Towards the molecular Borromean link with three unequal rings: double-threaded ruthenium(II) ring-in-ring complexes

Abstract

This study describes synthetic efforts towards the molecular Borromean link consisting of three unequal rings. The design and strategy involve step by step construction of a ruthenium(II) templated ring-in-ring complex doubly threaded with endocyclic ligands ready for the macrocyclization. The control over the topology is achieved by using specially designed directional building blocks based on 2,2′:6′,2′′-terpyridine. Preliminary macrocyclization attempts utilizing the copper-mediated Eglington reaction provide the mass spectrometric evidence consistent with the ruthenium(II) complex of the molecular Borromean link.

---

Introduction

The Borromean link comprises three topologically interlocked rings with the remarkable property that no two loops are mutually linked and the cleavage of any one loop results in separation of the other two.¹ Such links² have long fascinated chemists as a complex synthetic target.³ The knotted DNA superstructure⁴ with the Borromean topology, the Borromean networks,⁵ and the thermodynamic assembly of the molecular Borromean links consisting of equal rings⁶ have been reported. When all rings are either oriented or unequal the topological symmetry becomes a factor; thus, a directed synthesis of a molecular Borromean link controlling the nature of the three rings would be of special interest;⁷ however, despite several attempts, the synthesis involving three unequal rings remains a challenge.^2b,8,9 In this context, ring-in-ring complexes^8,10,11 may serve as strategic intermediates towards this goal.

A classic ring-in-ring strategy involves starting from the complexation of a macrocycle A, possessing endocyclic coordination sites, with the heteroleptic metal complex B to form an endo,endo-threaded macrocycle C (Fig. 1(a)).⁸ Subsequent coupling of C with linker D would afford a ring-in-ring complex E. If E contains endocyclic coordination sites, then threading with another complex F would give a double threaded ring-in-ring structure G. The closure of the final ring with linker H would form the Borromean link metal complex I. This strategy has led to the synthesis of Ru(II) ring-in-ring complexes having two 2,2′-bipyridine (bipy)^8a or 2,2′:6′,2′′-terpyridine (terpy)^8b pockets in the internal ring, which are topologically equivalent structures to E. All attempts to perform threading of the ring-in-ring complexes E with F have so far been unsuccessful, presumably because of an unfavorable strain inhibiting the necessary change of the bipy conformation upon complexation or the limited cavity available for the incoming motif F. This motivated a detailed structural study of the conformational flexibility of large macrocycles and ring-in-ring complexes, which supports the necessity for new strategies.^8b

Fig. 1 Towards the molecular Borromean link comprising unequal rings.

A modified approach implementing preformed Ru(II)-templated cap J combined with the threaded ring C could be used to afford directly a G analog (Fig. 1(b)). If cap J would already contain linkers suitable for the macrocyclization then it would take one step to convert G into the Borromean link I. Herein we report a successful synthesis of a double threaded ring-in-ring ruthenium(II) complex 1a (Fig. 2), which is the topological equivalent of G. This molecule contains linkers functionalized with terminal acetylenes, suitable for the macrocyclization to form the final third ring. The macrocyclization attempts provided mass spectrometric detection of the Borromean link Ru(II) complex.

Fig. 2 The double threaded macrocyclic ruthenium(II) ring-in-ring complex 1a.

Our molecular design involves the use of directional building blocks from a set of 2,2′:6′,2′′-terpyridine (terpy)¹² based ligands (Fig. 3), which mimic the linear-rod geometry of 5,5′-substituted 2,2′-bipyridine ligands – “linear bilateral extended terpy”¹³ – the missing unit for the construction of extended terpy based structures. For example, “V-terpy” is handy for the 60° turn motifs with respect to its coordination vector, but “extended terpy” is useful for orthogonal motifs, moving substituents further away from one another. The proper combination of such building blocks allows one to control how the molecular strands inter-weave in 1a (cf.Fig. 2).

Fig. 3 Terpy-based motifs as the topological control element.

Results and discussion

Synthesis of the necessary ligands starts from diiodoterpyridine 2 (Scheme 1),¹³ which couples to the corresponding acetylene 3 (a or b) via the Sonogashira reaction,¹⁴ to give either bromo- or methoxy-substituted bis-ethynyl terpyridines 4. A one-pot methoxymethyl (MOM) deprotection and cycloisomerization affords “linear bilateral extended terpy” ligands 5.¹³ The methoxy substituted ligand 5b was further converted to bis-hydroxy-substituted “extended terpy” 5c, which could be alkylated readily with 2-(2-chloroethoxy)ethan-1-ol to give ligand 5d.

