The efficient and stereoselective conversion of simple chemical building blocks, including olefins, alkynes, and aldehydes, into value-added products represents a modern challenge in synthetic organic chemistry. Significant research efforts have led to the discovery that rhodium-based catalysts can promote a variety of novel transformations. To this end, my co-workers and I have developed new synthetic methods, where we leveraged catalytically generated rhodium-hydride intermediates to achieve stereo- and regiocontrolled hydrofunctionalizations (Chapter 1) and cycloisomerizations (Chapter 2). Both processes are attractive methods that address the need for atom-economical and sustainable chemistry.
Typically, the hydrofunctionalization of alkynes yields achiral olefin products. We showed that rhodium-hydride catalysis can switch the regioselectivity of these processes to generate chiral products by (1) isomerization of an alkyne to an allene, (2) Rh–H reinsertion to generate an electrophilic Rh–allyl intermediate, and (3) allylic substitution with various (pro)nucleophiles. By careful choice of the catalyst, we developed an asymmetric alkyne hydroamination with amines (Chapter 1.1) and a regioselective decarboxylative hydroalkylation with β-keto acids (Chapter 1.2).
Despite the numerous catalysts available for asymmetric reduction, allenes are challenging substrates for stereo- and regiocontrolled reduction. In light of this challenge, we envisioned that our aforementioned strategy of alkyne hydrofunctionalization could be applied to achieve an asymmetric semireduction of allenes. We described a method that generates a Rh–allyl intermediate from a 1,1-disubstituted allene, which reacts with a Hantzsch ester (a hydride donor) to produce a chiral olefin product (Chapter 1.3). A designer Josiphos ligand was key to generate the products with high regio- and enanatioselectivity.
Desymmetrizations are powerful strategies to form multiple chiral centers in a single step. When coupled with cycloisomerizations, various carbocyclic motifs can be stereoselectively formed. We describe a desymmetrization of prochiral diketoaldehydes by ketone hydroacylation to generate chiral bicylic ketolactones (Chapter 2.1). In this process, a Rh–H intermediate generated from aldehyde C–H activation inserts into one of the carbonyl groups, and subsequent reductive elimination yields the product. By tuning the reaction conditions, we can selectively form each diastereomer of the product. Using aldehyde C–H activation, we showed that prochiral dienyl aldehydes can be transformed into chiral cyclohexenes (Chapter 2.2). This method complements the Diels-Alder cycloaddition and is a rare example of a cycloisomerization to generate six-membered rings.
Dynamic kinetic resolutons have emerged as an attractive to transform racemic building blocks into enantioenriched products. We described a dynamic kinetic resolution (DKR) of racemic α-allyl aldehydes by olefin hydroacylation to generate α,γ-disubstituted cyclopentanones with high enantio- and diastereoselectivity (Chapter 2.3). A bulky primary amine co-catalyst is important for selective racemization of the aldehyde substrate, and a bulky bisphosphine ligand is needed for cycloisomerization. Three different classes of aldehydes can be efficiently resolved with different amine and ligand combinations.
Fused cyclic ketones (e.g. tetralones) are commonly found in natural products and used as building blocks in chemical synthesis. Several tetralone derivatives were needed by Genentech scientists for the synthesis of drug candidates. However, no satisfactory general method existed for the preparation of these motifs. Toward a solution to this challenge, I collaborated with two process chemists to develop a strategy towards the synthesis of tetralones from aryl iodides and cyclopropane diesters (Chapter 3.1). This strategy proceeds via a homoconjugate addition between the corresponding aryl organocopper intermediate and the cyclopropyl electrophile followed by a decarboxylative Dieckmann annulation.