwise procedures, by sophisticatedly designing and tuning the geometries and conformations of the building blocks that allows the corresponding target products to represent the thermodynamic minima in self-assembly. One of the most successfully developed reversible reactions is metal-ligand coordination15,16,17,18,19,20,21,22,23,24,25,26,27,28. A variety of molecules with complex architectures and topologies have been successfully self-assembled29,30,31,32,33,34, some of which16,17,21,22,23,30 were obtained in close to quantitative yields. Here, transition metal cations with fixed coordination modes are able to dictate the corresponding organic ligands to orientate in specific manners that favor the formation of some specific products. Subtle changes of the organic ligands in geometry and/or size might lead to dramatic variation in self-assembly pathway35,36,37,38,39,40,41,42. The same level of success, however, has not been achieved in the systems containing purely covalent components that are often relatively more flexible, in which the intrinsic templates namely transition metals are absent. The implication is that, the self-assembled products via dynamic covalent chemistry8,9,10,11,12,13,14,43,44,45 such as imine formation46,47,48,49,50,51,52,53,54,55,56,57 are thus often less controllable compared to the coordinative counterparts, despite a few exceptions56,58,59,60,61,62,63. For example, Cooper64 et al. demonstrated that the amino precursors containing odd or even numbers of methylene units favored the formation of [2 + 3] or [4 + 6] cages, respectively. Mastalerz13 et al. discovered that an organic cage self-assembled via boronic ester bond formation could undergo dimerization and form a catenane in solid-state, when switching the constitution of the side chains in the tetraol precursor. More recently, the same group65 indicated that introducing methoxy or thiomethyl unit onto the framework of an imine cube led to occurrence of dimerization and trimerization, forming catenanes in solution. Mukherjee66 et al. employed intramolecular hydrogen bonding to direct self-sorting. The group led by Beuerle67 obtained a highly strained organic cage whose formation would be otherwise unlikely to occur without intramolecular driving forces namely hydrogen bonds.
In the present work, we employed directing groups to precisely control the self-assembly products based on imine formation. These directing groups mediate or determine the conformations of the building blocks, by providing hydrogen bonds with different modes to the latter. Each of these preorganized conformations of the building blocks favors one specific cage compound that is produced as the predominant product. To be more specific, a pseudo-linear tetraformyl precursor containing two isophthalaldehyde units and trans-cyclohexane-1,2-diamine (trans-CHDA), which is either enantiomerically pure or the racemic mixture, are combined for self-assembly. Subtle variation in the substituents located in each isophthalaldehyde residue switches the self-assembly products between two types of constitutionally different cage molecules, including a [3 + 6] chiral cage and a [2 + 4] achiral cage. In the case of OH unit whose acid proton acts as a hydrogen bond donor, intramolecular hydrogen bonds drive the two formyls and/or the resultant imine units on both sides to orientate in an exo-endo conformation. Such conformation favors the production of a [3 + 6] chiral cage, which is composed of three equivalents of the tetraformyl precursor and six equivalents of enantiomerically pure trans-CHDA. When the racemic trans-CHDA is used in self-assembly, narcissistic self-sorting occurs, generating a pair of enantiomers of the [3 + 6] chiral cage each containing only one type of enantiomer of trans-CHDA. As a comparison, an alkoxy substituent containing no acidic protons affords the two formyls and/or imines an exo-exo conformation68. When the tetraformyl precursor is combined with the racemic mixture of trans-CHDA in a 1:2 ratio, a meso [2 + 4] cage is produced as the only observable product, which is composed of two equivalents of the tetraformyl precursor and four equivalents of racemic trans-CHDA. More interestingly, when the substituent is an ester unit containing protons with modest acidity, both exo-endo and exo-exo conformations become thermodynamically feasible. The self-assembled products are then determined by the chirality of the bisamine precursors. That is, enantiomerically pure and racemic bisamine favor the [3 + 6] and [2 + 4] products, respectively. Physicochemical analysis based on NMR spectroscopic results, solid-state structures, as well as theoretical calculation results indicated that these self-assembly preferences stem from the self-assembly products attempting to minimize intramolecular steric hindrance, by keeping all the imine protons in the syn conformation69,70 relative to the corresponding adjacent methine proton in the cyclohexyl unit.
Results
Each of the five structurally analogous tetraaldehyde precursors 0-4 (Fig. 1), whose synthetic procedures are described in the Supplementary Information (Supplementary Figs. 5, 7, 9, 11 and 14), contains two isophthalaldehyde units. The differences between these five precursors lie in the substituents grafted in each of the isophthalaldehyde moiety between the two formyl units (Fig. 1), which are ‒H (0), ‒OH (1), ‒OC4H9 (2), ‒OCH2COOC2H5CH3 (3), and ‒OCH2COOC(CH3)3 (4), respectively. A pair of enantiomers of a chiral bisamine, namely (1S,2S)-cyclohexane-1,2-diamine ((S,S)-CHDA) and (1R,2R)-cyclohexane-1,2-diamine ((R,R)-CHDA), as well as ethylenediamine (EDA) were used as the amino partners for self-assembly (Fig. 1).
