Self-assembly in Supramolecular Systems

By Leonard F. Lindoy, Ian M. Atkinson

The Royal Society of Chemistry

Copyright © 2000 The Royal Society of Chemistry
All rights reserved.
ISBN: 978-0-85404-512-9

Contents

Chapter 1 Self-assembly: What Does it Mean?, 1,
Chapter 2 Intermolecular Interactions: The Glue of Supramolecular Chemistry, 7,
Chapter 3 Hydrogen-bonded and π-Stacked Systems, 19,
Chapter 4 Rotaxanes, 47,
Chapter 5 Catenanes, 87,
Chapter 6 Metal-directed Synthesis – Rotaxanes, Catenanes, Helicates and Knots, 119,
Chapter 7 Further Metal-containing Systems, 185,
Subject Index, 220,

CHAPTER 1

Self-assembly: What Does it Mean ?

1.1 Introduction

Supramolecular chemistry – broadly the chemistry of multicomponent molecular assemblies in which the component structural units are typically held together by a variety of weaker (non-covalent) interactions – has developed rapidly over recent years. ‘Typically’ is used since, in a considerable number of systems, metal-donor bonds – often essentially covalent in nature – have also been employed to ‘stitch’ together organic components into larger assemblies. Such metal-linked assemblies will be treated as part of the supramolecular realm in the present work (although not employed here, perhaps ‘supermolecular’ is a better term for this category).

With the development of supramolecular chemistry, there has been a concomitant shift in the mind-set of chemists working in the area. This has involved a change in focus from single molecules, often constructed step by step via the formation of direct covalent linkages, towards molecular assemblies, with their usual (see exception above) non-covalent weak intermolecular contacts. This change in focus is nicely encapsulated in Lehn’s description of supramolecular chemistry as ‘the designed chemistry of the intermolecular bond’.

As a consequence of the intense interest in the field, a very large number of synthetic supramolecular systems have now been synthesised, with many of the (non-polymeric) systems ranging in size from around a nanometre or so to tens of nanometres. Quite often, innovative design features have been required to achieve the desired structures – with the design and synthesis of individual systems often representing a very considerable intellectual and practical achievement. The field remains an exciting and fast moving one that continues to produce a range of new materials; many of which are endowed with aesthetically pleasing structures as well as unusual properties. The latter, for example, may include novel redoxactive, photoactive, conductive (including superconductive) and non-linear optical behaviour. Clearly, the area is one that continues to show considerable promise for underpinning the development of molecular scale components and devices, including opto-electronic devices. The promise of useful molecular devices remains a motivation for the continuing widespread interest in the field.

Much of the work in supramolecular chemistry has focused on molecular design for achieving complementarity between single molecule hosts and guests. Besides complementarity, recognition, self-assembly, preorganisation and even self-replication represent important ‘key words’ in the armoury of the supramolecular chemist. As a consequence, the practice of supramolecular chemistry tends to be a somewhat interdisciplinary activity, often requiring knowledge of a range of appropriate chemical, physical and biochemical procedures and techniques. Indeed, aspects of supramolecular chemistry now impinge on virtually all of the chemistry sub-disciplines.

Apart from the special case where metal ions are used as the ‘glue’, central to the supramolecular field is the use of a variety of weaker (non-covalent) interactions – including hydrogen bonding, π — π stacking, dipolar interactions, van der Waals forces and hydrophobic interactions – to hold molecular components together. These are the same forces that Nature uses to bind its molecular assemblies. Indeed, much of the activity in the area aims to mimic (but not necessarily copy directly) the way that Nature goes about things.

Creativity and challenge are, by necessity, key ingredients in any effort to devise and synthesise totally synthetic molecular systems that function like biological systems. To achieve such an aim, the elements of molecular recognition, self-assembly and (ultimately) self-synthesis, all ubiquitous in biology, need to be mastered. Further, the product of such a synthesis should be capable of being functionally active if it is truly to match the behaviour of a natural system. The work discussed in this and subsequent chapters documents the progress made, across a broad front, towards this goal. While some quite beautiful examples of self-assembled synthetic systems have now been produced (very often in good yield and under mild conditions), in general there is still a long way to go before individual systems match the biological ones in both subtlety and function. Therein lies the challenge! Indeed, the entire synthetic supramolecular enterprise so far tends to be dominated by the interaction of relatively simple molecular components that are associated with a limited number of bonding contacts on forming the aggregated product. In contrast, for larger biological assemblies, such as DNA, the tobacco mosaic virus, the enzymes or the respiratory proteins, the respective components are of high molecular weight and are of a quite complex nature. Indeed, the resulting assemblies typically incorporate hundreds, if not thousands, of intermolecular contacts. Many systems of this type are able to reassemble from their separated components; the amount of steric and electronic information stored in the latter, and which must be ‘read out’ during reassembly, is thus very large indeed. Such levels of information storage (and processing) have not yet been approached in the synthetic systems investigated so far. Inevitably, a move towards greater complexity will represent one direction for future development.

