You can perform SEM in the cross section to get more accurate thickness measurement. AFM might not be an appropriate way to measure thickness, since it only give you a surface peak-valley information, unless you use some kind of indentation to pierce through the film to the substrate.

It depends on the thickness range you expect. If the expected value is in the >100 nm range you may use standard SEM on sample cross section. In the 30-100 nm range a FESEM (Field Emission SEM) is better. For thinner samples tou will need a TEM cross sectional analysis, that however require a proper sample preparation. Of course, some special care must be taken in all the above measurements. First of all, the sample must be properly prepared even in the case of SEM analysis. For example. if the interested film is the topmost one, you will need to deposit a dummy thin coating (e.g. a thin Au layer) before cutting the cross-sectional sample. Any SEM or TEM experts may provide you with the best recipes to properly prepare your samples for the analysis.

In recommendation Curro Sir,  SEM and FESEM are the best candidates for your problem . 

I agree with Vinod and Giuseppe. Giuseppe has provided the requirement and limitation in more details. If the thickness is below 30-40 nm, it will be difficult to identify from cross sectional SEM, whether it is SEM or FESEM. Also, sample preparation for TEM cross sectional analysis is not an easy task.

As your films are grown via PLD, you may check with X-ray reflectivity (XRR). For this, deposit your material on a suitable substrate (say Si). Otherwise, ellipsometry would work better for finding the thickness.

Find the attachment for "XRR measurement".

Best,

Chandan

Dear Pravin Vaidya,

A very elegant and precise way to measure thickness of a film is using interference oscillations, if you observe it! If that is the case I can sent you a paper with a detailed explanation of the method.

Best regards,

Pedro 

use Rubidium back scattering 

I recommend Ellipsometry as a simple way.

Sometimes, synchrotron X-ray scattering may work (using micro- or nano-beam) but it often uses for other purposes than determines the thickness.

Ideally, ellipsometry should work as suggested above, although in practice it is pretty much user dependent if you're studying non-conventional systems. The thickness determination strongly depends on the model assumptions made for the system.

I wouldn't discard AFM so fast. There are some ways you can use AFM for thickness determination of your PLD film: 1) make a scratch with a needle or sharp object in order to create an interface between the film and the substrate beneath such that you can measure the height difference with AFM. 2) this is one is more tricky, analyze a cross-section of the sample with AFM. The difficulty with this option is that there is a risk of tip damage when scanning the PLD/air interface. In any case, what you care is to measure the distance between the air/PLD and PLD/substrate interfaces. You can check here something like that we did a couple of years ago. I could also give you some ideas on how to execute the experiment if you're interested.

Good luck!

Article Raman, AFM, and TEM profiling of QD multilayer structures

There are many methods to obtain film thicknesses.  Your best choice can only be determined from the thickness rage of your films, the film quality, the substrate quality, the accuracy you require, you budget and more.  I have discussed many methods in my book: Surfaces, Interfaces and Thin Films for Microelectronics, John Wiley, 2008, ISBN 978-0-470-17447-0. Before you make a choice you will need to better define your problem.

Well. There is a lot of different methods.

1.You can use a mask before your deposition and then to measure a step by a surface stylus profiler.

2. XRF method with a special soft for film thickness measurements. The same for EDX with variable electron beam energy and de-convolution.

3. Any thin film analysis method: RBS, SIMS, AES or XPS with ion sputtering...

4. Elipsometry and cross section analysis as it was mentioned above...

Depends on thickness and material analyzed.

For thickness