High-Throughput Experimentation and Machine Learning-Assisted Optimization of Iridium-Catalyzed Cross-Dimerization of Sulfoxonium Ylides.

A novel and convenient approach that combines high-throughput experimentation (HTE) with machine learning (ML) technologies to achieve the first selective cross-dimerization of sulfoxonium ylides via iridium catalysis is presented. A variety of valuable amide-, ketone-, ester-, and N-heterocycle-substituted unsymmetrical E-alkenes are synthesized in good yields with high stereoselectivities. This mild method avoids the use of diazo compounds and is characterized by simple operation, high step-economy, and excellent chemoselectivity and functional group compatibility. The combined experimental and computational studies identify an amide-sulfoxonium ylide as a carbene precursor. Furthermore, a comprehensive exploration of the reaction space is also performed (600 reactions) and a machine learning model for reaction yield prediction has been constructed.

A deep learning framework for accurate reaction prediction and its application on high-throughput experimentation data.

In recent years, it has been seen that artificial intelligence (AI) starts to bring revolutionary changes to chemical synthesis. However, the lack of suitable ways of representing chemical reactions and the scarceness of reaction data has limited the wider application of AI to reaction prediction. Here, we introduce a novel reaction representation, GraphRXN, for reaction prediction. It utilizes a universal graph-based neural network framework to encode chemical reactions by directly taking two-dimension reaction structures as inputs. The GraphRXN model was evaluated by three publically available chemical reaction datasets and gave on-par or superior results compared with other baseline models. To further evaluate the effectiveness of GraphRXN, wet-lab experiments were carried out for the purpose of generating reaction data. GraphRXN model was then built on high-throughput experimentation data and a decent accuracy (R2 of 0.712) was obtained on our in-house data. This highlights that the GraphRXN model can be deployed in an integrated workflow which combines robotics and AI technologies for forward reaction prediction.

HTE- and AI-assisted development of DHP-catalyzed decarboxylative selenation.

1,4-Dihydropyridine (DHP) derivatives play key roles in biology, but are rarely used as catalysts in synthesis. Here, we developed a DHP derivative-catalyzed decarboxylative selenation reaction that showed a broad substrate scope, with the assistance of high-throughput experimentation (HTE) and artificial intelligence (AI). The AI-based model could identify the key structural features and give accurate prediction of unseen reactions (R2 = 0.89, RMSE = 9.0%, and MAE = 6.3%). Our work not only developed the catalytic applications of DHP derivatives, but also demonstrated the power of the combination of HTE and AI to advance chemical synthesis.

Leveraging Regio- and Stereoselective C(sp3)-H Functionalization of Silyl Ethers to Train a Logistic Regression Classification Model for Predicting Site-Selectivity Bias.

The C-H functionalization of silyl ethers via carbene-induced C-H insertion represents an efficient synthetic disconnection strategy. In this work, site- and stereoselective C(sp3)-H functionalization at α, γ, δ, and even more distal positions to the siloxy group has been achieved using donor/acceptor carbene intermediates. By exploiting the predilections of Rh2(R-TCPTAD)4 and Rh2(S-2-Cl-5-BrTPCP)4 catalysts to target either more electronically activated or more spatially accessible C-H sites, respectively, divergent desired products can be formed with good diastereocontrol and enantiocontrol. Notably, the reaction can also be extended to enable desymmetrization of meso silyl ethers. Leveraging the broad substrate scope examined in this study, we have trained a machine learning classification model using logistic regression to predict the major C-H functionalization site based on intrinsic substrate reactivity and catalyst propensity for overriding it. This model enables prediction of the major product when applying these C-H functionalization methods to a new substrate of interest. Applying this model broadly, we have demonstrated its utility for guiding late-stage functionalization in complex settings and developed an intuitive visualization tool to assist synthetic chemists in such endeavors.

Dirhodium tetracarboxylates as catalysts for selective intermolecular C-H functionalization.

C-H Functionalization has become widely recognized as an exciting new strategy for the synthesis of complex molecular targets. Instead of relying on functional groups as the controlling elements of how molecules are assembled, it offers a totally different logic for organic synthesis. For this type of strategy to be successful, reagents and catalysts need to be developed that generate intermediates that are sufficiently reactive to functionalize C-H bonds but still capable of distinguishing between the different C-H bonds and other functional groups present in a molecule. The most well-established approaches have tended to use substrates that have inherently a favored site for C-H functionalization or rely on intramolecular reactions to control where the reaction will occur. A challenging but potentially more versatile approach would be to use catalysts to control the site-selectivity without requiring the influence of any directing group. One example that is capable of achieving such transformations is the C-H insertion chemistry of transient metal carbenes. Dirhodium tetracarboxylates have been shown to be especially effective catalysts for these reactions. This review will highlight the development of these dirhodium catalysts and illustrate their effectiveness to control both site-selective and stereoselective C-H functionalization of a wide variety of substrates.

