Forum Navigation
You need to log in to create posts and topics.

Polyhedral Template Matching for Defected Perovskite Structure

Dear Ovito Maintainers,

I'm trying to use the Polyhedral Template Matching (PTM) or the Common Neighbor Analysis (CNA) to identify bulk and grain boundary regions in my model, which seems feasible for systems like bcc alloys. However, for my model, these two methods seem not able to correctly identify the polyhedral coordinates in my model.

My model is relative complex for two reasons:

(1). It has a cubic perovskite structure. It seems that no existing template are for the perovskite materials.

(2). It contains some site vacancies. In the ABO3 perovskite structure, the A-site is defected with around 15% vacancies.

I'm opening this post to see if there is anything I can do to identify the perovskite polyhedral coordinate. And I want to understand which one or both of the above reasons lead to the failure of PTM and CNA to identify coordinates in the system. As reference, I have attached a snapshot of the Ovito interface analyzing my grain boundary model (interface in the middle, bulk on two ends with a rotation angle), in which PTM with existing template cannot identify the perovskite polyhedral coordinates in bulk regions. As our group just recently purchase the mini-group Pro account, my account is still not linked with any Pro license, and I'm quite inexperienced with Ovito. Please kindly let me know how to figure out this issue. Thanks a lot.

Uploaded files:
  • Screen-Shot-2022-02-26-at-8.16.38-PM.png

Dear admins, can I get some guidance or suggestions about issue in this post?  Thanks!

Hi Ji Qi,

I am not an Ovito admin but I developed PTM.

As you note, neither CNA nor PTM can detect perovskites. There is perhaps a trick you can try though.

If you delete the oxygen atoms, the remaining A and B atoms should form a BCC structure. This does not solve the problem of the vacancies but it might be sufficient to illustrate the grain structure.

Alternatively you could delete the A and B atoms. The remaining oxygen atoms should form a simple cubic structure, for which PTM has a template. You will not have the problem of vacancies here, although the cubic PTM template is often less reliable than the other templates because it only contains six atoms.

I haven't tried either of these proposals so I don't know how well they will work, but they may be worth trying.

Hi Ji Qi,

Lastly, you can remove center atoms of the perovskite structure during analysis leaving you with an FCC structure easily identifiable using PTM. Note, that the type select modifier and the "apply to selection" option of PTM may be helpful in your analysis.


New for our users in China: OVITO on WeChat 

Official OVITO WeChat channel operated by Foshan Diesi Technology Co., Ltd.