The Rise and Promise of Molecular Nanotopology

Open AccessCCS ChemistryMINI REVIEW1 Jul 2021The Rise and Promise of Molecular Nanotopology Qing-Hui Guo†, Yang Jiao†, Yuanning Feng† and J. Fraser Stoddart Qing-Hui Guo† *Corresponding authors: E-mail Address: email protected E-mail Address: email protected Department of Chemistry, Northwestern University, Evanston, IL 60208 Stoddart Institute of Molecular Science, Department of Chemistry, Zhejiang University, Hangzhou 310027 ZJU-Hangzhou Global Scientific and Technological Innovation Center, Hangzhou 311215 , Yang Jiao† Department of Chemistry, Northwestern University, Evanston, IL 60208 , Yuanning Feng† Department of Chemistry, Northwestern University, Evanston, IL 60208 and J. Fraser Stoddart *Corresponding authors: E-mail Address: email protected E-mail Address: email protected Department of Chemistry, Northwestern University, Evanston, IL 60208 Stoddart Institute of Molecular Science, Department of Chemistry, Zhejiang University, Hangzhou 310027 ZJU-Hangzhou Global Scientific and Technological Innovation Center, Hangzhou 311215 School of Chemistry, University of New South Wales, Sydney, NSW 2052 https://doi.org/10.31635/ccschem.021.202100975 SectionsAboutAbstractPDF ToolsAdd to favoritesTrack Citations ShareFacebookTwitterLinked InEmail Molecular nanotopology—a term we coined recently—is a rapidly developing field of research that is emerging out of the confluence of chemical topology with the mechanical bond. When perusing the increased research activities in this field, it is clear that a new discipline is ready to receive recognition in its own right. In this Mini-Review, we address the historical development of chemical topology and describe how the rational design and practical synthesis of molecular links and knots with mechanical bonds, together with interwoven extended frameworks, have led to the rapid establishment of molecular nanotopology as a discipline. Representative examples are highlighted to offer the reader an extensive overview of ongoing research. We spotlight the major challenges facing chemists and materials scientists and provide some indications as to how molecular nanotopology is going to develop in the years ahead. Download figure Download PowerPoint Introduction to Molecular Nanotopology A skill set in our everyday lives is to endow strands of materials with either artistic or practical properties by incorporating topological features based on the age-old techniques of tying and weaving. Be they functional or decorative, examples can be found ranging from shoelaces to fishing nets not to mention in traditional Chinese knots, which are decorative, handcrafted arts tied from pieces of red cord. Molecular architectures with topologies also exist in the microscopic world. Knotted strands that have been identified in DNA1–5 and proteins6–14 have been shown15–17 to form randomly in synthetic polymer chains. Molecules with topologies expressed on the nanoscale are beginning to attract greater attention in the chemical community. In an attempt to control molecular topology and understand its special features at the nanoscopic level, chemists have devoted substantial efforts to the rational design and synthesis of molecules with different kinds of topology during the past three decades. Thanks to the advent of the mechanical bond in chemistry,18 the emergence of mechanically interlocked molecules (MIMs),18–20 and a growing awareness of chemical topology,21 chemists have made significant progress in the syntheses of molecules exhibiting topology. The mechanical bond18,22,23 is a relatively new class of chemical bond that exists in MIMs in which components are bound together by the simple act of mechanical interlocking. The chemistry of the mechanical bond is already well established and has been applied with remarkable success to the design and synthesis of catenanes,24,25 rotaxanes,26 molecular shuttles,27 and switches,28 as well as MIMs-based artificial molecular machines29 (AMMs), pioneered by Sauvage30 and one of us,19 who were awarded the 2016 Nobel Prize in Chemistry conjointly with Feringa.31 The concept of chemical topology was introduced by Frisch and Wasserman32 in 1961 to explain the phenomenon of topological isomerism in which two molecules have the same molecular formulas, yet their structures cannot be interconverted by any kind of deformation. Chemical topology21 is of fundamental importance in distinguishing and describing the molecular structures of MIMs. For example, the intrinsic distinction between catenanes and rotaxanes is very clear when taking chemical topology into consideration. A catenane is topologically nontrivial since its molecular components cannot be separated by any continuous deformation without breaking at least one participating covalent bond. A rotaxane, on the one hand, is topologically trivial for the simple reason that its molecular components can be separated—in principle at least—by the slippage33–36 of the