Cu Auger Line Curve
Fitting CasaXPS 2.0.1
Problem:
Given an unknown spectrum, determine the relative contributions from Cu2O, CuO and Metallic Cu to the unknown spectrum.
Four
spectra are displayed in Figure 1, three of which are believed to contribute to
the fourth spectrum. The three know spectra can be used to construct
line-shapes forming the basis for a nonlinear optimisation, the results of
which estimate the relative proportions on these know spectra to the unknown data.
Figure 1: Cu Auger lines. Data supplied
by: Institute of Physics,
The
following steps are required to perform the analysis:
Construct background
subtracted line-shapes from the data.
For each of
the spectra required by the peak-fit, add an appropriate background and create
a new VAMAS block in the data file consisting of the background subtracted
data.
- Add a background using the
Regions property page on the Quantification Parameters dialog window
(Options menu “Quantify …”).
- Subtract the background from
the data using the Test Data property page on the Spectrum Processing
dialog window (Options menu “Processing …”).
- Copy the processed data to a
new VAMAS block by right-clicking over the right-hand-side of the Experiment
Frame and selecting the Data Only tick-box Figure 3.
- Adjust the names of the new
data using the Edit Info toolbar button
Figure 4. The Block Ids for the new
line-shape VAMAS block must start with a % character (Figure 5).
Figure 2: Preprocess the basis data using the Background Subtract option.
Figure 3: Copy the processed line-shape to a new VAMAS block.
Figure 4: Edit the line-shape VAMAS block and ensure the Block Id starts with a % character.
Figure 5: Right-hand-side of the Experiment Frame following creation and renaming of the Block Id, Element and Experimental Variable fields in Figure 4
Define a peak model on the
unknown spectrum in terms of the new line-shapes.
- Using each of the original
basis spectra fit the new line-shape derived from that data to the basis
spectra. This provides initial starting points for the peak parameters
ultimately used to create the model for the unknown spectrum. Create a
component on the basis spectrum and enter the Block Id for the appropriate
line-shape into the Line-Shape field on the Components property page (see
Figure 6 for an example).
- Fix the fwhm
parameters derived from fitting to the original basis data and also limit
the position parameters to a small interval around the value determined
for the basis data. Note the fwhm parameter
should be close to unity owing to the requirement that the line-shape from
data should not be deformed away from the shape of the basis data.
- Copy and Paste the components
from the basis spectra onto the unknown spectrum. Manual adjust the
components to guide the auto-fitting and press the Fit Components pushbutton
on the Components property page.
Figure 6: Model defined from data. Note the Block Id of
the line-shapes is entered into the Line-shape field.
Figure 7: A possible fit to the unknown spectrum using the line-shapes taken from known spectral forms.