FTIR同时测定多组分室内有机污染气体的方法研究

随着环保意识的增强，室内空气污染问题越来越引起人们的重视，尤其是挥发性有机化合物(VOCs)的污染。急需建立室内有机污染气体的多组分同时快速检测方法。文章采用红外光谱技术与化学计量学方法相结合，建立了同时测定室内主要有机污染气体苯、甲苯、二甲苯的分析方法。选择红外光谱中3 000～2 600 cm-1,1 100～600 cm-1两个谱段建立分析校正模型，对气体样品中苯、甲苯、二甲苯3种组分的计算浓度与标准浓度之间的复相关系数(r2)分别为0.970，0.955和0.946，校正集的均方根偏差(RMSEC)分别为0.074 2，0.081 9，0.087 7，预测集的均方根偏差(RMSEP)分别为0.132，0.134和0.033 3。 对未知样品的预测结果在误差允许范围内，用该方法同时测定多组分室内有机污染气体是可行的。采用偏最小二乘(PLS)算法所得分析校正模型的各性能指标略优于主成分回归(PCR)算法。

Study on Simultaneous Analysis of Indoor Air Multi-Component VOCs with FTIR

Recently,the indoor air pollution,especially the volatile organic compounds(VOCs) is attracting more and more(attention.) The determination of indoor air organic gases becomes the key issue.In the present paper,a simultaneous quantitative measuring method of indoor air multi-component VOCs was established based on FTIR combined with chemometrics.The(3 200)-(2 600) cm~(-1) and 1 100-600 cm~(-1) bands of IR spectrum were used to establish validation model,and excellent coefficients were obtained(r~2=0.970,0.955 and 0.946 for benzene,toluene and dimethylbenzene,respectively).The root mean standard error of calibration for benzene,toluene and dimethylbenzene is 0.074 2,0.081 9 and 0.087 7,respectively.The root mean standard error of prediction is 0.132,0.134 and 0.033 3,respectively.The error of unknown sample prediction is acceptable.The IR method is effective for simultaneous analysis of indoor air multi-component VOCs.The model established by partial least squares(PLS) is better than that by principal components analysis(PCR).