Hi all,
Anyone may suggest one nice and friendly open software to analyze and model ellipsometer measurements?
Thanks in advance
Jose-Maria
A good free software for modeling ellipsometry measurements, as for transmission and reflection spectra is RefFit by Alexey Kuzmenko (University of Geneva). It contains a lot of models. I use it often for our experimental data, especially for ellipsometry and reflection spectroscopy in far- and middle infrared regions (phonon and magnon modeling). Website of RefFit program below this text.
http://optics.unige.ch/alexey/reffit.html
Hi Jose Maria,
Although there is no 'free software' in ellipsometry, there are some articles and book chapters that can help you. For instance, this chapter of the Handbook of optics by Azzam could be very usefull (http://photonics.intec.ugent.be/education/IVPV/res_handbook/v2ch27.pdf). You can send me a message for further details. All the best!
What are you looking to do specifically? You can probably use a program like FreeSnell or OpenFilters to simulation layers and measure their respective ellipsometric parameters (Phi, Delta) and maybe a program like ReFIT would help with the modeling?
Hi Kaleem. Maybe I am confused but I was interested in one GNU (free) software, and your recommended link is not open. Thanks anyway
This (the link) may not be what you are looking for but it is not difficult to write one yourself with open source computing languages.
Article FILM THICKNESS DETERMINATION BY MATLAB ASSISTED ELLIPSOMETRY
A good free software for modeling ellipsometry measurements, as for transmission and reflection spectra is RefFit by Alexey Kuzmenko (University of Geneva). It contains a lot of models. I use it often for our experimental data, especially for ellipsometry and reflection spectroscopy in far- and middle infrared regions (phonon and magnon modeling). Website of RefFit program below this text.
http://optics.unige.ch/alexey/reffit.html
I have recently published the 2.0 release of Regress Pro, a free software application for spectroscopic ellipsometry and reflectometry.
It does have an easy to use user interface, many different optical models to describe the dispersion curves and use the Levenberg-Marquardt well-known algorithm using the GSL library.
In addition it is free software so that everyone can "look inside" and make improvements if needed. The project is hosted at github
https://github.com/franko/regress-pro
and the binary packages are available for Windows and Linux from the release page:
https://github.com/franko/regress-pro/releases
The project webpage gives some details about the software and points to the user manual:
http://franko.github.io/regress-pro/
I hope it will be useful for other people who want to analyse ellipsometry data and don't want to use closed source proprietary software.
Dr Francesco Abbate, can you add a sample file for learning to use your software. At this moment, your documentation is not proving helpful. [email protected]
Hi,
I am sorry the documentation for Regress Pro was not clear or complete enough. I will try to improve it but if you give me some hints about what is actually lacking I will be able to improve the documentation more effectively.
As for your specific problem, please be assured that Regress Pro is able to treat ellipsometer data from many common configurations. In the release package there is an "examples" directory with some useful examples that can be used as a starting point. For example in the directory examples/resist-ellipsometry you will find an example of an SE spectrum and a recipe file that you can open with regress pro.
In general, the steps to analyse data are:
- write the data in a format that regress pro can read (the software accept Tan(Psi) Cos(Delta) format but also SE alpha/beta)(*)
- with the software configure the film stack starting by defining the substrate and any additional film that is present on the sample
- choose appropriate dispersion model for each film. A library of common films is included with the software. Common software models are Cauchy, Harmonic oscillator, Tauc-Lorentz.
At the beginning you can work with the interactive fit to rapidly make some tests. In any case the goal is to fit the appropriate parameters to get a good agreement with experimental data.
(*) The format SE alpha/beta is much more convenient to perform fit because it does ease the convergence of the fit. I strongly advice to convert the data to SE alpha/beta format. This can be done in any case by using a "fake" analyzer angle of 25 degree.
Thank you for your message. Yes I did find the examples folder after messaging. I created a *.nk file for my substrate looking at example file thermal-siO2.nk
it has a first line 2 0.24621 0.9984 474. What is 2?
Yes I am giving SE alpha/beta format for my data file.
I think I have done things fine.
But my data is not getting fit. I am sending you the input files I created. Could you look and tell me if I am doing something wrong. I'll appreciate if you can show me the results of this. Most probs I am giving bad guess for TL model.
- Badges
- Science topic
- Similar topics
- Chemistry
- Nanotechnology
- Nanostructures
- Thin Films