Scholarly article on topic 'Catenanes: Fifty Years of Molecular Links'

Catenanes: Fifty Years of Molecular Links Academic research paper on "Chemical sciences"

0
0
Share paper
OECD Field of science
Keywords
{""}

Academic research paper on topic "Catenanes: Fifty Years of Molecular Links"

Angewandte. D. A. Leigh et al.

Reviews

Catenanes

International Edition: DOI : 10.1002/anie.201411619 German Edition: DOI: 10.1002/ange.201411619

Catenanes: Fifty Years of Molecular Links

Guzman Gil-Ramirez, David A. Leigh,* and Alexander J. Stephens

Keywords:

catenanes • interlocked molecules links • supramolecular chemistry • template synthesis

Angewandte

International Edition Chemie

Half a century after Schill and Luttringhaus carried out the first directed synthesis of a [2]catenane, a plethora of strategies now exist for the construction of molecular Hopf links (singly interlocked rings), the simplest type of catenane. The precision and effectiveness with which suitable templates and/or noncovalent interactions can arrange building blocks has also enabled the synthesis of intricate and often beautiful higher order interlocked systems, including Solomon links, Borromean rings, and a Star of David catenane. This Review outlines the diverse strategies that exist for synthesizing catenanes in the 21st century and examines their emerging applications and the challenges that still exist for the synthesis of more complex topologies.

1. Introduction

Over the last fifty years, research into the synthesis of interlocked molecules has evolved from a concept viewed with some scepticism to a reality in which ways to harness the properties afforded by mechanical bonding are now being sought. Catenanes are at the forefront of efforts to make artificial molecular machines and to exploit the dynamics of interlocked structures in polymers, MOFs, and other materials. Their study has led to significant developments that have implications not only in supramolecular chemistry but across a range of scientific disciplines from biology to soft matter physics. Here we review the synthetic tactics that have been employed in the formation of catenanes, including the various template techniques and methods used in the contextual synthesis of singly interlocked [2]catenanes (Section 2), strategies for higher order interlocked structures (Sections 3 and 4), the properties of catenanes (Section 5), and their incorporation into materials and onto surfaces (Section 6). The vast literature featuring interlocked molecular rings means that only examples of the most significant advances of the last fifty years could be covered. All of the cap-and-stick structures shown in the Review are original representations produced from coordinates taken from the Cambridge Structural Database (CSD).

From the Contents

1. Introduction 6111

2. Synthesis of Hopf Link (Singly

Interlocked) [2]Catenanes 6116

3. Higher Order Linear and Radial

[n]Catenanes 6125

4. Higher Order Entwined

[n]Catenanes 6130

5. Catenanes as Switches, Rotary

Motors, and Sensors 6134

6. Catenane Linkages

Incorporated into Polymer

Chains, Materials, and

Attached to Surfaces 6140

7. Conclusions and Outlook 6143

1.1. Historical Background

The earliest known mention of the possibility of mechanically linked cyclic molecules, structures later termed "cate-nanes"[1] is attributed'21 to the 1915 Nobel Laureate in Chemistry, Richard Willstätter, in a seminar in Zürich some time in the period 1906-1912. Unfortunately, the nature of that discussion is lost in the mists of time, but it is remarkable that the notion of interlocked molecules was raised at all in an era that predated the assertion of Staudinger (Willstätter's successor at the then newly named "Eidgenössische Technische Hochschule" in Zürich) that polymers were covalently bonded molecular chains and even before the existence of macrocycles—the putative components of catenanes—was known.'3] Half a century on, following pioneering studies on the synthesis of medium and large cyclic molecules by the

research groups of Ruzicka,[4] Ziegler,[5] Prelog,[6] Stoll,[7] and others,[8] and the discovery of cyclic molecules in various synthetic polymer-forming reactions,[9] speculation about the existence of catenanes—and their designed synthesis—gained fresh impetus. In 1953 Frisch, Martin, and Mark postulated that the liquid/waxy appearance of high-molecular-weight polysiloxanes might be due to the presence of large interlocked rings acting as plasticizers.[10] The following year Luttringhaus and Cramer began working on synthetic routes to catenanes (e.g. 1), building on efforts by Cramer's group to make rotaxanes as early as 1950.[11] An unsuccessful early attempt to make a catenane, based on the cyclization of an inclusion complex (2) of a para-disubstituted benzene (3) within a cyclodextrin (4), was published in 1958[11] (Scheme 1a).[12]

The first [2]catenane for which evidence was put forward in support of its structure was synthesized two years later by Wasserman.[1] IR spectroscopy indicated that a small (ca. 0.0001 %)[1b] fraction of the product obtained through the acyloin condensation of diester 6 in the presence of deuterated macrocycle 5 contained both deuterium atoms and the acyloin (-COCHOH-) group. It seems that this combination of functional groups in a single molecule can only result from this reaction through the formation of a catenane, 7 (Scheme 1b), although no definitive proof of the structure (e.g. from mass spectrometry) was ever presented.[1b] After oxidative cleavage of the acyloin group in the fraction of the reaction products containing the putative catenane, a small amount of 5 was recovered, again supporting the suggestion

Dr. G. Gil-Ramírez, Prof. D. A. Leigh, A. J. Stephens School of Chemistry, University of Manchester Oxford Road, Manchester, M13 9PL (UK) E-mail: david.leigh@manchester.ac.uk Homepage: http://www.catenane.net

Scheme 1. a) Attempted synthesis of a [2]catenane by Luttringhaus and Cramer (1958). Complexation between dithiol 3 and a-cyclodextrin 4 led to a threaded inclusion complex (2), but subsequent oxidation did not afford catenane 1.[11] b) Wasserman's 1960 synthesis of a [2]cate-nane (7) by statistical threading of diester 6 through macrocycle 5 during an acyloin condensation.[1]

that the deuterated cycloalkane was a component of a mechanically interlocked molecule. To date, 7 remains the only catenane synthesized that incorporates a fully saturated cycloalkane.

In their seminal 1961 discussion of "Chemical Topology"[2] Frisch and Wasserman began to consider ways of overcoming the limitations of statistical methods, and suggested the use of molecular scaffolds and directed synthesis to obtain mechanically interlocked links and knots.[13] The experimental realization of a covalent-bond-directed catenane synthesis by Schill and Luttringhaus followed shortly afterwards

Scheme 2. Schill and Luttringhaus's directed synthesis of a [2]catenane (1964).[14]

(Scheme 2).[14] In their stepwise route, the positioning of the amino group within the macrocyclic cavity of intermediate 8, together with the electrophilic alkyl chlorides situated above and below the plane of the macrocycle, is key in directing intramolecular cyclization to give the threaded structure 9. Cleavage of the aryl-nitrogen bond afforded [2]catenane 10 in 15 steps from a readily obtainable phosphonium salt. This first example remained one of the most efficient demonstrations of catenane synthesis for almost 20 years, only superseded by Sauvage's application of template methods (see Section 2.1). The ingenuity demonstrated in the Schill and Luttringhaus 1964 synthesis was the forerunner of chemists applying their imagination and skills to the synthesis of interlocked molecules for the next half-century.

Guzmán Gil-Ramirez obtained his BSc at Jaume I University (Castellón, Spain). He moved to Tarragona (Spain) in 2005 and completed his PhD under the supervision of Prof. Ballester at ICIQ studying anion-p interactions and self-assembled molecular capsules. In 2010 he moved to Oxford working with Prof. Anderson and Prof. Briggs developing N@C60-based materials for quantum information processing. In 2013 he joined Prof. Leigh's group. His research interests include organic functional materials and controlled molecular motion.

Alexander Stephens studied for an MChem at the University of Sheffield (UK), and was awarded the WebElements prize on graduating in 2011. He began his PhD in 2012 under the supervision of Prof. David Leigh, initially based at the University of Edinburgh (UK) but later relocated with the group to the University of Manchester (UK). He is working on the development of new synthetic tactics towards topologically complex molecules.

David Leigh obtained his BSc and PhD from the University of Sheffield. After postdoctoral research in Ottawa (1987-1989) he was appointed to a Lectureship at the University of Manchester Institute of Science and Technology (UK). After positions at the Universities of Warwick and Edinburgh, he returned to Manchester in 2012, where he currently holds the Sir Samuel Hall Chair of Chemistry. His research interests include chemical topology and synthetic molecular-level motors and machines.

synthetic routes to circumvent the formation of [3]catenane isomers, but these ultimately proved unsuccessful.[16]

In the late 1950s Frisch and Wasserman[2] and van Gulick[17] independently conceived an alternative synthetic strategy to knots, catenanes, and other topologically complex structures: the "Möbius strip approach". A Möbius strip is a paradromic ring with a half-twist: that is a surface with only one side and one boundary component, formed by twisting a strip and connecting the ends to form a loop. If a paradromic ring is cut along its center, a single ring, link, or knot results (a half-twist Möbius strip affords a ring, even numbers of half twists give links, odd numbers of half twists give knots). In chemical terms the macrocyclization of a ladder-shaped molecule can, in principle, generate loops with differing numbers of twists (Figure 1). Cleavage of the ladder "rungs" would afford various topoisomers (Figure 1 c).

Scheme 3. Schill's directed synthesis of a [3]catenane (1969).1

In the decade following the synthesis of [2]catenane 10, Schill et al. extended the "directed synthesis" approach to produce [3]catenanes (e.g. Scheme 3).[15] Macrocyclization of dibromide 11 with diamine 12 gives an inseparable mixture of precatenane isomers 13 and 14. Cleavage of the directing amino groups afforded a [3]catenane as a mixture of isomers 15 a and 15 b. Schill etal. also investigated alternative

Figure 1. The "Mobius strip" approach to generating different topological isomers. The macrocyclization of a ladder-shaped molecule (a) can form a range of cyclic molecules containing different numbers of twists (b). Cleavage of the rungs yields different topological isomers (c).

A "Möbius strip mechanism" was originally proposed to account for the apparent formation of catenanes during the metal-catalyzed olefin metathesis of various cycloalkenes.[18] However, this explanation fell out of favor when the mechanism of olefin metathesis was shown to involve metal-alkylidene and metallacyclobutane intermediates, incompatible with a paradromic ring mechanism for the formation of interlocked molecules.[13a] In the early 1980s, Walba et al. synthesized a number of molecular strips based on a tetraether of tetrahydroxymethylethylene (Scheme 4).[19] Cyclization of 16 yielded two products, identified as untwisted cylinder 17 and the half-twist Möbius strip 18. Scission of the strips was carried out by ozonolysis. Cylinder 17 afforded two molar equivalents of triketone macrocycle 19, while the half-twist Möbius strip 18 generated the hexaketone macrocycle 20. However, the double and triple twist paradromic rings, which would be precursors to a [2]catenane and trefoil knot, respectively, could not be isolated, probably because of their very low yields of formation.[19b]

Scheme 4. Walba's synthesis of a molecular Möbius strip,

mathematics of topology. Modern knot theory also includes the study of links, which are multiple loops (knotted or not) entwined and linked with one another, whereas knots are formally a single entwined loop. As the same definition used to describe a mathematical link can be applied to catenanes, it follows that all links (with the exception of those with zero crossings) expressed in molecular form are catenanes. Clearly an infinite number of interwoven molecular ring patterns are theoretically possible. Nevertheless, only a handful of different catenane topologies have been made to date.

The Alexander-Briggs notation[22] is commonly used to mathematically describe knots and links. It is written in the form Xy, in which x corresponds to the number of crossings within the system, y is the number of discrete components (loops), and z is the order of the knot used to distinguish a given topology from others with the same x and y descriptors. For example, the simplest type of [2]catenane with two crossings is the 21 link, also known as a Hopf link (Figure 2).

Although the early directed and statistical methods to catenanes showed that the synthesis of interlocked molecules was experimentally feasible, the low yields and long multistep routes meant that a radically new approach to their synthesis was necessary for catenanes to become more than academic curiosities. The revolution in catenane synthesis started in 1983 when Strasbourg chemist Jean-Pierre Sauvage recognized that the orthogonal arrangement of bidentate ligands in a tetrahedral Cu1 complex could be used to generate the crossing points necessary for catenane formation.[13c] Template synthesis has been the basis for forming the vast majority of catenanes ever since (Sections 2-4). However, before discussing the phenomenal advances in catenane assembly achieved during the "template synthesis era", it is first useful to connect chemical structure with the nomenclature used to describe the topology of links in mathematics (Section 1.2), and to put small-molecule catenanes into a broader context by giving a brief account of catenane macromolecules found in biology or assembled using biopolymer building blocks (Sections 1.3 and 1.4).

1.2. Link Topology

In the 19th century, interlocked and entwined rings captured the attention of mathematician Peter Guthrie Tait. Inspired by Lord Kelvin's proposal that elements were knotted structures in the aether,[20] Tait set about tabulating different types of knots in the hope that it might be related to the periodic table of elements.[21] Although the association between knots and atoms proved to be incorrect, Tait's work paved the way to the study of knots and the formulation of the

Figure 2. A selection of prime links with up to eight crossing points, and their Alexander-Briggs notation.

For a given link there can be multiple ways of graphically representing the topology, with interconversion between the different representations achieved through structural deformation and/or translation of the crossing points. For example, Solomon links (4^) consist of two loops interlocked twice with one another in the simplest depiction of the topology (Figure 3 A), but by deformation, other forms of the Solomon link can be created which are topologically equivalent (Figure 3B-D).

Figure 3. Four different representations of a Solomon link (4^)

1.3. Naturally Occurring Catenanes in Biopolymers

Unbeknownst to the chemists devising the early synthetic routes to catenanes, mechanically linked molecular rings occur naturally. In 1967 electron microscopy revealed catenanes made up of circular DNA isolated from the mitochondria of HeLa cells'23] and human leukaemic leucocytes'24 (Figure 4). Since then, many different interlocked DNA

Figure 4. a) Electron micrograph of circular DNA revealing a catenane topology. b,c) Highlighting the two component rings of the DNA catenane as a Hopf link. Modified from Ref. [23] with permission.

topologies have been both discovered and synthesized.'251 Topoisomerase enzymes are responsible for controlling DNA topology.'261 They function by binding to entangled DNA and cutting through the phosphate backbone to introduce gaps in the double helix, which subsequently unwind and disentangle the molecular threads. As enzymes responsible for transcription and replication require access to non-interlocked single strands of DNA to properly read the relevant genetic sequences, the formation and disentanglement of DNA catenanes and knots is of significance for both cell reproduction and gene expression.

Threaded entanglements in proteins, usually in the form of open knots,'27] can be crucial to their tertiary structures and therefore also important for their function. Protein catenanes have also been discovered in natural systems. The X-ray crystal structure of the bacteriophage HK97s capsid'28] shows it to be composed of a network of cyclic proteins interlocked with one another to create a "molecular chainmail" (Figure 5). The relatively thin bacteriophage capsid is thought to gain additional stability by interlocking proteins in this manner, thus providing a means of protecting the genetic information contained within the capsid necessary for virus replication, even in harsh environments. Four other natural protein catenanes are known: 1) Pyrobaculum aerophilum citrate synthase (PaCS) proteins,'29] in which the formation of intramolecular disulfide bonds result in an interlocked dimeric assembly of the biopolymers. This feature enhances the proteins thermal stability, which may be particularly important for Pyrobaculum aerophilum given that it is a thermophile (i.e. an organism that thrives at high temperatures) ; 2) two interlocked protein rings in mutant bovine mitochondrial peroxiredoxin III;'30] 3) Recombination R (RecR) proteins used for recombinatorial DNA repair in the Deinococcus radiodurans bacterium which form interlocked octamers,'31] an arrangement which it has been suggested may enable the RecR rings to open and close and

Figure 5. The "chainmail" arrangement of proteins found in bacteriophage HK97s capsid (colored sections highlight the individual protein rings). a) The repeating pattern of interlocking proteins which constitute the spherical capsid. b) A cross-section of the capsid in which three protein rings interlock with one another. c) Magnified view of the position at which protein rings overlap (cross-linking isopeptide bonds are highlighted). Reprinted from Ref. [28] with permission.

act as a DNA clamp during the repair process; 4) E. coli Class 1a ribonucleotide reductase (RNR) catenanes.'32] Additionally, there is evidence to support the discovery of naturally occurring catenanes in the CS2 hydrolase enzymes found in archaeon Acidianus A1-3.'33]

1.4. Catenanes Constructed Using Synthetic Biopolymers

In recent years the programmability of DNA recognition sequences and automated synthesis has enabled the preparation of catenanes and other more complicated topological structures from DNA. Synthetic routes to DNA catenanes'34] generally involve the use of single strands of DNA that contain specific sequences of nucleic acids designed to fold into double-stranded helical turns with a complementary sequence contained on a separate thread, thereby creating crossing points between the DNA strands. Catenanes are formed by ligation of the loose ends (Figure 6). An additional level of control can be imparted to the structural assembly by using B or Z DNA to form right- or left-handed helical twists, respectively. By utilizing these approaches Seeman and co-workers have synthesized an array of DNA structures including Borromean rings,'35] a truncated octahedron,'361

Figure 6. Forming catenanes from single strands of synthetic DNA. Complementary sequences of DNA (represented by colored spheres) fold into helicates, the subsequent ligation of the strands affords catenanes.

and DNA cubes.'37] Recently, efforts have been made to control the relative motion of the rings in synthetic DNA catenanes for their possible use as molecular machines.'38

Although their design and synthesis presents more challenges than their DNA counterparts, protein catenanes have also been successfully synthesized.'391 The dimeric folding of a tumour suppressor protein p53 to form bisecting concave peptide strands has been exploited such that intramolecular cyclization by native chemical ligation forms a catenane.