We first combined 1 (2.5 mM) and (S,S)-CHDA (Fig. 2A) in a 1:2 ratio in CDCl3. After heating the solution at 50 °C for 6 h, the 1H NMR spectrum (Fig. 2B) was recorded, in which a set of sharp resonances were observed, indicating a product with a symmetrical structure was obtained as the predominant product. Mass spectrum (Supplementary Fig. 28) indicated that this product is a [3 + 6] product, namely composed of three equivalents of 1 and six equivalents of (S,S)-CHDA. This product is referred to as 13S6. In the 1H NMR spectrum (Fig. 2B), each of the resonances corresponding to the protons in the isophthalaldehyde residues, including both the imine e/e’ and the phenyl c/c’, splits into two peaks. This observation indicated that in each isophthalaldehyde residue, the two imine units have two different orientations, namely either exo or endo with respect to the central OH group (Fig. 3B, middle). The framework of 13S6 is chiral, resulting from the stereo chirality of the (S,S)-CHDA precursor. The chirality of 13S6 was supported by 1H NMR spectrum (Supplementary Fig. 29) in which the two protons in the methylene f become diastereotopic. The solid-state sample of 13S6 was obtained by adding MeOH into its solution in chloroform and collecting the precipitate via filtration. The solid of 13S6 was re-dissolved in CDCl3, whose 1H NMR spectrum was essentially the same as the before precipitation, indicated that 13S6 was rather kinetically inert and did not undergo observable degradation during precipitation and re-dissolving. In the 1H NMR spectrum (Fig. 2B) of 13S6, a few small broad resonances were observed, indicating that some oligomeric or polymeric byproducts were also generated. These broad resonances were not removed, even after we attempted to purify 13S6 via precipitation. We thus used NMR yield to quantify the production of each cage in this article, by adding an internal standard in the NMR sample. The self-assembly yield of 13S6 was determined to be 60% (Supplementary Fig. 76). The cage 13R6, which is the enantiomer of 13S6, was also self-assembled in a similar procedure (Supplementary Fig. 16C), by condensing 1 and (R,R)-CHDA in CDCl3. The circular dichroism (CD) spectra (Supplementary Fig. 33B) of both 13S6 and 13R6 were recorded, showing mirror-like images. When 1 was combined with a racemic mixture of (R,R)-CHDA and (S,S)-CHDA in CDCl3, narcissistic self-sorting71,72,73 occurred (Supplementary Fig. 17), yielding a racemic mixture of 13S6 and 13R6 as the major product.
Single crystals of the cage 13S6 (Fig. 2C), as well as the racemic mixture (Supplementary Fig. 88) of 13S6 and 13R6, were obtained by vapor diffusion of methanol or acetonitrile into the corresponding solutions in CHCl3 or DMF, respectively. The solid-state structures of both 13S6 and 13R6 convinced the exo-endo conformation of the two imine units in each isophthalaldehyde residue, consistent with the 1H NMR spectroscopic results. This exo-endo conformation is favored by the formation of intramolecular hydrogen bonds (Fig. 3B, middle), namely that the oxygen atom in the central OH forms hydrogen bond with the imine proton on one side, while the OH proton forms hydrogen bond with the imine nitrogen atom on the other side. These two types of hydrogen bonds are clearly observed in the solid-state structures (Fig. 2C, black dashed lines). Such exo-endo conformation was also observed in many cage systems containing trans-CHDA61,74,75,76. The exo-endo conformation allows all the imine protons to orientate in a syn conformation relative to the corresponding adjacent methine protons (Fig. 2C, red double-head arrows). According to a report by Gawronski70, such syn conformer is thermodynamically more favored than the anti counterpart by 0.92 kcal/mol, due to larger steric hindrance in the latter conformer (Fig. 3E). We also combined 1 and EDA, which is a less preorganized counterpart of trans-CHDA. Heating the mixture in CDCl3 led to the generation of a library of oligomeric and polymeric precipitates, instead of a putative [3 + 6] cage namely 13(EDA)6. Such results indicate that conformation preorganization is also of importance in the bisamino precursors.
In order to strengthen our proposition that the OH group in 1 plays a predominant role in favoring the formation of the corresponding [3 + 6] cages namely either 13S6 or 13R6, we synthesized three analogs including 0, 2 as well as 3, in which the central OH substituents in 1 are replaced by other units namely H, alkyloxy and ester respectively. The precursor 0 can be considered as a counterpart of 1 without any directing groups. 0 and (S,S)-CHDA were combined in CDCl3 in a 1:2 ratio. After heating the corresponding solution at 50 °C for 4 h, the 1H NMR spectrum (Supplementary Fig. 23A) indicated that an analog of 13S6, namely 03S6 was produced as the major product, accompanied with a [2 + 4] product 02S4 as a kinetic product. The formation of both 03S6 and 02S4 was convinced by mass spectrometry (Supplementary Fig. 24). Further heating the reaction mixture for 120 h converted most 02S4 into 03S6, as inferred from the observation that the molecular ion peak corresponding to the [2 + 4] product weakened significantly (Supplementary Fig. 25). Even heating the solution for no less than 120 h, the resonances corresponding to 02S4 with small intensity were still observable in the 1H NMR spectrum (Supplementary Fig. 23C), indicating that the conversion from 02S4 into 03S6 was not complete. The NMR yield of 03S6 was determined to be 70% (Supplementary Fig. 84). Compared to 1, the precursor 0 is lack of the OH directing group, so that the 03S6 is thermodynamically less favored. Combining 0 and the racemic CHDA in a 1:2 ratio also produced the racemic mixture of 03S6 and 03R6 as major products. However, this narcissistic self-sorting was much less successful compared to the aforementioned system involving 1, i.e., more oligomeric or polymeric byproducts (Supplementary Fig. 27) were observed in the corresponding 1H NMR spectrum.
We then combined either 2 or 3 (2.5 mM) with a racemic mixture of trans-CHDA in a 1:2 ratio in CDCl3. After heating at 50 °C for 6 h, the 1H NMR spectra (Fig. 4B and Supplementary Fig. 48) and mass spectra (Supplementary Figs. 34 and 47) of both solutions were recorded. Mass spectrum indicated that in both cases, [2 + 4] products were generated. That is, each product is composed of two equivalents of tetraformyl precursors (i.e., 2 or 3) and four equivalents of trans-CHDA, namely either (S,S)-CHDA or (R,R)-CHDA.
