Intercomponent interactions and mobility in hydrogen-bonded rotaxanes
Jagesar, D.C.
Publication date 2010
Document Version Final published version
Link to publication
Citation for published version (APA):
Jagesar, D. C. (2010). Intercomponent interactions and mobility in hydrogen-bonded rotaxanes.
General rights
It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons).
Disclaimer/Complaints regulations
If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Ask the Library: https://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. You will be contacted as soon as possible.
D h i r e d j C. J a g e s a r
Intercomponent Interactions and Mobility
in Hydrogen-Bonded Rotaxanes
componen
tIn
ter
actionsandMobi
lityinHy
dr
og
en-Bonde
dR
ot
a
x
ane
sDhir
e
djC.Jag
ISBN 978-90-9025766-2
Intercomponent Interactions and
Mobility in Hydrogen-Bonded
Rotaxanes
Intercomponent Interactions and
Mobility in Hydrogen-Bonded
Rotaxanes
ACADEMISCH PROEFSCHRIFT
Ter verkrijging van de graad van doctor aan de Universiteit van Amsterdam,
op gezag van de Rector Magnificus Prof. Dr. D.C. van den Boom
ten overstaan van een door het college voor promoties ingestelde commissie,
in het openbaar te verdedigen in de Aula der Universiteit op vrijdag 26 november 2010, te 11.00 uur
door
Dhiredj Chandre Jagesar
geboren te Nickerie, SurinamePromotor: Prof. dr. A.M. Brouwer Co-promotor: Prof. dr. W.J. Buma Overige leden: Dr. E. Eiser
Prof. dr. C.J. Elsevier Prof. dr. F. Hartl Dr. G. Lodder Prof. dr. F. Paolucci Prof. dr. J.N.H. Reek Dr. S. Woutersen
Faculteit der Natuurwetenschappen, Wiskunde en Informatica
Het onderzoek beschreven in dit proefschrift werd uitgevoerd binnen het Van ’t Hoff Institute for Molecular Sciences, Faculteit der Natuurwetenschappen, Universiteit van Amsterdam.
Dit werk kwam tot stand met financiele ondersteuning van de Nederlandse Organisatie voor Wetenschappelijk Onderzoek (NWO).
Printed by Wöhrman Print Service
C h a p t e r 1 1
Introduction
Intercomponent Interactions and Mobility in Rotaxanes
1.1 Interlocked Molecules... 2
1.2 Intercomponent Interactions in Rotaxanes ... 3
1.2.1 Synthesis ... 3
1.2.2 Hydrogen-Bonded Rotaxanes ... 4
1.2.3 π-Stacking... 9
1.2.4 Hydrophobic Interactions ...10
1.2.5 Transition Metal Coordination...11
1.3 Intercomponent Dynamics in Rotaxanes... 11
1.3.1 Chemically Driven Molecular Shuttles...13
1.3.2 Electron-Driven Molecular Shuttles...14
1.3.3 Light-Driven Molecular Shuttles...17
1.4 Scope of this Thesis ... 21
1.5 References... 22
C h a p t e r 2 31
Photoinduced Shuttling Dynamics of Rotaxanes in Viscous Polymer
Solutions
2.1 Introduction... 322.2 Results and Discussion ... 35
2.2.1 Rheological Behavior ...35
2.2.2 Shuttling ...37
2.2.3 Hydrodynamic Scaling Model...41
2.2.4 Correlation with Macroscopic Viscosity...42
2.2.5 Power-Law Relationship...43
2.3 Conclusion ... 46
2.4 Photophysics of the Naphthalimide Rotaxane... 46
2.5 Experimental Details... 49
2.6 References... 51
C h a p t e r 3 57
Naphthalimide Rotaxanes
Infrared Study of Intercomponent Interactions in a Switchable Hydrogen-Bonded Rotaxane
3.1 Introduction... 583.2 Results and Discussion ... 61
3.2.1 Amide I...61
3.2.2 Solvent effect...66
3.2.3 NH stretching...68
3.2.4 Infrared Spectroelectrochemistry...69
3.3 Conclusion ... 75
3.4 Appendix: Calculations N-Methylacetamide ... 76
3.5 Experimental Details... 76
4.1 Introduction ...84
4.2 Results and Discussion...87
4.2.1 Infrared Spectra ... 87 4.2.2 Infrared Spectroelectrochemistry ... 89 4.3 Conclusion...99 4.4 Experimental Details ... 100 4.5 References ... 102 C h a p t e r 5 105
Perylene Diimide Rotaxanes
Hydrogen Bonding and Macrocycle Translocation
5.1 Introduction ... 1065.2 Results and Discussion... 108
5.2.1 Perylene Diimide Rotaxanes [2]8 and [3]8... 108
5.2.2 Perylene Diimide Shuttles [2]9 and [3]9... 111
5.3 Conclusion... 122
5.4 Experimental Details ... 123
5.5 References ... 124
C h a p t e r 6 129
Photoinduced Dynamics in Structurally Modified Naphthalimide Rotaxanes
6.1 Introduction ... 130 6.2 Co-conformer Distribution ... 132 6.3 Photophysical Behavior ... 134 6.4 Photoinduced Shuttling... 137 6.4.1 niGly Rotaxane 11... 137 6.4.2 C9 Rotaxane 12... 1396.5 Temperature Dependence of the Shuttling Rate... 140
6.6 Discussion ... 142 6.7 Conclusion... 147 6.8 Experimental Details ... 148 6.9 References ... 150 C h a p t e r 7 153
Fullerene Rotaxanes
Reverse Shuttling in Fullerene-Stoppered Rotaxane
7.1 Introduction ... 1547.2 Results and Discussion... 155
7.2.1 Solvent Induced Shuttling in C60R... 155
7.2.2 UV-Vis Absorption and Fluorescence ... 157
7.2.3 Transient Absorption ... 160
7.2.4 Driving Force for Shuttling ... 162
7.3 Conclusion... 163
7.4 Experimental Details ... 164
C h a p t e r 8 169
Pressure-Induced Absorption and Fluorescence Shifts of Solvatochromic
Probes
8.1 Introduction...170
8.1.1 Pressure-Viscosity Relationship ...170
8.1.2 Pressure-Induced Polarity Change...171
8.2 Results and Discussion ...173
8.2.1 Absorption Probe ET30...173
8.2.2 Fluorescence Probe 5PI...176
8.3 Conclusion ...178
8.4 Experimental Details...179
8.4.1 High-Pressure Setup...179
8.4.2 Absorption and Fluorescence...181
8.5 References...182
Summary 185 Samenvatting 189
C h a p t e r 1
Introduction
Intercomponent Interactions and Mobility in Rotaxanes
Abstract
Rotaxanes are among the most prominent members of the class of interlocked molecules. In the field of research on rotaxanes and other interlocked architectures, scientists are facing many challenges, but also many interesting opportunities are recognized. In this introductory chapter the structural aspects of rotaxanes in terms of mechanical bonding and interactions between components are discussed. The focus of this thesis is on hydrogen-bonded rotaxanes; therefore special attention is reserved for intercomponent hydrogen bonding interactions. Also the different approaches for the synthesis of rotaxanes are discussed.
Interlocked molecules possess unique properties arising from the restriction of degrees of freedom in comparison with their separate components. In appropriately designed systems, the mechanical movements of the components with respect to each other can be made to occur between different well-defined states in a controlled manner. These molecular scale switches are referred to as molecular motors, for they are able to convert chemical, electrochemical and photochemical energy into controllable molecular motion. The molecular motor functionality of rotaxanes is discussed and illustrated with several state-of-the art prototypes.
