激光与光电子学进展, 2016, 53 (11): 113001, 网络出版: 2016-11-14
基于SiPLS模型的稻壳中重金属铬LIBS检测 下载: 519次
Determination of Heavy Metal Chromium in Rice Husk by LIBS Coupled with SiPLS
光谱学 激光诱导击穿光谱 铬 稻壳 联合区间偏最小二乘法 spectroscopy laser-induced breakdown spectroscopy chromium rice husk synergy interval partial least squares
摘要
为了探索利用激光诱导击穿光谱(LIBS)对水田污染区稻壳中铬(Cr)元素含量进行绿色、快速检测的可行性,采用LIBS结合联合区间偏最小二乘法(SiPLS),对产自江西省某湖周边24个水田污染区稻壳样品中的Cr元素进行了定量分析。利用原子吸收光谱法(AAS)测得样品中Cr元素的真实浓度为32.51~510.33 μg/g,利用LIBS光谱获得的Cr元素三个特征谱线Cr I 425.43 nm、Cr I 427.48 nm和Cr I 428.97 nm清晰明显。对稻壳样品在422~446 nm波段的LIBS光谱数据进行九点平滑处理后,在采用SiPLS获得的最佳模型基础上,得出模型交叉验证均方根误差与预测均方根误差分别为26.1 μg/g和22.6 μg/g,训练集相关系数与预测集相关系数分别为0.9714和0.9840。对预测集样品进行相对误差及T检验分析,结果显示稻壳中Cr元素浓度的预测值与AAS法测量的真实值之间的平均相对误差为6.20%,且无显著性差异,表明模型具有较好的预测精度,可为自然条件下生长的农产品重金属安全绿色分析提供参考依据。
Abstract
To explore the feasibility of rapid and green determination of chromium (Cr) in rice husk from polluted paddy fields by laser-induced breakdown spectroscopy (LIBS), synergy interval partial least squares (SiPLS) combined with LIBS was employed for a quantitative analysis of Cr in 24 rice husk samples from different polluted paddy fields around a certain lake in Jiangxi Province. The real concentration in samples was determined by atomic absorption spectroscopy (AAS), and it ranged from 32.51 μg/g to 510.33 μg/g. The LIBS profiles of rice husk samples in the range of 422-446 nm were collected and three characteristic spectral lines, Cr I 425.43 nm, Cr I 427.48 nm, and Cr I 428.97 nm, were clear. The spectra from 422 nm to 446 nm were processed by the nine-point smoothing method, and then an optimal model was established with the SiPLS method. The results show that the root mean square error of cross validation and the root mean square error of prediction set were 26.1 μg/g and 22.6 μg/g, respectively, and the correlation coefficients for the training set and the prediction set were 0.9714 and 0.9840, respectively. The relative error was calculated and the T-test was carried out. The results indicate that the average relative error was 6.20% between predicted values from the SiPLS algorithm and real values from AAS, and there is no significant difference. The proposed model has high prediction accuracy, and provides reference for the rapid green determination of heavy metals in natural agricultural products.
王彩虹, 黄林, 刘木华, 陈添兵, 杨晖, 姚明印. 基于SiPLS模型的稻壳中重金属铬LIBS检测[J]. 激光与光电子学进展, 2016, 53(11): 113001. Wang Caihong, Huang Lin, Liu Muhua, Chen Tianbing, Yang Hui, Yao Mingyin. Determination of Heavy Metal Chromium in Rice Husk by LIBS Coupled with SiPLS[J]. Laser & Optoelectronics Progress, 2016, 53(11): 113001.