三维荧光光谱结合交替惩罚三线性分解同时对水中三种酚类测定

酚类化合物对动植物机理有着严重危害，利用三维荧光光谱结合交替惩罚三线性分解（APTLD）算法，完成了不含干扰物和干扰物共存时激发-发射荧光光谱重叠严重的麝香草酚、对苯二酚和苯酚的直接快速准确定性、定量分析。研究了温度对三种酚类化合物荧光强度的影响。对扫描所得激发-发射矩阵信号（EEM）进行二次去散射和光谱校正预处理，最大程度保留了原光谱信息，避免光谱严重失真。将APTLD算法与平行因子（PARAFAC）和交替三线性分解（ATLD）算法进行对比，突显该算法的优势。实验得出，APTLD算法能够较好的解析荧光光谱数据的重叠峰，分别得到三种目标分析物的荧光光谱，实现快速定性分析；定量分析时平均回收率为（97.4±4.5）%~（103.1±3.0）%；预测均方根误差（RMSEP）低于1.664×10-2 μg·mL-1，且检测限低于国家标准；处理过程简洁快速，为水环境中酚类化合物实现现场检测和在线实时监测提供了有力依据。

Determination of Phenols in Water by Three Dimensional Fluorescence Spectroscopy Combined with APTLD

Phenols have serious harm to animals and plants. The experiment is a direct, rapid and accurate method for the qualitative and quantitative analysis of phenols in the case of interference and no interference by the method of fluorescence spectrometer combined with APTLD algorithm. The effects of temperature on the fluorescence intensity of thymol, hydroquinone and phenol were studied. After the comprehensive consideration, the experiment was carried out at 20 degrees centigrade. The obtained spectral data array is processed by eliminating scattering and correction. The original spectrum information is preserved to avoid the serious distortion of spectrum. APTLD algorithm is compared with PARAFAC and ATLD algorithm, highlighting the advantages of the algorithm. Experimental results showed that the APTLD algorithm can well analyze the overlapped peaks of the fluorescence spectrum data and obtain the fluorescence spectra of three target analytes respectively, achieving rapid qualitative analysis; the average recovery rate of the quantitative analysis is 97.4%±4.5%~103.1%±3%; The root mean square error of prediction is less than 1.664×10-2 g·mL-1, and the detection limit is lower than the national standard; The treatment process is simple and rapid, which provides a powerful basis for on-site detection and on-line real-time monitoring of phenolic compounds in water environment.