In most of the profilometers, either Stylus type (example: Dektak) or Optical Profilometer (example: Zygo), you need to create a physical step in the film during deposition. This can be created quite simply as explained by Ross Birney. But the important point is the sharpness of the step boundary. The sharper is the step, the clearer is the scan height measurement and so accuracy is better in film thickness estimation. If the step boundary is blurred or has a gradient in material thickness in itself, the scan height elevation has a slope and sometimes it becomes difficult to determine the level of the film plane, hence inaccuracy results in thickness estimation.

Adithi Umamaheswara · Can you please elaborate a little on FTP(Film Thickness Profilometer) which, you said, does not require a step?

FTP(Film Thickness Profilometer)

The Dektak 3ST is quite good (for reasonably thick films, at least).

If I remember correctly, the step height range is about 100Å to 1300kÅ.

It's called an AFM :)

Have you tried interferometry? A simple optics apparatus and green laser is able to give you a good resolution (if the thin films have low roughness)

But for Dektak you need to have a step height. But profilometer does not step height

It's easy to create a step though, just mask an area of the substrate with photoresist, Kapton tape, small piece of Si wafer etc. during deposition.

Not necessarily the most accurate method of thickness determination, but can be useful.

Ellipsometry and Fabry-Perot (cheap+simple) resonances might help if the film is transparent.

Magnetic response may work if the substrate is metallic.

Impedance spectroscopy may work if the substrate is conductive.

Best idea is keeping track during the deposition process, if that's possible: Interferometric or by means of detuning an oscillation quarz.

Filmetrics may be of your choice: http://www.filmetrics.com/

In most of the profilometers, either Stylus type (example: Dektak) or Optical Profilometer (example: Zygo), you need to create a physical step in the film during deposition. This can be created quite simply as explained by Ross Birney. But the important point is the sharpness of the step boundary. The sharper is the step, the clearer is the scan height measurement and so accuracy is better in film thickness estimation. If the step boundary is blurred or has a gradient in material thickness in itself, the scan height elevation has a slope and sometimes it becomes difficult to determine the level of the film plane, hence inaccuracy results in thickness estimation.

Adithi Umamaheswara · Can you please elaborate a little on FTP(Film Thickness Profilometer) which, you said, does not require a step?

It is almost like a ellipsometer, where you don't need any step height. It can also give the refractive index. In ellipsometer you can get the thickness almost up-to 8 layers but in FTP 2 layers maximum. FTP is less costlier than ellipsometer.

Wyko NT1100 White Light Interferometer/profilometer

Thanks, Adithi. Is the principle same as Ellipsometer?

thanks alot guys..

@ Adithi - Can you suggest any good company or model of the FTP that you are talking about?

you might want to try the LEXT from Olympus. I have obtianed really nice data with the 3D images... the bad part is that it is $100K

FTP what i am using is SENTECH ftp adv 100 but i am not sure of the cost. Only thing I can tell you is it is less costlier than Ellipsometer

In some situation it could be useful the channeled spectrum effect.

It permits to evaluate the optical thickness of the (transparent) film with extremely high precision!

We use Ambios Profiometer for the thickness measurement in the nanosale resolution.

Take a look at this one, we use it and works pretty well.

http://www.sensofar.com/sensofar/products/thin-film/

After more than one decade devoting to organic device manufacturing (www.oog.escet.urjc.es), we use a KLA Tencor D120 contact profilometer to characterize thickness of our thin films, which range from 5 nm to upwards. It is perfect for this purpose. Vertical resolution of this equipment is higher (in the order of angstroms) but is dificult to measure thicknesses below 5 nm, either in organic films or oxide graphene multilayers, due to surface roughness. Lateral resolution depends on the stylus diameter. We use typically a 2 um one. You can find stylus with lower radius at higher prices. But in this case one should be cautious with the exerted pressure and the hardness of the material. Minimum force in this equipment is 0.03 mg. It does not cause dammage over a typical polymeric layer using a 2 um stylus. You can scan macroscopic areas up to 55 mm! (tremendous advantage over typical AFM). The Price in 2011 was 45000 eur.

To avoid the problems as stated by Angel L Alvarez, one can use Optical Profilometer which is completely non-destructive and unobtrusive. There are several good Optical Profilometers in the market like Zygo, Taylor Hobson, Bruker. The vertical resolution of this instrument is very high i.e. in Angstrom level like AFM. Lateral resolution is poor though with respect to AFM.

I use dektak 150 profilometer with 1 Armstrong of vertical resolution. Now, there is a new one:http://www.bruker.com/products/surface-analysis/stylus-profilometry/dektak-xt/learn-more.html

I agree with Adolfo, a Dektak 150 profilomter could be a best solution.

Hello again! Yes, AlphaStep D120 (by KLA Tencor) or Dektak 150 (by Bruker), are very similar and compete for the same niche. We considered both possibilities, and decided finally for the D120 due to minor details, for example, because the inspection camera had software controllable zoon (not manual) with autofocus in a larger range. This was important for us but is a minor detail. Apparently, resolution and reproducibility of the D120 were a little bit higher, but the salesperson from Bruker defended their control electronics was better. Both are good options.

If you deposit an epitaxial layer on a single crystal substrate you may use MEED (Medium Energy Electron Diffraction) oscillation method for thickness measurement, which consist of measuring medium energy electron diffraction intensity in specular direction at grazing incidence. Each oscillation of intensity would correspond to one atomic layer of deposited film.

in my opinion it is impossible to give a general response to your question, if I understand correctly your question. If you need indeed nanoscale profilometer, than your "probe" should have to stay in sub-nanoscale. You have to use for example AFM systems with enhanced resolution. For example assemble the CSWNT on the edge of AFM tip and focus a laser on CSWNT edge. This could give you sub-10nm resolution. However it is definitely not a cheap and simple option.

You have to describe in more details your experiment to get some reasonable advice. For example: if you have the NLO chromophore (molecules with large dipole moment) self-assembly, than you could use a methodology developed by Tobin Marks and Shlomo Yitchaik in mid 90s. If you surface have any kind of chemical functionality on the surface, than you could use affinity mapping of the surface, so... in some practical cases you might find solution of your problem, but it would not be easy. MEED is a good technique for most nanoscale application but it is not a low budget instrument. so good luck

You may use OceanOptics NanoCalc for low budget it max resolutin until 0.1 nm

http://oceanoptics.com/product/nanocalc/

Dear Sir,

Please use AFM for film thickness!!

you can use Dektak XT (Bruker)

Immanuel Paulraj what is the estimated price of this instrument?