Supramolecular Chemistry: A Brief Overview
The group works in the field of Supramolecular Chemistry, which has been defined as the chemistry beyond the molecule. Consequently, weak intermolecular interactions are our aim and a number of speciallized methods is required to address such weak interactions. The basis of our work is organic synthesis of the components which will later be combined in the supramolecule.
Special strategies are applied utilizing the non-covalent bond already during the synthesis: Hydrogen-bond mediated template effects help for example to position a string-like molecule within the cavity of a macrocycle in a way that the string threads through the wheel. Templates are thus the basis for the synthesis of interlocked molecules. Self-assembly on the other hand is a strategy which utilizes small and simple building blocks with suitably programmed binding sites. They find each other and form a larger, more complex species without the help of a chemist. For self-assembly, reversibility is important, since it ensures error correction. Among the self-assembled species examined in our group, hydrogen-bonded capsules and metallo-supramolecular polygons are the most prominent ones.
In the following, a few introductory remarks into Supramolecular Chemistry are given.
3.1.Molecular Recognition and Host-Guest Chemistry
3.2. Chelate Effects and Preorganization: Entropic Factors
3.3. Self-Assembly and Self-Organization
3.4. Cooperativity and Multivalency
3.5. Template Effects
3.6. Self-Replication and Supramolecular Catalysis
3.7. Molecular Devices and Machines: Implementing Function4. Conclusions
1. Some Historical Remarks on Supramolecular Chemistry
The fundaments of Supramolecular Chemistry date back to the late 19th century, when some of the most basic concepts for this research area were developed. In particular, the following ideas have been introduced quite early:
- coordination chemistry (A. Werner, Zeitschr. Anorg. Chem. 1893, 3, 267)
- the lock-and-key principle (E. Fischer, Ber. Deutsch. Chem. Ges. 1894, 27, 2985)
- cyclodextrins, the first host molecules (A. Villiers, C. R. Hebd, Seances Acad. Sci. 1891, 112, 435)
- receptor chemistry (P. Ehrlich, Studies on Immunity, Wiley, New York 1906): any molecule can only have an effect on the human body, if it is bound (“Corpora non agunt nisi fixata”)
These concepts were refined later. For example, Daniel Koshland formulated the induced-fit concept for binding events to biomolecules which undergo conformational changes in the binding event (D. E. Koshland, Jr., Proc. Natl. Acad. Sci. USA 1958, 44, 98). The induced fit model provides a more dynamic view on the binding event as compared to the rather static key-lock principle and is thus more easily able to explain phenomena such as cooperativity.
Why hasn't it been recognized earlier as a research area in its own? Why did it take more than 40 years from the introduction of the term "Übermolekül" for carboxylic acid dimers to Lehn's definition of supramolecular chemistry (J.-M. Lehn, Pure Appl. Chem. 1979, 50, 871) as the "chemistry of molecular assemblies and of the intermolecular bond"?
A first answer relates to the perception of the scientists involved in this area. As long as chemists accepted the paradigm that properties of molecules are properties of the molecules themselves, while the interactions with the environment are small and - to a first approximation - neglectable, there is no room for supramolecular chemistry as an independent field of research. With an increasing number of examples of the importance of the environment for a molecules properties, a paradigm shift occurred in the late 1960ies. Chemists started to appreciate that their experiments almost always provided data about molecules in a particular environment. It became clear that the surrounding almost always has a non-neglectable effect. Consequently, the intermolecular interactions became the focus of research and a new area was born.
The second answer is not less important, although somewhat more technical in nature. Supramolecules are often weakly bound and highly dynamic. Such features need specialized experimental methods, many of which still had to be developed in the early days of supramolecular chemistry. As observed quite often, the progress in a certain research area - here supramolecular chemistry - depends on the development of suitable methods. An emerging new method on the other hand leads to further progress in this research field, since it opens new possibilities for the experimenters.
Numerous non-covalent bonds form the topic of supramolecular chemistry. The following table summarizes the most important ones.