Scheme 1 Conditions: (a) 5b (1 eq.), Py·HCl (50 eq.), MW, 190 °C, 2 × 2 min; (b) 5c (1 eq.), 2-(2-chloroethoxy)ethan-1-ol (2.9 eq.), Cs₂CO₃ (3 eq.), DMF, 100 °C, 18 h.

Independent functionalization of bis-methoxy substituted “V-terpy” ligand 6a¹⁵ with appropriate linkers (Scheme 2) is achieved in a good yield by cleaving the methoxy groups of 6a under microwave irradiation in the presence of pyridine hydrochloride to obtain 6b, which is then alkylated with 3-chloropropan-1-ol to afford bis-alcohol 6c.

Scheme 2 Conditions: (c) 6a (1 eq.), Py·HCl (19 eq.), MW, 190 °C, 10 min; (d) 6b (1 eq.), 3-chloropropan-1-ol (2.2 eq.), Cs₂CO₃ (3.0 eq.), DMF, 100 °C, 20 h; (e) 5 (1 eq.), RuCl₃·3H₂O (1 eq.), EtOH, reflux, 18 h; (f) 7a (1 eq.), 6c (1.1 eq.), N-ethylmorpholine (2.3 eq.), EtOH, N₂, reflux, 24 h, then aq. KPF₆; (g) 7a (1 eq.) or 7b (1 eq.), 5d (1 eq.), N-ethylmorpholine (2.3 eq.), EtOH, N₂, reflux, 20–24 h, then aq. KPF₆; (h) 10a (1 eq.), 11 (2.55 eq.), Cs₂CO₃ (3 eq.), DMF, N₂, 75 °C, 5 h; (i) 10b (1 eq.), Et₃N (70 eq.), MsCl (36 eq.), THF/MeCN (20 : 3), rt, 1.5 h; (j) 9a (1 eq.), Et₃N (40 eq.), MsCl (20 eq.), THF/MeCN (7 : 3), rt, 1 h; (k) 8a (1 eq.), 11 (2.2 eq.), Cs₂CO₃ (3 eq.), DMF, N₂, 80 °C, 9.5 h; (l) 8b (1 eq.), Et₃N (60 eq.), MsCl (30 eq.), THF/MeCN (6 : 1), rt, 2 h. Py = pyridine, MW = microwave irradiation, Ms = methanesulfonyl.

Pretemplated Ru(II) caps of structure-type J (Fig. 1) come from the complexation of “extended terpy” ligands 5a or 5c with RuCl₃·3H₂O to furnish the ligand–ruthenium(II)Cl₃ complexes 7a/b (Scheme 2). Further complexation of 7a/b with “extended terpy” ligands 5a and 5d, or “V-terpy” ligand 6c in EtOH and N-ethylmorpholine¹⁶ as a base provides the heteroleptic ligand complexes 8a, 9a, or 10a, accordingly. The ¹H and ¹³C NMR spectra along with mass spectrometry (ESI) support the formation of 1 : 1 mixed ligand Ru(II) complexes. Slow diethyl ether vapor diffusion into an acetonitrile solution of the complex 9a gave red crystals, for which the crystal structure was solved in space group P2₁/n, revealing the coordination of two different ligands around the ruthenium(II) center, as anticipated from the solution NMR and MS data (Fig. 4A).¹⁷

Fig. 4 The molecular structures of: (A) the heteroleptic ruthenium(II) complex 9a; (B) the endo,endo-double threaded ruthenium(II) complex 13. (Solvent molecules and PF₆⁻ omitted for clarity.)

Next, complexes 8a and 10a were functionalized with the linkers containing terminal acetylene groups suitable for the final macrocyclization. The heteroleptic complexes 8a and 10a have two pairs of chemically distinct –OH groups, thus no alcohol protection was necessary to perform bis-alkylation with linker 11 (Scheme 2). The phenolic hydroxy groups are more reactive, exclusively affording products 8b or 10b. Further, the treatment of corresponding complexes with methanesulfonyl chloride resulted in bis-mesylate complexes 8c, 9b, and 10c.

The reaction of 12,^8a with “extended terpy” Ru(II)Cl₃ complex 7a furnished endo,endo-double threaded ruthenium(II) complex 13 (Scheme 3) – a topological equivalent of C (Fig. 1). Slow solvent Et₂O/MeCN diffusion afforded red block-like crystals of 13, for which the crystal structure was solved in space group P1̄ (Fig. 4B). Both threaded ligands have the expected endocyclic conformation necessary for the synthesis of the Borromean link. It reveals that the distance between both flanking hydroxy groups on the opposing ligands is 19.242(7) Å – twice as much as for the similar previously reported bis-“W-terpy”/macrocycle Ru(II) complex (∼8.3 Å).^8b The distance between the –OH groups of the same ligands is 23.084(9) Å, providing a significant void for the incorporation of the third ring.

