IntroductionSupramolecular chemistry is poised at a fascinating moment in its history. Advances in synthesis, structural techniques and computing allow us to devise and prepare complex systems at will, study their structures and dynamics in exquisite detail, and rationalise the observations afterwards. So why, amongst the myriad of new supramolecular building blocks and arrays, do we see so few effective catalysts? How is it that we can apparently understand so much and yet fail the practical test of producing even rudimentary catalysis in any reliable way?There are several motivations for attempting supramolecular catalysis: the most profound is probably that the essence of chemistry is to tame the molecule and so demonstrate our mastery over matter. It is only when we can predict behaviour and then demonstrate in the laboratory the accuracy of our prediction that we can truly claim to understand. In supramolecular chemistry, success and failure depend on the delicateÐand still unpredictableÐbalance between weak and opposing noncovalent interactions; in the absence of reliable prediction, we usually fail. Yet we know it can be done: enzymes represent the highest expression of chemical catalysis, and therefore they are a key source of inspiration for supramolecular chemists. They achieve astonishing selectivities and catalytic efficiencies by deploying intermolecular forces to guide captured substrates very precisely along a reaction pathway towards the transition state and beyond. Additional binding interactions in the transition state ensure that the activated complex of the reaction is stabilised to a larger extent than the enzyme ± substrate complex itself, so enzymes can be thought of as complementary in structure to the transition state of the reaction they catalyse. These binding properties, coupled with catalytic functionalities strategically placed within the enzyme active site, decrease the activation energy for reaction. If enzymes can achieve all this, surely so can we? In this brief review I attempt to highlight some recent successes, [1] explore why success is so elusive, and suggest some likely directions for future exploration.
DiscussionSuccessful design approaches: Figure 1 summarises the types of reactions one might hope to catalyse or influence. The simplest, and therefore the most common, are those that operate on a single substrate, catalyzing a chemical transformation such as an oxidation or a ring opening or closing reaction (Figure 1a). In such reactions, product binding is not likely to be stronger than substrate binding and catalytic turnover should be achievable except where the host molecule becomes covalently modified in the process. In practice it is proving much easier to achieve regio-or stereoselectivity in such reactions, by directing the substrate along one pathway rather than another, than it is to achieve catalytic turnover. For example, ring opening of cyclic phosphodiesters has been controlled in a specific direction (Figure 2) by Breslow, [2] who used modified cyclodextrin...