ring off its dumbbell by its passing over one of the stoppers. In a catenane, on the other hand, the mechanical bond is also a topological bond, which is not the case in a rotaxane. For this reason, we have advocated the use of the term mechanical bond,18 first suggested by Frisch et al. in 1953.37 In mathematical topology (Figure 1), a knot is defined as a closed loop embedded in three-dimensional (3D) Euclidean space, whereas a link is a collection of rings that are mechanically interlocked, one with another. Prime knots refer to those that cannot be represented as sums of other knots, in analogy with prime numbers, while combinations of prime knots generate (Figure 1a) composite knots. The definitions (Figure 1b) of prime and composite links can be expressed in similar ways. Figure 1 | Graphical representations of several links and knots in mathematical topology. In the Alexander–Briggs notation, which is featured in red, a link or a knot is denoted in the form x z y , where x is equal to the minimum number of crossings in the projection of a topology, y is the number of components (in a knot y = 1 and is usually omitted), and z represents the order of the given topology among its peers with the same x and y descriptors. (a) Topological isomerism in knots with 0, 3, 4, 5, 7, and 6 nodes or crossings, respectively. (b) Topological isomerism in links consisting of two or three mechanically interlocked rings with 2, 4, 6, 6, 6, and 9 nodes/crossings, respectively. The graphics were prepared using KnotPlot. Download figure Download PowerPoint The Alexander–Briggs notation38 has been used to classify different topologies. In this notation, a link or a knot is denoted in the form x z y —featured in red in Figure 1 and subsequent Figures for consistency—where x is equal to the minimum number of nodes or crossings in the projection of the topology, y is the number of components (in a knot y = 1 and is usually omitted), and z represents the order of the particular topology among its peers with the same x and y descriptors. Aesthetically appealing topologies have encouraged chemists to express their counterparts in molecules. The foremost challenge in the synthesis of molecules with topologies lies in how to precisely control the entanglement of the closed loops and the generation of crossover points. Because of the innovative research on molecular knots and links, pioneered by Sauvage et al.39–41 and Leigh et al.,42,43 a new and independent research field—one we have ventured to call molecular nanotopology44—is emerging (Figure 2) out of the potpourri of chemical topology, mechanical bonds, and MIMs such as links (catenanes) and knots. In the case of molecular nanotopology involving links and knots, not to mention interwoven frameworks, a mechanical bond is a fundamental requirement for the formation of these topologically nontrivial molecules, keeping in mind that the existence of topology in links and knots does not lay claim to the exclusive use of mechanical bonding, which is also present, for examples, in rotaxanes and suitanes. Figure 2 | A Venn diagram showing (1) a set of MIMs, including rotaxanes, daisy chains, suitanes, catenanes/links, and knots; (2) a set of molecules that exhibit molecular nanotopology, including rings, belts, Möbius strips, catenanes/links, knots, nets, and interwoven frameworks—interwoven supramolecular nets, woven fabrics, and COFs. As far as molecular nanotopology is concerned, a mechanical bond is a fundamental requirement of topologically nontrivial molecules, whereas molecules that are topologically trivial can also exhibit mechanical bonding. Download figure Download PowerPoint Given the recent developments21,42,43 and breakthroughs45–50 in molecular nanotopology,44 we believe that a sea change is afoot and that molecular nanotopology is waiting in the wings to be embraced by the wider community of chemists and other scientists. It is timely to review this rapidly emerging field to encourage more young scientists to promote molecular nanotopology by pursuing research into molecular links and knots, some of which will prove to be transformative in the years to come. In this Mini-Review, following our brief discussion of the historical development of molecular nanotopology, we will present a blueprint and roadmap for its implementation by placing the emphasis on the rational design, practical syntheses, and potential applications of molecular links and knots, as well as interwoven molecular frameworks. Rather than trying to discuss numerous examples from the past, we will focus mainly on a few seminal contributions39,40,51–56 and some recent breakthroughs45–50,57–59 to offer readers a comprehensive overview of ongoing research in the field. Many more examples can be found in other informative perspectives and reviews.21,42,43,60–66 We will start by summarizing the general design strategies for the efficient construction of molecules with increasingly sophisticated topologies. We will also