Synthetic Data Fitting- Goodness of Fit

[spekky_bandit]

Hi folks,

I am currently working on a project involving an unknown crystal sample. I have now identified the sample but I would like to quantify my findings by considering how well the structure I have determined fits other known structures. I have already ruled out fitting atomic positions to structures as this is very difficult unless the structures are very similar (same space group etc). I have instead opted to fitting a synthetic genetration of powder diffraction data for my crystal structure to synthetic generations from literature data.

My question is really about how to check the goodness of this fit? As both data sets are synthetic I have no associated errors so I cannot use a traditional chi squared tests. Does anyone have a suggestion on how this may be performed or any other method I may consider for comparing crystal structure quantitatively?

Cheers.

 

[mark726]

Hello,

Thank you for sharing your project with us. It sounds like you are on the right track by using synthetic data to compare your crystal structure to known structures. As you mentioned, using atomic positions can be difficult and may not provide an accurate comparison.

One method you could consider is using a program or software that calculates the Root Mean Square Deviation (RMSD) between two crystal structures. This measures the average distance between corresponding atoms in the two structures and can provide a quantitative measure of the fit.

Additionally, you could also look into using the R-factor or the weighted profile R-factor to compare the intensity values of your synthetic data to the literature data. This can provide a measure of how well your structure fits the diffraction data.

I hope these suggestions are helpful and best of luck with your project.