1.1
Interlocked Molecules
The structure of organic molecules is traditionally described in terms of the number and types of atoms they contain and the sequence and nature of the connecting bonds. Two molecules containing the same atoms, linked in a different sequence, are described as constitutional isomers, for example n-butane and 2-methylpropane. Even when two molecules contain the same atoms and sequence it is still possible for isomers to exist, originating from different spatial arrangements of atoms. This isomerism is referred to as stereoisomerism and arises in molecules containing substituted double bonds (E-Z isomerism) and asymmetric carbon atoms (optical isomerism). Over the years, as the range of molecules prepared became more complex, many special forms of isomerism (e.g. in octahedral metal complexes) have been identified. However, these are all variants upon one of the fundamental types of isomerism.
A new type of isomerism was first described in the early 1960’s, namely topological isomerism.[1] This type of isomerism comes into play when considering the structure of mechanically interlocked architectures such as catenanes and rotaxanes (Figure 1-1).
[2]rotaxane [2]catenane trefoil knot
Solomon knot Borromean rings
Figure 1-1 Schematic representation of some interlocked molecules.
An [n]catenane, from the Latin word "catena" meaning chain, consists of n rings which are mechanically interlocked. The two rings in the [2]catenane in Figure 1-1 do not differ from the unlinked rings in terms of the atoms or bonds they contain, yet they are chemically different. The two structures, the separate rings and the catenane, are called topological isomers, because it is impossible to convert the catenane into its two separate rings without breaking chemical bonds. Other examples of interlocked topological isomers, of which synthetically created molecular embodiments are known, are trefoil knots,[2-5] Solomon knots (doubly interlocked catenanes)[6-9] and molecular Borromean rings.[10,11]
[n]Rotaxanes, from the Latin words “rota”, meaning wheel and “axis”, meaning axle, consist of a dumbbell-shaped molecule encircled by n-1 macrocycles. The macrocycle is mechanically trapped on the axle due to the bulky stoppers on each end which prevent the
macrocycle to slip away. Despite the fact that the macrocycle is mechanically locked onto the thread, rotaxanes are not considered topological isomers of their components, because they can be separated without breaking of chemical bonds: dissociation of a rotaxane into its separate parts is in principle possible by simply slipping the macrocycle over a stopper.
In the literature, several examples of [n]rotaxanes with more than one macrocycle (n > 2) or thread have been reported. Some examples of these rotaxanes with more complex structures are (Figure 1-2): daisy chains[12-15] (A), polyrotaxanes[16-18] (B) and doubly threaded
rotaxanes[19-22] (C).
(A) (B) (C)
Figure 1-2 Examples of rotaxanes with more complex structures. (A): Daisy chain, (B): polyrotaxane and (C): doubly threaded rotaxane.
Daisy chains are referred to as “molecular muscles” because their components can perform contraction and extension movements, reminiscent of the shortening and lengthening of the functional elements in muscle fibres (see also Figure 1-10).
1.2
Intercomponent Interactions in Rotaxanes
1.2.1 Synthesis
In contrast with the synthesis of ordinary organic molecules, in which chemical bonds between atoms are formed, for the construction of rotaxanes one has to deal with the challenge of creating mechanical bonds. Two main strategies can be distinguished in the synthesis of rotaxanes: threading and clipping (Figure 1-3). In the threading approach, the axle without the end groups is threaded through the cavity of a macrocycle to form a so-called pseudorotaxane, which after endcapping affords the rotaxane. Another approach is clipping, in which the macrocycle is assembled around the axle. A third and less often used strategy is slipping (not shown). In this approach the macrocycle is slided onto an already existing thread, the slipping can be induced by heat[23,24] or pressure.[25]
The major event in the abovementioned approaches is reaching a specific spatial arrangement of the molecular components’ precursors with respect to each other. The first successful synthesis of a rotaxane relied on the low statistical probability of the thread precursor threading through a resin-immobilized macrocycle, followed by capping.[26] This
Figure 1-3 Two different approaches generally used in the template-directed synthesis of rotaxanes. (A): Threading and capping and (B): clipping. The template which serves to pre-organize the precursors is represented by the rectangle.
The recognition and development of templates in synthesis has dramatically improved the accessibility to a wide variety of rotaxanes. In template-directed synthesis the required spatial arrangement of the components is induced by noncovalent interactions between recognition sites incorporated into the precursors of the macrocycle and the thread. In the past two decades several templating mechanisms have been developed for rotaxane synthesis. In fact, all types of noncovalent interactions known in chemistry are suitable for this purpose and have indeed been utilized for the template-directed synthesis of rotaxanes. In the majority of the template-directed approaches, the noncovalent interactions that are used to correctly align the precursors, survive the chemical reactions necessary to link the components, and live on in the resulting rotaxanes. As will be discussed in section 1.3, the manipulation of these interactions is the starting point for the application of rotaxanes as molecular motors or switches. The pioneering work of several research groups, each specialized in a certain type of templating interaction, has resulted in a huge number of different rotaxanes. Three of the four major groups of rotaxanes rely on specific interactions between the thread and the macrocycle, namely hydrogen bonding, π-stacking, and transition metal ion coordination. Another important class is assembled through the hydrophobic effect. In the next sections, an overview is presented of the rotaxanes based on these different types of interactions.
1.2.2 Hydrogen-Bonded Rotaxanes
Hydrogen bonds are formed between a donor with an available “acidic” hydrogen atom, e.g. NH or OH groups, and an acceptor carrying available non-bonding lone pairs of electrons. The first example of a hydrogen-bonded rotaxane was reported in the literature by F. Vögtle and coworkers.[27] This rotaxane with hydrogen bonding between amide groups
in the macrocycle and the thread was synthesized using templating effects following the threading and capping approach. This research group also developed a trapping approach, based on hydrogen bonding of the macrocycle with an alkoxy anion and followed by reaction with an alkylbromide to form an ether bond.[28,29] A typical feature of these
rotaxanes with ether linkages is that the interaction that is applied to pre-align the axle precursors is lost once the rotaxane is formed. An example of such a rotaxane is discussed in Chapter 5.
In the 1990’s, the research group of D.A. Leigh developed another powerful synthetic strategy for the synthesis of a new group of hydrogen-bonded rotaxanes.[30-32] Their
five-component clipping strategy utilizes a dipeptide motif in the axle as template for the formation of the benzylic tetraamide macrocycle from its precursors (isophthaloyl dichloride and xylylene diamine, Figure 1-4A).
(A)
(B)
Macrocycles Binding motifs
amide glycylglycine (GlyGly) succinamide (succ)
fumaramide (fum) glycyl glycyl ester succinic amide ester
X = Y = CH, R = H X = N, Y = CH[33-35] X = CH, Y = N[36,37]
X = Y = CH, R = NO2[37,38] adipamide[39] nitrone squaraine
Figure 1-4 (A): Example of hydrogen-bond assisted rotaxane synthesis using a fumaramide motif as template. (B): Overview of the different possible binding motifs for the macrocycle and the structural variations of the macrocycle.
