Platinum Metals Rev., 1998, 42, (3), 100
Metallo-Based Cyclophanes and Catenanes
Towards Molecular-Scale Functional Assemblies
- By Andrew C. Benniston a
- Philip R. Mackie a
- Louis J. Farrugia a
- Simon Parsons b
- William Clegg c
- Simon J. Teat d
aDepartment of Chemistry, University of Glasgow, Scotland
bDepartment of Chemistry, University of Edinburgh, Scotland
cDepartment of Chemistry, University of Newcastle, England
dDaresbury Laboratory, Cheshire, England
The creation of systems that are capable of performing molecular-scale operations is currently under extensive investigation, particularly in the emerging field of nanotechnology. Towards this end, we have been actively involved in the construction and study of the properties of metallo-based assemblies based on electron-deficient cyclophanes and donor-acceptor catenanes. Metallic moieties, such as [Ru(bipy)2 (L)]6+ and [Os(bipy)2(L)]6+, where bipy = 2,2′-bipyridyl and L= tetracationic cyclophane ligand, form an integral part of the molecular structures of these assemblies and in these specific cases create photoactive complexes. The photoinduced electron transfer reactions which occur within these molecules have been extensively studied, especially in relation to the corresponding processes found in natural photosynthetic reaction centres. Here, we present a short review on relevant aspects of this work as well as the potential use of metal-based cyclophanes as binders for aromatics.
The ultimate aim of research in supramolecular chemistry is to control the intermolecular bond so that large-scale molecular aggregates can be constructed (1). Although supramolecular chemistry is still at an early stage of development, chemists are now able to build complex molecular assemblies, where the shape, size and functionality can, to a limited degree, be controlled. Functionality, for instance, could be the simple recognition of a specific molecule, with the resultant interaction being communicated to a remote part of the structure, such as in chemical sensors (2). Topologically intriguing rotaxane and catenane molecules stand out as prime examples of cases where external triggering, for example, by photons, ions or redox changes, can be used to drive large-scale conformational changes, see Figure 1 (3).
For an operational molecular-scale system, the required reaction-functionality is to control the bead positioning in the rotaxane (4), or the ring gliding motion in the catenane. As well as utilising the mechanical properties of rotax-ane/catenane assemblies these structures also contribute to the investigation of electron transfer processes in structurally well-defined models (5).
The two projects described below are some of our attempts to make use of the unique structural features of cationic N, N′ -bipyridinium (viologen) catenane and the counterpart cyclophane assemblies. In these examples, the most important design feature is the incorporation of metal binding sites directly into the molecular structures.
Artificial Models of Photosynthetic Reaction Centres
Without doubt the unique machinery that Nature has constructed to capture light and convert it into energy is a model of ingenuity and efficiency, the workplace for the energy conversion being termed the photosynthetic reaction centre (RC). The RC for purple bacteria, which are typical light-harvesting membrane protein complexes, consists of a “special pair” of bacteriochlorophyll, positioned close to two identical electronic relays (the co-factors) made up of bacteriochlorophyll and bacteriopheophytin, the latter having nearby quinone groups that act as the final destination of the electrons in the electron transfer process. Perhaps one of the most intriguing aspects of photosynthesis is the observed unidirectional electron transfer which occurs despite the almost two-fold symmetry in the RC co-factors (6).
The unidirectionality of the electron transfer is a fascinating phenomenon since discrimination by electrons between the two sets of co-factors is due to differences in the local dielectric environment. In an attempt to mimic RC asymmetry we have begun a programme in which redox differences are introduced into complexes by relatively simple means (7).
Some molecular assemblies which have features of the photosynthetic RC are illustrated in Figure 2. Here, the bipyridyl chelates of ruthenium(II) and osmium(II) function as ideal photosensitisers in the assemblies, with the viologen-based cyclophanes acting as electron-accepting moieties. Due to predetermined design criteria, the redox chemistries of the assemblies are quite different, either because of:
the introduction or removal of donor-acceptor interactions (that is 2 and 1a), respectively,
the exchange of the bound metal ion (that is ruthenium by osmium in 1b).
In particular, wrapping the electron-donating crown ether around one of the electron-accepting units in 2 results in a significant difference in the redox potential (of around 130 mV) between the more “open” (E = −0.32 V) and “wrapped-up” (E = −0.45 V) viologens. This small, but significant, difference in redox potential between the two chemically identical electron acceptors is enough to calculate that quenching of the ruthenium chromophore occurs via preferential electron transfer to the more exposed viologen (8). Although the analogy to the natural photosynthetic reaction centre is rather basic, altering the local environment synthetically could pave the way to controlling the pathways that the electron transfer takes.