2. Synthesis of Hopf Link (Singly Interlocked) [2]Catenanes

2.1. Catenane Synthesis Promoted by Noncovalent Interactions

The early statistical and directed synthetic approaches to mechanically interlocked molecules'401 suffered from low yields and/or lengthy synthetic procedures (Section 1.1). This meant that catenanes remained firmly in the realm of laboratory curiosities rather than molecular architectures that could be usefully exploited. The situation changed with the application of template synthesis to the preparation of entwined and interlocked structures.'411 This enabled cate-nanes to be made on a significant scale for the first time. The formation of an interlocked molecule requires the physical overlap of molecular strands to create crossing points prior to covalent capture, a task for which templates are ideal, as they act to gather molecules into preorganized assemblies.'421 In addition to the often high yields and relatively short synthetic pathways associated with template synthesis, predictability and reliability in template preorganization enables the rational design of synthetic pathways to more complex topologies. Over the past three decades these have afforded a number of fascinating (and often beautiful) mechanically interlocked molecular architectures of ever increasing complexity.

2.1.1. Passive Metal Templates

The organization of ligands about transition-metal cations has proven to be the one of the most practical, versatile, and reliable methods for creating molecular crossing points,'41,42b,-c,e,f,h,i] owing to the various well-defined coordination geometries that transition-metal complexes adopt. Sauvage's group was the first to exploit this in the synthesis of interlocked molecules and introduced metal template '2]cat-

Schemes. Sauvage's metal-template synthesis of a [2]catenane by Williamson ether macrocyclization of the tetrahedral coordination complex 23-Cu1, formed by coordination ofligand 21 and macrocycle 22 to CuI.[43]

enane synthesis in 1983 (Scheme 5).'43] Their original system employed a hydroxy-functionalized 2,9-diphenyl-1,10-phe-nanthroline (dpp) ligand 21 and related macrocycle 22 that adopt a threaded structure on coordination to a tetrahedral Cu1 ion. The perpendicular ligand arrangement generates a crossing point. Subsequent "clipping" by a Williamson ether macrocyclization reaction covalently captured the '2]cate-nane 24-Cu1 in 42 % yield. The copper ion could be removed quantitatively from catenate 24-Cu1 with potassium cyanide to afford the metal-free catenand 24 (Scheme 5).

Developments in covalent-bond-forming reactions are often quickly exploited in the synthesis of interlocked molecules. A double ring-closing olefin metathesis (RCM) macrocyclization reaction improved the '2]catenane yield from suitably functionalized dpp ligands'44] to 90% (Scheme 6).

A significant feature of metal-template catenane synthesis is the possibility of removing the metal ion(s) after covalent capture of the interlocked topology (e.g. catenane 24, Scheme 5). With the template removed, the macrocycle components are generally able to move with respect to each other with little or no intercomponent noncovalent interactions impeding the motion, within the topological con-straints.'45]

Since the original metal-template strategy with dpp ligands and CuI ions was introduced, catenanes have been synthesized with a range of different metal templates, including the use of octahedral,'461 trigonal bipydramidal,'47] square planar,'48] and linear'49] transition-metal templates.

Scheme 6. Sauvage's high-yielding olefin metathesis macrocyclization to form [2]catenane 26-Cu'.[44] Electrostatic interactions between the glycol ether oxygen atoms and the aromatic protons of the metal-coordinated dpp ligands may help promote intramolecular macrocycli-zation.

Recently, lanthanide metal ions have also been utilized to template the formation of mechanically interlocked molecules.'50!

An additional layer of control can be achieved in metal-template catenane synthesis by using reversible reactions to maximize the yield of the most thermodynamically favorable product, which can often be designed to be the catenane. Imine bond formation is a particularly attractive reaction for preparing metal-template-interlocked molecules; it occurs under mild conditions and its reversible formation allows "error-checking" during the assembly process (Scheme 7).'46d,51]

Ligand 27 is based on the structure successfully used to assemble hydrogen-bond-assembled benzylic amide cate-nanes and rotaxanes.'52] It is modified to incorporate a triden-tate 2,6-diiminopyridine coordination motif that allows two distinct routes for the preparation of '2]catenanes. One uses precatenane complex 28-Mn. Subsequent double macrocycli-zation by RCM of the terminal alkenes affords the '2]cate-nane 29-Mn in good yields. Alternatively, catenane formation can be carried out under thermodynamic control by exploiting the reversibility of imine bond formation between the bis(benzylamine) chain 30 and 2,6-diformylpyridine in the presence of a divalent metal salt (Scheme 7). The solid-state structures of several of the catenates show interligand aromatic-stacking interactions which are a feature of benzylic amide catenanes and rotaxanes and serve to favor the assembly of the interlocked structure rather than macrocyclic isomers.

The use of attractive metal-metal interactions to template the formation of catenanes has been investigated as an alternative to utilizing their coordination geometries. Beer and co-workers found that '2]catenane 32-CunCum (Scheme 8) was assembled through the rearrangement of

Scheme 7. Octahedral [2]catenanes 29-M11 formed around a range of divalent metal ions through double RCM reactions from precatenane 28-M11 or assembled by imine bond formation.[46d] The quoted yields refer to the double RCM route. Stabilizing p-p interactions between the phenyl and the pyridyl rings likely play a significant role in the assembly process.

Scheme 8. Beer's mixed-valence [2]catenane generated from partial oxidation of macrocycle 31-Cu".[53] Alkyl side chains have been omitted from the crystal structure for clarity.

macrocycle 31-Cu11 following its partial oxidation.'53] Chargetransfer interactions between the CuII and CuIII centers in the mixed-valence catenane probably provide the driving force for interlocking of the macrocycles. This strategy was later extended to the synthesis of mixed-metal CuII-AuIII catenanes

from the corresponding homometallic macrocycle precur-sors.'54]

Aurophilic interactions have also been employed to assemble small organometallic fragments into inorganic catenanes. The first such example was reported by Mingos et al. (Scheme 9 a),'55] wherein alkyne 33 was treated with 'Au(NH3)2]+ to generate '2]catenane 34-Au\2. The combination of h1-Au-h1, h2-Au-h1, and ^2-Au-^2 alkyne-Au1 coordination modes established between the components generates interlocked hexameric macrocycles that are stabilized by the multiple aurophilic interactions between the AuI centers. AuI-

2.1.2. Active Metal-Template Synthesis

Whilst passive metal templates often provide an effective means of gathering and organizing ligands into interlocked architectures (or precursors to such structures), they do not take advantage of another common feature of transition metals, namely the ability to catalyze covalent-bond-forming reactions. In the last decade, several "active metal-template" strategies to interlocked molecules have been developed.'57] In active template synthesis the metal ion plays a dual role, acting as a template to entwine or thread the building blocks while also actively catalyzing the bond-forming reaction that covalently traps the interlocked structure. In active metal template reactions substoichiometric amounts of metal can often be employed,'58] whilst the use of nonpermanent recognition motifs also allows for the traceless synthesis of interlocked molecules.'59] Initial investigations into the utilization of active metal templates focused on the synthesis of rotaxanes,'58-60] but these were later extended to include catenanes'61] and a trefoil knot.'62]

A commonly used reaction in active metal-template synthesis is the CuAAC click reaction.'63] The synthesis of a heterocircuit (comprised of two different rings) '2]catenane was achieved by macrocyclization of thread 37 around macrocycle 38'61a] (Scheme 10 a). First a CuI cation coordinates both to a chelating pyridyl within macrocycle 38 and the

Scheme 10. The active metal-template synthesis of a) heterocircuit [2]catenane 39 and b) homocircuit [2]catenane 41 by CuAAC macrocyclization reactions.[61a]

Scheme 9. Catenane formation driven by aurophilic interactions from a) Mingos et al. (34-Au'12)[55] and b) Che et al. (36-AuI11).[56] In the crystal structure of 36-Au'ln, the sugars have been omitted for clarity, whilst the sulfur atoms that constitute the pentameric ring are shown in blue.

thiolate interactions have also been employed to generate a '2]catenane (Scheme 9b).'56]

complementary azide and alkyne functional groups within 37 to gather the loose thread ends within the cavity of the macrocycle, and thus create a crossing point. Covalent capture of the interlocked product occurs by the Cu1-catalyzed cycloaddition reaction between the thread ends to generate [2]catenane 39 (53% yield). The synthesis of an analogous [2]catenane with a modified macrocycle containing a bipyridine chelating site was also demonstrated, although longer reaction times and higher concentrations were required.

The system was also adapted for a double macrocycliza-tion process using thread 40 (Scheme 10b), in which a macrocycle is formed in situ by macrocyclization of the thread. A second thread macrocyclization follows through the macrocycles cavity as before, producing a homocircuit (comprised of two identical rings) [2]catenane 41. The effectiveness of the active metal-template reaction in being able to catalyze two distinct covalent-bond-forming reactions and also act as a template is apparent, as the [2]catenane was formed in reasonable yield (46%) with little of the non-interlocked macrocycle by-product obtained (< 7 %).

In addition to the use of CuAAC click reactions in active metal-template synthesis, [2]catenanes have been prepared using the Cadiot-Chodkiewicz coupling between a terminal alkyne and a bromoalkyne,[61a] as well as an oxidative alkyne homocoupling, both Cu'-catalyzed reactions.[61b]

2.7.3. Catenanes Assembled through p-p Stacking Interactions

Stoddart et al. carried out a seminal research program on the synthesis of catenanes and rotaxanes by exploiting the interactions of p-electron-rich (donor) and p-electron-defi-cient (acceptor) aromatic rings to direct the threading of molecular strands through macrocycles.[64] The first example was reported in 1989, with the synthesis of a donor/acceptor [2]catenane achieved by combining thread 42 with macrocycle 43 and 1,4-bis(bromomethyl)benzene 44 in acetoni-trile[65] (Scheme 11). Given that dication 42 has no significant interaction with the electron-rich macrocycle 43, the proposed mechanism is that the thread first reacts with 1,4-bis(bromomethyl)benzene to produce a tricationic species containing an electron-deficient bipyridinium motif, which does bind strongly within the cavity of 43 (Scheme 11, intermediate shown in square brackets). The resulting [2]cat-enane, 45, is formed in 70% yield and can also be formed directly from 4,4'-bipyridine, 1,4-bis(bromomethyl)benzene, and macrocycle 43.[66] A large number of mechanically interlocked molecular systems have subsequently been developed based on this assembly motif.[67]

Sanders and co-workers have developed neutral donor-acceptor ligand sets that assemble to form catenanes[68] (Scheme 12). These have improved chemical stability and better solubility in organic solvents than Stoddart's cationic viologen-based macrocycles and catenanes. Combining two molecules of thread 46 with macrocycle 47, followed by intermolecular oxidative coupling of the terminal alkynes of 46, afforded [2]catenane 48.

Recently, electron-rich pillar[5]arene cyclophanes have been used in the diastereoselective synthesis of a [2]cate-

Scheme 11. Stoddart's first [2]catenane, featuring p-electron-rich/p-electron-poor aromatic stacking.[65]

Scheme 12. Sanders' use of neutral p-electron-rich and poor motifs to form a [2]catenane by oxidative coupling of alkynes.[68]

nane[69] (Scheme 13). Thread 49, which contains an electron-poor pyridinium unit, and pillar[5]arene 50 assemble to form a pseudorotaxane. Subsequent RCM under high dilution conditions (0.5 mM) generated the [2]catenane 51 (25%). Rotation of the alkoxybenzene units of the pillar[5]arene can potentially result in eight different diastereoisomers of catenane 51, but only one is observed (all the alkoxybenzene units have the same orientation relative to one another).[70]

2.1.4. Hydrogen and Halogen-Bond Templates

In addition to the use of transition-metal templates and p-stacking interactions to direct the assembly of mechanically interlocked molecules, hydrogen bonding has been widely used to promote the synthesis of catenanes. The discovery of

Scheme 13. Ogoshi's synthesis of a [2]catenane assembled by threading of a pyridinium salt through the cavity of a pillar[5]arene cyclo-phane.[69]

Figure 7. Hunter's benzoquinone receptor.

the first such system was serendipitous: Hunter and Purvis wanted to investigate 52 as a receptor for p-benzoquinone[71] (Figure 7). In an attempt to improve the yield (10%) of the macrocyclization to obtain 52, a two-step procedure was carried out in which diamine 54 was isolated prior to the intended macrocyclization (Scheme 14).[72,73] The reaction of diamine 54 with isophthaloyl dichloride led to the expected macrocycle, 52, along with two other products which both had twice the molecular mass of the targeted macrocycle. One of the side products was quickly confirmed as the tetrameric macrocycle 55 (Scheme 14). However, structural determination of the third product was more challenging. The molecule's complex 1H NMR spectra and low polarity hinted at the formation of an interlocked structure, which was verified by careful 1H NMR analysis as the [2]catenane 56 (Scheme 14). Remarkably, the [2]catenane was obtained in a one-pot double macrocyclization, in which the initial cyclization of 54 with isophthaloyl dichloride affords macrocycle 52 before a second cyclization proceeds through the cavity of the first macrocycle ring, thereby giving 56 in 34 % yield.

Around the same time, Vogtle's group was investigating the synthesis of similar amide macrocycles[74] and carried out the condensation of diamine 53 and 5-methoxyisophthaloyl dichloride to form a [2]catenane in 10% yield. A few days after Hunter's [2]catenane 56 was published, and presumably prompted by Hunter's detailed 1H NMR analysis proving the catenane structure, Vogtle submitted his own paper on the synthesis of a very similar [2]catenane.[75] In both cases the presence of bulky cyclohexyl side groups prevents free rotation of the macrocycles within the resulting [2]catenane.

Scheme 14. Hunter's synthesis of an amide-based [2]catenane 56 and its X-ray crystal structure (solvent molecules and non-amide hydrogen atoms omitted for clarity).'72,731 Vogtle reported the synthesis of a very similar [2]catenane shortly afterwards.'741

In terms of dynamic properties and their potential use in molecular machines, the Hunter/Vogtle amide catenanes were largely superseded by the benzylic amide [2]catenane system, serendipitously discovered whilst preparing a macro-cyclic receptor for CO2.[76] [2]Catenane 57 was prepared in a single step using dibenzylamine and isophthaloyl dichloride (Scheme 15) in a direct eight-molecule condensation reaction that afforded the catenane in 20 % yield. To this day, the one-step, chromatography-free, synthesis of 57 from commercially available chemicals remains one of the simplest ways of accessing a mechanically interlocked molecule. The X-ray structure, the first of an amide catenane, confirmed the amide-amide hydrogen-bonding arrays, as well as showing a network of p-p stacking interactions further stabilizing the structure (Scheme 15). Unlike the previous examples of amide catenanes, the macrocycle components of catenane 57 are able to rotate around one another in solution.[77]

Scheme 15. The eight-molecule condensation to form benzylic amide [2]catenane 57.[76]

Benzylic amide catenanes are generally straightforward to prepare, have high structural versatility as well as variable and controllable ring dynamics.[78]

The use of halogen bonding[79] in the template synthesis of [2]catenanes has been reported by the Beer group. [2]Cate-nane 60-Br~ was assembled in 24 % yield from RCM of 58 and 58-Br~ (Scheme 16).[80] When the macrocyclization was performed in the absence of the bromide ion, using only precursor 58, no [2]catenane was formed.

A pyridinium iodide/pyridine interaction may play a significant role in the assembly of [2]catenane 62 (Scheme 17).[81] The precursor pseudorotaxane complex 61 has a significantly higher association constant in dichloromethane (Ka = 180 ± 20 m_1) than when the iodine atom is not present (Ka = 30 m_1). Ring-closing olefin metathesis of 61 generated [2]catenane 62, albeit in modest yield (6.5%).

2.1.5. Catenane Synthesis Driven by Hydrophobic Effects

Solvophobic effects can be used to favor the threading of components to create interlocked molecules (or their precursors), acting to minimize the surface area of hydrophobic motifs that are exposed to a polar solvent by forming inclusion complexes.[82] Although Cramer and Luttringhaus's efforts[11] to synthesize a [2]catenane from cyclodextrin complexes in the 1950s were unsuccessful (as outlined in Section 1.1), in the 1990s Stoddart and co-workers were able to isolate an interlocked product by using a similar approach[12] (Scheme 18). The inclusion complex of 63 with heptakis(2,6-di-O-methyl)-p-cyclodextrin (DM-p-CD, 64) was cyclized with terephthaloyl dichloride to afford [2]cate-nane 65 in 3 % yield.

