Question about the formation energy calculations

[YongchenZhu]

Dear Prof. I have used this method to calculate the formation of 3D materials. But, I found the energy of the perfect supercell (just VASP calculations) and the defect materials (the energy obtained by the scpc method) with one negative charge has a big difference, nearly 100 eV, so I want to know is this normal? and for 3D materials, did we need set tne daming parameters? Thank you!

[isobestico]

Hello,

-- Could you please give more information about your issue? 
-- Could you send me the POSCAR and INCAR of your systems?
-- Precisely how did you get this 100eV? Please write down what you did to get this energy - the systems you used, and their energies.

[YongchenZhu]

The energy of the charged (-1) defect of my materials when I just used VASP is -502.69161 eV，but when I use SCPC method, the total energy becomes -436.163 eV，so the difference is nearly 66 eV，it's really a bigger difference. By the way, the correction energy in the SCPCOUT is 0.6 eV. I don't know how to use these energy to calculate the formation energy of this defect, what's the relationship between these energies.

[YongchenZhu]

You can see the details in the attached files test.zip

[isobestico]

Hi, sorry, with the information you are providing to me I cannot help at all. it would be nice to check the POSCAR files for the pristine and the defect system if it is possible. I am saying that because the SCPC can only be used with orthogonal cells and the material should be insulator or semi-conductor. Hence, it would be nice to check your POSCAR file. However, the SCPCOUT is pointing a correction of 22ev and a potential alignment of 1eV, probably your unit cell is too small or you are starting from scratch using HSE. For HSE you need to start the calculation using the previous WAVECAR of the uncorrected charged system as a starting point. In case, you are using hybrid functionals copy the WAVECAR and CHGCAR from the 'converged charged system without SCPC' to the folder you are going to run SCPC and start the method immediately IN=1.

[YongchenZhu]

Yes, I have used a preconverged WAVECAR and CHGCAR and set IN=1, but the same results I obtained. the lattice parameter I used is 11x11x12 A. The material is a semiconductor.

[isobestico]

Then, I don´t know what's going on and without the POSCAR it will be impossible to have a clue about if, sorry.

[isobestico]

Hi, in case you cannot share your POSCAR here due to some research aspect you can send me in my UnivLorraine email, check the original paper and send me your POSCAR directly to me in my UnivLorraine email

[isobestico]

Hi,

I got your file, and I deleted it from here because you sent some protected files. I will check your issue

using the defect in Vc(-2).

[isobestico]

Hello,

Regarding the hybrid functional, you should not use ALGO=Damped when use SCPC and initial WAVECAR from the uncorrected hse06 charged system. You can simple copy the WAVECAR from the fully uncorrected charged system into the scpc calculation and NOT use ALGO=D and TIME=XXX. I think this should fix the issue with the ALGO Damp and SCPC method. So

-- run your charged system without SCPC in HSE06 level and save the WAVECAR -- start your scpc in the HSE06 with the HSE06 wavecar from uncorreted charged system and do NOT use ALGO=D and TIME=XXX

[YongchenZhu]

Thank you. I have tested my system, It's fine now.