HS-SPME/GC×GC-Q-TOF MS结合化学计量学对回收聚对苯二甲酸乙二醇酯鉴别能力的研究

Study on the Ability of HS-SPME/GC×GC-Q-TOF MS Combined with Chemometrics to Discriminate Recycled Polyethylene Terephthalate

建立了顶空-固相微萃取(HS-SPME)结合全二维气相色谱-串联四极杆飞行时间高分辨质谱(GC×GC-Q-TOF MS)与化学计量学相结合的方法鉴别回收和原生的聚对苯二甲酸乙二醇酯(PET)。首先,对回收和原生PET中的挥发性有机物(VOCs)进行非靶向分析。然后根据正交偏最小二乘判别分析(OPLSDA)和T检验筛选出对鉴别贡献度高的标记物质,并基于所有物质、标记物质和高标准标记物质建立主成分分析(PCA)和线性判别分析(LDA)模型。结果表明,PCA模型很好地展示了回收PET组和原生PET组之间的差异,基于高标准标记物的LDA模型训练集和验证集的鉴别准确率分别达到100%和97.1%。该方法具有可行性、高稳定性和可预测性,能够达到鉴别回收PET材料的要求。此外,共检测和鉴定了468种挥发性有机物,其中31种挥发性有机物为对鉴别起重要作用的高标准标记物质。这些高标准标记物质的可能来源是食品、药品、化妆品、农药、塑料、工业和未知来源。

A method was established for the discrimination of recycled and virgin polyethylene terephthalate(PET)by head-space solid phase microextraction/comprehensive two-dimensional gas chromatography-quadrupole time-of-flight mass spectrometry(HS-SPME/GC×GC-Q-TOF MS)combined with chemometrics.Firstly,volatile organic compounds(VOCs)of recycled and virgin PET were detected to obtain their peak area data for building discrimination datasets.Then the markers and high-level markers were selected on the basis of orthogonal partial least squares discrimination analysis(OPLS-DA)and T-test.The principal component analysis(PCA)and linear discrimination analysis(LDA)models based on all compounds,markers and high-level markers were established.The results showed that PCA model could well explain the differences between recycled and virgin groups,while the accuracies for calibration and validation sets reached 100%and 97.1%respectively in LDA model.This study indicated that the approach above was accessible,high stable and predictable.Besides,a total of 468 VOCs were detected and identified,of which 31 VOCs were classified as high-level markers.The possible origins of these high-level markers could be food,medicine,cosmetics,pesticides,plastic,industry and unknown sources.