Dynamic clipping strategy to synthesize [2]rotaxanes consisting of borate ion-containing crown ether and ammonium ions with kinetic and thermodynamic control†

Takumi Takizawa, Shinobu Miyagawa and Yuji Tokunaga*
Department of Materials Science and Engineering, Faculty of Engineering, University of Fukui, Bunkyo, Fukui 910-8507, Japan. E-mail: tokunaga@u-fukui.ac.jp

First published on 10th July 2025

---

---

Introduction

Dynamic covalent chemistry is a powerful method for constructing mechanically interlocked molecules. Reversible covalent bond formation generates stable products in high yields with thermodynamic control.¹ Imine bond formation,² disulfide exchange reactions,³ and olefin metathesis⁴ have been utilized to synthesize rotaxanes, catenanes, and related interlocked molecules.⁵ In particular, dynamic macrocyclization enables the efficient construction of interlocked molecules from acyclic precursors (clipping), allowing otherwise tedious macrocyclization to proceed via template-directed clipping.⁶ Additionally, macrocyclization allows the synthesis of highly ordered rotaxanes,⁷ catenanes,⁸ knots,⁹ and links.¹⁰ Imine bond formation is an ideal and efficient dynamic covalent approach because the reversibility of a reaction can be controlled by the addition of an acid or a base and the nitrogen atom of the imine serves as a hydrogen-bond acceptor to stabilize the interlocked architecture.

Recently, we reported the formation of spiroborate from an acyclic bis-catechol and boric acid (trimethyl borate) in the presence of a secondary amine via a dynamic covalent clipping rotaxane synthesis approach (Fig. 1).¹¹ The borate anion in the macrocycle and the ammonium cation derived from the amine formed an ion pair that increased the binding affinity between the axle and macrocyclic components without the need for any other ion pairing. Moreover, this approach prevented the preparation of low-solubility ammonium salts prior to the key rotaxane-forming reaction.

Herein, we report the following: (1) Various (pseudo)rotaxanes were synthesized by using different amines, and the importance of hydrogen bonding in the formation of the borate-containing crown ether/ammonium rotaxanes was investigated by varying both the solvent and the amine used. (2) We investigated the size of functional groups that can prevent dethreading of 23-membered borate-containing crown ether and found that a phenyl group was not sufficiently bulky but that a cyclohexyl group prevented dethreading at room temperature. (3) An axle component in a [2]rotaxane was exchanged with a free amine under thermodynamic conditions; thus, the borate-forming clipping reaction is a reversible dynamic covalent approach. (4) Borate-forming reactions involving two competing amines with different basicities were performed. The more acidic ammonium ion was an effective template for [2]rotaxane formation, and the [2]rotaxane formed from the basic amine was thermodynamically stable.

Results and discussion

Synthesis of (pseudo)rotaxanes: Role of hydrogen bonding in stabilizing the rotaxane structure

We previously synthesized a few (pseudo)[2]rotaxanes (3a–3d) from bis-catechol 1, benzylamines (2a–2d), and boric acid or trimethyl borate (Scheme 1).¹¹ In this study, we used different secondary amines (2e–2h) as axle component precursors during rotaxane synthesis, namely, bis(4-methoxybenzyl)amine 2e, bis(3,5-di-tert-butylbenzyl)amine 2f, bis[3,5-di(trifluoromethyl)benzyl]amine 2g, and bis(cyclohexylmethyl)amine 2h, which bear electron-donating, bulky, bulky electron-withdrawing, and cyclic alkyl groups, respectively.

A suspension of 1, 2e, and boric acid in CH₃CN was heated overnight, as reported previously. The reaction mixture was concentrated, and a ¹H nuclear magnetic resonance (NMR) spectrum of the crude product formed from 2e indicated quantitative formation of pseudo[2]rotaxane 3e (Fig. 2a). The reaction of 1, 2e, and trimethyl borate in CDCl₃ was then monitored by ¹H NMR spectroscopy (Fig. S10a†). After the addition of 2e to a solution of 1 and trimethyl borate in CDCl₃, signals attributed to pseudo[2]rotaxane 3e appeared immediately in the ¹H NMR spectrum, along with other low-intensity signals (e.g., at 6.3 ppm). After 17.3 h, the low-intensity signals converged with those of 3e, suggesting formation of the most thermodynamically stable pseudo[2]rotaxane under the NMR spectroscopy conditions (Fig. 2b).