Several binding motifs have been found to be efficient templates for rotaxane formation following this approach, including even a simple mono-amide.[40] The best results are
obtained with templates containing at least two hydrogen bond accepting amide or ester groups: glycylglycine (GlyGly),[41] fumaramides (fum),[39,42-45] succinamides (succ)[37,39,42,46-49] and
succinic amide esters[50] (Figure 1-4B). The isophthaloyl dichloride is tolerant towards
introduction of meta-substituents onto the aromatic ring, such as nitro groups.[37,38]
Substitution of the benzene ring with a pyridine ring[33-37] allows further transformations
after the formation of the rotaxane, including protonation[36] and alkylation.[51]
The rotaxanes described in this thesis are Leigh-type hydrogen-bonded rotaxanes. These rotaxanes exhibit a great diversity in cooperative and multipoint hydrogen bonding between their components. Due to the tolerance of the used clipping reaction (Figure 1-4) towards different binding motifs and macrocycles, it is possible to tune the intercomponent hydrogen-bonding interactions, simply by selecting the appropriate components with the desired binding properties. The diverse binding can be traced back to the structural characteristics of the binding template and the macrocycle. The binding strengths depend on the hydrogen bond acceptor strengths, the spatial arrangements of the hydrogen bond accepting C=O and donating NH groups of the macrocycle and the binding motif, and on the conformational flexibility of the components.
Hydrogen Bonding Donor-Acceptor Strength
The acceptor strength of the binding motif can be modified by incorporation of electron-donating or withdrawing moieties on the amide nitrogen atoms. For example, additional alkylation of the amide groups results in decreased hydrogen bond affinity of the thus obtained tertiary amides compared to the secondary amides. Replacement of an amide group by an ester group is another way to influence the acceptor strength; in this case the hydrogen bonding becomes weaker.
Also other types of binding motifs with hydrogen bonding groups other than amides are capable of binding the macrocycle. An important category of such acceptors is that of functional groups containing negatively charged oxygen atoms. For instance, nitrones[52] and squaraines[24,34,35,53] are found to possess excellent affinity towards the macrocycle and have
been used to template the assembly of rotaxanes. Apart from these stabile oxides, another interesting group that can firmly bind the tetraamides macrocycle is that of “transient” oxides. These can be created chemically (e.g. N-oxides,[54] or alkoxy anions[40,48]), electro-chemically or photoelectro-chemically. Examples of the latter two classes are radical anions and dianions of aromatic imides. The corresponding neutral imides are generally poor hydrogen bond acceptors. This principle makes imide-based binding motifs attractive functional units to induce macrocycle translocation in hydrogen-bonded rotaxanes, because the hydrogen bond affinity of the imide can be switched on and off by the interconversion between the neutral and the anion states. Examples of such imide binding motifs (naphthalimides, pyromellitimides and perylene diimides), and their application in macrocycle translocation in rotaxanes are described in Chapters 3 – 6.
The hydrogen bonding affinity of the macrocycle can be tuned by modification of the isophthaloyl moiety. Replacement of a CH-unit in the phenyl ring by a nitrogen atom at the
2-position or the 5-position leads to the “endopyridine”[33-35] or “exopyridine”[36,37] analogs,
respectively. In these pyridines the acidity of the NH protons is increased due to the electron-withdrawing nitrogen atoms; hence these macrocycles bind more strongly to the thread. Alkylation or protonation of the nitrogen of the pyridine leads to a further increase of the hydrogen bonding affinity. Nitro groups on the phenyl ring induce the same effect.[37,38]
Conformational Flexibility
Due to its flexibility, the diamide binding motif can adopt several conformations. These internal degrees of freedom are to a large extent lost in the rotaxane.[55] It is likely that the
unfavorable loss of entropy in going from the flexible binding template to the much stricter conformational requirements of the rotaxane could be partly overcome by pre-organizing the hydrogen bonding sites of the template in a spatial arrangement already suited for binding the macrocycle. This is indeed the case. For example, the spatial arrangements of the amide groups in the succinamide and the fumaramide motifs are the same, but still the binding to the fumaramide motif is stronger. This is due to the rigidity of the latter and is reflected by the higher yields generally obtained with a fumaramide template (97%) compared to a succinamide template (53%).[39] Another illustrative example is provided by
[2]rotaxanes containing both a fumaramide and succinamide binding site. In these systems the macrocycle resides almost exclusively at the fumaramide site (> 95%).[39] These
examples demonstrate that the structural rigidity of the thread binding sites has a major influence on hydrogen bonding with the macrocycle.
The spatial arrangement of the amides in the binding motif determines the geometry of the hydrogen bonds and thus the binding strength. The best fit is obtained if the two amide C=O groups are separated by two bonds, like in the succinamide and fumaramide motifs. X-ray crystal structures show that in this case the macrocycle adopts a chair-like conformation and forms two sets of bifurcated hydrogen bonds with the diamide motif, as is depicted in Figure 1-4.[39] If the distance is larger, the macrocycle NH groups can hardly
reach both amides to form two sets of bifurcated hydrogen bonds. This is illustrated by the low yield of 8% if an adipamide template is used.[39] Another example of the importance of
the spatial orientation of the binding motif is a rotaxane in which complexation of one of the amide N-atoms to a transition metal forces the diamide binding motif to adopt a spatial arrangement which is unfavorable for binding the macrocycle.[50] The result of complexation is that the macrocycle can no longer bind to the amide, due to disturbed multipoint hydrogen bonds interactions.
Another important parameter that governs the hydrogen bond interactions is the conformational flexibility of the macrocycle. The flexibility of the macrocycle gives rise to a variety of binding geometries. In general, the conformation of the macrocycle is to a large extent determined by the binding motif, and vice versa, because the system will try to reach
a compromise between energy-minimizing hydrogen bonds and the extent of conformational strain of the different components. This mutual dependence is illustrated by several examples reported in the literature. In principle, the amides of the macrocycle can function as both hydrogen-bond donor (NH) and acceptor (C=O). In the fumaramide rotaxane depicted in Figure 1-4, the macrocycle adopts a chair-like conformation in order to facilitate two sets of bifurcated hydrogen bonds with the fumaramide template. In rotaxanes with a GlyGly binding motif, a variety of conformations of the macrocycle and different arrangements of the hydrogen bonds can be distinguished.[56] The X-ray crystal structure
reveals that the macrocycle amide groups also act as hydrogen-bond acceptors. In these instances, the C=O groups are rotated and point inwards to form hydrogen bonds with the NH groups of the GlyGly dipeptide motif (Figure 1-5). Macrocycles containing endo-pyridine moieties have been shown to adopt a boat-conformation when attached to a squaraine template (Figure 1-5).[35] The bifurcated arrangement of hydrogen bonds in
fumaramide rotaxanes is confirmed by the X-ray crystal structures.[39] This binding geometry
is also found in rotaxanes containing a succinamide binding motif.[57]
Figure 1-5 Different binding geometries of the tetra-amide macrocycle to different binding motifs.
It should be noted that the crystal structure is to a large extent determined by intermolecular interactions. In solution on the other hand, these interactions play a minor role and the intercomponent and solvent-solute interactions become dominant. Therefore it is likely that more than one conformation and hydrogen-bonding geometry can exist in solution, while generally only one conformation is found in the crystalline solid state.
