Dear Lu Xiaoke,

>>I tranform FCC into tetragonal structure, in which the direction of c axie is the right (111) direction of FCC structure.

Hi xiaoke,

you may try to calculate the energy versus different length of lattice length a (b). The basic idea is to using quadratic polynomial to fit the cure and find the lowest site, which gives the optimized lattice length. Besides, a shell script to do this will help a lot.

Dear V. V. Bannikov: first , thank you very much for your answer. but can you give a more detailed explantion ? I cannot understand that >>because (111) axis in cubic system is the axis of third order

Dear Lu Xiaoke,

Oh, perhaps, it was not a good expression... I meant that (111) axis in cubic system is 3-fold symmetry axis. As we apply the stress along that axis the cubic system become to be distorted to rhombohedral one keeping no 4-fold (tetragonal) symmetry axes.

Thank you! I get it .

 Dear V. V. Bannikov

Thank you very  much.

Hi Xiaoke,

I asked one of my colleagues who have related research experiences. He told me that you can customize the VASP source code for a strain-stress calculation.

1. There is a F file named “constr_cell_relax.F” in the subdirectory of vasp

Mine is ~vasp/vasp5.4.1/build/std

Note: ‘std’ can be changed to the certain type of VASP calculation of your interest, like vasp_ncl or vasp_gam

or using the following command to locate this file

find ~ -iname constr_cell_relax.F

2. Edit the F file

For a tensile strength calculation

Add FCELL(1,1)=0.0    

%%%it means when using ISIF=3, the X axis is fixed, yet stress in other directions are allowed to relax

For a shear strength calculation

Add FCELL(1,3)=0.0, and FCELL(3,1)=0.0 

3. Recompile the VASP

4. Write a shell script to start your calculation

Attention: The stress is along with the X axis, so rotate target direction to

X

Dear Cheng Zeng:

Thanks for your detailed explantation.The method you supported is very elaborative.

thanks for your help again. But ,do you konw the difference between the calculations of stress-strain in 2D and 3D material ? In particularly, the vacuume layer or the distance betweeen layers must be considered  in stress-strain calculations of  2D and 1D materials. In other words, the force calulated in low-demension material just is a 3D direction one. what is the relationship between the forces of low-demension and 3D materials?

In my opinion, if the distance between 2D layers is fixed, it is quite similar to a 3D material, i.e. one dimension elongated or compressed and the other dimension relaxed. However, it may not be the real case. Then you can adjust the vacuum distance, for example selecting several discrete points and searching for the most thermo-dynamically stable distance. It may be not right, but that is all I can tell. 

Cheng Zeng May I know what is the value of strength?? Which file can I find it?

Julio Cesar Gonzalez Hi Julio, I am sorry that I cannot help you with this as I have not used VASP for a very long time, hence I cannot recall what I mean.

Lu Xiaoke excuse me, May I ask where can we find the value of strength?? I know that the output file called as "OUTCAR" show something call pullay stress and external pressure.