Single crystals of these two self-assembled products were obtained by vapor diffusion of either methanol or diethyl ether into the corresponding solutions in CHCl3, respectively. The solid-state structures unambiguously indicated that two cage products, namely 22R2S2 and 32R2S2 were obtained (Fig. 4C–F). In the framework of both 22R2S2 and 32R2S2, the two imine bonds in each of the isophthalaldehyde residues orientate in an exo-exo manner. This conformation is in sharp contrast to the exo-endo conformation in the case of either 13S6 or 13R6 containing OH substituents. The exo-exo conformation is favored by the formation of intramolecular five-member ring hydrogen bonds between the central oxygen atom and the two imine protons on both sides (Fig. 3C, left), assisted by the repulsion between the central oxygen and two imine nitrogen atoms. In the framework of either 22R2S2 or 32R2S2, the four trans-CHDA residues distribute in an RSRS manner. That is, each isophthalaldehyde residue is connected by two different enantiomers of trans-CHDA. Such ligand distribution affords each cage namely either 22R2S2 or 32R2S2 two plane symmetries, affording both 22R2S2 and 32R2S2 meso structures. In addition, as expected, each imine proton adopts the more favored syn conformation with respect to the corresponding axial methine proton.
The 1H NMR spectra (Fig. 4B and Supplementary Fig. 48) of both 22R2S2 and 32R2S2 are similar. In the 1H NMR spectrum (Fig. 4B) of 22R2S2, a few singlets corresponding to the protons including e, c and g are observed. The simple patterns of both 1H NMR spectra convince that both 22R2S2 and 32R2S2 are highly symmetrical, consistent with the corresponding solid-state structure. It is noteworthy that because both (R,R)-CHDA and (S,S)-CHDA are involved in self-assembly, the cage containing four trans-CHDA residues has theoretically nine stereoisomers. However, the 1H NMR spectra unambiguously indicates that only the RSRS ones, namely 22R2S2 or 32R2S2, is produced selectively. This is because the RSRS isomer is the only one that can allow all the imine protons to adopt the more favored syn conformation with respect to corresponding axial methine protons. This unlikely occurring social self-sorting behavior was also reported by Mastalerz recently77. Both 22R2S2 and 32R2S2 were isolated as solid-state compounds by adding MeOH into the corresponding solutions in chloroform and collecting the precipitates via filtration. Again, NMR yields were used to quantify their production, given that oligomeric or polymeric impurities rendered the isolated yields less accurate. By using internal standard in the corresponding NMR samples, the yields (Supplementary Figs. 77 and 79) of 22R2S2 and 32R2S2 were determined to be 57% and 75%, respectively. We also combined either 2 or 3 (2.5 mM) with EDA (5.0 mM) in CDCl3, yielding two self-assembled products whose 1H NMR spectra (Supplementary Figs. 43 and 53) are very similar as those of 22R2S2 and 32R2S2. Mass spectra (Supplementary Figs. 42 and 52) confirmed that two cages namely 22(EDA)4 and 32(EDA)4 were self-assembled, whose yields (Supplementary Figs. 78 and 80) are 50 % and 85 %, respectively. It is still unclear why the yields of 32R2S2 and 32(EDA)4 are higher compared to 22R2S2 and 22(EDA)4. We hypothesize that the precursor 3 contains two ester side chains. The protons in the methylene units that are connected directly by the ester oxygen atoms are more acidic compared to the methylene protons in 2. The former protons form stronger CH-π interactions with the phenyl units in the cage framework.
When the racemic trans-CHDA was replaced by enantiomeric pure (S,S)-CHDA, self-assembly yielded totally different products. Heating a 1:2 mixture of 2 (2.5 mM) and (S,S)-CHDA in CDCl3 produced a library of mixture, whose 1H NMR spectrum (Supplementary Fig. 19B) showed a set of irregular resonances. Such observation indicated that the putative chiral cage namely 22S4 is not a thermodynamically favored product. The putative cage 22S4 can be considered as a chiral counterpart of 22R2S2, in which two (R,R)-CHDA residues are replaced by two (S,S)-CHDA residues. However, in 22S4, four of the eight imine protons adopt the thermodynamically disfavored anti conformation (Supplementary Fig. 19) with respect to the adjacent methine protons, making its formation less favored. Another putative product, namely 23S6, an analog of 13S6, was not observed in either 1H NMR spectrum or mass spectrum. This is not surprising, given that the [3 + 6] products require the exo-endo conformation, while the alkyloxy directing groups in 2, favor the exo-exo conformation in contrast.
