can we do by random fitting and choose the best? Because voigt fitted most the graphs using peak fit
Since a Voigt peak considers mixing a gaussian with a Lorentzian, it is the normal result that you will obtain a better fit with it.
I recommend checking the literature on each specific case. XRD might work better with Lorentzian peaks. Raman with gaussian. You should use a Voigt peak only if bad fitting is obtained for both gaussian and lorentzian.
Also you should check the theory involved. For example, in XPS fitting is preferable to have all peaks of same width, etc...
A Lorentzian best describes a perfectly homogeneous system, a Gaussian one in which in homogeneous broadening influences the lineshape.
A weighted mixture of the two (Voight) is probably best, allowing the weighting to be a fit parameter.
- Badges
- Science topic
- Similar topics
- Mathematics
- Algebra
- Matrix