Similarly, the rotaxane-forming reaction of 2f was performed in CH₃CN and CHCl₃ (Fig. 2c and d, respectively). In CH₃CN, some additional species were generated (Fig. 2c), whereas the reaction in CHCl₃ afforded predominantly rotaxane 3f (Fig. 2d).

When the reaction of 2h, a slightly bulky amine (Fig. S10b†), was performed in CH₃CN at 50 °C, both prominent rotaxane 3h signals and other distinct signals appeared in the ¹H NMR spectrum, but during the reaction in CDCl₃, only rotaxane 3h signals were observed in the ¹H NMR spectrum after 6 d. High temperatures and the CH₃CN solvent inhibited the preferential formation of rotaxane from bulky amines. However, the [2]rotaxane structure was stabilized in CHCl₃ at low temperatures, which was attributed to the formation of strong hydrogen bonds. As ion pairing can enhance the interactions between the components in face-to-face complexes and in rotaxanes, hydrogen bonding is expected to play a crucial role in stabilizing the rotaxane structure.

All reactions were performed independently in CHCl₃, and (pseudo)rotaxanes 3e–3h were all successively isolated; the isolated yields are listed in Table 1. The structures of all the new (pseudo)[2]rotaxanes were determined via two-dimensional (2D) NMR spectroscopy (Fig. S1 and S2 for 3e, Fig. S3 and S4 for 3f, Fig. S5 and S6 for 3g, and Fig. S7 and S8 for 3h†) and mass spectrometry (Fig. S9†).

---

The above data did not indicate whether 3a, 3d, 3e, and 3h are rotaxanes or pseudorotaxanes. However, after monitoring the decomplexation of these compounds, we concluded that 3a, 3d, and 3e are pseudo[2]rotaxanes and that 3h is a [2]rotaxane (described below).

To determine the stability of the prepared (pseudo)[2]rotaxanes, we monitored the borate-forming reaction of bis-catechol 1 (10 mM) and benzylamine 2a (20 mM) in the presence of excess trimethyl borate (100 mM) in CDCl₃ (Fig. S11†). The ¹H NMR spectra obtained during the reaction revealed that pseudo[2]rotaxane 3a formed immediately, along with small amounts of other species, and that only 3a was present in the reaction mixture after 85 min.

Next, we conducted the reaction with high concentrations of reactants (Fig. S12†). Because 1 had low-solubility in CDCl₃, CD₃CN was used as the solvent. The reaction of 1 (50 mM), 2a (60 mM), and trimethyl borate (60 mM) afforded 3a as a major product after 16 min via macrocyclization of one molecule each of 1 and trimethyl borate, along with other unidentified product(s). After 3 d, only signals of 3a and excess amine 2a were observed in the NMR spectrum, without any signals corresponding to oligomers and other macrocycles from two or more molecules of 1 and trimethyl borate. The high yield of 3a was attributed to the high stability of the pseudorotaxane and the thermodynamic conditions.

Distinguishing rotaxanes from pseudorotaxanes

Stoddart et al. reported that the reaction of a 24-membered crown ether with dibenzylammonium ions in an aprotic nonpolar solvent resulted in immediate formation of the corresponding pseudo[2]rotaxane,¹² and Huang et al. reported that phenyl groups hindered the reaction of 21-membered crown ethers.¹³ The relationship between ring size and the bulkiness of the terminal functional group of the axle component (molecule) affected the threading rate.¹⁴

In a previous study, we could not determine whether 3a was a rotaxane or a pseudo[2]rotaxane, consisting of a 23-membered macrocycle and an axle component with terminal phenyl groups, because the apparent decomplexation of the (pseudo)rotaxane involved dethreading and ring-opening decomplexation processes (Fig. 1).