Hydrophobic effects can also be used to generate a thermodynamic bias to favor the interlocking of macrocycles at equilibrium. This concept was discovered serendip-itously in the form of Fujita's "magic rings"[83] (Scheme 19). In basic aqueous solution, macrocycle 67-Pdn spontaneously

Scheme 16. Beer's synthesis of [2]catenane 60-Br directed by halogen bonding.[80]

Scheme 17. Beer's synthesis of a [2]catenane stabilized by a pyridinium iodide—pyridine interaction.'81

interlocks to form [2]catenane 68-Pdn. At low concentrations the equilibrium lies towards the macrocycle, however at high concentrations (or at high salt concentrations to increase the hydrophobic effect) the [2]catenane is formed almost quantitatively.^

Scheme 18. Stoddart's synthesis of [2]catenane 65, promoted by hydrophobic binding.'82'

Scheme 19. Fujita's "magic ring" [2]catenane synthesis. Reversible coordination ofligand 66 with Pd11 generates an interconverting mixture of macrocycle 67-Pd" and [2]catenane 68-Pd", with the [2]catenane energetically favored at high concentrations through hydrophobic binding.'83'841

2.1.6. Alkali Metal Cation Templates

Scheme 20. Chiu's Na'-template synthesis of a [2]catenane.

Although size-discrimination makes alkali metal cations Scheme 21. Beer's synthesis of a [2'catenane using a chloride anion excellent templates for different-sized crown ethers,[85] their template.'881

6122 www.angewandte.org

** These are not the final page numbers!

lack of well-defined 3D coordination geometries generally make them unsuitable templates for interlocked molecules. Nevertheless, an example of an alkaline cation template synthesis of a [2]catenane formed under thermodynamic control has been reported by Chiu and co-workers (Scheme 20).[86] The system combines diamine 69 with dialdehyde 70 through reversible imine bond formation in the presence of a sodium salt in a low polarity solvent to give the [2]catenane 72-Na1 (Scheme 20). To maximize the template effect, tetrakis[3,5-bis(trifluoromethyl)phenyl]borate (TFPB) was used as a weakly coordinating counterion for the sodium cation. In the presence of 1 equiv of NaTFPB, only macrocycle 71-Na1 is observed in the reaction mixture (65 % yield). However, when the reaction is carried out with 0.5 equiv NaTFPB, [2]catenane 72-Na1 is formed, which after reduction of the imines, affords [2]catenane 73 in 17 % yield over the two steps.

2.7.7. Anion Templates

The use of anions as templates for interlocked molecules has only developed over the last decade.[87] Whilst some anions (e.g. halides and oxoanions) are capable of forming strong electrostatic interactions due to their small size and high charge density, the absence of predictable coordination geometries in their association complexes often makes it difficult to design the formation of molecular crossing points through their use as templates. Beer and co-workers have overcome some of these problems by utilizing a combination of noncovalent interactions in conjunction with anion templates to impose control over the directionality of the assembly. For example, in the formation of the chloride-complexed [2]catenane precursor 76-Cl~ (Scheme 21), macrocycle 74 contains 1) an isophthalamide unit for hydrogen bonding, 2) hydroquinone groups for p-p stacking, and 3) a glycol chain for electrostatic interactions with electron-poor

groups and the ability to fold into a low-energy cyclic conformation because of multiple gauche effects. Accordingly, [2]catenane 77-Cr is obtained in 45 % yield (Scheme 21).[88] This strategy was further improved by using two diamidopyridinium-containing threads of 75 assembled around a single chloride anion template followed by double macrocyclization by RCM to afford a [2]catenane in 78% yield.[89]

2.1.8. Radical-Radical Interactions

The ability of stable cationic radical viologen species to dimerize in solution has been known for over fifty years,[90] with the pairing of the radicals driven by the resulting closed-shell electron configurations. Stoddart and co-workers

Scheme 22. Stoddart's synthesis of a [2]catenane through radical-pairing interactions.'91'

Scheme 23. Sanders' [2]catenane formed within a DCL and amplified by acetylcholine biding.[93]

Scheme 24. Sanders' partial control over the composition of [2]catenanes formed from a disulfide-based DCL can be achieved by choice of the DN building block. In both cases, the DCLs initially also contained non-interlocked macrocycles. Amplification of [2]catenane 85 was achieved by the addition of either cationic templates to intercalate between the catenanes inner NDI units, or polar salts to enhance hydrophobic effects. Amplification of [2]catenane 87 was achieved by the addition of polar salts, whilst amplification of 88 was achieved by employing a threefold excess of DN 86 relative to NDI 83, and increasing the solvent ionic strength.[95,96]

recently reported the use of this dimerization phenomenon to assemble [2]catenanes (Scheme 22).[91] Reduction of 782+ to the monoradical species is achieved with an excess of zinc dust, thereby resulting in a threaded dimeric assembly 792('+) (Scheme 22). Cyclization with 4,4'-bipyridine affords [2]cat-enane 804('+). Although the initial radical quenching occurs on exposure to air to give an equilibrium mixture of 802'6+ and 80'7+, the radical species persist for several weeks under ambient conditions, thus requiring forcing oxidation conditions to completely generate the nonradical 808+ species.

2.2. Catenanes from Dynamic Combinatorial Libraries

Dynamic combinatorial libraries (DCLs) are systems consisting of molecular building blocks that combine through reversible covalent bonds to produce a mixture ("library") of

[n]catenane network Figure 8. Classification of higher order [n]catenanes (n > 2). a) [3]Cate-nane topoisomers. b) The addition of macrocycles to a linear [3]cate-nane generates further topoisomers.

6124 www.angewandte.org

** These are not the final page numbers!

different species.[92] As the members of a DCL are in equilibrium, their relative thermodynamic stabilities determine the composition of the library, an equilibrium that can be shifted by changing the reaction conditions (e.g. by adding a substrate that noncovalently binds to, and thus lowers the free energy of, a particular species in the library).

Sanders, Otto, and co-workers discovered a [2]catenane[93] present in a DCL formed by hydrazone exchange of the peptide-based building block 81 (Scheme 23). The building block quickly forms linear oligomers that rearrange over 60 min into cyclic molecules of various sizes. The introduction of acetylcholine to the DCL causes predominant conversion of the library members into 82, a [2]catenane comprised of trimeric rings. Despite two diastereoisomers being possible (stereoisomers resulting from the directionality of the unsym-metrical rings), only one diastereoisomer is observed. Gagné and co-workers later modified the tripeptide building block to demonstrate the synthesis of a [2]catenane composed of tetrameric rings.[94] In this case, catenane formation is driven by hydrogen bonding, p stacking, and CH-p interactions rather than acetylcholine binding.

[2]Catenanes have also been discovered in DCLs where the components are interchanged through disulfide exchange reactions. Systems developed by Sanders and co-workers made use of components featuring naphthalenediimide (NDI) and dioxynaphthalene (DN) motifs (Scheme 24). Oxidation of dithiol-functionalized NDI 83 and cysteine-functionalized DN 84 in an aqueous solution containing NaNO3 produced a [2]catenane 85 composed of heterodimer rings (i.e. containing both an NDI and DN unit) arranged in a donor-acceptor-acceptor-donor (DAAD) fashion.[95] Modification of the DN building block to give 86 afforded the analogous DAAD catenane 87 along with [2]catenane 88 possessing a DADD arrangement of building blocks.[96] The DAAD conformation adopted by [2]catenanes 85 and 87 is such that p-p stacking between the components is maximized, whilst the otherwise unfavorable stacking of p-electron donors is overcome in 88 by hydrophobic forces together with favorable aromatic stacking interactions generated in the DAD stack. The absence of catenanes with the thermodynamically optimal DADA conformation was attributed to the cavity of the acceptor homodimer being too small to permit threading of a donor unit.

3. Higher Order Linear and Radial [n]Catenanes

3.1. [n]Catenanes

macrocycles around a single central macrocycle (a "radial" [4]catenane, Figure 8 b) is also possible. The number of discrete topologies possible increases rapidly with the number of components (e.g. [n]catenane networks, Figure 8 b).

3.1.1. Linear [n]Catenanes

The first example of a linear [3]catenane was reported by Schill et al. , who employed a lengthy multistep synthesis based on his original directed [2]catenane synthesis (discussed in Section 1.1).[15] Later, Sauvage and Weiss employed metal-template synthesis to link two CuI-complexed threaded pseudorotaxanes 23-Cu1 to form a [3]catenane 89-Cu'2, albeit in modest (2%) yield (Scheme 25).[97] The metal

Scheme 25. Sauvage's synthesis of a [3]catenane by linking two Cu1-complexed pseudorotaxanes.[97,98]

The interlocking of two macrocycles can afford a number of [2]catenane topological isomers (e.g. Hopf link, Solomon link, Star of David catenane etc. , see Section 4). The addition of an extra macrocycle allows for an additional type of topoisomerism, in which the connectivity of the linked components varies (e.g. linear and cyclic [3]catenanes, Figure 8 a). The addition of further macrocycles increases both the number and complexity of topological isomers that are possible (Figure 8 b). For example, whilst a [4]catenane can have a linear connectivity of rings, the interlocking of several

catenate was demetalated with potassium cyanide to give the metal-free [3]catenand 89.[98] A subsequent development saw the use of the oxidative coupling of terminal acetylenes in place of the Williamson ether synthesis for the intermolecular macrocyclization, thereby increasing the yield of the [3]cat-enane to 58%.[99]

By increasing the size of the electron-poor macrocycle employed, the Stoddart group were able to promote the linking of two electron-rich macrocycles through a tetracat-

ionic cyclophane.[100] The tricationic intermediate resulting from the addition of 90 to 91 is large enough to cyclize around two crown ether macrocycles containing either hydroquinone (43) or 1,5-dioxynaphthalene (47) rings (Scheme 26). The resulting [3]catenanes 92 and 93 were isolated in 20% and 31 % yields, respectively.

92 R = -C6H4- 20%

93 R = -C10H6- 31%

Scheme 26. Stoddart's assembly of [3]catenanes.[1001

A similar approach was used to assemble a linear [5]catenane the Stoddart group termed "Olympiadane" because of its topology being shared with the symbol of the Olympic games.[101] The additional electron-rich 1,5-dioxy-naphthalene groups in the large crown ether macrocycles of [3]catenane 96, allowed for further cyclization of dication 42 and 1,4-bis(bromomethyl)benzene 44 to afford [5]catenane 97 in 5 % yield and [4]catenane 98 in 31 % yield (Scheme 27). The efficiency of the catenane-forming reaction was improved by the use of ultrahigh pressure conditions (97: 30% at 12kbar), which also formed the nonlinear [6]- and [7]catenanes in 28% and 26% yields, respectively. The rationale for the improved catenane yields using ultrahigh pressure conditions is that the number of components decreases in going from reactants to products.

Although examples of discrete linear [4]- or [5]catenanes are still rare in the literature, other successful strategies for the synthesis of linear [3]catenanes include an amide-based [3]catenane rotary motor (see section 5.2),[102] Fujita's Pd11

Scheme 27. Stoddart's linear [5]catenane "Olympiadane".[101

[3]catenane,[103] Loeb's host-guest [3]catenane,[104] and Sanders' [3]catenane generated from a DCL.[105]

3.7.2. Radial [n]Catenanes: n-1 Macrocycles Threaded onto a Central Macrocycle

Sauvage and co-workers reported the first examples of radial [n]catenanes, which they referred to as "multicate-nanes", in 1991.[106] By utilizing the intermolecular cyclization of Cu'-complexed pseudorotaxanes as previously described for the synthesis of linear [3]catenanes (Scheme 25), alkyne-functionalized pseudorotaxane 99-Cu1 was submitted to oxidative alkyne homocoupling conditions (Scheme 28). Along with the expected [3]catenate 100-Cu'2, which was formed in 23 % yield, a trimetallic complex was also generated in 23 % yield, tentatively assigned as a [4]catenate consisting of a central 66-membered hexayne ring with three peripheral 30-membered rings. Higher order homologues of these catenates also appeared to be formed in the reaction, with electrospray mass spectrometry of the crude reaction

Scheme 28. Sauvage's radial [n]catenanes generated by the oxidative homocoupling of alkyne-functionalized pseudorotaxanes, detected by ESI-MS. The yields are approximate and are based on m/z signal intensities from electrospray mass spectrometry. Catenanes 100-Cu'2, 101-Cu'3, and 102-Cu'4 could be isolated by chromatography, structures 103-Cu'5 and 104-Cu'6 could not.[106]

mixture providing evidence of radial [n]catenates with up to six rings around a central macrocycle.

A radial [4]catenane 110-Zn11, comprised of a six-porphy-rin nanoring with three mechanically interlocked phenanthro-line macrocycles 106, was assembled by Anderson and coworkers^ (Scheme 29). The rotaxane porphyrin dimer 107-Zn11 was prepared in 61 % yield through an active metal-template Glaser coupling between a monosilyl-protected alkyne-functionalized porphyrin 105-Znn and a phenanthro-line macrocycle 106. Unmasking of the rotaxane alkyne afforded 108-Znn. Reaction of rotaxane 108-Znn under oxidative homocoupling conditions in the presence of hex-apyridyl template 109, gave [4]catenane 110-Znn in 62% yield. The template, 109, could be removed in 89 % yield by treatment with DABCO.

Kim connected three cucurbituril-diammonium pseudor-otaxanes[107] into a cyclic array through platinum-pyridine coordination to form a "molecular necklace" radial [4]cate-nane in 90% yield (Scheme 30).[108] Altering the substitution pattern of the pyridyl group from 4- to 3-substitution gave a radial [5]catenane in 84% yield (Scheme 31).[109] A thread containing a phenanthroline ligand was used to connect two pseudo[3]rotaxanes with Cu11 ions to form a related radial [5]catenane.[110]

Bohmer and co-workers demonstrated the formation of an [8]catenane-like structure by ring closure of the hydrogen-bonded calix[4]arene dimer formed between 115 and 116 (Scheme 32).[111] The synthesis of a related bis([3]catenane) was demonstrated on replacing 115 with a calix[4]arene in which intramolecular cyclization generates two annular cycles.

3.1.3. Other [3]Catenane Topologies

Gunnlaugsson and co-workers have carried out a lanthanide-template synthesis of an interlocked structure, tentatively assigned as a cyclic [3]catenane (Scheme 33).[50a] Three equivalents of pyridyldiamide ligands 118 coordinate to a single Eu111 cation to generate an assembly from which a [3]catenane can be produced following triple intracompo-nent cyclization through RCM. However, alternative topologies (macrocycles, [2]catenanes, or knots)[50b] may also be produced from the clipping procedure in the absence of sufficient ligand preorganization to direct the ring-closing reactions. Both 1H NMR spectroscopy and mass spectrometry indicated that full closure of the complex had occurred following the reaction, and the detection of [2]catenanes containing two equivalents of cyclized 118 as a side product provides evidence for the formation of 119-Eum, thus indicating a preference for intraligand cyclization within the lanthanide-template assembly.

3.1.4. [n]Catenane Networks

Nature abhors a vacuum and the interweaving and interlocking of rings upon crystallization to form infinitely extended catenane structures is widespread, including examples derived from macrocycles and cages that contain dynamic (e.g. coordination) bonds.[112] As interlocking arises as a means of maximizing van der Waals interactions, and not because of interactions between specific recognition motifs, solvation of the network removes the driving force for the formation of polycatenanes and usually triggers their disassembly. As the consequences of mechanical bonding, such

Scheme 29. Anderson's synthesis of the radial [4]catenane 110-ZnM.[61c]

as large amplitude motion of the components, are generally not apparent in close-packed crystalline structures, the utility of such architectures is currently unclear.

Some of the earliest suggestions for physical manifestations of mechanical bonding are attributed to polycatenane networks. The anomalous physical properties of polysiloxanes were attributed to the presence of interlocked cyclic polymers in the 1950s[10] as was the depolymerization behavior and rubberlike elasticity of polymeric phosphonitrile chloride.[113] Half a century later Endo et al. proposed[114] a polycatenane network (121) is formed from the bulk polymerization of 1,2-dithiane (120; Scheme 34 a). Significant differences in the thermal and physical properties of 121 in comparison to linear poly(1,2-dithiane) prepared with benzylmercaptan end groups (Scheme 30b), and cyclic poly(oxoethylene) 122 infer the formation of a new network topology. Following polymerization of 120 in the presence of 122, the inability to remove 122 from the resulting polymer mixture, despite the

two polymers having significantly different solubilities in methanol, suggests the possible formation of an interlocked network. An analogous investigation into the ring-opening polymerization of 1,4-dihydro-2,3-benzodithiine afforded a polymer with similar properties to 121.[115]

3.2. Interlocked Cage Molecules

The interlocking of cage molecules can lead to novel catenane topologies through the interpenetration of multiple faces of the cage. The first example of a cage-based catenane was described by Fujita et al. in a one-pot ten-component self-assembly reaction (Scheme 35).[116] In D2O, two equivalents of the two tripodal monodentate pyridyl-based ligands 124 and 125 combined with six equivalents of square-planar Pd11 or Pt11 cations to generate the interlocked cage 126-Mn (Scheme 35). Hydrophobic effects, along with intercomponent p-p stack-

Scheme 30. Kim's synthesis of a radial [4]catenane "molecular necklace" n2-Pt", assembled from threaded cucurbituril macrocycles and Pt11 connecting units.[108]

Scheme32. Bohmer's [8]catenane-like structure based on interlocked calix[4]arene dimers. Hydrogen bonding between the urea motifs is illustrated with dashed lines.[111]

Scheme31. Kim's radial [5]catenane

triply interlocked chiral cages based on Co11 and Zn11 ing, drive the formation of the catenane under thermody- coordination by Hardie and co-workers,[117] quadruply namic control. stranded Pd11 cages by Kuroda and co-workers,[118] and

Several interlocked cage topologies which utilize metal- interlocked PdII cages containing phenothiazine by Clever ligand coordination have since been reported, such as the and co-workers.[119]

118 119-Eu"'

Scheme 33 Gunnlaugsson's lanthanide-template synthesis of an apparent link.[50a]

Scheme34. a) Endo's polycatenane networks obtained from polymerization of 1,2-dithiane. b) Synthesis of linear poly(1,2-dithiane).[114]

The synthetic routes to interlocked cages are not, however, limited to systems with metal coordination bonds. Beer and co-workers employed a sulfate anion to template the formation of catenane cage complex 130-S042~ (Scheme 36).[120] The product topology was inferred by mass spectrometry, 1H NMR spectroscopy, and diffusion ordered NMR spectroscopy (DOSY) experiments.