Scheme 3 Conditions: (m) 12 (1 eq.), 7a (3 eq.), N-ethylmorpholine (1.6 eq.), EtOH, N₂, reflux, 24 h, yield 31%; (n) 13 (1 eq.), 8c (2.7 eq.), or 9b (2.3 eq.), or 10c (2.2 eq.), K₂CO₃ (25 to 28 eq.), 4 Å molecular sieve powder, DMF, Ar, 70 to 80 °C, 2 to 4 days.

Further, attempts were made to combine the threaded macrocycle 13 with the pretemplated caps 8c, 9b, or 10c by intermolecular alkylation to form the double threaded ring-in-ring complexes 1. This turned out to be challenging because of competing decomposition and polymerization pathways. After investigating various reaction parameters, it was identified that the best result for the synthesis of ring-in-ring complexes 1 was reached when the reaction was performed in high dilution using DMF as a solvent, K₂CO₃ as a base, and activated 4 Å molecular sieve powder as an additive (Scheme 3).

This reaction is extremely sensitive to temperature variations. The temperature must be increased gradually from 70 to 80 °C over the course of the reaction. Higher temperatures (80–83 °C) caused polymerization and decomposition of the starting materials giving only traces of the products 1. Heating at lower temperature (70 °C) resulted in an extremely slow conversion of the starting materials and accumulation of partially alkylated/cyclized intermediates, but it helped to decrease the rate of decomposition.

Three reaction pathways could take place, depending on how the orthogonal caps approach the macrocyclic complex 13 (Fig. 5), and three different stereochemical outcomes are foreseen – D_2h symmetric exo,exo-, C_2v symmetric endo,exo-, or the desired D_2h symmetric endo,endo-structures.

Fig. 5 Three theoretical trajectories of orthogonal caps towards the threaded macrocycle leading to endo- and exo-isomerism.

When heteroleptic complexes 8c or 9b consisting of only “extended terpy” ligands were used (Scheme 3), complex mixtures of ring-in-ring stereoisomers 1b or 1c appear to be obtained, as suggested by the low symmetry observed in the ¹H and ¹³C NMR spectra (see ESI,† pages S58 to S64). The significant overlap of the NMR signals prevented unambiguous determination of an exact ratio and geometry of these isomeric species. The expected composition of the complexes 1b or 1c was supported by HRMS-ESI.

When the mixed “V-shaped/extended terpy” cap 10c was used for grafting onto the threaded macrocycle 13, the ¹H and ¹³C NMR spectra showed the formation of a product of high symmetry (D_2h). As expected, the ¹H NMR spectrum of the putative 1a appears as a superimposition of the corresponding building blocks 10c and 13 (Fig. 6). The phenolic –OH signals observed for 13 and the –OMs signals of 10c disappear in the product's spectra. In addition, the –CH₂ signal corresponding to the m-protons is shifted upfield in the product in comparison to 10c, supporting the formation of ether bonds (Fig. 6 and Scheme 3). HRESI-MS data support the expected composition, charge (8+) and isotope pattern of the desired complex 1a (see ESI† and Fig. 7). The geometry of “V-shaped terpy” and relatively short 3-carbon linkers should prevent the exo-approach (Fig. 5) favoring the formation of an endo,endo-complex 1a. Thus, the product obtained seems to have the required threading analogous to intermediate G (Fig. 1) on the way towards the Borromean link consisting of three unequal rings.

Fig. 6 Selected regions of the ¹H NMR spectra of the building blocks 10c, 13, and the threaded ring-in-ring complex 1a.

Fig. 7 Mass spectrometric support for the intramolecular macrocyclization of 1a.

When complex 1a was subjected to the copper-mediated Eglington reaction to facilitate macrocyclization by acetylene homocoupling,¹⁸ mostly insoluble material formed, which presumably arises from the intermolecular polymerization reaction (Fig. 7). However, mass spectrometry (HRESI-MS) supports the formation of a molecular ion with the loss of four mass units and a composition consistent with the desired intramolecular macrocyclization product 14, providing hope that the Borromean link Ru(II) complex forms, albeit in minute quantities.

These results provide a new departure point for future investigations towards the synthesis of the molecular Borromean link consisting of three unequal rings. We anticipate that the optimization of the structure 1, with emphasis on the linker lengths, should provide the desired product in synthetically practical yields.