To our surprise, combining the tetraaldehyde 3 (2.5 mM) and (S,S)-CHDA in CDCl3 in a 1:2 ratio yielded a product whose 1HNMR spectrum (Fig. 5B) was in reminiscence of the aforementioned [3 + 6] product namely 13S6. Mass spectrum (Supplementary Fig. 57) confirms that a chiral cage 33S6 was self-assembled as the predominant product, whose yield was determined to be 63 % (Supplementary Fig. 81). As occurred in 13S6, the resonances corresponding to two imine protons in 33S6 also split into two peaks (Fig. 5B), indicating that the two imines within each isophthalaldehyde residue adopt an exo-endo conformation. The exo-endo conformation was also supported by the NOESY spectrum (Supplementary Fig. 61) of 33S6. That is, while the endo imine proton e’ undergoes coupling with one phenyl proton c’ in the isophthalaldehyde residue, the exo imine proton e undergoes coupling with the CH2 proton k in the central side chain. The exo-endo conformation in the case of 33S6 was also driven by the formation of two different types of hydrogen bonds (Fig. 3D, middle). That is, one hydrogen bond forms between exo imine proton e and the central oxygen atom, while the second one forms between the endo imine nitrogen atom and the proton in the methylene unit h/h’ in ethyl unit of the ester. The occurrence of the latter hydrogen bonding results from the electron-withdrawing inductive effect of the ester oxygen atom, which renders the protons h/h’ relatively acidic. Such relatively acidic protons are absent in the case of 2 containing n-butoxy chains. It is therefore not surprising that 2 did not form the putative cage 23S6. The occurrence of CH‒N hydrogen bonding forces in 33S6 is supported by its 1HNMR spectrum (Fig. 5B), in which the resonances corresponding to the methylene unit h/h’ in the ester split into two peaks, while the methylene unit k exhibits a singlet. In order to further strengthen our hypothesis that the acidic ester proton plays an important role in cage formation, we thus synthesized another tetraformyl precursor 4 (see its molecular formula in Fig. 1) containing tert-butyl acetate units. 4 contains ester functions while does not contain the protons with modest acidity as those in 3. Combining 4 and (S,S)-CHDA in a 1:2 ratio yielded a library of mixture (Supplementary Fig. 22), instead of 43S6. This control experiment unambiguously supported our proposition that hydrogen bonding involving the ester side chains in 3 plays a critical role in favoring the formation of 33S6. Addition of (R,R)-CHDA into 33S6 gradually transformed latter into the [2 + 4] achiral cage 32R2S2 (Fig. 5A). Furthermore, combining a mixture of pre-synthesized 33S6 and 33R6 with a 1:1 ratio also yielded the meso cage 32R2S2. Apparently, the achiral cage 32R2S2 is more favored in terms of entropy compared with 33S6, because the latter is composed of fewer building blocks compared to the former.
In order to confirm that the self-assembly preference results from the cage products attempting to keep all the imine protons in the syn conformation with respect to the corresponding methine protons, we further performed density functional theory (DFT) calculations at the BP86-D3/6-311G(d) level with the Gaussian 16 package78. Based on the solid-state structure of 22R2S2 obtained via crystallography, two putative cages, namely 22S4 and 22S2R2, were optimized (see details in Supplementary Data 1). Here, 22S4 is a putative counterpart of 22R2S2, whose two (R,R)-CHDA residues are replaced with two (S,S)-CHDA residues. 22S2R2 was obtained by replacing the two (R,R)-CHDA and two (S,S)-CHDA residues with (S,S)-CHDA and (R,R)-CHDA, respectively. In 22R2S2, 22S4 and 22S2R2, there are respectively zero, four and eight imine protons that adopt the less favored anti conformation with respect to the corresponding methine protons. The theoretical results revealed that the free energies of 22S4 and 22S2R2 are 5.4 kcal/mol and 9.1 kcal/mol with respect to that of 22R2S2, respectively (Fig. 6). This confirms that the syn conformer is thermodynamically more favored than the anti counterpart in the cage framework. 22R2S2, with all the imine protons in the syn conformation, is the most stable and favored product, which is fully consistent with our experimental results. Similar approaches were also used to calculate the free energies of 13S6 and a putative cage 13R6(anti) (Supplementary Fig. 89). It is noteworthy that 13S6 and 13R6(anti) are not a pair of enantiomers. The putative cage 13R6(anti) was obtained by replacing all (S,S)-CHDA residues in 13S6 with (R,R)-CHDA, while keeping the conformations of all the imine bonds (details see Supplementary Data 1). In the putative cage 13R6(anti), all imine protons adopt the anti conformation with respect to the adjacent methine protons. The calculations revealed that the free energy of 13R6(anti) is 37.4 kcal/mol higher than that of 13S6 (Supplementary Fig. 89).
Discussion
In summary, an ingenious approach to precisely control the self-assembly product based on imine condensation is developed, by introducing different directing groups to favor specific target products. When a tetraformyl precursor containing two isophthalaldehyde units and trans-CHDA are combined, the self-assembly products are very sensitive to the variation of the central substituents located in each isophthalaldehyde residue between two imine units. This is because the central directing groups provide hydrogen bonding with different modes to imine building blocks located on both sides. These intramolecular forces endow the imine units with specific conformations, each resembling and favoring a specific cage product that is produced selectively. To be specific, in the case of OH substituent group whose proton is rather acidic, the two imine bonds are preorganized in an exo-endo conformation, driven by two different hydrogen bonds in the form of either CH‒O or OH‒N. Such conformation favors the formation of a chiral [3 + 6] cage containing enantiomeric pure trans-CHDA residues. When the substituent group contains no acidic protons such as an alkoxyl unit, the two imine bonds have an exo-exo conformation driven by two identical hydrogen bonds namely CH‒O. Such conformation favors the formation of achiral [2 + 4] cages containing racemic trans-CHDA building blocks as the predominant product, implying the occurrence of social self-sorting. In the case of ester substituent group containing protons with modest acidity, both the exo-endo and exo-exo conformations are thermodynamically feasible. As a consequence, the self-assembly products are surprisingly determined by the chirality of the trans-CHDA, namely that enantiomeric pure and racemic trans-CHDA favor the [3 + 6] and [2 + 4] products respectively. Such preference results from a tendency that the imine protons attempt to adopt the syn conformation relative to the adjacent axial methine protons in the CHDA residues, in order to minimize steric hindrance. Our fundamental understanding of precisely controlling the thermodynamic stability of a target product by using directing group to regulate the intramolecular forces, is thus significantly improved. Such results help us to rule out the reliance of cationic transition metal templates that might be highly toxic. Future research includes self-assembly of cage molecules with larger cavities and water-solubility, so that these self-assembled hosts are employed for applications in some more challenging arenas, such as mimicking enzyme79,80,81,82 in artificial systems.
