Did You try XPS ? Please see this paper:

Article SiO2 film thickness metrology by x-ray photoelectron spectroscopy

@ Tomas Jarosz: Thank you. However, we need a fast way of mapping the thickness along and across relatively large substrates, in a large number of spots (tens to hundreds) and this might be needed sometimes on a daily basis. We thought more in the direction if there is a similar system to the one from Dektak or Tencor (which work with stylus profilometry), but working in non-contact and being able to cover the thickness range and the materials we use.

Have you looked into 3D optical surface profilers, e.g. Zygo (probably there are others too)? It won't work for every type of film on every type of substrate, but may complement mapping stylus profilometry.

Hey, Alan Savan! Thanks a lot, I am actually right now looking into this matter (3D optical profilometers). This is what I try to find out, which are the limitations of these kind of systems (the optical ones) and if they can replace the stylus ones for the kind of applications we need. I suspect that there should be some kind of drawback (resolution, price, etc.) . Have you worked with such system? Greetings from Linz!

Dear Cezarina

Are you sure the Alpha-step IQ profilemeter with precision of 1 angstrom  is not suitable for your thin film and large substrate?

Best,

Hamidreza

Hi Cezarina, I have used a Zygo, but many (20+) years ago so I hesitate to say the limitations and which applications it can and cannot do now. I'm sure that there have been many instrument and computational developments since. Also, at that time it was considerably more expensive than a stylus profilometer, but that gap may have meanwhile closed.

There are also x-ray fluorescence, eddy current monitors, prism coupler (made by Metricon) techniques for getting thickness, but these were all for fairly specialized applications rather than a general, wide range of materials, substrates and thicknesses.

Hi,

You can try RBS ( Rutherford Backscattering spectrometry ) , its widely used to determine thickness and compositions of this films and multi-layers of all materials in one shot measurement. 

RBS

provides composition Depth Information, Well Suited For Thin Film Analysis,Typical analysis times are 10 minutes or less.High Sensitivity,

You can also conduct ion channeling , and check sample crystal form.

this will provide the following information, Amount and depth distribution of lattice disorder,Location of impurity atoms in lattice site and  thickness of amorphous or defective epitaxial surface layers.

May be you can use a new type of ellipsometer, which is just under developement.This instrument is measuring a bigger area at same time (18 000 000 pixels in about 5 seconds). If you do a measurement before treatment and after treatment, you have a chance to calculate the thickness distrubution for the area. The thickness for metal films is limited. If you give me refrative index data (if availible) I can make calculation, if this instrument can solve your problems.

Dear Hamidreza,

Thank you for the suggestion. Since yesterday afternoon I have been in contact with the people from Tecor, they have already made some suggestions concerning contact - non-contact systems. I will check the specifications, possibly I will send some samples and, if the case, decide which kind would be suitable :).

Hi again, Alan! Yes, the optical ones are more expensive, I will try to figure out if is is worth the money ;). As you mentioned, there are other ways, specialized indeed. We have already used our XRF for thickness measurements, but not for light elements. We basically need to know the thickness of our libraries in a decently fast way and with the scanning possibility, so a stylus one would do it with no problems, but if there is something better (?) at a reasonable price, why not? :)

@ Makram Abd El Qader. Thanks, this is a precise, though fancy tool and we do not have access to it... :)

@ Lior Kornblum: Thanks, we do have a diffractometer and we will also try the reflectivity measurements, at least for the very thin single layers which have an uniform thickness distribution.

@ Udo Riß: Thank you for the answer and for the offer! However, for combinatorial thin films where you may have a continuous composition change the availability of optical data is indeed the biggest issue. One can only extrapolate assuming a combination of optical parameters from the 2, 3, etc pure components but unfortunately the results might not always lead to the true result.

For getting a trustful result with ellipsometry the type of mixure must be calculated. At first you must defined the expected type of mixture. So here three different ground  types are possible.

1. Mixture on atomic basis

2. Medium A solved in medium B

3. Macroscopic micture

With an adequate simulation software you can reduce the number of possible results significantly, so that you have a realistic chance to get a trustful result. 