Conclusions

In conclusion, the endo,endo-double threaded ruthenium(II) ring-in-ring complex 1a was successfully prepared in a stepwise synthesis by grafting preformed heteroleptic terpy cap 10c with double-threaded macrocycle ruthenium(II) complex 13. The shape of the directional terpy based ligands can be a defining control element for the exo-,endo-conformation of the metal-templated threaded macrocyclic architectures. The mass spectrometric evidence suggests that the herein described strategy towards the molecular Borromean link consisting of three unequal rings is conceptually feasible. In the future, optimization of the building blocks and linkers should provide the desired product in synthetically useful quantities.

Acknowledgements

We thank the Swiss National Science Foundation for financial support, P.D. Dr Laurent Bigler and Urs Stadler for measuring mass spectra, Andrew Raub for help with the synthesis of intermediates, and Dr Philip Pattison for collecting the diffraction data for compound 9a on the Swiss Norwegian Beamlines at the European Synchrotron Radiation Facility.

Notes and references

1. (a) C. Liang and K. Mislow, J. Math. Chem., 1994, 16, 27; (b) D. Huylebrouck, P. Cromwell, E. Beltrami and M. Rampichini, Math. Intell., 1998, 20, 53.
2. Towards the Borromean link: (a) H. L. Frisch and E. Wasserman, J. Am. Chem. Soc., 1961, 83, 3789; (b) T. J. Hubin, A. G. Kolchinski, A. L. Vance and D. H. Busch, Adv. Supramol. Chem., 1999, 6, 237; (c) J. S. Siegel, Science, 2004, 304, 1256; (d) S. J. Cantrill, K. S. Chichak, A. J. Peters and J. F. Stoddart, Acc. Chem. Res., 2005, 38, 1.
3. Recent examples of complex supramolecular architectures: (a) D. M. Engelhard, S. Freye, K. Grohe, M. John and G. H. Clever, Angew. Chem., Int. Ed., 2012, 51, 4747; (b) C. Schouwey, J. J. Holstein, R. Scopelliti, K. O. Zhurov, K. O. Nagornov, Y. O. Tsybin, O. S. Smart, G. Bricogne and K. Severin, Angew. Chem., Int. Ed., 2014, 53, 11261; (c) J. E. Beves, J. J. Danon, D. A. Leigh, J. F. Lemonnier and I. J. Vitorica-Yrezabal, Angew. Chem., Int. Ed., 2015, 54, 7555. Recent reviews: (d) J. E. Beves, B. A. Blight, C. J. Campbell, D. A. Leigh and R. T. McBurney, Angew. Chem., Int. Ed., 2011, 50, 9260; (e) G. Gil-Ramirez, D. A. Leigh and A. J. Stephens, Angew. Chem., Int. Ed., 2015, 54, 6110.
4. The DNA superstructure with the Borromean topology: C. Mao, W. Sun and N. C. Seeman, Nature, 1997, 386, 137.
5. The Borromean networks: (a) L. Carlucci, G. Ciani and D. M. Proserpio, CrystEngComm, 2003, 5, 269; (b) M. Pan and C.-Y. Su, CrystEngComm, 2014, 16, 7847; (c) F. L. Thorp-Greenwood, A. N. Kulak and M. J. Hardie, Nat. Chem., 2015, 7, 526.
6. The thermodynamic assembly of the molecular Borromean links: (a) K. S. Chichak, S. J. Cantrill, A. R. Pease, S. H. Chiu, G. W. Cave, J. L. Atwood and J. F. Stoddart, Science, 2004, 304, 1308; (b) A. J. Peters, K. S. Chichak, S. J. Cantrill and J. F. Stoddart, Chem. Commun., 2005, 3394; (c) C. D. Pentecost, A. J. Peters, K. S. Chichak, G. W. Cave, S. J. Cantrill and J. F. Stoddart, Angew. Chem., Int. Ed., 2006, 45, 4099; (d) S. L. Huang, Y. J. Lin, T. S. Hor and G. X. Jin, J. Am. Chem. Soc., 2013, 135, 8125; (e) S. L. Huang, Y. J. Lin, Z. H. Li and G. X. Jin, Angew. Chem., Int. Ed., 2014, 53, 11218.