Type of Non-Covalent Bond |
Range of Strengths (kJ/mol) |
Remarks |
coordinative bond |
usually strong |
close to the character of covalent bonds |
ion-ion interaction |
100 - 350 |
non-directional, distance between charges most important |
ion-dipole interaction |
50 - 200 |
suitable orientation of dipole important for maximizing the forces |
dipole-dipole interaction |
5 - 50 |
relative orientation of dipoles important |
hydrogen bond |
4 - 160 |
directional, strongest, if A...H...B in linear arrangement |
cation-pi interaction |
5 - 80 |
cation located usually above aromatic plane |
pi-pi stacking interaction |
usually weak |
molecules usually don't stack directly on top of each other, but shift slightly (exception: strong
electron-rich/electron-poor complexes) |
Van-der-Waals forces |
< 5 |
interactions between fluctuating dipoles |
hydrophobic effect |
- |
based on a maximization of interaction energies between solvent molecules |
3. Basic Concepts of Supramolecular Chemistry
3.1. Molecular Recognition and Host-Guest Chemistry
If a receptor is able to recognize a molecule, it can distinguish it from other potential guest molecules
with some specificity. An important factor is the complementarity of steric and electronic
properties between both binding partners. The higher the degree of complementarity, the stronger
the interaction between both molecules. Cations and neutral molecules are usually more easily recognized by a
designed host as compared to anions. Historically, cation recognition was developed early on during the development
of supramolecular chemistry due to the combination of two favorable factors: defined coordination geometries and
higher binding energies due to the charge. Highly selective anion recognition is however rather difficult to
achieve and this field is currently under quick progress.
3.2. Chelate Effects and Preorganization: Entropy Factors
A binding event in which one complex forms from two molecules is entropically disfavored. The entropic costs need to be paid from the reaction enthalpy released upon host-guest binding. However, strategies exist which permit to reduce these costs to a minimum.
One approach is to incorporate more than one binding site in one host molecule. When the first bond is formed, the entropic costs of combining two molecules are taken care of. The second and all following binding events between the same two partners will not suffer from this effect again and thus contribute more to the free enthalpy of binding. This effect is called the chelate effect and is long known from coordination chemistry, where ethylene diamine or 2,2'-bipyridine ligands easily replace ammonia or pyridine in a transition metal complexes. Bidentate binding generates rings and the chelate effect depends on their sizes. Optimal are five membered rings as formed by the ethylene diamine or bipyridine ligands discussed above. Smaller rings suffer from ring strain, larger rings need a higher degree of conformational fixation as compared to their open-chain forms and are thus entropically disfavored. The latter argument can be refined. If the same number of binding sites are incorporated in a macrocycle or even macrobicycle, guest binding will again become more favorable, because each cyclization reduces the conformational flexibility for the free host and thus the entropic costs stemming from conformational fixation during guest binding. These effects have entered the literature as the macrocyclic and macrobicyclic effect. Donald Cram developed these ideas into the preorganization principle (D. J. Cram, Angew. Chem. 1986, 98, 1041; Angew. Chem. Int. Ed. 1986, 25, 1039). A host which is designed to display the binding sites in a conformationally fixed way, perfectly complementary to the guest's needs, will bind significantly more strongly than a floppy host which needs to be rigidified in the binding event. This becomes strikingly clear, if one compares conformationally flexible 18-crown-6 with the spherand shown in Figure 1 which displays the six oxygen donor atoms in a preorganized manner. The alkali binding affinities between both host molecules differ by factors up to 10 to the power of 10!
Figure 1: Preorganization does matter. A comparison of 18-crown-6 and the spherand on the right with respect to alkali metal ion binding reveals that the spherand has an up to 10 orders of magnitude higher binding constant.
