为了鉴别残留包衣剂农药的玉米幼苗, 使用傅里叶变换红外光谱结合主成分分析研究了玉米幼苗的根和叶片。 包衣剂包裹的玉米种子和未经过包衣剂包裹的玉米种子在相同条件下种植, 并测试它们幼苗根和叶片的红外光谱, 进行平行对照试验。 同时测试纤维素和包衣剂的红外光谱用于参考。 包衣剂包裹种子的幼苗根和叶片的红外光谱在1 384 cm-1附近出现C—H的弯曲振动吸收峰, 而未经过包衣剂包裹种子的幼苗根和叶片的红外光谱中C—H的弯曲振动特征吸收峰出现在1 375 cm-1附近。 参考纤维素和包衣剂的红外光谱, 可以确定1 384 cm-1吸收峰源自包衣剂残留的吸收。 在根的红外光谱中, 1 384 cm-1的包衣剂农药残留吸收峰尤为明显, 与1 375 cm-1的峰形对比, 较为尖锐。 随着玉米植株的生长, 根中1 384 cm-1的特征峰相对强度有减弱的趋势, 这是由于包衣剂农药残留被不断输送到植株的地上器官, 导致根中的农药残留浓度降低。 在经过包衣剂包裹种子的幼苗叶片的红外光谱中, 除了1 384 cm-1的农药残留特征峰外, 酰胺Ⅱ带的吸收峰呈现明显的肩峰, 而这一肩峰在种子没有被包衣剂包裹的幼苗叶片中未被观察到。 光谱分析显示一些农药残留的特征吸收峰被较强的纤维素吸收峰所掩盖, 而纤维素的一系列特征吸收峰又造成了光谱信息的重叠和数据冗余, 因此主成分分析被用于挖掘光谱中的特征信息。 在根的主成分1和主成分2得分图中, 含农药残留的样本和未含农药残留的样本被聚为两类, 两类样本散点没有重叠, 正确识别率为100%。 在叶片的主成分1和主成分2得分图中, 含农药残留的样本和未含农药残留的样本虽然也分为两类, 但是少量样本散点存在重叠, 正确识别率为93%。 结果表明, 傅里叶变换红外光谱结合主成分分析可以作为一种客观、 便捷的方法鉴别含有包衣剂农药残留的玉米幼苗。

In order to identify the maize seedlings which contain pesticide residues of coating agent, the roots and leaves of maize seedlings were studied by Fourier transform infrared spectroscopy (FTIR) combined with principal component analysis (PCA). The uncoated and coated maize seeds were planted with the same conditions, and the infrared spectra of roots and leaves of these seedlings were tested for parallel control experiments. At the same time, the infrared spectra of coating agent and cellulose were tested for reference. The infrared spectra of roots and leaves of the seedlings whose seeds were coated by coating agent showed a peak of C—H bending vibration near at 1 384 cm-1, but the C—H bending vibration in the infrared spectra of roots and leaves of seedlings without coatings appeared near at 1 375 cm-1. Referring to the infrared spectra of cellulose and coating agent, it can be determined that 1 384 cm-1 originated from the coating agent. At the infrared spectra of roots, the absorption peaks of pesticide residues at 1 384 cm-1 are particularly evident, which are sharper than that at 1 375 cm-1. With the growth of maize plants, the relative intensity of the characteristic peak at 1 384 cm-1 in roots tend to decrease, which is due to the continuous transport of pesticide residues to the above-ground organs of seedling, resulting in the reduction of pesticide residues in roots. Besides the characteristic peaks at 1 384 cm-1 of pesticide residues, the amide II band also shows obvious shoulder peak at the infrared spectra of seedling leaves whose seeds are coated by the pesticide, and this shoulder peak is not observed in seedling leaves whose seeds are uncoated. The spectral analysis showed that the characteristic absorption peaks of pesticide residues are covered up by the strong absorption peaks of cellulose, and the characteristic absorption peaks of cellulose result in overlapping of spectral information and redundancy of data. Therefore, the PCA was used to mine the characteristic information in the spectra. In the score plot of principal component 1 (PC 1) and principal component 2 (PC 2) of the roots, the samples containing pesticide residues and those without pesticide residues are clustered into two groups respectively, the scatter points of the two types of samples do not overlap, and the correct recognition rate is 100%. Although the leaves containing pesticide residues and those without pesticide residues are also divided into two groups in the score plot of PC 1 and PC 2, a small number of samples are overlapped, and the correct recognition rate is 93%. The results demonstrated the feasibility of utilizing FTIR spectroscopy combined with PCA, as an objective and rapid method for identification of the maize seedlings containing residual coating agent.