紫外光谱法结合化学计量学用于环境水样中2,4,6-三硝基甲苯及其分解物的同时测定

Ultraviolet Spectroscopy Combined with Chemometrics for Simultaneous Quantitative Determination of 2,4,6-Trinitrotoluene and Its Degraded Products in Environmental Water Sample

结合紫外吸收和偏最小二乘法建立了环境水样中2,4,6-三硝基甲苯及其分解物的分析方法,在不经过任何预先分离的情况下,实现了环境水样中2,4,6-三硝基甲苯(TNT)及其分解物2,4-二硝基甲苯(2,4-DNT)和2,6-二硝基甲苯(2,6-DNT)的快速准确测定。采用正交试验设计(OAD)配制了25个训练集样本和15个独立预测集样本,采用变量种群分析(VCPA)方法挑选特征变量以提高偏最小二乘法(PLS)模型的预测效果,预测集中每个物质的相关系数R^2均达到0.99。将最优的VCPA模型用于真实水样中TNT以及DNT的检测,并采用HPLC方法进行了验证。实验结果显示,VCPA模型的回收率与HPLC法相近。紫外光谱法结合化学计量学方法可作为环境体系中多成分同时测定的一种简便、快速、有效的方法。

2,4,6-Trinitrotoluene （TNT） and its by-products dinitrotoluene （DNT） pose a significant threat to human health and other living organisms. However, the conventional analytical methods involved in bulky and expensive instruments are complicated and time- consuming, impeding quick and on-line determination. In this work, a facile yet effective strategy of utilizing UV-vis spectroscopy coupled with partial least squares （ PLS） was proposed, through which TNT and two isomers of DNT （2 ,4-DNT and 2 ,6-DNT） in nature water could be rapidly and simultaneously determined without any pre- separation. Variable combination population analysis （VCPA ） was utilized to select important feather variables and significantly improved the predictive performance of the PLS model. The calibration set contained 25 samples constructed by orthogonal array design （OAD）. The predictive ability of the models was validated by an independent prediction set including 15 samples, achieving up to 0. 99 of the determination coefficients （ R2） for each of the analytes. The optimized models were successfully applied to determine the 3 ingredients in 8 environmental samples involving in tap,lake and two kinds of river water with the recovery values of great than 97 % . Finally, the proposed method was further validated by high performance liquid chromatography method. UV-vis spectroscopy coupled with chemometrics may be used as simple and effective strategy with high potential in environmental monitoring.