While discussing entropic effects, it should not be forgotten that there exist examples for enthalpically
disfavored, entropy-driven host-guest binding. This is possible, if the free host contains more than one solvent
molecule as the guests which upon guest binding are replaced by one large guest as discussed for cyclodextrins
above. In this case, a host-solvent complex releases more molecules than it binds and the overall reaction benefits
entropically from the increase in particle number.
#
3.3. Self-Assembly and Self-Organization
Self-assembly (J. S. Lindsey, New J. Chem. 1991, 15, 153; G. M. Whitesides, J. P. Mathias, C. T. Seto, Science 1991, 254, 1312; D. Philp, J. F. Stoddart, Angew. Chem. 1996, 108, 1243; Angew. Chem. Int. Ed. 1996, 35, 1154; C. A. Schalley, A. Lützen, M. Albrecht, Chem. Eur. J. 2004, 10, 1072) is a strategy used by supramolecular chemists to reduce the efforts required for the generation of complex structures and architectures. Instead of tedious multi-step covalent syntheses, simple building blocks are programmed with the suitably positioned binding sites and upon mixing the right subunits, they spontaneously assemble without any additional contribution from the chemist. Several requirements must be met: (i) The building blocks must be mobile, but this requirement is almost always fulfilled with molecules in solution due to Brownian motion. (ii) The individual components must bear the appropriate information written into their geometrical and electronic structure during synthesis to provide the correct binding sites at the right places. Since their mutual recognition requires specificity, self-assembly is a matter of well-preorganized building blocks (see above). (iii) The bonds between different components must be reversibly formed. This means that the final aggregate is generated thermodynamically controlled under equilibrium conditions. This aspect is important, because kinetically controlled processes do not have the potential for error correction and thus usually lead to mixtures. The reversibility of self-assembly processes also results in quite dynamic aggregates prone to exchange reactions of their building blocks.
Self-assembly is ubiquitous in nature and often occurs on several hierarchy levels simultaneously in order to generate functional systems. For example, the shell-forming protein building blocks of the tobacco mosaic virus need to fold into the correct tertiary protein structure before they can be organized around a templating RNA strand. All these processes are mediated by non-covalent forces which guide the formation of secondary structure elements on the lowest hierarchy level. These form the tertiary structure on the next level which displays the necessary binding sites for the assembly of the virus from a total of 2131 building blocks to occur as programmed on the highest level. Other examples for hierarchical self-assembly are multienzyme complexes, the formation of cell membranes with all the receptors, ion channels, or other functional entities embedded into them, or molecular motors such as ATP synthase. Self-assembly is thus an efficient strategy to create complexity and - together with it - function in nature.
Figure 2: Self-assembling "softball". Right: Computer model of the softball bearing the hydroquinone spacer (side chains are omitted). Box (left): Different monomers which form dimers with cavities of volumes between 187 and 313 Å 3 depending on the spacer length. Left: A selection of good guest molecules which can occupy the cavity inside the capsule.
Self-assembly has also been applied to numerous different classes of complexes in supramolecular chemistry. Since we cannot discuss them all here, Figure 2 shows only one example of a capsule reversibly formed from two identical self-complementary monomers which are bound to each other by hydrogen bonding. The two monomers can encapsulate guests in the interior cavity of the capsule. Even more than one guest can be encapsulated, and reactions can be catalyzed inside.
Another term which is often used in the literature as synonymous with self-assembly is self-organization. However,
again, we should be precise with respect to the meaning of the terms we use. One suggestion for definitions would be
to distinguish processes which lead to the thermodynamic minimum and thus lead to chemical equilibria. These processes
should be called self-assembly processes. On the other hand is the broad variety of spontaneous organization which
occurs far away from the thermodynamic equilibrium. Many processes in living organisms are examples for self-organization
in this sense. The major difference between self-assembly and self-organization is that self-assembly occurs even in a
closed system while self-organization can be characterized as a steady state in which a systems remains without falling
into the thermodynamic minimum, because energy is constantly flowing through it. This definition has the advantage that
it makes a clear difference between the two terms. This advantage, however comes at the prize that it is experimentally
difficult to determine which is which by simple criteria.
