QuantumPioneer: Scalable generation of quantum chemical data for solution-phase hydrogen transfer reactions
Name
QP_main_v3.pdf
Size
4.24 MB
Format
Adobe PDF
Checksum (MD5)
f4880f1ac50133bd0cf874f1f039d1e5
Name
QP_SI_v3.pdf
Size
9.04 MB
Format
Adobe PDF
Checksum (MD5)
778ca83f2175f91f87cc7fc818444fb5
Author(s)
Wu, Haoyang
Zheng, Jonathan
Pang, Hao-Wei
Kao, Yu-Chi
Zalte, Akshat
Abreu, Charlles
Al, Emad
Biswas, Sayandeep
Byun, Chansup
Burns, Jackson
Date Issued
August 20, 2026
Journal
Journal of the American Chemical Society
Publisher
American Chemical Society
Citation
Haoyang Wu, Jonathan W. Zheng, Hao-Wei Pang, Yu-Chi Kao, Akshat Shirish Zalte, Charlles R. A. Abreu, Emad Al Ibrahim, Sayandeep Biswas, Chansup Byun, Jackson W. Burns, Chuangchuang Cao, Yunsie Chung, Xiaorui Dong, Anna C. Doner, Jeremy Kepner, Bonhyeok Koo, Shih-Cheng Li, Angiras Menon, Lauren Milechin, Nathan Morgan, Kariana Moreno Sader, Kevin Spiekermann, Florence Vermeire, Bryan Wang, Markus Kraft, Connor W. Coley, William H. Green; QuantumPioneer: Scalable Generation of Quantum Chemical Data for Solution-Phase Hydrogen Transfer Reactions. Journal American Chemical Society 2026.
Version
Author's final manuscript
Abstract
High-fidelity quantum chemical (QM) data sets that jointly resolve reaction thermochemistry, kinetics, and solvation at scale remain scarce, especially for radical chemistry. We introduce QuantumPioneer, an open-access reaction-centered QM database and workflow for small organic molecules, focused on peroxylmediated hydrogen atom transfer (HAT) and the corresponding homolytic bond dissociation reactions. QuantumPioneer contains 348,258 species (2−21 heavy atoms), 167,237 validated HAT transition states (TS) with corresponding reaction energies and homolytic bond dissociation energies (BDEs), and over 100 million COSMO-RS solvation free energies ( G* ) solv and enthalpies ( H* ) solv across 295 solvents. The workflow uses ωB97X-D/def2- SVP geometries, DLPNO−CCSD(T)-F12d/cc-pVTZ-F12 single-point energies, empirical thermochemical corrections, transitionstate theory, and COSMO-RS BP-TZVPD-FINE solvation in a single high-throughput pipeline. Our benchmarks show reliable accuracy, with mean absolute errors (MAEs) compared to experimental and high-level QM reference data of 0.82 kcal/mol for gasphase enthalpies of formation, 1.60 kcal/mol for C−H BDEs, 1.45 kcal/mol for HAT barriers, and 0.57 kcal/mol for G* solv values. We demonstrate two predictive applications. First, we show that combining BDE and HAT-barrier models identifies experimentally observed oxidative degradation sites in drug-like molecules with a 91% top-5 hit rate and 82% site-level recall. Second, we show that a QM-parametrized Abraham model enables rapid solvation energy estimates at near-COSMO-RS accuracy within its training domain, reproducing computed G* solv and H* solv values with MAEs of 0.16 and 0.18 kcal/mol, respectively, though performance on experimental G* solv values for unseen solutes was worse, with an MAE of 1.32 kcal/mol. This work provides a scalable template for other reaction families, unifying equilibrium species, validated TS, thermochemistry, kinetics, and solvation into one workflow. 1. INTRODUCTION Efficient exploration of chemical space is fundamental to molecular discovery across fields ranging from drug discovery and materials science to sustainable energy. However, even when considering only plausible “drug-like” small organic molecules containing H, C, N, O, and S atoms, the estimated space exceeds 1060 making unguided search impractical. Accelerated, data-driven strategies are therefore needed to navigate this space efficiently and better design functional molecules.1,2 Computational chemistry provides a scalable route to large, self-consistent data sets, overcoming the heterogeneity often found in aggregated experimental data and enabling robust data-driven models for predictive modeling of complex chemical phenomena.3−5 First-principles quantum mechanics (QM) is particularly effective for this purpose, as electronicstructure calculations combined with statistical mechanics yield the thermo-kinetic parameters required for mechanistic models of reaction engineering, degradation chemistry, lead identification, and synthesis planning.
MIT Department
Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science
Terms of Use
Article is made available in accordance with the publisher's policy and may be subject to US copyright law. Please refer to the publisher's site for terms of use.
Persistent DSpace Link
DOI of Published Version
https://doi.org/10.1021/jacs.6c10371