An alternative strategy to triply interlocked organic cages was serendipitously discovered by Cooper and co-workers.[121] In the presence of CF3CO2H, tri-aldehyde 131 and a suitable diamine (132 a-c) affords crystals of cage [2]catenanes 134 a, 134 b, or 134 c (from diamines 132 a, 132 b, or 132 c respectively; Scheme 37). Remarkably, formation of the interlocked product is achieved despite the absence of any apparent recognition motifs to associate the two cage components. It is likely that the catenane preferentially crystallizes from the reaction mixture (the interlocked structure presumably fills the void space more effectively and maximizes van der Waals interactions in the solid state), while the reversibility of imine bond formation produces more catenane in solution through Le Chatelier's principle during the several weeks of the

Scheme35. Fujita's triply interlocked cage complex formed by the self-assembly of four organic ligands with six square-planar-coordinating metal cations (Pd11 or Pt").[1161 The interplanar distance between the aromatic faces of the cages is ideal for generating strong p-p stacking interactions with a separate intercalated cage.

crystallization process. The resulting catenane cages are, however, relatively stable when taken up in solution under neutral conditions, kinetically trapped by the relatively slow dynamics of imine bond formation.

A [2]catenane cage was recently described by Mastalerz and co-workers,[123] in which reversible boronic ester formation was utilized in a 40-component assembly process to generate a quadruply interlocked catenane. The groups of Nitschke and Sanders have reported the synthesis of a [7]cat-enane 136-Fe11 (Scheme 38) based on a tetrahedral-metal/ organic cage complex with six macrocycles threaded around each of the cage vertices.[124]

4. Higher Order Entwined [n]Catenanes

The introduction of two additional crossings to a Hopf link (22 link) generates an inherently chiral doubly interlocked [2]catenane known colloquially as a "Solomon knot". However, as this topology is not a knot but a link (a 42 link in Alexander-Briggs notation), it is termed a "Solomon link" by chemists. Adding a further two crossings can generate a triply interlocked or "Star of David" [2]catenane (62 link). Cate-nanes in which the component rings are entwined about each other multiple times represent a significant challenge to synthetic chemists, with few examples of their successful synthesis reported to date.

Scheme 36. Beer's sulfate anion template synthesis of a catenane cage molecule.[120]

Scheme 37. Cooper's formation of discrete interlocked cages mediated by reversible imine bond formation.'1211 Non-interlocked cage monomers 133a-c are formed in the absence of CF3CO2H.[122]

4.1. Solomon Links

The Sauvage group carried out the first syntheses of Solomon links, through the macrocyclization of trimetallic linear double-stranded helicates (Scheme 39).[125] The cycli-zation of the entwined and threaded complex 137-Cu1 through

Scheme38. Nitschke and Sanders' [7]catenane based on the dynamic threading of macrocycles around each of the cage vertices.[124]

Williamson ether synthesis generated Solomon link 138 in 2 % overall yield after a demetalation step (Scheme 39 a). A more efficient synthesis involved the lithium helicate 139-Li1 which was cyclized by double ring-closing olefin metathesis to give the corresponding Solomon link 140-Cu1 in 30% yield after metal exchange (Scheme 39b).[126]

Although longer linear helicates could theoretically be used to increase the number of crossings, higher order molecular knots and links have yet to be accessed by this strategy. The problem is that as the linear helicate gets longer, the distance between the ends that have to be connected increases and rapidly becomes synthetically impractical.'1271

Scheme39. Sauvage's use of trimetallic double-stranded linear helicates to generate a Solomon link in a) a single Williamson ether macro-cyclization[125] and b) double macrocyclization by RCM.[126]

Circular helicates offer the advantage of shorter distances for cyclization, at the expense of a greater number of connections that must take place. Circular helicates have been successfully used to make higher order knots and links, including a Solomon link (Scheme 40).

Bisaldehyde 141 reacts with amines to form tris(biden-tate) ligand strands reminiscent of the tris(bipyridine) ligands found by Lehn and co-workers to form tetrameric circular helicates with transition-metal ions.[128] Solomon link 143-Fen is formed in 75 % yield by heating an equimolar mixture of dialdehyde 141, diamine 142, and FeCl2 (Scheme 40).[129] The in situ generation of the tris(bidentate) ligands allows for the self-assembly of a tetrameric circular helicate on coordination to FeII, within which parallel ligand strands are covalently connected through the newly formed diimine linkages. The effectiveness of the approach is attributed to the combination of dynamic and reversible imine bonds and metal-ligand coordination correcting any "mistakes" in connectivity, and gauche effects within the glycol linkers allowing the low energy turns necessary to cyclize the structure.

As with the Hopf link [2]catenanes, Solomon links have been synthesized which contain metal cations as an integral part of the molecular backbone. Puddephatt and co-workers synthesized Solomon link 145-Au1 from dialkynylgold polymer 144-Au1 on introduction of diphosphine ligands to compete for metal coordination (Scheme 41).[130] Aurophilic interactions between the gold cations in each component ring of the catenane drive the formation of the Solomon link 145-Au1.

Severin and co-workers recently reported the formation of a large Solomon link using CuI to template the assembly of pyridine-derivatized bipyridine ligands connected by PtII coordination (Scheme 42).[131] When ligand 146 was combined with [Pt(dppp)(OTf)2] (dppp = 1,3-bis(diphenylphosphino)-propane) and [Cu(MeCN)4]BF4 in a 2:2:1 stoichiometry, Solomon link 147-PtIICuI is formed. Twelve Pt11 connectors generate two hexanuclear bowed macrocycles, while six CuI-bipyridine complexes generate the interwoven topology. Other examples of Solomon links include systems closely related to Stoddart's Borromean rings,[132] and a recent report suggesting a Solomon link is formed in a dynamic combinatorial library.[133] A related "Solomon cube" topology has been reported by Hardie and co-workers.[134]

4.2. A Star of David [2]Catenane

In its simplest representation, the Star of David topology (612 link) consists of two rings entwined about each other three times, thereby creating six alternating crossings. In molecular terms it can be considered a triply interlocked [2]catenane. Like the Solomon link (4^ link), the topology is intrinsically chiral. The synthesis of a Star of David [2]catenane was reported by connecting the end groups of a hexameric circular helicate by RCM (Scheme 43).[135] The X-ray crystal structure confirms the structure and topology. Key to the synthesis is the use of the ortho-substituted phenyl group in the alkene linker. The twist of the phenyl group restricts the conforma-

143-Fe"

Scheme 40. The one-pot synthesis of a Solomon link based on a tetrameric circular helicate.[129]

tions of the linker directing formation of the correct connectivity. Demetalation of 150-Fe11 afforded the wholly organic Star of David '2]catenane.

4.3. Borromean Rings

Molecular Borromean rings, the simplest type of Brun-nian link, are '3]catenane topoisomers in which none of the

145-Au'

145-Au'

Scheme 41. Puddephatt's Solomon link generated from the rearrangement of a dialkyne-Au1 polymer on introduction of coordinating diphosphine ligands. Phosphophenyl groups are omitted from the crystal structure for clarity. Au-Au distances; 3.130(2) and 3.239(2) A.[130]

component rings are linked, but also cannot be separated without breaking one of the rings.[136]

The Stoddart group synthesized molecular Borromean rings 153-Zn11 through the one-pot reaction of six dialdehydes (151), six diamines (152), and six Zn11 ions (Scheme 44).[137] The Borromean ring topology was confirmed by X-ray crystallography, which also revealed p-stacking interactions between the phenoxy and bipyridyl rings that likely aid the assembly process. The ZnII templates could be removed from the structure by reduction of the imine bonds with NaBH4, followed by treatment with ethylenediaminetetraacetic acid (EDTA) to afford the demetalated Borromean rings.[138] The Borromean rings were shown to self-assemble from the same ligand set with other metal templates'1391 such as Cu11, Co11, MnII, and CdII, whilst employing equimolar amounts of Zn(OAc) 2 and Cu(OAc)2 in an attempt to form heterometal-lic Borromean rings resulted instead in a heterometallic Solomon link.[132a]

6x [Q(MeCN)4]BF4

Scheme 42. Severin's Solomon link containing a 2:1 mixture of square-planar and tetrahedral coordinated metal cations. Dppp ligands on the Pt11 metal centers are omitted from the crystal structure for clarity.[131]

Molecular Borromean rings were also discovered by chance by Jin and co-workers during their investigations of the synthesis of Cp*Rh (Cp* = h5-C5Me5) metallarectangles

(Scheme 45).[140] Ligands 154a, 154b, or 154c were coordinated to [(RuCp*Cl2)2]. Chloride abstraction with AgOTf, followed by addition of 156-Cu11 affords Borromean rings 157-RunCun(a-c). The relative lengths of the ligands was found to be crucial to the formation of Borromean rings; too long or too short, or with a non-optimal ratio of lengths, gave non-interlocked metallarectangles instead.

5. Catenanes as Switches, Rotary Motors, and Sensors

The interlocked architecture of catenanes can be exploited for function in several different ways. The dynamics of the large amplitude motions that their components can undergo make catenanes attractive candidates for utilization in molecular machines.[141] The change of the relative positioning of the rings can be used as a switch, or the net directionality of 360° rotation of one ring with respect to each other can be used as the basis for a rotary motor.[142] The cavity formed by interlocked rings can be used to hold functional groups in precise positions in 3D space, a property that can be used to bind substrates with exquisite specificity.

5.1. Catenane Switches

Sauvage and co-workers demonstrated both electrochemical and photochemical control over ring motions in hetero-[2]catenate 158 (Scheme 46 a).[143] Although many metal-

149-Fe"

Scheme 43. A Star of David [2]catenane synthesized by RCM of a hexameric circular helicate scaffold. A PF6~ ion occupies the central cavity of the helicate in the X-ray crystal structure of 150-Fe", oriented to direct the fluorine atoms towards the twelve electron-poor protons that line the walls of the cavity (CH-F distances 1.88-2.43 A).[135]

Scheme 44. Stoddart's Borromean rings, with the carbon atoms in each ring of the interlocked topology highlighted in different colors (light blue, orange, and gray). Additional anions occupying the sixth coordination site in the Zn11 cations have been omitted for clarity. The Zn11 centers are all crystallographically equivalent, with distorted octahedral geometries (cis-N-Zn"-N bond angles 72.5(2)-109.5(3)8). Extensive p stacking occurs between the phenyl rings and bipyridine groups (distances: 3.51 and 3.72A).[137]

ligand interactions can be rather kinetically stable, the difference in the preferred coordination geometries of different oxidation states can be exploited to bring about config-urational switching in [2]catenanes. Oxidation of the metal center in dpp,dpp-[158-Cu'] (by either chemical or electrochemical means) reverses the order of preference for coordination numbers (the preferred order for Cu11 is 6 > 5 > 4). Thus, a configurational change to the five-coordinate state obtained in dpp,terpy-[158-Cun] is observed. The kinetics of intercomponent motion in the switching process are relatively slow due to the generation of a metastable dpp,dpp-[158-Cun] state following oxidation. Whilst the process can be reversed on reduction to regenerate

Scheme 45. Jin's Borromean rings comprised of RhCp* metallarectan-gles. The Cp* ligands on the Rh metal centers are omitted from the crystal structure for clarity.[140]

dpp,dpp-[158-Cu'], the formation of a metastable dpp,terpy-[158-Cu1] state makes the switching slow once again.

The related homo-[2]catenate 159 (Scheme 46b), in which each ring contains a bidentate dpp unit and a tridentate terpy site, exhibits more complicated behavior.[144] In dpp,dpp-[159-Cu1] the copper ion coordinates to the two dpp units in the usual tetrahedral arrangement. Following oxidation of the metal cation, circumrotation of the rings proceeds to give the preferred hexacoordinated species terpy,terpy-[159-Cun]. It was demonstrated that this process occurs by the revolution of one ring with respect to the other to give an intermediate five-coordinate species (terpy,dpp-[159-Cun]).[144] In comparison to many related coordination complexes, the process is relatively fast, with the ligand rearrangement occurring on the timescale of tens of seconds. The process is completely reversible, via the same five-coordinate geometry, on reduction to Cu1 (that is, via dpp,terpy-[159-Cu']).

The Stoddart group demonstrated the co-conformational switching of [2]catenane 160 containing an electron-poor cyclobisparaquat(p-phenylene) (CBPQT4+) macrocycle, and an electron-rich crown ether macrocycle (Scheme 47).[145] In the initial 1604+ state, the tetrathiafulvalene (TTF) unit in the crown ether macrocycle is the more electron-rich motif, and is predominantly situated within the cavity of the tetracationic cyclophane. However, following TTF oxidation, either through chemical or electrochemical means, the dioxynaph-thalene unit becomes the more electron-rich motif and the structure switches to accommodate that unit within the crown

dpp,terpy-[158-Cu"] terpy,dpp-[159-Culj] terpy,terpy-[159-Cu"]

Scheme 46. a) Sauvage's switchable heterocatenate 158.[143] b) Oxidation-state-controlled switching of [2]catenate 159 between three distinct co-

conformations.[144]

Scheme 47. Stoddart's co-conformational switching of a [2]catenane nating between electron-rich/poor aromatic stacking interactions and

mediated by the relative strengths of intermacrocycle electron-rich/ radical-pairing interactions as dominant intermacrocycle forces, after

poor aromatic stacking interactions.[145] bipyridinium reduction and oxidation, respectively.[146]

Scheme 49. Sauvage's pH-induced switching between co-conformations following demetalation of a Cu'-template-directed [2]catenane.[147]

ether cavity (1605+/1606+). The process can be reversed through reduction of the TTF back to its neutral state.

A related [2]catenane features co-conformational switching mediated by electrostatic я-я interactions and radical pairing as the dominant intermacrocycle forces (Scheme 48).[146] The CBPQT4+ macrocycle resides over a dioxynaphthalene unit in catenane 1616+. On reduction of bipyridinium (bipy2+) units to bipy+', radical pairing interactions change the predominant structure to catenane 1613(+0. Oxidation regenerates 1616+ and reverses the switching process.

The pH-induced switching of a [2]catenane between two well-defined co-conformations has been demonstrated by Sauvage and co-workers (Scheme 49).[147] Reversible switching of the catenanes co-conformations between 162 and 162-H+ is achieved by protonation and deprotonation of the catenane respectively.

A combination of metal coordination/demetalation and hydrogen bonding results in reversible switching of [2]cate-nane 163-Pdn (Scheme 50).[148] In both 163-Pd" and 163-H2, the pyridine groups of each macrocycle are held in proximity through their involvement in metal coordination and hydrogen bonding, respectively. However, addition of [PdCl2-(CH3CN)2] to 163-H2 generates 163-H2-PdCl2(CH3CN) which has a preferred co-conformation in which one ring is rotated 180° with respect to the other. The switching process can be reversed through deprotonation of the amide nitrogen atoms with NaH, or by demetalation and reintroduction of a basic PdII salt.

163-H2-PdCI2(CH3CN)

Scheme 50. A three-state switching process between two co-conformations for a [2]catenane utilizing variable coordination modes of suitable Pd"

cations.[148]

Angew. Chem. Int. Ed. 2015, 54, 6110-6151 © 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.angewandte.org 6137

These are not the final page numbers! **

binding affinity order of A > B > C, and returns the benzylic amide macrocycle to its original site.

Switching in other catenanes has been achieved using anion binding[149] and photochemical stimuli.[150]

5.2. Catenanes as Rotary Motors

Introducing kinetic barriers to restrict the pathways of Brownian motion available to a macrocycle in a catenane can be used to produce net-directional rotary motion and a molecular motor. [3]catenane 165 (Scheme 52a)[102a] differs from [2]catenane 164 (Scheme 51 a) only by the addition of a further benzylic amide macrocycle, and it features the same tunable station-binding affinities to control the position of the

Scheme 51. A [2]catenane in which modifying the binding affinity of three different stations to a benzylamide macrocycle allows switching between three discrete co-conformations. a) Structure of [2]catenane 164-(E,E). b) Illustration of the switching process.'102"1

Nondirectional switching between three different co-conformations can be brought about with [2]catenane 164 (Scheme 51a).[102a] Three "stations" are incorporated into the larger macrocycle ring: A (green), B (red), and C (yellow). Each station has a different binding affinity for the smaller benzylic amide macrocycle. The affinity of fumaramide groups at A and B can be drastically reduced by photo-isomerization to the maleamide (Z) form, which occurs at different wavelengths for the two stations because of the proximity of the benzophenone group to A.

