卷烟主流烟气粒相物硅烷衍生化GC-MS数据的解析

应用化学计量学方法对卷烟的主流烟气粒相物中极性非挥发物硅烷衍生化的CC-MS数据进行了解析。首先应用主成分分析法判断各色谱峰中的化学成分数。对于其中含有多种成分的重叠峰,应用多变量曲线分辨-变换最小二乘法(MCR- ALS)进行解析。用最大不相关法选取MCR-ALS的一组初始迭代矢量。最后根据总离子流图对解析得到的色谱峰曲线进行标度化。对选定样品的GC-MS数据进行判别,发现17％的色谱峰是2种或3种成分的重叠峰,经过解析可得到较好的色谱图和质谱图。结果表明了此方法在复杂样品的GC-MS数据解析中的有效性。

Resolution of GC/MS data of total particulate matter in mainstream smoke of tobacco after silylating

Chemometric methods are used to resolve GC/MS data of total particulate matter in mainstream smoke of tobacco after sily-lating. PCA is applied to estimate the number of chemical compositions in the chromatographic peaks. Multivariate Curve Resolution-Alternating Least Squares (MCR-ALS) method is employed to resolve the overlapped chromatograms. The initial estimate for MCR-ALS is collected by picking the most dissimilar chromatographic profiles. And the resolved chromatographic profiles are scaled based on the total ion chromatogram. The resolution method is applied to CC/MS data of a selected total particulate matter in mainstream smoke of tobacco sample. 17% of the chromatographic peaks are overlapped chromatograms containing two or three components. After resolution , pure chromatographic profiles and mass spectra are obtained. Satisfactory results illustrate that the resolution methods are powerful approaches for the determination of the number of chemical compositions contained in the investigated chromatographic peaks and the resolution of overlapped chromatograms.