Quantitative spectral analysis, using Fourier transforms

EEG is quantitatively analyzed by spectral analysis using a fast Fourier transform algorithm. Spectra are calculated from epochs of 4 sec duration. The mean total spectral power between 1.5 and 64 Hz and the power in 5 subfrequency bands (1.5 4, 5-8, 8.5-12.5, 13-35 and 36-64 Hz) are calculated for the periods during which the treadmill is on. 

Infrared analyses are conducted on dispersive (scanning) and Fourier transform spectrometers. Non-dispersive industrial infrared analysers are also available. These are used to conduct specialised analyses on predetermined compounds (e.g. gases) and also for process control allowing continuous analysis on production lines. The use of Fourier transform has significantly enhanced the possibilities of conventional infrared by allowing spectral treatment and analysis of microsamples (infrared microanalysis). Although the near infrared does not contain any specific absorption that yields structural information on the compound studied, it is an important method for quantitative applications. One of the key factors in its present use is the sensitivity of the detectors. Use of the far infrared is still confined to the research laboratory. 

For qualitative analysis, two detectors that can identify compounds are the mass spectrometer (Section 22-4) and the Fourier transform infrared spectrometer (Section 20-5). A peak can be identified by comparing its spectrum with a library of spectra recorded in a computer. For mass spectral identification, sometimes two prominent peaks are selected in the electron ionization spectrum. The quantitation ion is used for quantitative analysis. The confinnation ion is used for qualitative identification. For example, the confirmation ion might be expected to be 65% as abundant as the quantitation ion. If the observed abundance is not close to 65%, then we suspect that the compound is misidentified. 

Enhancement of the accuracy of quantitative infrared determinations by use of mathematical manipulation of the spectral data as performed by Fourier Transform (FTIR) and Coiqruterlzed Dispersive (GDIS) Infrared Spectroscopy has been compared. Cotton-polyester blends and cotton treated with THPOH-NHo and with dimethyloldihydroxy-ethylene urea (DMDHEU) were analyzed by FTIR and GDIS. The mathematical techniques used Included direct spectral subtraction attd spectral subtraction combined with analysis of peak areas. 

---