In its initial state (E,E), the binding affinity of the stations to the small macrocycle is in the order A > B > C, and thus the benzylic amide macrocycle is predominantly located over station A. Following E to Z isomerization of A, the relative affinities change to B > C > A and the small macrocycle moves predominantly to station B (State I to State II, Scheme 51 b). A further change in the position of the benzylic amide macrocycle occurs upon isomerization of station B, thereby locating the macrocycle preferentially at station C (State II to State III, Scheme 51 b), with station binding affinity now in the order C > A « B. Re-isomerizations of A and B to fumaramide groups restores the initial station

Scheme 52. A [3]catenane rotary motor. a) Structure of [3]catenane 165-(E,E).[102a] b) Illustration of the directional switching process. Conditions: i) UV (350 nm); ii) UV (254 nm); iii) heat.

6138 www.angewandte.org

** These are not the final page numbers!

smaller macrocycles. The isolated amide group (dark green) acts as a fourth binding station, D. The station binding affinities for the benzylic amide macrocycles follow the order A > B > C > D in 165-(E,E), and the two macrocycles are accordingly positioned primarily over the A and B stations.

Following photoisomerization of station A, the blue macrocycle moves from station A to C (KB > KC > KD > KA). Importantly, translocation of the ring can only occur in one direction around the larger ring (anticlockwise from Scheme 52 b) because of the presence of the second macrocycle on station B. Likewise, on E to Z isomerization of station B, translation of the purple macrocycle can only proceed in one direction (anticlockwise) to station D (KC > KD > KA « KB). Hence, the small rings move in a "follow-the-leader" process directionally around the larger ring. The entire reaction sequence must be repeated for a full directional 360° rotation of each macrocycle to occur.

Selective rotation in either direction is possible using [2]catenane 166 (Scheme 53).[102b] In this case, the route of travel of the smaller ring around the larger one is determined by the order in which blocking groups either side of the succinamide binding site are released. Starting from 166-Succ-(Z), isomerization of the maleamide station (purple) to its fumaramide form (green) makes it the preferred binding station for the benzylic amide macrocycle. Removing one of the blocking groups (tBuSiMe2 for counterclockwise rotation; CPh3 for clockwise rotation) allows for directional movement of the ring to give 166-Fum-(E). The fumaramide station can

then be isomerized to the maleamide form, thereby creating a thermodynamic driving force for movement of the macrocycle back to the succinamide station in 166-Mal-(Z). Removal of the blocking group opposite to that used in the first change (i.e. CPh3 for counterclockwise rotation; tBu-SiMe2 for clockwise rotation) results in a full 360° directional rotation of the benzylic amide macrocycle around the large macrocycle.

5.3. Catenanes as Selective Hosts for Anions and Cations

As is apparent from this Review, the driving force for catenane formation is often the attractive noncovalent interactions between the components and/or a template. These interactions often "live on" after the interlocked structure is assembled, and any cavity in the 3D space formed by such a catenane then has functional groups in precise positions that are often well-matched to bind substrates with high specificity. These features have proved particularly useful in the case of catenanes synthesized with anion templates, with the resulting catenanes in some cases being able to selectively bind particular anions.[151] This, in turn, has led to catenanes being modified to signal binding events photo- or electrochemically.

Beer and co-workers used ferrocene-derivatized [2]cate-nane 167 (Scheme 54) to selectively complex chloride anions over oxoanions and electrochemically report the binding.[152]

Scheme 53. Cleavage and switching sequences in rotary motor [2]catenane 166, which is capable of directional and reversible circumrotation.

[102b]

Scheme 54. Beer's incorporation of a ferrocene motif into a catenane allows electrochemical sensing of anion binding in 167, which was found to selectively bind chloride anions. The corresponding non-interlocked macrocycles of 167 were found to display binding selectivity dictated by the basicity of the anions (BzO" > H2PO4" > Cl" > HSO,f ).[152]

Scheme 55. Beer's selective binding of chloride and bromide anions detected by optical sensing of their influence on the naphthalene emission spectra.[80]

Scheme 56. Yashima's cation-sensing catenane which undergoes a switch from a "locked" state to one in which macrocycles can freely rotate.11541

(309 nm) concomitant with the appearance of a new band (445 nm). No changes in the emission spectrum were detected with a range of alternative anions. Jeong and co-workers have also used fluorescence to measure selective chloride binding

using a [2]catenane containing indolocarbazole fluoro-phores.[153]

[2]Catenane 168 has been employed by Yashima and co-workers[154] to detect cation binding through changes in circular dichroism (CD; Scheme 56). The chiral R-phenyl-ethyl substituents on the amidinium side of the salt bridge induce twisting of the m-terphenyl motifs in both macrocycles of the catenane, thereby creating distinct Cotton effects in the CD spectrum. On the addition of Zn(ClO4)2, the salt bridge is disrupted through Zn11 coordination, which results in significant changes in the CD spectrum.

The binding of chloride facilitates the oxidation of the ferrocene to ferrocinium and can be detected by cyclic voltammetry. The X-ray structure of 167-C1" suggests that the selectivity in anion binding is probably due to the catenane cavity being too small to accommodate larger anions.

An optical response occurs upon anion binding to [2]catenane 60 (Scheme 55, the synthesis is outlined in Section 2.1.4).[80] Chloride or bromide anions significantly decrease the intensity of the naphthalene emission band

6. Catenane Linkages Incorporated into Polymer Chains, Materials, and Attached to Surfaces

Given the high cost of introducing new chemical building blocks, it seems likely that future generations of commercial polymers will be derived from assembling existing cheap monomers in new ways. Incorporating catenanes into a polymer backbone offers the possibility of introducing flexible, mobile, or switchable linkages which could influence the

rheological,[155] mechanical,[156] and dynamic[157] properties of

the resulting material. Harnessing the coordinated motion of the components of catenanes in ordered arrays could be useful for solid-state devices, and could arise from embedding catenanes onto or within a support material, such as a metal-organic framework (MOF), or on a functionalized surface.

6.1. [2]Catenanes in Polymers

6.1.1. [2]Catenanes as Main-Chain Linkages

Main-chain poly[2]catenanes consist of [2]catenanes connected in a linear manner by covalent bonds (Figure 9). Polycondensation of catenate 169-Cu1, or the corresponding

Figure 9. A main-chain poly[2]catenane.

metal-free catenand 169, with dicarboxylic acid 170 afforded polymers 171-CuI and 171,[158] although the degree of polymerization in these systems is probably modest (Scheme 57).[159] Related poly[2]catenanes based on Cu'-dpp [2]catenane motifs have also been reported by Shimada et al.[160]

To probe the influence of mechanical links in polymer chains, benzylic amide [2]catenanes have been used to dope commercial polymers (polycarbonate and poly(ethylene tere-

Scheme 57. Geert and Sauvage's synthesis of main-chain poly[2]cate-nanes incorporating dpp-Cu1 catenanes or their demetalated ana-logues.[158,159]

phthalate)).[161] The phenol groups of catenane 172 provide sites for transesterification into the polymers, whilst methyl-ation of the amide nitrogen atoms removes the possibility of intercomponent hydrogen bonding and allows relatively free movement of the rings for amplification of any physical properties endowed by the mechanical bond (Scheme 53). Solid-state polymerization of catenane 172 and bisphenol A polycarbonate oligomers (Scheme 58) was conducted to

Solid-State Polymerization

N Me ' Q

Scheme 58. The solid-state polymerization of [2]catenane monomer 172 and bisphenol A polycarbonate oligomers to generate a main-chain poly[2]catenane copolymer.[161a]

generate copolymers consisting of 10,20, or 30 wt % catenane linkages.[161a] Homogeneous dispersion of the [2]catenane was observed at low levels of incorporation (10wt%), but heterogeneity probably occurs at higher levels of incorporation. Incorporating the catenane links had a modest effect on the glass transition temperature (Tg) of the resulting polymer. Hagiwara et al. have utilized alternatively function-alized derivatives of catenane 172 for polymerization with rigid dialkyne complexes through Sonogashira cross cou-pling[162] and CuAAC "click" reactions.[163] Other examples of poly[2]catenanes have also been prepared.[164]

6.1.2. Catenanes as Polymer Side Chains

Unlike main-chain poly[2]catenanes, creating polymers with catenane side chains requires functionalization of only a single macrocycle within the catenane. As the catenane's mechanical bond no longer constitutes part of the polymer backbone (Figure 10), it is expected to induce different

Figure 10. Side chain poly[2]catenanes in which (A) a single macrocycle forms part of the polymer main chain or (B) the entire catenane is a pendent group.

properties in the resulting material. Polymers with catenane side chains have been synthesized by polymerization across a single macrocycle of the catenane,[165] or grafting of catenanes onto a preformed polymer.[166]

Side-chain poly[2]catenane 176 was prepared by treating diol catenane 174 with bis(4-isocyanatophenyl)methane (175, Scheme 59).[165a] A pendent poly[2]catenane was prepared by Simone and Swager[165b] by electrochemical polymerization.

Scheme 59. Stoddart's side-chain poly[2]catenane in which a single macrocycle of the [2]catenane constitutes part of the polymer backbone.^

6.1.3. Catenanes Incorporated into Metal-Organic Frameworks

The highly ordered arrangement of molecules in crystalline solids, combined with the space available for component movements, has led to metal-organic frameworks (MOFs) being investigated as scaffolds for interlocked molecules. Whilst rotaxanes have been incorporated into different types of MOFs,[167] catenanes present a significant challenge because of their size and shape. Nevertheless, MOFs containing [2]catenanes have been described by Stoddart, Yaghi, and co-workers.[168] The catenane-bearing molecular strut 180 forms a 2D array after heating with Cu(NO3)22.5H2O (Scheme 61a). Within the MOF, each Cu1 cation coordinates to the carboxylate groups of two struts, and to the acetylene group of an additional 180 in an h2 fashion. Accommodating such large molecules requires the network to form an alternating sequence of catenane orientations positioned above and below the plane of the 2D sheet, with further long-range-ordering observed between the stacked arrays. A more regular 2D MOF was formed using a related catenane incorporated into a longer coordinating strut (Scheme 61 b),[169] whilst a related 3D MOF has also been

reported.[170]

6.2. Catenanes Attached to Surfaces

Attaching catenanes to surfaces is an alternative approach to obtaining ordered arrays of interlocked molecules. Cate-nanes can be fastened to surfaces in different ways: 1) attached through a tether (Figure 11, type A); 2) chemisorbed

An alternative strategy to pendent poly[2]catenanes was demonstrated by Bria et al.[166] Functionalization of [2]cate-nane monomer 177 (Scheme 60) with a single alkyne group allows its incorporation into copolymer 178 through CuAAC "click" reactions to give 179.

Scheme 60. Bria's side-chain [2]catenane polymer.[1

Figure 11. Three different forms of [2]catenanes attached to surfaces.

such that the surface forms an intrinsic part of one ring (Figure 11, type B); or 3) physisorbed (Figure 11, type C).[171]

The process of tethering a [2]catenane to a gold surface is illustrated with dithiol [2]catenane 184 (Scheme 62).[172] X-Ray photoelectron spectroscopy (XPS) indicates that 184 initially forms a monolayer in which the catenanes are arranged perpendicular to the surface, chemisorbed through only one of the two thiol groups. Given enough time, however, (14 h in the case of 184) the catenane finds a lower energy co-conformation in which both thiol groups bind to the gold, thereby holding the catenanes parallel to the surface.

Beer's chloride-binding catenane with an electrochemi-cally active ferrocene reporting group (see Section 5.3) has also been adapted for incorporation onto a surface.[152]

6142 www.angewandte.org

** These are not the final page numbers!

Scheme 61. Stoddart and Yaghi's synthesis of metal-organic frameworks containing [2]catenane organic struts.'168,169' The interlocked macrocycles which do not form part of the MOF backbone have been colored blue, and represented as space-filled structures for clarity.

Exposure of the dithiol pseudorotaxane complex 187 (Scheme 63) to a Au(111) surface led to formation of a self-assembled monolayer (188) after 8 h that is effectively a [2]catenane in which the Au(111) surface forms part of one ring. This type of strategy for forming [2]catenanes at a surface was originally introduced by Gokel, Kaifer, and coworkers.^7^ Other catenane monolayers have been described by the Sauvage group.[174]

Co-conformational switching has also been demonstrated in a mixed Langmuir monolayer of [2]catenane 160 and phospholipid dimyristoylphosphatidic acid (DMPA) sandwiched between an Si electrode and a Ti/Al electrode (Scheme 64, see also Section 5.1.1).[175] Application of a — 2V or + 2V bias to the system results in oxidation of 1604+ or partial reduction of 1605+, respectively. A significant change in the junction resistance was observed on cycling between 1604+ and 160®+.

Co-conformational switching can occur with catenanes tethered to a surface. Stoddart and co-workers[176] demonstrated that the TTF-based catenane 1894+ (Scheme 65) still undergoes switching when chemisorbed on metal nanoparti-cles (Au, Pd, or Pt). Reversible oxidation and reduction of the surface-grafted catenanes between 1894+-NP and 1896+-NP was achieved with Fe(ClO4)3 and ascorbic acid, respectively.

Benzylic amide catenanes have been extensively studied as chemisorbed and physisorbed thin films on various

surfaces. Well-ordered films of catenanes such as 57 can be grown by sublimation under ultrahigh vacuum, and catenanes of the first monolayer were chemisorbed on a Au surface through some of their carbonyl groups. Their electronic and vibrational structures have been characterized by XPS and high-resolution electron energy loss spectroscopy (HREELS) as well as by infrared reflection absorption spectroscopy.[177] Physisorbed surface catenanes have also been demonstrated

by the Sauvage group.[178]

Despite these successes, introducing catenanes into polymers or assembling them into ordered arrays on surfaces or within crystalline solids remains a very significant challenge.

Exploiting the dynamic properties of catenanes-inherent

flexibility, rotation, or switching of the relative positions of

the rings-in such environments has yet to be achieved in

any genuinely useful manner.

7. Conclusions and Outlook

The End of an Era....

The template synthesis of catenanes is unquestionably one of the great triumphs of synthetic supramolecular chemistry. It provides the means through which chemists can today routinely make Hopf link catenanes in one step, in high yields,

Scheme 63. Beer's chloride-containing pseudorotaxane which forms a [2]catenane with a Au(111) surface.'1521

Scheme 62. A thiol-functionalized [2]catenane tethered to a Au(111) surface.'1721

from readily available building blocks. Such a situation was simply unimaginable fifty years ago in an era when there was little or no understanding of template synthesis, host-guest chemistry, self-assembly processes, or the utilization of non-covalent interactions, nor any of the highly effective modern tools for covalent-bond formation. Indeed, the success of interlocked molecule synthesis is testament to just how dramatically chemistry can progress. The brilliance of Schill and Luttringhaus demonstrated how chemists could design synthetic pathways that feature spatial awareness in all three dimensions, a necessity to bring about the controlled crossing of molecular strands and/or the threading of an organic chain through a macrocycle. Sauvage's great contribution was to realize that the synthesis of such structures could be greatly facilitated by using the preferred coordination geometry of transition-metal ions to assemble and orientate the molecular building blocks, whilst Stoddart's (as well as pioneering much of the synthetic chemistry on catenanes and rotaxanes in the 1990s) was to recognize the potential of mechanically bonded molecules as architectures for molecular machines. In articulating that vision, he provided an incentive for a generation

Scheme 64. Stoddart's switchable [2]catenane layered between electrodes.'1751

of chemists to use their imagination and creativity to come up with ever more inventive and successful routes to interlocked molecules.

The resulting accomplishments—in template syntheses that exploit metal coordination, hydrogen bonding, aromatic stacking interactions, the hydrophobic effect, halogen bonding, and ever-improving methods of covalent capture such as active template synthesis, dynamic covalent chemistry, olefin metathesis, and "click" chemistries—-mean that in 2015 the problem of how to form Hopf link [2]catenanes is effectively solved.

6144 www.angewandte.org

** These are not the final page numbers!

Metal Nanoparticle

Scheme 65. Stoddart's switchable [2]catenane tethered to a metal nanoparticle. TOAB = tetraoctylammonium bromide.'1761

....and the Start of a New One!

The rest of the field of chemical topology remains almost completely virgin territory. The synthesis of catenanes is straightforward only for the simplest Hopf link topology. Only a handful of the other links listed in Figure 2 (and there are countless other possibilities)'211 have been prepared thus far (Solomon links, Borromean rings, and one Star of David catenane)[179] and the dynamics of the components in those topologies, and how to control or exploit them, remain unknown. Whilst it is often suggested that given enough time and resources today's chemists can make almost any molecule,'180 that is simply not true of topologically complex structures. No general method yet exists to make catenanes more complicated than a Hopf link, nor is there any strategy for controlling stereochemistry (over or under) at crossing points (such control is necessary to make all but the most trivial links and knots), nor methods to effectively incorporate catenanes into materials (polymers, MOFs, or at surfaces and interfaces) and exploit their interlocked architectures and

dynamics. These are major unsolved problems-in synthesis,

design, and application-that currently preclude the exploration and exploitation of a vast region of chemical space. Indeed, many of the unmet challenges in chemical topology today seem even more perplexing than the problems faced by the pioneers of catenane chemistry fifty years ago!