Methods
Self-assembly of 13S6
A 1:2 mixture of 1 (2.77 mg, 0.005 mmol) and (S,S)-CHDA (1.14 mg, 0.01 mmol) was combined and dissolved in CDCl3 (2 mL). The corresponding reaction mixture was heated at 50 °C for 6 h. 13S6 was self-assembled as the major product in the corresponding 1H NMR spectrum, without further manipulation. The solid-state sample of 13S6 was obtained by adding MeOH into its solution in chloroform and collecting the precipitate via filtration. However, it is unsuccessful to purify 13S6 via precipitation, accompanied by the generation of some oligomeric or polymeric byproducts. We thus used a NMR yield (60 %) to quantify the production of 13S6, by adding an internal standard in the NMR sample.
Self-assembly of 22R2S2, 32R2S2 and 42R2S2
A 1:2 mixture of the corresponding tetraformyl precursor (2, 3 or 4, 0.005 mmol) and racemic trans-CHDA (1.14 mg, 0.01 mmol) was combined and dissolved in CDCl3 (2 mL). The corresponding reaction mixture was heated at 50 °C for 6 h. A [2 + 4] achiral cage (22R2S2, 32R2S2 or 42R2S2) was self-assembled as the major product in the 1H NMR spectrum, without further manipulation. The solid-state sample of each cage was obtained by adding MeOH into its solution in chloroform and collecting the precipitate via filtration. However, it is also unsuccessful to purify these cages via precipitation. We thus used NMR yields to quantify the production of 22R2S2 (57 %), 32R2S2 (75 %) and 42R2S2 (71 %), by adding an internal standard in the NMR sample.
Self-assembly of 22(EDA)4, 32(EDA)4 and 42(EDA)4
A 1:2 mixture of the corresponding tetraformyl precursor (2, 3 or 4, 0.005 mmol) and EDA (0.60 mg, 0.01 mmol) was combined and dissolved in CDCl3 (2 mL). The corresponding reaction mixture was heated at 50 °C for 6 h. A [2 + 4] achiral cage (22(EDA)4, 32(EDA)4 or 42(EDA)4) was self-assembled as the major product in the 1H NMR spectrum, without further manipulation. The NMR yield of 22(EDA)4, 32(EDA)4 and 42(EDA)4 was determined to be 50%, 85% and 77%, respectively.
Self-assembly of 33S6
A 1:2 mixture of 3 (3.63 mg, 0.005 mmol) and (S,S)-CHDA (1.14 mg, 0.01 mmol) was combined and dissolved in CDCl3 (2 mL). The corresponding reaction mixture was heated at 50 °C for 12 h. 33S6 was self-assembled as the major product in the corresponding 1H NMR spectrum, without further manipulation. The NMR yield of 33S6 was determined to be 63%.
Self-assembly of 03S6
A 1:2 mixture of 0 (3.13 mg, 0.006 mmol) and (S,S)-CHDA (1.37 mg, 0.012 mmol) was combined and dissolved in CDCl3 (2 mL). The corresponding reaction mixture was heated at 50 °C for 24 h. 03S6 was self-assembled as the major product accompanied with a [2 + 4] product as a kinetic product in the corresponding 1H NMR spectrum. As the reaction proceeding, the [2 + 4] kinetic product would transfer to the [3 + 6] chiral cage mostly, but not completely. The NMR yield of 03S6 was determined to be 70%.
General methods
Nuclear magnetic resonance (NMR) spectra were recorded at ambient temperature using Bruker AVANCE III 400, Bruker AVANCE III 500, or Agilent DD2 600 spectrometers, with working frequencies of 400/500/600 and 100/125/150 MHz for 1H and 13C, respectively. Chemical shifts are reported in ppm relative to the residual internal non deuterated solvent signals (CDCl3: δ = 7.26 ppm, DMSO-d6: δ = 2.50 ppm). High-resolution mass spectra (HRMS) were measured by using a SHIMADZU liquid chromatograph mass spectrometry ion trap time of flight (LCMS-IT-TOF) instrument and Bruker Daltonics Autoflex III (MALDI-TOF). X-ray crystallographic data were collected on a Bruker D8 Venture diffractometer. CD spectra were recorded on a Circular Dichroism Spectrometer (Chirascan V100, Applied Photophysics Ltd).
Theoretical calculations
All investigated cage structures were optimized by using the density functional theory (DFT) at the BP86-D3/6-311G(d) level with the Gaussian 16 package78. The solvent effect of chloroform was included with the polarizable continuum model (PCM)83. All the optimized structures were verified by the phonon frequencies calculated at the same level (namely no imaginary frequency should exist).
Data availability
The authors declare that all other data supporting the findings of this study are available from the article and its Supplementary Information. Cartesian coordinates and energies of all investigated cage structures are provided in Supplementary Data 1. The X-ray crystallographic coordinates for structures reported in this study have been deposited at the Cambridge Crystallographic Data Centre (CCDC), under deposition numbers 2201823, 2201825, 2203208 and 2203209. These data can be obtained free of charge from The Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/structures. Additional data are available from the authors upon request.
References
Grimes, J. M. et al. The atomic structure of the bluetongue virus core. Nature 395, 470–478 (1998).
Wikoff, W. R. et al. Topologically linked protein rings in the bacteriophage HK97 capsid. Science 289, 2129–2133 (2000).
Ahnert, S. E., Marsh, J. A., Hernández, H., Robinson, C. V. & Teichmann, S. A. Principles of assembly reveal a periodic table of protein complexes. Science 350, aaa2245 (2015).
Thorp, H. H. Proteins, proteins everywhere. Science 374, 1415 (2021).
