I would like to know it is possible to find out thin film thickness from the reflectance pattern that have been taken from the UV-Vis spectrophotometer. I have read FFT numerical technique is there, I hope some relevant answer with references for this question.
Thank you all in advance.
How much expect the think film thickness ? why not try to use SEM of cross section ?
Dear mustafa,
The expected thickness is about ~30nm. Actually, SEM is not available in our department, that is why I'm trying to get film thickness from UV-Vis spectrophotometer interference data for this purpose. I hope you will help me in this context
Dear Shamjid Palappr, check again your measurements. If the thickness of your film is 30 nm it is not possible an interference pattern in R or T to be observed in UV-VIS-NIR spectral range. What type is your substrate?
Swanepoel route is very famous and efficient when Transmittance pattern contains more than 3 fringes. Numerical methods like PUMA can help you much.
If the film is deposited onto known substrate (I mean you know substrate's refractive index and extinction coefficient) you can use non linear minimization of the discrepancies between measured and calculated R spectra for deriving optical constants and thickness d. The simplest way is to measure R in the transparent region of the film. We use non linear fitting for determination of n, k and d of sol-gel derived Nb2O5 film - see the attached paper.
Dear Ali Abdolahzadeh Ziabari, I absolutely agree with you that Swanepoel's method is very useful and accurate method. The only disadvantage is that the studied films should be thick enough in order interference pattern to be observed. My experience has shown that you need films thicker than 500 nm to be able to apply Swanepoel method. In the case of thinner films single-wavelength methods (see the attached file) or non linear curve fitting should be used (see the answer above). Regards
Dear Tsvetanka Babeva, the main problem for all nonlinear curve fitting is to guess the initial values. In your works (e.g. in "Optical characterization of sol–gel derived Nb2O5 thin films") you've implied a method that it doesn't need any dispersion model, but you need manage a detailed experiment that may be impossible elsewhere. I do appreciate your route. But by an appropriate modeling, u can get very reliable results. Some strong algorithms like PSO, GA or IC can do that very easily, though you need to have the basic knowledge about them.
- Badges
- Science topic
- Similar topics
- Chemistry
- Nanotechnology
- Nanostructures
- Thin Films