But you must make sure that you have no optical depolarisation.

SEM

for the determination of the thickness of a thin film, you can use:

- the optical method: this method is very simple if the transmission or reflectivity spectra present  interference fringes. Using this method you can access to the thickness and the refractive index of  the film in the transparency region. http://www.chem.agilent.com/Library/applications/uv90.pdf

- Scanning Electron Microscope (SEM) : cross section observation:

http://static.ifp.tuwien.ac.at/homepages/Personen/duenne_schichten_alt/pdf/t_p_ds_chapter4.pdf

- Atomic Force Microscope : if the thickness is in the order of few tens of nm, this method consists to scan a region in the edge of the film, the film thickness of the film is extracted from the z-profile  (difference the Zmax corresponding to the film surface and Zmin corresponding the substrate)

http://www.intechopen.com/books/physics-and-technology-of-silicon-carbide-devices/ion-synthesis-of-sic-and-its-instability-at-high-temperatures

Please, page 18.

Hi,

We've been using Rigaku's Z??400 with multiple crystals to give us sub 0.5Å sigmas in 30s to 5min for Mg through Te using 10 to 30mm spot sizes.  You will need to have custom holders made for your samples. Calibration from thin up to 0.5µ is pretty straight-forward, but as you go thicker, there are non-linearities that should be checked with standards of similar composition and thickness. Also, remember that you are measuring metal atoms/unit area so that if films density is changing, you are not measuring a physical thickness.  Always RBS to characterize your standards.

We also have a reasonably modern Zygo optical profilier.  It is fast, but I think you should see if it really gives accurate results without good knowledge of the optical properties of substrate and films.

Hi Bryan Hendrix,

Thank you for your time!I guess before deciding if we will buy a profilometer (either with stylus or optical), we will send our samples to be measured by the producer of these equipments and have a really detailed discussion about which parameters are required as input for calculating the thickness with the optical ones (the manuals and brochures the providers are so general and non-specific!). XRD for mapping the thickness is not really an option for us since our films density would most probably change and we do not have the means of measuring so fast and with so small spot size anyway :).

Sorry.  I didn't make it specific.  The ZSX 400 is an XRF not XRD.

The Zygo might do just what you need.

Imaging spectroscopic ellipsometry is a nice non destructive tool to characterize microstructured thin films (even up to 5 microns). We useing the Ep4 here at the Institute as well.

Best regards

Christian

Eddy current is a nondestructive technique that allows you to measure the thickness of metallic plates without contact. A thickness of several micrometers can be measured. An absolue tool, Conducsens(tm), can determine the thickness without complex calibration (it only needs 1 point reference). See the Sciensoria web site.

Best regards

An example of determining the film thickness by X-ray reflectometry is shown in [1]. This method can determine the film thickness in the range of 2-300 nm.

http://download.springer.com/static/pdf/720/art%253A10.1134%252FS1063783414110237.pdf?originUrl=http%3A%2F%2Flink.springer.com%2Farticle%2F10.1134%2FS1063783414110237&token2=exp=1457157606~acl=%2Fstatic%2Fpdf%2F720%2Fart%25253A10.1134%25252FS1063783414110237.pdf%3ForiginUrl%3Dhttp%253A%252F%252Flink.springer.com%252Farticle%252F10.1134%252FS1063783414110237*~hmac=46005f4368f1dfe6d2bf52f62c00b054a6c78d064762deb3a22b49358c4a87fb

The thickness of the step between the film and the substrate can be determined by atomic force microscopy, semi-contact method. We determined 50-1000 nm.

Which instrument did you bought for the thickness measurement of your combinatorial thin film area of 100 of spots.

In the end, we have not bought any instrument. We are using a DektakXT stylus profilometer from Bruker from a partner institution.

I tried a 3D optical microscope (one still needs to have access to the uncoated substrate though), but the results were not as accurate for our thicknesses ranges and materials as from contact profilometry.