Creating and Using Spectra Libraries
Comparing Spectra

To compare two spectra with each other, the curve of the normalized spectrum can be compared to the curves of single library spectra. In some cases, better results are achieved when the first or second derivations are compared instead of the spectra themselves.

Use the comparison function (see Check Derivative) to determine the curve form to be used for the comparison.

In the next step, determine how the single curves are compared with each other. Select one of three mathematical methods; they are referred to as Match Criterion.

The combination of comparison function and standard of comparison results in a "similarity value" between 0 and 1000 that expresses the match degree between the search spectrum and various library spectra. The similarity value is known as Match Factor. A perfect match has the value 1000.

Depending on the method (hit list or peak tracking), a similarity list is displayed or each peak is assigned the most similar spectrum. If you enter a minimum similarity value (threshold), only the hits above the threshold value are displayed.

For more information about how to search spectra, refer to Displaying and Using UV Spectra  Searching Single Reference Spectra.

Result Presentation in a Hit List

Using the hit list, a similarity list sorted by the match factor and a representation of the original spectrum and the library spectrum with the best match are displayed. Entering a minimum similarity value excludes dissimilar spectra.

Due to the frequently insufficient characteristics of UV spectra, it may be necessary to further limit the hit lists. Hit Criteria are available as additional filters. Only the spectra fulfilling the selected criteria will be displayed.

For information about how to search reference spectra, refer to  Searching Reference Spectra.