Real spectrum

There is one important point, however, that we have neglected so far. Real spectra in the frequency domain do not look like those in Figures 3.7(b)-3.9(b) the lines in the spectra are not stick-like and infinitely sharp but have width and shape. 

A complete description of the effects of a crystal field upon a d" ion would include similar analysis of the behavours of all terms arising for that d" configuration. Box 3-7 summarizes the case for d, and in Box 3-8, we illustrate a method of using Fig. 3-19 to determine Dq and B values from real spectra. 

Note that mosaic artifacts can also occur physically in real spectra when a real powder sample of a model compound exhibits microcrystallinity and thus contains too few different molecular orientations. This phenomenon is rare in X-band EPR and is usually easily solved by grinding the sample in a mortar it is, however, not at all uncommon even for extensively ground samples in high-frequency EPR with single-mode resonators where the sample size is orders of magnitude less than that of an X-band sample. 

Figure 16 Comparison of real and synthetic spectra of 0.3 M DDAB/DTAB mixtures in the fingerprint region illustrating the band splitting in real spectra is not accounted for by spectral overlap (T = 25°C).

The overall aggreement between the synthetic and experimental spectra is remarkably good, providing support for the stated assumption. Slight discrepancies between the synthetic and real spectra may be attributed to changes in absorptivity of some of the bands with pH, which were ignored in this simple analysis. 

Real spectra are rarely perfect. Samples often contain traces of water, giving weak absorptions in the O — H region. 

In real spectra, there will be several peaks, and the time series appear much more complex than in Figure 3.16, consisting of several superimposed curves, as exemplified in Figure 3.17. The beauty of Fourier transform spectroscopy is that all the peaks can... 

The difficulty is that real spectra always contain noise. Figure 3.25 represents a noisy time series, together with the exponentially filtered data. The filtered time series amplifies noise substantially, which can interfere with signals. Although the peak width of the new transform has indeed decreased, the noise has increased. In addition to making peaks hard to identify, noise also reduces the ability to determine integrals and so concentrations and sometimes to accurately pinpoint peak positions. 

Now that we have had the chance to look at a selection of experimental results we can discuss some more general points associated with the spectra and the theory. Using the techniques described above in Section 2 we obtain UP-spectra of adsorbate-covered surfaces either as real spectra or as difference spectra. The most important change compared to the clean, uncovered surface is the observation of adsorbate resonances as positive peaks in the difference spectra. 

For the example in Fig. 2, the Fourier transformed NMR spectra (variables or descriptors being intensity as a function of frequency) were utilized for the creation of the data matrix D. It should be noted that many different descriptors can be used to create D, with the descriptor selection depending on the analysis method and the information to be extracted. For example, in the spectral resolution methods (Section 6), the desired end result is the determination of the true or pure component spectra and relative concentrations present within the samples or mixtures [Eq. (4)]. For this case, the unmodified real spectra Ij co) are commonly used for the chemometric analysis. In contrast, for the non-supervised and supervised methods described in Sections 3 and 4, the classification of a sample into different categories is the desired outcome. For these types of non-supervised and supervised methods the original NMR spectrum can manipulated or transformed to produce new descriptors including... 

It can be seen that the spectra are very similar. This was also observed within the experimental spectra. For HD and VX theoretical spectra were also used for the cluster analysis. The Clnster Analysis was done by making the assnmption that the theoretical spectra of GB, HD and VX are more closely related to the real spectra of GB, HD and VX than the experimental spectra of the simnlants and degradation products. This assumption is based on the comparisons that were done before and on the fact that this is the procednre nsed by most researchers where the theoretical and experimental spectra are compared nsing DFT °. 

Figure 2.4 shows a comparison of the results obtained from the PCA and real chemical models for a Raman emulsion image [30]. The latter image is formed by four constituents related to the drop phase, the interphase, an additive and the off-drop phase. The two models (real and PCA) resemble each other when considering general trends for example, the score maps are reminiscent of the real distribution maps, although the information seems to be more mixed, and the most salient spectral features in the real spectra can also be found in the different... 

We assume that the dipole-dipole interaction term can be ignored for this calculation. However Sackman and Trauble (14,15,16) and Devaux and McConnell (17) believe there is a contribution from this term in real spectra. The problem appears tractable if measurements are made at temperatures high enough to average out the dipole-dipole terms. 

This semi-deterministic approach is interesting because the basis of reference spectra used for the modelling of real spectra is made up of spectra mixtures and specific mineral or organic compounds, the optical properties of which often explain a part of the UV spectra. The reference spectra of mixtures are statistically representative of the different heterogeneous fractions of wastewater, because they are selected from wastewater fractionation. For each wastewater sample, several nitrations are carried out (1 and 0.025 ixm), and the spectra of the filtrate are acquired. The differential spectra are then considered for the basis constitution (Table 2). 

Figure 22 shows real spectra from industrial, wastewater and natural sample containing hexavalent chromium at different concentrations. 

The abstract spectra in matrix X represent a mathematical solution to Equation 3.44 but their interpretation and identification are not clear. True spectra can be obtained by rotating the abstract spectra to best match suspected (target) real spectra. This is the process of target transformation and is a powerful technique in factor analysis since it allows real factors to be identified individually. 

Section 8 applies the adaptive wavelet algorithm to two sets of data in an attempt to further illustrate the mechanics behind the procedures. The first set of data is simulated, whilst the second considers real spectra of various kinds of minerals. The classifier that we use is Bayesian linear discriminant analysis [15]. 

The best fit was obtained with the sum of two bands. At low concentration of calcofluor, one band is obtained at a maximum of 326 nm and a bandwidth of 23 nm. The second band has a maximum at 347 nm and a bandwidth of 40 nm (Fig. 8.29). Although the positions of the maxima are significant and correspond to those found for the hydrophobic and surface Trp residues, respectively (Fig. 8.26), the values of the bandwidths do not correspond to those known for Trp residues (48 and 59 nm for the hydrophobic and surface residues, respectively). This means that it is difficult to separate the real spectra one from each other, i.e., in presence of low calcofluor concentration, a i-acid glycoprotein has kept its folded structure. 

---