Catalyst-Controlled Selective Functionalization of Unactivated C-H Bonds in the Presence of Electronically Activated C-H Bonds.

A new chiral dirhodium tetracarboxylate catalyst, Rh2( S-2-Cl-5-BrTPCP)4, has been developed for C-H functionalization reactions by means of donor/acceptor carbene intermediates. The dirhodium catalyst contains four ( S)-1-(2-chloro-5-bromophenyl)-2,2-diphenylcyclopropane-1-carboxylate ligands, in which all four 2-chloro-5-bromophenyl groups are on the same face of the catalyst, leading to a structure, which is close to C4 symmetric. The catalyst induces highly site selective functionalization of remote, unactivated methylene C-H bonds even in the presence of electronically activated benzylic C-H bonds, which are typically favored using earlier established dirhodium catalysts, and the reactions proceed with high levels of diastereo- and enantioselectivity. This C-H functionalization method is applicable to a variety of aryl and heteroaryl derivatives. Furthermore, the potential of this methodology was illustrated by sequential C-H functionalization reactions to access the macrocyclic core of the cylindrocyclophane class of natural products.

Design of catalysts for site-selective and enantioselective functionalization of non-activated primary C-H bonds.

C-H functionalization represents a promising approach for the synthesis of complex molecules. Instead of relying on modifying the functional groups present in a molecule, the synthetic sequence is achieved by carrying out selective reactions on the C-H bonds, which traditionally would have been considered to be the unreactive components of a molecule. A major challenge is to design catalysts to control both the site- and stereoselectivity of the C-H functionalization. We have been developing dirhodium catalysts with different selectivity profiles in C-H functionalization reactions with donor/acceptor carbenes as reactive intermediates. Here we describe a new dirhodium catalyst capable of the functionalization of non-activated primary C-H bonds with high levels of site selectivity and enantioselectivity.

Site-selective and stereoselective functionalization of non-activated tertiary C-H bonds.

The synthesis of complex organic compounds usually relies on controlling the reactions of the functional groups. In recent years, it has become possible to carry out reactions directly on the C-H bonds, previously considered to be unreactive. One of the major challenges is to control the site-selectivity because most organic compounds have many similar C-H bonds. The most well developed procedures so far rely on the use of substrate control, in which the substrate has one inherently more reactive C-H bond or contains a directing group or the reaction is conducted intramolecularly so that a specific C-H bond is favoured. A more versatile but more challenging approach is to use catalysts to control which site in the substrate is functionalized. p450 enzymes exhibit C-H oxidation site-selectivity, in which the enzyme scaffold causes a specific C-H bond to be functionalized by placing it close to the iron-oxo haem complex. Several studies have aimed to emulate this enzymatic site-selectivity with designed transition-metal catalysts but it is difficult to achieve exceptionally high levels of site-selectivity. Recently, we reported a dirhodium catalyst for the site-selective functionalization of the most accessible non-activated (that is, not next to a functional group) secondary C-H bonds by means of rhodium-carbene-induced C-H insertion. Here we describe another dirhodium catalyst that has a very different reactivity profile. Instead of the secondary C-H bond, the new catalyst is capable of precise site-selectivity at the most accessible tertiary C-H bonds. Using this catalyst, we modify several natural products, including steroids and a vitamin E derivative, indicating the applicability of this method of synthesis to the late-stage functionalization of complex molecules. These studies show it is possible to achieve site-selectivity at different positions within a substrate simply by selecting the appropriate catalyst. We hope that this work will inspire the design of even more sophisticated catalysts, such that catalyst-controlled C-H functionalization becomes a broadly applied strategy for the synthesis of complex molecules.

Site-selective and stereoselective functionalization of unactivated C-H bonds.

The laboratory synthesis of complex organic molecules relies heavily on the introduction and manipulation of functional groups, such as carbon-oxygen or carbon-halogen bonds; carbon-hydrogen bonds are far less reactive and harder to functionalize selectively. The idea of C-H functionalization, in which C-H bonds are modified at will instead of the functional groups, represents a paradigm shift in the standard logic of organic synthesis. For this approach to be generally useful, effective strategies for site-selective C-H functionalization need to be developed. The most practical solutions to the site-selectivity problem rely on either intramolecular reactions or the use of directing groups within the substrate. A challenging, but potentially more flexible approach, would be to use catalyst control to determine which site in a particular substrate would be functionalized. Here we describe the use of dirhodium catalysts to achieve highly site-selective, diastereoselective and enantioselective C-H functionalization of n-alkanes and terminally substituted n-alkyl compounds. The reactions proceed in high yield, and functional groups such as halides, silanes and esters are compatible with this chemistry. These studies demonstrate that high site selectivity is possible in C-H functionalization reactions without the need for a directing or anchoring group present in the molecule.