Spectra digitized

Dynamic range Range determined by the number of bits in the output of the analog-to-digital converter represents the ratio between the largest and smallest signals observable in the digitized spectrum. 

Fig. 5. Digital Spectrum Stripping of an Infrared Spectrum, a 0.05% butyl methacrylate in carbon tetrachloride b Residual spectrum after stripping that of carbon tetrachloride c Spectrum of pure butyl methacrylate in 0.1 mm path length cell for comparison. (Adapted from Ref. IJI)...

The scattered light is detected with a photomultiplier tube and photon-counting electronics. The digital spectrum is then recorded with a multichannel analyzer and analyzed with a computer. A typical experimental arrangement is shown in Figure 2. 

For simulated IR and C-NMR spectra, no intensities were provided and all intensities were set to 1. The sampling interval (I) determines the level of detail that is preserved in the digitized spectrum. Finally, the spectrum is normalized to obtain the same total intensity for each spectrum, independent of the size of the molecule. 

Just as with real continuous signals, real digital signals exhibit conjugate symmetry in the frequency domain. Hence in a 512 point DFT the first 256 values will be conjugate symmetric with the second 256. For this reason only half a digital spectrum is normally plotted, and this will have half the number of values used to calculated the DFT. 

Full spectrum search methods use, as the name implies, a fully digitized version of the spectrum. When used in this form, the spectra are usually stored at reduced resolution because at full resolution the storage overhead and the time required for processing become excessive. The matching of the unknown spectrum is usually based on a correlation criterion, such as the Dot Product between the library spectrum and the unknown. In practice, the only real advantage of this method is the availability of a digitized spectrum for display purposes. 

---