In the present study, we utilized NMR spectroscopy to monitor the decomplexation of 3a in the presence of [2]rotaxane 3b with sufficiently bulky 3,5-dimethylphenyl functional groups on both ends of the axle component within a 24-membered crown ether (Fig. 3 and S13a†).¹⁵ Because the rate of the ring-opening process is expected to be sensitive to the amount of water, acidic conditions, and temperature,¹⁶ we performed the decomplexation experiments using a mixture of the (pseudo)rotaxanes. After mixing 3a and 3b in deuterated dimethyl sulfoxide (DMSO-d₆), the intensity of the 3a signals (approximately 7.2 ppm) decreased, and the intensity of the protonated 2a (dibenzylammonium ion) signals (approximately 7.4 ppm) increased. After 55 min of reaction at 21 °C, 93% of 3a was converted to the axle and macrocyclic molecules, and after 184 min of reaction, the decomplexation of 3a was complete. In contrast, 93% and 80% of 3b remained after 55 min and 184 min, respectively (6.86 ppm and 6.76 ppm for 3b and 7.03 ppm and 7.01 ppm for protonated 2b). The difference in decomplexation rates resulted from dethreading of 3a. These decomplexation experiments indicated that phenyl groups are not sufficiently bulky to inhibit dethreading of the 23-membered borate ion–containing crown ether. Similarly, 3e (bearing 4-methoxyphenyl terminal groups) was dethreaded when 3b and 3e were mixed in DMSO-d₆ (Fig. S13c†). When 3b and 3d (bearing 4-nitrophenyl terminal groups) were mixed in DMSO-d₆, however, an immediate spectrum revealed that decomplexation of 3d was complete, with only signals attributed to the macrocycle and axle ammonium species appearing in the spectrum, as confirmed by the spectrum of a solution of 3d in DMSO-d₆ (Fig. S13b†). In contrast, when 3h (bearing cyclohexyl terminal groups) and 3b were mixed in DMSO-d₆, the decomplexation rate of 3h was slower than that of 3b (Fig. S13d†). Thus, the cyclohexyl group hindered the reaction of the 23-membered borate crown ether at ambient temperature.

To determine the effect of a functionality in phenyl groups on the dethreading rate of pseudorotaxanes, solutions of pseudorotaxanes in DMSO-d₆ were independently monitored via ¹H NMR spectroscopy. Decomplexation of pseudo[2]rotaxanes 3a and 3e proceeded similarly in DMSO-d₆ at 21 °C (Fig. S14†). Plots of the concentrations of 3a and 3e were consistent with a first-order process. The decomplexation rate constant k was determined from the slope of the linear plots of ln[(integration of pseudorotaxane)/(all integration)] versus time (t): the k of 3a was 5.3 × 10⁻⁴ (s⁻¹), and the k of 3e was 4.1 × 10⁻⁴ (s⁻¹) (Fig. 4). As described above, dethreading was the main mechanism of decomplexation, and decomplexation of 3d, which possessed electron-withdrawing nitro groups, immediately proceeded to completion in DMSO-d₆. The presence of more acidic hydrogen-bond donors led to strong interactions between the two components in the pseudorotaxane, but the stability was not as expected. The stabilization was expected to result from the stability of the ammonium ions, as described below. Thus, we concluded that 3a, 3d, and 3e are pseudorotaxanes and that 3h is a rotaxane.

