Refining nuclide identification results
As identification of components is not an easy task, fully automatic analysis is not always achievable, thus manual intervention is regularly necessary to improve the results.
Checking Nuclide identification details
To see the details of the nuclide identification, click to tab Nuclide Id.
On the left, the components listed which has been included in the final activity fit. On the right, there is the list of all possible components (those where at least some of the gamma lines are found in the spectrum).
At the left side, there are two main groups under the Determinables (components which has meaningful lines): the Single components, where the activity of a decay could be determined could be determined on its own; and the Component groups at the bottom of the list, where the tightly coupled, highly correlated components can be found.
Node of each component can be opened by a double click on its name, and all the searched radiations are listed under it. If a library gamma found in the spectrum, it is designated by a colorful gamma photon sign. If no spectrum peak was found at its place, a pale icon is drawn, and the “Bkg” prefix is used. In this latter case, the amount of missing counts is also displayed which would be required at the point of the missing peak, but not found.
Regardless of the status of the radiation node, double-clicking it opens the Spectrum tab, and  jumps to the position of the clicked line.
Removing an unnecessary nuclide
HyperLab provides two methods for removing a component (decay) from the analysis:
This latter method can be demonstrated on the region which is shown on the picture: the isomer 71Zn is unnecessary, but automatically could not be removed, as it overlaps with 511keV annihilation peak, and the 389keV peak of 214Bi.
Right-click over it and remove from the nuclide identification fit.
Adding back an improperly removed component
If a necessary component has been removed from the Nuclide Id fit some way, and appears on the list of the All possible components, just select it, and click the arrow button pointing to the left. Thus the abandoned component will be included in the nuclide list again, and a nuclide identification re-fit occurs.
Energy calibration shift
When a slight difference can be observed in the position of the virtual peaks (displayed by the nuclide identification module) and the measured peaks, probably there is a small energy calibration problem.
Update your calibration, by adding this peak to the energy calibration points, and check if the nonlinearity part of the calibration is in effect.
There is a small chance, too, that the nuclear data is not perfect, but this is rarely the case.
Non-fitted spectrum peak found by Nuclide Id
Due to statistical nature of the measured counts, the spectrum peak deconvolution algorithm sometimes misses a peak. In several cases, those peaks are pointed out by the nuclide identification algorithm, as it utilizes the information from all peaks of the nuclide.
The picture shows the case when a small 214Bi peak is missed by spectrum analysis, but the nuclide identification found it.
Insert the missing peak manually to the region, and re-run nuclide identification.
Problem with the nuclear data
In some very rare cases, the nuclear libraries are not perfect: may contains errors for the energy positions or the intensities of library peaks.
One such example can be seen on this figure. As we see that while intensities and the positions of other 214Bi peaks are properly matching the measured peaks, there is a true outlier at 1416keV. This is probably comes from an intensity problem in the original nuclear data.
 
 
Copyright © 1998-2021 by HyperLabs Software