Normalized Descriptor for Unbiased Screening of Second-Order Nonlinear Optical Materials

Second-order nonlinear optical materials enable frequency doubling of light (second-harmonic generation, SHG), which is essential for optoelectronic applications ranging from materials characterization to quantum technologies. However, comparing SHG performance across materials remains challenging as the second-order nonlinear susceptibility, χ(2), spans several orders of magnitude and strongly depends on the band gap, Eg. To address this, we empirically validate a theoretical upper bound on χ(2) using new databases of ab initio-computed nonlinear optical (NLO) properties. We then formulate a normalized descriptor, d̂, which expresses the NLO response of a material relative to the band gap-dependent physical limit. We show that d̂ exhibits a similar distribution across a wide range of band gap energies. We also demonstrate that the formalism used to compute χ(2) leads to a competition with methodological uncertainty driven by band gap predictions to systematically alter the prediction of the second-order nonlinear response. The universality of d̂ supports its use as a robust, generalizable metric for chemistry-informed machine-learning models and quantification of NLO performance, enabling accelerated materials discovery and optimization across broad application frequencies.二阶非线性光学材料（second-order nonlinear optical materials）可实现光的倍频效应（second-harmonic generation, SHG），该效应在从材料表征到量子技术的各类光电子应用中均不可或缺。然而，由于二阶非线性极化率（second-order nonlinear susceptibility, χ(2)）的数值跨度可达数个数量级，且强烈依赖于带隙（band gap, Eg），跨材料对比SHG性能仍颇具挑战。为解决这一难题，本研究借助从头算（ab initio）得到的非线性光学（nonlinear optical, NLO）性质新数据库，通过实验验证了χ(2)的理论上限。随后，我们构建了归一化描述符d̂，该描述符将材料的非线性光学响应表示为相对于带隙依赖物理极限的归一化形式。我们发现，在宽泛的带隙能量区间内，d̂的分布具有高度一致性。我们还证实，用于计算χ(2)的理论框架会与由带隙预测驱动的方法学不确定性形成竞争，进而系统性地改变二阶非线性响应的预测结果。d̂的普适性证明其可作为化学信息机器学习模型（chemistry-informed machine-learning models）和非线性光学性能量化的可靠、可推广指标，从而在宽泛的应用频率范围内加速材料的发现与优化。