3.4. Cooperativity and Multivalency
Cooperativity and multivalency are phenomena arising in molecular recognition at hosts with more than one binding site. In order to avoid misunderstandings, one should clearly distinguish the two terms. Cooperativity describes the influence of binding a guest at the host's binding site A on the second binding step occurring at site B of the same host. Cooperativity can be positive, which means that binding strength of the second guest is increased by the first one and the sum of both binding energies is higher than two times the binding energy of the first guest. Cooperativity can also be negative, if the first binding event decreases the binding of the second guest. Many examples for cooperativity are known from biochemistry, the most prominent one certainly oxygen binding at hemoglobin (W. A. Eaton, E. R. Henry, J. Hofrichter, A. Mozzarelli, Nat. Struct. Biol. 1999, 6, 351). This protein is a a 2 : 2 tetramer with four oxygen binding hems as the prostethic groups, one in each subunit. Upon binding the first oxygen molecule to one of the hem groups, conformational changes are induced in the protein tertiary structure which also affect the other subunits and prepare them for binding oxygen more readily. From this example, it becomes clear that cooperativity not necessarily relies on interactions between a multivalent host and a multivalent guest, but that there may well be mechanisms to transmit the information of the first binding event to the second one, even if both are monovalent interactions. The concept of cooperativity has been applied to supramolecular chemistry and was recently discussed in the context of self-assembly (G. Ercolani, J. Am. Chem. Soc. 2003, 125, 16097).
Conceptually related to the chelate effect, multivalency (N. Röckendorf, T. K. Lindhorst, Top. Cur. Chem. 2002, 217, 201; S.-K. Choi, Synthetic Multivalent Molecules, Wiley-Interscience, Hoboken, USA, 2004; A. Mulder, J. Huskens, D. N. Reinhoudt, Org. Biomol. Chem. 2004, 2, 3409; J. D. Badjic, A. Nelson, S. J. Cantrill, W. B. Turnbull, J. F. Stoddart, Acc. Chem. Res. 2005, 38, 723) describes the unique thermodynamic features arising from binding a host and a guest each equipped with more than one binding site. Although sometimes not used in a stringent way in the chemical literature, one should use the term "multivalency" only for those host-guest complexes, in which the dissociation into free host and guest requires at least the cleavage of two recognition sites. The concept of multivalency has been introduced to adequately describe the properties of biomolecules (M. Mammen, S.-K. Choi, G. M. Whitesides, Angew. Chem. 1998, 110, 2908; Angew. Chem. Int. Ed. 1998, 37, 2754). For example, selectivity and high binding strengths in recognition processes at cell surfaces usually requires the interaction of multivalent receptors and substrates. Due to the complexity of many biological systems, limitations exist for a detailed analysis of the thermochemistry and kinetics of multivalent interactions between biomolecules. For example, the monovalent interaction is usually unknown and thus, a direct comparison between the mono- and multivalent interaction is often not feasible. The sometimes surprisingly strong increase of binding energy through multivalency is thus not fully understood in terms of enthalpy and entropy.
Figure 3: Molecular elevator synthesized by utilizing multivalency. The position of the wheel component can be controlled by protonation/deprotonation.
Recently, this concept was applied convincingly to artificial supramolecules. The examination of artificial,
designable, and less complex multivalent systems provides an approach which easily permits to analyze the
thermochemical and kinetic effects in great detail. As an example, the binding of a divalent calixarene ligand
bearing two adamantane endgroups on each arm binds more strongly to a cyclodextrin by a factor of 260 as compared
to the monovalent interaction - a much higher increase than expected for merely additive interactions. If offered
many cyclodextrin hosts on a surface, the binding constant again increases by 3 orders of magnitude (J. Huskens,
M.A. Deij, D.N. Reinhoudt, Angew. Chem. Int. Ed. 2002, 41, 4467; Angew.