Hydrogen-Bonded Crown Ether Rotaxanes
Another important group of hydrogen-bonded rotaxanes is that of the secondary ammonium/crown ether type. These systems are characterized by the formation of hydrogen bonds between a protonated secondary amine in the thread and a crown ether macrocycle (Figure 1-6B). These rotaxanes are easy to synthesize with high yields using the
threading and capping approach. Several examples of this type of rotaxanes are reported in the literature, with structural variations in the crown ether macrocycles and threads.[58-61] An
elegant example of a system containing this binding type is the molecular cage-based rotaxane depicted in Figure 1-6B.[60]
(A) (B)
Figure 1-6 (A): Binding motif in rotaxanes based on hydrogen bonding between a secondary ammonium and a crown ether macrocycle. (B): Molecular cage-based rotaxane.[60]
The hydrogen bond interaction between the thread and macrocycle can be manipulated easily by deprotonation of the ammonium ions, which leads to weaker hydrogen bonds with the crown ether. This principle can be used to induce translational motions of the macrocycle along the thread.[13,60-63]
1.2.3
ππππ
-StackingAromatic π-stacking interaction between electron-deficient and electron-rich com-ponents is another important binding type in many rotaxanes. This type of rotaxanes was developed in the research group of Stoddart during the 1980’s. The synthesis of these rotaxanes comprises the formation of a pseudorotaxane host-guest complex by π-π interactions (Figure 1-7), followed by endcapping to obtain the [2]rotaxane. This approach has led to the synthesis of numerous rotaxanes and catenanes. Various electron-deficient and electron-rich compounds have proven to be efficient combinations for creating interlocked structures. An overview of the diversity of electron-rich structures capable of binding the electron-deficient cyclobis(paraquat-p-phenylene) macrocycle (CBPQT4+) is
presented in Figure 1-7A.
Of special interest is the tetrathiafulvalene unit (TTF). Among the summarized electron-rich guests, TTF exhibits the strongest binding affinity towards the CBPQT4+ macrocycle. It also displays a very favorable electrochemical behavior. The TTF2+ cation obtained after
electrochemical oxidation is an extremely weak binding station for the tetra-cationic macrocycle due to electrostatic repulsion. Therefore, TTF and derivates thereof have been used in a large number of rotaxanes with the purpose of inducing movement of the macrocycle along the thread. Applications of this principle are discussed in section 1.3. The
binding type with the electron-rich component in the macrocycle (Figure 1-7B) is also tolerant towards several structural modifications. For example, substitution of the phenyl ring with other aromatic ring systems, e.q. 1,5-naphthalene,[70,71,89] can be applied to tune the
strength of the intercomponent interactions. Other examples of electron-deficient guest are the N,N-dialkyl-2,7-diazapyrenium dication[84] and naphthalene diimides.[89] Besides π
-stacking interactions, due to the presence of the crown ether part, this macrocycle is also capable of forming stable host-guest complexes with secondary ammonium ions through hydrogen bonding (see also Figure 1-6). This dual binding mode is illustrated by the rotaxane depicted in Figure 1-9.
(A) (B)
[64,65] [66-72] [65,66,68-75] [76-79]
[80] [81,82] [65,83] [84]
[85,86] [87,88]
Figure 1-7 Two different types of binding motifs in rotaxanes based on π-stacking. (A): Overview of the different electron-rich guests capable of binding the electron-deficient cyclobis(paraquat-p-phenylene) macrocycle. (B): Electron-rich dibenzo-crown ether macrocycle and examples of electron-deficient guests. The reference numbers are given between square brackets.
1.2.4 Hydrophobic Interactions
The most common example for hydrophobic interaction driven formation of rotaxanes is provided by cyclodextrins. Cyclodextrins are cyclic oligosaccharides, characterised by a hydrophilic exterior and a hydrophobic interior. The hydrophilic nature of the exterior is due to the presence of hydroxyl groups, while the aliphatic carbons are located in the cavity, which is the origin of the hydrophobicity. The two most familiar and frequently used cyclodextrins for rotaxane synthesis are α-cyclodextrins and β-cyclodextrins. The property
of cyclodextrins of forming inclusion complexes with various substrates has led to the synthesis of numerous rotaxanes, with different axles.[12,90-95] An example of a rotaxane based
on hydrophobic interactions with cyclodextrins is shown in Figure 1-12.
1.2.5 Transition Metal Coordination
Due to the variety in coordination number and geometry, transition metal ions are powerful templates in various reactions where a specific alignment of reactants or reaction intermediates is required to obtain selectivity. Transition metal ion templating strategies for the synthesis of interlocked molecules were developed during the 1980’s by the research group of Sauvage.[96] Sauvage used the tetrahedral coordination geometry of a CuI ion which
provides the basis for a correct alignment of the axle within the macrocycle. An example of such a rotaxane synthesized with the assistance of a transition metal ion is depicted in Figure 1-10. In the course of the years, several other transition metals have been employed in this approach, including iron,[20] cobalt,[20,97] copper,[98] ruthenium,[99] palladium[16,100,101] and
gold.[102]
An interesting development in this field is the so-called active-metal template approach. In this approach the transition metal is not only used to assist the correct alignment of the thread precursors in the macrocycle cavity, but it also functions as an active catalyst for the coupling of the precursors. Examples of such couplings are CuI catalyzed azide-alkyne
1,3-cycloadditions,[103] PdII catalyzed coupling of terminal alkynes[104] and PdII catalyzed oxidative
Heck cross-couplings.[105]
1.3
Intercomponent Mobility in Rotaxanes
Initially, the design and synthesis of rotaxanes were purely performed because of the challenge to develop synthetic approaches for the construction of these appealing structures. However, in the course of the past two decades, the perspective on interlocked molecules, and in particular rotaxanes, has undergone a transformation from novel to functional molecules. The interlocked nature of rotaxanes has been recognized as a tool for the expression of functionalities that are not achievable with “ordinary” molecules.
This idea was prompted by the interesting property of mechanically interlocked molecules that they offer a unique set of additional degrees of freedom, arising from motions of the components with respect to each other. Three different types of intercomponent motions can be distinguished in rotaxanes. The macrocycle can undergo translocation along the axle, the so-called shuttling motion. The rotational motion of the macrocycle along the axle is referred to as pirouetting. Also pivoting motions, changing the angle between the axle and the plane of the macrocycle, are possible.
The position of the macrocycle in rotaxanes is determined by the relative strength of the possible intercomponent interactions with the different parts of the thread. In rotaxanes
containing two binding sites (also referred to as binding stations), an equilibrium will exist between the states in which the ring is bound to station 1 or to station 2, depending on the Gibbs free energy difference (∆G) of both co-conformers (Figure 1-8). As illustrated in the previous section, in appropriately designed systems the equilibrium can be adjusted, simply by incorporating different types of binding sites with different binding strengths. The system will minimize its energy by adopting a co-conformation in which the macrocycle predominantly resides on the station with the highest binding strength, i.e. corresponding to the lower Gibbs energy.
Figure 1-8 Macrocycle shuttling in a rotaxane containing two different binding stations, induced by an external stimulus.
