Abstract The current study presents a novel methodology to quantify lead in Turmeric using Raman spectroscopy. In this study, Partial Least Squares Regression (PLSR) was used for the quantification of lead. For calibration purposes, different amounts of lead were added to Turmeric samples encompassing a concentration range between 4 and 25 ​μg ​g−1. Since lead does not show any Raman band, for the purposes of this study, a complex was formed, its solvent was evaporated and the complex solid samples were registered with a Raman instrument. Raman measurements were performed in two different modes, -diffuse reflectance and transmission-. The PLSR models developed from Raman spectra of two data acquisition modes were evaluated in order to determine the suitability of both acquisition modes for quantifying lead content. The results indicated that diffuse reflectance showed better performance in terms of accuracy and robustness with a bias of 0.55 ​μg ​g−1, a relative standard error of prediction (RSEP) of 8.5% and a correlation between the predicted and reference values (R2) of 0.967. Despite the low lead concentration in the samples, the proposed model allows the quantification of the lead content in a fast and simple way.

中文翻译：

---

使用拉曼光谱检测姜黄中铅含量的定量模型

摘要 目前的研究提出了一种使用拉曼光谱定量姜黄中铅的新方法。在本研究中，偏最小二乘回归 (PLSR) 用于量化铅。出于校准目的，将不同量的铅添加到姜黄样品中，浓度范围在 4 到 25 微克 g-1 之间。由于铅不显示任何拉曼谱带，出于本研究的目的，形成了复合物，蒸发了其溶剂，并用拉曼仪器记录了复合固体样品。拉曼测量在两种不同的模式下进行，-漫反射和透射。对根据两种数据采集模式的拉曼光谱开发的 PLSR 模型进行了评估，以确定两种采集模式对铅含量定量的适用性。结果表明，漫反射在准确性和鲁棒性方面表现出更好的性能，偏差为 0.55 μg g-1，预测的相对标准误差 (RSEP) 为 8.5%，预测值和参考值之间的相关性 (R2 ) 的 0.967。尽管样品中的铅浓度很低，但所提出的模型允许以快速简单的方式量化铅含量。