Martín, T., Obst, U. & Rebek, J. Molecular assembly and encapsulation directed by hydrogen-bonding preferences and the filling of space. Science 281, 1842–1845 (1998).
Trembleau, L. & Rebek, J. Helical conformation of alkanes in a hydrophobic cavitand. Science 301, 1219–1220 (2003).
Yang, D. et al. Air-and light-stable P4 and As4 within an anion-coordination-based tetrahedral cage. J. Am. Chem. Soc. 139, 5946–5951 (2017).
Ponnuswamy, N., Cougnon, F. B. L., Clough, J. M., Pantoş, G. D. & Sanders, J. K. M. Discovery of an organic trefoil knot. Science 338, 783–785 (2012).
Li, H. et al. Quantitative self-assembly of a purely organic three-dimensional catenane in water. Nat. Chem. 7, 1003–1008 (2015).
Rowan, S. J., Cantrill, S. J., Cousins, G. R. L., Sanders, J. K. M. & Stoddart, J. F. Dynamic covalent chemistry. Angew. Chem. Int. Ed. 41, 898–952 (2002).
Wang, L., Vysotsky, M. O., Bogdan, A., Bolte, M. & Böhmer, V. Multiple Catenanes derived from calix[4]arenes. Science 304, 1312–1314 (2004).
Jin, Y., Yu, C., Denman, R. J. & Zhang, W. Recent advances in dynamic covalent chemistry. Chem. Soc. Rev. 42, 6634–6654 (2013).
Zhang, G., Presly, O., White, F., Oppel, I. M. & Mastalerz, M. A shape-persistent quadruply interlocked giant cage catenane with two distinct pores in the solid state. Angew. Chem. 126, 5226–5230 (2014).
Lee, S., Yang, A., Moneypenny, T. P. & Moore, J. S. Kinetically trapped tetrahedral cages via alkyne metathesis. J. Am. Chem. Soc. 138, 2182–2185 (2016).
Fujita, M. et al. Self-assembly of ten molecules into nanometre-sized organic host frameworks. Nature 378, 469–471 (1995).
Olenyuk, B., Whiteford, J. A., Fechtenkotter, A. & Stang, P. J. Self-assembly of nanoscale cuboctahedra by coordination chemistry. Nature 398, 796–799 (1999).
Takeda, N., Umemoto, K., Yamaguchi, K. & Fujita, M. A nanometre-sized hexahedral coordination capsule assembled from 24 components. Nature 398, 794–796 (1999).
Biros, S. M., Yeh, R. M. & Raymond, K. N. Design and formation of a large tetrahedral cluster using1,1’-binaphthyl ligands. Angew. Chem. Int. Ed. 47, 6062–6064 (2008).
Percástegui, E. G., Ronson, T. K. & Nitschke, J. R. Design and applications of water-soluble coordination cages. Chem. Rev. 120, 13480–13544 (2020).
Leininger, S., Olenyuk, B. & Stang, P. J. Self-assembly of discrete cyclic nanostructures mediated by transition metals. Chem. Rev. 100, 853–908 (2000).
Umemoto, K., Yamaguchi, K. & Fujita, M. Molecular paneling via coordination: guest-controlled assembly of open cone and tetrahedron structures from eight metals and four ligands. J. Am. Chem. Soc. 122, 7150–7151 (2000).
Seidel, S. R. & Stang, P. J. High-symmetry coordination cages via self-assembly. Acc. Chem. Res. 35, 972–983 (2002).
Nitschke, J. R. Mutual stabilization between imine ligands and copper(I) ions in aqueous solution. Angew. Chem. Int. Ed. 43, 3073–3075 (2004).
Fiedler, D., Leung, D. H., Bergman, R. G. & Raymond, K. N. Selective molecular recognition, C-H bond activation, and catalysis in nanoscale reaction vessels. Acc. Chem. Res. 38, 349–358 (2005).
Cook, T. R., Zheng, Y. R. & Stang, P. J. Metal-organic frameworks and self-assembled supramolecular coordination complexes: Comparing and contrasting the design, synthesis, and functionality of metal-organic materials. Chem. Soc. Rev. 113, 734–777 (2013).
Smulders, M. M. J., Riddell, I. A., Browne, C. & Nitschke, J. R. Building on architectural principles for three-dimensional metallosupramolecular construction. Chem. Soc. Rev. 42, 1728–1754 (2013).
Han, M., Engelhard, D. M. & Clever, G. H. Self-assembled coordination cages based on banana-shaped ligands. Chem. Soc. Rev. 43, 1848–1860 (2014).
Cook, T. R. & Stang, P. J. Recent developments in the preparation and chemistry of metallacycles and metallacages via coordination. Chem. Soc. Rev. 115, 7001–7045 (2015).
Cui, Z. & Jin, G.-X. Construction of a molecular prime link by interlocking two trefoil knots. Nat. Synth. 1, 635–640 (2022).
Zhang, H.-N., Lin, Y.-J. & Jin, G.-X. Selective construction of very large stacking-interaction-induced molecular 818 metalla-knots and borromean ring using curved dipyridyl ligands. J. Am. Chem. Soc. 143, 1119–1125 (2021).
Cui, Z., Gao, X., Lin, Y.-J. & Jin, G.-X. Stereoselective self-assembly of complex chiral radial [5]catenanes using half-sandwich rhodium/iridium building blocks. J. Am. Chem. Soc. 144, 2379–2386 (2020).
Ashbridge, Z. et al. Vernier template synthesis of molecular knots. Science 375, 1035–1041 (2022).
August, D. P., Jaramillo-Garcia, J., Leigh, D. A., Valero, A. & Vitorica-Yrezabal, I. J. A chiral cyclometalated iridium Star of David [2]catenane. J. Am. Chem. Soc. 143, 1154–1161 (2021).