Binding to the other station can be made favorable after modification by an external stimulus of station 1, leading to a relatively lower binding affinity, or station 2, leading to a relatively higher binding affinity. This external stimulus can induce changes in geometric configuration (e.g. E-Z isomerization) or electronic arrangement (e.g. electrochemical reduction) of the binding stations, or changes in the environmental properties (e.g. polarity, temperature) which influence the noncovalent intercomponent interactions. In both cases, after modification the energetically favorable situation is the macrocycle residing at station
2*. This situation is reached by translocation of the macrocycle. The rate of the translocation process is determined by the Gibbs energy of activation (
∆
G‡). The height of this free energy barrier is influenced by internal (e.g. attractive forces between the macrocycle and station 2) and external factors (e.g. the polarity and viscosity of the medium). If the stimulus has altered the relative binding strength of the station irreversibly,the endpoint is co-conformer C. If, on the other hand, the change is reversible, the original stations can be restored. The macrocycle then shuttles back to its initial position, completing the shuttling cycle. A rotaxane in which reversible ring translocation between two distinct binding stations is possible, is referred to as a molecular shuttle.
A wide variety of activation methods, comprising chemical, electrochemical and photonic stimuli, can be employed to modulate the noncovalent interactions. The suitability of these stimuli depends much on the nature of the binding stations and the interactions involved. Chemical stimuli may be less elegant from a practical point of view, because they require manipulation with solvents. In the majority of molecular shuttles designed so far, the shuttling motion is made to occur by a change in redox states of the binding station. Redox states can be changed electrochemically or photochemically. Also, (reversible) photoisomerization is an attractive approach to induce and control the macrocycle shuttling. In the next paragraphs some examples of chemically, electrochemically and photoinduced shuttling in rotaxanes are discussed.
1.3.1 Chemically Driven Molecular Shuttles
The modification of the relative binding affinities of different stations on the thread and subsequent macrocycle translocation can be achieved by using a chemical reaction. Examples of chemical reactions suitable for reversible activation of rotaxane-based shuttles are formation of imine bonds[106] and Diels-Alder reactions.[107] However, the most often
used chemical approach for the reversible activation of rotaxane-based shuttles is protonation-deprotonation of secondary amines in rotaxanes containing crown ether macrocycles.[48,60,61,76,82,108-110] A spectacular example of such a pH-switchable system is the
[4]rotaxane depicted in Figure 1-9. [77,78]
This system contains thrice the features of a previously reported pH-switchable [2]rotaxane.[76] It contains three legs, each containing two different types of recognition
sites, a dialkylammonium centre (―RR’NH2
+―) and a bipyridinium unit (BIPY2+). The three
2,3-naphtho[24]crown-8 macrocycles fused onto a hexaoxytriphenylene core can interact with ― RR’NH2
+― or the BIPY2+ stations via hydrogen bonding or π-stacking, respectively.
The position switching of the platform containing the three [24]crown-8 macrocycles from the ―RR’NH+― station to the BIPY2+ stations was induced chemically by deprotonation of
the dialkylammonium stations using a strong phosphazene base (t-BuNP(NMe2)3). In the
deprotonated state (―RR’NH―) the hydrogen bonding interaction is lost and the macrocycles move to the BIPY2+ where donor-acceptor interactions and hydrogen bonding
become stabilizing. The shuttling is fully reversible, upon protonation of the ―RR’NH― with trifluoroacetic acid, the platform moves back to its original position.
Figure 1-9 Base-acid controlled operation of a molecular elevator. Upon deprotonation of the dialkylammonium ions, the platform switches position to bind with the energetically more attractive BIPY2+ stations. The picture was redrawn from reference [77].
The coherent movement of the three macrocycles in this rotaxane resembles the operation of an elevator. The shuttling in both directions was monitored with 1H NMR, cyclic voltammetry and UV-Vis spectroscopy.
1.3.2 Electron-Driven Molecular Shuttles
Switching between redox states, by electrochemical oxidation and reduction is a versatile tool to induce translational motions in rotaxanes. In rotaxanes containing more than one binding station, it can be employed to modify the initial binding station such that binding becomes energetically less favorable. Alternatively, electrochemical oxidation or reduction can also be used to enhance the binding affinity of an unoccupied station, such that binding to this station becomes more attractive. In both cases, the macrocycle will switch position to form predominantly the co-conformer with the lowest energy. The requirement for the operation of electron-driven molecular shuttles is the presence of at least one redox-active component in the molecule.
Transition metals form one group of such redox-active units. Oxidation or reduction of transition metals can change the affinity towards different ligands. The change of affinity can generally be attributed to a change of the electron configuration, which determines the coordination geometry around the metal centre. This is illustrated by the copper ion based molecular shuttles designed by Sauvage and co-workers;[98,111,112] an example is shown in
Figure 1-10.[98] In these systems, the binding geometry can be switched between a
four-coordinate geometry around a CuI centre to a five-coordinate CuII complex. Upon
electrochemical oxidation of the CuI centre, the metal ion and the bidentate macrocycle
migrate from the bidentate phenanthroline to the tridentate terpyridine station. The rate constants of the shutting process were determined from cyclic voltammetry experiments, values of 0.4 s-1 and 50 s-1 were found for the forward and backward process, respectively.
This switching principle has also been applied for controlling the pirouetting motion of the macrocycle in rotaxanes.[113]
Figure 1-10 A reversible, electron-switchable molecular shuttle based on oxidation and reduction of a coordinated transition metal ion.[98]
Redox-active components based on organic molecules form the second group which has been extensively used in electron-driven molecular shuttles. The majority of these molecules contain an aromatic core or conjugated double bonds. This is no surprise, because due to their ability to accommodate an excess charge by delocalization, the oxidation or reduction potentials of these molecules are well within the potential window in which commonly used solvents are electrochemically inert (e.g. for acetonitrile ca. -2.5 – 3.5 V vs. NHE).[114] In the
literature, a large number of electron-driven shuttles have been reported.
The research group of Stoddart has created several examples in which electrons are used to modulate π-stacking interactions between a binding station and the macrocycle, leading to ring movement. Among the redox-active stations in these examples are benzidines[82] and
tetrathiafulvalenes (TTF).[68,72,74] Upon electrochemical oxidation of these neutral
electron-rich stations, the ring is electrostatically repelled and migrates along the thread to bind with another station further up the axle (e.g. naphthalene, NP). An example of such a system,
which contains the typical features of the Stoddart-type shuttles, is the [3]rotaxane depicted in Figure 1-11.[66,115] In solution, the rings of this shuttle can be switched between the TTF
and NP units by oxidation and reduction of the TTF unit. The movement of the macrocycle from the TTF2+ unit is driven by electrostatic repulsion and the π-donor ability
of the NP station. The return to the recovered TTF station is a thermally activated diffusive process. The shuttling of both macrocycles in solution was confirmed by UV-Vis spectro-electrochemical methods and cyclic voltammetry. The electron-induced switching in rotaxanes between TTF and NP units has also been exploited as the working mechanism in the operation of nanovalves[116,117] and for data storage purposes.[74]
(A)
(B)
Figure 1-11 (A): Reversible electron-driven shuttle that functions as a molecular muscle.[66,115] After oxidation of the TTF units, the tetra-cationic CBPQT4+ macrocycles switch position to bind with the naphthalene stations. Upon back-reduction of the TTF2+ units, the rings move back to their initial position at the TTF station. (B): Schematic representation of the operation of the molecular muscle anchored on a gold-coated flexible silicon cantilever.
