Multivariate and univariate analysis of infrared imaging data for high-throughput studies of NH(sub 3) decomposition and NO(sub x) storage and reduction catalysts

The application of Fourier transform infrared (FTIR) spectroscopic imaging for the analysis of the reaction products from parallel reactors has been extended to the quantitative analysis of complex infrared (IR) spectra. Multivariate factor-based and univariate calibration models were developed to extract quantitative concentration information from highly overlapped IR spectra. The three multivariate factor-based models of principal component regression (PCR) and partial least squares 1 and 2 (PLS-1 and PLS-2) were employed. The effects of the number of coadded mirror scans used in the data collection and the number of factors used in the data analysis on the predictive ability of this multivariate approach were characterized. The effectiveness of these approaches is demonstrated through application to the high-throughput study of ammonia decomposition and NO(sub x) storage and reduction catalysts.