A
Supramolecular Catalysis: An Introduction

Matthieu Raynal1 and Piet W. N. M. van Leeuwen2

1 Sorbonne Université, CNRS, IPCM, UMR 8232, 4 Place Jussieu, 75252 Paris, Cedex 05, France

2 LPCNO, Institut National des Sciences Appliquées-Toulouse, Laboratoire de Physique et Chimie des Nano-Objets, Toulouse F-31077, France

A.0 Introduction

During the major part of its history, chemistry as a science was concerned with the separation of mixtures into pure compounds or the constituting elements, while the Greek chimia was interpreted as "forging together," alloying, etc. This key activity of chemistry was underscored by the Flemish mathematician and engineer Simon Stevin at the end of the sixteenth century, who worked a great deal of his life in the Northern provinces of the then Netherlands, by introducing the Dutch word scheikunde, science of separating as he did for many words with Latin/Greek origin.1 Separation of mixtures has retained its importance in chemistry in spite of the tremendous expansion in synthetic methodologies for making new compounds in the most efficient way. This huge synthetic machinery is mainly based on the construction of covalent atom-atom bonds and to a lesser extent on ionic bonds. In the last 50?years, we have witnessed the development of supramolecular chemistry, which is concerned with the weak interactions of unlike molecules in a highly specific manner leading to organized, well-defined agglomerates [1]. The weak interactions include van der Waals forces, hydrogen bonding [2], p-p stacking [3], hydrophobic interactions [4], anion-p interactions [5], cation-p interactions [6], Lewis acid metal-ligand interactions, ion-dipole interactions [7], cation-anion attractions, Dewar-Chatt-Duncanson metal-ligand interactions, charge-transfer interactions, halogen bonding [8], etc. Some covalent bonds also share similar features because of their reversible nature [9]. Several of these forces were discovered in the last decades, but van der Waals interactions were defined already as early as 1873. The importance of non-bonding interactions in living matter was also recognized more than a century ago in 1894, when Emil Fischer described for the first time the lock-and-key interaction between enzyme and substrate [10]. The discovery of hydrogen bonding in 1920 [11] was an important asset (see quote from Pauling [12]) for the later developments of our knowledge on the structures of peptides and DNA, and thus, supramolecular phenomena were abundant and paramount before "supramolecular chemistry" as we know it today started to develop in the 1970s, when the concepts known from biological systems were transferred to synthetic systems. The synthesis of crown ethers and the identification of their complexes with weakly binding metal ions by Pedersen, while working for Du Pont, mark the beginning of supramolecular chemistry, although Pedersen classified his work under coordination chemistry [13]. Two years later, cyclic and open-chain crown ethers were joined by bicyclic cryptands reported by Lehn et al. [14]. Cryptands form very stable complexes with selected ions, provided there is a good match with sizes of the cavities and ions, and orientation of donor atoms, as also holds for crown ethers. Although the molecules were relatively simple, yet the concepts introduced were far reaching already, such as host-guest, selective capturation, catalysis, and the like. The boom really started in the 1970s with the work on spherands by Cram's group [15], an area that was joined by many chemists, and the synthetic version of supramolecular chemistry was born. Spherands, cryptands, etc. on the man-made side and biological systems on the other side have been a source of inspiration for the design (by CPK models!) of numerous host molecules to serve as components in future applications in sensing, catalysis, light harvesting, materials, etc. Pedersen, Lehn, and Cram received the Nobel prize in 1987 for their germinal contributions, and by then, the concepts of supramolecular chemistry had been widely embraced: host-guest chemistry, molecular recognition, self-assembly, folding, to mention just the earliest ones.

Catalysis concerns the phenomenon of a substance, the catalyst, that accelerates a chemical reaction. More precisely, a catalyst is a substance (inorganic, organic, biological) that enhances the rate by which a reaction approaches equilibrium without itself being permanently incorporated. By definition, the catalyst will do so a number of times (the turnover number, TON) within a certain time, expressed as turnover frequency (TOF). A catalyst lowers the free energy barrier of a reaction by interacting with the substrates and bringing them together (if more than one reactant is involved). Catalyzed intramolecular reactions and simple substitution reactions may follow the same, if accelerated, pathway as the uncatalyzed reaction, but especially metal catalysts and organocatalysts may follow a completely different pathway [16]. Selectivity is an important additional property of a catalyst system, and this is achieved by a preferential acceleration of the pathway that leads to the desired product.

Catalyst-substrate interaction may involve the broad range of non-bonding interactions mentioned earlier in our brief introduction to supramolecular forces, but it may also include temporary covalent bonds between substrates and catalysts. As a general characteristic, catalyst and substrate "self-assemble," and thus, it would seem that catalysis is always a supramolecular phenomenon. One may ask the question then what is meant by supramolecular catalysis. While this appears a tough question, all catalysis specialists must know the answer, as out of 40+ invited contributors to this book, only one invitee had doubts about whether or not his/her catalytic chemistry would belong to this field. Lehn, who coined the name supramolecular catalysis in 1983, described it initially as "molecular catalysis" taking place in a "supermolecule," i.e. a host-guest complex in which the host carries the catalyst, inside or at the rim of the cavity [17]. In 2008, taking into consideration that non-covalent interactions have been incorporated in numerous catalytic systems without being formally involved in substrate binding, the scope of the discipline has been expanded [18]. "Supramolecular catalysis" was thus referred to as any catalytic system that contains supramolecular interactions not included in the "basic" catalytic reaction, which, admittedly, leaves the matter ill-defined in a number of cases. Later [19], some of the key features expected for a catalyst to be ranked as "supramolecular" have been specified: (i) a certain degree of design has to be implemented in its structure such as recognition unit(s) dedicated to bind any of the members of the catalytic reaction, and (ii) assembly of the catalytic subunits must preferably operate by in situ mixing of the different building units. A catalyst of any nature (organometallic, organic, enzymatic) and operating through any activation mode will be defined as "supramolecular" as long as it fulfils one of the aforementioned requisites. This is the case of all catalytic systems present in the 41 chapters of this book.

Man-made supramolecular catalysis started with host-guest chemistry mimicking enzyme catalysis, with the simplified characteristic that the host containing the catalyst will convert the guest to the product, the guest being complexed via a lock-and-key principle in the host. The reactions studied were often taken from known enzymatic reactions, such as ester hydrolysis, aldehyde condensations, but also Diels-Alder reactions have also led to nice examples of accelerations and selectivity changes [20]. Most of these reactions also occur without catalyst, and the effectiveness of the supramolecular catalyst is usually measured in this field by comparing its rate with that of the uncatalyzed reaction, often present as a background reaction. The rate accelerations often stem from enhancements of local concentrations or decrease of the entropy of activation, as most of man-made supramolecular catalysts of the host-guest type do not show the typical properties of enzymes, highly active in creating "activated" complexes [21]. Organometallic catalysts cannot be compared with uncatalyzed reactions, as alkene polymerization, hydrogenation, and hydroformylation have no uncatalyzed counterparts apart from radical mechanisms in several instances. The performance of these catalysts is compared with those of simple, routine catalyst systems for the same reaction. Breslow introduced the use of natural host molecules, viz. cyclodextrins equipped with catalytically active groups as hosts for supramolecular catalysts [22]. Around 1950, 20 years well before the concepts of supramolecular chemistry and catalysis were recognized in the early 1970s, Friedrich Cramer, in his work on inclusion compounds, discovered that molecules incorporated in cyclodextrins may exhibit unusual reactivities [23]. He reported not only catalytic reactions taking place in cyclodextrin,...