Fourier transform infrared deconvolution

We conclude this chapter by presenting several examples of deconvolution of real data. Most of these examples represent deconvolutions of data that were used as part of a spectral analysis rather than generated as deconvolution examples or tests. The examples include high-resolution grating spectra, tunable-diode-laser (TDL) spectra, a Fourier transform infrared spectrum (FTIR), laser Raman spectra, and a high-resolution y-ray spectrum. 

This work was supported by a grant from the National Science Foundation, t Abbreviations used are as follows. FTIR Fourier transform infrared spectroscopy, ATR attenuated total reflectance, IRE internal reflection element, SATR solution ATR FTIR, FSD Fourier self-deconvolution, PLS partial least-squares analysis, PRESS prediction residual sum of squares from PLS. SECV standard error of calibration values from PLS, PLSl PLS analysis in which each component is predicted independently, PLS2 PLS analysis in which all components are predicted simultaneously. 

This paper reports a study to verify the relationship between functional group distribution and thermal decomposition behavior. A Fourier transform infrared spectrometer (FTIR) has been employed to obtain quantitative infrared spectra of the coals, chars, and tars produced in the devolatilization experiments. The spectra have been deconvoluted by using a computerized spectral synthesis routine to obtain functional group distributions, which are compared to the model parameters. 

Highly concentrated protein solutions, thin protein films, or precipitate suspensions in various solvents can be used for Fourier transform infrared spectroscopy (FTIR). Great care should be taken in interpreting deconvoluted spectra the basic data is still only one curve. Using IR difference spectra avoids the possible errors of deconvolution, but the protein environment during the measurement must be absolutely controlled to obtain meaningful data. 

Figure 3 Fourier transform infrared (FTIR) regions correlating with A, the exotherm (105°C-120°C), and, B, the postexotherm (125°C-140°C) events for the lactose-lysine system. Spectra were deconvoluted and normalized. They were obtained at 5°C intervals with the highest temperature at the top of the stack.

The previous sections have dealt primarily with infrared absorption spectra, although the conclusions can in general be applied to other types of spectra. Here additional uses of deconvolution will be demonstrated. In the first example, a Fourier transform spectrum is simulated and several attempts to deconvolve this spectrum show limited success. In the second example, pressure-broadening effects in an infrared absorption spectrum and a Raman spectrum are simulated. An attempt at removing these effects by deconvolution shows some promise. 

Fig. 26 Fourier transform spectrum of v2 of ammonia. Trace (a) is a section of the infrared absorption spectrum of ammonia recorded on a Digilab Fourier transform spectrometer at a nominal resolution of 0.125 cm-1. In this section of the spectrum near 848 cm-1 the sidelobes of the sine response function partially cancel, but the spectrum exhibits negative absorption and some sidelobes. Trace (b) is the same section of the ammonia spectrum using triangular apodiza-tion to produce a sine-squared transfer function. Trace (c) is the deconvolution of the sine-squared data using a Jansson-type weight constraint.

The interferogram is actually a series of data points (retardation, intensity) collected during the smooth movement of the mirror. Using a mathematical function known as a Fourier transform, the spectrometer computer is able to deconvolute ( Fourier transform ) all the individual cosine waves that contribute to the interferogram, and so produce a plot of intensity against wavelength, or more usually the frequency in cm that is, the infrared single beam spectrum. All the... 

---