内置杂化泛函分子修正的分子固体中机器学习预测的NMR屏蔽常数
Machine-Learned NMR Shieldings in Molecular Solids with Built-In Hybrid-Functional Molecular Corrections
浏览论文内容
中文总结 AI 辅助
本研究推出内置杂化泛函分子修正的ShiftML4模型，用于预测分子固体NMR屏蔽常数，其在1H、13C、15N等核的预测精度优于前代模型及GIPAW-PBE，采用PET-MOLS优化结构后误差进一步降低。
中文摘要 AI 辅助
快速且准确的化学屏蔽估计器对屏蔽驱动的核磁共振（NMR）晶体学至关重要。用于屏蔽预测的机器学习模型已显著成熟，如今其主要限制在于训练所使用的电子结构参考数据。在此，我们推出ShiftML4，这是一种直接基于近似PBE0的单体修正计算训练的屏蔽张量模型，而非早期ShiftML模型所针对的PBE参考。ShiftML4在包含分子有机固体中12种最常见NMR核的多样化结构集上进行训练。在实验基准集上，针对相同几何结构，与GIPAW-PBE相比，ShiftML4对实验的13C各向同性均方根误差（RMSE）为1.67 ppm；ShiftML4给出的1H预测RMSE与ShiftML3相近（0.5 ppm），并将15N的RMSE从7.24 ppm降至6.08 ppm。该模型还降低了屏蔽张量各向异性的误差，针对13C CSA主成分，其RMSE为4.63 ppm，而GIPAW的RMSE为5.85 ppm。在更优几何结构上，预测精度的提升得以保留。采用近期的机器学习原子间势PET-MOLS优化的结构作为基础，该势可在数秒内达到近似杂化DFT几何结构，进一步将ShiftML4的误差降至0.48 ppm（1H）、1.49 ppm（13C）和3.66 ppm（15N）。
英文摘要
Fast and accurate chemical shielding estimators are essential for shielding-driven Nuclear Magnetic Resonance (NMR) crystallography. Machine-learning models for shielding predictions have matured significantly and today are primarily limited by the electronic structure reference data they are trained on. Here, we introduce ShiftML4, a shielding-tensor model trained directly on monomer-corrected calculations that approximate PBE0, rather than the PBE reference targeted by earlier ShiftML models. ShiftML4 is trained on a diverse set of structures containing 12 of the most common NMR nuclei in molecular organic solids. On experimental benchmark sets, the 13C isotropic RMSE against experiment is 1.67 ppm, compared with 2.34 ppm for GIPAW-PBE on the same geometries. ShiftML4 gives a similar 1H prediction RMSE to ShiftML3 (0.5 ppm) and improves the 15N RMSE from 7.24 to 6.08 ppm. The model also reduces errors in the shielding-tensor anisotropy, with an RMSE of 4.63 ppm on 13C CSA principal components against 5.85 ppm for GIPAW. The improvements in prediction accuracy are retained on better geometries. Basing shift predictions on structures relaxed with PET-MOLS, a recent machine-learned interatomic potential that reaches approximate hybrid-DFT geometries in seconds, lowers the ShiftML4 errors further to 0.48 ppm (1H), 1.49 ppm (13C) and 3.66 ppm (15N).
