Elastic constants of a disordered rocksalt

[Inaoton]

Hi,
           I want to calculate the elastic constants of disordered rock salt structure crystallized in fcc. The structure has vacancies and the space group changes while construction. But i want to get only the three cubic elastic constants. How can I model in such a way that the detected symmetry is 225  so that only three elastic constants are calculated?

[Jess Wellendorff]

The ElasticConstants analysis object has the default option enable_symmetry=True, which means that only linearly independent elastic constants are calculated. For an FCC lattice with space group 225, only 3 elastic constants are therefore calculated. Example: (script attached)

Code

+------------------------------------------------------------------------------+
| Calculating elastic constants for the given bulk configuration               |
+------------------------------------------------------------------------------+
| Detected space group number:    225                                          |
| Detected lattice symmetry:      Cubic                                        |
| This lattice symmetry has 3 independent elastic constants:                   |
| C11  C12  C44                                                                |
+------------------------------------------------------------------------------+

As long as the configuration that you give as input for ElasticConstants analysis has space group 225, only those 3 elastic constants will be calculated. If the input lattice does not reconstruct when subjected to small strains, I think your calculation should be ok.

[Inaoton]

Thank you sir .
The ordered structure has space group 225. But I need to calculate elastic constants for the disordered structure  which has vacancies in it.  Obviously, the space group will change .  So when I feed the structure for elastic constants, it will try to calculate more than three elastic constants.
1. Can I modify the script or something else so that  the disordered structure will calculate only the three elastic constants of 225?
2. If not possible to calculate only three , can I  take only the three elastic constants from the output and analyse it? Will be the values reliable for analysis?  Can you cite some references?

[Petr Khomyakov]

For an amorphous material, the symmetry is much lower than that for an ordered phase, or very likely there is no point symmetry, so that you will have to calculate all the elastic constants, assuming no symmetry for your amorphous structure.

If you think that the symmetry is approximately preserved (or you want to enforce it) for the amorphous structure, you may do averaging (in a post-processing manner) over the elastic constants that are supposed to be identical for the ordered material. In this case, you will get three effective elastic constants.

[Inaoton]

Thank you sir
         The disordered structure  has tetragonal symmetry as detected with  space group 123 . Does the same averaging  method applicable for this?

[Petr Khomyakov]

You can certainly do it, but whether this is an adequate description of the elastic properties of your disordered material of interest, this is something you have to figure out yourself, as I do not know why you really want to describe the disordered materials with the 3 effective elastic constants instead of using the full elasticity tensor.

[Steven]

For an amorphous material, the symmetry is much lower than that for an ordered phase, or very likely there is no point symmetry, so that you will have to calculate all the elastic constants, assuming no symmetry for your amorphous structure.

If you think that the symmetry is approximately preserved (or you want to enforce it) for the amorphous structure, you may do averaging (in a post-processing manner) over the elastic constants that are supposed to be identical for the ordered material. In this case, you will get three effective elastic constants.

My question is how to perfrom averaging ...Can u give some references??

[Petr Khomyakov]

See a table with elasticity tensor for different cases at the very end at this web page, https://serc.carleton.edu/NAGTWorkshops/mineralogy/mineral_physics/tensors.html

[Steven]

               After calculation, I got 21 independent elastic constants (bcoz symmetry was not detected by the tool).
 Are saying that from 21 elastic contants, I have to ignore the zero component of the particular crystal system and to take the average of the equal components?

 

[Petr Khomyakov]

Yes if you are sure that your system has that particular symmetry.