Frequency discriminator, spectral

The unusual feature of their approach was in the frequency stabilisation, signal recovery and background discrimination, which are problems inherent in all spectrometer designs. The BWO frequency was stabilised using a reference cell containing a sample of the gas under test. Its spectral line was used as a frequency discriminator (Section 3.3). The important feature was its resolution of phase... 

Experimental studies of liquid crystals have been used for many years to probe the dynamics of these complex molecules [12]. These experiments are usually divided into high and low-frequency spectral regions [80]. This distinction is very important in the study of liquid crystalline phases because, in principle, it can discriminate between inter- and intramolecular dynamics. For many organic materials vibrations above about 150 cm are traditionally assigned to internal vibrations and those below this value to so-called lattice modes . However, the distinction is not absolute and coupling between inter- and intramolecular modes is possible. 

From the outset acoustic chemometrics is fully dependent upon the powerful ability of chemometric full spectrum data analysis to elucidate exactly where in the spectral range (which frequencies) the most influential information is found. The complete suite of chemometric approaches, for example PCA, PLS regression, SIMCA (classification/discrimination) are at the disposition of the acoustic spectral data analyst there is no need here to delve further into this extremely well documented field. (See Chapter 12 for more detail.)... 

Principal component analysis (PCA) [61] was first used to determine the number of independently varying chemical species present and to provide initial estimates of the spectral shapes resulting from these species and of their concentration profiles. Reference ATR FTIR spectra for several components (the solvent acetonitrile and water, the reagent cyclopentyltrichlorosilane and the product o7h3) were measured to assist in the deconvolution of the data. Frequency windows were selected that allowed the best discrimination between the reference compounds (725-775 cm-1 for acetonitrile, 850-900 cm-1 for water and the silsesquioxane). Finally, the MCR technique was applied to the data in the selected frequency windows to find the component spectra and relative concentration profiles that best fit the observed spectra. 

Raman spectroscopy is a bulk technique, although the depth of the analyzed volume is limited. The information depth and thus the spectra depend on the excitation frequency and the absorption coefficient and crystallinity of the sample (Cardona, 1983). To characterize catalyst surfaces and their interactions with reactants, the spectral contributions from the surface have to be discriminated from those of the catalyst bulk. This complication has to be considered when applying Raman spectroscopy to working catalysts (Banares, 2005). 

As we saw in Section 3.4, quadrature phase detection discriminates between frequencies higher and lower than the pulse frequency, but it does not prevent foldover from frequencies higher than the Nyquist frequency. For a desired spectral width FT, there are two common methods for carrying out quadrature phase detection, as was indicated in Section 3.4. One method uses two detectors and samples each detector at FT points per second, thus acquiring 2 FT data in the form of FT complex numbers. The other (commonly called the Redfield method ) requires only a single detector and samples at 2 FT points per second while incrementing the phase of the receiver by 90° after each measurement. (In two-dimensional NMR studies, a variant of this method is usually called the rime-proportional phase incrementation, or TPPI, method.) Because these methods result in quite different treatment of folded resonances, we now consider these approaches in more detail. 

While different types of bioinformatics tools can be used to discriminate samples based on their source, the initial paper by Petricoin and Liotta used ProteomeQuest , a software tool developed by Correlogics of Bethesda, Maryland. This software combines elements from genetic algorithm methods and cluster analysis. For the analysis of mass-spectral data, each of the input files is composed of m/z values on the x axis along with their corresponding amplitudes on the y axis. The output of the algorithm is the most robust subset of amplitudes at dehned frequency values that best separates the preliminary data acquired from the samples obtained from either healthy or diseased patients [21]. 

Spin-lattice relaxation is most sensitive to fast segmental fluctuations such as side-chain rotation and backbone oscillations with correlation times of around 10 s < Tc < lO s. A combination of different relaxation measurements makes it possible to discriminate among the different spectral densities [2, 15]. For a complete description of the frequency dependence, however, a series of measurements needs to be carried out at several different field strengths. 

The range of frequencies correctly represented in the spectrum (i.e. not folded) is called the spectral width, fsw As quadrature detection is used, the frequency scale on the spectrum runs from — j/sw to +ffsw i.e. positive and negative frequencies are discriminated. 

---