Spectral Data Banks

One of the advantages of the Internet is that information is constantly updated. Information such as spectral data banks or analytical instrumentation manufacturers can be freely obtained. 

Spectral data banks contain all sorts of information about a particular substance in the form of tables. The requested field of a table can be accessed by the user either via abbreviations (e.g., MF for molecular formula) or via input masks, which place the necessary denominations of the field at the user s disposal. Some fields may contain searchable information only as alphanumerical text (e.g., compound names), others may be searched only numerically (e.g., molecular weight). In the case of numerical fields, some numerical operators (e.g., < = > or-for area allocation) may also be applied. 

It is often possible to search for bands of particular intensity in selected wavelength ranges. During the search all fields may be interconnected logically (search masks) or by logic operators (and or not proximity operators). A summary of information contained in a spectral data bank for a given compound is given in Tab. 22.3. 

Table 22.3 Searchable information in spectral data banks.

The identification of compounds comprising more than 1 % in the oils can be also carried out by C-NMR and computer-aided analysis.The chemical shift of each carbon in the experimental spectmm can be compared with those of the spectra of pure compounds. These spectra are listed in the laboratory spectral data bank, which contains approximately 350 spectra of mono-, sesqui- and diterpenes, as well as with literature data. Each compound can be unambiguously identified, taking into account the number of identified carbons, the number of overlapped signals, as well as the difference between the chemical shift of each resonance in the mixture and in the reference. 

Spectral data banks There are several which offer real spectra of several tens of thousands of compormds in infrared, proton and carbon-13 NMR, and mass spectrometry. For example The Aldridi/ACD Library of FT-NMR Spectra (12 000 spectra) Nicolet/Aldridi FT-IR Condensed Phase Library (18500 spectra) NIST/EPA/NIH Mass Spectral Database (130000 spectra). 

The spectrum of Figure lb is a fingerprint of the presence of a CO molecule, since it is different in detail from that of any other molecule. UPS can therefore be used to identify molecules, either in the gas phase or present at surfaces, provided a data bank of molecular spectra is available, and provided that the spectral features are sufficiently well resolved to distinguish between molecules. By now the gas phase spectra of most molecules have been recorded and can be found in the literature. Since one is using a pattern of peaks spread over only a few eV for identification purposes, mixtures of molecules present will produce overlapping patterns. How well mixtures can be analyzed depends, obviously, on how well overlapping peaks can be resolved. For molecules with well-resolved fine structure (vibrational) in the spectra (see Figure lb), this can be done much more successfiilly than for the broad. 

Focus identification on the odor-active regions as determined by GC-sniffing and RI values as published in the Flavomet (http //www.nysaes.cornell.edu/flavornet). The fragmentation pattern obtained for the compounds can be compared with those in data banks like the Wiley/NBS Registry of Mass Spectral data (McLafferty and Stauffer, 2000) or the NIST 98 library (National Institute of Science and Technology). 

The process of measuring the difference between the two Raman parent spectra (right and left) is shown on the flow chart of the ROA data acquisition program in Figure 13. This program is written in Array Basic supported in Spectra Calc software. First, three spectral memory banks are created for the current, Raman parent and ROA spectra. Then several coadded spectra at four different quarter-wave plate positions are taken to complete one QWP cycle. The... 

In the ADAS data bank [23,24] one can find further spectral transitions. The so-called X-Paschen program, which was brought in as module 603, allows us to display the Zeeman pattern for a variety of elements like Bn, Bei-m, Ci-V, Hel, Ol, Krl, Mgn, Nai, Cai-ll, Nel-ll, Sii-iv, and 60 lines from 115nm to 910 nm. The variation of the pattern can be studied for different T[, B, field direction, observation angle, and apparatus function. 

The evaluation of spectra will be discussed separately for qualitative and quantitative analysis. Particular emphasis will be laid (i) on state-of-the-art methods for searching spectra in spectral libraries or searching for spectroscopic information in data banks and on (ii) procedures for multivariate data analysis. 

In the case of a spectral search by comparison of peak tables, the system first computes the necessary peak table from the experimental spectmm of the analyte. For the subsequent search both the position and the intensity of the peaks in the spectrum of the unknown and in the spectra of the compounds contained in the data bank are compared. After the search has been completed, the top position in the hit list is assigned to the substance with least differences in peak positions and peak intensities. 

In the case of a full spectra search, the complete set of spectral features (absorbance values at p wavelength positions) is compared between the spectrum of the unknown and aU spectra contained in the data bank. So-called similarity measures are computed for each individual comparison. In the case of the mostly employed similarity measure, the Euclidean distance, the spectram is regarded as a p-dimen-sional spectral vector (data points at p wavelength positions). The comparison... 

Spectral data are taken from the PLASUS SpecLine software library for atoms, ions and molecules as well as directly from the corresponding literature [108,122], PLASUS Specline is a commercially available program, but all the information is based exclusively on public sources and data banks [20,84,115,139,158]. For further details about PLASUS SpecLine contact PLASUS higenieurbitio, Dr.-Ing. Thomas Schiitte, Robert-Koch-StraBe 8, 86343 Konigsbrunn, Germany or visit the web on http //www.plasus.de. 

The development of such integrated approaches to spectral analysis will benefit greatly from the large data bases and reference banks developed with the Mossbauer Effect Data Indexes covering 1958-1976 and the Mossbauer Effect Reference and Data Journal database (MERDJ Vols. 1-19) covering 1976 to the present (http //www.unca.edu/medc/Journal.html). 

---