Exchange of an axle component via dynamic covalent chemistry

To prove the dynamic nature of the covalent B–O bond, exchange of an axle component in the rotaxane was performed. As 4-tert-butyphenyl¹⁷ and 3,5-dimethylphenyl¹⁵ functional groups are sufficiently bulky for rotaxanes bearing a 24-membered crown ether as a macrocycle, we treated [2]rotaxane 3c bearing a 23-membered crown with amine 2b. A solution of 3c (5 mM) and 2b (5 mM) in CDCl₃ was prepared, and the reaction progress at 50 °C was monitored by ¹H NMR spectroscopy (Fig. 5 and S15a†). NMR signals attributed to [2]rotaxane 3b (2.16 ppm for Me) appeared and gradually increased in intensity, and the intensity of the signal attributed to [2]rotaxane 3c (1.24 ppm for ^tBu) decreased with time; simultaneously, the intensity of the signals attributed to free amines of 2c (1.31 ppm) and 2b (2.29 ppm) increased and decreased, respectively.

After the system reached equilibrium, the ratio of [2]rotaxanes 3c and 3b was approximately 1:1, and these rotaxanes exhibited similar thermodynamic stability. We independently performed a rotaxane-forming reaction of 1 and boric acid in the presence of an equimolar mixture of amines 2b and 2c. The ¹H NMR spectrum of the product was consistent with a 1:1 ratio of [2]rotaxanes 3b and 3c (Fig. S15b†). The formation of a B–O bond and exchange of the axle components were reversible and were accompanied by protonation of the free outer amine and deprotonation of the inner ammonium ion.

Thermodynamic and kinetic control of (pseudo)[2]rotaxane formation

In the Stoddart approach (template-directed synthesis via reversible imine formation), secondary ammonium ions were used as templates, and the nature of the hydrogen-bond donor controlled the kinetic and thermodynamic stabilities of the dynamic [2]rotaxanes.¹⁸ More acidic ammonium ions promoted rotaxane formation and increased the thermodynamic stability of the rotaxane.

We performed borate-forming reactions with two different amines with different basicities. A mixture of 1 (1.0 equiv.), trimethyl borate (1.2 equiv.), basic bis(methoxybenzyl)amine 2e (1.0 equiv.), and less basic bis(nitrobenzyl)amine 2d (1.0 equiv.) was monitored (Fig. 6 and S16†). At the beginning (5 min), a 1.1:1 mixture of the pseudorotaxanes was observed (NMR signals at 6.09 ppm for 3e and 6.02 ppm for 3d), and some other signals not assigned to the pseudorotaxanes also appeared in the NMR spectrum. The intensity of the 3e and 3d signals increased and decreased, respectively, over time until a 20:1 mixture of 3e and 3d was present when the reaction system reached equilibrium (8 d).

The equilibrium constant K for the conversion of bis-catechol 1, amine 2, and trimethyl borate into (pseudo)rotaxane 3 was determined by using eqn (1). The relative stabilities of the two (pseudo)rotaxanes was determined by the ratio [eqn (2)] of the K values for each (pseudo)rotaxane-forming reaction, which depended on the concentrations of free amines and (pseudo)rotaxanes.

	(1)

	(2)

However, the final ratio (20:1) of pseudorotaxanes seemed too high to allow the exact concentrations of the four individual species to be determined; in particular, the concentration of the reactive free amine greatly affected this ratio. To validate the thermodynamic selectivity of the amines, the same reaction was performed in the presence of 2e (1.5 equiv.) and 2d (1.5 equiv.). The final ratio of 3e and 3d was 70:1 (Fig. S18†). We calculated ΔΔG° values of −14.6 kJ mol⁻¹ and −13.0 kJ mol⁻¹ by integrating the ¹H NMR spectra after both experiments reached equilibrium (Table 2). In this system, the less acidic ammonium ion axle was predominantly encircled by a borate-containing crown ether, but the reason was unclear.

---

In Stoddart et al.'s work, ammonium axles were used initially, so the hydrogen bonding ability determined the kinetic and thermodynamic formation of rotaxanes.¹⁸ In the current study, amines were used as precursors of the axle components, and these amines were protonated to form the actual axle molecules. Therefore, more basic amine 3e was predominantly protonated to form an ammonium ion axle. The stability of the ammonium ions mainly determined the thermodynamic stability of the pseudorotaxanes, as the predicted pK_a values of the protonated ammonium ions of 2e and 2d were 9.55 and 6.98, respectively.¹⁹ In addition, the initial ratio of pseudorotaxanes 3e and 3d was 1.1:1. The template effect of the acidic ammonium ion of 2d was expected to be much stronger than that of 2e, considering the basicity of both amines.

