Enclosed you see the fluorescence detected from a steel sample alloyed with Ti, Cr and Mn, and the exciatation energy is varied. As you see, the Ti and Cr fluorescence is always excited, but the Mn fluorescence changes in intensity if the primary energy is increased from about 6.5 keV to 6.7 keV. And while the Fe fluorescence is not excited for these energies, it is excited at 7.1 keV photon energy. Since you excite with Cu K-radiation of 8.04 keV in a standard laboratory diffractometer, you alwys excite the fluorescence radiation of these elements.

In the lower part of the figure, you can see the intensity variation of the Mn K-radiation as the energy is scanned around the energy of the Mn K-edge. You can clearly see the distinct increase of the intensity at the edge - and the fine structures above the edges (this is the X-ray absorption fine structure).

O.k.? Regards to all, Dirk

Is it PXRD or GIXRD ?. If you have thin film sample and doing xrd in normal PXRD mode then you will get this kind of results. If it is a film, then I would suggest you need to optimize the Glancing angle while doing xrd first and then only you will solve your problem

This is PXRD only and not a thin film sample.

There can be several reason for this

Ist thing I can suggest is that crystallinity of your material is poor. you may try heating it at higher temperature.

2nd The amount of sample taken is not enough which would result in addition of large background (from glass sample holder).

Sample preparation is another issue. try keeping surface of the sample flat/uniform.

3rd x ray fluoroscence might be occuring which enhances the background. If the sample contains an element that can fluoresce absorbing a given xray wavelength this will occur. Change of X-Ray source may help.

1. I have prepared this sample at very higher temperature and it is crystalline enough.

2. The amount of sample is less as you mentioned.

And I have used Cu source.Thanks for ur answer Gohil . Will try now

According to my view this XRD pattern due to low sample and impurity you can take XRD again with the proper amount of sample because the low amount of sample cause the noise peak. The sample quantity for XRD is min 2 gram u can calcinate the sample in various temp and check the report

Make use of a "zero-background holder", and align the sample properly in the beam.

I have the feeling that the sample position is not correct - if you are working with a divergent beam (i.e. no Goebel-mirror in the incident beam optics of your diffractometer, then the correct height adjustment of the sample is a rather critical issue. If the sample is mounted too high or too low with respect to the correct position, then you may get such artifacts.

I have measured clear diffractograms from less than mg amounts, and the quality was good enough to do Rietveld refinements etc, using a rather simple powder diffractometer in a theta-2theta scan. But of course, asymmetric grazing incidence XRD would also increase the sensitivity to small sample amounts, but the powder must be dispersed on a very smooth substrate.

Sample rotation also helps to improve counting statistics. Or may use of a capillary instead of a flat sample holder!

All the above mentioned should help! Good luck, Dirk

@ Shanmugaselvan: It is definitely not true that you need at least 2 grams of material for powder XRD - instead 1 or 2 milligrams are enough! See the attachment, the data were obtained on a zero-background holder (tilted Si-single-crystal), the SnO powder material was simply dispersed on the flat sample holder which was rotated during scanning. The entire diffractogram took about 20 minutes only. Regards, Dirk

Very nice pattern Sir @Dirk . What would you say about changing X-ray source?

What are the elements your sample consist of?

Cu, Fe, O

So if you have large amounts of Fe, I would suggest to use a Cr - anode, because the Fe fluorescence is not excited in this case. Fe fluorescence might be a reason for the large backround in your diffractogram as well. If you do not have a Cr-anode, make use of a monochromator in front of the detector to suppress the Fe-related background! Good luck, Dirk

Thank u @Dirk

Your material contains magnetic elements in it which causes the fluorescence. But use large collection time per step so that atleast you will get all the peaks resoluted well and use background correction...

@ Karu- can u elaborate your statement " magnetic elements causes fluroscence"?

@Karu - the magnetic properties do not play any role here, it is simply the fact that their fluorescence may be excited using the material of the anode.

@satish

The magnetic properties of the material won't affect the xrd pattern. But whenever we use the materials having elements such as Cu, Fe, Co, Ni, Mn and Cr, we will get the pattern like this only, since the fluorescence is more when we are using Cu target. If u could not change the target, use lowest step width and high collection time per step, the peaks will be very high in intense compared to the background..

Yes this looks like Fe fluorescence from the CuKa radiation. If you do not have Co tube then it may require a different detector. I have an energy dispersive detector that will remove fluorescence (SOL-XE). Also depending on your detector you may be able to adjust the discriminator setting to omit the Fe fluorescence. I have a Bruker LynxEye detector and by simply changing the Discriminator Lower Level to 0.18 from the default of 0.11 pretty much all Fe fluorescence is removed - much easier than changing tubes! You do need to increase the counting times to compensate for loss of counts.

I also agree with comments about using a Si low background holder.

@Ian Slipper. Thank q. I would like to know the exact phenomenon of this Fe fluorescence from CuKa radiation.

The energy of the CuKa at 8.04keV is sufficiently high to excite the Fe atoms in your samples, though not very efficiently, at 6.4keV, and you then have a combination of both CuKa and Fe radiations with different wavelengths plus the continuum radiation entering your detector - this all adds up to a very high background which can swamp the signal from your peaks of interest. If you have Mn and Cr in the sample these will also fluoresce.

Enclosed you see the fluorescence detected from a steel sample alloyed with Ti, Cr and Mn, and the exciatation energy is varied. As you see, the Ti and Cr fluorescence is always excited, but the Mn fluorescence changes in intensity if the primary energy is increased from about 6.5 keV to 6.7 keV. And while the Fe fluorescence is not excited for these energies, it is excited at 7.1 keV photon energy. Since you excite with Cu K-radiation of 8.04 keV in a standard laboratory diffractometer, you alwys excite the fluorescence radiation of these elements.

In the lower part of the figure, you can see the intensity variation of the Mn K-radiation as the energy is scanned around the energy of the Mn K-edge. You can clearly see the distinct increase of the intensity at the edge - and the fine structures above the edges (this is the X-ray absorption fine structure).

O.k.? Regards to all, Dirk

@Dirk & Ian . Thank q... There is no different detector as well as different target other than Cu. So, now I should run the sample for long hours it seems...

No chance to manipulate detector electronics (i.e. discriminator settings of your counter) as Ian has proposed above? Check your diffractometer software in depth, usually there is a way to do this. Good luck, and/or much time to integrate ;-)

I wonder y I am getting this kind of noisy pattern for the other sample which is very shiny and also crystalline .. :(

And of course this has Cr now :P @Dirk

So this again is a proof that RG makes sense ;.)

With regard to my previous comment - just to motivate you to increase your integration time for summing up all the XRD data!

Well ! I had a tough time for the preparation of this compound itself, and now it's Cr 's turn :( Thanx @Dirk

Another solution to decrease the noise is to use a larger exit/divergence/detector slit and sacrifice some resolution. You will get more counts and better s/n.

one way to get the neat XRD pattern is to use different x-ray tube targets and find which target is suitable for getting clean pattern.

The best solution is to change the anode (Co), other than that u can use difracted beam monochomator, but it also reduce the overall signal.