Rapid Determination of BTEX in Vehicles by Mid-infrared (MIR) Technique Combined with Chemometric Methods

The quality of in-vehicle air and its impact on the human health are gaining more and more concern recently. BTEX (benzene, toluene and xylene) is an important group of air contaminants in vehicles, and it dramatically threatens people's health. It is necessary to establish a rapid determination method for BTEX in vehicles to match the demand of routine monitor. In this paper, a rapid analytical method for BTEX in vehicles was developed by mid-infrared spectroscopy (MIR) technique combined with chemometric method. Partial least squares (PLS) regression was used to establish the calibration models between the MIR spectra data and the concentrations. The spectral pre-processing and the number of PLS factors were optimized to achieve the best prediction of the BTEX concentration. The root mean square error of calibration (RMSEC), root mean square error of prediction (RMSEP) and root mean square error of cross validation (RMSECV) of the optimal model were 0.008 9, 0.020 0 and 0.111 5 for benzene; 0.011 6, 0.001 1 and 0.139 8 for toluene; 0.013 7, 0.003 7 and 0.139 0 for xylene, respectively. Excellent correlations between the calculated concentrations of the gas samples and their actual values were observed, and the correlation coefficients (R-2) for benzene, toluene and xylene were all over 0.98. The relative error of test gas samples were between -5.69% and 2.05% when determined with the established MIR method. The MIR Method will significantly improve the efficiency of the routine monitor of the air quality in vehicles.