How might higher molecular weight assemblies be produced? One (of many possible) modus operandi would be to mimic Nature by stringing together complementary molecular units in predetermined sequences such that two matched strands are produced that will induce self-assembly over many tens, or even hundreds of nanometers of strand length. In such an approach, the characteristics of the final assembly would be set by the nature, positioning in the strand sequence, and frequency of incorporation of the individual complementary molecular units together with their relative ‘cross-strand’ orientations. So far, the use of such a ‘modular unit’ approach for the construction of larger synthetic assemblies has been little exploited.

Of course, the above suggestion ignores the problem of possible supramolecular functionality. Whereas the natural systems are invariably characterised by high functionality in terms of their biochemical roles, in contrast, the functionality of the majority of synthetic assemblies so far investigated has very often been either minimal or, indeed, absent altogether. The incorporation of designed functionality into supramolecular systems will thus undoubtedly continue to attract increased attention in future studies.

1.2 Self-assembly

In the present context, self-assembly may be defined as the process by which a supramolecular species forms spontaneously from its components. For the majority of synthetic systems it appears to be a beautifully simple convergent process, giving rise to the assembled target in a straightforward manner.

It must be emphasised that self-assembly is very far from a unique feature of supramolecular systems – it is ubiquitous throughout life chemistry. Biological systems aside, self-assembly is also commonplace throughout chemistry. The growth of crystals, the formation of liquid crystals, the spontaneous generation of synthetic lipid bilayers, the synthesis of metal coordination complexes, and the alignment of molecules on existing surfaces are but a few of the many manifestations of self-assembly in chemical systems.

A distinctive feature of using weak, non-covalent forces, or for that matter metal–donor bonds, in molecular assemblies is that such interactions are normally readily reversible so that the final product is in thermodynamic equilibrium with its components (usually via its corresponding partially assembled intermediates). This leads to an additional property of most supramolecular systems: they have an in-built capacity for error correction not normally available to fully covalent systems. Such a property is clearly of major importance for natural systems with their multitude of intermolecular contacts. It is a factor that will assume increasing importance for the construction of the new (larger) synthetic systems mentioned previously – as both the number of intermolecular contacts present and overall structural complexity are increased.

Nevertheless, thermodynamic reversibility may prove a disadvantage in particular systems. A classic example is provided by the synthesis of individual rotaxanes (‘bead on a thread’ compounds), discussed in detail in Chapter 4. After threading of the macrocyclic ‘bead’ on to an open-chain component by self-assembly, it has been found desirable to block the reverse (unthreading) pathway by subsequent covalent attachment of branched alkane groups to each end of the ‘thread’ so that any tendency for separation of the components is blocked. That is, the final covalent step is equivalent to tying a knot in each end of the ‘thread’ to stop the ‘bead’ from slipping off.

A related example of this type is given by the facile synthesis of catenanes – supramolecular compounds incorporating mechanically interlocked rings (these are discussed in Chapters 5 and 6). An efficient procedure for synthesising these compounds involves an initial self-assembly process in which an open-chain component is threaded through a macrocyclic component, then orientated such that a subsequent ‘template’ ring-closing reaction results in a structure containing the required mechanically locked rings.

Both of the above examples correspond to assembly processes that involve non-covalent followed by covalent bond formation; of course, other sequences are also possible as is totally covalent self-assembly – strategies discussed by Lindsey as early as 1991.

It needs to be noted that supramolecular systems may also form under kinetic rather than thermodynamic control. This situation will tend to be more likely for larger supramolecular assemblies incorporating many intermolecular contacts, especially when moderately rigid components are involved. It may also tend to occur when metal ions, and especially kinetically inert metal ions, are incorporated in the framework of the resulting supramolecular entity or when, for example, an intermediate product in the assembly process precipitates out of solution because of its low solubility.

There are some difficulties in discussing the concept of self-assembly in the present context. First there is a problem of nomenclature. The host–guest convention, largely derived from simple macrocyclic chemistry, was defined at the outset as representing species with concave and convex binding sites, respectively. The term is now also commonly used in relation to the assembly of supramolecular systems; however, host–guest suggests the presence of two or more ‘unequal’ partners. As such, this original connotation is not necessarily appropriate for describing the components of individual supramolecular systems. In view of this, although sometimes used within the limits of their original meanings, we have very often made no attempt to distinguish hosts from guests in our discussion of molecular assemblies.

