Towards faster FT-IR imaging by reducing noise

Fourier transform infrared (FT-IR) imaging is a powerful technique that can be used to obtain spatially resolved chemical information from a large sample area in a relatively short time. However, temporal resolution of fast FT-IR imaging is limited by rapid degradation of data quality (due to increased noise) with faster image acquisition. We present various coaddition schemes to reduce noise and improve the quality of images acquired from such systems. The application of the proposed schemes allows for improved signal-to-noise ratio (SNR) characteristics in the resulting data. These schemes are tested by monitoring the dissolution of a polymer film [poly(α-methyl styrene)] by a low-molecular-weight solvent [methyl isobutyl ketone (MIBK)]. Pseudo coaddition improved the SNR by approximately 45%, while the SNR for sampling coaddition was found to scale as approximately N^0.5 where N is the number of coadded pixels. A total acquisition time of about 100 s was achieved, allowing the dissolution process to be monitored by using image acquisitions separated by 3 min. Low noise concentration profiles, linear solvent penetration rate, and polymer dissolution rate were measured. Detection limits of approximately 5% and quantification limits of approximately 20% were achieved by using optimal coaddition strategies. This result represents an order of magnitude improvement over untreated data.