This [3]rotaxane is able to deliver work and is one of the rare examples of artificial molecular motors whose performance is expressed on the macroscopic level. The shuttle was anchored to a gold-coated (thickness 20 nm) flexible silicon cantilever via a self-assembly process. The disulfide tethers, which are covalently attached to the macrocycle, form Au-S bonds with the surface, resulting in a monolayer of rotaxanes. Oxidation of the
TTF unit with four equivalents of Fe(ClO4)3 and subsequent reduction with ascorbic acid
causes the coated cantilever to bend and relax (Figure 1-11B). The bending could be detected by measuring the difference in deflection angle of a light beam in the oxidized and reduced state. The fact that the bending can be detected on the macroscopic level is remarkable, because the amplitude of the motion of the macrocycles is only 2.8 nm. The macroscopic physical change is explained by the fact that the bending is a cumulative effect of individual nanoscale movements.
1.3.3 Light-Driven Molecular Shuttles
The most elegant way to operate molecular motors is probably by using photons as fuel. One of the advantages of using photoactivation instead of the chemical or electrochemical methods is that the stimulus can be applied without the need of making physical contact with the sample. In this sense, photons are genuine external stimuli. Other advantages are the high time and spatial resolutions that can be achieved with light excitation: it leads to a fast response and can be performed in small spaces. The mechanism of photoactivation can be divided into two categories which are related to the events following the absorption of a photon.
The excited state can be the endpoint of the activation process if it already has the desired difference in binding affinity compared to the ground state. In this case shuttling occurs in the excited state. Examples of this type are rare, because the lifetime of the excited state is generally too short (due to fast radiative or non-radiative decay) to allow completion of the shuttling motion which generally occurs on a much longer timescale. Therefore, the application of this principle is restricted to systems in which small-amplitude motions occur. The only known example is the hydrogen-bonded rotaxane with an active anthracene-carboxamide chromophore. In this rotaxane, the macrocycle moves towards an amide binding station close to the photo-excited chromophore. The higher affinity for binding to this station is due to a change in charge distribution and a geometrical change of the anthracene-amide in the excited state.[46] The translation occurs in less than 5
nanoseconds and the travelled distance is only ca. three C-C bonds.
Generally, in photoactivation strategies for inducing motions in rotaxane-based shuttles, the excited state only serves as a high-energy intermediate in subsequent processes leading to the desired modified binding station. An example of the first kind is the E-Z isomerization of C=C double bonds in fumaramide[118] or stilbene,[119] induced by direct
UV-excitation or by sensitization with a triplet energy donor. In this case, the formation of the
Z-isomer leads to translation of the macrocycle, which moves to an energetically more attractive binding station located on the thread. Alternatively, the chromophore in the excited state can act as a reactant and undergo a chemical reaction. These photochemical reactions may sometimes require the addition of reagents (e.g. reductants). The activation of the majority of the photo-controlled molecular shuttles is based upon this mechanism.
E-Z Photoisomerization
Photoisomerization of C=C or N=N double bonds are often used photochemical reactions for the operation of molecular shuttles and switches. The mechanism of activation involves change of the binding affinity due to substantial geometric changes upon isomerization. The two main examples of the first type, isomerization of C=C bonds, are fumaramides and stilbenes. E-fumaramides are excellent binding stations for the Leigh-type tetraamide macrocycles, while the Z-isomers bind the macrocycle only poorly. Photoisomerization is therefore an easy tool to reversibly change the binding affinity of fumaramides and has been applied in several cases to induce macrocycle translation.[33,39,120,121]
Stilbenes and azobenzenes (the latter contains N=N double bonds) are often associated with cyclodextrin containing rotaxane shuttles based on hydrophobic interactions. A nice example of a shuttling induced by photoisomerization is depicted in Figure 1-12.[93] This light-driven shuttle contains two isomerizable units, a stilbene and an azobenzene, and two different fluorescent naphthalimide stoppers. The α-cyclodextrin (α-CD) macrocycles of this [3]rotaxane can be switched selectively by selecting the appropriate wavelength of the excitation light. In this way the stilbene or azobenzene can be isomerized selectively and thus each of the macrocycles can be moved individually to the centre part of the thread. The positions of the rings along the thread are signalled by the fluorescence of the two different naphthalimide stoppers. The fluorescence intensity of either station increases drastically if the adjacent α-CD ring moves away in the Z-isomer.
Figure 1-12 A light-driven molecular shuttle.[93] The directions of shuttling motions of the α -cyclodextrin macrocycles can be controlled by selective isomerization of the stilbene or azobenzene unit.
Although photoisomerization is a clean way to activate molecular shuttles, because no auxiliary reactants are required, a drawback can be that a selective and complete conversion to each of the isomers is usually difficult to achieve. This is because the isomers have strongly overlapping absorption bands, which makes the selective excitation of either one of the isomers difficult. Better selectivity can sometimes be achieved by avoiding the singlet excited state, and allowing the isomerization to occur via the triplet state (T1). The T1 levels
of the E and Z isomers can be at distinctly different energies, so the enrichment of the Z-isomer (with the higher triplet state energy) is possible with the use of triplet sensitizers with an energy between those of the two isomers.[122-124] This principle has been applied in
molecular shuttles in an elegant way by incorporating the sensitizer in the molecular structure of the shuttle, close to the isomerizable double bond.[118,119]
Photoinduced Electron Transfer
Photoinduced electron transfer is the most basic photochemical reaction because only a single electron is transferred from a donor to an acceptor (Figure 1-13). The excited state (singlet or triplet) of a chromophore, incorporated in the molecular shuttle, can function as electron donor or acceptor. In appropriately designed systems electron transfer can occur on a very short timescale, ranging from picoseconds to nanoseconds. Also, in the majority of the cases the process is fully reversible.
Due to these characteristics, photoinduced electron transfer can be a powerful tool for fast and reversible generation of radical anions and cations, suitable for the activation of molecular shuttles. Yet, this process has not been exploited very often in photoactivation strategies for inducing large-amplitude and controllable motions in molecular shuttles. The main reason is the rapid decay of the charge-separated state by charge recombination, in intramolecular cases or in contact ion pairs. The lifetime of the activated state might be too short to allow completion of the shuttling process. Therefore, in the few examples that use photoinduced electron transfer as activation process, it is associated with fast shuttling on a timescale of roughly < 100 µs (see the examples below).
Figure 1-13 The mechanism of photoinduced electron transfer from a electron donor (D) to an acceptor (A). D* represents a singlet or triplet excited state.
One of the rare working examples of systems driven by photoinduced electron transfer is based on single-step photoinduced electron transfer from a Ru2+ metal centre to a
bipyridinium binding station (Figure 1-14).[125,126] Upon photoactivation with visible light,
the dibenzo-crown ether ring undergoes displacement from the bipyridinium to the 3,3’-dimethylbipyridinium station with a time constant of 48 µs at 303 K, but only 14% of the macrocycles can complete with the fast charge recombination recovery process (7.7 µs). The charge recombination process could be slowed down by reduction of the Ru3+ using an
auxiliary electron donor. This resulted in a higher shuttling efficiency of 76%.
(A)
(B)
Figure 1-14 (A): Molecular structure of a rotaxane-based molecular shuttle driven by photoinduced electron transfer.[125,126] (B): Operation mechanism.
The lifetime of the charge-separated state can be extended by retardation of the back-electron transfer process. This can be achieved in triad systems in which sequential back-electron transfer steps occur. The dissociation of the radical ion pair is mimicked by making the distance between the ions as large as possible in order to inhibit charge recombination. A drawback of this approach is an inherent one generally associated with multi-step electron transfer processes: the low quantum efficiency. That means that the number of absorbed photons which are actually converted into mechanical movement, will be low. This principle has been proven to be effective to create long-lived charge-separated states in rotaxanes, but its actual application for inducing controlled intercomponent motions
remains a challenge. Recently, a prototype of a light-driven shuttle was presented, based on multi-step electron transfer and containing the required ingredients to function in a ‘stand alone’ manner.[127] The photoinduced shuttling process in this shuttle, containing an array of
electron donors and acceptors, was postulated on the basis of the properties and arrangements of the functional components.
