Electron proton-coupled transfers in [NH4][H3N]n (n = 1, 2) Rydberg clusters: A machine learning-path integral study.

Machine learning-accelerated screening of hydroquinone analogs for proton-coupled electron transfer

Proton-coupled electron transfer (PCET) mediated by hydroquinone and related molecules is key to natural and artificial energy conversion. The reactivity of these molecules depends on their bond dissociation free energy (BDFE), but studying the relationship between structure and thermochemistry across this chemical space has been limited by challenging experimental setup and high computational expense. Here, we present the first use of the AIMNet2 neural network potential to calculate average BDFE (BDFEavg) values for the 2H+/2e− dehydrogenation of about 200 000 hydroquinone-like compounds, including vicinal diamines, diols, and dithiols. Benchmarking against DFT calculations for 168 substituted ortho-phenylenediamines (opda) shows good agreement (R2 ∼ 0.84). Our analysis finds that the BDFEavg of diamines ranges from 50 to 80 kcal mol−1 and can be systematically tuned by modifying the backbone and N-substitution: electron-withdrawing groups raise BDFEavg by up to 15 kcal mol−1, while lower aromaticity in furan and thiophene backbones decreases BDFEavg by approximately 10 kcal mol−1 compared to the phenyl systems (∼65 kcal mol−1). Validation through cyclic voltammetry and reactivity studies with quinone oxidants for selected compounds supports the computational results. This extensive thermochemical database and a web-based prediction tool developed as a result of this work will offer valuable resources for designing PCET reagents for catalysis, energy storage, and biomedical uses.

Hydrogen-Bond Network in Equimolar N-Methylacetamide-Water: Integrated Neutron Scattering, Molecular Dynamics, DFT-NBO-AIM, and Machine Learning Analysis.

An equimolar N-methylacetamide-water (NMA-W) mixture was re-examined by combining molecular dynamics (MD) simulations, DFT-based Natural Bond Orbital (NBO) and Atoms-in-Molecules (AIM) analyses, and data-driven reconstruction of previously reported neutron scattering functions of the NMA-W liquid. MD simulations, using an AMBER-based force field for NMA and the SPC/E model for water, reproduce reasonably well the experimental structure factors and pair correlation functions of the liquid. In parallel, a Random Forest Regressor is employed as a nonparametric data-driven reconstruction tool, showing that the main features of the structural data can be represented as a continuous numerical representation of the scattering vector Q, providing a continuous numerical representation for comparison purposes. Three representative NMA-water clusters previously identified as the most probable local arrangements governing the liquid structure are further analyzed using NBO and Atoms-in-Molecules (AIM) approaches. The analysis provides additional information on hydrogen-bond electronic structure, including charge-transfer effects, electron density at bond critical points, and topological characteristics of intermolecular interactions. Comparison with MD-derived lifetimes shows that stronger electronic interactions do not always correspond to longer-lived hydrogen bonds, reflecting the role of local network topology and dynamical fluctuations.

Cooperative stability, many-body expansion, and σ-aromaticity of (LiH)n clusters (n = 1-6): A CCSD(T) study at the complete basis set limit.

We investigated the energetics and bonding of lithium hydride clusters (LiH)n (n = 1-6) using a composite ab initio scheme inspired by W2 theory to achieve sub-kcal/mol accuracy. This approach combines CCSD(T) results extrapolated to the complete basis set limit with a rigorous treatment of core-valence correlation, scalar relativistic effects, and the diagonal Born-Oppenheimer correction. Our results show that while Hartree-Fock theory captures the primary electrostatic binding, correlation effects are crucial for determining the energetic preference of compact isomers over cyclic rings. A parallel density functional theory study shows that while standard hybrid functionals like B3LYP-D4 and M06-2X exhibit larger deviations, the double-hybrid revDSD-PBEP86-D4 functional closely matches our benchmarks, delivering sub-kcal/mol accuracy. Structural and chemical bonding analyses, including intrinsic bond orbital, nucleus-independent chemical shift, and many-body expansion (MBE) methods, reveal high ionic character and multi-center bonding (3c-2e and 4c-2e) within the (LiH)n clusters. MBE analysis of the interaction energy of the monocyclic clusters with respect to the LiH molecules reveals that the two- and three-body terms are consistently negative (stabilizing), while all higher-order terms are negligible. We find that σ-aromaticity in these systems is predominantly local and bond-centered. As the rings expand, the interior becomes magnetically decoupled from the σ-skeleton, precluding the formation of a global ring current. These results establish definitive benchmarks for the stability of prototypical electron-deficient clusters.

Assessing Constrained Nuclear–Electronic Orbital Theory for the Floppy CH 5 + Ion: Geometry-Dependent Zero-Point Effects on Structure, Hydrogen Rearrangement Dynamics, and Simulated IR Spectra

Recently, constrained nuclear–electronic orbital (CNEO) theory has been developed to incorporate nuclear quantum delocalization and zero-point effects into quantum chemistry calculations and ab initio molecular simulations. Motivated by discussions with John Stanton, we apply CNEO methods to protonated methane, CH5 +, as a stringent test case for the quantum-corrected effective-potential framework. CH5 + is a prototypical penta-coordinated nonclassical carbonium ion with a highly anharmonic potential energy surface, many low-lying geometries connected by low barriers, and extensive fluxional motion. Using CNEO, we examine the effective structural features, hydrogen rearrangement dynamics, and simulated IR spectra of CH5 +. The CNEO effective potential energy surface exhibits a minimum with C2v symmetry, which is more symmetric than the minimum-energy eclipsed-Cs structure on the conventional potential energy surface. CNEO molecular dynamics predicts more frequent hydrogen rearrangement than conventional ab initio molecular dynamics, with pronounced rearrangement observed starting from 50 K. A Fourier-filtered analysis of C–H bond-length distributions along the trajectories, combined with a Gaussian mixture model, reveals three structural and dynamical motifs: equilibrium-like, fluxional, and transition-state hovering configurations. The simulated IR spectra, which are obtained from either harmonic analysis or classical dynamics on the CNEO surface, are compared qualitatively with experimentally observed spectral features. Overall, this work assesses how the computationally efficient CNEO framework captures qualitative structural, dynamical, and spectroscopic trends in the highly fluxional CH5 +, while also highlighting the challenges and shortcomings of this classical-trajectory-based method in a system where fully quantum nuclear effects beyond zero-point effects are important.

Solvent effects on the structures and binding energies of arsine clusters (AsH3)n (n = 2–6): a DFT study with implicit solvation

Mapping Proton-Coupled Electron Transfer With Real Space Coordinates.

Both electron transfer (ET) and proton transfer (PT) are common steps in energy conversion across chemistry and biology. Coupling of these two transfer events reduces the energy demand relative to either individual process, but introduces a fundamentally new process-proton-coupled electron transfer (PCET)-with its own demands for a suitable theoretical description. Conceptualization of PCET usually involves a square scheme representing ET, PT, and PCET steps, each generating states with different energies. While intuitive, these square schemes do not offer structural insights such as the identities and contributions of nuclear motions to PCET. Herein, we present a computational approach that maps these square schemes onto real space coordinates (Å), from which ground and excited-state potential energy surfaces can be generated. This mapping involves the identification of PT and ET coordinates and reconstruction of the potential energy surfaces in orthogonalized coordinates. We find qualitative differences in key features of the surfaces for two distinct processes within PCET, namely concerted proton-electron transfer and hydrogen atom transfer, which may help to distinguish different PCET scenarios.