Fourier transform infrared imaging: theory and practice.

The signal-to-noise ratio (SNR) of spectral data obtained from a microimaging Fourier transform infrared (FT-IR) spectrometer assembly, employing a step-scan interferometer and focal plane array detector, is analyzed. Based on the methodology of data collection, a theoretical description for the performance characteristics is proposed and quantitative effects of the acquisition parameters on the SNR are explained theoretically and compared to experiment. To obtain the best strategy for achieving either the highest SNR in a given time interval or for attaining a given SNR in the shortest time period, the concept of characteristic plots is introduced. The theoretical analysis is extended to FT-IR microimaging employing continuous scan interferometers in which the advantages of fast image collection are enumerated, while SNR limitations arising from mirror positioning errors are discussed. A step-scan method is suggested for faster data collection in which an optimal detector response and SNR benefits are retained. Theoretically obtained SNRs based upon the expressions proposed in this paper predict experimentally determined values quite well and can be used to obtain an understanding of the required developments for improved performance. Finally, SNRs for both microimaging systems and conventional microspectroscopic instrumentation are compared.