As described above, 3d and 3e are pseudorotaxanes, and the exchange of an axle component between pseudorotaxanes and free amines was expected to proceed quickly. Therefore, the real kinetic selectivity induced by more acidic ammonium ions was expected to exceed the experimental results (1.1:1). Competitive experiments involving basic amine 2b (1.0 equiv.) and less basic amine 2g (1.0 equiv.), which possessed bulky functional groups, were conducted in CDCl₃ (Fig. 7 and S17†). After 6 min, a 1:3 mixture of [2]rotaxanes 3b and 3g was produced, indicating complete reversal of the kinetic selectivity of [2]rotaxane formation. After this system reached equilibrium (5.7 d), a 30:1 mixture of 3b and 3g was observed by ¹H NMR (by integrating the signals at 6.14 ppm for 3b and 6.09 ppm for 3g). Thus, [2]rotaxane 3b derived from the more basic amine 2b formed predominantly under thermodynamic conditions (the predicted pK_a values of the protonated ammonium ions of 2b and 2g were 9.17 and 7.07, respectively¹⁹). We also determined the final selectivity (ratio of 3b to 3g > 80:1) by using 2b (1.5 equiv.) and 2g (1.5 equiv.) (Fig. S19†). The ΔΔG° values (−16.6 and <−13.3) were calculated by using eqn (2) (Table 2). The thermodynamic stability might be the preferential protonation of the more basic amine, as described above.

Conclusion

We synthesized (pseudo)[2]rotaxanes from bis-catechol 1 and secondary amines 2 in the presence of boric acid or trimethyl borate by conducting borate-forming reactions via a clipping approach. In CH₃CN at high temperatures, the yield of (pseudo)rotaxane from bulky amines was poor, but in a less polar CHCl₃ solvent, the (pseudo)rotaxanes were obtained in good yields. Phenyl groups were not sufficiently bulky to prevent dethreading of the 23-membered borate crown ether, but cyclohexyl groups inhibited this process. As the axle component of [2]rotaxane 3b was exchanged in the presence of amine 2c under heating, this borate-forming reaction was considered a dynamic clipping approach. Finally, we monitored competitive rotaxane-forming reactions involving two different amines. The acidic ammonium ion derived from the less basic amine was predominantly encircled by the borate crown ether under kinetic conditions, even though a less acidic ammonium ion was mainly produced. After the system reached equilibrium, a thermodynamically stable rotaxane derived from the basic amine formed as the major product because of the stability of the ammonium ion. The present results suggest selective and time-dependent formation of translational isomers of rotaxanes from origo- and polyamines.

Author contributions

Y. T. conceptualized the research. T. T. and S. M. performed the experiments. Y. T. wrote original draft. T. T. and Y. T. prepared the ESI.† All authors reviewed and edited the manuscript and ESI.†

Conflicts of interest

There are no conflicts to declare.

Data availability

The data supporting this article have been included as part of the ESI.†

Acknowledgements

This study was supported by JSPS KAKENHI [JP24K08392].

