Multiple spectra

Mass Spectroscopy. A coUection of 125,000 spectra is maintained at Cornell University and is avaUable from John WUey Sons, Inc. (New York) on CD-ROM or magnetic tape. The spectra can be evaluated using a quaHty index algorithm (63,76). Software for use with the magnetic tape version to match unknowns is distributed by Cornell (77). The coUection contains aU avaUable spectral information, including isotopicaUy labeled derivatives, partial spectra, and multiple spectra of a single compound. 

The rate-limiting step of MALDI-MS analysis (Basic Protocol 3) is the sample preparation, which involves depositing matrix and sample solutions onto the probe, allowing the solvent to evaporate, and introducing the probe into the ion source. This process may require up to 10 min per sample. From this point, analysis can be carried out in less than 1 min per sample. If multiple spectra are to be collected for each sample, accordingly more time will be required. Manipulation of spectra and peak labeling can be achieved in less than 5 min. 

IV. STRUCTURE ELUCIDATION RESEARCH BASED UPON MULTIPLE SPECTRA... 

Much of the research and development on computer-assisted structure elucidation (CASE) has been reported in recent decades. The key problem with CASE is that the system should integrate different pieces of information from multiple spectra (such as one- and two-dimensional H, C NMR spectra, IR spectra, and mass spectroscopy spectra), filter redundant information, generate a reasonable number of structural candidates, verify the candidates, and eventually suggest the best structure(s) for the chemist. 

To formalize a rigorous structure elucidation procedure, the data structures for multiple spectra should be well defined. The data structure represents the relationship between structure and spectrum (or spectra). Based upon the data structure, we can build CASE applications. The data characteristics of spectra routinely used for structure elucidation are summarized in Table V. 

As previously discussed, different types of spectra have different data structure and interpretation rules. To deduce the structure from these multiple spectra, we need to choose one of the spectra as the base spectrum and then start our analysis. Typically, the base spectrum should have an explicit correlation between substructure and subspectra. For instance, the NMR spectrum provides explicit structure-spectrum relations. We now select the one-dimensional NMR speetrum as the base spectrum to show one of the structure elucidation strategies. Other spectra, such as infrared or two-dimensional NMR spectra, are used as constraints to reduce the search space. This strategy is outlined in Fig. 19. 

FIGURE 19 One of the general schemes for structure elucidation from multiple spectra. 

A different but effective approach for removing cosmics is based on comparing several repetitive spectra of the same sample. Provided readout noise is not significant, there is no SNR penalty for obtaining multiple spectra before averaging. If these multiple spectra are added directly, the result is a simple average or sum, and cosmics will only be diluted, as noted earlier. However, the multiple spectra may be compared before adding, and cosmics may be detected. There are several mathematical methods for this process, a few of which are available in commercial instruments. 

The second means of collecting multiple spectra involves collection from numerous locations within the vessel at one or more times. Multiple spectra drawn at one time from various locations in a blender may be compared to themselves or to spectra collected at different times, with similarity indicating content uniformity. Parameters for comparison can again include common statistical factors such as standard deviation, relative standard deviation, variance, a host of indices based on standard deviation, or the results from pattern recognition routines. Of course, spectra collected in this manner can also be compared to a library of spectra from previous homogenous blends. 

Figure 11.11 (a) Simulation of a macro measurement. A spectrum from a sample covering a large area (large circle) (b) Simulation of micro measurements. Multiple spectra from a sample each covering a small area (small circles). 

Determining the system s parameters, and especially the activation energy E, requires an analysis of multiple spectra obtained with different heating rates. 

CID spectra of sulfadimethoxine at collision energies of 10-50 volts. Scanning the voltage that provides the collision energy is used to generate multiple spectra with structural information. 

---