Chem. 2002, 114, 4647; A. Mulder, T. Auletta, A. Sartori, A. Casnati, R. Ungaro,
J. Huskens, D. N. Reinhoudt, J. Am. Chem. Soc. 2004, 126, 6627; T. Auletta,
M. R. de Jong, A. Mulder, F. C. J. M. van Veggel, J. Huskens, D. N. Reinhoudt, S. Zou, S. Zapotoczny, H.
Schönherr, G. J. Vancso, L. Kuipers, J. Am. Chem. Soc. 2004, 126, 1577).
Another example of which is shown in Figure 3 (J. D. Badjic, V. Balzani, A. Credi, S. Silvi, J. F. Stoddart,
Science 2004, 303, 1845). A three-armed guest is capable of forming a triply
threaded pseudorotaxane with the tris-crown derivative. Attachment of stoppers at the ends of each arm prevents
deslippage of the axle components. The trivalent interaction increases the yield of the synthesis through favorable
entropic contributions. At the same time, the function of a "molecular elevator" is implemented: Depending
on protonation and deprotonation of the dialkyl amines, the crown ethers move back and forth between two different
stations along the axle.
One way to control the outcome of a reaction is templation. Like in the macroscopic world, a chemical
template organizes reaction partners and thus allows the chemist to control their reactivity to achieve the
formation of a desired product. However, it is almost impossible to give a concise definition of the term
“template” (D. H. Busch, N. A. Stephensen, Coord. Chem. Rev. 1990, 100,
119; R. Cacciapaglia, L. Mandolini, Chem. Soc. Rev. 1993, 22, 221; N. V.
Gerbeleu, V. B. Arion, J. Burgess, Template Synthesis of Macrocyclic Compounds, Wiley-VCH, Weinheim
1999; T. J. Hubin, A. G. Kolchinski, A. L. Vance, D. H. Busch, Adv. Supramol. Chem. 1999,
5, 237; F. Diederich, P. J. Stang (eds.) Templated Organic Synthesis, Wiley-VCH, Weinheim 2000).
Templates span the whole range from biochemistry with its complex apparatus for DNA replication to the formation
of structured inorganic materials to the templated synthesis of macrocycles to the preparation of supramolecular
catalysts - just to name a few examples. Nevertheless, all these have in common that a template must serve different
purposes: (i) It organizes reaction partners for the formation of a desired product whose synthesis cannot be
achieved in the absence of the template. Thus, a template controls reactivity and produces form. (ii) The template
needs to bind to the reaction partners. Molecular recognition is thus a necessary prerequisite for template
syntheses and the binding sites of the components must be complementary to each other. Usually, binding is due
to non-covalent bonds, although examples for covalent templates exist. (iii) The control of reactivity and the
recognition of the reaction partners imply information to be programmed into the template which is transferred
to the product of the reaction.
3.6. Self-Replication and Supramolecular Catalysis
While multivalency, self-assembly, and template effects provide strategies aiming at generating more and more complex architectures, supramolecular chemistry can also be utilized for controlling reactivity and even catalyzing reactions. Closely related to organocatalysis, supramolecular catalysts accelerate reactions by lowering the barriers. The principles by which they fulfil the task are very different. Increasing the local concentration of the reactands by encapsulation is one example (see Figure 2 above), increasing the intrinsic reactivity of carbonyl compounds through hydrogen bonding is another and many more exist.
Figure 4: A minimal self-replicating system. In the presence of template A , the two reactands on the left are organised in a way suitable for a 1,3-dipolar cycloaddition reaction. The pyridineamide part of the template recognizes the acid substituent in the reactand, while the second reactand is recognized by the carboxylic acid incorporated in the template. Particularly interesting is the fact that template A favors its own formation, while the other stereoisomer B is formed only in low amounts.
Originating from the question how the living organisms came into existence, self-replication is a special,
but certainly intriguing case of supramolecular catalysis. If one thinks about the complex ribosome, which
nowadays transscribes genetic information stored in nucleic acids into proteins, which then become involved
in the duplication of DNA, it immediately is clear that this is a much to complex apparatus to self-organize
accidentially at the beginning of life. Instead, much simpler mechanisms must have existed in the early world.