Biology shows us time and again that the organization of matter with topological complexity at the molecular level offers tremendous potential for achieving novel molecular and material properties (strength, lightweight, flexibility, dynamics). Chemists can uncover the ways and means to tap into this hitherto unrealized wealth of form and function if they can make inroads into what is, perhaps, one of the last great unexplored frontiers of synthetic chemistry. Exactly what the next half-century of catenane research will uncover is impossible to foresee, but surely it will be as rich, diverse, and full of extraordinary surprises as the first fifty years.

We thank the EPSRC for supporting our research program on molecular knots and links and Steffen L. Woltering and Alina L. Nussbaumer for useful discussions and the German translation of the Review.

[1] a) E. Wasserman, J. Am. Chem. Soc. 1960,82,4433 - 4434; b) E. Wasserman, Sci. Am. 1962, 207, 94-102.

[2] H. L. Frisch, E. Wasserman, J. Am. Chem. Soc. 1961, 83, 37893795.

[3] The cyclic nature of many of the first macrocycles to be synthesized went unappreciated for decades. In an early example, Posner condensed 2-aminobenzaldehyde in the pres-

ence of Zn11 in 1898 [a) T. Posner, Ber. Dtsch. Chem. Ges. 1898, 31, 656 - 660], but the isolated product was not recognized to be a macrocycle for another 30 years [b) F. Seidel, Ber. Dtsch. Chem. Ges. B 1926, 59, 1894-1908].

[4] a) L. Ruzicka, M. Stoll, H. Schinz, Helv. Chim. Acta 1926, 9, 249-264; b) L. Ruzicka, M. Hurbin, M. Furter, Helv. Chim. Acta 1934,17, 78-87.

[5] a) K. Ziegler, H. Eberle, H. Ohlinger, Justus Liebigs Ann. Chem. 1933,504,94 -130 ; b) K. Ziegler, A. Lüttringhaus, Justus Liebigs Ann. Chem. 1934, 511, 1 -12; c) K. Ziegler, K. Weber, Justus Liebigs Ann. Chem. 1934, 512,164-171; d) A. Lüttringhaus, K. Ziegler, Liebigs Ann. Chem. 1937, 528, 155-161.

[6] V. Prelog, L. Frenkiel, M. Kobelt, P. Barman, Helv. Chim. Acta 1947, 30, 1741-1749.

[7] a) M. Stoll, A. Rouve, Helv. Chim. Acta 1947, 30, 1822-1836;

b) M. Stoll, J. Hulstkamp, Helv. Chim. Acta 1947, 30, 18151821.

[8] a) J. van Alphen, Recl. Trav. Chim. Pays-Bas 1937, 56, 343 -350; b) V. L. Hamsley, U.S. Patent 2,228,268, 1941.

[9] a) W. Pathnode, D. F. Wilcock, J. Am. Chem. Soc. 1946, 68, 358-363; b) C. J. Brown, A. Hill, P. V. Youle, Nature 1956,177, 128.

[10] H. Frisch, I. Martin, H. Mark, Monatsh. Chem. 1953, 84, 250256.

[11] A. Lüttringhaus, F. Cramer, H. Prinzbach, F. M. Henglein, Justus Liebigs Ann. Chem. 1958, 613, 185-198.

[12] In 1992, a similar approach was successfully used to form a [2]catenane by Stoddart and co-workers, albeit with different coupling chemistry and a different cyclodextrin derivative (see Section 2.1.5). D. Armspach, P. R. Ashton, C. P. Moore, N. Spencer, J. F. Stoddart, T. J. Wear, D. J. Williams, Angew. Chem. Int. Ed. Engl. 1993,32, 854 - 858; Angew. Chem. 1993,105, 944948.

[13] a) D.M. Walba, Tetrahedron 1985, 41, 3161-3212; b) D. B. Amabilino, J. F. Stoddart, Chem. Rev. 1995, 95, 2725 - 2829;

c) Molecular Catenanes, Rotaxanes and Knots (Eds.: J.-P. Sauvage, C. O. Dietrich-Buchecker), Wiley-VCH, Weinheim, 1999; d) G. A. Breault, C. A. Hunter, P. C. Mayers, Tetrahedron 1999, 55, 5265 - 5293; e) N. H. Evans, P.D. Beer, Chem. Soc. Rev. 2014, 43, 4658-4683.

[14] G. Schill, A. Lüttringhaus, Angew. Chem. Int. Ed. Engl. 1964, 3, 546-547; Angew. Chem. 1964, 76, 567 - 568.

[15] a) G. Schill, C. Zürcher, Angew. Chem. Int. Ed. Engl. 1969, 8, 988-988; Angew. Chem. 1969, 81, 996-997; b) G. Schill, E. Logemam, W. Vetter, Angew. Chem. Int. Ed. Engl. 1972, 11, 1089-1090; Angew. Chem. 1972,84,1144-1145; c) G. Schill, C. Zürcher, Chem. Ber. 1977,110, 2046-2066.

[16] a) G. Schill, K. Murjahn, Liebigs Ann. Chem. 1970, 740,18 - 30; b) G. Schill, K. Risler, H. Fritz, W. Vetter, Angew. Chem. Int. Ed. Engl. 1981,20,187-189; Angew. Chem. 1981, 93,197 - 201.

[17] N. van Gulick, New J. Chem. 1993,17, 619-625.

[18] a) R. Wolovsky, J. Am. Chem. Soc. 1970, 92, 2132 - 2133; b) D. A. Ben-Efraim, C. Batich, E. Wasserman, J. Am. Chem. Soc. 1970, 92, 2133 - 2135.

[19] a) D. M. Walba, R. M. Richards, R. C. Haltiwanger, J. Am. Chem. Soc. 1982, 104, 3219-3221; b) D. M. Walba, J. D. Armstrong, A. E. Perry, R. M. Richards, T. C. Homan, R. C. Haltiwanger, Tetrahedron 1986, 42, 1883-1894.

[20] Lord Kelvin and many other scientists in the 19th century supported the theory of "vortex atoms", a hypothesis that the chemical elements were knotted aether. Mathematicians began listing knots in the belief that they were listing chemical elements. The model was quickly discarded following the Geiger-Marsden experiments, carried out under Rutherford's direction at the University of Manchester, which gave information about the internal structure of the atom [a) H. Geiger, E. Marsden, Proc. R. Soc. London Ser. A 1909, 82, 495 - 500].

For a historical perspective on the vortex theory of atoms, see b) A. B. Sossinsky, Knots: mathematics with a twist, Harvard University Press, Cambridge, 2002.

[21] C. C. Adams, The knot book : an elementary introduction to the mathematical theory of knots, Amer. Math. Soc., Providence, 2004.

[22] J. W. Alexander, G. B. Briggs, Ann. Math. 1926, 28, 562-586.

[23] B. Hudson, J. Vinograd, Nature 1967, 216, 647 - 652.

[24] D. A. Clayton, J. Vinograd, Nature 1967, 216, 652 - 657.

[25] F. B. Dean, A. Stasiak, T. Koller, N. R. Cozzarelli, J. Biol. Chem. 1985, 260, 4975-4983.

[26] J. J. Champoux, Annu. Rev. Biochem. 2001, 70, 369-413.

[27] a) W. R. Taylor, Comput. Biol. Chem. 2007, 31, 151-162; b) A. L. Mallam, FEBS J. 2009,276, 365 - 375 ; c) W. R. Taylor, Nature 2000, 406, 916-919.

[28] W. R. Wikoff, L. Liljas, R. L. Duda, H. Tsuruta, R. W. Hendrix, J. E. Johnson, Science 2000, 289, 2129-2133.

[29] D. R. Boutz, D. Cascio, J. Whitelegge, L. J. Perry, T. O. Yeates, J. Mol. Biol. 2007, 368, 1332-1344.

[30] Z. B. Cao, A. W. Roszak, L. J. Gourlay, J. G. Lindsay, N. W. Isaacs, Structure 2005,13, 1661-1664.

[31] B. I. Lee, K. H. Kim, S. J. Park, S. H. Eom, H. K. Song, S. W. Suh, EMBO J. 2004, 23, 2029-2038.

[32] C. M. Zimanyi, N. Ando, E. J. Brignole, F. J. Asturias, J. Stubbe, C. L. Drennan, Structure 2012, 20, 1374-1383.

[33] M. B. van Eldijk, I. van Leeuwen, V. A. Mikhailov, L. Neijen-huis, H. R. Harhangi, J. C. M. van Hest, M. S. M. Jetten, H. J. M. O. den Camp, C. V. Robinson, J. Mecinovic, Chem. Commun. 2013, 49, 7770 - 7772.

[34] N. C. Seeman, Angew. Chem. Int. Ed. 1998, 37, 3220-3238; Angew. Chem. 1998,110, 3408-3428.

[35] C. D. Mao, W. Q. Sun, N. C. Seeman, Nature 1997, 386, 137138.

[36] Y. W. Zhang, N. C. Seeman, J. Am. Chem. Soc. 1994, 116, 1661 -1669.

[37] J. H. Chen, N. C. Seeman, Nature 1991, 350, 631-633.

[38] a) J. Elbaz, Z. G. Wang, F. A. Wang, I. Willner, Angew. Chem. Int. Ed. 2012, 51, 2349-2353; Angew. Chem. 2012,124, 23992403; b) C. H. Lu, A. Cecconello, J. Elbaz, A. Credi, I. Willner, Nano Lett. 2013, 13, 2303 - 2308; c) X. J. Qi, C. H. Lu, X. Q. Liu, S. Shimron, H. H. Yang, I. Willner, Nano Lett. 2013, 13, 4920-4924.

[39] L. Z. Yan, P. E. Dawson, Angew. Chem. Int. Ed. 2001, 40,36253627; Angew. Chem. 2001,113, 3737-3739.

[40] For statistical and directed methods to interlocked structures, see a) G. Schill, Catenanes, Rotaxanes and Knots, Academic Press, New York, 1971; b) G. Schill, C. Zürcher, Naturwissenschaften 1971, 58, 40 - 45 ; for more recent examples, see c) A. Godt, Eur. J. Org. Chem. 2004, 1639-1654; d) T. Umehara, H. Kawai, K. Fujiwara, T. Suzuki, J. Am. Chem. Soc. 2008, 130, 13981-13988.

[41] a) J.-P. Sauvage, Acc. Chem. Res. 1990, 23, 319-327; b) C. O. Dietrich-Buchecker, J.-P. Sauvage, Chem. Rev. 1987, 87, 795810; c) J.-C. Chambron, C. Dietrich-Buchecker, C. Hemmert, A. K. Khemiss, D. Mitchell, J.-P. Sauvage, J. Weiss, Pure Appl. Chem. 1990, 62, 1027-1034; d) C. Dietrich-Buchecker, J.-P. Sauvage, Tetrahedron 1990, 46, 503 -512; e) J.-C. Chambron, C. O. Dietrich-Buchecker, J.-F. Nierengarten, J.-P. Sauvage, Pure Appl. Chem. 1994, 66, 1543-1550; f) J.-C. Chambron, C. Dietrich-Buchecker, J.-P. Sauvage in Comprehensive Supra-molecular Chemistry, Vol. 9 (Eds. : J.-P. Sauvage, M. W. Hos-seini), Elsevier, Oxford, 1996, pp. 43-83.

[42] For general reviews on the use of templates to form interlocked structures, see a) R. Hoss, F. Vögtle, Angew. Chem. Int. Ed. Engl. 1994, 33, 375-384; Angew. Chem. 1994, 106, 389-398; b) M. Fujita in Comprehensive Supramolecular Chemistry, Vol. 9 (Eds. : J.-P. Sauvage, M. W. Hosseini), Elsevier, Oxford,

1996, pp. 253 - 282; c) M.-J. Blanco, M. C. Jimenez, J.-C. Chambron, V. Heitz, M. Linke, J.-P. Sauvage, Chem. Soc. Rev.

1999, 28, 293 - 305; d) D. A. Leigh, A. Murphy, Chem. Ind. 1999,178-183; e) T. J. Hubin, D. H. Busch, Coord. Chem. Rev.

2000, 200, 5-52; f)J.-P. Collin, C. Dietrich-Buchecker, C. Hamann, D. Jouvenot, J.-M. Kern, P. Mobian, J.-P. Sauvage in Comprehensive Coordination Chemistry II, Vol. 7, (Eds.: J. A. McCleverty and T. J. Meyer), Elsevier, Oxford, 2004, pp. 303326; g) F. Aric, J. D. Badjic, S. J. Cantrill, A. H. Flood, K. C. F. Leung, Y. Liu, J. F. Stoddart, Top. Curr. Chem. 2005,249, 203259; h) P. Gaviña, S. Tatay, Curr. Org. Synth. 2010, 7, 24-43; i) J. E. Beves, B. A. Blight, C. J. Campbell, D. A. Leigh, R. T. McBurney, Angew. Chem. Int. Ed. 2011,50,9260-9327; Angew. Chem. 2011,123, 9428-9499.

[43] a) C. O. Dietrich-Buchecker, J.-P. Sauvage, J.-P. Kintzinger, Tetrahedron Lett. 1983, 24, 5095-5098; b) C. O. DietrichBuchecker, J.-P. Sauvage, J. M. Kern, J. Am. Chem. Soc. 1984, 106, 3043 - 3045.

[44] a) B. Mohr, M. Weck, J.-P. Sauvage, R. H. Grubbs, Angew. Chem. Int. Ed. Engl. 1997,36,1308-1310; Angew. Chem. 1997, 109,1365 -1367; b) for the use of reversible olefin metathesis to assemble catenanes under thermodynamic control, see T. J. Kidd, D. A. Leigh, A. J. Wilson, J. Am. Chem. Soc. 1999, 121, 1599-1600.

[45] M. Cesario, C. O. Dietrich-Buchecker, J. Guilhem, C. Pascard, J.-P. Sauvage, J. Chem. Soc. Chem. Commun. 1985, 244-247.

[46] a) J. C. Loren, P. Gantzel, A. Linden, J. S. Siegel, Org. Biomol. Chem. 2005, 3, 3105-3116; b) P. Mobian, J.-M. Kern, J.-P. Sauvage, J. Am. Chem. Soc. 2003, 125, 2016-2017; c)J.-P. Sauvage, M. Ward, Inorg. Chem. 1991,30, 3869-3874; d) D. A. Leigh, P. J. Lusby, S. J. Teat, A. J. Wilson, J. K. Y. Wong, Angew. Chem. Int. Ed. 2001, 40, 1538-1543; Angew. Chem. 2001,113, 1586-1591; e) D. A. Leigh, P.J. Lusby, R. T. McBurney, A. Morelli, A. M. Z. Slawin, A. R. Thomson, D. B. Walker, J. Am. Chem. Soc. 2009,131, 3762-3771; f) F. Aricó, P. Mobian, J.-M. Kern, J.-P. Sauvage, Org. Lett. 2003, 5, 1887-1890; g) P. Mobian, J.-M. Kern, J.-P. Sauvage, Inorg. Chem. 2003, 42, 8633 - 8637.

[47] a) G. Kaiser, T. Jarrosson, S. Otto, Y. F. Ng, A. D. Bond, J. K. M. Sanders, Angew. Chem. Int. Ed. 2004, 43, 1959-1962; Angew. Chem. 2004, 116, 1993-1996; b) C. Hamann, J.-M. Kern, J.-P. Sauvage, Inorg. Chem. 2003, 42, 1877-1883.

[48] A. M. L. Fuller, D. A. Leigh, P. J. Lusby, A. M. Z. Slawin, D. B. Walker, J. Am. Chem. Soc. 2005,127, 12612-12619.

[49] S. M. Goldup, D. A. Leigh, P. J. Lusby, R. T. McBurney, A. M. Z. Slawin, Angew. Chem. Int. Ed. 2008, 47, 6999-7003; Angew. Chem. 2008,120,7107 - 7111.

[50] a) C. Lincheneau, B. Jean-Denis, T. Gunnlaugsson, Chem. Commun. 2014, 50, 2857-2860; b) for a related molecular trefoil knot, see J.-F. Ayme, G. Gil-Ramírez, D. A. Leigh, J.-F. Lemonnier, A. Markevicius, C. A. Muryn, G. Zhang, J. Am. Chem. Soc. 2014,136, 13142-13145.

[51] a) L. Hogg, D. A. Leigh, P. J. Lusby, A. Morelli, S. Parsons, J.K.Y. Wong, Angew. Chem. Int. Ed. 2004, 43, 1218-1221; Angew. Chem. 2004, 116, 1238-1241; b) M. Hutin, C. A. Schalley, G. Bernardinelli, J. R. Nitschke, Chem. Eur. J. 2006, 12, 4069-4076; c) R. Price, J. K. Clegg, R. R. Fenton, L. F. Lindoy, J. C. McMurtrie, G. V. Meehan, A. Parkin, D. Perkins, P. Turner, Aust. J. Chem. 2009, 62, 1014-1019.

[52] a) D. A. Leigh, A. Murphy, J. P. Smart, A. M. Z. Slawin, Angew. Chem. Int. Ed. Engl. 1997, 36, 728-732; Angew. Chem. 1997, 109, 752 -756; b) C. Seel, A. H. Parham, O. Safarowsky, G. M. Hubner, F. Vögtle, J. Org. Chem. 1999, 64,7236 - 7242; c) A. G. Johnston, D. A. Leigh, A. Murphy, J. P. Smart, M. D. Deegan, J. Am. Chem. Soc. 1996, 118, 10662-10663; d) A. Altieri, G. Bottari, F. Dehez, D. A. Leigh, J. K. Y. Wong, F. Zerbetto, Angew. Chem. Int. Ed. 2003, 42, 2296 - 2300; Angew. Chem.