Related to the above is the question of size. At what point does a host–guest complex become ‘supramolecular’? Currently, at least in terms of common usage, the answer seems to lie in the eye of the beholder. We will maintain this tradition.

While a considerable amount of data is now available covering the thermody-namic aspects of assembly formation, in general, very little corresponding information is available concerning their detailed mechanisms of formation and dissociation. In particular, little insight exists into the co-operative nature of individual host–guest contacts in directing the course of the assembly processes. As an aid to efficient supramolecular assembly design, this is clearly an area requiring further attention.

1.3 Molecular Recognition

Like self-assembly, molecular recognition processes are found widely throughout both natural and synthetic systems. In particular, molecular recognition is, of course, of central importance in a range of biological and medical areas including, for example, fields as diverse as immunology, pharmacology and genetics. Likewise, it is clearly of fundamental importance to a number of chemical areas. These range from sensor and other analytical applications, through separation science, to aspects of catalysis. Molecular recognition is also crucial to organic templating effects which, in themselves, represent a specialised form of self-assembly process.

As discussed more fully in Chapter 2, it is the degree of electronic and steric complementarity between host and guest that, in general, dictates the magnitude of any molecular recognition that occurs for a given supramolecular system. However, other subtleties may also influence recognition behaviour. When a discrete molecular assembly has a stoichiometry other than 1:1, the existence of more than one host–guest binding domain raises the prospect that co-operative binding may occur – akin to allosteric behaviour in natural systems.

The presence of chirality in host and guest will likewise affect the interaction between them. Chirality can perhaps be seen as a ‘second order’ source of stored structural information that is available for exploitation, often with dramatic effect, for achieving an additional type of host–guest recognition based on ‘handedness’.

The idea of host preorganisation, first proposed by Cram, provides a means for rationalising much guest recognition behaviour as well as the (observed) relative binding strengths of many supramolecular complexes. Essentially, the preorganisation effect implies that the more closely the binding sites of a host molecule are arranged for binding to a guest, the larger will be the association constant for the corresponding host–guest complex. In such a case, there will thus be minimal change in the degrees of conformational freedom of the host on binding to the guest, especially if the host’s backbone structure is rigid. As a consequence, this lower ‘loss of disorder’ of the host is expected to be manifested as a favourable entropic contribution to the overall free energy of host–guest complex formation.

It is instructive to consider the relationship between molecular preorganisation, recognition and self-assembly in relation to the formation of a supramolecular complex. Classically, all three will play a sequential role in complexation. Namely, appropriate preorganisation of the bonding sites in the host for receiving the guest thus predisposes the former for guest recognition. This, in turn, promotes spontaneous self-assembly of the required supramolecular entity. In part, the stability of the guest depends upon the degree of preorganisation of host with respect to guest since the forces acting in the recognition step will also, in essence, continue to act in the product after formation.

Besides the above, the overall stability of a supramolecular complex will clearly also depend upon both the number and the nature of the available binding sites in each component. As discussed in Chapter 2, solvation effects may often also play an important role in determining the strength of host–guest binding.

1.4 Scope of the Present Treatment

Owing to the enormous range of what can now be described as supramolecular chemistry, the present work will chiefly focus on discrete molecular assemblies. In particular, no attempt has been made to discuss the burgeoning field of ‘crystal engineering’. Similarly, unless of particular relevance to the topic under discussion, higher oligomeric and polymeric systems, including most metal cluster and related compounds, have also been excluded from the discussion.

In the next chapter, we will look in more detail at the nature of the weak inter-molecular forces that act between the assembled components of supramolecular systems.

CHAPTER 2

Intermolecular Interactions: The Glue of Supramolecular Chemistry

2.1 Introduction

In the past, synthetic chemists have largely focused on the reaction of molecules rather than on their interaction. However, at least for the supramolecular chemist, this no longer holds true. Increasingly, attention has been given to the formation of molecular assemblies that are held together by a range of relatively weak intermolecular interactions, much as Nature holds its molecular assemblies together. These non-covalent interactions are often dominated by hydrogen bonding and, if aromatic components are present, by π — cloud interactions. Also, other weak forces (both attractive and/or repulsive) may act. These include dispersion, polarisation and charge-transfer interactions – combinations of which make up van der Waals forces. While the use of hydrogen bonding and π — π interactions have tended to receive most attention in the design of individual supramolecular systems, van der Waals considerations are often also of crucial importance.

(Continues…)Excerpted from Self-assembly in Supramolecular Systems by Leonard F. Lindoy, Ian M. Atkinson. Copyright © 2000 The Royal Society of Chemistry. Excerpted by permission of The Royal Society of Chemistry.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.