References

1. Selected reviews on dynamic covalent bond, see: (a) S. J. Rowan, S. J. Cantrill, G. R. L. Cousins, J. K. M. Sanders and J. F. Stoddart, Dynamic covalent chemistry, Angew. Chem., Int. Ed., 2002, 41, 898–952 CrossRef ; (b) T. Takata, Polyrotaxane and polyrotaxane network: Supramolecular architectures based on the concept of dynamic covalent bond chemistry, Polym. J., 2006, 38, 1–20 CrossRef CAS ; (c) Y. Jin, C. Yu, R. J. Denman and W. Zhang, Recent advances in dynamic covalent chemistry, Chem. Soc. Rev., 2013, 42, 6634–6654 RSC ; (d) S. P. Black, J. K. M. Sanders and A. R. Stefankiewicz, Disulfide exchange: exposing supramolecular reactivity through dynamic covalent chemistry, Chem. Soc. Rev., 2014, 43, 1861–1872 RSC ; (e) J. Yu, M. Gaedke and F. Schaufelberger, Dynamic covalent chemistry for synthesis and co-conformational control of mechanically interlocked molecules, Eur. J. Org. Chem., 2023, 26, e202201130 CrossRef CAS .
2. (a) P. T. Glink, A. I. Oliva, J. F. Stoddart, A. J. P. White and D. J. Williams, Template-directed synthesis of a [2]rotaxane by the clipping under thermodynamic control of a crown ether like macrocycle around a dialkylammonium ion, Angew. Chem., Int. Ed., 2001, 40, 1870–1875 CrossRef CAS ; (b) X. Han, G. Liu, S. H. Liu and J. Yin, Synthesis of rotaxanes and catenanes using an imine clipping reaction, Org. Biomol. Chem., 2016, 14, 10331–10351 RSC ; (c) C. D. Meyer, C. S. Joiner and J. F. Stoddart, Template-directed synthesis employing reversible imine bond formation, Chem. Soc. Rev., 2007, 36, 1705–1723 RSC .
3. (a) Y. Furusho, T. Oku, T. Hasegawa, A. Tsuboi, N. Kihara and T. Takata, Dynamic covalent approach to [2]- and [3]rotaxanes by utilizing a reversible thiol–disulfide interchange reaction, Chem. – Eur. J., 2003, 9, 2895–2903 CrossRef CAS PubMed ; (b) N. Ponnuswamy, F. B. L. Cougnon, G. D. Pantos and J. K. M. Sanders, Homochiral and meso figure eight knots and a Solomon link, J. Am. Chem. Soc., 2014, 136, 8243–8251 CrossRef CAS PubMed .
4. (a) B. Mohr, M. Weck, J.-P. Sauvage and R. H. Grubbs, High-yield synthesis of [2]catenanes by intramolecular ring-closing metathesis, Angew. Chem., Int. Ed. Engl., 1997, 36, 1308–1310 CrossRef CAS ; (b) D. G. Hamilton, N. Feeder, S. J. Teat and J. K. M. Sanders, Reversible synthesis of π-associated [2]catenanes by ring-closing metathesis: towards dynamic combinatorial libraries of catenanes, New J. Chem., 1998, 1019–1021 RSC ; (c) T. J. Kidd, D. A. Leigh and A. J. Wilson, Organic “magic rings”: The hydrogen bond-directed assembly of catenanes under thermodynamic control, J. Am. Chem. Soc., 1999, 121, 1599–1600 CrossRef CAS ; (d) A. F. M. Kilbinger, S. J. Cantrill, A. W. Waltman, M. W. Day and R. H. Grubbs, Magic ring rotaxanes by olefin metathesis, Angew. Chem., Int. Ed., 2003, 42, 3281–3285 CrossRef CAS PubMed ; (e) S. Dasgupta and J. Wu, Formation of [2]rotaxanes by encircling [20], [21] and [22]crown ethers onto the dibenzylammonium dumbbell, Chem. Sci., 2012, 3, 425–432 RSC ; (f) D. Quaglio, G. Zappia, E. De Paolis, S. Balducci, B. Botta and F. Ghirga, Olefin metathesis reaction as a locking tool for macrocycle and mechanomolecule construction, Org. Chem. Front., 2018, 5, 3022–3055 RSC .
