Using Data Science To Guide Aryl Bromide Substrate Scope Analysis in a Ni/Photoredox-Catalyzed Cross-Coupling with Acetals as Alcohol-Derived Radical Sources

Abstract

Ni/photoredox catalysis has emerged as a powerful platform for C(sp 2 )-C(sp 3 ) bond formation. While many of these methods typically employ aryl bromides as the C(sp 2 ) coupling partner, a variety of aliphatic radical sources have been investigated. In principle, these reactions enable access to the same product scaffolds, but it can be hard to discern which method to employ because nonstandardized sets of aryl bromides are used in scope evaluation. Herein we report a Ni/photoredox-catalyzed (deutero)methylation and alkylation of aryl halides where benzaldehyde di(alkyl) acetals serve as alcohol-derived radical sources. Reaction development, mechanistic studies, and late-stage derivatization of a biologically-relevant aryl chloride, fenofibrate, are presented. Then, we describe the integration of data science techniques, including DFT featurization, dimensionality reduction, and hierarchical clustering, to delineate a diverse and succinct collection of aryl bromides that is representative of the chemical space of the substrate class. By superimposing scope examples from published Ni/photoredox methods on this same chemical space, we identify areas of sparse coverage and high versus low average yields, enabling comparisons between prior art and this new method. Additionally, we demonstrate that the systematically-selected scope of aryl bromides can be used to quantify population-wide reactivity trends with supervised machine learning.

Data-driven multi-objective optimization tactics for catalytic asymmetric reactions

“…A potential drawback of such dimensionality-reduced maps, especially those encompassing numerous descriptors, is that their axes can be inherently challenging to interpret, as is the case for our bisphosphine descriptor library. 8 Therefore, we set out to develop a chemical space representation that would include more readily interpretable axes akin to a classical Tolman map, 32 while also incorporating multi-dimensional information of modern PCA plots. To accomplish this, parameters were selected from recognizable bins: steric, electronic, and geometric.…”

“…These workflows include machine learning (ML) algorithms such as multivariate linear regression (MLR) to regress structure to function, 5 classification tools to explore reactivity cliffs, 6 and dimensionality reduction techniques to map chemical space to visualize reactivity patterns. 7,8,9,10 More sophisticated algorithms can also be employed to explore the multi-dimensionality of reaction optimization including the exploitation of Bayesian optimization tools. 11 However, limited effort thus far has been reported for the simultaneous optimization of multiple objectives.…”

Guidelines for Machine Learning Yield Prediction from Small-Size Literature Dataset

“…The models exhibit poor or modest performances for coupling partners with less than 70 reactions reported. Adequate coverage of chemical space is crucial for building ML models 36 . Those results show that a dataset of a much smaller size than NiCOlit can be used to build a predictive model, provided that all reactions belong to the same coupling partner or substrate category (Fig.…”

A reactivity model for oxidative addition to palladium enables quantitative predictions for catalytic cross-coupling reactions

“…4 While much has been done to develop and understand new cross-coupling reactions and catalysts, less is known about how the specific molecular structures of complex building blocks affect the likelihood of successful coupling. 5–7 As a result, time- and resource–intensive reaction screening and optimization campaigns are often required for each new synthetic target. These involve many iterations and can still result in failure to find appropriate conditions for a given transformation, impeding access to potentially promising new medicines and materials.…”