When collecting spectra from your sample the volume from which X-rays are generated is defined by the electron interaction volume, the size of which depends on the material being analyzed and acceleration voltage being used. If the interaction volume is completely enclosed in a single phase or particle, then the spectrum observed will contain data just from that material. However, if the phase or particle is smaller than the electron interaction volume, electrons will penetrate into surrounding material (matrix) and generate X-rays. As a result the spectrum observed will contain X-rays from both the phase/particle being analysed and the material (matrix) surrounding it. If you suspect this to be the case, a second spectrum can be collected from the material surrounding the phase/particle being analyzed to act as a reference spectrum.

Such spectra can be presented as a comparison by overlaying one on top of the other. This is achieved in the ‘Compare’ steps of the ‘Analyzer Navigator’ and ‘Point and ID Navigator’. The overlay shows the differences between the spectra, but you need to look for the differences to make the correct interpretation. The spectra shown to the right are those of an inclusion (red) contained in a steel matrix (yellow).

Another way to present these spectra is to subtract one spectrum from the other, to display only the differences between the two spectra. This can result in the removal of the matrix peaks from the subtraction spectrum. ‘Subtract Spectrum’ is accessed through the tab at the bottom of the ‘Compare’ screen. 

First select a spectrum from the Spectrum 1 list (A) and then select a second spectrum from the Spectrum 2 list (B). The spectrum subtraction performed will be:

Spectrum 1 – (Spectrum 2 * Multiplication Factor) = Subtraction Spectrum

Both Spectrum 1 and Spectrum 2 are displayed as line overlays and the result of the subtraction spectrum is displayed as a solid yellow spectrum. Spectra can be dragged into the Spectrum 1 and Spectrum 2 lists from other Projects, Samples or ‘Sites of Interest’ displayed in the ‘Data tree’.

By default the multiplication factor is set to 1 for the subtraction, but can be varied to account for differences in intensities between the spectra being subtracted. With the correct value set, it can lead to the elimination of matrix peaks in the subtraction result. In the example to the right, the Matrix spectrum is subtracted from the Inclusion spectrum using the default multiplication factor of 1. The resulting subtraction spectrum (yellow) shows an over subtraction of the matrix spectrum, producing ‘negative peaks’ associated with the main matrix component, iron.

By adjusting the multiplication factor to scale the proportion of the matrix spectrum being subtracted, the matrix peaks can be eliminated from the subration result. The multiplication factor can be adjusted manually and the result will be shown in the spectrum display. Alternatively, INCA can calculate the appropriate multiplication factor to set the subtraction result to zero at a point that you select in the spectrum. Simply click on the point in the spectrum where you wish to set the result to zero and press the ‘Zero Result Spectrum at Cursor' button (see image to side). In this case, we have assumed the Fe peak is exclusively from the matrix, therefore has been scaled to be completely removed from the subtraction result. To do this select a point in the spectrum at the centre of the FeK? peak, by clicking on it with the mouse, then press the ‘Zero Result Spectrum at Cursor' button and INCA will calculate the appropriate multiplication factor and apply it to the subtraction.

Having selected the centre of the Fe K? peak, INCA calculated that a multiplication factor of 0.53 will bring the subtraction spectrum result to zero at this point. This results in the removal of the Fe peak, the main matrix component, from the resulting subtraction spectrum. INCA will automatically generate a name for the subtraction spectrum and this appears in the result box, pressing ‘Save’ will save the subtraction spectrum to the Data Tree. Alternatively, the name in the result box can be edited before the subtraction spectrum is saved.