5. (a) Y. Furusho, G. A. Rajkumar, T. Oku and T. Takata, Synthesis of [2]rotaxanes by tritylative endcapping of in situ formed pseudorotaxanes having thiol or hydroxyl functionality on the axle termini, Tetrahedron, 2002, 58, 6609–6613 CrossRef CAS ; (b) Y. Tokunaga, T. Ito, H. Sugawara and R. Nakata, Dynamic covalent chemistry of a boronylammonium ion and a crown ether: Formation of a C₃-symmetric [4]rotaxane, Tetrahedron Lett., 2008, 49, 3449–3452 CrossRef CAS ; (c) T. Fujino, H. Naitoh, S. Miyagawa, M. Kimura, T. Kawasaki, K. Yoshida, H. Inoue, H. Takagawa and Y. Tokunaga, Formation of [2]- and [3]rotaxanes through bridging under kinetic and thermodynamic control, Org. Lett., 2018, 20, 369–372 CrossRef CAS PubMed .
6. J. Wu, K. C.-F. Leung and J. F. Stoddart, Efficient production of [n]rotaxanes by using template-directed clipping reactions, Proc. Natl. Acad. Sci. U. S. A., 2007, 104, 17266–17271 CrossRef CAS PubMed .
7. (a) J. Yin, S. Dasgupta and J. Wu, Synthesis of [n]rotaxanes by template-directed clipping: The role of the dialkylammonium recognition sites, Org. Lett., 2010, 12, 1712–1715 CrossRef CAS PubMed ; (b) O. A. Bozdemir, G. Barin, M. E. Belowich, A. N. Basuray, F. Beuerle and J. F. Stoddart, Dynamic covalent templated-synthesis of [c2]daisy chains, Chem. Commun., 2012, 48, 10401–10403 RSC ; (c) X. Zhou, Z. Hu and X. Ji, Synthesis of adhesive polyrotaxanes through sequential self-assembly via supramolecular interactions and dynamic covalent interactions, Chem. – Eur. J., 2024, 30, e202402156 CrossRef CAS PubMed .
8. (a) L. Wang, M. O. Vysotsky, A. Bogdan, M. Bolte and V. Böhmer, Multiple catenanes derived from calix[4]arenes, Science, 2004, 304, 1312–1314 CrossRef CAS PubMed ; (b) R. T. S. Lam, A. Belenguer, S. L. Roberts, C. Naumann, T. Jarrosson, S. Otto and J. K. M. Sanders, Amplification of acetylcholine-binding catenanes from dynamic combinatorial libraries, Science, 2005, 308, 667–669 CrossRef CAS PubMed ; (c) S. Dasgupta and J. Wu, Template-directed synthesis of kinetically and thermodynamically stable molecular necklace using ring closing metathesis, Org. Biomol. Chem., 2011, 9, 3504–3515 RSC ; (d) Z. Li, F. Hu, G. Liu, W. Xue, X. Chen, S. H. Liu and J. Yin, Photo-responsive [2]catenanes: synthesis and properties, Org. Biomol. Chem., 2014, 12, 7702–7711 RSC ; (e) K. Caprice, D. Pál, C. Besnard, B. Galmés, A. Frontera and F. B. L. Cougnon, Diastereoselective amplification of a mechanically chiral [2]catenane, J. Am. Chem. Soc., 2021, 143, 11957–11962 CrossRef CAS PubMed .
9. Selected reviews on knots and links, see: (a) R. S. Forgan, J.-P. Sauvage and J. F. Stoddart, Chemical topology: Complex molecular knots, links, and entanglements, Chem. Rev., 2011, 111, 5434–5464 CrossRef CAS PubMed ; (b) J.-F. Ayme, J. E. Beves, C. J. Campbell and D. A. Leigh, Template synthesis of molecular knots, Chem. Soc. Rev., 2013, 42, 1700–1712 RSC ; (c) S. D. P. Fielden, D. A. Leigh and S. L. Woltering, Molecular knots, Angew. Chem., Int. Ed., 2017, 56, 11166–11194 CrossRef CAS PubMed ; (d) W.-X. Gao, H.-J. Feng, B.-B. Guo, Y. Lu and G.-X. Jin, Coordination-directed construction of molecular links, Chem. Rev., 2020, 120, 6288–6325 CrossRef CAS PubMed .
