Advances in asymmetric organocatalysis over the last 10 years

Beyond esoteric interest, organocatalysis has now become one major pillar of asymmetric catalysis. Here, we discuss how new activation modes are conquering challenging stereoselective transformations and the recent integration of organocatalysis with emerging photo- and electrocatalysis, as well as artificial intelligence.

Activating carbonyl group with chiral secondary amine underpins the working principle of enamine catalysis for enantioselective functionalization of aldehydes or ketones, whereas carbonyl catalysis represents the reversal of this strategy. By analogy to pyridoxal-dependent enzymes 22 in carbonyl catalysis, Zhao's group innovatively disclosed a class of organic small molecule-carbonyl catalysts in 2018 (Fig. 1b, reaction i) 23 . This well-defined strategy enabled asymmetric Mannich reaction of glycinate 6 with N-diphenylphosphinyl imines 7 in the presence of an N-quaternized pyridoxal catalyst, furnishing synthetically useful a,β-diamino acid esters 8 in high efficiency with remarkable stereocontrol. Since then, substantial research efforts culminated in a wide array of asymmetric transformations by exploiting chiral aldehydes as the reliable amine activation catalysts 24 .
List group introduced organotextile catalysis built upon photochemical fixation of organocatalysts on textile nylon via a radical course (Fig. 1b, reaction ii) 25 . The organocatalysts exhibited excellent compatibility with polyamide to provide heterogeneous organocatalysts harbouring Lewis bases, Brønsted acids and chiral cinchona alkaloid derivatives. Excellent stability and efficiency comparable to the corresponding homogeneous catalysts were evidenced. This approach addressed the recyclability issue in classic organocatalysis with near-perfect enantioselectivity preserved after >250 asymmetric catalytic cycles. This impressive robustness of organotextile catalysts portends successful applications in industrial production.
Asymmetric organocatalytic nucleophilic additions are usually undertaken by hydrogen-bond catalysis. Given the modest acidity of chiral H-donors such as thioureas, squaramides, and guanidinium ions, the incompatibility of some substrate classes persists, which could be rectified by introduction of a Lewis acid as co-catalyst. Inspired by the reactivity enhancement through interaction with more weakly coordinating ligands, an extremely active Lewis acid catalyst generated by in situ silylation from chiral C-H acids was designed by List and coworkers to promote asymmetric Diels-Alder reaction 26 . One year later, a H-bond catalyst-assisted Lewis acid enhancement strategy was ingeniously utilized by Jacobson's group with squaramide as a weak H-donor (Fig. 1c) 27 . Control experiments demonstrated the high reactivity of silyl triflate-squaramide ion pair and NMR analysis shed light on the interaction between TESOTf (triethylsilyl trifluoromethanesulfonate) and squaramide wherein the latter establishes an adequate spatial environment for stereoinduction. These chiral active species readily mediate processes with heteroatom-stabilized cations to deliver Mukaiyama-type aldol reaction and [4 + 3] cycloaddition of enolates 11 with furan derivatives 12 in a highly stereoselective manner.

Target-oriented developments in organocatalysis
The catalytic enantioselective formation of fully substituted carbon stereocenters remains a challenging synthetic pursuit.    COMMENT NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-020-17580-z Contrary to a S N 2-type nucleophilic substitution reaction, enantio-differentiating addition in S N 1-type pathway is challenged by the high reactivity of the prochiral carbocationic intermediates, as it appears evident in organocatalytic nucleophilic substitution of racemic tertiary electrophiles in an enantioconvergent fashion. To this end, the capability of the silyl triflate-squaramide ion pair to abstract an acetoxy anion 27 was favorably leveraged in a catalytic enantioselective S N 1 reaction by Jacobson's group using racemic substrates with an acetoxy leaving group (Fig. 2a, reaction i) 28 . This work highlighted a stabilization of carbocation intermediate by non-covalent interaction with the chiral squaramide-embedded anion instead of a specific heteroatom interaction. While engendering effective enantiodiscrimination for nucleophilic attack, the chiral ion pair also circumvents undesired elimination and rearrangement byproducts. Tan's group made a further stride in 2019 employing a chiral cationic pentanidium catalyst, which involves a different catalytic process (Fig. 2a, reaction ii) 29 . While both methods share a similar key formation of stabilized chiral ion pair, in Tan's work a halogenophilic S N 2X pathway with halogen leaving group is operative. The encumbered carbon center directs nucleophilic attack to electrophilic halogen atom, aided by halogen-bonding effect provided by two germinal electronwithdrawing groups.