It is worth noting that the synthesis of the metallo-assemblies which are shown in Figure 2 is not trivial and relies on the building of molecular units, such as those illustrated in Figure 3. In this case, the two key intermediates, 3 and 4, are required in the preparation of rutheniumcontaining cyclophane 1a and the catenane 2, respectively. The structures of 3 and 4 have been verified by single-crystal X-ray crystallographic studies, see Figure 4. The conformation of 3 reveals that the two bipyridinium units are prearranged in a manner that facilitates cyclisation to form the metallo-cyclophane 1a. It is noteworthy that in the structure of 4 the two nitrogens of the 2,2′-bipyridyl unit are almost transoid (diagonally opposite each other) and not predisposed for chelation. In view of the metal ion binding properties of the ligand, it is reasoned that structural alterations occur which help flip the bipy unit to the required cisoid conformation, where the nitrogens approach the closest together.
Host-guest interactions offer the possibility of selectively and reversibly binding a substrate and reaction centre adjacent to each other, in order to enhance and catalyse reactions, for example oxidation/reduction, see Figure 5. The binding site in Figure 5 has not been specified, but in principle it could be a chelator or macro-cyclic ligand, which would enable a redox-active metal ion to be positioned close to the reactive site on the bound substrate.
As a first step towards creating functional systems, the cyclophane 5 in Figure 6 has been prepared. This incorporates the necessary design criteria so that it can act not only as a metal ion binder but also can encapsulate electron rich aromatics in its central cavity.
X-ray crystallography has also been invaluable here in the unequivocal verification of molecular structure (9). The structure of cyclophane 5 is shown in Figure 7 and clearly shows a slight twisting of the N,N′ -bipyridinium groups and more importantly the cisoid conformation of the bipy unit. It should also be remembered that in the counterpart  catenane structure the bipy unit was transoid, suggesting mat interlocking of the two rings induces steric strain into the molecule.
Although still at the early development stage, the binding of p -dimethoxybenzene, p -phenylenediamine and other functionalised aromatics inside the central cavity of 5 has been verified by using UV-visible and 1H NMR spectroscopies. The obtained binding constants are modest (around 0.5 to 100 M-1) but nevertheless encouraging for future work using metal ion redox reactions to transform substrate functionality.
Future Work Using Terdentate Chelators
In view of the success in the construction of systems incorporating bipy binders, attention has now been turned to the well known terdentate chelator 2,2′:6′,2″-terpyridine (terpy). Ligand 6 represents the first step towards this goal from the functionalisation of the terpy backbone with tolyl groups at the highly unusual 3,3″ locations (10). The chelator readily forms a 2:1 complex with a first row transition metal such as iron (11), with the ligand co-ordinating around the meridian, as expected. The crystal structure of the cation [Fe(6)2]2+ is illustrated in Figure 8.
By functionalisation of the methyl groups we expect to be able to “build from the back” of the chelator in a similar fashion to the bipy cyclophanes and catenanes. In particular, because of the increase in size in moving to the terpy chelator it is expected that larger cavities can be constructed with the potential for binding larger substrates.
The chemistry of catenanes and counterpart cyclophanes which incorporate metal binding groups such as bipy and terpy is still very much at the development stage. Our studies have demonstrated the potential application of these systems as artificial models for photosynthesis, with extensions towards redox-active and photoactive catalysts. In particular, we are now exploring ways to induce greater redox asymmetry into assemblies by altering both the type of electron acceptor and size of the crown ether donor. By attaching high oxidation state ruthenium-oxo species to the cyclophane bipyridyl group the production of oxidation catalysts is also being explored. Of course, our ultimate aim will be to produce working systems that will have a real commercial benefit, though at this stage this is still a long way off.
- 1J-M. Lehn, Supramolecular Chemistry-Concepts and Perspectives, VCH. Weinheim, 1995
- 2L. Fabrizzi and A. Poggi, Chem. Soc. Rev ., 1995, 24, 197
- 3A. C. Benniston, Chem. Soc. Rev ., 1996, 25, 427
- 4A. C. Benniston,, A. Harriman and V. M. Lynch, J.Am. Chem. Soc ., 1995, 117, 5279
- 5J.-P. Sauvage and A. Harriman, Chem. Soc. Rev ., 1996, 25, 41
- 6M. A. Steffen,, K. Loa, and S. G. Boxer, Science, 1994, 264, 810
- 7A. C. Benniston,, P. R. Mackie and A. Harriman, Tetrahedron Lett ., 1997, 38, 3577
- 8A. C. Benniston,, P. R. Mackie and A. Harriman, Angew. Chem. Int. Ed ., 1998, 37, 354
- 9A. C. Benniston,, P. R. Mackie,, L. J. Farrugia,, S. Parsons and D. S. Yufit, manuscript in preparation
- 10A. C. Benniston, Tetrahedron Lett ., 1997, 38, 8279
- 11A. C. Benniston,, L. J. Farrugia,, P. R. Mackie,, W. Clegg and S. J. Teat, manuscript in preparation
This research is supported by the EPSRC (PRM and WC), Nuffield Foundation, Johnson Matthey precious metals loans scheme (ACB) and CLRC (SJT). We would also like to thank the EPSRC mass spectrometry service at Swansea. The above work would not be possible without the fast kinetic spectroscopy supplied by Professor Anthony Harriman (France).