10. (a) C. Dietrich-Buchecker and J.-P. Sauvage, Lithium templated synthesis of catenanes: efficient synthesis of doubly interlocked [2]-catenanes, Chem. Commun., 1999, 615–616 RSC ; (b) C. D. Meyer, R. S. Forgan, K. S. Chichak, A. J. Peters, N. Tangchaivang, G. W. V. Cave, S. I. Khan, S. J. Cantrill and J. F. Stoddart, The dynamic chemistry of molecular borromean rings and Solomon knots, Chem. – Eur. J., 2010, 16, 12570–12581 CrossRef CAS PubMed ; (c) D. A. Leigh, R. G. Pritchard and A. J. Stephens, A Star of David catenane, Nat. Chem., 2014, 6, 978–982 CrossRef CAS PubMed ; (d) G. Gil-Ramírez, D. A. Leigh and A. J. Stephens, Catenanes: Fifty years of molecular links, Angew. Chem., Int. Ed., 2015, 54, 6110–6150 CrossRef PubMed .
11. T. Takizawa, K. Ohtani, M. Naito, S. Miyagawa and Y. Tokunaga, Rotaxane formation from borate ion-containing crown ether and ammonium ion: Enhancement of their association through ion-pairing, Org. Lett., 2024, 26, 8211–8215 Search PubMed .
12. P. R. Ashton, E. J. T. Chrystal, P. Glink, T. S. Menzer, C. Schiavo, N. Spencer, J. F. Stoddart, P. A. Tasker, A. J. P. White and D. J. Williams, Pseudorotaxanes formed between secondary dialkylammonium salts and crown ethers, Chem. – Eur. J., 1996, 2, 709–728 CrossRef CAS .
13. C. Zhang, S. Li, J. Zhang, K. Zhu, N. Li and F. Huang, Benzo-21-crown-7/secondary dialkylammonium salt [2]pseudorotaxane- and [2]rotaxane-type threaded structures, Org. Lett., 2007, 9, 5553–5556 CrossRef CAS PubMed .
14. (a) P. R. Ashton, I. Baxter, M. C. T. Fyfe, F. M. Raymo, N. Spencer, J. F. Stoddart, A. J. P. White and D. J. Williams, Rotaxane or pseudorotaxane? That is the question!, J. Am. Chem. Soc., 1998, 120, 2297–2307 Search PubMed ; (b) Y. Tokunaga, M. Yoshioka, T. Nakamura, T. Goda, R. Nakata, S. Kakuchi and Y. Shimomura, Do dibenzo[22–30]crown ethers bind secondary ammonium ions to form pseudorotaxanes?, Bull. Chem. Soc. Jp., 2007, 80, 1377–1382 CrossRef CAS .
15. H. Kawasaki, N. Kihara and T. Takata, High yielding and practical synthesis of rotaxanes by acylative end-capping catalyzed by tributylphosphine, Chem. Lett., 1999, 1015–1016 CrossRef CAS .
16. S. Green, A. Nelson, S. Warriner and B. Whittaker, Synthesis and investigation of the configurational stability of some dimethylammonium borate salts, J. Chem. Soc., Perkin Trans. 1, 2000, 4403–4408 RSC .
17. S. J. Rowan and J. F. Stoddart, Precision molecular grafting: Exchanging surrogate stoppers in [2]rotaxanes, J. Am. Chem. Soc., 2000, 122, 164–165 CrossRef CAS .
18. M. Horn, J. Ihringer, P. T. Glink and J. F. Stoddart, Kinetic versus thermodynamic control during the formation of [2]rotaxanes by a dynamic template-directed clipping process, Chem. – Eur. J., 2003, 9, 4046–4054 CrossRef CAS PubMed .
19. Predicted pK_a values were obtained from SciFinder.

---

Footnote

† Electronic supplementary information (ESI) available: Synthetic procedures and analytical data. See DOI: https://doi.org/10.1039/d5qo00880h

This journal is © the Partner Organisations 2025