Some symmetry problems in nonlinear optical calculations

Running Abinit Other tools and postprocessing in ABINIT
1

Hello everyone,

I am a beginner in using ABINIT and recently, I have been learning how to calculate the second-order nonlinear optical properties of materials with it. However, I have encountered some issues during this process, and after several unsuccessful attempts to resolve them on my own, I am hoping someone can assist me.

I did not encounter any problems when running the examples provided on the official website, but I faced symmetry issues while testing SnP2S6. The structural parameters of SnP2S6 were obtained using VASP optimization. Upon reviewing the run files, I found that ABINIT was unable to correctly recognize the space group (P146), and instead, it used the space group P1.

When I specified the space group as 146 using the Spgroup parameter and reran the calculation, ABINIT issued warnings about large errors when identifying equivalent atoms through symmetry. Initially, I did not pay much attention to this. However, after computing the entire second-order susceptibility tensor, I noticed that in theory, certain elements (such as in the ZXY direction) should not be zero, but ABINIT calculated them to be nearly zero.

I earnestly hope that someone with more experience can help me resolve this issue. Any assistance would be greatly appreciated!

I’m sorry, the system prompted that newly registered users are unable to upload files, so I provided my files in the form of pictures.

optic.abi(without Spgroup)
optic.abi(without Spgroup)1710×3999 300 KB

topic(without Spgrou).log
topic(without Spgrou).log1708×4355 240 KB

optic(Spgroup=146).abi
optic(Spgroup=146).abi1810×4113 241 KB

topic(Spgroup=146).log
topic(Spgroup=146).log1701×1164 40.5 KB

2

Hello YFF,
Which version of ABINIT do you use ?
It looks like you are using an old version, in which the tolerance on the input of atomic positions corresponded to tolsym 10e-8 . A couple of years ago, it was changed to the less stringent value 10e-5 .
Your VASP coordinates are not very precise, actually …
So: if you are using an old version of ABINIT, either switch to the latest available, or set tolsym to 10e-5 .
Best woshes,
Xavier

3

Hello Xavier,
Thank you for your valuable advisement!
The ABINIT version I am currently using is 9.0.4, which does seem a bit outdated. I will try installing a new version. May I ask which version is more suitable for me to install? I optimized the structure of this material with high precision using VASP and used it after confirming the spatial group is correct. May I ask why you would say that this structure is not very precise?
Thank you very much for your help!
YFF