Anion abstraction
Beyond imine and carbonyl substrates, the weak basicity and lack of binding site of unbiased alkenes often counterpoise the potency of chiral Brønsted acid catalysis. Consequently, a highly acidic and structurally well-defined chiral organic acid, which can functionalize this class of poorly active molecules is enthralling. List's group devised a chiral imidodiphosphorimidate (IDPi) catalyst, which accomplished the straightforward enantioselective hydroalkoxylation of sterically and electronically unbiased alkenol 20 30 . Density functional theory calculations supported a mechanism involving initiation of the reaction by alkene protonation, whereas the rigid and narrow chiral cavity is beneficial for enantiocontrolled cyclization. Catalytic single aldolizations of acetaldehyde enolates 23 were realized asymmetrically shortly after by virtue of IDPi catalyst without competing oligomerization (Fig. 2b, reaction i) 31 . Remarkable enantiopurity of βhydroxy aldehyde 24 building blocks was maintained at 0.5 mol% catalyst loading. In the context of the more exacting arene functionalization, our group overrode the high dearomatization barrier through judicious installation of an azo group. The interaction of azo with CPAs via H-bond boosted the system electrophilicity and allowed arene activation via conjugation transmission (Fig. 2b, reaction ii) 32 .
Intensive research efforts further expanded the boundary of organocatalysis to generate atropisomers. Our group has been devoted toward asymmetric construction of axially chiral frameworks including the privileged NOBIN 27 and BINAM 28 by means of organocatalysis 33   of organocatalysis, bringing in efficient solutions to longstanding challenges. These include multicomponent reactions (MCRs) 35 such as four-component Ugi reaction and Passerini reaction known for high step and atom economy, the much valued attributes in diversity-oriented synthesis of complex molecules. The asymmetric variants have long remained unknown owing to the complexity of the mechanism as well as strong background and competing reactions 36 . Recently, enantiocontrol of the classical Ugi reaction was tackled with CPAs 37,38 . The H-bond interaction between CPA and acid component not only enhances the acidity of CPA and inhibits background reaction, but also strengthens the nucleophilicity of acid for attack of imine.

Outlook
This 10 years' advent of organocatalysis witnessed huge achievements in resolving enduring synthetic challenges as well as discovery of new catalytic modes, reaction types and advantageous integration with disparate reaction types. A sustained growth of this field is anticipated with refinement along several directions. First, most organocatalysts remain substandard compared with metal catalysts particularly for substrates without binding point. In this regard, stronger chiral Brønsted acids or Lewis acid enhancement tactics could enhance the potential of typical H-bond catalysis. More promisingly, asymmetric halogen, chalcogen, and pnictogen bonding catalysis may provide a new avenue in activation of inert substrates. Alternatively, merging organocatalysis with other synthetic strategies could develop into novel synthetic approaches. Parallel to the tremendous success in photocatalysis, the potential of electrocatalysis as another source of radical partners is likely to be implemented in organocatalytic reactions, however, the practical realization still remains in its infancy. Moreover, translating organocatalysis into industrial settings require efforts to improve both catalyst efficiency (in order to decrease the catalyst loading) and recyclability. Last but not least, introduction of machine learning into organocatalysis is underway. Although Denmark's group has realized an accurate prediction of the selectivity of chiral phosphoric acid catalysts, reported models were confined to thio-and N-acylimine substrates 39 . There remains a long way before machine learning will be established as handy daily tool for chemists working in this field.