Another approach for extending the lifetime of the charge-separated state, and successfully implemented in molecular shuttles, is by designing systems in which back-electron transfer is spin-forbidden. Following this approach, in our research group a hydrogen-bonded molecular shuttle, based on a naphthalimide chromophore as binding station, has been studied. This shuttle exhibits the fastest shuttling dynamics ever observed for controlled large-amplitude macrocycle shuttling in artificial molecular shuttles.[128] The
key element in this system is a naphthalimide chromophore that undergoes rapid intersystem crossing to the triplet state after which it is reduced by an electron donor present in solution. The back electron transfer is a slow process because of its spin-forbidden nature. Due to this, the dissociation of the ion-pair is very efficient. The lifetime of the naphthalimide radical anion (which functions as a new binding station for the macrocycle) is very long: in the range of hundreds of microseconds. This is long enough to allow macrocycle shutting; this process occurs on a timescale of several microseconds. The activation mechanism, shuttling dynamics and its detection are described in Chapter 2. The performance of these shuttles in response to environmental variations (viscosity) and some structural modifications is the scope of Chapters 2 and 6, respectively.
1.4
Scope of this Thesis
This thesis describes the operation of molecular shuttles based on hydrogen-bonded rotaxanes. The main topic is the correlation between the intercomponent mobility in these shuttles and the internal noncovalent interactions between their components, and external interactions with the surrounding medium. So, the main objective of this thesis is to provide an answer to the question: How does the structure of the rotaxane components and its environment influence the intercomponent dynamics? In order to answer this question, the shuttling process in photo- and electron-driven hydrogen-bonded rotaxane shuttles was studied after structural modification of their components and direct surroundings.
Chapter 2 describes the effect of viscosity on the photoinduced shuttling in a naphthalimide-based rotaxane. This molecular shuttle is operated in viscous polymer solutions and the retardation effects on the shuttling dynamics are evaluated in the context of several theoretical and empirical models that describe the viscosity effect on the transport properties of solutes.
The topic of Chapters 3 – 5 is hydrogen-bond interaction in rotaxanes and its effect on the macrocycle switching behavior. The influence of these interactions on the co-conformer
distribution is evaluated in terms of hydrogen-bond affinity of the binding stations towards the macrocycle. For this purpose, the macrocycle shutting in [2] and [3]rotaxanes with the same basis structure, namely consisting of a succinamide and a redox-active aromatic imide station connected via an alkyl spacer, was investigated with infrared and UV-Vis spectroscopy, and with cyclic voltammetry. Three different imides are incorporated in the rotaxanes in these studies: naphthalimide (ni, Chapter 3), pyromellitimide (pmi, Chapter 4) and perylene diimide (pdi, Chapter 5). The basic principles of detection of the shuttling by means of infrared spectroscopy are established in Chapter 3.
In Chapter 6, the mechanism of the photoinduced shuttling dynamics in naphthalimide-based molecular shuttles is discussed. The activation parameters of the shuttling process in structurally modified naphthalimide rotaxanes are used as basis to propose a model for the shuttling process. The structural variations comprise the distance between the binding stations and the binding affinity of the naphthalimide station. The consequences of these variables for the activation parameters of the shuttling process were surprising, but could be explained in the context of the proposed model.
Chapter 7 describes the macrocycle switching in a molecular shuttle containing a functional fullerene stopper. Spectroscopic and cyclic voltammetry data revealed an unexpected shuttling behavior which can be ascribed to the presence of the fullerene stopper.
The work described in Chapter 8 is to some extent a continuation of Chapter 2. The idea is to use hydrostatic pressure to tune the viscosity of a medium suitable for the operation of imide-based molecular shuttles. In contrast with the approach used in Chapter 2, this alternative approach guarantees a microscopically homogenous increase of the viscosity. In this chapter some preliminary experiments are described with a home-built high-pressure set-up, which unfortunately was not available in time to realize the planned measurements. The preliminary experiments at least allowed to obtain a picture of the polarity changes of alkyl nitriles associated with the increase of pressure, which are shown to counteract the expected effect of viscosity.
1.5
References
[1] Breault, G. A.; Hunter, C. A. and Mayers, P. C. Supramolecular Topology. Tetrahedron 1999,
55, 5265-5293.
[2] Carina, R. F.; Dietrich-Buchecker, C. and Sauvage, J. P. Molecular Composite Knots. J. Am.
Chem. Soc. 1996, 118, 9110-9116.
[3] Brüggemann, J.; Bitter, S.; Müller, S.; Müller, W. M.; Müller, U.; Maier, N. M.; Lindner, W. and Vögtle, F. Spontaneons Knotting - from Oligoamide Threads to Trefoil Knots. Angew.
[4] Guo, J.; Mayers, P. C.; Breault, G. A. and Hunter, C. A. Synthesis of a Molecular Trefoil Knot by Folding and Closing on an Octahedral Coordination Template. Nat. Chem. 2010, 2, 218-222.
[5] Lukin, O. and Vögtle, F. Knotting and Threading of Molecules: Chemistry and Chirality of Molecular Knots and Their Assemblies. Angew. Chem. Int. Ed. 2005, 44, 1456-1477.
[6] Ibukuro, F.; Fujita, M.; Yamaguchi, K. and Sauvage, J. P. Quantitative and Spontaneous Formation of a Doubly Interlocking [2]Catenane Using Copper(I) and Palladium(II) as Templating and Assembling Centers. J. Am. Chem. Soc. 1999, 121, 11014-11015.
[7] McArdle, C. P.; Vittal, J. J. and Puddephatt, R. J. Molecular Topology: Easy Self-Assembly of an Organometallic Doubly Braided [2]Catenane. Angew. Chem. Int. Edit. 2000, 39, 3819-3822. [8] Pentecost, C. D.; Chichak, K. S.; Peters, A. J.; Cave, G. W. V.; Cantrill, S. J. and Stoddart, J. F.
A Molecular Solomon Link. Angew. Chem. Int. Edit. 2007, 46, 218-222.
[9] Peinador, C.; Blanco, V. and Quintela, J. M. A New Doubly Interlocked [2]Catenane. J. Am.
Chem. Soc. 2009, 131, 920-921.
[10] Loren, J. C.; Yoshizawa, M.; Haldimann, R. F.; Linden, A. and Siegel, J. S. Synthetic Approaches to a Molecular Borromean Link: Two-Ring Threading with Polypyridine Templates. Angew. Chem. Int. Edit. 2003, 42, 5702-5705.
[11] Cantrill, S. J.; Chichak, K. S.; Peters, A. J. and Stoddart, J. F. Nanoscale Borromean Rings.
Acc. Chem. Res. 2005, 38, 1-9.
[12] Dawson, R. E.; Lincoln, S. F. and Easton, C. J. The Foundation of a Light Driven Molecular Muscle Based on Stilbene and α-Cyclodextrin. Chem. Commun. 2008, 3980-3982.
