The objective of this Account is to summarize the first five years of anion−π catalysis. The general idea of anion−π catalysis is to stabilize anionic transition states on aromatic surfaces. This is complementary to the stabilization of cationic transition states on aromatic surfaces, a mode of action that occurs in nature and is increasingly used in chemistry. Anion−π catalysis, however, rarely occurs in nature and has been unexplored in chemistry. Probably because the attraction of anions to π surfaces as such is counterintuitive, anion−π interactions in general are much younger than cation−π interactions and remain under-recognized until today. Anion−π catalysis has emerged from early findings that anion−π interactions can mediate the transport of anions across lipid bilayer membranes. With this evidence for stabilization in the ground state secured, there was no reason to believe that anion−π interactions could not also stabilize anionic transition states. As an attractive reaction to develop anion−π catalysis, the addition of malonic acid half thioesters to enolate acceptors was selected. This choice was also made because without enzymes decarboxylation is preferred and anion−π interactions promised to catalyze selectively the disfavored but relevant enolate addition. Concerning anion−π catalysts, we started with naphthalene diimides (NDIs) because their intrinsic quadrupole moment is highly positive. The NDI scaffold was used to address questions such as the positioning of substrates on the catalytic π surface or the dependence of activity on the π acidity of this π surface. With the basics in place, the next milestone was the creation of anion−π enzymes, that is, enzymes that operate with an interaction rarely used in biology, at least on intrinsically π-acidic or highly polarizable aromatic amino-acid side chains. Electric-field-assisted anion−π catalysis addresses topics such as heterogeneous catalysis on electrodes and remote control of activity by voltage. On π-stacked foldamers, anion–(π)_n–π catalysis was discovered. Fullerenes emerged as the scaffold of choice to explore contributions from polarizability. On fullerenes, anionic transition states are stabilized by large macrodipoles that appear only in response to their presence. With this growing collection of anion−π catalysts, several reactions beyond enolate addition have been explored so far. Initial efforts focused on asymmetric anion−π catalysis. Increasing enantioselectivity with increasing π acidity of the active π surface has been exemplified for enamine and iminium chemistry and for anion−π transaminase mimics. However, the delocalized nature of anion−π interactions calls for the stabilization of charge displacements over longer distances. The first step in this direction was the formation of cyclohexane rings with five stereogenic centers from achiral acyclic substrates on π-acidic surfaces. Moreover, the intrinsically disfavored exo transition state of anionic Diels–Alder reactions is stabilized selectively on π-acidic surfaces; endo products and otherwise preferred Michael addition products are completely suppressed. Taken together, we hope that these results on catalyst design and reaction scope will establish anion−π catalysis as a general principle in catalysis in the broadest sense.

A visible-light-induced anti-Morkovnikov hydrosulfonation reaction of styrene derivatives with sodium sulfinates has been developed, featuring mild reaction conditions, good functional-group tolerance, good yields and high regioselectivity.

Epoxide-opening ether cyclizations are shown to occur on π-acidic aromatic surfaces without the need of additional activating groups and with autocatalytic amplification. Increasing activity with the intrinsic π acidity of benzenes, naphthalenediimides (NDIs) and perylenediimides (PDIs) support that anion−π interactions account for function. Rate enhancements maximize at 270 for anion−π catalysis on fullerenes and at 5100 M^–1 for autocatalysis. The occurrence of anion−π autocatalysis is confirmed with increasing initial rates in the presence of additional product. Computational studies on autocatalysis reveal transition state and product forming a hydrogen-bonded noncovalent macrocycle, like holding their hands and dancing on the active π surface, with epoxide opening and nucleophile being activated by anion−π interactions and hydrogen bonds to the product, respectively.

An intermolecular Dowd–Beckwith ring expansion/C–C bond formation is achieved through light-induced palladium catalysis. Not only styrenes but also the electron-rich alkenes such as silyl enol ethers and enamides were also compatible in this reaction.