7. Symmetry of the Borromean link: (a) S. J. Tauber, J. Res. Natl. Bur. Stand., Sect. A, 1963, 67, 591; (b) D. M. Walba, Tetrahedron, 1985, 41, 3161; (c) D. M. Walba, T. C. Homan, R. M. Richards and R. C. Haltiwanger, New J. Chem., 1993, 17, 661.
8. (a) J. C. Loren, M. Yoshizawa, R. F. Haldimann, A. Linden and J. S. Siegel, Angew. Chem., Int. Ed., 2003, 42, 5702; (b) J. K. Klosterman, J. Veliks, D. K. Frantz, Y. Yasui, M. Loepfe, E. Zysman-Coleman, A. Linden and J. S. Siegel, Conformations of Large Macrocycles and Ring-in-Ring Complexes, Org. Chem. Front. 2016, in press.
9. (a) R. S. Forgan, J. M. Spruell, J.-C. Olsen, C. L. Stern and J. F. Stoddart, J. Mex. Chem. Soc., 2009, 53, 134; (b) R. S. Forgan, C. Wang, D. C. Friedman, J. M. Spruell, C. L. Stern, A. A. Sarjeant, D. Cao and J. F. Stoddart, Chem. – Eur. J., 2012, 18, 202.
10. The ring-in-ring metal complexes: (a) M. Schmittel, A. Ganz and D. Fenske, Org. Lett., 2002, 4, 2289; (b) J.-H. Fu, Y.-H. Lee, Y.-J. He and Y.-T. Chan, Angew. Chem., Int. Ed., 2015, 54, 6231.
11. Selected examples of the ring-in-ring complexes exploiting donor–acceptor interactions: (a) A. I. Day, R. J. Blanch, A. P. Arnold, S. Lorenzo, G. R. Lewis and I. Dance, Angew. Chem., Int. Ed., 2002, 41, 275; (b) J. F. Stoddart, A. J. P. White and D. J. Williams, Angew. Chem., Int. Ed., 2002, 41, 270; (c) T. Iwanaga, R. Nakamoto, M. Yasutake, H. Takemura, K. Sako and T. Shinmyozu, Angew. Chem., Int. Ed., 2006, 45, 3643; (d) Y. Liu, Tetrahedron Lett., 2007, 48, 3871; (e) J. Sun, M. Frasconi, Z. Liu, J. C. Barnes, Y. Wang, D. Chen, C. L. Stern and J. F. Stoddart, Chem. Commun., 2015, 51, 1432.
12. 2,2′:6′,2′′-Terpyridine (terpy): (a) S. G. Morgan and F. H. Burstall, J. Chem. Soc., 1931, 20; (b) E. C. Constable and J. Lewis, Polyhedron, 1982, 1, 303; (c) H. Hofmeier and U. S. Schubert, Chem. Soc. Rev., 2004, 33, 373; (d) U. S. Schubert, H. Hofmeier and G. R. Newkome, Modern Terpyridine Chemistry, Wiley-VCH-Verl., Weinheim, 2006.
13. The “linear bilateral extended terpy” ligands: (a) J. Veliks, J.-C. Tseng, K. I. Arias, F. Weisshar, A. Linden and J. S. Siegel, Chem. Sci., 2014, 5, 4317; (b) J. Veliks, O. Blacque and J. S. Siegel, Inorg. Chem., 2014, 53, 12122.
14. The Sonogashira coupling: (a) K. Sonogashira, Y. Tohda and N. Hagihara, Tetrahedron Lett., 1975, 16, 4467; (b) R. Chinchilla and C. Najera, Chem. Soc. Rev., 2011, 40, 5084.
15. J. C. Loren and J. S. Siegel, Angew. Chem., Int. Ed., 2001, 40, 754.
16. The synthesis of heteroleptic ruthenium(II) complexes: M. Maestri, N. Armaroli, V. Balzani, E. C. Constable and A. M. W. C. Thompson, Inorg. Chem., 1995, 34, 2759.
17. The crystals of 9a appeared to be of good quality, but they diffracted poorly on the in-house X-ray diffractometer. Synchrotron radiation was used to obtain better diffraction data, allowing the elucidation of the structure (see ESI†).
18. Examples of acetylene homocoupling: (a) G. Eglinton and A. R. Galbraith, J. Chem. Soc., 1959, 889; (b) A. R. Karim, A. Linden, K. K. Baldridge and J. S. Siegel, Chem. Sci., 2010, 1, 102; (c) K. I. Arias, E. Zysman-Colman, J. C. Loren, A. Linden and J. S. Siegel, Chem. Commun., 2011, 47, 9588.

---

Footnote

† Electronic supplementary information (ESI) available: Full experimental synthesis procedures and characterization data, including ¹H NMR and ¹³C NMR spectra of all new compounds and CIF files. CCDC 1431458 and 1431459. For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/c6qo00025h

---