2003,115, 2398 - 2402; e) R. Ahmed, A. Altieri, D. M. D'Souza, D. A. Leigh, K. M. Mullen, M. Papmeyer, A. M. Z. Slawin, J. K. Y. Wong, J. D. Woollins, J. Am. Chem. Soc. 2011, 133, 12304-12310; f) A. Altieri, V. Aucagne, R. Carrillo, G.J. Clarkson, D. M. D'Souza, J. A. Dunnett, D. A. Leigh, K. M. Mullen, Chem. Sci. 2011, 2, 1922-1928.

[53] M. E. Padilla-Tosta, O. D. Fox, M. G. B. Drew, P. D. Beer, Angew. Chem. Int. Ed. 2001, 40, 4235-4239; Angew. Chem. 2001,113, 4365-4369.

[54] W. W. H. Wong, J. Cookson, E. A. L. Evans, E. J. L. McInnes, J. Wolowska, J. P. Maher, P. Bishop, P. D. Beer, Chem. Commun. 2005, 2214-2216.

[55] D. M. P. Mingos, J. Yau, S. Menzer, D. J. Williams, Angew. Chem. Int. Ed. Engl. 1995,34,1894-1895; Angew. Chem. 1995, 107, 2045-2047.

[56] S. S.-Y. Chui, R. Chen, C.-M. Che, Angew. Chem. Int. Ed. 2006, 45, 1621-1624; Angew. Chem. 2006,118, 1651-1654.

[57] J. D. Crowley, S. M. Goldup, A. L. Lee, D. A. Leigh, R. T. McBurney, Chem. Soc. Rev. 2009, 38, 1530-1541.

[58] V. Aucagne, K. D. Hanni, D. A. Leigh, P. J. Lusby, D. B. Walker, J. Am. Chem. Soc. 2006,128, 2186 - 2187.

[59] J. D. Crowley, K. D. Hänni, A. L. Lee, D. A. Leigh, J. Am. Chem. Soc. 2007,129, 12092-12093.

[60] V. Aucagne, J. Berna, J. D. Crowley, S. M. Goldup, K. D. Hänni, D. A. Leigh, P. J. Lusby, V. E. Ronaldson, A. M. Z. Slawin, A. Viterisi, D. B. Walker, J. Am. Chem. Soc. 2007, 129, 1195011963.

[61] a) S. M. Goldup, D. A. Leigh, T. Long, P. R. McGonigal, M. D. Symes, J. Wu, J. Am. Chem. Soc. 2009,131,15924-15929; b) Y. Sato, R. Yamasaki, S. Saito, Angew. Chem. Int. Ed. 2009, 48, 504-507; Angew. Chem. 2009,121, 512-515; c) M. J. Langton, J. D. Matichak, A. L. Thompson, H. L. Anderson, Chem. Sci. 2011, 2, 1897-1901.

[62] P. E. Barran, H. L. Cole, S. M. Goldup, D. A. Leigh, P. R. McGonigal, M. D. Symes, J. Y. Wu, M. Zengerle, Angew. Chem. Int. Ed. 2011, 50, 12280-12284; Angew. Chem. 2011, 123, 12488-12492.

[63] K. D. Hänni, D. A. Leigh, Chem. Soc. Rev. 2010,39,1240-1251.

[64] For reviews of the use of p-electron-rich/p-electron-poor interactions to form interlocked structures, see a) J. F. Stoddart, H.-R. Tseng, Proc. Natl. Acad. Sci. USA 2002, 99, 4797-4800; b) K. E. Griffiths, J. F. Stoddart, Pure Appl. Chem. 2008, 80, 485-506; c) S. J. Loeb, Chem. Soc. Rev. 2007, 36, 226-235.

[65] P. R. Ashton, T. T. Goodnow, A. E. Kaifer, M. V. Reddington, A. M. Z. Slawin, N. Spencer, J. F. Stoddart, C. Vicent, D. J. Williams, Angew. Chem. Int. Ed. Engl. 1989, 28, 1396-1399; Angew. Chem. 1989,101, 1404-1408.

[66] C. L. Brown, D. Philp, J. F. Stoddart, Synlett 1991, 459-461.

[67] G. Barin, A. Coskun, M. M. G. Fouda, J. F. Stoddart, Chem-PlusChem 2012, 77, 159-185.

[68] D. G. Hamilton, J. K. M. Sanders, J. E. Davies, W. Clegg, S. J. Teat, Chem. Commun. 1997, 897-898.

[69] K. Kitajima, T. Ogoshi, T. Yamagishi, Chem. Commun. 2014, 50, 2925-2927.

[70] T. Ogoshi, D. Yamafuji, T. Aoki, K. Kitajima, T. Yamagishi, Y. Hayashi, S. Kawauchi, Chem. Eur. J. 2012,18, 7493-7500.

[71] C. A. Hunter, D. H. Purvis, Angew. Chem. Int. Ed. Engl. 1992, 31, 792-795; Angew. Chem. 1992,104, 779-782.

[72] C. A. Hunter, J. Am. Chem. Soc. 1992,114, 5303-5311.

[73] H. Adams, F. J. Carver, C. A. Hunter, J. Chem. Soc. Chem. Commun. 1995, 809-810.

[74] a) F. Vögtle, T. Dünnwald, T. Schmidt, Acc. Chem. Res. 1996, 29, 451-460; b) B. Dung, F. Vogtle, J. Inclusion Phenom. 1988, 6, 429-442.

[75] F. Vögtle, S. Meier, R. Hoss, Angew. Chem. Int. Ed. Engl. 1992, 31, 1619-1622; Angew. Chem. 1992,104, 1628-1631.

[76] A. G. Johnston, D. A. Leigh, R. J. Pritchard, M. D. Deegan, Angew. Chem. Int. Ed. Engl. 1995, 34, 1209-1212; Angew. Chem. 1995,107, 1324-1327.

[77] M. S. Deleuze, D. A. Leigh, F. Zerbetto, J. Am. Chem. Soc. 1999,121, 2364-2379.

[78] a) D. A. Leigh, A. Venturini, A. J. Wilson, J. K. Y. Wong, F. Zerbetto, Chem. Eur. J. 2004, 10, 4960-4969; b) J. Berna, G. Bottari, D. A. Leigh, E. M. Perez, Pure Appl. Chem. 2007, 79, 39-54.

[79] a) A. C. Legon, Phys. Chem. Chem. Phys. 2010,12,7736 - 7747; b) P. Metrangolo, F. Meyer, T. Pilati, G. Resnati, G. Terraneo, Angew. Chem. Int. Ed. 2008, 47, 6114-6127; Angew. Chem. 2008,120, 6206-6220.

[80] A. Caballero, F. Zapata, N. G. White, P. J. Costa, V. Félix, P. D. Beer, Angew. Chem. Int. Ed. 2012, 51, 1876-1880; Angew. Chem. 2012,124, 1912-1916.

[81] L. C. Gilday, T. Lang, A. Caballero, P. J. Costa, V. Félix, P. D. Beer, Angew. Chem. Int. Ed. 2013, 52, 4356-4360; Angew. Chem. 2013,125, 4452 - 4456.

[82] a) F. Biedermann, W. M. Nau, H.-J. Schneider, Angew. Chem. Int. Ed. 2014, 53, 11158-11171; Angew. Chem. 2014, 126, 11338-11352; b) C. Tanford, The Hydrophobic Effect, 2nd ed., Wiley, New York, 1980 ; c) D. Chandler, Nature 2005, 437, 640647.

[83] a) M. Fujita, F. Ibukuro, H. Hagihara, K. Ogura, Nature 1994, 367, 720-723; b) for wholly organic "magic ring" catenane synthesis, see Ref. [44b].

[84] A "Möbius strip mechanism" has been proposed for the interconversion between a [2]catenane and macrocycle: M. Fujita, Acc. Chem. Res. 1999, 32, 53-61.

[85] C. J. Pedersen, Angew. Chem. Int. Ed. Engl. 1988, 27, 1021 -1027; Angew. Chem. 1988,100, 1053-1059.

[86] S. T. Tung, C. C. Lai, Y. H. Liu, S. M. Peng, S. H. Chiu, Angew. Chem. Int. Ed. 2013, 52, 13269-13272; Angew. Chem. 2013, 125, 13511-13514.

[87] For reviews on the use of anion templates to form interlocked structures, see a) M. S. Vickers, P. D. Beer, Chem. Soc. Rev. 2007, 36, 211-225; b) K. M. Mullen, P.D. Beer, Chem. Soc. Rev. 2009,38,1701 -1713 ; for other examples, see c) C. Reuter, W. Wienand, G. M. Hüner, C. Seel, F. Vögtle, Chem. Eur. J. 1999, 5, 2692-2697; d) J. M. Mahoney, R. Shukla, R. A. Marshall, A. M. Beatty, J. Zajicek, B. D. Smith, J. Org. Chem. 2002, 67, 1436-1440; e) P. Ghosh, O. Mermagen, C. A. Schalley, Chem. Commun. 2002, 2628-2629.

[88] M. R. Sambrook, P. D. Beer, J. A. Wisner, R. L. Paul, A. R. Cowley, J. Am. Chem. Soc. 2004,126, 15364-15365.

[89] K. Y. Ng, A. R. Cowley, P. D. Beer, Chem. Commun. 2006, 3676 - 3678.

[90] E. M. Kosower, J. L. Cotter, J. Am. Chem. Soc. 1964, 86, 55245527.

[91] J. C. Barnes, A. C. Fahrenbach, D. Cao, S. M. Dyar, M. Frasconi, M. A. Giesener, D. Benitez, E. Tkatchouk, O. Chernyashevskyy, W. H. Shin, H. Li, S. Sampath, C. L. Stern, A. A. Sarjeant, K. J. Hartlieb, Z. C. Liu, R. Carmieli, Y. Y. Botros, J. W. Choi, A. M. Z. Slawin, J. B. Ketterson, M. R. Wasielewski, W. A. Goddard, J. F. Stoddart, Science 2013, 339, 429 - 433.

[92] For reviews on synthesizing interlocked molecules by using reversible reactions, see a) C. D. Meyer, C. S. Joiner, J. F. Stoddart, Chem. Soc. Rev. 2007, 36, 1705-1723; b) P. C. Haussmann, J. F. Stoddart, Chem. Rec. 2009, 9, 136-154.

[93] R. T. S. Lam, A. Belenguer, S. L. Roberts, C. Naumann, T. Jarrosson, S. Otto, J. K. M. Sanders, Science 2005,308,667 - 669.

[94] M. K. Chung, P. S. White, S. J. Lee, M. R. Gagné, Angew. Chem. Int. Ed. 2009, 48, 8683-8686; Angew. Chem. 2009,121, 88398842.

[95] H. Y. Au-Yeung, G. D. Pantos, J. K. M. Sanders, Proc. Natl. Acad. Sci. USA 2009,106, 10466-10470.

[96] H. Y. Au-Yeung, G. D. Pantos, J. K. M. Sanders, J. Am. Chem. Soc. 2009,1S1, 16030-16032.

[97] J.-P. Sauvage, J. Weiss, J. Am. Chem. Soc. 19S5,107, 6108-6110.

[98] J. Guilhem, C. Pascard, J.-P. Sauvage, J. Weiss, J. Am. Chem. Soc. 19SS, 110, 8711-8713.

[99] C. O. Dietrich-Buchecker, A. Khemiss, J.-P. Sauvage, J. Chem. Soc. Chem. Commun. 198б, 1376-1378.

[100] P. R. Ashton, C. L. Brown, E. J. T. Chrystal, T. T. Goodnow, A. E. Kaifer, K. P. Parry, A. M. Z. Slawin, N. Spencer, J. F. Stoddart, D. J. Williams, Angew. Chem. Int. Ed. Engl. 1991, S0, 1039-1042; Angew. Chem. 1991,10S, 1055-1058.

[101] a) D. B. Amabilino, P. R. Ashton, A. S. Reder, N. Spencer, J. F. Stoddart, Angew. Chem. Int. Ed. Engl. 1994, SS, 1286-1290; Angew. Chem. 1994,106,1316-1319; b) D. B. Amabilino, P. R. Ashton, A. S. Reder, N. Spencer, J. F. Stoddart, Angew. Chem. Int. Ed. Engl. 1994, SS, 433-437; Angew. Chem. 1994,106,450453; c) D. B. Amabilino, P. R. Ashton, V. Balzani, S. E. Boyd, A. Credi, J. Y. Lee, S. Menzer, J. F. Stoddart, M. Venturi, D. J. Williams, J. Am. Chem. Soc. 199S, 120, 4295 - 4307.

[102] a) D. A. Leigh, J. K. Y. Wong, F. Dehez, F. Zerbetto, Nature 2003,424,174 -179 ; b) J. V. Hernández, E. R. Kay, D. A. Leigh, Science 2004, S06, 1532-1537.

[103] A. Hori, K. Kumazawa, T. Kusukawa, D. K. Chand, M. Fujita, S. Sakamoto, K. Yamaguchi, Chem. Eur. J. 2001, 7, 4142-4149.

[104] A. L. Hubbard, G. J. E. Davidson, R. H. Patel, J. A. Wisner, S. J. Loeb, Chem. Commun. 2004, 138-139.

[105] F. B. L. Cougnon, N. A. Jenkins, G. D. Pantos, J. K. M. Sanders, Angew. Chem. Int. Ed. 2012, 51, 1443-1447; Angew. Chem. 2012,124, 1472-1476.

[106] F. Bitsch, C. O. Dietrich-Buchecker, A. K. Khémiss, J.-P. Sauvage, A. Vandorsselaer, J. Am. Chem. Soc. 1991, 11S, 4023 - 4025.

[107] For a review of cucurbituril rotaxane and catenane systems, see K. Kim, Chem. Soc. Rev. 2002, S1, 96-107.

[108] D. Whang, K.-M. Park, J. Heo, P. Ashton, K. Kim, J. Am. Chem. Soc. 199S, 120, 4899-4900.

[109] K.-M. Park, S.-Y. Kim, J. Heo, D. Whang, S. Sakamoto, K. Yamaguchi, K. Kim, J. Am. Chem. Soc. 2002,124, 2140-2147.

[110] S.-G. Roh, K.-M. Park, G.-J. Park, S. Sakamoto, K. Yamaguchi, K. Kim, Angew. Chem. Int. Ed. 1999, S8, 637 - 641; Angew. Chem. 1999,111,671-675.

[111] L. Wang, M. O. Vysotsky, A. Bogdan, M. Bolte, V. Böhmer, Science 2004, S04, 1312-1314.

[112] a) C.-M. Jin, H. Lu, J. Huang, Chem. Commun. 200б, 50395041; b) X. Kuang, X. Wu, R. Yu, J. P. Donahue, J. Huang, C.-Z. Lu, Nat. Chem. 2010, 2, 461-465; c) J. Heine, J. Schmedt auf der Günne, D. Dehen, J. Am. Chem. Soc. 2011, 1SS, 1001810021; d) Q. Chen, F. Jiang, D. Yuan, G. Lyu, L. Chen, M. Hong, Chem. Sci. 2014, 5, 483 - 488.

[113] F. Patat, P. Derst, Angew. Chem. 1959, 71, 105-110.

[114] K. Endo, T. Shiroi, N. Murata, G. Kojima, T. Yamanaka, Macromolecules 2004, S7, 3143 - 3150.

[115] H. Ishida, A. Kisanuki, K. Endo, Polym. J. 2009, 41, 110-117.

[116] M. Fujita, N. Fujita, K. Ogura, K. Yamaguchi, Nature 1999, 400, 52-55.

[117] A. Westcott, J. Fisher, L. P. Harding, P. Rizkallah, M. J. Hardie, J. Am. Chem. Soc. 200S, 1S0, 2950-2951.

[118] a) M. Fukuda, R. Sekiya, R. Kuroda, Angew. Chem. Int. Ed. 200S, 47, 706 - 710; Angew. Chem. 200S, 120, 718-722; b) R. Sekiya, M. Fukuda, R. Kuroda, J. Am. Chem. Soc. 2012, 1S4, 10987-10997.

[119] M. Frank, J. Hey, I. Balcioglu, Y. S. Chen, D. Stalke, T. Suenobu, S. Fukuzumi, H. Frauendorf, G. H. Clever, Angew. Chem. Int. Ed. 2013, 52, 10102-10106; Angew. Chem. 2013, 125, 1028810293.

[120] Y. T. Li, K. M. Mullen, T. D. W. Claridge, P. J. Costa, V. Felix, P. D. Beer, Chem. Commun. 2009, 7134-7136.

[121] T. Hasell, X. F. Wu, J. T. A. Jones, J. Bacsa, A. Steiner, T. Mitra, A. Trewin, D. J. Adams, A. I. Cooper, Nat. Chem. 2010,2, 750755.

[122] T. Tozawa, J. T. A. Jones, S. I. Swamy, S. Jiang, D. J. Adams, S. Shakespeare, R. Clowes, D. Bradshaw, T. Hasell, S. Y. Chong, C. Tang, S. Thompson, J. Parker, A. Trewin, J. Bacsa, A. M. Z. Slawin, A. Steiner, A. I. Cooper, Nat. Mater. 2009, 8, 973 - 978.