In order to find an answer, several research groups provided evidence that short DNA oligomers are indeed able
to self-replicate in the presence of the appropriate template (L. E. Orgel, Nature 1992,
358, 203; T. Li, K. C. Nicolaou, Nature 1994, 369, 218; D. Sievers,
G. von Kiedrowski, Nature 1994, 369, 221). Later, suitable a -helical peptides
have been shown to self-replicate as well (K. Severin, D. H. Lee, J. A. Martinez, M. R. Ghadiri, Chem. Eur. J.
1997, 3, 1017). In the context of supramolecular chemistry most interesting are
however organic minimal-replicators (A. Robertson, A. J. Sinclair, D. Philp, Chem. Soc. Rev. 2000,
29, 141) which are not based on biomolecules. Figure 4 shows an example for a minimal self-replicating system,
which operates even in a chiroselective way. One given enantiomer of the template catalyzes its own formation, while the
other enantiomer is by and large suppressed.
3.7. Molecular Devices and Machines: Implementing Function
Early supramolecular chemistry certainly focussed on the non-covalent bond and the beauty of structures which
can be generated employing it. This is certainly the case for topologically interesting molecules such as rotaxanes,
catenanes, knotanes and Borromean rings. It also holds for the generation of self-assembling capsules, helicates, or metallo-supramolecular tetrahedra, octahedra and the like. However, the focus has shifted in contemporary supramolecular
chemistry towards the implementation of function into non-covalent architectures. The scope of function is broad and
ranges from light-induced energy and electron transfer processes and molecular wires to switches, molecular
"motors", and devices for the active pH-driven transport of molecules through membranes. This area is too
broad to give a satisfying introduction here and thus, the reader is referred to the literature cited.
The short and simplyfied considerations make clear that one aim of supramolecular chemistry is to mimick natural processes. The above sections deliberately chose examples from biochemistry as well as the multitude of artificial supramolecules in order to point to the relations which exist between both fields. Understanding the details of non-covalent binding is much more difficult in a complex biomolecule, and thus simple model systems provide the basis for a more profound analysis. However, supramolecular chemistry goes beyond merely creating model systems for naturally occurring species. In contrast to biomolecules, supramolecular chemistry can utilize the whole range of conditions achievable, for example with respect to the use of organic solvents, in which many biomolecules would loose their integrity, because they are designed for an aqueous surrounding. Higher or lower temperatures or different pressures can also be applied. Supramolecules may even find their applications under conditions where biomolecules would not have the necessary long-term stability. The implementation of function also aims at new functions which are not realized in nature. In particular, the latter two aspects lead us to the second research are to which supramolecular chemistry significantly contributes: material sciences. Self-assembly, for example, is a strategy to create long-range order and has even been applied to particles on a micro- to millimeter scale (N. B. Bowden, M. Weck, I. S. Choi, G. M. Whitesides, Acc. Chem. Res. 2001, 34, 231).
If one thinks about function, in particular switches, logic gates, and molecular wires, it becomes clear that supramolecular chemistry is also about information processing. However, it is not only its potentially upcoming use in microelectronics, information processing begins at a much more fundamental level. Templates transfer spatial information between molecules; in order to achieve correctly self-assembling species, the building blocks of the assembly need to be programmed with the appropriate binding sites. Information transfer and information processing already starts at the molecular level.
A view back on the last few decades makes perfectly clear: Supramolecular chemistry has become a highly diverse field which requires the interdisciplinary use of a huge variety of methods to answer the scientific questions addressed. Diversity however is not the only challenge for the methods that are needed. The complexity of the architectures meanwhile realized requires sophisticated structure analysis tools. The highly dynamic features of supramolecules need kinetic methods able to address many different time scales. Gathering evidence for the functions implemented is impossible without a sound methodological basis.