摘要
分别采用中红外衰减全反射技术和近红外光谱分析透射技术对毒死蜱乳油中违禁添加的敌敌畏乳油含量进行了快速定量测定。利用偏最小二乘法(partial least squares,PLS)建立敌敌畏的定量模型并进行优化,再分别用独立检验集和不同批次考核样本对模型适应性进行评价。结果表明,中红外和近红外光谱分析技术均可以准确测定毒死蜱乳油中敌敌畏的含量。中红外校正集标准差(RMSEC)为0.013,近红外为0.020,模型测定系数(R2)二者均为1.000。对于考核样品,近红外模型适用性更好,不需要进行模型修正,预测标准差(RMSEP)为0.09;中红外模型需要进行修正,修正后预测标准差(RMSEP)为0.03。红外光谱分析技术因前处理方法简单,测量迅速、方便,对于监督部门进行农药质量控制具有重要意义。
In the present study,the content of dichlorvos in chlorpyrifos was rapidly determined by mid-infrared and near-infrared spectroscopy.The quantitative models were established by partial least squares（PLS）method and optimized.The independent validation sets and 7 test samples were used to evaluate the model accuracy.The results showed that mid-infrared and nearinfrared spectroscopy can accurately determine the content of dichlorvos in chlorpyrifos.The RMSEC（the root mean square error of calibration）of the mid-infrared model and near-infrared model was 0.013 and 0.020,respectively.R2（determination coefficient）both were 1.000.For 7 test samples,RMSEP before model recalibration is 0.22（MIRS）and 0.09（NIRS）.The adaptability of the near-infrared model was much better and model updating was unnecessary.To sum up,MIR and NIR are both rapid and easy-operation method with simple pretreatment.
出处
《光谱学与光谱分析》
SCIE
EI
CAS
CSCD
北大核心
2010年第11期2962-2966,共5页
Spectroscopy and Spectral Analysis
基金
国家自然科学基金项目(20575076)资助
关键词
红外
近红外
衰减全反射
敌敌畏
毒死蜱
偏最小二乘法
Infrared
Near infrared
Attenuated total reflection
Dichlorvos
Chlorpyrifos
Partial least squares