illustrate the existence of nanotopology in naturally occurring macromolecules, including its presence in DNA and proteins. Thereafter, proof-of-concept demonstrations and applications of molecular nanotopology will be discussed in some depth. To conclude, we will venture to offer our opinions in the context of recent and current developments and provide suggestions as to what we think lies ahead in what we predict will become an increasingly rapid and evolving field of research in the coming years. The Design and Synthesis of Molecular Nanotopology In this section, we will discuss one-by-one the design principles employed in the synthesis of molecular links and knots regarding four strategies—linear and circular helicates, interwoven grids, and all-in-one approaches—in addition to describing how to synthesize a variety of molecules with topology by applying these four strategies, with the emphasis placed on particular synthetic routes. Since the statistical synthesis reported in 1960 by Wasserman67 of a 2catenane—also described as a Hopf or 2 1 2 link—the template-directed synthesis68,69 of molecular links and knots—starting with the seminal work by Schill et al.,70,71 who employed covalent templation in the synthesis of catenanes and knots—exhibiting topological features has represented a significant advance18 in the realms of the mechanical bond. When designing routes to synthesize molecules exhibiting topology, three key factors need to be considered: (1) the generation of crossing points by loops, (2) the appropriate selection of connectivities between the loops, and (3) the formation of covalent bonds to complete the closure of the loops. Although loop closure is kinetically controlled, the all-in-one strategy employs dynamic covalent chemistry72–75 and is conducted under thermodynamic control. Templates, often aided and abetted by noncovalent bonding interactions, can be employed to entice building blocks to adopt entwined, preorganized co-conformations.76 Among all the known templates, transition metal ions77 are especially valuable and powerful in gathering and organizing organic strands into well-defined arrangements and crossings, largely on account of the specific coordination geometries dictated by these ions. Sauvage et al.24 described the first metal-templated synthesis in 1983 by employing the tetrahedral coordination between a Cu(I) ion and two bidentate phenanthroline ligands to produce, in their orthogonal disposition with respect to each other, the required crossing geometry for generating a catenane. Subsequently, metal templation, relying on coordination bonds and multifunctional ligands, has become77 a powerful and efficient approach to the rational synthesis of molecules with complex molecular topologies. By invoking traditional metal templation, Leigh et al.78 pioneered the concept of active-metal template synthesis in 2006. In the first reported example of this approach,78 Cu-mediated alkyne–azide cycloaddition (CuAAC) was employed in the synthesis of a rotaxane. The Cu(I) ion was bound to a pyridine ligand embedded in the backbone of a macrocycle. The alkyne and azide moieties of two half-axle components, on opposite faces of the macrocycle, coordinated to the Cu(I) ion and were activated to undergo a 1,3-dipolar cycloaddition to afford the rotaxane. The yields of rotaxane, based on the macrocycle, are very high—namely, up to 94% with stoichiometric Cu(I) and 82% with 20 mol % of Cu(I). In addition, the experimental procedure involved in running the reaction is extremely simple because neither an inert atmosphere nor a dried solvent is required. During the process of active-metal template synthesis, the metal ion plays a dual role, serving (1) as a template for entwining ligands and (2) as a catalyst for facilitating the formation of covalent bonds. The active-template process is driven kinetically simply because the bond formation occurring in the cavity of the macrocycle is faster than that taking place in bulk solution. The use of this approach has expanded79,80 rapidly and become a highly efficient strategy for the synthesis of a wide range of molecular links, that is, catenanes. While Leigh et al.78,81–84 were reporting the use of the active-metal template approach to the synthesis of 2rotaxanes, they were joined by Saito et al.85 who investigated the catalytic reactions of a macrocyclic copper complex to synthesize 2rotaxanes. Saito et al.86 reported the synthesis of 2catenanes by oxidative intramolecular diyne coupling mediated by macrocyclic copper(I) complexes in early 2009. Soon thereafter in 2009, Leigh et al.87 published the active-metal template synthesis of 2catenanes by employing appropriately functionalized pyridine ether or bipyridine ligands and either the CuAAC click reaction of azides with terminal alkynes or the Cu(I)-mediated Cadiot-Chodkiewicz heterocoupling of an alkyne halide with a terminal alkyne. Subsequently in 2011, Leigh et al.88 extended the active-template CuAAC