Leigh, D. A. et al. A molecular endless (74) knot. Nat. Chem. 13, 117–122 (2021).
Benchimol, E., Nguyen, B. T., Ronson, T. K. & Nitschke, J. R. Transformation networks of metal-organic cages controlled by chemical stimuli. Chem. Soc. Rev. 51, 5101–5135 (2002).
Jansze, S. M. et al. Ligand aspect ratio as a decisive factor for the self-assembly of coordination cages. J. Am. Chem. Soc. 138, 2046–2054 (2016).
Zou, Y.-Q. et al. Sterics and hydrogen bonding control stereochemistry and self-sorting in BINOL-based assemblies. J. Am. Chem. Soc. 143, 9009–9015 (2021).
Li, R.-J., Marcus, A., Fadaei-Tirani, F. & Severin, K. Orientational self-sorting: formation of structurally defined Pd4L8 and Pd6L12 cages from low-symmetry dipyridyl ligands. Chem. Commun. 57, 10023–10026 (2021).
Tominaga, M. et al. Finite, spherical coordination networks that self-organize from 36 small components. Angew. Chem. Int. Ed. 43, 5621–5625 (2004).
Sun, Q. F. et al. Self-assembled M24L48 polyhedra and their sharp structural switch upon subtle ligand variation. Science 328, 1144–1147 (2010).
Fujita, D. et al. Self-assembly of M30L60 icosidodecahedron. Chem 1, 91–101 (2016).
Fujita, D. et al. Self-assembly of tetravalent Goldberg polyhedral from 144 small components. Nature 540, 563–566 (2016).
Au-Yeung, H. Y., Pantos, G. D. & Sanders, J. K. M. Dynamic combinatorial synthesis of a catenane based on donor-acceptor interactions in water. Proc. Natl. Acad. Sci. USA 106, 10466–10470 (2009).
Klotzbach, S. & Beuerle, F. Shape-controlled synthesis and self-sorting of covalent organic cage compounds. Angew. Chem. Int. Ed. 54, 10356–10360 (2015).
Wang, Q. et al. Solution-phase dynamic assembly of permanently interlocked aryleneethynylene cages through alkyne metathesis. Angew. Chem. Int. Ed. 54, 7550–7554 (2015).
Liu, X., Liu, Y., Li, G. & Warmuth, R. One-pot, 18-component synthesis of an octahedral nanocontainer molecule. Angew. Chem. Int. Ed. 45, 901–904 (2006).
Quan, M. L. C. & Cram, D. J. Constrictive binding of large guests by a hemicarcerand containing four portals. J. Am. Chem. Soc. 113, 2754–2755 (1991).
Xu, D. & Warmuth, R. Edge-directed dynamic covalent synthesis of a chiral nanocube. J. Am. Chem. Soc. 130, 7520–7521 (2008).
Mastalerz, M. One-pot synthesis of a shape-persistent endo-functionalised nano-sized adamantoid compound. Chem. Commun. 38, 4756–4758 (2008).
Tozawa, T. et al. Porous organic cages. Nat. Mater. 8, 973–978 (2009).
Hasell, T. et al. Triply interlocked covalent organic cages. Nat. Chem. 2, 750–755 (2010).
Jin, Y., Voss, B. A., Noble, R. D. & Zhang, W. A Shape-persistent organic molecular cage with high selectivity for the adsorption of CO2 over N2. Angew. Chem. Int. Ed. 49, 6348–6351 (2010).
Mastalerz, M., Schneider, M. W., Oppel, I. M. & Presly, O. A salicylbisimine cage compound with high surface area and selective CO2/CH4 adsorption. Angew. Chem. Int. Ed. 50, 1046–1051 (2011).
Schneider, M. W. et al. Periphery-substituted [4+6] salicylbisimine cage compounds with exceptionally high surface areas: Influence of the molecular structure on nitrogen sorption properties. Chem. Eur. J. 18, 836–847 (2012).
Hong, S. et al. Porphyrin boxes: rationally designed porous organic cages. Angew. Chem. Int. Ed. 54, 13241–13244 (2015).
Wang, X. et al. Assembled molecular face-rotating polyhedra to transfer chirality from two to three dimensions. Nat. Commun. 7, 12469 (2016).
Acharyya, K. & Mukherjee, P. S. Organic imine cages: molecular marriage and applications. Angew. Chem. Int. Ed. 58, 8640–8653 (2019).
Cougnon, F. B. L., Ponnuswamy, N., Jenkins, N. A., Pantos, D. & Sanders, J. K. M. Structural parameters governing the dynamic combinatorial synthesis of catenanes in water. J. Am. Chem. Soc. 134, 19129–19135 (2012).
Teng, B. et al. Synthesis of a large, shape-flexible, solvatomorphic porous organic cagecryst. Growth Des. 19, 3647–3651 (2019).
Greenaway, R. et al. From concept to crystals via prediction: multi-component organic cage pots by social self-Sorting. Angew. Chem. Int. Ed. 58, 16275–16281 (2019).
Zhang, L. et al. A facile solution-phase synthetic approach for constructing phenol-based porous organic cages and covalent organic frameworks. Green. Chem. 22, 2498–2504 (2020).
Qu, H. et al. Molecular face-rotating cube with emergent chiral and fluorescence properties. J. Am. Chem. Soc. 139, 18142–18145 (2017).
Hu, Q.-P. et al. Chirality gearing in an achiral cage through adaptive binding. J. Am. Chem. Soc. 144, 6180–6184 (2022).
Jelfs, K. E. et al. In silico design of supramolecules from their precursors: odd-even effects in cage-forming reactions. J. Am. Chem. Soc. 135, 9307–9310 (2013).