[13] Fang, L.; Hmadeh, M.; Wu, J. S.; Olson, M. A.; Spruell, J. M.; Trabolsi, A.; Yang, Y. W.; Elhabiri, M.; Albrecht-Gary, A. M. and Stoddart, J. F. Acid-Base Actuation of [c2]Daisy Chains. J. Am. Chem. Soc. 2009, 131, 7126-7134.
[14] Clark, P. G.; Day, M. W. and Grubbs, R. H. Switching and Extension of a [c2]Daisy-Chain Dimer Polymer. J. Am. Chem. Soc. 2009, 131, 13631-13633.
[15] Jiménez, M. C.; Dietrich-Buchecker, C. and Sauvage, J. P. Towards Synthetic Molecular Muscles: Contraction and Stretching of a Linear Rotaxane Dimer. Angew. Chem. Int. Edit.
2000, 39, 3284-3287.
[16] Fuller, A. M. L.; Leigh, D. A. and Lusby, P. J. One Template, Multiple Rings: Controlled Iterative Addition of Macrocycles onto a Single Binding Site Rotaxane Thread. Angew. Chem.
Int. Edit. 2007, 46, 5015-5019.
[17] Wu, J.; Leung, K. C. F. and Stoddart, J. F. Efficient Production of [n]Rotaxanes by Using Template-Directed Clipping Reactions. Proc. Natl. Acad. Sci. U. S. A. 2007, 104, 17266-17271. [18] Harada, A.; Hashidzume, A.; Yamaguchi, H. and Takashima, Y. Polymeric Rotaxanes. Chem.
Rev. 2009, 109, 5974-6023.
[19] Klotz, E. J. F.; Claridge, T. D. W. and Anderson, H. L. Homo- and Hetero-[3]Rotaxanes with Two π-Systems Clasped in a Single Macrocycle. J. Am. Chem. Soc. 2006, 128, 15374-15375. [20] Prikhod'ko, A. I. and Sauvage, J. P. Passing Two Strings through the Same Ring Using an
Octahedral Metal Center as Template: A New Synthesis of [3]Rotaxanes. J. Am. Chem. Soc.
2009, 131, 6794-6807.
[21] Lee, C. F.; Leigh, D. A.; Pritchard, R. G.; Schultz, D.; Teat, S. J.; Timco, G. A. and Winpenny, R. E. P. Hybrid Organic-Inorganic Rotaxanes and Molecular Shuttles. Nature 2009, 458, 314-318.
[22] Goldup, S. M.; Leigh, D. A.; McGonigal, P. R.; Ronaldson, V. E. and Slawin, A. M. Z. Two Axles Threaded Using a Single Template Site: Active Metal Template Macrobicyclic [3]Rotaxanes. J. Am. Chem. Soc. 2010, 132, 315-320.
[23] Raymo, F. M. and Stoddart, J. F. Slippage - a Simple and Efficient Way to Self-Assemble [n]Rotaxanes. Pure Appl. Chem. 1997, 69, 1987-1997.
[24] Lee, J. J.; White, A. G.; Baumes, J. M. and Smith, B. D. Microwave-Assisted Slipping Synthesis of Fluorescent Squaraine Rotaxane Probe for Bacterial Imaging. Chem. Commun.
2009, 46, 1068-1069.
[25] Tokunaga, Y.; Wakamatsu, N.; Ohbayashi, A.; Akasaka, K.; Saeki, S.; Hisada, K.; Goda, T. and Shimomura, Y. The Effect of Pressure on Pseudorotaxane Formation by Using the Slipping Method. Tetrahedron Lett. 2006, 47, 2679-2682.
[26] Harrison, I. T. and Harrison, S. Synthesis of a Stable Complex of a Macrocycle and a Threaded Chain. J. Am. Chem. Soc. 1967, 89, 5723-5724.
[27] Vögtle, F.; Händel, M.; Meier, S.; Ottens-Hildebrandt, S.; Ott, F. and Schmidt, T. Template Synthesis of the First Amide-Based Rotaxanes. Liebigs Ann. 1995, 739-743.
[28] Hübner, G. M.; Gläser, J.; Seel, C. and Vögtle, F. High-Yielding Rotaxane Synthesis with an Anion Template. Angew. Chem. Int. Edit. 1999, 38, 383-386.
[29] Seel, C. and Vögtle, F. Templates, "Wheeled Reagents", and a New Route to Rotaxanes by Anion Complexation: The Trapping Method. Chem. Eur. J. 2000, 6, 21-24.
[30] Johnston, A. G.; Leigh, D. A.; Murphy, A.; Smart, J. P. and Deegan, M. D. The Synthesis and Solubilization of Amide Macrocycles Via Rotaxane Formation. J. Am. Chem. Soc. 1996, 118, 10662-10663.
[31] Leigh, D. A.; Murphy, A.; Smart, J. P. and Slawin, A. M. Z. Glycylglycine Rotaxanes - the Hydrogen Bond Directed Assembly of Synthetic Peptide Rotaxanes. Angew. Chem. Int. Edit.
Engl. 1997, 36, 728-732.
[32] Clegg, W.; Gimenez-Saiz, C.; Leigh, D. A.; Murphy, A.; Slawin, A. M. Z. and Teat, S. J. "Smart" Rotaxanes: Shape Memory and Control in Tertiary Amide Peptido[2]Rotaxanes. J.
Am. Chem. Soc. 1999, 121, 4124-4129.
[33] Bottari, G.; Dehez, F.; Leigh, D. A.; Nash, P. J.; Pérez, E. M.; Wong, J. K. Y. and Zerbetto, F. Entropy-Driven Translational Isomerism: A Tristable Molecular Shuttle. Angew. Chem. Int.
Edit. 2003, 42, 5886-5889.
[34] Arunkumar, E.; Forbes, C. C.; Noll, B. C. and Smith, B. D. Squaraine-Derived Rotaxanes: Sterically Protected Fluorescent near-IR Dyes. J. Am. Chem. Soc. 2005, 127, 3288-3289.
[35] Fu, N.; Baumes, J. M.; Arunkumar, E.; Noll, B. C. and Smith, B. D. Squaraine Rotaxanes with Boat Conformation Macrocycles. J. Org. Chem. 2009, 74, 6462-6468.
[36] Cecchet, F.; Rudolf, P.; Rapino, S.; Margotti, M.; Paolucci, F.; Baggerman, J.; Brouwer, A. M.; Kay, E. R.; Wong, J. K. Y. and Leigh, D. A. Structural, Electrochemical, and Photophysical Properties of a Molecular Shuttle Attached to an Acid-Terminated Self-Assembled Monolayer. J. Phys. Chem. B 2004, 108, 15192-15199.
[37] Leigh, D. A.; Morales, M. A. F.; Pérez, E. M.; Wong, J. K. Y.; Saiz, C. G.; Slawin, A. M. Z.; Carmichael, A. J.; Haddleton, D. M.; Brouwer, A. M.; Buma, W. J.; Wurpel, G. W. H.; Leon, S. and Zerbetto, F. Patterning through Controlled Submolecular Motion: Rotaxane-Based Switches and Logic Gates That Function in Solution and Polymer Films. Angew. Chem. Int.
Edit. 2005, 44, 3062-3067.
[38] Bermudez, V.; Gase, T.; Kajzar, F.; Capron, N.; Zerbetto, F.; Gatti, F. G.; Leigh, D. A. and Zhang, S. Rotaxanes - Novel Photonic Molecules. Opt. Mater. 2003, 21, 39-44.