approach to the synthesis of a molecular knot by combining the passive and active templating portions. In this ingenious approach, the first Cu(I) ion assists in the formation of a looped intermediate, and the second Cu(I) ion mediates the CuAAC reaction to close the knotted structure. Notably, the resulting trefoil knot was the smallest one to have been reported at that time. The relatively good yield of this tight knot was ascribed to the kinetically controlled bond formation promoted by the active CuAAC template.88 Linear helicates In the linear helicate strategy (Figures 3a–3e), templates, often accompanied by noncovalent bonding interactions, initiate the twisting of two linear strands to generate crossings and eventually form a double helicate. The final topology of a link or knot is determined by the number of crossings in the double helicate. If the number of crossings is odd, a molecular knot—the trefoil knot 31 or the pentafoil knot 51—will be formed, otherwise a molecular link, for example, the 2catenane 2 1 2 , the Solomon link 4 1 2 , or the Star of David 2catenane 6 1 2 , will be generated. Several examples (Figures 3f–3j) of molecular knots and links employing metal coordination,39,89 donor–acceptor interactions,51,90,91 hydrogen-bonding,92–95 π···π interactions, and hydrophobic effects55,56,96 to assemble linear strands have been accomplished by some of the pioneers in the field. In the following contents of this subsection, the rational synthesis of catenanes, trefoil knots, Solomon links, and figure-eight knots based on this linear helicate strategy will be introduced. Figure 3 | Graphical representations of the linear helicate strategy for constructing links and knots. (a–e) In the case of this strategy, templates initiate twisting of the linear strands to generate crossings and eventually form double helices. The final topology of a molecule is determined by its number of crossings. (f–j) Trefoil knots, including 1, 3, and 4; Solomon links, including 2; and a figure-eight knot, have all been synthesized and isolated as molecular compounds employing this strategy. (a–e) Adapted with permission from ref 42. Copyright 2013 Royal Society of Chemistry. (j) Adapted with permission from ref 56. Copyright 2014 American Chemical Society. Download figure Download PowerPoint In 1989, Sauvage et al.40,89 reported his seminal investigation on the metal-templated synthesis of molecular trefoil knots and a molecular Solomon link based on double-stranded helicates (Figures 3f and 3g). In the presence of Cu(I) ions, a ligand comprising two 1,10-phenanthroline units connected by a tetramethylene linker, formed a double-stranded helicate with two crossings. The subsequent ring-closure step, involving alkylation of this double helicate with hexaethylene glycol chains under highly dilute conditions, afforded the molecular trefoil knot 1 (Figure 3f). By increasing the number of phenanthroline units to three, and extending the linkers to hexamethylenes, a double helicate with three crossings was generated, affording a molecular Solomon link 2 after cyclization with heptaethylene glycol chains (Figure 3g). Donor–acceptor interactions have been employed by one of us51,97–100 as the driving force in the preparation of topological molecules (Figure 3h), including a trefoil knot 3 with six positive charges. A starting material with two electron-rich dioxynaphthalene units was threaded by another starting material containing three electron-poor bipyridinium units, forming a double helical precursor that underwent alkylation under high-pressure to afford the molecular trefoil knot 3. During his attempts to synthesize a macrocycle, Vögtle et al.92 discovered serendipitously that a hydrogen-bonded molecular trefoil knot 4 was formed during a one-pot reaction (Figure 3i). The structure of this knot was identified unambiguously by single-crystal X-ray diffraction analysis. Vögtle believed a network of hydrogen-bonding interactions was responsible for directing the assembly of 4 from no less than 12 precursors. The ease of the synthesis and the ready availability of hydrogen-bonded knots led to the preparation of other topologically complex compounds that were resolved and isolated as their enantiomers.101 In 2012, Sanders et al.55,56 developed a strategy (Figure 3j) for directing the synthesis of entwined architectures from naphthalenediimide-based building blocks, employing a disulfide-based dynamic combinatorial library (DCL) in an aqueous buffer. The topologies of the assembled end-products depended to a considerable extent on the number of aromatic components—1,4,5,8-naphthalenediimide (NDI) units—connected by flexible hydrophilic amino acid linkers. The driving force for the formation of these topological molecules in water was most likely a hydrophobic effect, which ensured that the NDI units were buried in hydrophobic cavities, while hydrophilic carboxyl groups on the amino acids were exposed to the aqueous environment. A molecular