[123] G. Zhang, O. Presly, F. White, I. M. Oppel, M. Mastalerz, Angew. Chem. Int. Ed. 2014, 53, 5126-5130; Angew. Chem. 2014,126, 5226-5230.

[124] S. P. Black, A. R. Stefankiewicz, M. M. J. Smulders, D. Sattler, C. A. Schalley, J. R. Nitschke, J. K. M. Sanders, Angew. Chem. Int. Ed. 2013, 52, 5749-5752; Angew. Chem. 2013,125, 58615864.

[125] J. F. Nierengarten, C. O. Dietrich-Buchecker, J.-P. Sauvage, J. Am. Chem. Soc. 1994,116, 375-376.

[126] C. Dietrich-Buchecker, J.-P. Sauvage, Chem. Commun. 1999, 615-616.

[127] C. Dietrich-Buchecker, B. Colasson, D. Jouvenot, J.-P. Sauvage, Chem. Eur. J. 2005,11, 4374-4386.

[128] B. Hasenknopf, J.-M. Lehn, N. Boumediene, A. Dupont-Gervais, A. Van Dorsselaer, B. Kneisel, D. Fenske, J. Am. Chem. Soc. 1997,119, 10956-10962.

[129] J. E. Beves, C. J. Campbell, D. A. Leigh, R. G. Pritchard, Angew. Chem. Int. Ed. 2013, 52, 6464-6467; Angew. Chem.

2013.125, 6592 - 6595.

[130] C. P. McArdle, J. J. Vittal, R. J. Puddephatt, Angew. Chem. Int. Ed. 2000,39, 3819-3822; Angew. Chem. 2000,112, 3977-3980.

[131] C. Schouwey, J. J. Holstein, R. Scopelliti, K. O. Zhurov, K. O. Nagornov, Y. O. Tsybin, O. S. Smart, G. Bricogne, K. Severin, Angew. Chem. Int. Ed. 2014, 53, 11261-11265; Angew. Chem.

2014.126, 11443-11447.

[132] a) C. D. Pentecost, K. S. Chichak, A. J. Peters, G. W. V. Cave, S. J. Cantrill, J. F. Stoddart, Angew. Chem. Int. Ed. 2007, 46, 218-222; Angew. Chem. 2007, 119, 222-226; b) T. Prakasam, M. Lusi, M. Elhabiri, C. Platas-Iglesias, J.-C. Olsen, Z. Asfari, S. Cianférani-Sanglier, F. Debaene, L. J. Charbonnière, A. Tra-bolsi, Angew. Chem. Int. Ed. 2013, 52, 9956 - 9960; Angew. Chem. 2013,125, 10140-10144.

[133] N. Ponnuswamy, F. B. L. Cougnon, G. D. Pantos, J. K. M. Sanders, J. Am. Chem. Soc. 2014,136, 8243 - 8251.

[134] T. K. Ronson, J. Fisher, L. P. Harding, P. J. Rizkallah, J. E. Warren, M. J. Hardie, Nat. Chem. 2009,1, 212-216.

[135] D. A. Leigh, R. G. Pritchard, A. J. Stephens, Nat. Chem. 2014, 6, 978-982.

[136] C. Z. Liang, K. Mislow, J. Math. Chem. 1994,16, 27-35.

[137] K. S. Chichak, S. J. Cantrill, A. R. Pease, S. H. Chiu, G. W. V. Cave, J. L. Atwood, J. F. Stoddart, Science 2004, 304, 13081312.

[138] A. J. Peters, K. S. Chichak, S. J. Cantrill, J. F. Stoddart, Chem. Commun. 2005, 3394-3396.

[139] C. D. Meyer, R. S. Forgan, K. S. Chichak, A. J. Peters, N. Tangchaivang, G. W. V. Cave, S. I. Khan, S. J. Cantrill, J. F. Stoddart, Chem. Eur. J. 2010,16, 12570-12581.

[140] a) S.-L. Huang, Y.-J. Lin, T. S. A. Hor, G.-X. Jin, J. Am. Chem. Soc. 2013,135, 8125-8128; b) S.-L. Huang, Y.-J. Lin, Z.-H. Li, G.-X. Jin, Angew. Chem. Int. Ed. 2014, 53, 11218-11222; Angew. Chem. 2014,126, 11400-11404.

[141] E. R. Kay, D. A. Leigh, Top. Curr. Chem. 2005, 262, 133-177.

[142] For reviews on metal-template interlocked systems as molecular machine prototypes, see a) J.-P. Sauvage, Acc. Chem. Res. 1998, 31, 611-619; b)J.-P. Collin, C. Dietrich-Buchecker, P. Gaviña, M. C. Jiménez-Molero, J.-P. Sauvage, Acc. Chem. Res. 2001, 34, 477 - 487; c) C. Dietrich-Buchecker, M. C. Jiménez-Molero, V. Sartor, J.-P. Sauvage, Pure Appl. Chem. 2003, 75,

1383-1393; d) J.-P. Collin, V. Heitz, S. Bonnet, J.-P. Sauvage, Inorg. Chem. Commun. 2005, 8, 1063-1074; e) J.-P. Collin, V. Heitz, J.-P. Sauvage, Top. Curr. Chem. 2005, 262, 29 - 62; f) S. Bonnet, J.-P. Collin, M. Koizumi, P. Mobian, J.-P. Sauvage, Adv. Mater. 2006, 18, 1239-1250; g) B. Champin, P. Mobian, J.-P. Sauvage, Chem. Soc. Rev. 2007, 36, 358-366; h) E. R. Kay, D. A. Leigh, F. Zerbetto, Angew. Chem. Int. Ed. 2007, 46, 72191; Angew. Chem. 2007,119,72-196; i) S. Durot, F. Reviriego, J.-P. Sauvage, Dalton Trans. 2010, 39, 10557-10570.

[143] a) A. Livoreil, C. O. Dietrich-Buchecker, J.-P. Sauvage, J. Am. Chem. Soc. 1994,116, 9399-9400; b) F. Baumann, A. Livoreil, W. Kaim, J.-P. Sauvage, Chem. Commun. 1997, 35-36; c) A. Livoreil, J.-P. Sauvage, N. Armaroli, V. Balzani, L. Flamigni, B. Ventura, J. Am. Chem. Soc. 1997, 119, 12114-12124.

[144] D. J. Cárdenas, A. Livoreil, J.-P. Sauvage, J. Am. Chem. Soc. 1996, 118, 11980-11981.

[145] M. Asakawa, P. R. Ashton, V. Balzani, A. Credi, C. Hamers, G. Mattersteig, M. Montalti, A. N. Shipway, N. Spencer, J. F. Stoddart, M. S. Tolley, M. Venturi, A. J. P. White, D. J. Williams, Angew. Chem. Int. Ed. 1998, 37,333-337; Angew. Chem. 1998, 110, 357-361.

[146] Z. X. Zhu, A. C. Fahrenbach, H. Li, J. C. Barnes, Z. C. Liu, S. M. Dyar, H. C. Zhang, J. Y. Lei, R. Carmieli, A. A. Sarjeant, C. L. Stern, M. R. Wasielewski, J. F. Stoddart, J. Am. Chem. Soc. 2012,134, 11709-11720.

[147] D. B. Amabilino, C. O. Dietrich-Buchecker, A. Livoreil, L. Pérez-García, J.-P. Sauvage, J. F. Stoddart, J. Am. Chem. Soc. 1996,118, 3905-3913.

[148] D. A. Leigh, P. J. Lusby, A. M. Z. Slawin, D. B. Walker, Chem. Commun. 2005, 4919-4921.

[149] a) A. Andrievsky, F. Ahuis, J. L. Sessler, F. Vögtle, D. Dudat, M. Moini, J. Am. Chem. Soc. 1998,120,9712 - 9713 ; b) K. Y. Ng, V. Felix, S. M. Santos, N. H. Rees, P. D. Beer, Chem. Commun. 2008, 1281-1283; c) N. H. Evans, C. J. Serpell, P.D. Beer, Chem. Eur. J. 2011,17, 7734-7738.

[150] P. Mobian, J.-M. Kern, J.-P. Sauvage, Angew. Chem. Int. Ed. 2004, 43, 2392 - 2395; Angew. Chem. 2004,116, 2446-2449.

[151] a) A. Caballero, F. Zapata, P. D. Beer, Coord. Chem. Rev. 2013, 257, 2434 - 2455; b) M. J. Langton, P. D. Beer, Acc. Chem. Res. 2014, 47, 1935-1949.

[152] N. H. Evans, H. Rahman, A. V. Leontiev, N. D. Greenham, G. A. Orlowski, Q. Zeng, R. M. J. Jacobs, C. J. Serpell, N. L. Kilah, J. J. Davis, P.D. Beer, Chem. Sci. 2012, 3, 1080-1089.

[153] M. K. Chae, J. M. Suk, K. S. Jeong, Tetrahedron Lett. 2010, 51, 4240-4242.

[154] Y. Nakatani, Y. Furusho, E. Yashima, Angew. Chem. Int. Ed. 2010, 49, 5463-5467; Angew. Chem. 2010,122, 5595 - 5599.

[155] C. J. G. Plummer, N. Cudré-Mauroux, H. H. Kausch, Polym. Eng. Sci. 1994, 34, 318-329.

[156] S. Wu, J. Polym. Sci. Part B 1989, 27, 723-741.

[157] J. S. Shaffer, J. Chem. Phys. 1994,101, 4205-4213.

[158] J.-L. Weidmann, J.-M. Kern, J.-P. Sauvage, Y. Geerts, D. Muscat, L. Müllen, Chem. Commun. 1996, 1243-1244.

[159] J.-L. Weidmann, J.-M. Kern, J.-P. Sauvage, D. Muscat, S. Mullins, W. Köhler, C. Rosenauer, H. J. Räder, K. Martin, Y. Geerts, Chem. Eur. J. 1999, 5, 1841-1851.

[160] S. Shimada, K. Ishikawa, N. Tamaoki, Acta Chem. Scand. 1998, 52, 374-376.

[161] a) C.-A. Fustin, C. Bailly, G. J. Clarkson, P. De Groote, T. H. Galow, D. A. Leigh, D. Robertson, A. M. Z. Slawin, J. K. Y. Wong, J. Am. Chem. Soc. 2003, 125, 2200 - 2207; b) C. A. Fustin, G. J. Clarkson, D. A. Leigh, F. Van Hoof, A. M. Jonas, C. Bailly, Macromolecules 2004, 37, 7884-7892.

[162] T. Hagiwara, Y. Murano, Y. Watanabe, T. Hoshi, T. Sawaguchi, Tetrahedron Lett. 2012, 53, 2805 - 2808.

[163] T. Hagiwara, M. Yamazaki, T. Suzuki, T. Sawaguchi, S. Yano, Polymer 2011, 52, 5426-5430.

[164] a) S. R. Batten, R. Robson, Angew. Chem. Int. Ed. 1998, 37, 1460-1494; Angew. Chem. 1998, 110, 1558-1595; b) L. Car-lucci, G. Ciani, D. M. Proserpio, Coord. Chem. Rev. 2003, 246, 247 - 289; c) Z. B. Niu, H. W. Gibson, Chem. Rev. 2009, 109, 6024-6046; d) K. Pangilinan, R. Advincula, Polym. Int. 2014, 63, 803-813.

[165] a) C. Hamers, F. M. Raymo, J. F. Stoddart, Eur. J. Org. Chem. 1998, 2109-2117; b) D. L. Simone, T. M. Swager, J. Am. Chem. Soc. 2000,122, 9300 - 9301.

[166] M. Bria, J. Bigot, G. Cooke, J. Lyskawa, G. Rabani, V. M. Rotello, P. Woisel, Tetrahedron 2009, 65, 400 - 407.

[167] a) S. J. Loeb, Chem. Commun. 2005, 1511-1518; b) V. N. Vukotic, K. J. Harris, K. L. Zhu, R. W. Schurko, S. J. Loeb, Nat. Chem. 2012, 4, 456-460.

[168] Q. W. Li, W.Y. Zhang, O.S. Miljanic, C.B. Knobler, J. F. Stoddart, O. M. Yaghi, Chem. Commun. 2010, 46, 380-382.

[169] D. Cao, M. Juriccek, Z. J. Brown, A. C.-H. Sue, Z. Liu, J. Lei, A. K. Blackburn, S. Grunder, A. A. Sarjeant, A. Coskun, C. Wang, O. K. Fahra, J. T. Hupp, J. F. Stoddart, Chem. Eur. J. 2013,19, 8457-8465.

[170] Q. W. Li, C. H. Sue, S. Basu, A. K. Shveyd, W. Y. Zhang, G. Barin, L. Fang, A. A. Sarjeant, J. F. Stoddart, O. M. Yaghi, Angew. Chem. Int. Ed. 2010, 49, 6751-6755; Angew. Chem. 2010,122, 6903 - 6907.

[171] E. Coronado, P. Gavina, S. Tatay, Chem. Soc. Rev. 2009, 38, 1674-1689.

[172] C. De Nadaï, C. M. Whelan, C. Perollier, G. Clarkson, D. A. Leigh, R. Caudano, P. Rudolf, Surf. Sci. 2000, 454, 112-117.

[173] T. B. Lu, L. Zhang, G. W. Gokel, A. E. Kaifer, J. Am. Chem. Soc. 1993,115, 2542 - 2543.

[174] a) L. Raehm, C. Hamann, J.-M. Kern, J.-P. Sauvage, Org. Lett. 2000, 2, 1991-1994; b) L. Raehm, J.-M. Kern, J.-P. Sauvage, C. Hamann, S. Palacin, J.-P. Bourgoin, Chem. Eur. J. 2002, 8, 2153 -2162.

[175] C. P. Collier, G. Mattersteig, E. W. Wong, Y. Luo, K. Beverly, J. Sampaio, F. M. Raymo, J. F. Stoddart, J. R. Heath, Science 2000, 289, 1172-1175.

[176] R. Klajn, L. Fang, A. Coskun, M. A. Olson, P. J. Wesson, J. F. Stoddart, B. A. Grzybowski, J. Am. Chem. Soc. 2009, 131, 4233-4235.

[177] a) C. A. Fustin, P. Rudolf, A. F. Taminiaux, F. Zerbetto, D. A. Leigh, R. Caudano, Thin Solid Films 1998, 329, 321-325; b) C. M. Whelan, F. Cecchet, G. J. Clarkson, D. A. Leigh, R. Caudano, P. Rudolf, Surf. Sci. 2001, 474,71-80; c) C.-A. Fustin, R. Gouttebaron, C. De Nadaï, R. Caudano, P. Rudolf, F. Zerbetto, D. A. Leigh, Surf. Sci. 2001, 474, 37-40; d) C. A. Fustin, R. Gouttebaron, R. Caudano, P. Rudolf, D. A. Leigh, M. Fanti, A. Krug, F. Zerbetto, Surf. Sci. 2002,515, 45 - 52 ; e) C. A. Fustin, S. Haq, A. Wingen, C. Grégoire, R. Raval, P. Dumas, J. S. Hannam, D. A. Leigh, P. Rudolf, Surf Sci. 2005,580,57-62.

[178] D. Payer, S. Rauschenbach, N. Malinowski, M. Konuma, C. Virojanadara, U. Starke, C. Dietrich-Buchecker, J.-P. Collin, J.-P. Sauvage, N. Lin, K. Kern, J. Am. Chem. Soc. 2007, 129, 15662-15667.

[179] While this Review was in press, two impressive syntheses of 633 links were reported: a) R. Zhu, J. Lübben, B. Dittrich, G. H. Clever, Angew. Chem. Int. Ed. 2015, 54, 2796 - 2800; Angew. Chem. 2015, 127, 2838-2842; b) C. S. Wood, T.K. Ronson, A. M. Belenguer, J. J. Holstein, J. R. Nitschke, Nat. Chem. 2015, 7, 354-358; and a 2x2 interwoven grid was used to assemble a Solomon link: c) J. E. Beves, J. J. Danon, D. A. Leigh, J.-F. Lemonnier, I. J. Vitorica-Yrezabal, Angew. Chem. Int. Ed. DOI:10.1002/anie.201502095; Angew. Chem. DOI:10.1002/ ange.201502095..

[180] a) C. H. Heathcock, Proc. Natl. Acad. Sci. USA 1996, 93, 14323-14327; b) K. C. Nicolaou, E.J. Sorensen, Classics in Total Synthesis, Wiley-VCH, Weinheim, 2006.

Received: December 2, 2014

6150 www.angewandte.org © 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Angew. Chem. Int. Ed. 2015, 54, 6110-6151

** These are not the final page numbers!

Reviews

Catenanes

G. Gil-Ramírez, D. A. Leigh,*

A. J. Stephens - llll - III!

Catenanes: Fifty Years of Molecular Links

Half a century after Schill and Luttringhaus carried out the first directed synthesis of a [2]catenane, a plethora of strategies now exist for the construction of interlocked molecular rings. Effective template synthesis enables the synthesis

of higher order interlocked systems. This Review outlines the diverse strategies that exist for forming catenanes, their applications, and the important challenges that remain in the field of chemical topology.