Benke, B. P., Kirschbaum, T., Graf, J., Gross, J. H. & Mastalerz, M. Dimeric and trimeric catenation of giant chiral [8+12] imine cubes driven by weak supramolecular interactions. Nat. Chem. 15, 413–423 (2023).
Acharyya, K. & Mukherjee, P. S. Hydrogen-bond-driven controlled molecular marriage in covalent cages. Chem. Eur. J. 20, 1646–1657 (2014).
Schäfer, N., Bühler, M., Heyer, L., Röhr, M. I. S. & Beuerle, P.-D. F. Endohedral hydrogen bonding templates the formation of a highly strained covalent organic cage compound. Chem. Eur. J. 27, 6077–6085 (2021).
Schneider, M. W., Oppel, I. M., Griffin, A. & Mastalerz, M. Post-modification of the interior of porous shape-persistent organic cage compounds. Angew. Chem. Int. Ed. 52, 3611–3615 (2013).
Gawroński, J., Kołbon, H., Kwit, M. & Katrusiak, A. Designing large triangular chiral macrocycles: efficient [3+3] diamine-dialdehyde condensations based on conformational bias. J. Org. Chem. 65, 5768–5773 (2000).
Kwit, M. & Gawronski, J. Chiral calixsalen-type macrocycles from trans-1,2-diaminocyclohexane. Tetrahedron. Asymmetry 14, 1303–1308 (2003).
Gidron, Q. et al. Homochiral [2]catenane and bis[2]catenane from alleno-acetylenic helicates - A highly selective narcissistic self-sorting process. J. Am. Chem. Soc. 137, 12502–12505 (2015).
Rizzuto, F. J. & Nitschke, J. R. Narcissistic, integrative, and kinetic self-sorting within a system of coordination cages. J. Am. Chem. Soc. 142, 7749–7753 (2020).
Chen, Q. et al. Narcissistic self-sorting and enhanced luminescence via catenation in water. Mater. Today Chem. 23, 100679 (2022).
Zhang, L. et al. From discrete molecular cagesto a network of cages exhibiting enhanced CO2 adsorption capacity. Angew. Chem. Int. Ed. 56, 7787–7791 (2017).
Yang, W., Sun, N., Wang, X., Yu, B. & Wang, H. Racemic porous organic cage crystal with selective gas adsorption behaviors. Z. Anorg. Allg. Chem. 648, e2021003 (2022).
Skowronek, P., Warżajtis, B., Rychlewska, U. & Gawrońskia, J. Self-assembly of a covalent organic cage with exceptionally large and symmetrical interior cavity: the role of entropy of symmetry. Chem. Commun. 49, 2524–2526 (2013).
Wagner, P. et al. Chiral self-sorting of giant cubic [8+12] salicylimine cage compounds. Angew. Chem. Int. Ed. 60, 8896–8904 (2021).
Frisch, M. J. et al. Gaussian 16 Revision B. 01 (Gaussian Inc., Wallingford CT, 2016).
Yoshizawa, M., Tamura, M. & Fujita, M. Diels-Alder in aqueous molecular hosts: unusual regioselectivity and efficient catalysis. Science 312, 251–254 (2006).
Pluth, M. D., Bergman, R. G. & Raymond, K. N. Acid catalysis in basic solution: a supramolecular host promotes orthoformate hydrolysis. Science 316, 85–88 (2007).
Kaphan, D. M., Toste, F. D., Bergman, R. G. & Raymond, K. N. Enabling new modes of reactivity via constrictive binding in a supramolecular-assembly-catalyzed Aza-Prins cyclization. J. Am. Chem. Soc. 137, 9202–9205 (2015).
Mondal, B., Acharyya, K., Howlader, P. & Mukherjee, P. S. Molecular cage impregnated palladium nanoparticles: efficient, additive-free heterogeneous catalysts for cyanation of aryl halides. J. Am. Chem. Soc. 138, 1709–1716 (2016).
Scalmani, G. & Frisch, M. J. Continuous surface charge polarizable continuum models of solvation. I. General formalism. J. Chem. Phys. 132, 114110 (2010).
Acknowledgements
The research at Zhejiang University was supported by the Starry Night Science Fund of Zhejiang University Shanghai Institute for Advanced Study (No. SN-ZJU-SIAS-006 to H.L.). H.L. also want to thank the support from the Leading Innovation Team grant from Department of Science and Technology of Zhejiang Province (2022R01005). L.W. acknowledges support from the National Natural Science Foundation of China (No. 22273082) and the High Performance Computing Center in Department of Chemistry, Zhejiang University.
Author information
Authors and Affiliations
Contributions
H.L. conceived the project. Q.C., Z.L., Y.L., Y.C., S.Z., F.H., L.W. and H.L. prepared the manuscript. Q.C., Y.L., H.T., B.S. and Y.W. synthesized the molecules studied in this work. Q.C., Y.C., G.W., T.J. and S.Z. performed characterization of the key compounds. Q.C. and H.L analyzed experimental data and draw conclusions. Under the supervision of L.W., Z.L. performed theoretical calculations and analysis. All authors discussed the results and commented on the manuscript.
Corresponding authors
Ethics declarations
Competing interests
The authors declare no competing interests.
Peer review
Peer review information
Nature Communications thanks the anonymous, reviewer(s) for their contribution to the peer review of this work. A peer review file is available.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Chen, Q., Li, Z., Lei, Y. et al. The sharp structural switch of covalent cages mediated by subtle variation of directing groups. Nat Commun 14, 4627 (2023). https://doi.org/10.1038/s41467-023-40255-4
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41467-023-40255-4
Comments
By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.