trefoil knot was formed55 from the trimeric building block in nearly quantitative yield, whereas topological isomers, including a 2catenane, a Solomon link, and a figure-eight knot were generated56 starting from a dimeric building block. Notably, in this DCL of molecular links and knots, the stereogenic centers present in the building blocks played an important role in the of the The amino acids employed be either in the case of the two to four stereogenic centers in one building block. one of these centers in the building block any of links or knots. during the synthesis of the molecular trefoil A of knots was when a of trimeric building blocks was whereas a topologically figure-eight knot was when starting with a of dimeric building the with the that dynamic covalent In this linear helicate strategy has in the construction of a of topological molecules, including catenanes, trefoil knots, Solomon links, and figure-eight knots. The synthesis, of molecules exhibiting as the pentafoil knots and Star of David not been most likely because of the that exist between the two of the double helicate loops. helicates The of the linear helicate strategy, as by its in (Figure to molecules with topological structures of have to use a circular helicate strategy. In this strategy (Figures templates or noncovalent bonding interactions can and molecular strands in a circular to the efficient synthesis of molecules with complex topologies. The number of strands and crossings the final topology of the trefoil Solomon pentafoil and Star of David In the circular helicate between the of the loops are relatively in with those involving linear helicates, the formation of covalent bonds that to loop the of the with recognition their synthesis strategy has been applied to the syntheses of molecules with topologies that are using the linear helicate strategy. Figure 4 | Graphical representations of the circular helicate strategy for constructing links and knots. In the case of this strategy, templates can and molecular strands in a circular to the efficient synthesis of molecular links and knots, including trefoil knots, Solomon links, pentafoil knots, and Star of David The number of strands and crossings the topology of the A diagram with three strands and the circular strategy used in the construction of an Synthesis of a molecular knot, employing the circular The reaction of with by with an circular A subsequent links the of the ligands in the to afford a molecular knot 6 that a as a template in its cavity as by single-crystal X-ray by the organic knot that a flexible Adapted with permission from ref 42. Copyright 2013 Royal Society of Chemistry. and Adapted with permission from ref Copyright American for the of Download figure Download PowerPoint In the et serendipitously discovered circular helicates that form double woven metal centers of or ions. these Leigh et to form molecular links and knots by the ligand strands and the appropriate By employing the circular helicate strategy, a trefoil a Solomon interlocked a pentafoil and a Star of 2catenane interlocked have all been synthesized from and circular helicates, respectively. The coordination defined by the bond and with each important in the assembly and of molecular topologies on the nanoscale The rational synthesis of a molecular knot, based on the circular helicate strategy, is discussed in in the following In Leigh et extended the circular helicate strategy from to (Figure to synthesize molecules with topologies. molecular knot (Figure was prepared by three strands with coordinated ions. of ligand with by with a circular In this the of the three strands at each crossing were controlled by ions, while the were determined by on the The subsequent connected the of ligands, affording the knot The in the preorganized at the reaction that the can between strands coordinated to the tight molecular knot with loops containing crossing was with relatively The knot topological which is expressed in the circular of the strategy considerable in the building of topologies from some and simple molecular The principle for strands is to be applied to the synthesis of a range of topological molecules exhibiting For example, a composite molecular three trefoil of the same with a of well as a 9 3 link have been from a circular helicate by the with the In this subsection, the rational design and synthesis of an interwoven supramolecular molecular woven molecular and covalent organic based on the interwoven strategy will be introduced following a discussion of supramolecular weaving. To synthesize more complex molecular an interwoven strategy has been applied to the of molecular In one of introduced the concept of supramolecular (Figure and an extended interwoven topology using a acid A as a supramolecular a term introduced by et in The interwoven supramolecular was (Figure by noncovalent synthesis, starting from the building blocks and ions. The of two formed by the double of two the of with each other to