Development of General Methods for Nucleophilic Functionalization

Chapter 1 describes development of a visible-light photoredox-catalyzed method that enables nucleophilic amination of primary and secondary benzylic C(sp3)–H bonds. A novel amidyl radical precursor and organic photocatalyst operate in tandem to transform primary and secondary benzylic C(sp3)–H bonds into carbocations via sequential hydrogen atom transfer (HAT) and oxidative radical-polar crossover (ORPC). The resulting carbocation can be intercepted by a variety of N-centered nucleophiles, including nitriles (Ritter reaction), amides, carbamates, sulfonamides, and azoles for the construction of pharmaceutically-relevant C(sp3)–N bonds under unified reaction conditions. Mechanistic studies indicate that HAT is amidyl radical-mediated and that the photocatalyst operates via a reductive quenching pathway. These findings establish a mild, metal-free, and modular protocol for the rapid diversification of C(sp3)–H bonds to a library of aminated products.

Chapter 2 describes the discovery and characterization of multiple N-(hetero)aryl, Nbenzyl and N-alkyl derivatives of 9-mesityl-3,6-di-tert-butyl-10-phenyl acridinium photocatalyst. The catalytic performances of these catalysts as photo-oxidant or photo-reductant (via in situ generated acridine radical) were compared in three model reactions. We also identified improved catalytic conditions for a previous cyanoarene-catalyzed nucleophilic amination reaction using a synthesized N- ycloheptyl acridinium catalyst (up to 98% yield).

Chapter 3 describes the discovery and development of several new (hetero)aryl sulfonyl fluoride reagents that have enhanced deoxyfluorination reactivity, improved physical properties, and excellent safety profiles compared to those of PyFluor and other fluorination reagents such as PBSF and DAST. To select structurally diverse reagents, we computed a virtual library of (hetero)aryl sulfonyl fluorides and leveraged training set design principles to broadly survey structure–activity relationships in a model deoxyfluorination reaction. We developed predictive models to optimize sulfonyl fluoride reagents for the deoxyfluorination of a key intermediate used in the synthesis of RIPK1 inhibitor GDC-8264. The top-performing reagents demonstrated broad applicability across diverse alcohol substrate classes, including complex natural products and active pharmaceutical ingredients, highlighting the power of data science-enabled approaches in reagent development.

Chapter 4 discusses and reflects on the design of reaction datasets in ways that are conducive to data-driven modeling, emphasizing the idea that training set diversity and model generalizability rely on the choice of molecular or reaction representation. As demonstrated throughout developments in chapters 1–3, models can codify our understanding of chemical reactivity and serve a useful purpose in the development of new synthetic processes via, for example, evaluating hypothetical reaction conditions or in silico substrate tolerance. Perhaps the most determining factor is the composition of the training data and whether it is sufficient to train a model that can make accurate predictions over the full domain of interest. We additionally discuss the experimental constraints associated with generating common types of chemistry datasets and how these considerations should influence dataset design and model building.

• Request a Brochure
• Request more Info
• Request a Quote