修复PBE虚假金属性以用于HSE06级别的二维光催化剂筛选生产绿色氢气
Repairing PBE-Spurious Metallicity for HSE06-Level Screening of 2D Photocatalysts for Green Hydrogen Production
AI总结：
本研究结合泄漏感知的PBE虚假金属性修复与HSE06-PBE Δ学习，利用XGBoost回归器校正二维材料数据库中的带隙，筛选出符合要求的绿色氢气光催化剂候选物。
AI中文摘要：
半局部PBE计算会将窄带隙半导体标记为金属，从而在筛选前排除可行的光催化剂。我们通过将泄漏感知的PBE虚假金属性修复与HSE06-PBE Δ学习相结合，解决了计算二维材料数据库（C2DB）中的这种失效模式。第一阶段使用结构、化学、磁性和稳定性描述符对HSE06未知的PBE金属进行分类，同时排除HSE06/GW量和PBE电子捷径。第二阶段针对校正后的绝缘群体学习E_g^HSE - E_g^PBE。精选的XGBoost回归器重建HSE06带隙，平均绝对误差为0.108 eV（R²=0.989），而原始PBE的平均绝对误差为1.036 eV。第一阶段分类器仅用于分类，因为标记的真实金属类别包含29种材料；其最佳留出性能给出87.5%的准确率、0.286的真实金属召回率和0.643的平衡准确率。校正后的pH 0电子筛选产生10种严格和29种初始松弛的绿色氢气光催化剂候选物。在PBE水平下，4种严格和18种松弛候选物将无法通过1.6-2.8 eV的带隙窗口。对6种机器学习预测化合物的针对性VASP HSE06计算给出了材料级平均绝对误差：C2DB原生模型为0.417 eV，Magpie+结构模型为0.182 eV，精选模型为0.110 eV；\feat{1AgBr-1}移至带隙上限以上，留下28种保留的松弛候选物。该工作流程表明，高保真校正必须通过候选成员资格而非仅通过全局回归误差来评估。
英文摘要：
Semilocal PBE calculations can remove viable photocatalysts before screening by labeling narrow-gap semiconductors as metals. We address this failure mode in the Computational 2D Materials Database (C2DB) by combining leakage-aware repair of PBE-spurious metallicity with HSE06--PBE $Δ$-learning. Stage~I classifies HSE06-unknown PBE metals using structural, chemical, magnetic, and stability descriptors, while excluding HSE06/GW quantities and PBE electronic shortcuts. Stage~II learns $E_g^{HSE} - E_g^{PBE}$ for the corrected insulating population. The curated XGBoost regressor reconstructs HSE06 gaps with a mean absolute error of 0.108~eV ($R^2=0.989$), compared with 1.036~eV for raw PBE. The Stage~I classifier is used only for triage because the labeled true-metal class contains 29 materials; its best holdout performance gives 87.5\% accuracy, 0.286 true-metal recall, and 0.643 balanced accuracy. The corrected pH~0 electronic screen yields 10 strict and 29 initial relaxed green-hydrogen photocatalyst candidates. Four strict and 18 relaxed candidates would fail the same 1.6--2.8~eV gap window at the PBE level. Targeted VASP HSE06 calculations for six ML-predicted compounds give material-level MAEs of 0.417, 0.182, and 0.110~eV for the C2DB-native, Magpie+structural, and curated models, respectively; \feat{1AgBr-1} shifts above the upper gap cutoff, leaving 28 retained relaxed candidates. The workflow shows that high-fidelity correction must be evaluated by candidate membership, not only by global regression error.
