I'm looking for a way to measure the uncertainty (standard deviation) on the quantification (area) of the components of an XPS spectrum using CasaXPS.
I found these options in the software, but they don't satisfy me:
1) From the "Quantify (F7)" window, in the "Regions" tab, clicking on "Calculate Error Bars" but it is independent of the fit and changes with each click.
2) From the "Quantify (F7)" window, in the "Components" tab, by clicking on "Monte Carlo", I obtain only the relative value and not the absolute one. But above all, the values do not follow the goodness of the fit: even with components that are not fitted and clearly incorrect, the value is low.
As I have not found these methods to be reliable, my idea is to use the RMS as an estimation of the error on the sum of the areas of the components and then obtain the percentage error of the individual components.
My scope is to provide the composition of my sample, and also the percentages of the various components, for each element, all with their measurement error.
Does anyone know if there is a more automatic and reliable method?