Hi,

As far as I know, u can do it in two ways.

1. Obviously, the best and better way: grazing XRD

2. Rough way but still works and u have to sacrifice your sample: Scratch the coating and collect all the flakes/powder then do powder XRD

Hi,

in my opinion if you do not get XRD reflections from your film it means that there is no well-ordered crystallites. Or you have very specific crystallographic texture. Of course you can try grazing or offset XRD measurements.  However, in my work I measure films as thick as 10 nm and I observe "normal" XRD reflections.

Yes u can us GXRD technique.

For known thickness and packing u cud calculate depth of penetration, further incident angle. But if not than u can try to vary the incident angle until the substrate peak disappear.

I suspect that the signal of your thin layer is simply too weak to be detected in the presence of stronger reflections (and background) due to the substrate, granted that the reflections you are interested in are not superimposed with those due to the bulk phase.

You might consider one of the following strategies:

(i) You should perform a grazing incidence XRD experiment. This is absolutely the best choice in my opinion, as it would allow you to disentangle just the diffraction contribution of the deposited layer from the signal due to the bulk;

(ii) You might also consider to perform electron diffraction experiments. For example, the LEED technique would give you information on the surface structure, from which the bulk crystal structure of the deposited material could be inferred. Also the diffraction pattern due to backscattered electrons recorded in a SEM (EBSD) could help.

I think that the thickness  of your film  could be sufficient to see an XRD pattern with classical powder diffraction. Is it possible that the deposited metal could form an amorphous oxide layer?  

Probably the time per step is not enough to see the peaks, in my opinion 0.26 s/step is too low.  Anyway, the best techinique to analyse thin films is the grazing x-ray diffraction, the angle of incidence depends on the absorption coefficient of the material and the film thickness, two degrees is in general a good choice (but depends on the material). The time for step depends on your diffractometer and the intensity of the peaks, remember that in general the deposition process of thin films generates preferred orientation of some peaks. 

Please post your data with all the experimental details. Also share the chemistry of the substrate and the film for the best feedback from the expert RG membership.

Some very useful comments!

One caveat is that for grazing incidence it is essential that your sample surface is optically flat and is at the correct height, so please take extra care to make certain this is the case.

Thank you all very much for your helpful advice!

So, a little update on the situation, in case it is helpful: we've observed a cross section of our sample with SEM-EDX and it appears that the deposited layer is not as thick and regular as expected. So I guess that's definitely problematic for powder XRD.

In any case, I would like to attempt doing grazing angle incidence XRD on our samples. However, I've read that grazing XRD requires a parallel beam geometry, whereas our powder XRD has a Bragg-Brentano geometry. Is it still possible to extract useful information using this equipment, or are parallel beam optics an indispensable condition?

Again, many thanks!

"10-70° 2θ (step size 0.017°, 0.26 s/step)" - Number of steps=(60/0.017)=3529!

At 0.26 sec/step that amounts to around 15+ min (I'll double check the calculation again) and someone thinks "Probably the time per step is not enough to see the peaks, in my opinion 0.26 s/step is too low"? Probably right with the conventional tools. I've collected reasonable resolution conventional diffractograms at 1o/min many years ago.

Back to the future fellows! You are at least a century behind. The sensitivity of your XRD data acquisition tool is lousy. LOL!

On a positive note, observing the formation of the intermetallic (if  your hypothesis is affirmed) may be done in situ with XRD using real time transmission Laue techniques and other reflection methods as well. It all depends on the signal to noise ratio (SNR) of the reflection of interest and the sensitivity of the detector you use.

Nothing wrong in using the B-B geometry for feasibility. Generally, the focussing optics should yield higher SNR. However, in GI application the X-ray flux would be significantly diminished due to defocusing at the sample surface, as you correctly concluded.

https://www.flickr.com/photos/85210325@N04/10221065324/in/set-72157645018820696

Sornadurai! Very much interested in collaborating! What XRD equipment and detectors do you have access to at  Indira Gandhi Centre for Atomic Research?

Parallel beam or focusing geometry? What detector do you use? 0D conventional point counter?

Just to clarify! The general term XRD includes XRR, SAXS, WAXS, GIXRD (GIWAXS), GISAX and all other methods as well. Now in this case of a thin film my approach would be:

https://www.flickr.com/photos/85210325@N04/7920887956/in/set-72157632728981912