xps

Python tool for xps peak fitting

Auxiliary:

• gauss(x, C, mu, sigma), Gaussian function with coefficient C

s_orb-class:

• s_orbital(self, x, C, mu, sig), s-orbital constructed from one Gaussian

p_orb-class:

• dsep(self), get doublet separation
• split_p_orbital(self, x, C, mu1, mu2, sig), split p-orbital (p1/3, p2/3) constructed from two Gaussians

d_orb-class:

• dsep(self), get doublet separation
• split_d_orbital(self, x, C, mu1, mu2, sig), split d-orbital (d2/5, d3/5) constructed from two Gaussians

XPS-class:

• init(self, peaks, background, x_data, y_data), setup for guess of inital peak position and size, data to fit, and background.
• analyticPeaks(self, x, *args), function used to curve fit the peaks. Adds peaks to fit according to information in self.peaks.
• peakfit(self), does the peak fitting. Returns popt and pcov so the user may make their own plots if they wish.
• plotPeakfit(self), plots the peak fitting data.
• shirley_passive(self), xps passive shirley background.
• linear_background(self, x, a, b), xps linear background.

Usage

An example of how to use the program can be found in example.py. This code can be copied, and with a few small changes, used to curve fit other XPS-data. Firstly, change the variables path and filename to specify the desired peak to be fitted. The file must be of the type .txt and contain only x- and y-values separated by whitespace.

path = "XPS_data/"
filename = "Rb3d.txt"

Change the peaks variable according to what peaks are expected to exist. This variable is a dictionary where the keys are the strings 's', 'p' and 'd' describing the orbital type. The values are lists of lists with the initial guesses for height and position for the peaks, also containing the doublet separation (in eV) in the case of p and d orbitals. In example.py we have looked for two separated d-orbitals:

peaks = {'d': [[111, 4.0e+4, 1.48],[113, 1.0e+4, 1.48]]}

Here we have guessed that the tallest peak of the first doublet is located at 111 eV, with a height of 40000. For the second doublet we guess the location of the tallest peak to be around 113 eV, with a height of 10000. In both cases we are assuming we will find split d-orbitals of Rb, so we constrain both to have a doublet separation of 1.48 eV.

Resulting peak fit: