Calibrating StoeStadiP Cu Diffractometer with Si NIST Standard

sdk_

Hi,

This seems to be a simple issue but I could not find a detailed guide on how to correctly calibrate a powder X-ray diffractometer with a NIST standard. The main issue I am facing is how to refine the 2theta offset. In my mind it does not make sense to refine the lattice parameter of the standard, nor does it make sense to refine the wavelength of the X-rays. Refining only a zero error, however, does not yield good matching of observed vs calculated reflection positions. I then used some commands I found on this forum but it was used in the case of having an internal standard:

prm !zero 0.04103
prm !step_error 0.96455
prm !square 0.00011
prm !data_middle 61 'this is average 2-theta value e.g. for a 10-90 degree scan
th2_offset = (Xo-data_middle)²*(square/1000) + (Xo-data_middle) (step_error/1000) + zero;

Refining the 2th offset of the machine in this way gives a good fit to the data, but I am not sure if this the correct way to do it. I attached the powder file and the input file (topas 6). Advise is welcome!

Thanks,
Simon

cu-rietveld-standard.txt

8kB

joa-x-si-nist.txt

160kB

johnsoevans

It depends what you're trying to do.
If you're happy to be purely empirical the approach you give above is probably ok. It basically says "I'm going to apply a correction to put the peaks of Si in the right place and I don't really care what it means". Providing your sample is measured in an equivalent way you can probably apply the same correction. That's why it will work with an internal standard. If you do this, you could plot the th2_offset equation in e.g. excel and see how bad the offsets are.
If you want to take a more rigorous approach then you need to think what the origin of the offset is. Depending on the experiment and instrument you might need to refine a zero point and/or (usually or) a sample height correction. If it's a capillary you might need to refine a capillary offset parameter. If the values are too large you might need to go back and align the instrument.

sdk_

Hi John,

Thank you for your quick reply. I attached the plotted th2_offset function from the initial try. I dont know if this is any good.

I got similar refinement results when using a combination of
Zero_Error(zero_err, 0.14683) Cylindrical_2Th_Correction(@, 0.15751)
Specimen_Displacement(@, 0.24136`)

We are using a capillaries for measurements. The goal is to create an input file for day to day use that provides a good intrument profile. Moreover, the idea is to refine as meaningful lattice parameters of new phases as possible without the use of an internal standard.

What would the best approach for this be? Especially when building a standard input file for refinement, these parameters, e.g. the th2_offset values or in the second case Zero_Error, Cylindrical_2Th_Correction, Specimen_Displacement should not be refined anymore, am I correct?

Thanks,
Simon

johnsoevans

The offsets in your plot look large to me. We'd usually have peaks within 0.01 degrees over the full 2-theta range on a reflection instrument. The diffractometer will have almost certainly have come with a specification sheet which states the accuracy of peak positions expected. You should try to match that.
You should look through tech ref or topas.inc for the meaning of the macros you're using. They're also discussed in Chapter 2 of the text Robert, Andreas and I wrote.
Speciment_Displacment() is specifically for reflection geometry and shouldn't be used in your case. It's also very highly correlated with Zero_Error() so you can rarely refine both together. If you put the displacement to zero you'll see the zero_err will just change to a much smaller number.
The Cylindrical_2Th_Correction() is a theta-dependent peak shift correction due to absorption as described by Sabine.
You'd be better starting with a displacement correction. There are a couple that are relevant for Debye-Scherrer geometry on the wiki: https://topas.awh.durham.ac.uk/doku.php?id=macros. They're similar to the e.g. Gozzo formalism.
I'd strongly recommend making sure the instrument alignment is good before applying too many corrections.

sdk_

Thank you for the clarification. I achieve a comparable fit to the data by just using a TCHZ profile function and the following macro: DS_Capillary_SD Capillary displacement correction (Debye-Scherrer) Pamela Whitfield

I will look into instrument alignment, unfortunately, I am just the guy trying to create an input file and cannot align myself..

Something that I notice is that there is a 2th dependent peak shape anisotropy that I cannot model. Can this be treated in some way, or is this also related to a possible misalignement of the instrument?