The transmission of substituent effects in isomeric dichloroethano-bridged anthracene derivatives

Electron-Poor Acridones and Acridiniums as Super Photooxidants in Molecular Photoelectrochemistry by Unusual Mechanisms

Electron-deficient acridones and in situ generated acridinium salts are reported as potent, closed-shell photooxidants that undergo surprising mechanisms. When bridging acyclic triarylamine catalysts with a carbonyl group (acridones), this completely diverts their behavior away from open-shell, radical cationic, 'beyond diffusion' photocatalysis to closed-shell, neutral, diffusion-controlled photocatalysis. Brønsted acid activation of acridones dramatically increases excited state oxidation power (by +0.8 V). Upon reduction of protonated acridones, they transform to electron-deficient acridinium salts as even more potent photooxidants (*E1/2 =+2.56-3.05 V vs SCE). These oxidize even electron-deficient arenes where conventional acridinium salt photooxidants have thusfar been limited to electron-rich arenes. Surprisingly, upon photoexcitation these electron-deficient acridinium salts appear to undergo two electron reductive quenching to form acridinide anions, spectroscopically-detected as their protonated forms. This new behaviour is partly enabled by a catalyst preassembly with the arene, and contrasts to conventional SET reductive quenching of acridinium salts. Critically, this study illustrates how redox active chromophoric molecules initially considered photocatalysts can transform during the reaction to catalytically active species with completely different redox and spectroscopic properties.

MolNet: A Chemically Intuitive Graph Neural Network for Prediction of Molecular Properties

Most graph neural networks (GNNs) in deep-learning chemistry collect and update atom and molecule features from the fed atom (and, in some cases, bond) features, basically based on the two-dimensional (2D) graph representation of 3D molecules. However, the 2D-based models do not faithfully represent 3D molecules and their physicochemical properties, exemplified by the overlooked field effect that is a "through-space" effect, not a "through-bond" effect. We propose a GNN model, denoted as MolNet, which accommodates the 3D non-bond information in a molecule, via a noncovalent adjacency matrix A ‾ , and also bond-strength information from a weighted bond matrix B . Comparative studies show that MolNet outperforms various baseline GNN models and gives a state-of-the-art performance in the classification task of BACE dataset and regression task of ESOL dataset. This work suggests a future direction for the construction of deep-learning models that are chemically intuitive and compatible with the existing chemistry concepts and tools.

Quantifying Through-Space Substituent Effects

The description of substituents as electron donating or withdrawing leads to a perceived dominance of through-bond influences. The situation is compounded by the challenge of separating through-bond and through-space contributions. Here, we probe the experimental significance of through-space substituent effects in molecular interactions and reaction kinetics. Conformational equilibrium constants were transposed onto the Hammett substituent constant scale revealing dominant through-space substituent effects that cannot be described in classic terms. For example, NO2 groups positioned over a biaryl bond exhibited similar influences as resonant electron donors. Meanwhile, the electro-enhancing influence of OMe/OH groups could be switched off or inverted by conformational twisting. 267 conformational equilibrium constants measured across eleven solvents were found to be better predictors of reaction kinetics than calculated electrostatic potentials, suggesting utility in other contexts and for benchmarking theoretical solvation models.